Research Article

Quantitative live-cell imaging and computational modeling shed new light on endogenous WNT/CTNNB1 signaling dynamics

Developmental, Stem Cell and Cancer Biology, Swammerdam Institute for Life Sciences, University of Amsterdam, Netherlands
Biosystems Data Analysis, Swammerdam Institute for Life Sciences, University of Amsterdam, Netherlands
Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam, Netherlands
van Leeuwenhoek Centre for Advanced Microscopy, Swammerdam Institute for Life Sciences, University of Amsterdam, Netherlands

Jun 30, 2021

https://doi.org/10.7554/eLife.66440

Version of Record

Accepted for publication after peer review and revision.

Download
Cite
Share
CommentOpen annotations (there are currently 0 annotations on this page).

Version of Record published: August 5, 2021 (This version)
Accepted Manuscript published: June 30, 2021 (Go to version)
Accepted: June 29, 2021
Received: January 11, 2021

1. Of interest
Purinergic GPCR-integrin interactions drive pancreatic cancer cell invasion

Elena Tomas Bort, Megan Daisy Joseph ... Richard Philip Grose

Research Article Mar 21, 2023
Further reading

Article
Figures and data
Abstract
Introduction
Results
Discussion
Materials and methods
Data availability
References
Decision letter
Author response
Article and author information
Metrics

Abstract

WNT/CTNNB1 signaling regulates tissue development and homeostasis in all multicellular animals, but the underlying molecular mechanism remains incompletely understood. Specifically, quantitative insight into endogenous protein behavior is missing. Here, we combine CRISPR/Cas9-mediated genome editing and quantitative live-cell microscopy to measure the dynamics, diffusion characteristics and absolute concentrations of fluorescently tagged, endogenous CTNNB1 in human cells under both physiological and oncogenic conditions. State-of-the-art imaging reveals that a substantial fraction of CTNNB1 resides in slow-diffusing cytoplasmic complexes, irrespective of the activation status of the pathway. This cytoplasmic CTNNB1 complex undergoes a major reduction in size when WNT/CTNNB1 is (hyper)activated. Based on our biophysical measurements, we build a computational model of WNT/CTNNB1 signaling. Our integrated experimental and computational approach reveals that WNT pathway activation regulates the dynamic distribution of free and complexed CTNNB1 across different subcellular compartments through three regulatory nodes: the destruction complex, nucleocytoplasmic shuttling, and nuclear retention.

Introduction

WNT signaling is one of the most ancient pattern-forming cell signaling cascades. It drives many biological processes from the onset of embryogenesis until adulthood in all multicellular animals (reviewed in van Amerongen and Nusse, 2009; Holstein, 2012; Loh et al., 2016). WNT signaling remains important throughout the lifespan of the organism and controls stem cell maintenance in many mammalian tissues, including the intestine (Barker et al., 2007). Disruption of the pathway causes disease, with hyperactivation being a frequent event in human colorectal and other cancers (reviewed in Nusse and Clevers, 2017; Wiese et al., 2018).

The key regulatory event in WNT/CTNNB1 signaling (traditionally known as ‘canonical WNT signaling’) is the accumulation and nuclear translocation of the transcriptional co-activator β-catenin (Catenin beta-1, hereafter abbreviated as CTNNB1) (Figure 1A). In the absence of WNT signaling, rapid turnover by the so-called destruction complex maintains low levels of CTNNB1. This cytoplasmic complex consists of the scaffold proteins Adenomatous Polyposis Coli Protein (APC) and Axis Inhibition Proteins 1 and 2 (AXIN), which bind CTNNB1, and the serine/threonine kinases Casein kinase I isoform alpha (CSNK1A1) and Glycogen Synthase Kinase-3 alpha and beta (GSK3), which subsequently phosphorylate residues S45, T41, S37, and S33 (Amit et al., 2002; Liu et al., 2002). This primes CTNNB1 for ubiquitination by E3 Ubiquitin Protein Ligase beta-TrCP 1 and 2 (BTRC and FBXW11) and subsequent proteasomal degradation (Aberle et al., 1997; Latres et al., 1999). In the current working model for WNT/CTNNB1 signaling, binding of WNT ligands to the Frizzled (FZD) and low-density lipoprotein receptor-related protein 5 and 6 (LRP) receptor complex sequesters and inhibits the destruction complex at the membrane in a process that involves Disheveled (DVL) (Bilic et al., 2007; Kan et al., 2020; Ma et al., 2020; Schwarz-Romond et al., 2007). This allows newly synthesized CTNNB1 to accumulate and translocate to the nucleus, where CTNNB1 binds to TCF/LEF transcription factors (TCF7, TCF7L1, TCF7L2, and LEF1) to regulate target gene transcription as part of a larger transcriptional complex (Behrens et al., 1996; Fiedler et al., 2015; Molenaar et al., 1996; van Tienen et al., 2017).

Figure 1 with 2 supplements see all

Download asset Open asset

Generation of HAP1^SGFP2-CTNNB1 cell lines.

(A) Cartoon depicting the current model of the WNT/CTNNB1 pathway. In the absence of WNT ligands (left, ‘OFF’), free cytoplasmic CTNNB1 is captured by the destruction complex consisting of AXIN, APC, CSNK1A1, and GSK3, which leads to its phosphorylation, BTRC-mediated ubiquitination and subsequent proteasomal degradation, resulting in low levels of CTNNB1 in the cytoplasm and nucleus. Binding of the WNT protein (right, ‘ON’) to the FZD and LRP receptors inhibits the destruction complex through DVL. CTNNB1 accumulates in the cytoplasm and subsequently translocates to the nucleus, where it promotes the transcription of target genes, such as *AXIN2*, as a co-activator of TCF/LEF transcription factors. (B) Cartoon depicting exon 2 of the *CTNNB1* locus, which contains the start codon, and the CTNNB1 protein before (top) and after (bottom) introduction of the SGFP2 by CRISPR/Cas9-mediated homology directed repair. (C) Schematic of the experimental workflow and timeline for generating HAP1^SGFP2-CTNNB1 clones. Cas9, gRNA and repair templates are transfected as plasmids. The repair template contains the coding sequence of SGFP2 surrounded by 800 bp homology arms on either side and lacks the gRNA recognition site (see supplement 2 of this figure). A short puromycin selection step is included from 24 to 48 hr after transfection to enrich for transfected cells. Haploid, GFP-positive cells are sorted, and single cell clones are expanded for further analysis. (**D–F**) FACS plots illustrating control (D) and SGFP2-CTNNB1-tagged cells (**E–F**). (D) Cells transfected with Cas9 and gRNA in the absence of a repair template were used to set the gate for SGFP2-positive events. (E) A small population of cells expressing low levels of SGFP2 can be detected when cells are transfected with Cas9, gRNA, and repair template. (F) Treatment for 24 hr of cells similar to those depicted in (D) with 8 μM CHIR99021 does not change the amount of cells that are SGFP2 positive, but increases the SFP2 signal, most likely reflecting an increase in SGFP2-tagged beta catenin levels on a per cell basis and supporting the notion that the gated events indeed represent successfully tagged cells.

The working model for WNT/CTNNB1 signaling described above is the result of almost 40 years of research. The use of traditional genetic and biochemical approaches has allowed identification of the core players, as well as dissection of the main signaling events. However, multiple aspects of WNT/CTNNB1 signaling remain poorly understood. For instance, the exact molecular composition of the destruction complex as well as the mechanism for its inhibition remain unclear (reviewed in Tortelote et al., 2017). How WNT/CTNNB1 signaling regulates the subcellular distribution of CTNNB1 also requires further scrutiny.

Most biochemical techniques lead to loss of spatial information and averaging of cell-to-cell heterogeneity, since proteins are extracted from their cellular context. Additionally, temporal information is usually limited to intervals of several minutes or hours. Live-cell microscopy offers better spatiotemporal resolution. Currently, however, many of these studies are conducted by overexpressing the protein(s) of interest. This can severely affect activation, localization, and complex formation (Gibson et al., 2013; Mahen et al., 2014). Although stabilization of CTNNB1 by WNT signaling has been extensively studied, very few studies have focused on the spatiotemporal dynamics of this process – especially at the endogenous level (Chhabra et al., 2019; Massey et al., 2019; Rim et al., 2020).

Here, we use CRISPR/Cas9-mediated genome editing in haploid cells to generate clonal cell lines that express fluorescently tagged CTNNB1. Using confocal imaging and automated cell segmentation, we quantify the dynamic subcellular increase of endogenous CTNNB1 upon WNT stimulation. Moreover, using Fluorescence Correlation Spectroscopy (FCS) and Number and Brightness (N and B) analysis, we measure the mobility and concentration of CTNNB1, providing detailed information on CTNNB1 containing complexes in the cytoplasm and nucleus. Next, we use these biophysical parameters to build a computational model of WNT/CTNNB1 signaling that predicts the levels and subcellular distribution of CTNNB1 across its cytoplasmic and nuclear pools. Using this integrated experimental and computational approach, we find that WNT regulates the dynamic distribution of CTNNB1 across different functional pools by modulating three regulatory nodes: cytoplasmic destruction, nucleocytoplasmic shuttling, and nuclear retention. Finally, we strengthen the link between our data and the model via specific experimental perturbations, which shows that the regulatory nodes responsible for nuclear retention and nuclear shuttling of CTNNB1 are equally important under physiological and oncogenic conditions.

Results

Generation and functional validation of clonal HAP1^SGFP2-CTNNB1 cell lines

To visualize and quantify the spatiotemporal dynamics of WNT/CTNNB1 signaling at the endogenous level, we fluorescently tagged CTNNB1 in mammalian cells using CRISPR/Cas9-mediated homology directed repair (Ran et al., 2013; Figure 1). To preserve the existing (epi)genetic control mechanisms of CTNNB1 expression, only the coding sequence for SGFP2, a monomeric, bright and photostable green fluorescent protein (Kremers et al., 2007), was seamlessly inserted at the starting ATG of the CTNNB1 coding sequence in HAP1 cells, a WNT-responsive near haploid cell line (Figure 1B, Figure 1—figure supplement 1A; Andersson et al., 1987; Carette et al., 2011; Kotecki et al., 1999; Lebensohn et al., 2016). The choice for this haploid cell line ensured tagging of the complete CTNNB1 protein pool (Figure 1C), thus overcoming the limitations of polyploid cell lines where genome editing often results in a combination of correctly and incorrectly edited alleles (Canaj et al., 2019).

We isolated clonal cell lines with the desired modification by FACS sorting (Figure 1D–F) with a gating strategy that specifically selected for haploid cells (Figure 1—figure supplement 2), since HAP1 cells can become diploid or polyploid over time (Essletzbichler et al., 2014; Yaguchi et al., 2018). Genome editing of wild-type HAP1 (HAP1^WT) cells resulted in a small population with low SGFP2 fluorescence (0.2%) (Figure 1D–E). The intensity, but not the number of cells in this population increased upon treatment with CHIR99021, a potent and selective GSK3 inhibitor (Bain et al., 2007), providing a strong indication that these fluorescent events corresponded to HAP1 cells in which the SGFP2 sequence was successfully knocked into the endogenous CTNNB1 locus (HAP1^SGFP2-CTNNB1) (Figure 1F). While scarless tagging of endogenous genes in HAP1 cells was relatively cumbersome (only 0.2% gated events), PCR-based screening and sanger sequencing revealed that the desired repair occurred with almost 90% efficiency within this population (Figure 1—figure supplement 1).

To verify that the SGFP2 tag did not interfere with CTNNB1 function, three clonal HAP1^SGFP2-CTNNB1 cell lines were further characterized using established experimental readouts for WNT/CTNNB1 signaling (Figure 2 and Figure 2—figure supplement 1). Western blot analysis confirmed that the HAP1^SGFP2-CTNNB1 clones did not contain any untagged CTNNB1 but only expressed the SGFP2-CTNNB1 fusion protein (Figure 2A). Moreover, the total levels of SGFP2-CTNNB1 in tagged cell lines increased to the same extent as wild-type CTNNB1 in untagged cells in response to CHIR99021 treatment (Figure 2A–B). Similarly, untagged and tagged CTNNB1 induced target gene expression in equal measure, as measured by a TCF/LEF responsive luciferase reporter (Hu et al., 2007; Figure 2C), and increased transcription of the universal WNT/CTNNB1 target AXIN2 (Lustig et al., 2002; Figure 2D). Finally, while unstimulated cells mainly showed SGFP2-CTNNB1 localization at adherens junctions, treatment with purified WNT3A protein (Figure 2E) and CHIR99021 (Figure 2—figure supplement 1E) increased SGFP2-CTNNB1 levels in the cytoplasm and nucleus consistent with its signaling function.

Figure 2 with 1 supplement see all

Download asset Open asset

Functional validation of three independent HAP1^SGFP2-CTNNB1 clones.

(A) Western blot, showing CTNNB1 (HAP1^WT) and SGFP2-CTNNB1 (HAP1^SGFP2-CTNNB1 clones 1, 2, and 3) accumulation in response to CHIR99021 treatment. All panels are from one blot that was cut at the 70 kDa mark and was stained with secondary antibodies with different fluorophores for detection. Top: HAP1^WT cells express CTNNB1 at the expected wild-type size. Each of the three clonal HAP1^SGFP2-CTNNB1 cell lines only express the larger, SGFP2-tagged form of CTNNB1, that runs at the expected height (~27 kDa above the wild-type CTNNB1). Middle: Only the tagged clones express the SGFP2-CTNNB1 fusion protein, as detected with an anti-GFP antibody at the same height. Bottom: alpha-Tubulin (TUBA) loading control. A representative image of n=3 independent experiments is shown. (B) Quantification of Western blots from n=three independent experiments, including the one in (A), confirming that the accumulation of CTNNB1 in response to WNT/CTNNB1 pathway activation is comparable between HAP1^WT and HAP1^SGFP2-CTNNB1 cells. Horizontal bar indicates the mean. (C) Graph depicting the results from a MegaTopflash dual luciferase reporter assay, showing comparable levels of TCF/LEF reporter gene activation for HAP1^WT and HAP1^SGFP2-CTNNB1 cells in response to CHIR99021 treatment. Data points from n=3 independent experiments are shown. Horizontal bar indicates the mean. Values are depicted relative to the DMSO control, which was set to one for each individual cell line. (D) Graph depicting *AXIN2* mRNA induction in response to CHIR99021 treatment, demonstrating that induced expression of an endogenous target gene is comparable between HAP1^WT and HAP1^SGFP2-CTNNB1 cells. Data points represent n=3 independent experiments. Horizontal bar represents the mean. *HPRT* was used as a reference gene. Values are depicted relative to the HAP1^WT DMSO control, which was set to 1. (E) Representative confocal microscopy images of the three HAP1 ^SGFP2-CTNNB1 clones after 4 hr vehicle control or 100 ng/ml WNT3A treatment from n=1 biological experiment, revealing intracellular accumulation of SGFP2-CTNNB1 (green). Nuclei were counterstained with SiR-DNA dye (magenta). Scale bar is 10 μm.

Figure 2—source data 1 Numerical data for Figure 2B,C,D and Figure 2—figure supplement 1A,C,D.: https://cdn.elifesciences.org/articles/66440/elife-66440-fig2-data1-v2.xlsx
Download elife-66440-fig2-data1-v2.xlsx

Taken together, WNT-responsive changes in CTNNB1 levels, localization and activity are preserved after CRISPR/Cas9-mediated fluorescent tagging of the entire CTNNB1 protein pool. Although there is some variation between the three clones with respect to CTNNB1 stabilization and target gene activation, this is likely due to the sub-cloning of these cell lines rather than the targeting per se.

Live imaging of endogenous SGFP2-CTNNB1 during WNT pathway activation

To better understand the temporal dynamics of endogenous CTNNB1 stabilization, we performed live-cell imaging over 12 hr in HAP1^SGFP2-CTNNB1 clone 2 (Figure 3, Video 1; Video 2; Video 3) with different levels of WNT stimulation. Unstimulated cells showed a stable CTNNB1 signal at the cell membrane throughout the imaging time course (Figure 3A, Video 1). The membrane localization of CTNNB1 is consistent with its structural role in adherens junctions (Valenta et al., 2012; Yap et al., 1997), which we will not consider further in the current study. Stimulation with different concentrations of purified WNT3A resulted in a heterogeneous response pattern, with some cells in the population showing a far more prominent increase in CTNNB1 levels in the cytoplasm and nucleus than others (Figure 3A, Figure 3—figure supplement 1A–B, Video 2).

Figure 3 with 3 supplements see all

Download asset Open asset

Live imaging of HAP1SGFP2-CTNNB1.

(A) Representative stills from confocal time-lapse experiments corresponding to Videos 1–2, showing an increase of SGFP2-CTNNB1 after treatment with 100 ng/ml WNT3A (bottom) relative to a vehicle control (BSA)-treated sample (top). Scale bar = 20 μm. (**B–D**) Quantification of time-lapse microscopy series, using the segmentation pipeline shown in (E). Arrow indicates the moment of starting the different treatments (T, see legend in B for details). (**B–C**) Graph depicting the normalized intensity of SGFP2-CTNNB1 in the cytoplasm (B) or nucleus (C) over time. Solid lines represent the mean normalized fluorescence intensity and shading indicates the 95% confidence interval. n=155–393 cells for each condition and time point, pooled data from n=three independent biological experiments. (D) Graph depicting the nuclear/cytoplasmic ratio of SGFP2-CTNNB1 over time, calculated from raw intensity values underlying (B) and (C). (E) Segmentation of nuclei (top) and cytoplasm (bottom) based on the SiR-DNA signal and SGFP2-CTNNB1 signal. Scale bar = 20 μm.

Video 1

Download asset

posterframe for video — Representative video of confocal time-lapse experiments, showing SGFP2-CTNNB1 (left, green), SiR-DNA staining (middle, magenta), and transmission image (right, gray) after treatment with vehicle control (BSA).

Time of addition is at 00:00:00 (indicated at the top left). Scale bar in the lower right represents 20μm.

Video 2

Download asset

To quantify these dynamic changes, we developed a custom-built automated segmentation pipeline in CellProfiler (Figure 3D). Quantification showed that the temporal dynamics of CTNNB1 accumulation were independent of the dose of WNT3A (Figure 3B–C, Videos 4–5), although this could be different for lower doses and other cell types (Massey et al., 2019). Treatment with 100 ng/ml WNT3A increased SGFP2-CTNNB1 fluorescence 1.74-fold (mean, 95% CI 1.73–1.76) in the cytoplasm and 3.00-fold (mean, 95% CI 2.97–3.03) in the nucleus, with similar results in the other two HAP1^SGFP2-CTNNB1 clones (Figure 3—figure supplement 2).

Video 3

Download asset

Video 4

Download asset

Video 5

Download asset

Our quantification further shows that nuclear accumulation of CTNNB1 is favored over a cytoplasmic increase (compare the fold-changes in Figure 3B–C). Moreover, the first statistically significant increases in fluorescence intensity in the cytoplasm could be detected after ~45 min of treatment (Video 4, Figure 3—figure supplement 1C), whereas in the nucleus an increase was first statistically significant after ~30 min (Video 5, Figure 3—figure supplement 1D). To examine the relation between the cytoplasmic and nuclear CTNNB1 pools more closely, we calculated the ratio between nuclear and cytoplasmic intensities of SGFP2-CTNNB1 (Figure 3D, Video 6). In untreated cells, the nuclear/cytoplasmic ratio was 0.652 (mean [3–5 hr], 95% CI 0.649–0.657), showing that SGFP2-CTNNB1 was preferentially localized to the cytoplasm (Figure 3D, Figure 3—figure supplement 3). For the first 3 hr after WNT3A, nuclear CTNNB1 levels rose considerably faster than cytoplasmic CTNNB1 levels until the nuclear/cytoplasmic ratio showed a slight nuclear enrichment of 1.08 (mean [3–5 hr] 95% CI 1.07–1.10) for 100 ng/ml WNT3A. This indicates that not only the turn-over, but also the subcellular localization of CTNNB1 is actively regulated both before and after WNT pathway activation.

Video 6

Download asset

Establishing a fitting model for SGFP2-CTNNB1 diffusion

Having measured the relative changes in the cytoplasmic and nuclear levels of CTNNB1 in response to WNT3A stimulation, we next sought to exploit our experimental system to quantify additional molecular properties of CTNNB1 in each of these subcellular compartments using Fluorescence Correlation Spectroscopy (FCS). FCS is a powerful method to measure the mobility and absolute levels of fluorescent particles in a nanomolar range, compatible with typical levels of signaling proteins in a cell (reviewed in Hink, 2014). It has for instance been used to gain insight into the assembly of DVL3 supramolecular complexes (Yokoyama et al., 2012), the endogenous concentrations and mobility of nuclear complexes (Holzmann et al., 2019; Lam et al., 2012), and most recently, to quantify ligand-receptor binding reactions in the WNT pathway (Eckert et al., 2020). In point FCS, the fluorescence intensity is measured in a single point (Figure 4A,D–E). Diffusion of labeled particles, in this case SGFP2-CTNNB1, causes fluctuation of the fluorescence signal over time (Figure 4B). By correlating the fluorescence intensity signal to itself over increasing time-intervals, an autocorrelation curve is generated (Figure 4C). To extract relevant biophysical parameters, such as mobility (a measure for size) and the absolute numbers of the fluorescent particles (corresponding to their concentration), this autocorrelation curve is fitted with an appropriate model.

Figure 4 with 1 supplement see all

Download asset Open asset

Two diffusion-component fit-model for SGFP2-CTNNB1 FCS measurements.

(A) Schematic representation of the point FCS technique, depicting the confocal volume with fluorescent particles diffusing in and out. Particles in FCS are defined by their coherent movement; therefore, a particle can be made up of monomers or multimers in isolation or complexed to unlabeled molecules. (B) Schematic representation of intensity fluctuations over time as measured in the confocal volume. Fluctuations are the result of both photo-physics (e.g. blinking of the fluorophore), diffusion and the number of particles in the confocal volume. (C) Graphical representation of the two diffusion-component fitting model used for our autocorrelation curves. T_trip describes the blinking of the SGFP2 fluorophore and the after-pulsing artefact. T_diff1 and T_diff2 describe the monomeric and complexed form of SGFP2-CTNNB1, respectively. Details of all fitting parameters are described in Materials and methods. (D) Representative confocal images of HAP1^SGFP2-CTNNB1 cells treated for 4 hr with BSA (left) or 100 ng/ml WNT3A (right). (E) Zoom in of the white rectangle in (D), with representative locations of FCS measurement points for cytoplasm (C) and nucleus (N) indicated with white crosses in the SGFP2-CTNNB1 channel and transmission channel. (**F–G**) Fitting of a representative autocorrelation curve with one unfixed diffusion-component (F) or a two diffusion-component model (G), where the first diffusion component was fixed to the speed of free monomeric SGFP2-CTNNB1 (14.9 µm²/s) and the second diffusion component was unfixed. The black line represents the autocorrelation curve generated from the FCS measurement; the red line represents the fitted model. The residuals after fitting of 25 individual curves are shown in the upper right corner of the graphs.

Figure 4—source data 1 Numerical data for Figure 4F,G and Figure 4—figure supplement 1.: https://cdn.elifesciences.org/articles/66440/elife-66440-fig4-data1-v2.xlsx
Download elife-66440-fig4-data1-v2.xlsx

We first attempted to fit the autocorrelation curves obtained with point FCS measurements in HAP1^SGFP2-CTNNB1 cells with a one-component model (i.e. containing one single diffusion speed for SGFP2-CTNNB1). This model was unable to fit most of our data (Figure 4F). The current literature suggests that while a large portion of CTNNB1 is present as a monomer (Gottardi and Gumbiner, 2004; Maher et al., 2010), CTNNB1 is also present in multiprotein complexes in the cytoplasm and in the nucleus (reviewed in Gammons and Bienz, 2018). We therefore tested the fit of a two-component model. To this end, we deduced the theoretical diffusion speed of monomeric, unbound SGFP2-CTNNB1 to be 14.9 µm²/s. This theoretical speed was confirmed by fitting an unbiased two-component model to our experimental data (Figure 4—figure supplement 1). To limit variability due to noise in the measurements, we proceeded with the two component model in which the first diffusion component was fixed to the theoretically determined diffusion speed of monomeric SGFP2-CTNNB1 (14.9 µm²/s) and with the second diffusion component limited to slower speeds compatible with point-FCS imaging (see Materials and methods for details). This model provided good fits for our autocorrelation curves obtained in both cytoplasmic and nuclear point FCS measurements (Figure 4G). Together this is consistent with the presence of free monomeric CTNNB1 (first, fast component) and larger CTNNB1 containing complexes (second, slow component) in both the nucleus and cytoplasm.

Quantification of absolute SGFP2-CTNNB1 concentrations

Using this fitting model, we determined, for the first time, the absolute concentrations of endogenous CTNNB1 in living cells in presence and absence of a physiological WNT stimulus (Figure 5A, Table 1). In the absence of WNT3A, we determined the total concentration of SGFP2-CTNNB1 to be 180 nM (median, 95% CI 127–218) in the cytoplasm and 122 nM (median, 95% CI 91–158) in the nucleus. This is consistent with the nuclear exclusion we observed with confocal imaging (Figure 3).

Figure 5 with 2 supplements see all

Download asset Open asset

Abundance and mobility of SGFP2-CTNNB1 molecules in living cells after 4 hr WNT3A treatment or control.

Details on sample size and statistics can be found in Supplementary file 1. (A) Graph depicting the total concentration of SGFP2-CTNNB1 particles (monomeric plus complexed) as measured with FCS. (B) Graph depicting the concentration of SGFP2-CTNNB1 particles with the fast diffusion component (i.e. free monomeric). (C) Graph depicting the concentration of SGFP2-CTNNB1 containing particles with the slow diffusion component (i.e. complex associated). (**D–E**) Graphs depicting the fraction (D) and speed (E) of the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS. (F) Example of typical regions of interest in two cells used in N and B analysis. Solid line represents the analysis ROI, dashed line, marks the outline of the nuclear envelope. (G) Schematic representation of a confocal volume with different brightness species. On the left are six monomers with a brightness of 1, on the right two trimers with a brightness of 3, both result in a fluorescence of 6. N and B analysis is able to extract the number and the brightness of such samples, for more detail see supplement 1 of this figure. (H) Graph depicting the molecular brightness of SGFP2-CTNNB1 in the cytoplasm and nucleus relative to controls as measured with N and B in the same subcellular compartments. EGFP monomer was used for normalization and EGFP dimer as a control for N and B measurements.

Figure 5—source data 1 Numerical data for Figure 5 (all graphs) and Figure 5—figure supplement 1 (all panels).: https://cdn.elifesciences.org/articles/66440/elife-66440-fig5-data1-v2.xlsx
Download elife-66440-fig5-data1-v2.xlsx

Table 1

Total number of SGFP2-CTNNB1 molecules in the confocal volume and corresponding calculated concentrations obtained from FCS measurements in n=3 independent experiments.

The concentration is calculated from the number of molecules in the confocal volume and the calibrated confocal volume (see Materials and methods). The number of molecules is consistent with those measured with N and B analysis (Figure 5—figure supplement 1A, Supplementary file 1).

			Number of molecules		Concentration (nM)
Compartment	Treatment	N	Median	95% CI	Median	95% CI
Cytoplasm	BSA	21	80	70–116	180	127–218
Cytoplasm	WNT3A	21	95	85–122	221	144–250
Nucleus	BSA	21	63	53–72	122	91–158
Nucleus	WNT3A	18	135	127–150	240	217–325

In the presence of WNT3A, we measured a 1.2-fold increase in the total SGFP2-CTNNB1 concentration to 221 nM (median, 95% CI 144–250 nM) in the cytoplasm. This increase was smaller than expected from fluorescence intensity measurements (Figure 3B). We excluded that this was caused by photo-bleaching and other photophysical effects, and currently have no explanation for this discrepancy (Figure 5—figure supplement 1, see also Materials and methods and discussion). In the nucleus the concentration increased 2.0-fold to 240 nM (median, 95% CI 217–325) upon pathway activation. Nuclear concentrations of SGFP2-CTNNB1 therefore exceed cytoplasmic concentrations after WNT3A treatment, consistent with the nuclear accumulation observed with live imaging (Figure 3). These concentrations are in a similar range as those previously determined by quantitative mass spectrometry in different mammalian cell lines (Kitazawa et al., 2017; Tan et al., 2012). Of note, the exact concentrations will likely vary between cell types and their calculated values may also be dependent on the intricacies and assumptions that underlie each individual measurement technique.

Our two-component fitting model also allowed us to discriminate between pools of SGFP2-CTNNB1 with different mobility (Table 2), that is fast diffusing monomeric CTNNB1 (Figure 5B) and slow diffusing complexed CTNNB1 (Figure 5C). In the nucleus, the concentration of fast moving CTNNB1 increased 2.0-fold from 87 nM (median, 95% CI 78–119) to 170 nM (median, 95% CI 147–214), while slow moving CTNNB1 concentration increased 3.9-fold from 22 nM (median, 95% CI 4–40) to 86 nM (median, 95% CI 67–114). This is also reflected by the increase in the bound fraction of SGFP2-CTNNB1 the nucleus (Figure 5D). The preferential increase of the slow-moving fraction is consistent with the notion that upon WNT stimulation CTNNB1 will become associated with the chromatin in a TCF-dependent transcriptional complex (or ‘WNT enhanceosome’).

Table 2

Number of SGFP2-CTNNB1 molecules in the confocal volume and corresponding calculated concentration of SGFP2-CTNNB1 molecules with the fast or slow diffusion coefficient obtained from FCS measurements in n=3 independent experiments.

The concentration is calculated from the number of molecules and the calibrated confocal volume (see Materials and methods).

			Fast SGFP2-CTNNB1				Slow SGFP2-CTNNB1
			Number of molecules		Concentration (nM)		Number of molecules		Concentration (nM)
Compartment	Treatment	n	Median	95% CI	Median	95% CI	Median	95% CI	Median	95% CI
Cytoplasm	BSA	21	51	40–63	91	66–139	29	20–37	57	38–76
Cytoplasm	WNT3A	21	60	47–80	145	76–168	35	30–41	68	57–76
Nucleus	BSA	21	48	41–66	87	78–119	13	2–22	22	4–40
Nucleus	WNT3A	18	96	81–101	170	147–214	47	37–49	86	64–104

Of note, in the cytoplasm, the concentration of both fast and slow SGFP2-CTNNB1 increased upon WNT3A treatment (Figure 5B–C), with the fraction of bound SGFP2-CTNNB1 remaining equal between stimulated (median 0.38, 95% CI 0.29–0.46) and unstimulated cells (median 0.34, 95% CI 0.31–0.4) (Figure 5D). The fact that a large portion of CTNNB1 remains in a complex after WNT stimulation, challenges the notion that mainly monomeric CTNNB1 accumulates, as commonly depicted in the textbook model (Figure 1A).

Quantification of SGFP2-CTNNB1 mobility

While we cannot determine the exact composition of the SGFP2-CTNNB1 complex, we do obtain biophysical parameters that are linked to its size. For instance, the diffusion coefficient of the nuclear SGFP2-CTNNB1 complex was 0.17 μm²s⁻¹ (median, 95% CI 0.14–0.22) in cells treated with purified WNT3A (Figure 5E). This is comparable to the diffusion coefficients measured for other chromatin-bound transcriptional activators (Lam et al., 2012), which further supports that this pool represents the WNT enhanceosome.

In the cytoplasm, we determined the second diffusion coefficient of SGFP2-CTNNB1 to be 0.13 μm²s⁻¹ (median, 95% CI 0.13–0.17) in the absence of WNT3A stimulation (Figure 5E). This is indicative of very large complexes containing SGFP2-CTNNB1 that move with diffusion kinetics comparable to those previously observed for the 26S proteasome (Pack et al., 2014). Of note, the speed of the cytoplasmic complex increased 3.5-fold to 0.46 μm²s⁻¹ (95% CI of the median 0.37–0.57) after WNT3A treatment. Because changes in diffusion coefficient are typically indicative of much larger changes in molecular weight (i.e. three-dimensional protein complex size, see Materials and methods section FCS data acquisition and analysis for details), this indicates that the size of the cytoplasmic CTNNB1 complex drastically changes when the WNT pathway is activated. Thus, although the fraction of CTNNB1 that resides in a complex remains the same (34–38%), the identity of the cytoplasmic complex is quite different in unstimulated and WNT3A-stimulated cells.

Determining the multimerization status of SGFP2-CTNNB1

Recent work suggests that the CTNNB1 destruction complex (also known as the ‘degradosome’) is a large and multivalent complex, mainly as the result of AXIN and APC multimerization (reviewed in Schaefer and Peifer, 2019). The speed of the slow CTNNB1 component, determined by the second diffusion coefficient in our FCS measurements (Figure 5E), is consistent with this model. Such a large, multivalent destruction complex would be expected to have multiple CTNNB1-binding sites. To measure the number of bound SGFP2-CTNNB1 molecules within this cytoplasmic complex, we performed Number and Brightness (N and B) analysis (Figure 5F–G, Figure 5—figure supplement 2). N and B is a fluorescence fluctuation spectroscopy technique similar to point FCS, but it makes use of image stacks acquired over time rather than individual point measurements (Digman et al., 2008). By quantifying the variance in fluorescence intensity of this stack, not only the number of particles but also their brightness can be determined.

Because brightness is an inherent property of a fluorophore, a change in brightness is a measure of the number of fluorophores per particle. In our case, the brightness is indicative of the number of SGFP2-CTNNB1 molecules per complex. As N and B does not incorporate diffusion kinetics, we cannot differentiate between monomeric (which would have a brightness of one) and complexed CTNNB1 (which would have a brightness exceeding one if multiple CTNNB1 molecules reside in a single complex). Therefore, the measured brightness of SGFP2-CTNNB1 in our N and B analysis is an average of both fractions.

First, we confirmed that the number of particles we determined using N and B, were highly similar to those obtained with FCS (compare Figure 5—figure supplement 1A with Table 1). Second, we observe that the total pool of SGFP2-CTNNB1 in both the cytoplasm and nucleus has a brightness similar to EGFP and SGFP2 monomers (Figure 5H, Table 3). Because we found a substantial fraction (34–38%) of SGFP2-CTNNB1 to reside in a large complex using point FCS (Figure 5C–D), this suggests that few, if any, of these complexes contain multiple SGFP2-CTNNB1 molecules. If the cytoplasmic SGFP2-CTNNB1 containing complex indeed represents a large, multivalent destruction complex, this would imply that under physiological conditions, quite unexpectedly, most CTNNB1-binding sites are unoccupied in both the absence and presence of WNT3A.

Table 3

Brightness of SGFP2-compared relative to EGFP-monomer and -dimer controls in n=2 independent experiments. N is the number of analyzed cells.

p-Values were calculated using PlotsOfDifferences that uses a randomization test (Goedhart, 2019).Note that only the EGFP-dimer is significantly different to the EGFP-monomer control, while SGFP2-CTNNB1 is not.

Fluorophore	Compartment	Treatment	N	Median	95 CI median	p-value to matched control (EGFP monomer in the nucleus or cytoplasm)
EGFP-monomer	Cytoplasm	NA	15	1	0.79–1.34	1.000
EGFP-dimer	Cytoplasm	NA	14	1.4	1.29–1.60	0.011*
SGFP2-CTNNB1	Cytoplasm	BSA	69	0.92	0.83–1.00	0.738
SGFP2-CTNNB1	Cytoplasm	100 ng/ml WNT3A	46	1.01	0.93–1.11	0.919
EGFP-monomer	Nucleus	NA	15	1	0.91–1.07	1.000
EGFP-dimer	Nucleus	NA	14	1.62	1.44–1.69	<0.001*
SGFP2-CTNNB1	Nucleus	BSA	69	0.87	0.78–0.96	0.192
SGFP2-CTNNB1	Nucleus	100 ng/ml WNT3A	46	1.05	0.95–1.15	0.578

A minimal computational model of WNT/CTNNB1 signaling

Quantitative measurements and physical parameters of WNT pathway components and their interactions remain limited (Kitazawa et al., 2017; Lee et al., 2003; Tan et al., 2012), especially in living cells. As we obtained absolute measurements of different functional pools of CTNNB1, we next sought to integrate these biophysical parameters in a minimal computational model of WNT signaling to identify the critical nodes of regulation of subcellular CTNNB1 distribution (Figure 6A, Tables 4–5, Materials and methods). This minimal model is based on a previous model of Kirschner and colleagues (Lee et al., 2003), and incorporates the new data obtained in our study, supplemented with parameters from the literature (Lee et al., 2003; Tan et al., 2012).

Figure 6

Download asset Open asset

Computational model of WNT/CTNNB1 based on FCS concentrations for free and complexed CTNNB1 (Tables 1–2).

(A) Schematic overview of the model. DC=destruction complex, DC* = DVL-inactivated DC, CB=cytoplasmic CTNNB1, CB*=phosphorylated CB, NB=nuclear CTNNB1, TCF=TCF/LEF transcription factors, DVL=WNT-activated DVL. In WNT OFF, w=0, and therefore k5/k4 does not play a role and no inactivated destruction complex is formed. In WNT ON, w=1, and k5/k4 is put into action, resulting in an increase in DC* at the expense of DC. Under the assumption that k3 remains equal and given that CB*-DC was experimentally determined to be the same in WNT ON and WNT OFF, removal of DC, results in an increase in CB. Changes in k6/k7 and k9/8 further increase NB and NB-TCF in WNT ON. Note that CB* is degraded and therefore plays no role in the model. (B) Graph depicting the modeled concentrations of cytoplasmic components over time. The black line indicates total concentration of cytoplasmic CTNNB1, corresponding to Figure 3B. (C) Graph depicting the modeled concentrations of nuclear components over time. The black line indicates total concentration of nuclear CTNNB1, corresponding to Figure 3C. (D) Graph depicting the ratio of total nuclear and cytoplasmic CTNNB1 over time, corresponding to the measurements in Figure 3D. (E) Graph depicting the DC-bound CTNNB1 fraction ratio over time. (F) Graph depicting the TCF-bound CTNNB1 fraction ratio over time.

Table 4

Variables Minimal Model of WNT signaling.

Model name	Variable	Compound	Values obtained from	WNT OFF (nM)	WNT ON (nM)
CB	$x_{1}$	Free cytoplasmic CTNNB1	FCS data this report	91	145
DC	$x_{2}$	Free destruction complex	Model equations	82.4	52
CB*-DC	$x_{3}$	DC-bound phosphorylated CTNNB1	FCS data this report*	62.5*	62.5*
DC*	$x_{4}$	Inactivated destruction complex	Model equations	0**	30.5**
NB	$x_{5}$	Free nuclear CTNNB1	FCS data this report	87	170
TCF	$x_{6}$	Free TCF	Model equations	81	17
NB-TCF	$x_{7}$	TCF-bound nuclear CTNNB1	FCS data this report	22.2	86
TCF⁰	${T C F}^{0}$	Total TCF	$x_{7}$ and Tan et al., 2012 - Figure 11	103	103

^*Under the assumption that k3 does not change, the levels of CB*-DC remain equal. Since there was no significant difference between the concentration of slow SGFP2-CTNNB1 in the absence or presence of WNT3A (57 nM versus 68 nM, not significant, Table 2) the average of both medians (62.5 nM) was used. ** In WNT OFF, w=0, and no inactivated destruction complex is formed. In WNT ON, w=1, which induces the formation of inactivated destruction complex at the expense of free destruction complex (see Equations 7a and 9 in the model description in Materials and methods).

Table 5

Equilibrium conditions for the Minimal Model of WNT signaling.

All rates are multiplied with factor R=20, so that the equilibrium is reached at 4.5 hr according to Figure 4C–D.

Rate constant		Biological process	Values based on	Wnt off	Wnt on
b	nMmin⁻¹	CTNNB1 synthesis	$v_{12}$ from Lee	0.423	0.423
$\frac{k_{2}}{k_{1}}$	nM	Binding to and phosphorylation by the destruction complex of cytoplasmic CTNNB1	$K8$ from Lee	120	120
$k_{3}$	min⁻¹	Dissociation and degradation of phosphorylated CTNNB1 from the destruction complex	Deduced from $b$ and $x_{3}$	0.0068	0.0068
$\frac{k_{5}}{k_{4}}$	nM	Inactivation of the destruction complex by activated DVL	Fitted to $x_{1}$ and $x_{7}$	N.A.*	1.7
$\frac{k_{6}}{k_{7}}$		Ratio between nuclear import and export of CTNNB1	Deduced from $x_{1}$ and $x_{5}$	0.96	1.17
$\frac{k_{9}}{k_{8}}$	nM	Dissociation of nuclear CTNNB1 from TCF	Deduced from $x_{5}$ , ${T C F}^{0}$ , $x_{7}$	320	33.6

^*In WNT OFF, w=0, and no inactive destruction complex is formed. Only in WNT ON, w=1, which induces the formation of inactivated destruction complex at the expense of free destruction complex (see Equations 7a and 9 in the model description in Materials and methods).

Our model diverges from the model presented by Lee et al. on two major points. First, the model is simplified by replacing the details of the destruction complex formation cycle and the individual actions of APC and AXIN with a single, fully formed destruction complex. We chose this option because our study does not provide new quantitative data on the formation and dynamics of the destruction complex, but does provide absolute concentrations of CTNNB1 in a bound state in the cytoplasm. Second, we explicitly include shuttling of CTNNB1 between the cytoplasm and nucleus in both directions (Schmitz et al., 2013; Tan et al., 2014).

Thus, our model (Figure 6A) describes the binding of cytoplasmic CTNNB1 (‘CB’) to the destruction complex (‘DC’) leading to its phosphorylation and degradation (described by k3), which releases the DC. Transport of CTNNB1 from the cytoplasm to the nucleus, allows nuclear CTNNB1 (‘NB’) to bind to TCF/LEF forming a transcriptional complex (‘NB-TCF’). When WNT is present in the system, we describe the inactivation of the destruction complex (‘DC*’) by DVL through the parameter $w$ (see Materials and methods section model description). The model is available as interactive app at https://wntlab.shinyapps.io/WNT_minimal_model/ and allows users to explore the effects of modulating different equilibria and constants in an intuitive way.

Our model faithfully recapitulates the dynamic changes observed with functional imaging (compare Figure 6B–F to Figures 3 and 5). Moreover, it reveals two critical regulatory nodes in addition to the requisite inactivation of the destruction complex (described by k5/k4). The first additional node of regulation is nuclear import and export (or ‘shuttling’, described by k6/k7). Upon WNT stimulation, the ratio of k6/k7 (nuclear shuttling) needs to increase in order for the model to match the free CTNNB1 concentrations we measured by FCS (Table 5, Figure 5B). Thus, the balance shifts from nuclear export before WNT, to nuclear import after WNT. The second additional node of regulation is the association of CTNNB1 with the TCF transcriptional complex (or ‘retention’), described by k9/k8. Upon WNT stimulation, the ratio of k9/k8 (nuclear retention) needs to decrease by almost a factor of 10 in order for the model to reproduce the concentrations of free and bound CTNNB1 in the nucleus as measured by FCS (Table 5, Figure 6F, Figure 5B–C). Thus, association of CTNNB1 to the TCF transcriptional complex is favored after WNT stimulation. In summary, our model suggests that WNT/CTNNB1 signaling is regulated at three distinct levels of the signal transduction pathway: destruction complex inactivation, nucleocytoplasmic shuttling and nuclear retention. How WNT signaling influences nuclear shuttling and nuclear retention is an open question and both are areas of active research (Anthony et al., 2020; Söderholm and Cantù, 2020).

Perturbing the system to mimic oncogenic WNT signaling

WNT signaling is often disrupted in cancer (reviewed in Polakis, 2000; Zhan et al., 2017), frequently due to inactivating mutations in negative regulators or due to activating mutations in CTNNB1 itself (Bugter et al., 2021). One of the earliest identified mutations in CTNNB1 was a substitution of serine-45 for a phenylalanine (S45F) (Morin et al., 1997). This mutation removes the CSNK1A1 priming phosphorylation site on CTNNB1 that is needed for sequential phosphorylation by GSK3, and thus blocks its proteasomal degradation (Amit et al., 2002; Liu et al., 2002).

We generated the S45F mutation in one of our HAP1^SGFP2-CTNNB1 cell lines through a second step of CRISPR/Cas9-mediated genome editing (Figure 7—figure supplement 1A–D). As expected, the mutation resulted in higher CTNNB1 levels (Figure 7—figure supplement 1E–F) and constitutive downstream activation of the pathway (Figure 7—figure supplement 1G–H). Next, we used this cell line for two purposes. First, we used FCS and N and B to compare the complex-state of wild-type and mutant CTNNB1 in the cytoplasm (Figure 7). Second, we reproduced the same perturbation in silico to strengthen the link between our experimental data and the computational model (Figure 8).

Figure 7 with 1 supplement see all

Download asset Open asset

Cytoplasmic complex characteristics in absence of SGFP2-CTNNB1 N-terminal phosphorylation.

The S45F mutant was introduced using CRISPR (see Figure 7—figure supplement 1) and CHIR treated and control cells were measured after 24 hr. Details on sample size and statistics can be found in Supplementary file 1. (A) Graph depicting the fraction of particles with the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS for S45F mutant (B) Graph depicting the speed of the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS for S45F mutant. (C).Graph depicting the fraction of particles with the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS after 24 hr treatment with CHIR99021 (C) Graphs depicting the speed of the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS after 24 hr treatment with CHIR99021. (**E–F**) Graphs depicting the molecular brightness of SGFP2-CTNNB1 in the cytoplasm relative to controls as measured with N and B in the same subcellular compartments for S45F mutant CTNNB1 (E) or after 24 hr of CHIR99021 treatment (F). EGFP monomer was used for normalization and EGFP dimer and trimer as controls for N and B measurements.

Figure 8 with 2 supplements see all

Download asset Open asset

In silico and experimental perturbation of WNT signaling.

Details on experimental sample size and statistics can be found in Supplementary file 1. (A) Representative confocal images of HAP1-SGFP2WT (WT, top) and HAP1-SGFP2S45F (S45F, bottom) cells acquired with the same image settings. The S45F mutation leads to the accumulation and nuclear enrichment of CTNNB1 in the cell. (B) Graph depicting the total concentration of SGFP2-CTNNB1 particles (monomeric plus complexed) as measured with FCS. (**C–D**) Inhibition of CTNNB1 degradation is modelled as a reduction in the value of k3. (C) Graph depicting the predicted total cytoplasmic CTNNB1 concentration as a function of k3. A reduction in k3 from 0.0068 (Table 5, WNT ON and WNT OFF conditions) to ~0.0043 (dotted line) corresponds to the cytoplasmic concentration observed (solid line). (D) Graph depicting the predicted total nuclear CTNNB1 concentration as a function of k3. The solid horizontal line indicates the concentration measured for the S45F mutant by FCS. Note that the value of k3 that matches the observed cytoplasm concentration (dotted line) does not match the experimentally determined concentration in the nucleus (solid line). (**E–F**) Graphs depicting the fraction of particles with the second diffusion component (i.e. SGFP2-CTNNB1 containing complex) measured by FCS for wild-type and S45F mutant (E) and after 24 hr CHIR99021 treatment (F). The increase in the bound fraction in the oncogenic mutant or after GSK3 inhibition we find, is comparable to what we observed in WNT3A stimulated cells (Figure 5D). (G) Graph showing the predicted nuclear bound fraction of CTNNB1as a function of k3 with the TCF/CTNNB1 binding affinity of the model (Table 4) for WNT OFF (k9/k8 = 320, pink line) and for WNT ON (k9/k8=33.5, blue line). Note that for WNT ON, the value for the nuclear bound fraction approximates the experimentally determined slow fraction for the S45F mutant (solid line, panel E) at the value for k3 that matches the cytoplasmic concentration of CTNNB1 (dotted line). (H) Graph showing the predicted nuclear/cytoplasmic (N/C)-ratio as a function of k3 with TCF/CTNNB1 binding affinity of the model (Table 4) for WNT OFF (k9/k8=320, pink line) and WNT ON (k9/k8=33.5, blue line). Note that, although for WNT ON the value of the N/C-ratio increases with k3, there is still nuclear exclusion (N/C-ratio lower than 1, dashed line) at the value of k3 that matches the cytoplasmic CTNNB1-concentration (dotted line). (**I–K**) The N/C-ratio as measured by FCS for wild-type and S45F mutant (I), after 24 hr CHIR99021 treatment (J) and after 4 hr WNT3A treatment (K). Note that all perturbations lead to nuclear accumulation (N/C-value exceeding 1). (L) Graph showing the predicted N/C-ratio as a function of k3 with the WNT ON value for k9/k8 with the nuclear shuttling ratio of the model (k6/k7 Table 4),corresponding to WNT OFF (k6/k7=0.96, pink line) and WNT ON (k6/k7=1.17, blue line), respectively. Note that the WNT ON value of k6/k7 increases the N/C-ratio to nuclear accumulation at the value for k3 that matches the cytoplasmic concentration (dotted line). (M) Graph depicting the predicted total cytoplasmic CTNNB1 concentration as a function of k3 with WNT ON and WNT OFF values for k9/k8 and k6/k7. Note that modulation of k9/k8 and k6/k7 has virtually no effect on the predicted cytoplasmic concentration of CTNNB1, resulting in overlapping points and lines in the graph. The horizontal solid line is the experimentally determined cytoplasmic CTBNN1 concentration (cf. panel B); the vertical dotted line is at the value of k3 that best reproduces this experimental finding in the model. (N) Graph depicting the predicted total nuclear CTNNB1 concentration as a function of k3 for WNT ON and WNT OFF values for k9/k8 and k6/k7. Note that if both k9/k8 and k6/7 are changed from their WNT OFF values the predicted nuclear concentration of CTBNN1 better matches the experimentally determined concentration (horizontal solid line) at the value for k3 that matches the cytoplasm concentration (vertical dotted line).

Similar to the situation detected under physiological conditions (Figure 5D, slow fraction: median 0.38), we find a large fraction (median 0.402, 95% CI 0.363–0.471) of SGFP2-CTNNB1^S45F to reside in a cytoplasmic complex (Figure 7A). As observed for physiological stimulation with WNT3A (Figure 5E), the speed of this complex is significantly increased (approximately threefold) in SGFP2-CTNNB1^S45F (median 0.589 μm²s⁻¹, 95% CI 0.585–0.691) compared to unstimulated HAP1^SGFP2-CTNNB1 cells (median 0.191 μm²s⁻¹, 95% CI 0.115–0.29) (Figure 7B). We find similar behavior when we block the GSK3-mediated phosphorylation of wild-type CTNNB1 using CHIR99021 (Figure 7C–D). The reduction in cytoplasmic complex size therefore must occur downstream of CTNNB1 phosphorylation. Intriguingly, our N and B analyses suggest that these smaller S45F mutant cytoplasmic complexes have a higher occupancy of CTNNB1 (Figure 7E) than the those in WNT3A (Figure 5H) or CHIR99021 (Figure 7F) stimulated wildtype cells. The S45F mutant (median 1.304, 95% CI 1.139–1.418, p=0.002) was significantly brighter than the SGFP2 monomer control (median 0.866, 95% CI 0.573–0.949), where the WT-tagged HAP1 cells again did not diverge from the monomer (0.886, 95% CI 0.722–1, p=0.845) (file 1). Thus, under oncogenic conditions more binding sites in the cytoplasmic CTNNB1 complex may be occupied than in physiological circumstances.

The S45F mutant shows a substantial increase in SGFP2-CTNNB1 levels in the cytoplasm and nucleus (Figure 8A). As this constitutive mutation does not provide any kinetic information, we also measured the dynamic response of SGFP2-CTNNB1 to CHIR99021-mediated GSK3 inhibition (Figure 8—figure supplement 1, Video 3). We see similar initial kinetics as for WNT3A stimulation. However, in contrast to what is observed for WNT3A treatment, no plateau was reached at the highest concentration of CHIR99021 (8 µM). Of note, the quantification also confirms that there is cell to cell heterogeneity in the response, regardless of whether WNT/CTNNB1 signaling is activated at the level of the receptor (WNT3A treatment) or at the level of the destruction complex (CHIR99021 treatment), as can be seen from the spread of intensities measured from individual cells (Figure 3—figure supplement 1A–B).

Finally, we compared our biological measurements from these perturbation experiments to our computational model predictions. Both the S45F mutation and CHIR99021 treatment disrupt degradation of phosphorylated CTNNB1 (corresponding to k3, Figure 6A). With FCS and N and B we quantified the accumulation of CTNNB1 levels of mutant SGFP2-CTNNB1^S45F (Figure 8B, Figure 8—figure supplement 2A) and wild-type SGFP2-CTNNB1 upon CHIR99021 treatment (Figure 8—figure supplement 2B–C). Both exceeded the levels observed with physiological WNT3A stimulation (Figure 3A–C, Figure 5A). Specifically, the absolute concentration of SGFP2-CTNNB1^S45F in the cytoplasm (median 351 nM, 95% CI 276–412) exceeded that of SGFP2-CTNNB1 in WNT3A-treated cells (median 221 nM, 95% CI 144–250). In the nucleus, the concentration of SGFP2-CTNNB1^S45F reached 429 nM (median, 95% CI 387–481), as opposed to 240 nM (median, 95% CI 217–325) for SGFP2-CTNNB1 in WNT3A-treated cells as a result of losing its priming phosphorylation site. While this further increase in concentration is evident, it should be noted that in both the cytoplasm and the nucleus CTNNB1 levels thus rise less than twofold in an oncogenic setting compared to WNT3A treatment.

In our computational model, we simulated reduced degradation by lowering the value of k3. A reduction in k3 from its initial value (k3=0.0068, Table 5) to k3=0.0043, accurately predicted the higher cytoplasmic concentration measured for the S45F mutant (Figure 8C), but a further reduction to k3=0.0038 was needed to match the measured nuclear concentration (Figure 8D). However, reducing k3 alone was not sufficient to reproduce either the fraction of CTNNB1 that is bound in the nucleus (Figure 8E–G) or the overall nuclear enrichment of CTNNB1 (Figure 8H). The latter requires a predicted nuclear/cytoplasmic (N/C) ratio greater than one, as observed in both physiological and constitutively active WNT/CTNNB1 signaling (Figure 8I–K).

The fraction of bound SGFP2-CTNNB1 in the nucleus was comparable between our HAP1^{SGFP2-CTNNB1(S45F)} mutant cell line (median 0.29, 95% CI 0.27–0.33) (Figure 8E), 8 μM CHIR99021 (median 0.38, 95% CI 0.29–0.46) (Figure 8F) and WNT3A (median 0.32, 95% CI 0.30–0.34) (Figure 5D) treated wild-type HAP1^SGFP2-CTNNB1 cells. This experimental observation can be matched by adjusting the k9/k8 (nuclear retention) ratio, as was also required for physiological WNT3A signaling (Table 5, Figure 8G). This shows the importance of this regulatory node not only in physiological, but also in oncogenic signaling. At the same time, the adjustment of nuclear retention (k9/k8) on top of a reduction in degradation (k3) still does not predict the observed nuclear enrichment of CTNNB1 (Figure 8H). After changing the nuclear shuttling ratio (k6/k7) to the ratio we fitted for the WNT ON situation (Table 5), the model now also reproduces the nuclear enrichment of CTNNB1 (Figure 8L). In Figure 8M–N, we show that these additional changes in nuclear shuttling (k6/k7) and nuclear retention (k9/k8) have little effect on the CTNNB1 concentration in the cytoplasm, but do substantially affect the nuclear concentrations of CTNNB1. This suggest that processes downstream of CTNNB1 degradation play a significant and active role in the CTNNB1 dynamics of the cell.

Taken together, our computational model can describe both physiological and oncogenic signaling. Moreover, it underlines the importance of CTNNB1 regulation downstream of destruction complex activity and confirms a critical role for nuclear import and nuclear retention.

Discussion

WNT signaling is critical for tissue development and homeostasis. Although most core players and many of their molecular interactions have been uncovered, dynamic spatiotemporal information with sufficient subcellular resolution remains limited. As both genome editing approaches and quantitative live-cell microscopy have advanced further, the goal of studying WNT/CTNNB1 signaling at endogenous expression levels in living cells now is within reach. Maintaining endogenous expression levels is important, as overexpression may lead to altered stoichiometry of signaling components, as well as changes in subcellular localization (Gibson et al., 2013; Mahen et al., 2014). Indeed, it has been shown that exogenously expressed CTNNB1 is less signaling competent, probably due to its post-translational modification status (Hendriksen et al., 2008).

Here, we generated functional HAP1^SGFP2-CTNNB1 knock-in cell lines to study the dynamic behavior and subcellular complex state of endogenous CTNNB1 in both a physiological and oncogenic context. Importantly, this allowed us to measure hitherto unknown biophysical parameters of WNT/CTNNB1 in individual living human cells for the first time. Using live-cell microscopy and automated cell segmentation, we observe that endogenous CTNNB1 levels increase only 1.7-fold in the cytoplasm and 3.0-fold in the nucleus after WNT3A treatment, which is consistent with the literature (Jacobsen et al., 2016; Kafri et al., 2016; Massey et al., 2019). Next, we used state-of-the-art, quantitative microscopy to measure the absolute concentration of CTNNB1 within different subcellular compartments and in different complex states in living cells. The findings from these experiments definitively challenge the still prevailing view that mainly monomeric CTNNB1 accumulates upon WNT pathway stimulation (Nusse and Clevers, 2017). Moreover, our integrative approach of quantitative imaging and computational modeling revealed three critical nodes of CTNNB1 regulation, namely CTNNB1 degradation, nuclear shuttling and nuclear retention, which together describe the CTNNB1 turnover, subcellular localization and complex status under both physiological and oncogenic conditions.

Cytoplasmic regulation of CTNNB1

Using FCS, we determined that in unstimulated HAP1 cells a substantial fraction (~30–40%) of SGFP2-CTNNB1 is associated with a very large, slow-diffusing cytoplasmic complex (Figures 4–5 and 7). The main known cytoplasmic complex containing CTNNB1 is the destruction complex. The combined weight of the individual destruction complex components (AXIN, APC, CSNK1A1, and GSK3) would be expected to result in a much higher mobility than that displayed by the cytoplasmic CTNNB1-containing complex we observed. However, evidence is growing that the destruction complex forms large phase separated assemblies (also termed biomolecular condensates) (reviewed in Schaefer and Peifer, 2019). Oligomerization of AXIN and APC underlies the formation of these assemblies, and this in turn appears to be required for efficient degradation of CTNNB1 (Fiedler et al., 2011; Kunttas-Tatli et al., 2014; Pronobis et al., 2017; Spink et al., 2000). There is some evidence that these biomolecular condensates form at (near) endogenous levels (Fagotto et al., 1999; Faux et al., 2008; Mendoza-Topaz et al., 2011; Pronobis et al., 2015; Schaefer et al., 2018; Thorvaldsen et al., 2015), but it is still an open question what the exact composition and size of the destruction complex is in a physiological context. It should be noted that our imaging does not visualize punctae, which are typically associated with these biomolecular condensates (Figure 3A). In addition, our N and B data indicate that most of the slow diffusing CTNNB1 complexes contain one or very few SGFP2-CTNNB1 molecules in either the absence or presence of WNT3A stimulation. In view of the above-mentioned destruction complex oligomerization and its presumed multivalency, this finding was quite unexpected. Several mechanisms could explain this apparent discrepancy. On the one hand, destruction complex multimerization at endogenous levels might be more subtle than previously thought. For example, quantification of AXIN polymerization in vitro showed that even at exceedingly high concentration (24 μM), AXIN polymers typically contained only eight molecules (Kan et al., 2020). On the other hand, even if the multivalent destruction complex offers multiple CTNNB1 binding sites, occupancy at any one time might be low, due to the continuous and high turnover of CTNNB1. In this respect, the CTNNB1 bindings sites in the destruction complex could be envisioned to act similar to the wooden vanes in the paddle wheel of an old-fashioned watermill: like the water in the analogous example, CTNNB1 would be continuously scooped up (for phosphorylation) and dropped off (for degradation).

Only following the introduction of an S45F mutation, which results in constitutive inhibition of CTNNB1 phosphorylation and degradation, we observe a brightness increase that would be compatible with the accumulation of multiple SGFP2-CTNNB1^S45F molecules in a single cytoplasmic complex. This indicates that while the destruction complex might be multivalent in both a physiological and an oncogenic context, CTNNB1 occupancy of the complex is low under physiological conditions, but increased in oncogenic signaling. This has major impacts on how we conceptualize the workings of the CTNNB1 destruction machinery – especially in the context of cancer, since mutations in CTNNB1 (affecting occupancy) may have very different biochemical consequences than mutations in APC (affecting multimerization and valency of the destruction complex itself).

The mechanism on destruction complex deactivation remains controversial (Tortelote et al., 2017; Verkaar et al., 2012). The current literature suggests that the destruction complex is sequestered to the FZD-LRP receptor complex upon WNT pathway stimulation. Several models exist for how membrane sequestration inhibits CTNNB1 degradation, including LRP-mediated GSK3 inhibition (Stamos et al., 2014), sequestration of GSK3 in multi vesicular bodies (Taelman et al., 2010), (partial) dissociation of the destruction complex (Liu et al., 2005; Tran and Polakis, 2012), and saturation of CTNNB1 within an intact destruction complex (Li et al., 2012). Our data clearly show that a substantial fraction of CTNNB1 in the cytoplasm remains bound upon pathway stimulation (Figure 5D). This is not captured by textbook models and cartoons that typically still depict the view that mainly monomeric CTNNB1 accumulates. It is, however, in line with the notion that CTNNB1 still accumulates in a destruction complex after WNT signaling. One line of evidence proposes that WNT traps CTNNB1 in a complex by inhibiting its transfer to the E3 ligase, rather than by inhibiting its phosphorylation and release (Li et al., 2012; Pronobis et al., 2015; Schaefer et al., 2020). However, other reports strongly suggest that CTNNB1 phosphorylation is at least partially inhibited during the WNT response (Hernández et al., 2012; Mukherjee et al., 2018).

The substantial fraction of slow diffusing CTNNB1 that remains upon physiological and oncogenic stimulation of the pathway, is also consistent with the previously proposed role for cytoplasmic retention by the destruction complex (Krieghoff et al., 2006; Roberts et al., 2011; Yamulla et al., 2014). These studies also show that both nuclear and cytoplasmic retention have an important role in determining the subcellular distribution of CTNNB1. Moreover, cytoplasmic retention of CTNNB1 can contribute to reducing downstream pathway activation even in the presence of oncogenic APC mutations (Kohler et al., 2008; Li et al., 2012; Schneikert et al., 2007; Yang et al., 2006), which further highlights the importance of this process.

We show that the cytoplasmic CTNNB1 complex in WNT3A or CHIR99021 treated cells as well as in S45F mutant cells has an over threefold increased mobility compared to control cells. Therefore, while the diffusion coefficient is still very low (indicating that the remaining complex is still very large), this implies it is a vastly different complex than that observed in the absence of WNT stimulation. The precise nature of these complexes remains unknown, but could be consistent with a reduced destruction complex size after WNT treatment, as also recently observed in Drosophila for AXIN complexes (Schaefer et al., 2018), or with the formation of inactivated destruction complexes (‘transducer complexes’) in response to WNT/CTNNB1 pathway activation (Hagemann et al., 2014; Lybrand et al., 2019). The fact that cells in which GSK3 phosphorylation is inhibited through S45F mutation or CHIR99021 treatment show similar behavior, suggests that the size of the cytoplasmic complex is directly linked to the phosphorylation status of CTNNB1. The destruction complex has been shown to associate with (parts of) the ubiquitin and proteasome machinery (Li et al., 2012; Lui et al., 2011; Schaefer et al., 2020; Thorvaldsen et al., 2015). One interesting possibility, therefore, is that phosphorylated CTNNB1 is required for coupling the destruction complex to the ubiquitination and proteasome machinery. In fact, although not explicitly mentioned in the main text, supplementary table 1 of Li et al., 2012 shows that in HEK293 cells, which harbor no mutation in the core components of the WNT pathway, CTNNB1 was found to interact with subunits of the proteasome, whereas in the S45F-CTNNB1 mutant cell line Ls174T these interactions were not detected. In conclusion, although we do not directly determine its identity, our measured biophysical parameters of the cytoplasmic CTNNB1 complex are consistent with it representing a large, multivalent destruction complex that is coupled to the proteasome as long as CTNNB1 is being phosphorylated.

Nuclear regulation of CTNNB1

The key function of CTNNB1 downstream of WNT is to regulate transcription of TCF/LEF target genes (Doumpas et al., 2019; Schuijers et al., 2014). Proteomic analyses have shown that the WNT enhanceosome consists of CTNNB1, TCF/LEF, Pygopus Homologs 1 and 2 (PYGO) and B-cell CLL/lymphoma nine protein (BCL9) and several other large proteins (Fiedler et al., 2015; van Tienen et al., 2017). Using FCS, we show that CTNNB1 resides in a nuclear complex with a diffusion coefficient that is compatible with such a DNA-bound transcriptional complex (Figure 5E; Lam et al., 2012).

Although CTNNB1 is known to associate with TCF/LEF factors in response to WNT/CTNNB1 signaling to drive transcription (Franz et al., 2017; Schuijers et al., 2014), we also detect low levels of nuclear CTNNB1 complex in the absence of a WNT stimulus (Figure 5C). The diffusion coefficient of the nuclear CTNNB1 complex does not change upon the addition of WNT3A (Figure 5E), suggesting that some CTNNB1 is already associated with the DNA even in the absence of a WNT stimulus. At this point, we cannot exclude the contribution of TCF/LEF independent DNA binding (Armstrong et al., 2012; Essers et al., 2005; Kormish et al., 2010), or anomalous subdiffusion in the nucleus, either due to physical obstruction, transient DNA-binding events protein or protein complex formation (Dross et al., 2009; Kaur et al., 2013; Wachsmuth et al., 2000), as FCS only allows us to probe the speed of this complex.

However, upon pathway activation through WNT3A, CHIR99021 or S45F mutation we see a consistent increase in the fraction and absolute levels of this slow-diffusing nuclear CTNNB1 complex (Figure 5E, Figure 8E–F), compatible with increased CTNNB1 binding to its target sites. Upon WNT stimulation, the concentration of bound SGFP2-CTNNB1 in the nucleus increased to ~90 nM, which corresponds to something in the order of 20,000 bound CTNNB1 molecules in one nucleus, assuming a nuclear volume of 0.36 picoliter (Tan et al., 2012). Published CHIPseq studies report many CTNNB1 DNA binding sites, ranging from several hundred to several thousand sites in mammalian cells (Cantù et al., 2018; Doumpas et al., 2019; Schuijers et al., 2014). It is therefore highly likely that at least part of the slow-diffusing CTNNB1 particles we measure indeed represents CTNNB1 that is associated with the WNT enhanceosome.

Regulation of CTNNB1 nuclear accumulation

In HAP1 cells, endogenous CTNNB1 is excluded from the nucleus in the absence of WNT. Our live imaging data reveal an immediate and preferential increase in nuclear CTNNB1 upon WNT3A stimulation, until an equilibrium is reached between the cytoplasmic and nuclear levels (Figure 3D). This is consistent with previous observations in HEK293 cells stably overexpressing low levels of YFP-CTNNB1 (Kafri et al., 2016).

Intriguingly, CTNNB1 does not contain nuclear import or export signals and can translocate independently of classical importin and exporter pathways (Fagotto et al., 1998; Wiechens et al., 2001; Yokoya et al., 1999). Hence, the molecular mechanism of CTNNB1 subcellular distribution remains incompletely understood. Evidence from Fluorescence Recovery After Photobleaching (FRAP) studies suggest that the increase in nuclear CTNNB1 is due to changes in binding to its interaction partners in the cytoplasm and nucleus (retention) rather than active changes in nuclear import and export rates (shuttling) (Jamieson et al., 2011; Krieghoff et al., 2006). We argue that the two are not mutually exclusive, as our experimental data and computational model show that WNT regulates both nucleocytoplasmic shuttling and nuclear retention of CTNNB1. Indeed, we see an increase of nuclear CTNNB1 complexes in the nucleus (Figure 5C–D) and the dissociation of CTNNB1 from TCF is reduced almost 10-fold in WNT signaling conditions in our computational model (Table 5). Our model predicts that this increased nuclear retention indeed also increases the nuclear/cytoplasmic ratio (Figure 8H). However, to reconcile our computational prediction with our experimental observations, we additionally need to include a shift from nuclear export to nuclear import upon pathway activation (Figure 6, Figure 8). Our integrated experimental biology and computational modeling approach thus reveals that WNT signaling not only regulates the absolute levels of CTNNB1 through destruction complex inactivation, but also actively changes its subcellular distribution through nuclear retention and shuttling. The fact that direct inhibition of GSK3-mediated phosphorylation of CTNNB1 results in the same behavior, indicates that the phosphorylation status of CTNNB1 plays a critical role. This further emphasizes the importance of posttranslational modifications and conformational changes in CTNNB1 for its subcellular localization and function (Gottardi and Gumbiner, 2004; Sayat et al., 2008; Valenta et al., 2012; van der Wal and van Amerongen, 2020; Wu et al., 2008).

Differences and similarities in physiological and oncogenic WNT signaling

As discussed above, several behaviors of CTNNB1 are conserved between the different modes of stimulation. For instance, WNT3A treatment, GSK3 inhibition and the oncogenic S45F mutation all result in (1) increased overall levels of CTNNB1; (2) a substantial fraction of CTNNB1 in a faster, albeit still very large complex in the cytoplasm; (3) increased nuclear accumulation of CTNNB1; and (4) increased retention of CTNNB1 in nuclear complexes. Our computational modeling further confirms that in addition to regulation of CTNNB1 turnover – either by removal of activated destruction complex or through inhibition of phosphorylation and ubiquitination – nuclear shuttling and nuclear retention are equally important regulatory nodes in oncogenic (CHIR99021 treatment or S45F mutation) and physiological (WNT3A stimulation) signaling.

However, the absolute levels of CTNNB1 and the resulting transcriptional activation are distinct in each of these conditions: Cells treated with a GSK3 inhibitor continue to accumulate CTNNB1 after 4 hr, when WNT3A treated cells reach a plateau. The latter is likely due to the fact that negative feedback mechanisms kick in, such as reconstitution of the destruction complex by AXIN2 or internalization of WNT-bound receptor complexes (Agajanian et al., 2019; Lustig et al., 2002), both of which function upstream of GSK3. Alternatively, it could reflect the notion that physiological WNT signaling does not turn the destruction complex off completely, but rather ‘turns it down’, as our N and B data support the fact that under physiological conditions the destruction complex itself provides a surplus reservoir of CTNNB1-binding sites that may only become occupied when WNT signaling is hyperactivated. As a combined result, GSK3 inhibition or S45F mutation of CTNNB1 can result in higher total intracellular levels of CTNNB1. Indeed, concentrations of SGFP2-CTNNB1 in S45F mutated cells exceed those in WNT3A treated cells in both the cytoplasm (1.6-fold) and the nucleus (1.8-fold). This subtle increase in CTNNB1 levels is likely amplified at the transcriptional level (Jacobsen et al., 2016), consistent with the well-known fact that constitutive activation of the pathway through different mechanisms, including APC mutation, results in higher pathway activation than physiological stimuli.

Challenges and opportunities for fluorescence fluctuation spectroscopy techniques

Using fluorescence fluctuation spectroscopy techniques (FCS and N and B), we have quantified endogenous CTNNB1 concentrations and complexes in living cells for the first time, which provided novel and long-awaited biophysical parameters for computational modeling. Moreover, our approach has also yielded novel insights into CTNNB1 regulation that challenge current dogmas in the field. If we are correct, this has important consequences. First, if only part of the cytoplasmic CTNNB1 pool is present in an uncomplexed (i.e. free or monomeric) state, regardless of whether the WNT/CTNNB1 pathway is off or on (either via physiological WNT3A stimulation or via oncogenic activation), this is a rewrite of the textbook model. Second, if the slow-diffusing cytoplasmic CTNNB1 complex indeed represents a proteasome-associated destruction complex, this would fuel a debate that has remained unresolved for many years (Li et al., 2012; Verkaar et al., 2012). As more studies will use image-based techniques to determine biophysical properties of WNT/CTNNB1 signaling events (Ambrosi et al., 2020; Eckert et al., 2020), the field will undoubtedly learn how to interpret these findings.

While it is tempting to speculate about the implications of our findings, as with any technique, there are several limitations to consider. First of all, to deduce absolute concentrations several variables need to be considered, including autofluorescence, bleaching and dark states. Although the FCS data are corrected for autofluorescence and bleaching (see Equation 3 in FCS data acquisition and analysis section in the Materials and methods), this could potentially introduce some errors. It should also be noted that a small portion of SGFP2-CTNNB1 could be in a non-fluorescent state. Although our FCS analysis model already accounts for dynamic dark states such as the triplet state, non-matured fluorophores could lead to a slight underestimation of our concentrations. However, this is expected to be a very small fraction as SGFP2 has very good maturation kinetics (Kremers et al., 2007). Secondly, our findings concerning the diffusion kinetics are limited by the assumptions we make in the FCS fitting model. Although obvious mistakes in underlying assumptions immediately become clear due to bad fitting results and can therefore be excluded, not every wrong assumption will stand out accordingly. Our data clearly shows that assuming only one diffusion speed for CTNNB1 in HAP1 cells would be incorrect (Figure 4). However, whether with the second, slower diffusion speed we measure a single distinct, large complex, or rather an average of multiple different CTNNB1 containing complexes cannot be determined in our current set-up. Moreover, the measured diffusion coefficients do not reveal the identity of the complexes. Previous studies have shown that a significant pool of CTNNB1 is associated with destruction complex components in presence and absence of WNT signaling (Gerlach et al., 2014; Kitazawa et al., 2017; Li et al., 2012), and it is therefore likely that at least part of the slow fraction of CTNNB1 we measure does indeed represent this destruction complex bound pool. As we discuss above, association and dissociation of the destruction complex with the proteasome offers one potential explanation for the different diffusion coefficients measured in the cytoplasm in WNT ‘ON’ and ‘OFF’ conditions. However, in both conditions other processes, such as transient association with intracellular structures (e.g. vesicular membranes or cytoskeletal components), could contribute to the diffusion coefficients we observe.

In addition, we assume that CTNNB1 is present as a free-floating monomer (as fixed for our first component), based on previous observations (Gottardi and Gumbiner, 2004; Maher et al., 2010) and further supported by unbiased fitting of our data. However, at least one report suggests that CTNNB1 is not present as a monomer but rather in small cytoplasmic complexes of ~200 kDa (Gerlach et al., 2014). As diffusion speed is relatively insensitive to differences in size (e.g. an eightfold increase in protein mass is expected to result in only a twofold reduction of the diffusion coefficient for a spherical particle), it is possible that we do not measure truly free-floating CTNNB1, but rather one or more of these smaller complexes. In addition, point FCS is limited to a single position in the cell. Therefore, in addition to the intercellular differences in the WNT signaling response of individual cells, our measurements also sample intracellular heterogeneity caused by the presence of organelles and molecular crowding. Notwithstanding these limitations, we have been able to show that a large portion of CTNNB1 is present in a very large complex in both stimulated and unstimulated conditions and that this complex has a statistically and biologically significant different speed after WNT3A treatment and upon oncogenic mutation of CTNNB1.

The biophysical parameters we obtained from point FCS and N and B have taught us more about the speed and occupancy of the SGFP2-CTNNB1 complexes in living cells. Moreover, using different stimuli and perturbations of the pathway we have been able to link this to the phosphorylation status of CTNNB1. However, FCS and N and B do not provide conclusive evidence on the identity and composition of these complexes. An exciting possibility would be to label additional components presumed to be present in the CTNNB1-containing complexes at the endogenous level to uncover the precise composition and stoichiometry of protein complexes involved in WNT signaling. For instance, Fluorescence Cross Correlation Spectroscopy (FCCS) could be employed to test if two proteins reside within the same complex (Elson, 2011; Hink, 2014; Macháň and Wohland, 2014). Ultimately, a combination of such quantitative functional imaging techniques, biochemical and proteomic approaches, together with additional perturbations will need to be employed to further our understanding of the dynamic composition of endogenous CTNNB1 complexes, as well as to help us resolve the molecular mechanism underlying nucleocytoplasmic shuttling and nuclear retention. As both genome editing and live cell imaging techniques continue to improve, additional possibilities will open up to address longstanding questions in cellular signaling in a physiological context with high spatial and temporal resolution. New opportunities and challenges await as these investigations extend to 3D organoid cultures, developing embryos and living organisms.

Materials and methods

Key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Cell line (Homo sapiens)	HAP1	Whitehead Institute	Cellosaurus: CVCL_Y019	kind gift from Thijn Brummelkamp (NKI)
Cell line (Homo sapiens)	HAP1^SGFP2-CTNNB1	This paper
Cell line (Homo sapiens)	HAP1^{SGFP2-CTNNB1(S45F)}	This paper
Transfected construct (Homo sapiens)	pSpCas9(BB)−2A-Puro (PX459) V2.0	Ran et al., 2013	RRID:Addgene_62988
Transfected construct (Homo sapiens)	MegaTopflash	Hu et al., 2007		kind gift from Christophe Fuerer and Roel Nusse, Stanford University
Transfected construct (Homo sapiens)	CMV Renilla	Promega	E2261
Transfected construct (Homo sapiens)	pSGFP2-C1	Kremers et al., 2007	RRID:Addgene_22881	kind gift from Dorus Gadella
Transfected construct (Homo sapiens)	pmScarlet-i_C1	Bindels et al., 2017	RRID:Addgene_85044	kind gift from Dorus Gadella
Transfected construct (Homo sapiens)	pSYFP2-C1	Kremers et al., 2006	RRID:Addgene_22878	kind gift from Dorus Gadella
Transfected construct (Homo sapiens)	mTurquoise2-C1	Goedhart et al., 2012	RRID:Addgene_54842	kind gift from Dorus Gadella
Transfected construct (Homo sapiens)	pEGFP	Clontech
Transfected construct (Homo sapiens)	pEGFP₂	Pack et al., 2006		kind gift from Masataka Kinjo
Transfected construct (Homo sapiens)	pEGFP₃	Pack et al., 2006		kind gift from Masataka Kinjo
Transfected construct (Homo sapiens)	pBluescript II KS(+)	Stratagene
Transfected construct (Homo sapiens)	pX459-CTNNB1-ATG	This paper	RRID:Addgene_153429
Transfected construct (Homo sapiens)	pX459-CTNNB1-S45	This paper	RRID:Addgene_164587
Transfected construct (Homo sapiens)	pRepair-SGFP2-CTNNB1	This paper	RRID:Addgene_153430
Recombinant DNA reagent	pRepair-mScI-CTNNB1	This paper	RRID:Addgene_153431
Recombinant DNA reagent	pRepair-SYFP2-CTNNB1	This paper	RRID:Addgene_153432
Recombinant DNA reagent	pRepair-mTq2-CTNNB1	This paper	RRID:Addgene_153433
Chemical compound, drug	CHIR99021	Biovision	1677–5	6 mM stock in DMSO
Peptide, recombinant protein	Recombinant Mouse Wnt-3a	R and D systems	1324-WN-002	10 µg/ml stock solution in 0.1% BSA in PBS
Chemical compound, drug	Dapi	Invitrogen	D1306
Chemical compound, drug	Vybrant DyeCycle Violet Stain	Invitrogen	V35003
Chemical compound, drug	Vybrant DyeCycle Ruby Stain	Invitrogen	V10273
Antibody	Non-phosphorylated (Active) β-catenin clone D13A1 (Rabbit monoclonal)	Cell Signaling	8814S RRID:AB_11127203	WB (1:1000)
Antibody	Total β-catenin clone 14 (mouse monoclonal)	BD	610153 RRID:AB_397554	WB (1:2000)
Antibody	α-Tubulin clone DM1A (mouse monoclonal)	Sigma-Aldrich	T9026 RRID:AB_477593	WB (1:1000)
Antibody	GFP Antibody (Rabbit polyclonal)	Invitrogen	A-6455 RRID:AB_221570	WB (1:1000)
Antibody	IRDye 680LT Goat anti-Rabbit IgG	LI-COR	926–68021 RRID:AB_10706309	WB (1:20,000)
Antibody	IRDye 800CW Donkey anti-Mouse IgG	LI-COR	926–32212 RRID:AB_621847	WB (1:20,000)
Chemical compound, drug	SiR-DNA	Spirochrome	SC007
Chemical compound, drug	Alexa Fluor 488 NHS Ester	Molecular probes	A20000
Software, algorithm	FlowJo
Software, algorithm	CellProfiler pipeline	This paper		available at https://osf.io/6pmwf/
Software, algorithm	FIJI/ImageJ
Software, algorithm	FFS Dataprocessor version 2.3	SSTC
Software, algorithm	ptu converter	Crosby et al., 2013
Software, algorithm	ImageJ macro script	modified from Crosby et al., 2013		available at https://osf.io/ys5qw/
Software, algorithm	PlotsOfDifferences	Goedhart, 2019		https://huygens.science.uva.nl/PlotsOfDifferences/
Software, algorithm	RStudio
Software, algorithm	R script	This paper		available at https://osf.io/sxakf/
Software, algorithm	R shiny app	This paper	WNT_minimal_model_v2.4.R	app available at https://wntlab.shinyapps.io/WNT_minimal_model/, source script available at https://osf.io/27ya6/

DNA constructs

Request a detailed protocol

The following constructs were used: pSpCas9(BB)−2A-Puro (PX459) V2.0 (Ran et al., 2013, a kind gift from Feng Zhang, available from Addgene, plasmid #62988), MegaTopflash (Hu et al., 2007, a kind gift from Christophe Fuerer and Roel Nusse, Stanford University), CMV Renilla (E2261, Promega, Madison, WI), pSGFP2-C1 (Kremers et al., 2007, a kind gift from Dorus Gadella, available from Addgene, plasmid #22881), pmScarlet-i_C1 (Bindels et al., 2017, a kind gift from Dorus Gadella, available from Addgene, plasmid #85044), pSYFP2-C1 (Kremers et al., 2006, a kind gift from Dorus Gadella, available from Addgene, plasmid #22878), mTurquoise2-C1 (Goedhart et al., 2012, a kind gift from Dorus Gadella, available from Addgene, plasmid #54842), pEGFP (Clontech, Mountain View, CA), pEGFP₂ and pEGFP₃ (Pack et al., 2006, a kind gift from Masataka Kinjo), and pBluescript II KS(+) (Stratagene, La Jolla, CA).

The gRNA targeting the start codon in exon2 of human CTNNB1 was designed using the MIT webtool (crispr.mit.edu) and cloned into pX459. Oligos RVA567 and RVA568 (Table 6), encoding the gRNA, were annealed and ligated into BbsI-digested pX459 plasmid as previously described (Ran et al., 2013) to obtain pX459-CTNNB1-ATG. The gRNA targeting codon 3 of CTNNB1 for mutagenesis of Serine 45 to Phenylalanine (S45F) was similarly designed and cloned by introducing RVA561 and RVA562 (Table 6) into pX459, yielding pX459-CTNNB1-S45.

Table 6

primers/oligonucleotides used in this study.

primer	sequence
RVA24	CAAGTTTGTTGTAGGATATGCCC
RVA25	CGATGTCAATAGGACTCCAGA
RVA124	AGTGTGAGGTCCACGGAAA
RVA125	CCGTCATGGACATGGAAT
RVA555	GCCAAACGCTGGACATTAGT
RVA558	AGACCATGAGGTCTGCGTTT
RVA561	CACCGTTGCCTTTACCACTCAGAGA
RVA562	AAACTCTCTGAGTGGTAAAGGCAAC
RVA567	CACCGTGAGTAGCCATTGTCCACGC
RVA568	AAACGCGTGGACAATGGCTACTCAC
RVA581	tgctcaccatggtggGATTTTCAAAACAGTTGTATGGTATACTTC
RVA582	actgttttgaaaatcCCACCATGGTGAGCAAGGGC
RVA583	agtagccattgtccaCTTGTACAGCTCGTCCATGCCG
RVA584	gacgagctgtacaagTGGACAATGGCTACTCAAGGTTTG
RVA618	atacgactcactatagggcgaattggagctGATGCAGTTTTTTTCAATATTGC
RVA619	ttctagagcggccgccaccgcggtggagctCTCTCTTTTCTTCACCACAACATTTTATTTAAAC
RVA622	AAGAGAGAGCTCCACCGCGGTGGCGGCCG
RVA623	TGCATCAGCTCCAATTCGCCCTATAGTGAGTCG
RVA1616	tgtccacgctgGATTTTCAAAACAGTTGTATGG
RVA1617	atacaactgttttgaaaatccagcgtggacaATGGTGAGCAAGGGCGAG
RVA1618	cacaaaccttgagtagccatCTTGTACAGCTCGTCCATGC
RVA1619	ATGGCTACTCAAGGTTTGTGTCATTAAATC
RVA2540	CTTACCTGGACTCTGGAATCCATTCTGGTGCCACTACCACAGCTCCTTTCCTGTCCGGTAAAGGCAATCCTGAGGAAGAGGATGTGGATACCTCCCAAGT

The repair plasmid for SGFP2-CTNNB1 (pRepair-SGFP2-CTNNB1) was cloned using Gibson cloning (Gibson et al., 2009). SGFP2 was chosen for its favorable brightness, maturation and photo-stability (Kremers et al., 2007). First, a repair plasmid including the Kozak sequence from the pSGFP2-C1 plasmid was generated (pRepair-Kozak-SGFP2 -CTNNB1). For this, 5’ and 3’ homology arms were PCR amplified from genomic HEK293A DNA with primers RVA618 and RVA581 (5’ arm) or RVA619 and RVA584 (3’ arm). SGFP2 was amplified with Gibson cloning from pSGFP2-C1 with primers RVA582 and RVA583 and the backbone was amplified from SacI digested pBlueScript KS(+) with primers RVA622 and RVA623. The final repair construct (pRepair-SGFP2-CTNNB1) contains the endogenous CTNNB1 Kozak sequence before the SGFP2 ATG. To obtain pRepair-SGFP2-CTNNB1, the backbone and homology regions were amplified from pRepair-SGFP2-Kozak-CTNNB1 with primers RVA1616 and RVA1619, and an SGFP2 without the Kozak sequence was amplified from pSGFP2-C1 with primers RVA1617 and RVA1618. To generate color variants of the repair plasmid SYFP2, mScarlet-i and mTurquoise2 (not used in this publication, but available from Addgene, see below) were also amplified from their respective C1 vectors with primers RVA1617 and RVA1618. PCR products were purified and assembled with a Gibson assembly master mix with a 1:3 (vector:insert) molar ratio. Gibson assembly master mix was either purchased (E2611S, NEB) or homemade (final concentrations: 1x ISO buffer (100 mM Tris-HCL pH 7.5, 10 mM MgCl2, 0.2M dNTPs (R0181, Thermo Scientific), 10 mM DTT (10792782, Fisher), 5% PEG-8000 (1546605, Sigma-Aldrich, St Louis, MO), 1 mM NAD+ (B9007S, NEB)), 0.004 U/µl T5 exonuclease (M0363S, NEB), 0.5 U/µl Phusion DNA Polymerase (F-530L, Thermo Scientific) and 4 U/µl Taq DNA ligase (M0208S, NEB)).

The following plasmids are available from Addgene: pX459-CTNNB1-ATG (#153429), pX459-CTNNB1-S45 (#164587), pRepair-SGFP2-CTNNB1 (#153430), pRepair-mScI-CTNNB1 (#153431), pRepair-SYFP2-CTNNB1 (#153432), pRepair-mTq2-CTNNB1 (#153433).

Cell culture, treatment, and transfection

Request a detailed protocol

HAP1 cells (a kind gift from Thijn Brummelkamp (NKI), acknowledging the Whitehead Institute as the source of the material) were maintained in full medium (colorless IMDM (21056023, Gibco, Thermo Fisher Scientific, Waltham, MA) supplemented with 10% FBS (10270106, Gibco) and 1X Glutamax (35050061, Gibco)) under 5% CO₂ at 37°C in humidifying conditions and passaged every 2–3 days using 0.25% Trypsin-EDTA (25200056, Gibco). Cells were routinely tested for mycoplasma. We verified the haploid identity of the parental HAP1^WT by karyotyping of metaphase spreads. To maintain a haploid population, cells were resorted frequently (see below) and experiments were performed with low passage number cells.

Where indicated, cells were treated with CHIR99021 (6 mM stock solution in DMSO) (1677–5, Biovision, Milpitas, CA) or Recombinant Mouse Wnt-3a (10 µg/ml stock solution in 0.1% BSA in PBS) (1324-WN-002, R and D systems, Bio-Techne, Minneapolis, MN) with DMSO and 0.1% BSA in PBS as vehicle controls, respectively. In Figure 3, the range of WNT3A used was based on previous experiments in HEK293T cells (Jacobsen et al., 2016).

Cells were transfected using Turbofect (R0531, ThermoFisher, Thermo Fisher Scientific, Waltham, MA), X-tremeGene HP (6366546001, Roche, Basel, Switzerland), or Lipofectamine 3000 (L3000001, Invitrogen, Thermo Fisher Scientific, Waltham, MA) in Opti-MEM (31985070, Gibco) according to the manufacturer’s instructions.

HAP1^SGFP2-CTNNB1 and HAP1^{SGFP2-CTNNB1(S45F)} generation

Request a detailed protocol

800,000 HAP1 cells/well were plated on six-well plates. The following day, cells were transfected with Turbofect and 2000 ng DNA. pX459-CTNNB1-ATG and pRepair-SGFP2-CTNNB1 were transfected in a 2:1, 1:1, or 1:2 ratio. pSGFP2-C1, pX459, or pX459-CTNNB1-ATG were used as controls. From 24 to 48 hr after transfection, cells were selected with 0.75 µg/ml puromycin (A1113803, Gibco). Next, cells were expanded and passaged as needed until FACS sorting at day 9. For FACS analysis and sorting, cells were washed, trypsinized, resuspended with full medium and spun down at 1000 rpm for 4 min. For sorting, cells were stained with 1 µg/ml Dapi (D1306, Invitrogen) in HF (2% FBS in HBSS (14175053, Gibco)), washed with HF and resuspended in HF. To determine the haploid population, a separate sample of cells was stained with 5 μM Vybrant DyeCycleTM Violet Stain (V35003, Invitrogen) in full medium for 30 min at 37°C and kept in vibrant containing medium. Cells were filtered with a 70 µm filter and then used for FACS sorting and analysis on a FACSARIA3 (BD, Franklin Lanes, NJ). Vybrant-stained cells were analyzed at 37° and used to set a size gate only containing haploid cells. Dapi-stained cells were single cell sorted at 4°C into 96-well plates, that were previously coated overnight with 0.1% gelatin (G9391, Sigma-Aldrich) in MQ and contained full medium supplemented with 1% penicillin/streptomycin (15140122, Gibco) and 0.025 M HEPES (H3375 Sigma-Aldrich, 1 M stock solution, pH 7.4, filter sterilized). The three independent clones used in this study were obtained from separate transfections of the same parental cell line. Clones were genotyped and sanger sequenced using primers RVA555 and RVA558 (Table 6).

HAP1^{SGFP2-CTNNB1(S45F)} were generated from HAP1^SGFP2-CTNNB1 clone 1. The same procedure as above was followed with slight adaptations; Cells were transfected 1000 ng pX459-CTNNB1-S45 or pX459 with 2 or 4 µl 10 mM repair oligo (RVA 2540) with Turbofect, selected with puromycin and expanded as described above. Haploid single cells were sorted after 11 days as described above. For haploid size discrimination Vybrant DyeCycle Ruby Stain (V10273, Invitrogen) was used. The five clones used in this study were obtained from two separate transfection (clones 2, 3, 16, 24 from the same transfection, clone 27 from a second transfection). Clones were genotyped using primers RVA555 and RVA558 (Table 6), followed by HpaII (ER0511, ThermoFisher) restriction as per the manufacturer’s instruction. RVA555 was used for sanger sequencing.

Resorting of the cell lines was also performed with the same FACS procedure, with collection of cells in 15 mL tubes containing full medium with 1% penicillin and 0.025 M HEPES.

FACS data were analyzed and visualized with FlowJo.

Luciferase assay

Request a detailed protocol

For luciferase assays, 100,000 cells per well were seeded on a 24-well plate. Cells were transfected with 1 µl X-tremeGene HP and 400 ng MegaTopflash reporter and 100 ng CMV-Renilla, or 500 ng SGFP2-C1 as a negative control 24 hr later. Cells were treated with the indicated concentration of CHIR99021 24 hr after transfection and after another 24 hr medium was removed and the cells were harvested with 50 µl Passive Lysis Buffer (E1941, Promega). Luciferase activity was measured on a GloMax Navigator (Promega), using 10 µl lysate in a black OptiPlate 96-well plate (6005279, Perkin Elmer, Waltham, MA) and 50 μL homemade firefly and luciferase reagents (according to Fuerer et al., 2014; Hampf and Gossen, 2006).

For luciferase assays, three technical replicates (i.e. three wells transfected with the same transfection master mix) were pipetted and measured for each sample in each experiment. For each technical triplicate, the average MegaTopflash activity was calculated and depicted as a single dot in Figure 2C and Figure 7—figure supplement 1G. Three independent biological experiments, each thus depicted as an individual dot, were performed. To calculate MegaTopflash activity, Renilla and Luciferase luminescence values were corrected by subtracting the average background measured in the SGFP2-transfected control. MegaTopflash activity was calculated as the ratio of corrected Firefly and Renilla luminescence and normalized to the average reporter activity of the relative DMSO control (Figure 2C) or WT DMSO control (Figure 7—figure supplement 1G).

Western blot

Request a detailed protocol

The remaining lysates from the technical triplicates of the luciferase assay were combined and they were cleared by centrifugation for 10 min at 12,000 g at 4°C. Western blot analysis was performed and quantified as previously described (Jacobsen et al., 2016). Antibodies were used with the following dilutions, 1:1000 Non-phosphorylated (Active) β-catenin clone D13A1 (8814S, Cell Signaling, Danvers, MA), 1:2000 total β-catenin clone 14 (610153, BD), 1:1000 α-Tubulin clone DM1A (T9026, Sigma-Aldrich), 1:1000 GFP polyclonal (A-6455, Invitrogen), 1:20,000 IRDye 680LT Goat anti-Rabbit IgG (926–68021, LI-COR, Lincoln, NE), 1:20,000 IRDye 800CW Donkey anti-Mouse IgG (926–32212, LI-COR). Raw data for all blots have been made available at https://osf.io/vkexg/.

qRT-PCR

Request a detailed protocol

For qRT-PCR analysis, 100,000 HAP1 cells per well were seeded on a 24-well plate. After 48 hr, cells were treated with indicated concentrations of CHIR99021. Cells were harvested 24 hr after treatment. RNA was isolated with Trizol (15596018, Invitrogen) according to the manufacturer’s instructions. cDNA was synthesized using SuperScriptIV (18090010, Invitrogen) according to the manufacturer’s instructions. qRT-PCR was performed with SyberGreen (10710004, Invitrogen). The endogenous WNT target gene AXIN2 was amplified using primers RVA124 and RVA125, and HPRT housekeeping control was amplified using primers RVA24 and RVA25.

For qRT-PCR experiments, three technical replicates (i.e. three reactions with the same cDNA) were pipetted and measured for each sample in each experiment. For each technical triplicate, the mean fold-change in AXIN2 expression was calculated and depicted as a single dot in Figure 2D and Figure 7—figure supplement 1H. Three independent biological experiments, each thus depicted as an individual dot, were performed. Relative expression levels of AXIN2 were calculated using the comparative Delta-Ct method (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008). Briefly, AXIN2 expression was normalized for HPRT expression and then the relative fold-change to a WT DMSO sample was calculated for all clones and conditions.

Time-lapse imaging

Request a detailed protocol

The day before imaging, 88,000 cells/well were seeded on an eight-well chamber slide with glass bottom (80827–90, Ibidi, Gräfelfing, Germany). HAP1^SGFP2-CTNNB1 clone 2 was used for the main Figure 3, all three clones were used for Figure 3—figure supplement 2. HAP1^{SGFP2-CTNNB1(S45F)} clone 2 was imaged for Figure 8A. Approximately 6 hr before imaging, medium was replaced with full medium supplemented with 1% penicillin/streptomycin, 0.025M HEPES and 500 nM SiR-DNA (SC007, Spirochrome, Stein am Rhein, Switzerland). Time lapse experiments were performed on an SP8 confocal microscope (Leica Microsystems, Wetzlar, Germany) at 37°C with a HC PL APO CS2 63x/1.40 oil objective (15506350, Leica), 488 and 633 lasers, appropriate AOBS settings, using HyD detectors for fluorescent signal with a 496–555 bandpass for SGFP2-CTNNB1 and 643–764 bandpass for SiR-DNA, and a transmission PMT. Using multi-position acquisition, up to 24 images were captured every 5 min. Focus was maintained using AFC autofocus control on demand for every time point and position. Automated cell segmentation and intensity quantification was performed using a custom CellProfiler pipeline (made available at https://osf.io/6pmwf/). Output data was further analyzed in R/RStudio. Cells with a segmented cytoplasmic area of less than 10 pixels were excluded. Intensities were normalized per position to the average intensity in the cellular compartment (nucleus or cytoplasm) for that position before the addition of the compounds. The imaging settings resulted in low signal in regions not occupied by cells (~10% of the nuclear intensity, and ~5% of the cytoplasmic intensity in untreated cells), and the data was therefore not background corrected. The nuclear cytoplasmic ratio was calculated by dividing the raw nuclear intensity by the raw cytoplasmic intensity. Videos and still images were extracted with FIJI/ImageJ.

FCS and N and B cell preparation and general settings

Request a detailed protocol

Two days before FCS and N and B experiments, 44,000 cells/well were seeded on an eight-well chamber slide with a glass bottom (80827–90, Ibidi). For low, FFS-compatible expression of control samples, HAP1^WT cells were transfected with ~5 ng pSGFP2-C1, pEGFP (monomer), pEGFP₂ (dimer), or pEGFP₃ (trimer) and ~200 ng pBlueScript KS(+) per well with Turbofect, X-tremeGene HP or Lipofectamine 3000 the day before the experiment. Lipofectamine 3000 yielded the best transfection efficiency. For Figures 4, 5 and 8K and accompanying supplements, HAP1^SGFP2-CTNNB1 clone 2 was used. For Figures 7 and 8 and accompanying supplements, CHIR99021 data was recorded and pooled for all three HAP1^SGFP2-CTNNB1 clones, and S45F data was recorded and pooled from HAP1^{SGFP2-CTNNB1(S45F)} clones 2, 24, and 27 and HAP1^SGFP2-CTNNB1 clone 1 (the parental line for these S45F mutant clones) was used as the wild-type control.

FCS and N and B measurements were performed on an Olympus FV-1000 equipped with SepiaII and PicoHarp 300 modules (Picoquant, Berlin, Germany) at room temperature. An Olympus 60x water immersed UPLS Apochromat (N.A. 1.2) objective was used for FCS acquisition and Figure 3—figure supplement 2E, and an Olympus 60x silicon immersed UPLS Apochromat (N.A. 1.4) objective was used for N and B measurements. Green fluorophores were excited with a 488 nm diode laser (Picoquant) pulsing at 20 MHz and detected through a 405/480-488/560/635 nm dichroic mirror (Chroma, Bellows Falls, VT) and 525df45 nm bandpass filter (Semrock, Rochester, NY) with an Avalanche Photodiode (APD) (MPD, Bolzano, Italy). For Figure 2—figure supplement 1E and for FCS and N and B reference images, the same laser and dichroic were used, but the signal was detected through a 505–540 bandpass filter with an internal PMT of the FV-1000 Olympus microscope.

FCS data acquisition and analysis

Request a detailed protocol

For FCS measurements, a confocal image was recorded. In this reference image, a single pixel was set as region of interest (ROI), based on its localization in the cytoplasm or nucleus as judged by the transmission image. In this ROI, the fluorescence intensity was measured over time using an APD, for typically 120 s.

FCS measurements were analyzed in FFS Dataprocessor version 2.3 (SSTC, Minsk, Belarus). The autocorrelation curve ( $G ()$ ) was calculated from the measured intensity (I) according to Equation 1. Intensity traces with significant photobleaching, cell movement or focal drift were excluded from further analysis (see Supplementary file 1 – tab FCS measurements and fitting). From other traces, a portion of the trace with minimal (less than 10%) intensity drift or bleaching was selected to generate autocorrelation curve (AC).

G () = 1 + \frac{< δ I (t) * δ I (t + τ) >}{< I >^{2}}

The resulting AC was fitted with a Triplet-state-diffusion model, described in Equation 2. G_∞ accounts for offset in the AC for example by intensity drift. N is the average of the number of particles that reside in the confocal volume. F_trip and τ_trip describe the fraction of molecules in the dark state and the relaxation of this dark state respectively. Of note, in this case, F_trip and τ_trip account both for blinking of the fluorescent molecules and for the afterpulsing artefact of the APD. τ_diff,i describes the diffusion rate of the fluorescent molecules with the corresponding fraction, F_i. This diffusion time depends on the structural parameter (sp), which is defined as the ratio of the axial (ω_z) over the radial axis (ω_xy) of the observation volume.

G (τ) = G_{\infty} + \frac{1}{< N >} * \frac{F_{t r i p}}{1 - F_{t r i p}} e^{\frac{- τ}{τ_{t r i p}}} * \sum_{j} \frac{F_{i}}{(1 + \frac{τ}{τ_{d i f f, i}}) \sqrt{1 + \frac{τ}{τ_{d i f f, i} * s p^{2}}}}

The apparent particle numbers (N_apa) for SGFP2-CTNNB1 were corrected for autofluorescence and bleaching (Equation 3). The autofluorescence (I_{autofluorescence}) of HAP1 cells in the nucleus and cytoplasm was measured in untransfected HAP1 cells using the same settings as for FCS measurements. The correction for moderate bleaching is based on the intensity of the selected portion of the intensity trace for AC calculation (I_ana) and the intensity at the start of the measurement (I_start).

The size and shape of the observation volume was calibrated daily by measuring Alexa Fluor 488 NHS Ester (A20000, Molecular probes, Thermo Scientific, stock dilution in MQ) in PBS in a black glass-bottom cell imaging plate with 96 wells (0030741030, Eppendorf, Hamburg, Germany). From the FCS measurements of Alexa488, the $τ_{d i f f}$ and sp were determined by fitting with a single diffusion and blinking component. The diffusion coefficient (D) of Alexa488 in aqueous solutions at 22.5°C is 435 μm²s⁻¹ (Petrásek and Schwille, 2008). From these parameters, the axial diameter can be determined with Equation 4 and the volume can be approximated by a cylinder (Equation 5). This allows for transformation of particle numbers to concentrations (Equation 6) and diffusion times to diffusion coefficients (Equation 4) that are independent of measurement settings and small daily changes in alignment of the microscope.

{N_{c o r r} = N_{a p a} * [1 - \frac{I_{a u t o f l u o r e s c e n c e}}{I_{t o t a l}}]}^{2} * [\frac{I_{s t a r t}}{I_{a n a}}]

τ_{d i f f} = \frac{{ω_{x y}}^{2}}{4 D}

V = 2 π {ω_{x y}}^{3} * s p

C = \frac{N_{c o r r}}{V * N_{A}}

The model to fit SGFP2-CTNNB1 measurements contained two diffusion components. The first diffusion component was fixed to the speed of monomeric SGFP2-CTNNB1. To estimate the speed of monomeric SGFP2-CTNNB1, the speed of free floating SGFP2, transfected in HAP1 cells, was measured to be 24.1 µm²s⁻¹ using FCS. Subsequently, this speed was used to calculate the speed of monomeric SGFP2-CTNNB1 with Einstein-Stokes formula (Equation 7).

D = \frac{k_{B} T}{6 π η r}

As the temperature (T), dynamic viscosity (η) and Boltzmann’s constant (k_B) are equal between SGFP2 and SGFP2-CTNNB1 measurements, the expected difference in diffusion speed is only caused by the radius (r) of the diffusing molecule assuming a spherical protein. The difference in radius was approximated by the cubic root of the ratio of the molecular weight of the SGFP2-CTNNB1 fusion protein (88 + 27=115 kDa) and the size of the SGFP2 protein (27 kDa), thus expecting a 1.62 times lower diffusion coefficient (compared to free floating SGFP2) of 14.9 µm²s⁻¹ for SGFP2-CTNNB1.

It must be noted that, especially for larger protein complexes, the linearity between the radius of the protein and the speed is not ensured, if the shape is not globular, and due to other factors such as molecular crowding in the cell and hindrance from the cytoskeletal network. We therefore did not estimate the size of the measured CTNNB1 complexes (as this would inescapably introduce errors, given that the ideal circumstances underlying the Einstein-Stokes formula are not met in the cellular environment for a complex of this size), but rather compared them to measurements from other FCS studies. However, it is likely that the 3.5-fold change in the second diffusion coefficient of SGFP2-CTNNB1 in response to WNT3A treatment is indicative of a larger than 3.5-change to complex size.

In the fitting model, the structural parameter was fixed to the one determined by the Alexa488 measurements of that day. To ensure good fitting, limits were set for other parameters; G_∞[0.5–1.5], N [0.001, 500], τ_trip [1*10⁻⁶-0.05 ms], τ_diff2[10–150 ms]. This model was able to fit most Autocorrelation Curves from FCS measurements. In case of clear misfits, as judged by the distribution of residuals around the fitted curves, the measurement was excluded (see Supplementary file 1 – tab FCS measurements and fitting).

To validate the obtained first diffusion coefficient of 14.9 µm²s⁻¹ for SGFP2-CTNNB1, the data were tested with an unfixed two-component model where both the first and the second diffusion coefficient were fitted (shown in Figure 4—figure supplement 1). The following limits were set; G_∞[0.5–1.5], N [0.001, 500], τ_trip [1*10⁻⁶-0.05 ms], τ_diff1[0.5–10 ms], τ_diff2[10–150 ms]. This resulted in a median diffusion time for the first component of 14.8 µm²s⁻¹ (Figure 4—figure supplement 1), which was in line with our calculated diffusion coefficient of 14.9 µm²s⁻¹. All analyses were performed with the two-component model with the fixed first component to reduce variability.

N and B data acquisition and analysis

Request a detailed protocol

As a control, and to optimize acquisition settings, HAP1 cells transfected with SGFP2, EGFP monomer, dimer or trimer were measured alongside HAP1^SGFP2-CTNNB1 cells treated with BSA, WNT3A, DMSO or CHIR99021, or HAP1^{SGFP2-CTNNB1(S45F)} cells. SGFP2 and EGFP are highly similar in sequence (with only four amino acid changes) and in spectral and biochemical characteristics. SGFP2 has a slightly higher quantum yield (+7%), lower extinction coefficient (−27%), and enhanced protein expression and maturation compared to EGFP (Kremers et al., 2007). The resulting brightness of monomeric SGFP2 in comparison to monomeric EGFP is slightly lower (−5%). In cellular measurements, this difference is within the biological and technical variation and therefore SGFP2 and EGFP controls can be considered synonymous in these experiments.

For N and B analysis, 50 images were acquired per measurement with a pixel time of 100 µs/pixel and a pixel size of 0.138–0.207 µm. The fluorescent signal was acquired with the APD described above for the FCS measurements. APD readout was converted to a TIF stack using a custom build .ptu converter (Crosby et al., 2013). This TIF stack was further analyzed using an ImageJ macro script (modified from Crosby et al., 2013, made available at https://osf.io/ys5qw/) based on Digman et al., 2008. Within the script, average brightness and particle numbers were calculated for nuclear or cytoplasmic ROIs, which were set based on transmission image (see Figure 5F). Static or slow-moving particles, including membrane regions, were excluded by thresholding and/or ROI selection, since they can severely impact the brightness measured.

Data were further analyzed in R/RStudio. Brightness was normalized to the median value of the EGFP-monomer brightness measured on the same day in the same cellular compartment (nucleus/cytoplasm). Of note, FCS and N and B analysis models assume a different confocal volume. In FCS we assume a cylinder with factor γ=1, whereas in N and B we assume a 3D-Gauss with factor γ=0.3536. To be able to compare particle numbers obtained with both techniques, particle numbers obtained with N and B were divided by the factor γ=0.3536.

Data representation and statistical analysis

Request a detailed protocol

Data processing and representation were performed in RStudio (version 1.1.456 running R 3.5.1 or 3.6.1). 95% confidence intervals of the median mentioned in the text and shown in Tables 1–2 and Supplementary file 1 were calculated using PlotsOfDifferences (Goedhart, 2019). The p-values in Table 3 and Supplementary file 1 were also calculated using PlotsOfDifferences, which uses a randomization test and makes no assumption about the distribution of the data. Representation of the imaging data in Figure 3—figure supplement 1 and in Videos 4-6 were generated in RStudio using a script based on PlotsOfDifferences (made available at https://osf.io/sxakf/).

Model description

We developed a minimal model for WNT signaling based on a previous model from the Kirschner group (Lee et al., 2003). The model is available as an interactive app at and the R https://wntlab.shinyapps.io/WNT_minimal_model/ source code of the model is available at https://osf.io/27ya6/ (WNT_minimal_model_v2.4.R).

Our minimal model comprises the following reactions:

C B + D C \underset{k 1}{\overset{k 2}{⇋}} C B^{*} - D C B i n d i n g o f c y t o p l a s m i c C T N N B 1 (C B) t o d e s t r u c t i o n c o m p l e x

C B^{*} - D C \overset{k 3}{\to} D C + C B^{*} R e l e a s e o f p h o s p h o r y l a t e d C B (C B *) a n d r e c y c l i n g o f t h e d e s t r u c t i o n c o m p l e x

D V L + D C \underset{k 4}{\overset{k 5}{⇋}} D C^{*} I n a c t i v a t i o n o f t h e d e s t r u c t i o n c o m p l e x b y D V L

C B \underset{k 6}{\overset{k 7}{⇋}} N B N u c l e o c y t o p l a s m i c s h u t t l i n g o f C B t o a n d f r o m t h e n u c l e u s

N B + T C F \underset{k 8}{\overset{k 9}{⇋}} N B - T C F B i n d i n g o f N B t o T C F

Below, we show the differential equations that govern the concentrations of the different compounds over time for the reactions described above. Table 4 in the main text gives the correspondence between the variables (i.e. $x_{1}$ ) in the differential equations and the model name (i.e. CB) in the reactions. The parameter $w$ in Equation 7a and Equation 9 is $w = 0$ in the absence of WNT and $w = 1$ if WNT is present, that is in our minimal model the inactive form of the destruction complex (DC*) is only present if WNT is present. The parameter $b$ in Equation 6a represents the constant production of CTNNB1, corresponding to $v_{12}$ in Lee et al., 2003.

\frac{d x_{1}}{d t} = - k_{1} x_{1} x_{2} + k_{2} x_{3} - k_{6} x_{1} + k_{7} x_{5} + b

\frac{d x_{2}}{d t} = - k_{1} x_{1} x_{2} + (k_{2} + k_{3}) x_{3} - w (k_{4} x_{2} - k_{5} x_{4})

\frac{d x_{3}}{d t} = k_{1} x_{1} x_{2} - (k_{2} + k_{3}) x_{3}

\frac{d x_{4}}{d t} = w (k_{4} x_{2} - k_{5} x_{4})

\frac{d x_{5}}{d t} = k_{6} x_{1} - k_{7} x_{5} - k_{8} x_{5} x_{6} + k_{9} x_{7}

\frac{d x_{6}}{d t} = - k_{8} x_{5} x_{6} + k_{9} x_{7}

\frac{d x_{7}}{d t} = k_{8} x_{5} x_{6} - k_{9} x_{7}

Equilibrium conditions without WNT

Request a detailed protocol

The parameters in our model can in part be determined from our measurements of the equilibrium concentrations of CB, NB and their complexes, see Tables 4–5 in the main text. Where we could not determine the parameters from our measurements, we used published values as indicated.

Under equilibrium conditions, the concentrations of the compounds do not change with time and the left-hand side of Equations 6a - 12 is zero. From Equations 10 and 11, we can determine the ratio of the rate constants $k_{6}$ and $k_{7}$ from the measured values of $x_{1}$ and $x_{5}$ :

k_{6} x_{1} = k_{7} x_{5} \Leftrightarrow \frac{k_{6}}{k_{7}} = \frac{x_{5}}{x_{1}} = \frac{87}{91} = 0.96

From Equations (6a), (8), (10) and (11) we have:

- k_{3} x_{3} + b = 0 \Leftrightarrow k_{3} = \frac{b}{x_{3}} = \frac{0.423}{62.5} = 0.0068 m i n^{- 1}

Our reaction (1) corresponds closely to step 8 in Lee et al. therefore, we use the value of the dissociation constant $K_{8} = 120$ nM from Lee et al. for our dissociation constant $K_{1} = \frac{k_{2}}{k_{1}}$ .

The concentration of the destruction complex is obtained from Equation (1a) under equilibrium conditions using Equations (6a), (8), (10), (11) and (14)

- k_{1} x_{1} x_{2} + k_{2} x_{3} + b = 0

The value of $b$ is assumed to be small compared to the two other terms, so we calculate the concentration of the destruction complex as:

$x_{2} = K_{1} \frac{x_{3}}{x_{1}} = 120 \frac{62.5}{91} = 82.4 n M$ . It was then verified in our interactive app that this value for the destruction complex is indeed consistent with the equilibrium conditions without WNT stimulation.

To calculate the dissociation constant for the NB-TCF complex, we estimate an equilibrium concentration for free TCF ( $x_{6}$ ) from Tan et al., 2012. From their Figure 11, it is seen that the bound TCF concentration in equilibrium in the presence of WNT has about the same value as the initial free TCF concentration and that no initial bound TCF is present. However, we measured NB-TCF also in the initial state. Therefore, we consider the free TCF concentration value from Tan et al. as a lower bound for the estimate of total TCF. Also, from Figure 11 of Tan et al., 2012, we estimate that of the initial free TCF, a fifth remains in the nucleus as free TCF after WNT is turned on. We measured 86 nM NB-TCF in the nucleus after the application of WNT. This leads to an estimate of the total concentration of TCF, TCF⁰, in the nucleus of: $[T C F^{0}] = 86 + 0.2 \times 86 = 103 n M$ . If we assume that the total TCF concentration does not change by the application of WNT, we calculate the dissociation constant of the NB-TCF complex from Equation 12:

k_{8} x_{5} (T C F^{0} - x_{7}) = k_{9} x_{7} \Rightarrow \frac{k_{9}}{k_{8}} = K_{2} = \frac{x_{5} (T C F^{0} - x_{7})}{x_{7}} = \frac{87 * 81}{22} = 320 n M

Equilibrium conditions with WNT

Request a detailed protocol

We model the action of WNT by deactivation of the destruction complex by DVL through reaction 3 by setting $w = 1$ in Equations 7a and 9. The dissociation constant of CB*-DC, $K_{1}$ , is assumed not to change in the presence of WNT. The measurements of free CB and NB in equilibrium (see Table 2) give for the ratio of $k_{6}$ and $k_{7}$ :

k_{6} x_{1} = k_{7} x_{5} \Leftrightarrow \frac{k_{6}}{k_{7}} = \frac{x_{5}}{x_{1}} = \frac{170}{145} = 1.17

The value of the rate of decay of the phosphorylated complex CB*-DC, $k_{3}$ , is found to be the same for the 'without WNT' situation:

- k_{3} x_{3} + b = 0 \Leftrightarrow k_{3} = \frac{b}{x_{3}} = \frac{0.423}{62.5} = 0.0068 m i n^{- 1}

To uniquely determine the ratio of $k_{4}$ and $k_{5}$ , we need the concentrations of the destruction complex DC and DC* neither of which we have access to. We can, however, fit this ratio with our model to the measured values of $x_{1}$ and $x_{7}$ and find $k_{4} / k_{5} = 1.7$ .

We again calculate the dissociation constant of the NB-TCF complex from Equation (12), using the concentrations for NB and NB-TCF obtained with FCS.

k_{8} x_{5} (T C F^{0} - x_{7}) = k_{9} x_{7} \Rightarrow \frac{k_{9}}{k_{8}} = K_{2} = \frac{x_{5} (T C F^{0} - x_{7})}{x_{7}} = \frac{170 * 17}{86} = 33.6 n M

Notice that we determined the ratios of the rate constants from the measured equilibrium values of free and bound CTNNB1 in the cytoplasm and the nucleus. This means that our rate constants are determined up to a multiplicative factor: the equilibrium equations do not change if all rate constants $k_{i}$ and the parameter $b$ are multiplied by the same factor, $R a t e$ . The factor $R a t e$ determines how fast our model system reaches equilibrium. By comparing the times equilibrium was reached by the cytoplasmic and nuclear SGFP2-CTNNB1 signals (Figure 3B, C) of about 4.5 hr, we fitted a factor $R a t e = 20$ for our model.

Our model shows that the ratios of $k_{6} / k_{7}$ and $k_{9} / k_{8}$ are different for the conditions without and with WNT stimulation, suggesting a change in mechanism for nuclear shuttling of CTNNB1 and nuclear retention of CTNNB1 in going from the WNT ‘off’ situation to the WNT ‘on’ situation. It seems likely that such changes do not occur instantaneously. In our model we therefore allow a gradual rise in $k_{5} / k_{4}$ and a gradual transition of the ratios of $k_{6} / k_{7}$ and $k_{9} / k_{8}$ from WNT ‘off’ to the WNT ‘on’. In our model, this is included by setting a parameter (‘Steep’) that indicates the time after application of WNT the transition from WNT ‘off’ parameter values to WNT ‘on’ parameter values is complete. The value that gives a good approximation of the experimentally observed concentration curves is Steep = 150 minutes (Figure 6B-F).

Data availability

Source data: for numerical data points in Figures 2-5,7-8 are attached to this article. In addition a comprehensive overview of all numerical data (summary statistics; median, mean and 95% CI's) for the FCS and N&B experiments depicted in Figures 5, 7, 8 plus accompanying supplements and in Tables 1, 2 and 3 is provided in summary tables as Supplementary File 1. Raw data: Original FACS data (.fcs), Western blot data (.tif), confocal images (.tif), FCS data (.ptu- and. oif reference images), N&B data (.ptu/.tif and. oif reference images) have been provided on Open Science Framework (https://osf.io/dczx8/). Source code: scripts for the following have been made publicly available on Open Science Framework (https://osf.io/dczx8/), as referenced in the materials and methods section: Cell profiler segmentation pipeline (Figure 3), R script based on PlotsOfDifference to generate Figure 3 supplement 2 and supplementary movies 4-6, ImageJ N&B analysis script (Figures 5,7 and 8), R source code for the computational model (Figure 6).

The following data sets were generated

(2021) Open Science Framework
ID 10.17605/OSF.IO/DCZX8. Quantitative Imaging of CTNNB1.

https://osf.io/dczx8/

References

1. Aberle H
2. Bauer A
3. Stappert J
4. Kispert A
5. Kemler R
(1997) beta-catenin is a target for the ubiquitin-proteasome pathway
The EMBO Journal 16:3797–3804.
https://doi.org/10.1093/emboj/16.13.3797
- PubMed
- Google Scholar
1. Agajanian MJ
2. Walker MP
3. Axtman AD
4. Ruela-de-Sousa RR
5. Serafin DS
6. Rabinowitz AD
7. Graham DM
8. Ryan MB
9. Tamir T
10. Nakamichi Y
11. Gammons MV
12. Bennett JM
13. Couñago RM
14. Drewry DH
15. Elkins JM
16. Gileadi C
17. Gileadi O
18. Godoi PH
19. Kapadia N
20. Müller S
21. Santiago AS
22. Sorrell FJ
23. Wells CI
24. Fedorov O
25. Willson TM
26. Zuercher WJ
27. Major MB
(2019) WNT activates the AAK1 kinase to promote Clathrin-Mediated endocytosis of LRP6 and establish a negative feedback loop
Cell Reports 26:79–93.
https://doi.org/10.1016/j.celrep.2018.12.023
- PubMed
- Google Scholar
Preprint
(2020) Allele-specific endogenous tagging and quantitative analysis of β-catenin in Colon cancer cells
bioRxiv.
https://doi.org/10.1101/2020.06.18.159616
- Google Scholar
1. Amit S
2. Hatzubai A
3. Birman Y
4. Andersen JS
5. Ben-Shushan E
6. Mann M
7. Ben-Neriah Y
8. Alkalay I
(2002) Axin-mediated CKI phosphorylation of beta-catenin at ser 45: a molecular switch for the wnt pathway
Genes & Development 16:1066–1076.
https://doi.org/10.1101/gad.230302
- PubMed
- Google Scholar
(1987) Ph-positive chronic myeloid leukemia with near-haploid conversion in vivo and establishment of a continuously growing cell line with similar cytogenetic pattern
Cancer Genetics and Cytogenetics 24:335–343.
https://doi.org/10.1016/0165-4608(87)90116-6
- PubMed
- Google Scholar
1. Anthony CC
2. Robbins DJ
3. Ahmed Y
4. Lee E
(2020) Nuclear regulation of wnt/β-Catenin signaling: it’s a Complex Situation
Genes 11:886.
https://doi.org/10.3390/genes11080886
- Google Scholar
(2012) Maternal wnt/β-catenin signaling coactivates transcription through NF-κB binding sites during Xenopus Axis formation
PLOS ONE 7:e36136.
https://doi.org/10.1371/journal.pone.0036136
- PubMed
- Google Scholar
1. Bain J
2. Plater L
3. Elliott M
4. Shpiro N
5. Hastie CJ
6. Mclauchlan H
7. Klevernic I
8. Arthur JSC
9. Alessi DR
10. Cohen P
(2007) The selectivity of protein kinase inhibitors: a further update
Biochemical Journal 408:297–315.
https://doi.org/10.1042/BJ20070797
- Google Scholar
1. Barker N
2. van Es JH
3. Kuipers J
4. Kujala P
5. van den Born M
6. Cozijnsen M
7. Haegebarth A
8. Korving J
9. Begthel H
10. Peters PJ
11. Clevers H
(2007) Identification of stem cells in small intestine and Colon by marker gene Lgr5
Nature 449:1003–1007.
https://doi.org/10.1038/nature06196
- PubMed
- Google Scholar
1. Behrens J
2. von Kries JP
3. Kühl M
4. Bruhn L
5. Wedlich D
6. Grosschedl R
7. Birchmeier W
(1996) Functional interaction of beta-catenin with the transcription factor LEF-1
Nature 382:638–642.
https://doi.org/10.1038/382638a0
- PubMed
- Google Scholar
1. Bilic J
2. Huang YL
3. Davidson G
4. Zimmermann T
5. Cruciat CM
6. Bienz M
7. Niehrs C
(2007) Wnt induces LRP6 signalosomes and promotes dishevelled-dependent LRP6 phosphorylation
Science 316:1619–1622.
https://doi.org/10.1126/science.1137065
- PubMed
- Google Scholar
1. Bindels DS
2. Haarbosch L
3. van Weeren L
4. Postma M
5. Wiese KE
6. Mastop M
7. Aumonier S
8. Gotthard G
9. Royant A
10. Hink MA
11. Gadella TW
(2017) mScarlet: a bright monomeric red fluorescent protein for cellular imaging
Nature Methods 14:53–56.
https://doi.org/10.1038/nmeth.4074
- PubMed
- Google Scholar
(2021) Mutations and mechanisms of WNT pathway tumour suppressors in Cancer
Nature Reviews Cancer 21:5–21.
https://doi.org/10.1038/s41568-020-00307-z
- PubMed
- Google Scholar
Preprint
1. Canaj H
2. Hussmann JA
3. Li H
4. Beckman KA
5. Goodrich L
6. Cho N
7. Li H
8. Santos YJ
9. McGeever DA
(2019) Deep profiling reveals substantial heterogeneity of integration outcomes in CRISPR knock-in experiments
bioRxiv.
https://doi.org/10.1101/841098
- Google Scholar
1. Cantù C
2. Felker A
3. Zimmerli D
4. Prummel KD
5. Cabello EM
6. Chiavacci E
7. Méndez-Acevedo KM
8. Kirchgeorg L
9. Burger S
10. Ripoll J
11. Valenta T
12. Hausmann G
13. Vilain N
14. Aguet M
15. Burger A
16. Panáková D
17. Basler K
18. Mosimann C
(2018) Mutations in Bcl9 and pygo genes cause congenital heart defects by tissue-specific perturbation of Wnt/β-catenin signaling
Genes & Development 32:1443–1458.
https://doi.org/10.1101/gad.315531.118
- PubMed
- Google Scholar
1. Carette JE
2. Raaben M
3. Wong AC
4. Herbert AS
5. Obernosterer G
6. Mulherkar N
7. Kuehne AI
8. Kranzusch PJ
9. Griffin AM
10. Ruthel G
11. Cin PD
12. Dye JM
13. Whelan SP
14. Chandran K
15. Brummelkamp TR
(2011) Ebola virus entry requires the cholesterol transporter Niemann–Pick C1
Nature 477:340–343.
https://doi.org/10.1038/nature10348
- Google Scholar
1. Chhabra S
2. Liu L
3. Goh R
4. Kong X
5. Warmflash A
(2019) Dissecting the dynamics of signaling events in the BMP, WNT, and NODAL cascade during self-organized fate patterning in human gastruloids
PLOS Biology 17:e3000498.
https://doi.org/10.1371/journal.pbio.3000498
- PubMed
- Google Scholar
(2013) Quantitative analysis of Self-Association and mobility of annexin A4 at the plasma membrane
Biophysical Journal 104:1875–1885.
https://doi.org/10.1016/j.bpj.2013.02.057
- Google Scholar
(2008) Mapping the number of molecules and brightness in the laser scanning microscope
Biophysical Journal 94:2320–2332.
https://doi.org/10.1529/biophysj.107.114645
- PubMed
- Google Scholar
1. Doumpas N
2. Lampart F
3. Robinson MD
4. Lentini A
5. Nestor CE
6. Cantù C
7. Basler K
(2019) TCF/LEF dependent and independent transcriptional regulation of wnt/β-catenin target genes
The EMBO Journal 38:e98873.
https://doi.org/10.15252/embj.201798873
- PubMed
- Google Scholar
1. Dross N
2. Spriet C
3. Zwerger M
4. Müller G
5. Waldeck W
6. Langowski J
(2009) Mapping eGFP oligomer mobility in living cell nuclei
PLOS ONE 4:e5041.
https://doi.org/10.1371/journal.pone.0005041
- PubMed
- Google Scholar
1. Eckert AF
2. Gao P
3. Wesslowski J
4. Wang X
5. Rath J
6. Nienhaus K
7. Davidson G
8. Nienhaus GU
(2020) Measuring ligand-cell surface receptor affinities with axial line-scanning fluorescence correlation spectroscopy
eLife 9:e55286.
https://doi.org/10.7554/eLife.55286
- Google Scholar
1. Elson EL
(2011) Fluorescence correlation spectroscopy: past, present, future
Biophysical Journal 101:2855–2870.
https://doi.org/10.1016/j.bpj.2011.11.012
- PubMed
- Google Scholar
(2005) Functional interaction between beta-catenin and FOXO in oxidative stress signaling
Science 308:1181–1184.
https://doi.org/10.1126/science.1109083
- PubMed
- Google Scholar
(2014) Megabase-scale deletion using CRISPR/Cas9 to generate a fully haploid human cell line
Genome Research 24:2059–2065.
https://doi.org/10.1101/gr.177220.114
- PubMed
- Google Scholar
(1998) Nuclear localization signal-independent and importin/karyopherin-independent nuclear import of beta-catenin
Current Biology 8:181–190.
https://doi.org/10.1016/S0960-9822(98)70082-X
- PubMed
- Google Scholar
1. Fagotto F
2. Jho E
3. Zeng L
4. Kurth T
5. Joos T
6. Kaufmann C
7. Costantini F
(1999) Domains of axin involved in protein-protein interactions, wnt pathway inhibition, and intracellular localization
Journal of Cell Biology 145:741–756.
https://doi.org/10.1083/jcb.145.4.741
- PubMed
- Google Scholar
1. Faux MC
2. Coates JL
3. Catimel B
4. Cody S
5. Clayton AH
6. Layton MJ
7. Burgess AW
(2008) Recruitment of adenomatous polyposis Coli and beta-catenin to axin-puncta
Oncogene 27:5808–5820.
https://doi.org/10.1038/onc.2008.205
- PubMed
- Google Scholar
(2011) Dishevelled interacts with the DIX domain polymerization interface of axin to interfere with its function in down-regulating β-catenin
PNAS 108:1937–1942.
https://doi.org/10.1073/pnas.1017063108
- PubMed
- Google Scholar
(2015) An ancient Pygo-dependent wnt enhanceosome integrated by chip/LDB-SSDP
eLife 4:e09073.
https://doi.org/10.7554/eLife.09073
- Google Scholar
1. Franz A
2. Shlyueva D
3. Brunner E
4. Stark A
5. Basler K
(2017) Probing the canonicity of the wnt/Wingless signaling pathway
PLOS Genetics 13:e1006700.
https://doi.org/10.1371/journal.pgen.1006700
- PubMed
- Google Scholar
(2014) Nodal·Gdf1 heterodimers with bound prodomains enable serum-independent nodal signaling and endoderm differentiation
Journal of Biological Chemistry 289:17854–17871.
https://doi.org/10.1074/jbc.M114.550301
- PubMed
- Google Scholar
1. Gammons M
2. Bienz M
(2018) Multiprotein complexes governing wnt signal transduction
Current Opinion in Cell Biology 51:42–49.
https://doi.org/10.1016/j.ceb.2017.10.008
- PubMed
- Google Scholar
(2014) Wnt signalling induces accumulation of phosphorylated β-catenin in two distinct cytosolic complexes
Open Biology 4:140120.
https://doi.org/10.1098/rsob.140120
- PubMed
- Google Scholar
1. Gibson DG
2. Young L
3. Chuang RY
4. Venter JC
5. Hutchison CA
6. Smith HO
(2009) Enzymatic assembly of DNA molecules up to several hundred kilobases
Nature Methods 6:343–345.
https://doi.org/10.1038/nmeth.1318
- PubMed
- Google Scholar
(2013) The transience of transient overexpression
Nature Methods 10:715–721.
https://doi.org/10.1038/nmeth.2534
- PubMed
- Google Scholar
(2012) Structure-guided evolution of cyan fluorescent proteins towards a quantum yield of 93%
Nature Communications 3:751.
https://doi.org/10.1038/ncomms1738
- PubMed
- Google Scholar
Preprint
1. Goedhart J
(2019) PlotsOfDifferences - a web app for the quantitative comparison of unpaired data
bioRxiv.
https://doi.org/10.1101/578575
- Google Scholar
1. Gottardi CJ
2. Gumbiner BM
(2004) Distinct molecular forms of beta-catenin are targeted to adhesive or transcriptional complexes
Journal of Cell Biology 167:339–349.
https://doi.org/10.1083/jcb.200402153
- PubMed
- Google Scholar
1. Hagemann AI
2. Kurz J
3. Kauffeld S
4. Chen Q
5. Reeves PM
6. Weber S
7. Schindler S
8. Davidson G
9. Kirchhausen T
10. Scholpp S
(2014) In-vivo analysis of formation and endocytosis of the wnt/β-Catenin signaling complex in zebrafish embryos
Journal of Cell Science 127:3970–3982.
https://doi.org/10.1242/jcs.148767
- PubMed
- Google Scholar
1. Hampf M
2. Gossen M
(2006) A protocol for combined Photinus and Renilla luciferase quantification compatible with protein assays
Analytical Biochemistry 356:94–99.
https://doi.org/10.1016/j.ab.2006.04.046
- PubMed
- Google Scholar
(2008) Plasma membrane recruitment of dephosphorylated beta-catenin upon activation of the wnt pathway
Journal of Cell Science 121:1793–1802.
https://doi.org/10.1242/jcs.025536
- PubMed
- Google Scholar
(2012) Kinetic responses of β-catenin specify the sites of wnt control
Science 338:1337–1340.
https://doi.org/10.1126/science.1228734
- PubMed
- Google Scholar
1. Hink MA
(2014) Quantifying intracellular dynamics using fluorescence fluctuation spectroscopy
Protoplasma 251:307–316.
https://doi.org/10.1007/s00709-013-0602-z
- PubMed
- Google Scholar
1. Holstein TW
(2012) The evolution of the wnt pathway
Cold Spring Harbor Perspectives in Biology 4:1–17.
https://doi.org/10.1101/cshperspect.a007922
- Google Scholar
1. Holzmann J
2. Politi AZ
3. Nagasaka K
4. Hantsche-Grininger M
5. Walther N
6. Koch B
7. Fuchs J
8. Dürnberger G
9. Tang W
10. Ladurner R
11. Stocsits RR
12. Busslinger GA
13. Novák B
14. Mechtler K
15. Davidson IF
16. Ellenberg J
17. Peters JM
(2019) Absolute quantification of Cohesin, CTCF and their regulators in human cells
eLife 8:e46269.
https://doi.org/10.7554/eLife.46269
- PubMed
- Google Scholar
1. Hu M
2. Kurobe M
3. Jeong YJ
4. Fuerer C
5. Ghole S
6. Nusse R
7. Sylvester KG
(2007) Wnt/beta-catenin signaling in murine hepatic transit amplifying progenitor cells
Gastroenterology 133:1579–1591.
https://doi.org/10.1053/j.gastro.2007.08.036
- PubMed
- Google Scholar
(2016) Construction and experimental validation of a petri net model of wnt/β-Catenin signaling
PLOS ONE 11:e0155743.
https://doi.org/10.1371/journal.pone.0155743
- Google Scholar
(2011) Regulation of β-catenin nuclear dynamics by GSK-3β involves a LEF-1 positive feedback loop
Traffic 12:983–999.
https://doi.org/10.1111/j.1600-0854.2011.01207.x
- PubMed
- Google Scholar
1. Kafri P
2. Hasenson SE
3. Kanter I
4. Sheinberger J
5. Kinor N
6. Yunger S
7. Shav-Tal Y
(2016) Quantifying β-catenin subcellular dynamics and cyclin D1 mRNA transcription during wnt signaling in single living cells
eLife 5:e16748.
https://doi.org/10.7554/eLife.16748
- PubMed
- Google Scholar
1. Kan W
2. Enos MD
3. Korkmazhan E
4. Muennich S
5. Chen DH
6. Gammons MV
7. Vasishtha M
8. Bienz M
9. Dunn AR
10. Skiniotis G
11. Weis WI
(2020) Limited dishevelled/Axin oligomerization determines efficiency of wnt/β-catenin signal transduction
eLife 9:e55015.
https://doi.org/10.7554/eLife.55015
- PubMed
- Google Scholar
1. Kaur G
2. Costa MW
3. Nefzger CM
4. Silva J
5. Fierro-González JC
6. Polo JM
7. Bell TD
8. Plachta N
(2013) Probing transcription factor diffusion dynamics in the living mammalian embryo with photoactivatable fluorescence correlation spectroscopy
Nature Communications 4:1637.
https://doi.org/10.1038/ncomms2657
- PubMed
- Google Scholar
1. Kitazawa M
2. Hatta T
3. Ogawa K
4. Fukuda E
5. Goshima N
6. Natsume T
(2017) Determination of Rate-Limiting factor for formation of Beta-Catenin destruction complexes using absolute protein quantification
Journal of Proteome Research 16:3576–3584.
https://doi.org/10.1021/acs.jproteome.7b00305
- PubMed
- Google Scholar
1. Kohler EM
2. Derungs A
3. Daum G
4. Behrens J
5. Schneikert J
(2008) Functional definition of the mutation cluster region of adenomatous polyposis coli in colorectal tumours
Human Molecular Genetics 17:1978–1987.
https://doi.org/10.1093/hmg/ddn095
- PubMed
- Google Scholar
(2010) Interactions between SOX factors and wnt/β-catenin signaling in development and disease
Developmental Dynamics : An Official Publication of the American Association of Anatomists 239:56–68.
https://doi.org/10.1002/dvdy.22046
- PubMed
- Google Scholar
(1999) Isolation and characterization of a near-haploid human cell line
Experimental Cell Research 252:273–280.
https://doi.org/10.1006/excr.1999.4656
- PubMed
- Google Scholar
(2006) Cyan and yellow super fluorescent proteins with improved brightness, protein folding, and FRET förster radius
Biochemistry 45:6570–6580.
https://doi.org/10.1021/bi0516273
- PubMed
- Google Scholar
(2007) Improved green and blue fluorescent proteins for expression in Bacteria and mammalian cells
Biochemistry 46:3775–3783.
https://doi.org/10.1021/bi0622874
- PubMed
- Google Scholar
(2006) Nucleo-cytoplasmic distribution of beta-catenin is regulated by retention
Journal of Cell Science 119:1453–1463.
https://doi.org/10.1242/jcs.02864
- PubMed
- Google Scholar
(2014) Self-association of the APC tumor suppressor is required for the assembly, stability, and activity of the wnt signaling destruction complex
Molecular Biology of the Cell 25:3424–3436.
https://doi.org/10.1091/mbc.e14-04-0885
- PubMed
- Google Scholar
1. Lam CS
2. Mistri TK
3. Foo YH
4. Sudhaharan T
5. Gan HT
6. Rodda D
7. Lim LH
8. Chou C
9. Robson P
10. Wohland T
11. Ahmed S
(2012) DNA-dependent Oct4-Sox2 interaction and diffusion properties characteristic of the pluripotent cell state revealed by fluorescence spectroscopy
Biochemical Journal 448:21–33.
https://doi.org/10.1042/BJ20120725
- PubMed
- Google Scholar
(1999) The human F box protein β-Trcp associates with the Cul1/Skp1 complex and regulates the stability of β-catenin
Oncogene 18:849–854.
https://doi.org/10.1038/sj.onc.1202653
- Google Scholar
1. Lebensohn AM
2. Dubey R
3. Neitzel LR
4. Tacchelly-Benites O
5. Yang E
6. Marceau CD
7. Davis EM
8. Patel BB
9. Bahrami-Nejad Z
10. Travaglini KJ
11. Ahmed Y
12. Lee E
13. Carette JE
14. Rohatgi R
(2016) Comparative genetic screens in human cells reveal new regulatory mechanisms in WNT signaling
eLife 5:e21459.
https://doi.org/10.7554/eLife.21459
- PubMed
- Google Scholar
1. Lee E
2. Salic A
3. Krüger R
4. Heinrich R
5. Kirschner MW
(2003) The roles of APC and axin derived from experimental and theoretical analysis of the wnt pathway
PLOS Biology 1:E10.
https://doi.org/10.1371/journal.pbio.0000010
- PubMed
- Google Scholar
1. Li VS
2. Ng SS
3. Boersema PJ
4. Low TY
5. Karthaus WR
6. Gerlach JP
7. Mohammed S
8. Heck AJ
9. Maurice MM
10. Mahmoudi T
11. Clevers H
(2012) Wnt signaling through inhibition of β-catenin degradation in an intact Axin1 complex
Cell 149:1245–1256.
https://doi.org/10.1016/j.cell.2012.05.002
- PubMed
- Google Scholar
1. Liu C
2. Li Y
3. Semenov M
4. Han C
5. Baeg GH
6. Tan Y
7. Zhang Z
8. Lin X
9. He X
(2002) Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism
Cell 108:837–847.
https://doi.org/10.1016/S0092-8674(02)00685-2
- PubMed
- Google Scholar
1. Liu X
2. Rubin JS
3. Kimmel AR
(2005) Rapid, Wnt-induced changes in GSK3beta associations that regulate beta-catenin stabilization are mediated by galpha proteins
Current Biology 15:1989–1997.
https://doi.org/10.1016/j.cub.2005.10.050
- PubMed
- Google Scholar
1. Livak KJ
2. Schmittgen TD
(2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta delta C(T)) Method
Methods 25:402–408.
https://doi.org/10.1006/meth.2001.1262
- PubMed
- Google Scholar
(2016) Generating cellular diversity and spatial form: wnt signaling and the evolution of multicellular animals
Developmental Cell 38:643–655.
https://doi.org/10.1016/j.devcel.2016.08.011
- PubMed
- Google Scholar
1. Lui TT
2. Lacroix C
3. Ahmed SM
4. Goldenberg SJ
5. Leach CA
6. Daulat AM
7. Angers S
(2011) The ubiquitin-specific protease USP34 regulates axin stability and wnt/β-Catenin signaling
Molecular and Cellular Biology 31:2053–2065.
https://doi.org/10.1128/MCB.01094-10
- PubMed
- Google Scholar
1. Lustig B
2. Jerchow B
3. Sachs M
4. Weiler S
5. Pietsch T
6. Karsten U
7. van de Wetering M
8. Clevers H
9. Schlag PM
10. Birchmeier W
11. Behrens J
(2002) Negative feedback loop of wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors
Molecular and Cellular Biology 22:1184–1193.
https://doi.org/10.1128/MCB.22.4.1184-1193.2002
- PubMed
- Google Scholar
1. Lybrand DB
2. Naiman M
3. Laumann JM
4. Boardman M
5. Petshow S
6. Hansen K
7. Scott G
8. Wehrli M
(2019) Destruction complex dynamics: wnt/β-catenin signaling alters Axin-GSK3β interactions in vivo
Dev 146:dev164145.
https://doi.org/10.1242/dev.164145
- Google Scholar
1. Ma W
2. Chen M
3. Kang H
4. Steinhart Z
5. Angers S
6. He X
7. Kirschner MW
(2020) Single-molecule dynamics of dishevelled at the plasma membrane and wnt pathway activation
PNAS 117:16690–16701.
https://doi.org/10.1073/pnas.1910547117
- PubMed
- Google Scholar
1. Macháň R
2. Wohland T
(2014) Recent applications of fluorescence correlation spectroscopy in live systems
FEBS Letters 588:3571–3584.
https://doi.org/10.1016/j.febslet.2014.03.056
- PubMed
- Google Scholar
1. Mahen R
2. Koch B
3. Wachsmuth M
4. Politi AZ
5. Perez-Gonzalez A
6. Mergenthaler J
7. Cai Y
8. Ellenberg J
(2014) Comparative assessment of fluorescent transgene methods for quantitative imaging in human cells
Molecular Biology of the Cell 25:3610–3618.
https://doi.org/10.1091/mbc.e14-06-1091
- PubMed
- Google Scholar
1. Maher MT
2. Mo R
3. Flozak AS
4. Peled ON
5. Gottardi CJ
(2010) Beta-catenin phosphorylated at serine 45 is spatially uncoupled from beta-catenin phosphorylated in the GSK3 domain: implications for signaling
PLOS ONE 5:e10184.
https://doi.org/10.1371/journal.pone.0010184
- PubMed
- Google Scholar
1. Massey J
2. Liu Y
3. Alvarenga O
4. Saez T
5. Schmerer M
6. Warmflash A
(2019) Synergy with tgfβ ligands switches WNT pathway dynamics from transient to sustained during human pluripotent cell differentiation
PNAS 116:4989–4998.
https://doi.org/10.1073/pnas.1815363116
- PubMed
- Google Scholar
(2011) The adenomatous polyposis coli tumour suppressor is essential for axin complex assembly and function and opposes axin's interaction with Dishevelled
Open Biology 1:110013.
https://doi.org/10.1098/rsob.110013
- PubMed
- Google Scholar
(1996) XTcf-3 transcription factor mediates beta-catenin-induced Axis formation in Xenopus embryos
Cell 86:391–399.
https://doi.org/10.1016/S0092-8674(00)80112-9
- PubMed
- Google Scholar
1. Morin PJ
2. Sparks AB
3. Korinek V
4. Barker N
5. Clevers H
6. Vogelstein B
7. Kinzler KW
(1997) Activation of beta-catenin-Tcf signaling in Colon cancer by mutations in beta-catenin or APC
Science 275:1787–1790.
https://doi.org/10.1126/science.275.5307.1787
- PubMed
- Google Scholar
1. Mukherjee A
2. Dhar N
3. Stathos M
4. Schaffer DV
5. Kane RS
(2018) Understanding how wnt influences destruction complex activity and β-Catenin dynamics
iScience 6:13–21.
https://doi.org/10.1016/j.isci.2018.07.007
- PubMed
- Google Scholar
1. Nusse R
2. Clevers H
(2017) Wnt/β-Catenin signaling, disease, and emerging therapeutic modalities
Cell 169:985–999.
https://doi.org/10.1016/j.cell.2017.05.016
- PubMed
- Google Scholar
1. Pack C
2. Saito K
3. Tamura M
4. Kinjo M
(2006) Microenvironment and effect of energy depletion in the nucleus analyzed by mobility of multiple oligomeric EGFPs
Biophysical Journal 91:3921–3936.
https://doi.org/10.1529/biophysj.105.079467
- PubMed
- Google Scholar
1. Pack CG
2. Yukii H
3. Toh-e A
4. Kudo T
5. Tsuchiya H
6. Kaiho A
7. Sakata E
8. Murata S
9. Yokosawa H
10. Sako Y
11. Baumeister W
12. Tanaka K
13. Saeki Y
(2014) Quantitative live-cell imaging reveals spatio-temporal dynamics and cytoplasmic assembly of the 26S proteasome
Nature Communications 5:3396.
https://doi.org/10.1038/ncomms4396
- PubMed
- Google Scholar
1. Petrásek Z
2. Schwille P
(2008) Precise measurement of diffusion coefficients using scanning fluorescence correlation spectroscopy
Biophysical Journal 94:1437–1448.
https://doi.org/10.1529/biophysj.107.108811
- PubMed
- Google Scholar
1. Polakis P
(2000)
Wnt signaling and cancer

Genes & Development 14:1837–1851.
- PubMed
- Google Scholar
(2015) A novel GSK3-regulated APC:axin interaction regulates wnt signaling by driving a catalytic cycle of efficient βcatenin destruction
eLife 4:e08022.
https://doi.org/10.7554/eLife.08022
- PubMed
- Google Scholar
(2017) Reconstituting regulation of the canonical wnt pathway by engineering a minimal β-catenin destruction machine
Molecular Biology of the Cell 28:41–53.
https://doi.org/10.1091/mbc.e16-07-0557
- PubMed
- Google Scholar
1. Ran FA
2. Hsu PD
3. Wright J
4. Agarwala V
5. Scott DA
6. Zhang F
(2013) Genome engineering using the CRISPR-Cas9 system
Nature Protocols 8:2281–2308.
https://doi.org/10.1038/nprot.2013.143
- PubMed
- Google Scholar
1. Rim EY
2. Kinney LK
3. Nusse R
(2020) β-catenin-mediated wnt signal transduction proceeds through an endocytosis-independent mechanism
Molecular Biology of the Cell 31:1425–1436.
https://doi.org/10.1091/mbc.E20-02-0114
- PubMed
- Google Scholar
(2011) Deconstructing the ßcatenin destruction complex: mechanistic roles for the tumor suppressor APC in regulating wnt signaling
Molecular Biology of the Cell 22:1845–1863.
https://doi.org/10.1091/mbc.e10-11-0871
- PubMed
- Google Scholar
1. Sayat R
2. Leber B
3. Grubac V
4. Wiltshire L
5. Persad S
(2008) O-GlcNAc-glycosylation of beta-catenin regulates its nuclear localization and transcriptional activity
Experimental Cell Research 314:2774–2787.
https://doi.org/10.1016/j.yexcr.2008.05.017
- PubMed
- Google Scholar
1. Schaefer KN
2. Bonello TT
3. Zhang S
4. Williams CE
5. Roberts DM
6. McKay DJ
7. Peifer M
(2018) Supramolecular assembly of the beta-catenin destruction complex and the effect of wnt signaling on its localization, molecular size, and activity in vivo
PLOS Genetics 14:e1007339.
https://doi.org/10.1371/journal.pgen.1007339
- PubMed
- Google Scholar
1. Schaefer KN
2. Pronobis M
3. Williams CE
4. Zhang S
5. Bauer L
6. Goldfarb D
7. Yan F
8. Major MB
(2020) Wnt regulation: exploring Axin-Disheveled interactions and defining mechanisms by which the SCF E3 ubiquitin ligase is recruited to the destruction complex
Molecular Biology 31:992–1014.
https://doi.org/10.1091/mbc.E19-11-0647
- Google Scholar
1. Schaefer KN
2. Peifer M
(2019) Wnt/Beta-Catenin signaling regulation and a role for biomolecular condensates
Developmental Cell 48:429–444.
https://doi.org/10.1016/j.devcel.2019.01.025
- PubMed
- Google Scholar
1. Schmittgen TD
2. Livak KJ
(2008) Analyzing real-time PCR data by the comparative C(T) method
Nature Protocols 3:1101–1108.
https://doi.org/10.1038/nprot.2008.73
- PubMed
- Google Scholar
(2013) Analysing the impact of nucleo-cytoplasmic shuttling of β-catenin and its antagonists APC, axin and GSK3 on wnt/β-catenin signalling
Cellular Signalling 25:2210–2221.
https://doi.org/10.1016/j.cellsig.2013.07.005
- PubMed
- Google Scholar
(2007) Truncated APC regulates the transcriptional activity of beta-catenin in a cell cycle dependent manner
Human Molecular Genetics 16:199–209.
https://doi.org/10.1093/hmg/ddl464
- PubMed
- Google Scholar
1. Schuijers J
2. Mokry M
3. Hatzis P
4. Cuppen E
5. Clevers H
(2014) Wnt-induced transcriptional activation is exclusively mediated by TCF/LEF
The EMBO Journal 33:146–156.
https://doi.org/10.1002/embj.201385358
- PubMed
- Google Scholar
1. Schwarz-Romond T
2. Fiedler M
3. Shibata N
4. Butler PJ
5. Kikuchi A
6. Higuchi Y
7. Bienz M
(2007) The DIX domain of dishevelled confers wnt signaling by dynamic polymerization
Nature Structural & Molecular Biology 14:484–492.
https://doi.org/10.1038/nsmb1247
- PubMed
- Google Scholar
1. Söderholm S
2. Cantù C
(2020) The WNT/β-catenin dependent transcription: a tissue-specific business
Wiley Interdisciplinary Reviews. Systems Biology and Medicine 13:1511.
https://doi.org/10.1002/wsbm.1511
- Google Scholar
(2000) Structural basis of the Axin-adenomatous polyposis coli interaction
The EMBO Journal 19:2270–2279.
https://doi.org/10.1093/emboj/19.10.2270
- PubMed
- Google Scholar
1. Stamos JL
2. Chu ML-H
3. Enos MD
4. Shah N
5. Weis WI
(2014) Structural basis of GSK-3 inhibition by N-terminal phosphorylation and by the Wnt receptor LRP6
eLife 3:e01998.
https://doi.org/10.7554/eLife.01998
- Google Scholar
(2010) Wnt Signaling Requires Sequestration of Glycogen Synthase Kinase 3 inside Multivesicular Endosomes
Cell 143:1136–1148.
https://doi.org/10.1016/j.cell.2010.11.034
- Google Scholar
1. Tan CW
2. Gardiner BS
3. Hirokawa Y
4. Layton MJ
5. Smith DW
6. Burgess AW
(2012) Wnt Signalling Pathway Parameters for Mammalian Cells
PLOS ONE 7:31882.
https://doi.org/10.1371/journal.pone.0031882
- Google Scholar
1. Tan C
2. Gardiner BS
3. Hirokawa Y
4. Smith DW
5. Burgess AW
(2014) Analysis of Wnt signaling Î²-catenin spatial dynamics in HEK293T cells
BMC Systems Biology 8:44.
https://doi.org/10.1186/1752-0509-8-44
- Google Scholar
1. Thorvaldsen TE
2. Pedersen NM
3. Wenzel EM
4. Schultz SW
5. Brech A
6. LiestÃ¸l K
7. Waaler J
8. Krauss S
9. Stenmark H
(2015) Structure, Dynamics, and Functionality of Tankyrase Inhibitor-Induced Degradasomes
Molecular Cancer Research 13:1487–1501.
https://doi.org/10.1158/1541-7786.MCR-15-0125
- Google Scholar
(2017) Complexity of the Wnt/Î²â€‘catenin pathway: Searching for an activation model
Cellular Signalling 40:30–43.
https://doi.org/10.1016/j.cellsig.2017.08.008
- Google Scholar
1. Tran H
2. Polakis P
(2012) Reversible Modification of Adenomatous Polyposis Coli (APC) with K63-linked Polyubiquitin Regulates the Assembly and Activity of the Î²-Catenin Destruction Complex
Journal of Biological Chemistry 287:28552–28563.
https://doi.org/10.1074/jbc.M112.387878
- Google Scholar
(2012) The many faces and functions of Î²-catenin
The EMBO Journal 31:2714–2736.
https://doi.org/10.1038/emboj.2012.150
- Google Scholar
1. van Amerongen R
2. Nusse R
(2009) Towards an integrated view of wnt signaling in development
Development 136:3205–3214.
https://doi.org/10.1242/dev.033910
- PubMed
- Google Scholar
1. van der Wal T
2. van Amerongen R
(2020) Walking the tight wire between cell adhesion and WNT signalling: a balancing act for Î²-catenin
Open Biology 10:200267.
https://doi.org/10.1098/rsob.200267
- Google Scholar
(2017) Constitutive scaffolding of multiple wnt enhanceosome components by legless/BCL9
eLife 6:e20882.
https://doi.org/10.7554/eLife.20882
- PubMed
- Google Scholar
(2012) Celebrating 30 Years of Wnt Signaling
Science Signaling 5:mr2.
https://doi.org/10.1126/scisignal.2003714
- Google Scholar
(2000) Anomalous diffusion of fluorescent probes inside living cell nuclei investigated by spatially-resolved fluorescence correlation spectroscopy
Journal of Molecular Biology 298:677–689.
https://doi.org/10.1006/jmbi.2000.3692
- Google Scholar
(2001) CRM1- and Ran-independent nuclear export of Î²-catenin
Current Biology 11:18–28.
https://doi.org/10.1016/S0960-9822(00)00045-2
- Google Scholar
(2018) Wnt signalling: conquering complexity
Development 145:dev165902.
https://doi.org/10.1242/dev.165902
- Google Scholar
1. Wu X
2. Tu X
3. Joeng KS
4. Hilton MJ
5. Williams DA
6. Long F
(2008) Rac1 Activation Controls Nuclear Localization of Î²-catenin during Canonical Wnt Signaling
Cell 133:340–353.
https://doi.org/10.1016/j.cell.2008.01.052
- Google Scholar
1. Yaguchi K
2. Yamamoto T
3. Matsui R
4. Tsukada Y
5. Shibanuma A
6. Kamimura K
7. Koda T
8. Uehara R
(2018) Uncoordinated centrosome cycle underlies the instability of non-diploid somatic cells in mammals
Journal of Cell Biology 217:2463–2483.
https://doi.org/10.1083/jcb.201701151
- Google Scholar
1. Yamulla RJ
2. Kane EG
3. Moody AE
4. Politi KA
5. Lock NE
6. Foley AVA
7. Roberts DM
(2014) Testing Models of the APC Tumor Suppressor/Î²-Catenin Interaction Reshapes Our View of the Destruction Complex in Wnt Signaling
Genetics 197:1285–1302.
https://doi.org/10.1534/genetics.114.166496
- Google Scholar
1. Yang J
2. Zhang W
3. Evans PM
4. Chen X
5. He X
6. Liu C
(2006) Adenomatous Polyposis Coli (APC) Differentially Regulates Î²-Catenin Phosphorylation and Ubiquitination in Colon Cancer Cells
Journal of Biological Chemistry 281:17751–17757.
https://doi.org/10.1074/jbc.M600831200
- Google Scholar
(1997) MOLECULAR AND FUNCTIONAL ANALYSIS OF CADHERIN-BASED ADHERENS JUNCTIONS
Annual Review of Cell and Developmental Biology 13:119–146.
https://doi.org/10.1146/annurev.cellbio.13.1.119
- Google Scholar
(1999) Î²-Catenin Can Be Transported into the Nucleus in a Ran-unassisted Manner
Molecular Biology of the Cell 10:1119–1131.
https://doi.org/10.1091/mbc.10.4.1119
- Google Scholar
(2012) Assembly of Dishevelled 3-based supermolecular complexes via phosphorylation and Axin
Journal of Molecular Signaling 7:8.
https://doi.org/10.1186/1750-2187-7-8
- Google Scholar
(2017) Wnt signaling in cancer
Oncogene 36:1461–1473.
https://doi.org/10.1038/onc.2016.304
- Google Scholar

Decision letter

Felix Campelo

Reviewing Editor; The Barcelona Institute of Science and Technology, Spain
Aleksandra M Walczak

Senior Editor; École Normale Supérieure, France
Mark Peifer

Reviewer; University of North Carolina at Chapel Hill, United States

Our editorial process produces two outputs: (i) public reviews designed to be posted alongside the preprint for the benefit of readers; (ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.

Acceptance summary:

Wnt signaling plays critical roles in cell fate determination in essentially every tissue in all animals, regulates tissue homeostasis in many adult tissues, and is inappropriately activated in many human cancers. It has been the focus of research for decades, and we have an outline of signal transduction. However, remarkably, key questions remain controversial. Central among these are questions about the nature of the negative regulatory destruction complex, its mechanism of action and how it is turned down by Wnt signaling. Here Saskia and colleagues take a novel and very exciting approach to these questions, combining innovative quantitative live-cell imaging and computational modelling.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Quantitative live-cell imaging yields novel insight into endogenous WNT/CTNNB1 signaling dynamics" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

The authors investigate how cells respond to WNT signaling by altering β catenin (CTNNB1) dynamics. They generated a number of cell lines in which they use different light microscopy techniques (such as FCS and number and brightness measurements) to quantitatively investigate the diffusion behavior and complex formation of intracellular CTNNB1. The results are in general well explained, reasoned and technically well-controlled (except for some, which raised concerns that were pointed out by the reviewers). The main finding of the paper is that CTNNB1 seems to reside in slow-moving complexes (that exist both in the presence and absence of WNT) that become slightly more mobile after WNT addition. As pointed out by the reviewers, these results can be interpreted in different ways, and it is not clear whether these complexes represent the destruction complex (cytoplasm) or enhanceosome (nucleus). In summary, yet the work shows some technical proficiency which could address some critical issues in Wnt signaling, the authors would need to identify the issues that could be resolved by the technique and then design experiments to resolve them in the future.

So, given the relatively large number of major points of the reviews, in particular those that would require extensive extra work (namely, Rev. 1's point 2and3; Rev 2's specific comments 2and3; and Rev 3's points 1and3), we proceed at this point with a formal rejection of the manuscript so then the authors can decide whether they want to work out all these points and proceed with a new submission in the future, or in case they want this paper to be published in a more timely manner, they can submit elsewhere with the reviews herein included, which will be publicly posted in the biorxiv.

Reviewer #1:

CTNNB1 is a core component of canonical Wnt signalling that is frequently mutated in cancers. A constitutively active destruction complex (degradosome) binds and phosphorylates CTNNB1 earmarking it for proteasomal degradation, this complex is inactivated upon Wnt3a/GSK3β inhibition leading to CTNNB1 stabilisation and nuclear translocation. The authors have successfully employed CRISPR mediated endogenous tagging of CTNNB1 and determined its cellular concentration and diffusion dynamics in HAP1 cells, in both the cytoplasm and nucleus by live-cell imaging and analysis. They provide the relative subcellular CTNNB1 concentration for the nucleus and cytoplasm, like previous studies in other cell lines (Tan et al., 2012) and in Xenopus (Lee et al., 2003). In addition their results suggest CTNNB1 resides in slow moving complexes that persist upon Wnt but become slightly more mobile, these results are intriguing but raise several unanswered questions, such as whether these complexes represent the destruction complex (cytoplasm) or enhanceosome (nucleus). The work has been completed to a high standard but I have several concerns listed below.

1) The authors acknowledge significant cell-cell heterogeneity. This is particularly noticeable in Figure 4A upon Wnt3a and CHIR99021 treatment. Figure 4B suggests all cells are analysed regardless of heterogeneity and the only exclusion criteria mentioned in the methodology is cells with a cytoplasm of less than 10pixels. Figure 4C/D does not seem to reflect the variation observed in Figure 4A? What is the spread pre-normalisation before and after treatment? How is the relative increase in nuclear/cytoplasmic intensity affected by cell size? Nuclear and cytoplasmic area? This may affect the relative fold increase and the cytoplasmic area seems highly variable at the confluence of cells shown.

2) Using point FCS the authors determined two diffusion speeds corresponding to monomer and complexed CTNNB1 in both the nucleus and cytoplasm. A modest increase in cytoplasmic diffusion speed of complexed CTNNB1 was observed after Wnt3a (0.461μm2/s^-1) but far from the speed of the monomer (14.9μm2/s^-1) suggesting it remains complexed upon Wnt3a. In addition the fraction of complexed CTNNB1 (~40%) remains largely unaltered. Is the same true under CHIR299021 treatment? Point FCS samples a very small area of the cell cytoplasm/nucleus and therefore gives a small representation of the subcellular pool (which is likely heterogeneous), only a single point appears to have been analysed per-cell and within the 21 cells analysed clear outliers can be observed (Figure 6A/B), this has not been adequately discussed. What is the variation in diffusion measured at different points within a single cell? Some discussion has been made as to these complexes reflecting the destruction complex/proteasome or the enhanceosome but this really needs to be tested in order to make any conclusions about these observations. Especially as cytoplasmic complexes are maintained under Wnt conditions, this would challenge the notion that CTNNB1 disassociates from the destruction complex upon Wnt. Ideally endogenous tagging of other destruction complex components with a different flourophore would be done to address this, if these complexes do represent the destruction complex and remain bound after Wnt this would have significant implications for our understanding of complex inactivation and greatly enhance the manuscript.

3) The NandB analysis averages out monomeric and complexed CTNNB1 intensity across an image stack around a single ROI within each cell. The authors interpret Figure 6C to mean SGFP2-CTNNB1 is present as a monomer whether in a complex or not. This is based on the fact the relative brightness averages at 1.0 similar to a monomeric GFP control. However, the spread of relative brightness is large, and often less than <1 so a relative brightness of 1 cannot refer to a monomeric SGFP2-CTNNB1? Does cellular concentration affect relative brightness? If so, transiently expressed monomer and dimer GFP may not be the best controls. Aggregation is spatially homogeneous and limited by the diffusion rate of protein/complexes – which your FCS measurements suggest is consistent with a large complex. Thus a single average may not represent the diversity of protein complexes, eNandB could be used (Cutrale et al., 2019). As mentioned in point 3, like FCS, you are only sampling a small region of the cell, which may or may not contain a destruction complex for example. Super-resolution imaging techniques such a STORM or LLSM may help with visualisation of cell complex heterogeneity and give a different impression of complex occupancy. I don't think the NandB data is sufficient to say complexes don't exist that contain more than one SGFP2-CTNNB1 molecule.

4) The computational model relies on a number of assumptions determined in other studies that may not reflect the HAP1 cells used in this study. Lee et al., was performed in Xenopus and Tan et al., 2012 found a number of differences in their mammalian cell studies. Important information regarding the concentration of destruction complex components has also been omitted, this information is important for future comparisons of cell-type specific behaviours.

Reviewer #2:

The manuscript by S.M.A. de Man et al., presents a study on the cellular response to Wnt activation and on the intracellular kinetics of β catenin (CTNNB1). The authors have developed cell lines expressing GFP reporters of CTNNB1 using CRISPR CAS9. They present different convincing controls on the specificity of the reporter and decided to analyze the temporal behavior of the best reacting clone. Then, they investigate the temporal evolution of fluorescent signal in the cell cytoplasm and nucleus upon Wnt signaling activation. They quantify the kinetics of the relocalization of CTNNB1 from the cytoplasm to the nucleus upon different strength of activation of the Wnt signaling and GSK3 inhibition. Using FCS, they identify that a dual diffusion model fit better the experimental data than a classical single diffusion model, suggesting the presence of complexes of different sizes. They measure the diffusion parameters and concentrations of the complexes in the nucleus and in the cytoplasm.

Using a dynamical model, the authors reveal that, to recapitulate the experimental observations, the regulation of CTNNB1 upon Wnt signaling has to be controlled at three levels, the destruction complex, the nuclear transport and the binding affinity to the chromatin.

Overall, the study is solid, presenting novel information on the kinetics of CTNNB1 during Wnt signaling. The results are consistent with the classical view on the regulation of β catenin during Wht signaling. I believe it is of interest for a general cell and developmental biology readership such as the one of eLife. I have few comments essentially on the methodology.

– The authors have designed a new cell line allowing for tracing the kinetics of betacatenin over time following Wnt signaling activation. They follow the relative changes in concentration in the nucleus and cytoplasm upon activation of Wnt signaling. Normalized changes render difficult to evaluate if the difference in the increase in the cytoplasm and the nucleus is due to a higher increase in the nucleus or simply due the absence of betacatenin in the nucleus at the onset of the process therefore enhancing the quantification. A non-normalized plot showing the increase in grey levels in the nucleus and cytoplasm should be added to complement the quantification and identify the differences between nuclear and cytoplasmic β catenin. It would also help the reader to compare with the results of concentrations extracted from the FCS.

– The response in figure 4 upon Wnt signaling activation and GSK3 inhibition are different (with the absence of a plateau in the case of GSK3 inhibition). The explanation of this difference is unclear as it is. I would suggest the authors to details a bit more their thoughts on the reason of the difference.

Could this simply be that Wnt activation clusters just a subset of GSK3 at the membrane and that inhibition can reach higher level of depletion of GSK3 in the cytoplasm?

– How GSK3 inhibition treatment affects the FCS measurements, particularly concentrations and different complexes compositions? The differences with Wnt3 activation could provide additional information on the nature of the identified complexes.

– The dynamical model presented in the paper shows a non-monotonous change in the concentration of betacatenin in the cytoplasm after activation. This seems to be due to the kinetics of nuclear transport and does not seems to be present in the experimental observations. Can the authors comment on this point? Is there a way by modulating parameters associated to transport to suppress this discrepancy?

– Finally, the model is consistent with the experimental observations but the authors di not checked with any type of perturbation how the model would compare with the experiments. For instance, how does the model would compare with experiments in the case of GSK3 inhibition, or when nuclear transport is affected. Adding a perturbation case would significantly strengthen the connection between model and experiment and the message of the manuscript.

Reviewer #3:

1) As I state below the paper is carefully done using a difficult and sophisticated biophysical technique, FCS to assess the changes in β catenin diffusion within the cell following Wnt signaling. So it passes the test on being an original piece of work executed well. However what has been learned is quite limited. A few interactions, such as the slow diffusion in the cytoplasm can be interpreted several ways. It is very helpful to have concentrations in the nucleus and cytoplasm for β catenin for future modeling. They could have tried to use single cross correlation with labeled APC or axin or the proteasome to derive more important information about the path through the destruction sequence. But that may be too hard to ask for at this stage. They could have combined their measurements with appropriate mutants or knockouts. I come down close to the line, high on the importance of the problem and the methods and execution; lower on the current take home lesson.

2) The support for the somewhat limited conclusions is strong as it is.

3) There are some technical issues. There is some concern with the FCS data. Itself. Figure 5F and 5G are of some concern. The curve doesn't drop to 1 at long correlation time (>100ms) and there are big fluctuations in the region of short correlation times (<0.1 ms). This could be due to the very long time course (120s) used in the experiment. Have the authors tried to image the same spot multiple times in short interval (etc 10s), or try to analyze 10s sub-trace of the original long trace to see if the conclusions hold? This type of error could influence the calculation of the diffusion coefficient of complexes of CNNTB1. They also affect the quantification of concentration. In line 352-353 the authors mentioned the nuclear concentration of CNNTB1 increases 2.1 fold based on FCS measurement, which is smaller than the fluorescent intensity change. Is this the result of errors such as this?

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Quantitative live-cell imaging and computational modelling shed new light on endogenous WNT/CTNNB1 signaling dynamics" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Aleksandra Walczak as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Mark Peifer (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential Revisions:

1) For free β-catenin, the authors have calculated the theoretical diffusion coefficient based on the measured diffusion coefficient of SGFP2 and the size difference between them. But it is possible that β-catenin will interact with other proteins in a specific or non-specific manner, which will make the β-catenin diffuse substantially slower than SGFP2. The free β-catenin diffusion coefficient will affect the fitted value of the slow diffusion component. I wonder if the authors could fit the autocorrelation curve directly with 2 component model without assuming a free β-catenin diffusion coefficient. On the other hand, if a predetermined free β-catenin diffusion coefficient is preferred to fit the autocorrelation curves, maybe it is better to make a ΔARM-β-catenin and use the diffusion coefficient of that protein as the diffusion coefficient of the fast component, or measure the diffusion coefficient of SGFP2-β-catenin directly in the cell extract. In addition, some of the free receptors have a surface diffusion coefficient in the range of 0.1 um2/s. Have the authors considered the possibility that the β-catenin complexes bind some large structure inside the cell, like ER membrane, showing a very slow diffusion coefficient?

2) The model that the author proposed has some logical flaws. In line 1042, equation (9), the authors write the equation w(k4 χ2 – k5 x4) as the mathematical form of Wnt activation, where w is applied to both k4 χ2 term and the k5 x4 term. The χ2 term represents the active destruction complex, and the x4 term represents the inactive destruction complex. This means Wnt will regulate both forms of the destruction complex the same way. And the ratio of χ2/x4 will not change after Wnt pathway activation. This way to model is logically flawed. The authors do show a response of the model after Wnt treatment, but this is because the destruction complex also has a substrate-bound fraction: x3. This fraction is not controlled directly by Wnt input.

3) It's better to use SGFP2 dimer and trimer as control instead of EGFP, in Figure 5H and 7E/F. The dimer and trimer constructs are important control for the average size and the percentage of bright fluorescent proteins will vary for different fluorescent proteins. I have no idea how SGFP2 compares to EGFP.

4) The data on the concentrations of β-catenin in the cytoplasm and nucleus before and after Wnt signaling and after inhibition of GSK3 are a major advance. However, I share the opinion of previous reviewer 2 (comment 1) that too much attention is paid to relative differences and not enough to absolute differences. There is strong evidence in the literature that one function of the destruction complex is to retain β-catenin in the cytoplasm (e.g. https://pubmed.ncbi.nlm.nih.gov/21471006/), and thus reduce the amount that can enter the nucleus. Their quantitative imaging (e.g. Figure 3A) reveals this quite clearly and combined with their FCS data suggest that the cytoplasmic pool includes this "bound" β-catenin and that this bound pool remains after Wnt stimulation. I would suggest making this clearer and putting it in the context of earlier work on cytoplasmic retention.

5) One model currently in the field is that the key regulated step in β-catenin destruction is not its phosphorylation but instead its transfer to the E3 ligase, an event that may be controlled by GSK3 phosphorylation of APC (e.g. https://pubmed.ncbi.nlm.nih.gov/22682247/;. https://pubmed.ncbi.nlm.nih.gov/26393419/; https://pubmed.ncbi.nlm.nih.gov/32129710/). In talking about this, the authors state: "Our data clearly show that a substantial fraction of CTNNB1 in the cytoplasm remains bound upon pathway stimulation (Figure 5D). This is not predicted by any of the above-mentioned models and challenges the long-held view that mainly monomeric CTNNB1 accumulates." However, in my mind the retention of high MW complexes after Wnt signaling seems like it is consistent with the model in the manuscripts cited above. Further, while model cartoons (I use them myself) indicate accumulation of monomeric β-catenin in the cytoplasm, in fact what seems more likely is that monomeric β-catenin that accumulates after Wnt signaling rapidly enters the nucleus and is "trapped" as part of high MW transcriptional complexes like those their data appear to document, and thus monomeric cytoplasmic levels may be relatively low. I think a more nuanced description of these data would be useful.

6) One major controversy in the field is the nature of the active destruction complex and its multimeric state. Some data (e.g. https://pubmed.ncbi.nlm.nih.gov/26124443/; https://pubmed.ncbi.nlm.nih.gov/26393419/; https://pubmed.ncbi.nlm.nih.gov/29641560/) suggest large multimeric assemblies while in vitro studies have suggested much smaller assemblies (e.g. https://pubmed.ncbi.nlm.nih.gov/32297861/). It was quite surprising that the authors did not review this literature-in fact the latter paper isn't even in the reference list. Their data, while not definitive, certainly has something to say about this discussion. That having been said, however, I do NOT agree with Reviewer 1 that extending this work to visualize tagged destruction complex components is essential. Of course, this is a very important next step, but this manuscript is already rich in novel data. The reduction in "size" of the potential destruction complexes after Wnt treatment is also consistent with recent observations in Drosophila (https://pubmed.ncbi.nlm.nih.gov/29641560/)

7) I think Reviewer 2 comment 2 (previous submission) and the authors response should be more clearly laid out in the manuscript. While many of us said for years that Wnt signaling "turns the destruction complex off", it is now abundantly clear that it only "turns the destruction complex down". For example, in both Drosophila and in cultured mammalian cells complete loss of GSK3 or APC leads to much higher levels of β-catenin than are seen after Wnt signaling.

8) The fact that their Number and Brightness data suggest few destruction complexes bind multiple β-catenins is very surprising. It doesn't fit my preconceptions and that is OK! I like the idea that usually throughput is rapid enough that multiple β-catenins don't accumulate (waterwheel model). It is surprising that Wnt signaling doesn't alter this but those are the data.

https://doi.org/10.7554/eLife.66440.sa1

Author response

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

The authors investigate how cells respond to WNT signaling by altering β catenin (CTNNB1) dynamics. They generated a number of cell lines in which they use different light microscopy techniques (such as FCS and number and brightness measurements) to quantitatively investigate the diffusion behavior and complex formation of intracellular CTNNB1. The results are in general well explained, reasoned and technically well-controlled (except for some, which raised concerns that were pointed out by the reviewers). The main finding of the paper is that CTNNB1 seems to reside in slow-moving complexes (that exist both in the presence and absence of WNT) that become slightly more mobile after WNT addition. As pointed out by the reviewers, these results can be interpreted in different ways, and it is not clear whether these complexes represent the destruction complex (cytoplasm) or enhanceosome (nucleus). In summary, yet the work shows some technical proficiency which could address some critical issues in Wnt signaling, the authors would need to identify the issues that could be resolved by the technique and then design experiments to resolve them in the future.

So, given the relatively large number of major points of the reviews, in particular those that would require extensive extra work (namely, Rev. 1's point 2and3; Rev 2's specific comments 2and3; and Rev 3's points 1and3), we proceed at this point with a formal rejection of the manuscript so then the authors can decide whether they want to work out all these points and proceed with a new submission in the future, or in case they want this paper to be published in a more timely manner, they can submit elsewhere with the reviews herein included, which will be publicly posted in the biorxiv.

After careful consideration, we felt increasingly confident that we could adequately and comprehensively address all of the reviewer comments – including the main criticisms that led to a formal rejection of our first submission. We have therefore decided to perform the extra work required and to proceed with a new submission for consideration by eLife.

Before addressing each of the specific reviewer comments in point-by-point detail (Rev.1: p.3-18, Rev.2: p.19-27, Rev.3 p.28-38) , we want to highlight the following:

We have performed extensive extra work to address Rev.1’s point’s 2and 3, Rev.2’s specific comments 2and 3 and Rev. 3’s points 1 and 3. We specifically want to highlight the following new data, which are shown in Figures 7 and 8 (and accompanying supplements):

1. To directly compare physiological (Wnt_off/Wnt_on) and oncogenic (i.e. constitutively active) signaling, we generated a second cell line using CRISPR/Cas9 genome editing, harboring an oncogenic point mutant form of CTNNB1 (SGFP2-CTNNB1^S45F).

2. To further quantify the levels, complex state and multimerization status when WNT/CTNNB1 signaling is hyperactivated, we performed additional FCS and NandB experiments in the new mutant cell line and upon GSK3B inhibition by CHIR99021 treatment (as requested by reviewers 1 and 2).

3. We use these same perturbations to strengthen the link between our experimental data and the computational model (as suggested by reviewer 2) and provide access to the model in the form of an interactive app (available at https://wntlab.shinyapps.io/WNT_minimal_model/).

We also suspect that we didn’t sufficiently highlight the novel aspects of our work, which caused some questions about the knowledge increase and biological insights. We want to emphasize the technical and conceptual advances our study brings about. In this current submission, we have clearly highlighted the newly obtained biological insights that are supported by the new data mentioned above. While we do not want to claim that this study provides the final answer as to the working mechanism of WNT/CTNNB1 signaling, we are confident that our study pushes the technical and conceptual boundaries in the field. For your convenience, we summarize our main findings (and how they relate to current views) below:

1. Our study is the first to measure changes in the absolute concentration of endogenous CTNNB1 in its relevant functional pools in living human cells in response to physiological or oncogenic WNT signaling. This work is at the forefront of what is technically possible.

2. We use our newly acquired biophysical parameters to build a computational model that predicts the levels and subcellular distribution of CTNNB1 across the different cytoplasmic and nuclear pools. This wholistic picture reveals that, in addition to destruction complex inactivation, nuclear shuttling and nuclear retention are critical regulatory nodes within the pathway in both physiological and oncogenic settings.

3. Our data challenge current views of the core WNT/CTNNB1 signaling mechanism (i.e. turnover of CTNNB1 by the destruction complex) in three ways.

a. First, we show that upon pathway 35-40% of cytoplasmic CTNNB1 remains bound within a complex, regardless of whether signaling is activated via WNT3A (physiological stimulation), GSK3 inhibition (hyperactivation) or a CTNNB1^S45F point mutation (oncogenic activation). These data challenge (and essentially overthrow) a longstanding dogma in the field, namely that CTNNB1 accumulates as a free and monomeric protein in the cytoplasm upon WNT stimulation.

b. Second, we show that the cytoplasmic CTNNB1 complex undergoes a major reduction in size depending on the phosphorylation status of CTNNB1. The nature of destruction complex inactivation has been debated for almost a decade. Our data are consistent with a model in which the destruction complex is bound to the ubiquitination and proteasome machinery when CTNNB1 is phosphorylated, but in which this association is lost when CTNNB1 phosphorylation is blocked.

c. Third, despite the large size of this cytoplasmic complex (compatible with a multivalent destruction complex) our data do not support high-occupancy binding of CTNNB1 under physiological conditions. However, occupancy does increase when a phospho-dead form (CTNNB1^S45F) is present. This has major impacts on how we conceptualize the workings of the CTNNB1 destruction machinery – especially in the context of cancer, since mutations in CTNNB1 (affecting occupancy) may have different biochemical consequences than mutations in APC (affecting multimerization and valency of the destruction complex itself).

Reviewer #1:

CTNNB1 is a core component of canonical Wnt signalling that is frequently mutated in cancers. A constitutively active destruction complex (degradosome) binds and phosphorylates CTNNB1 earmarking it for proteasomal degradation, this complex is inactivated upon Wnt3a/GSK3β inhibition leading to CTNNB1 stabilisation and nuclear translocation. The authors have successfully employed CRISPR mediated endogenous tagging of CTNNB1 and determined its cellular concentration and diffusion dynamics in HAP1 cells, in both the cytoplasm and nucleus by live-cell imaging and analysis. They provide the relative subcellular CTNNB1 concentration for the nucleus and cytoplasm, like previous studies in other cell lines (Tan et al., 2012) and in Xenopus (Lee et al., 2003). In addition their results suggest CTNNB1 resides in slow moving complexes that persist upon Wnt but become slightly more mobile, these results are intriguing but raise several unanswered questions, such as whether these complexes represent the destruction complex (cytoplasm) or enhanceosome (nucleus). The work has been completed to a high standard but I have several concerns listed below.

We thank Reviewer 1 for their critical reading and evaluation of our manuscript and we are happy that they found our work to have been completed to a high standard. We will take the opportunity to address their concerns in more detail below. Before doing so, however, we want to point out two things. First, unlike previous studies (as the reviewer mentions), we perform the first absolute measurements of subcellular CTNNB1 concentrations in living human cells, for both the unbound and complexed pools of the protein. This, in turn, allowed us to build a computational model based on these measured biophysical parameters, which is a major improvement over existing analysis. Second, the shift in complex mobility upon Wnt stimulation is, in fact, far from ‘slight’ or ‘modest’ (see Reviewer 1, point 2).

1) The authors acknowledge significant cell-cell heterogeneity. This is particularly noticeable in Figure 4A upon Wnt3a and CHIR99021 treatment. Figure 4B suggests all cells are analysed regardless of heterogeneity and the only exclusion criteria mentioned in the methodology is cells with a cytoplasm of less than 10pixels. Figure 4C/D does not seem to reflect the variation observed in Figure 4A? What is the spread pre-normalisation before and after treatment? How is the relative increase in nuclear/cytoplasmic intensity affected by cell size? Nuclear and cytoplasmic area? This may affect the relative fold increase and the cytoplasmic area seems highly variable at the confluence of cells shown.

The reviewer poses several questions concerning the heterogeneity of the response to the WNT3A and CHIR99021 treatments previously shown in Figure 4. Of note, these data are now found in Figure 3 (WNT3A) and accompanying supplements and in Figure 8 supplement 1 (CHIR99021). To clearly address all the reviewer’s questions, we have organized them as follows and will address them in this order:

A) Apparent lack of variation in Figure 4C/D (now Figure 3B/C and Figure 8 supplement 1B/C)

B) Scaling of variation after normalization

C) Influence of cell size on CTNNB1 accumulation

A. Apparent lack of variation: We agree that Figure 4C/D (now Figure 3B/C) does not fully show the cell-to-cell heterogeneity that is visible in Figure 4A (now Figure 3A). The reason for this apparent discrepancy is that Figure 4C/D (now Figure 3B/C) shows the 95% confidence interval of the mean of all individually segmented cells. Due to the large sample size (n=155-400 cells per timepoint for each condition) the 95% confidence interval becomes very determined even though there is cell to cell variation. We have chosen this representation because it allows an easy comparison of the general trends of CTNNB1 subcellular accumulation between different conditions.

The cell-to-cell heterogeneity that Reviewer 1 refers to is better observed in Figure supplement 3A/B (previously Figure 4 supplement 2A/B and depicted in Author response image 1), where the data for all individual cells are plotted after 4 hours of WNT3A or CHIR99021 stimulation. From these graphs it is clear that there is indeed a large spread in the extent to which individual cells accumulate CTNNB1 at any given time point, but that almost all cells do react, which results in combination with the high number of cells to a very determined confidence interval of the mean in Figure 3B/C (previously Figure 4C/D).

B. Scaling of variation after normalization: In our previous submission, we indeed only showed normalized fluorescence data for Figure 4 (now Figure 3). Note that this normalization was needed to combine different biological experiments in one graph. However, when we plot an example from an individual biological experiment (i.e. part of the data of Figure 4 (now Figure 3)), we find that the spread in the unnormalized data is similar to the spread of the normalized data (Author response image 1). The reason for this is that for each imaging position (i.e. a specific field of view of several cells) we normalize to the average intensity in the relevant cellular compartment (i.e. cytoplasm or nucleus) prior to treatment – this allows normalization per position but preserves intensity differences within the position so as not to mask cell-to-cell heterogeneity.

Author response image 1

Download asset Open asset

Boxplots and individual datapoints from a single biological experiment (that is part of Figure 3 previously Figure 4) at 0 and 4 hours after treatment.

The upper panels (raw) show the unnormalized intensity and the bottom panels (normalized) the normalized data. Note that although the axis is very different (0-750 compared to 0-5), the spread in the data is highly similar before and after normalization.

C. Influence of cell size: As per Reviewer 1’s suggestion, we have looked at the relation between cell size and WNT responsiveness (Author response image 2). When we divide the cells into 3 different size categories based on their cytoplasmic area (small: <500 pixels, middle: 500-1500 pixels, large: >1500 pixels), we do not find any meaningful differences in the cytoplasmic intensity, nuclear intensity and nuclear/cytoplasmic ratio after 4 hours of treatment between the different size categories.

Of course, there still could be more subtle effects of cell shape and size on CTNNB1 levels (and vice versa) and lots of other data can still be extracted from these live cell imaging experiments as well. We think this is a very interesting research direction, which will certainly get follow-up in our lab, but this is beyond the scope of the current study, which focusses on quantifying the subcellular behavior of CTNNB1 rather than on the causes underlying cell-to-cell heterogeneity.

Author response image 2

Download asset Open asset

Cells were divided into 3 size categories (small: <500 pixels, middle: 500-1500 pixels, large: >1500 pixels).

Graphs show the normalized cytoplasmic intensity (top), normalized nuclear intensity (middle) and the nuclear/cytoplasmic ratio (bottom) for cells treated for 4 hours with BSA (control) 100 ng/ml WNT3A or 8 µM CHIR99021.

2) Using point FCS the authors determined two diffusion speeds corresponding to monomer and complexed CTNNB1 in both the nucleus and cytoplasm. A modest increase in cytoplasmic diffusion speed of complexed CTNNB1 was observed after Wnt3a (0.461μm2/s^-1) but far from the speed of the monomer (14.9μm2/s^-1) suggesting it remains complexed upon Wnt3a. In addition the fraction of complexed CTNNB1 (~40%) remains largely unaltered. Is the same true under CHIR299021 treatment? Point FCS samples a very small area of the cell cytoplasm/nucleus and therefore gives a small representation of the subcellular pool (which is likely heterogeneous), only a single point appears to have been analysed per-cell and within the 21 cells analysed clear outliers can be observed (Figure 6A/B), this has not been adequately discussed. What is the variation in diffusion measured at different points within a single cell? Some discussion has been made as to these complexes reflecting the destruction complex/proteasome or the enhanceosome but this really needs to be tested in order to make any conclusions about these observations. Especially as cytoplasmic complexes are maintained under Wnt conditions, this would challenge the notion that CTNNB1 disassociates from the destruction complex upon Wnt. Ideally endogenous tagging of other destruction complex components with a different flourophore would be done to address this, if these complexes do represent the destruction complex and remain bound after Wnt this would have significant implications for our understanding of complex inactivation and greatly enhance the manuscript.

We thank Reviewer 1 for their detailed and critical evaluation of our FCS experiments and we fully understand that our unexpected findings raise several questions. Again, we have subdivided the different discussion points so we can address each of these more clearly in that same order.

A) Diffusion speed and bound fraction under WNT3A and CHIR99021 stimulation

B) Heterogeneity of FCS measurements in living cells

C) Implications of our findings for the destruction complex

A. Diffusion characteristics of WNT3A vs CHIR99021: First, the reviewer is indeed correct that our data show that a large fraction of CTNNB1 remains complexed upon WNT3A stimulation, which directly challenges the textbook model for WNT/CTNNB1 signaling. The reviewer is also correct that this complex remains much slower than monomeric CTNNB1.

However, we want to point out that a 3-fold change in mobility (“a modest increase”, according to Reviewer 1) is, in fact, quite substantial: the speed of a protein is proportional to the cube root of its size (i.e. a 3-fold increase in speed means a 3³ = 27-fold decrease in size) for spherical particles. Therefore, the cytoplasmic complex present after WNT3A stimulation clearly has a different identity than the one present before.

In our manuscript, we have purposely refrained from estimating complex sizes, since the shape of the complex, molecular crowding and the meshwork of the cytoskeleton in the cytoplasm (and the chromatin in the nucleus) will also influence the speed of larger proteins. Therefore, the cube root relationship likely cannot be applied across our entire range of measurements. Nevertheless, in our new submission we now emphasize these findings as follows (page 18, lines 339-344): “Because a 3.5-fold change in speed would result in 3.53-change in size for a spherical particle (assuming Einstein-Stokes, see equation 7 in the material and methods section for details), this indicates that the size of the cytoplasmic CTNNB1 complex drastically changes when the WNT pathway is activated. Thus, although the fraction of CTNNB1 that resides in a complex remains the same, the identity of the cytoplasmic complex is quite different in unstimulated and WNT3A stimulated cells.”

At Reviewer 1’s request (and as also suggested by Reviewer 2, see our reply on below), we have now also performed FCS experiments upon CHIR99021 treatment (new Figure 7). Interestingly, we find that under these circumstances a large portion of CTNNB1 is also retained in a large cytoplasmic complex. The same is true in cells in which we introduced an S45F mutation in CTNNB1, resulting in constitutive activation of the pathway. Moreover, we find that the speed of this cytoplasmic complex is similarly increased as observed with WNT3A treatment, both upon mutational (S45F) or pharmacological (CHIR99021) inhibition of GSK3. We will further address the implications on the identity of the cytoplasmic CTNNB1 complex at point C on page 10 below.

B. Heterogeneity of FCS measurements in living cells: The reviewer is correct that there is variation in our data and our current datasets indeed contain some outliers. However, we do not think it is warranted (or informative for the reader) to discuss individual datapoints. That being said, we have always checked these outliers and we think we can explain some of them (For instance: In cases where the slow-moving fraction of CTNNB1 is very small, the diffusion times are less well defined. Other cases might be related to the intra- and intercellular heterogeneity the reviewer referred to). We have explicitly chosen to include (and show) all data for transparency and we have only excluded outliers if there was an obvious technical reason to do so (e.g. a clear misfit of the fitting model, which was very rare but has now been included in supplementary file 1).

In an ideal world, we would both have measured multiple points within the same cell and many more cells overall. However, an inherit limitation of point FCS is that it is quite laborsome and time-consuming. This is the reason why we only record a single point within either cellular compartment, especially since in case of WNT3A stimulation we have a small time window in which to record cells (as stated in the methods, all recordings were made 3-5 hours after stimulation and it takes approximately 10 minutes to measure the cytoplasm and the nucleus in a single cell when including all technical actions). While measuring multiple points per cell would indeed allow us to determine the intracellular variation, we found it more important to obtain sufficient independent measurements of both control and stimulated cells within the same experiment and our data demonstrate that this was the right choice: in our current set-up we are able to determine clear and statistically significant changes in the concentration, fractions of bound and unbound CTNNB1 and the relative diffusion speeds of CTNNB1 in unstimulated and stimulated cells despite including both inter- and intracellular variation. This has allowed us to focus on general differences – and meaningful biological changes – between different experimental conditions. The addition of new FCS data on S45F mutant or CHIR99021 treated cells (Figure 7), further strengthen our conclusions.

In summary, we only discuss biological differences that are upheld between different conditions in spite of inter- and intracellular variation and some outliers. In the Discussion section “Challenges and opportunities for fluorescence fluctuation spectroscopy techniques” we have now included a statement of the impact of using a single point measurement on the variability in our measurements to discuss this more adequately as requested by the reviewer: (page 37-38, lines 730-736): “In addition, point FCS is limited to a single position in the cell. Therefore, in addition to the intercellular differences in the WNT signaling response of individual cells, our measurements also sample intracellular heterogeneity caused by the presence of organelles and molecular crowding. Notwithstanding these limitations, we have been able to show that a large portion of CTNNB1 is present in a very large complex in both stimulated and unstimulated conditions and that this complex has a statistically and biologically significant different speed after WNT3A treatment.”.

C. Implications of our findings for the destruction complex: Reviewer 1 rightfully recognizes that our data challenge a long-standing dogma in the field, namely that CTNNB1 disassociates from the destruction complex upon WNT stimulation. Even without knowing the precise identity of the cytoplasmic complex, our data are very clear on this point: We find that 35-40% of cytoplasmic CTNNB1 is bound within a cytoplasmic complex, regardless of whether the WNT/CTNNB1 pathway is activated via WNT3A (physiological stimulation), CHIR99021 treatment (hyperactivation) or an CTNNB1S45F point mutation (oncogenic activation). This alone is an exciting novel biological insight that requires an update of the WNT signaling mechanism.

We also share the reviewer’s enthusiasm and interest in determining the precise nature of the slow-moving CTNNB1 complexes. As also suggested by the reviewer, we have begun the generation of double-tagged cell lines to further pinpoint the composition of the cytoplasmic complex. While we do already have a cell line in which we tagged GSK3B with a red fluorophore (mScarlet-i) in addition to our SGFP2-CTNNB1 allele (see Author response image 3), our genome editing approach has not yet been successful for other destruction complex components and we suspect that low expression levels are a bottleneck. Unfortunately, this is especially true for AXIN1, which would be the most relevant candidate to tag in order to confidently identify the destruction complex.

Author response image 3

Download asset Open asset

These images show a double-tagged line that has endogenously tagged SGFP2-CTNNB1 (left panel, green) and endogenously tagged mScI-GSK3B (middle panel, magenta).

In addition, we have also run into some of the technical challenges that are well known to complicate in vivo FCCS measurements, including “various sources of errors caused by instrumental or optical limitations such as imperfect overlap of detection volumes or detector cross talk” (Štefl et al., 2020). Thus, while ideal on paper for determining cooccupancy of CTNNB1 and other destruction complex components, in reality we have encountered roadblocks that prevent us from obtaining meaningful biological insights about the cytoplasmic complex identity using an FCCS approach. Although this may change in the future, at present these challenges unfortunately outweigh the time and resources we can invest in this area.

We do want to highlight, however, that by performing additional FCS measurements on the S45F CTNNB1 mutant and CHIR99021 treated cells (new Figure 7), as briefly discussed on page 8 above, we obtained more information on the identity of this cytoplasmic complex that further challenge the textbook model for WNT/CTNNB1 signaling: The fact that the increase in speed (i.e. reduction in size) of this cytoplasmic complex is consistent between WNT3A stimulation, CHIR99021 treatment and S45F mutational activation of the pathway, allows us to pinpoint that the size of this complex is correlated to the phosphorylation status of CTNNB1. We think that these biophysical parameters we measure are in agreement with a model in which the destruction complex is bound to the ubiquitination and proteasome machinery when CTNNB1 is being phosphorylated, but loses this association when CTNNB1 phosphorylation is blocked. See page 33, lines 627-636: “One interesting possibility, therefore, is that phosphorylated CTNNB1 is required for coupling the destruction complex to the ubiquitination and proteasome machinery. In fact, although not explicitly mentioned in the main text, supplementary table 1 of Li et al., 2012 shows that in HEK293 cells, which harbor no mutation in the core components of the WNT pathway, CTNNB1 was found to interact with subunits of the proteasome, whereas in the S45F-CTNNB1 mutant cell line Ls174T these interactions were not detected. In conclusion, although we do not directly determine its identity, our measured biophysical parameters of the cytoplasmic CTNNB1 complex are consistent with it representing a large, multivalent destruction complex that is coupled to the proteasome as long as CTNNB1 is being phosphorylated.”

3) The NandB analysis averages out monomeric and complexed CTNNB1 intensity across an image stack around a single ROI within each cell. The authors interpret Figure 6C to mean SGFP2-CTNNB1 is present as a monomer whether in a complex or not. This is based on the fact the relative brightness averages at 1.0 similar to a monomeric GFP control. However, the spread of relative brightness is large, and often less than <1 so a relative brightness of 1 cannot refer to a monomeric SGFP2-CTNNB1? Does cellular concentration affect relative brightness? If so, transiently expressed monomer and dimer GFP may not be the best controls. Aggregation is spatially homogeneous and limited by the diffusion rate of protein/complexes – which your FCS measurements suggest is consistent with a large complex. Thus a single average may not represent the diversity of protein complexes, eNandB could be used (Cutrale et al., 2019). As mentioned in point 3, like FCS, you are only sampling a small region of the cell, which may or may not contain a destruction complex for example. Super-resolution imaging techniques such a STORM or LLSM may help with visualisation of cell complex heterogeneity and give a different impression of complex occupancy. I don't think the NandB data is sufficient to say complexes don't exist that contain more than one SGFP2-CTNNB1 molecule.

We thank the reviewer for their comments and suggestions concerning our NandB data. Below we address all of them, and expect that this should alleviate their concerns. We also understand the reservation the reviewer might have, as our finding that the cytoplasmic CTNNB1 complex usually holds only one or very few CTNNB1 molecules seems to be in direct contrast with the notion that the CTNNB1 destruction complex a very large, phase-separated, multivalent complex. We also further discuss these implications on page 14-16.

For clarity we have subdivided the comment in several points and will address them in that order

(A) Brightness quantification and controls

(A1) Spread in the data

(A2) Relation number of particles and brightness

(A3) Spatial heterogeneity

(A4) eNandB and super resolution

(B) Implications of the NandB data on the occupancy of the destruction complex

A1. A1. Spread in the data: As with any technique, the measurements of relative brightness are subject to experimental variation, which is seen as the spread Reviewer 1 refers to. This is especially true for NandB measurements in living cells, compared to isolated proteins in solution. For normalization purposes, we set the average of the control measurements to 1. By definition, some relative-brightness values will therefore be <1 and others will be >1, both in our control sample as well as in our experimental sample.

However, when we look at the statistics of these measurements, we can see that the brightness of SGFP2-CTNNB1 (in either the cytoplasm or the nucleus and with or without WNT3A treatment) is not statistically different from the EGFP monomer control. This is in contrast to the EGFP dimer, the brightness of which is statistically different from the EGFP monomer (indicated with * in the table below). We are therefore confident that we can use these controls to draw conclusions on the multimerization status of SGFP2-CTNNB1. Although these statistics were already available in the Supplementary File 1 of the previous and current version of the manuscript, we have now added this in the main text as Table 3 to strengthen the conclusions within the body of the text.

A2.A2. Relation number of particles and brightness: Reviewer 1 is completely correct that high concentrations are incompatible with NandB measurements (as is true for any FFS technique). At high concentrations, fluctuations between the pixels in time will become very small, reducing the apparent (i.e. measured) relative brightness. To avoid this, we have been careful to use very low amounts of DNA in our transfections as described in the Materials and methods section (page 46-47, lines 924-927): “For low, FFS-compatible expression of control samples, HAP1WT cells were transfected with ~5 ng pSGFP2-C1, pEGFP (monomer), pEGFP2 (dimer) or pEGFP3 (trimer) and ~200 ng pBlueScript KS(+) per well with Turbofect, X-tremeGene HP or Lipofectamine 3000 the day before the experiment.”

To further appease the reviewer, we have plotted the relative (i.e. normalized) brightness versus the number of particles to verify that the levels obtained with our transfection are indeed within a range where they do not influence the relative brightness. This plot (Author response image 4) confirms that we do not see any dampening of the relative brightness even in measurements with the highest number of particles (200-250 particles). Thus, while the reviewer’s concern is theoretically valid, we have verified that the low plasmid-based transient expression of our controls is suitable for our analysis.

Author response image 4

Download asset Open asset

Figure displaying the relation between the particle number (x-axis) and the normalized brightness (y-axis).

The number of particles is up to 3fold higher than on average in our endogenously tagged SGFP2CTNNB1 cell lines. However, we see no apparent effect of these higher particle numbers on the normalized brightness.

Spatial heterogeneity: Reading the full comment of reviewer 1, we assume that they meant to write that aggregation can be spatially heterogenous (rather than homogenous). In case we misunderstand this, we kindly ask the reviewer to clarify their concern so we can address it more accurately.

In our NandB analysis, we include much larger ROI’s than the point measurements we use in FCS. We apologize that this was not made sufficiently clear in our previous submission. In our new submission, we have added an additional panel showing typical NandB measurement ROI’s to the new Figure 5 (was Figure 6). We hope the reviewer agrees that a sufficiently large ROI of the nucleus and cytoplasm is included in the NandB analysis and therefore this covers the spatial heterogeneity more so than an FCS measurement.

A4. eNandB and super resolution: Reviewer 1 also suggests two additional approaches (eNandB and super resolution imaging) to investigate the brightness of SGFP2-CTNNB1 taking into account the potential heterogeneity of brightness between different complexes. One of us (Mark Hink) has in fact tried the eNandB method in our lab for known ratios of EGFP monomer, dimer, trimer or tetramer. However, the eNandB analysis was unable to extract these known brightness ratios and we were thus unable to reproduce the result of the original publication ((Cutrale et al., 2019)– also cited by Reviewer 1). As a control, the relative brightness obtained with our current NandB analysis did produce the expected average brightness in these experiments. So, while this approach looked promising, it did not work in our hands and we therefore decided to stick with the NandB analysis that provides a single average brightness, since this works reliably in our experience. We also think that our findings with the S45F mutation (see below) further confirm that we can detect brightness changes where they exist.

Although we agree that the use of super resolution imaging techniques offers exciting opportunities, the use of these techniques for brightness quantification is still very much in development and far less straightforward and commonplace than Reviewer 1 seems to imply.

B. Implications of the NandB data on the occupancy of the destruction complex: We agree with the reviewer that we cannot completely exclude that there are complexes that contain more than one SGFP2-CTNNB1 molecule. However, we think we have been careful not to overinterpret our findings (page 19-20, lines 367-369: “Because we found a large fraction of SGFP2-CTNNB1 to reside in a complex using point FCS (Figure 5C-D), this suggests that few, if any, of these complexes contain multiple SGFP2-CTNNB1 molecules.”).

In support of this interpretation, we also want to point out the following: if complexes with multiple SGFP2-CTNNB1 were present in our measurements, these would very quickly increase the average normalized brightness (as well as the apparent (i.e. measured) fraction that is in complex in FCS). This is due to the fact that in both FCS and NandB the intensity is multiplied by itself in the analysis, so it is the squared brightness of the fluorescent species that will contribute to the apparent brightness (and apparent fractions). This is best illustrated with simulations, of which we show some examples (see Author response table 1).

For example, if we assume that 50% of SGFP2-CTNNB1 resides in a complex, with a brightness (i.e. occupancy) of 1, 2 or 3, the average apparent brightness will increase from 1, to 1.67 and 2.5 respectively (Scenarios 1, 2 and 4 in Author response table 1). This immediately shows that it is unlikely that most of the complexes we measure with FCS contain more than one SGFP2-CTNNB1: in that case the average should increase compared to our monomeric EGFP control, which was not the case. This still holds true if we take into account that the brightness of the complex will also affect the apparent fraction of the complex (Scenarios 3 and 5 in Author response table 1).

Although this still leaves the possibility that some CTNNB1 complexes carry more than 1 SGFP2-CTNNB1, it is unlikely that even one complex would contain many CTNNB1 molecules. If, for example, only 1% of the complexes would carry 10 molecules of SGFP2CTNNB1, the apparent brightness would already increase to 1.83 (Scenario 6) – something we would have been able to detect in our NandB measurements. This scenario is also not consistent with the raw FCS traces, since the presence of a low number of complexes with very high brightness would be expected to give rather prominent spikes in the intensity traces, which is something we did not observe.

Summarizing, if highly saturated CTNNB1 complexes exist within the cell, these must be present in very low numbers such that we somehow missed them in both our FCS and NandB measurements. Even if this would be the case, this means that there is another cytoplasmic complex, which is much more abundant and has a low occupancy rate of SGFP2-CTNNB1 and which, moreover, changes in size upon WNT3A stimulation.

Author response table 1

This table describes different scenarios of how different brightness species would impact the average apparent brightness.

Scenario 1	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	50	50	-
complex containing 1 SGFP2-CTNNB1	1	50	50	-
				1


Scenario 2	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	50	20	-
complex containing 2 SGFP2-CTNNB1	2	50	80	-
				1.67


Scenario 3	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	80	50	-
complex containing 2 SGFP2-CTNNB1	2	20	50	-
				1.33

Scenario 4	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	50	10	-
complex containing 3 SGFP2-CTNNB1	3	50	90	-
				2.5

Scenario 5	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	90	50	-
complex containing 3 SGFP2-CTNNB1	3	10	50	-
				1.5

Scenario 6	actual brightness	actual fraction (%)	apparant fraction (%)	average apparent brightness
monomer SGFP2-CTNNB1	1	99	50	-
complex containing 10 SGFP2-CTNNB1	10	1	50	-
				1.83

Therefore, our data simply are not compatible with the prominent presence of a high occupancy complex (i.e. more than 1 or 2 – possibly 3 molecules of SGFP2-CTNNB1) under physiological conditions. We were equally surprised by this result (this is why we performed the simulations depicted in the Table shown above to begin with), as this is not what we would have expected from a large multivalent destruction complex, as appears to be supported by many current reports in the WNT field.

A possible explanation of this discrepancy is that CTNNB1 does reside in such multivalent complexes, but only very transiently. This would result in a situation where, on average, most of the potential binding sites for CTNNB1 are typically unoccupied – somewhat analogous to the wooden vanes in the paddle wheel of an old-fashioned watermill, or the rotating brushes on a Roomba vacuum cleaning robot: like the water and dust in these examples, CTNNB1 would continuously be scooped up (for phosphorylation) and dropped off (for degradation).

Interestingly, in the case of S45F mutation we do find an effect on the brightness (as shown in Figure 7), underscoring that our experimental approach is suitable for detecting these changes if and when they occur. This further suggests that in the presence of this oncogenic mutation the cytoplasmic complex does contain multiple SGFP2-CTNNB1 molecules.

Author response image 5

Download asset Open asset

4) The computational model relies on a number of assumptions determined in other studies that may not reflect the HAP1 cells used in this study. Lee et al., was performed in Xenopus and Tan et al., 2012 found a number of differences in their mammalian cell studies. Important information regarding the concentration of destruction complex components has also been omitted, this information is important for future comparisons of cell-type specific behaviours.

We appreciate the concern of the reviewer. Indeed, there are multiple parameters for which we had to rely on the literature. This underscores the need for studies like ours, which are dedicated to measuring absolute concentrations of proteins and protein complexes in different cell types. We also feel that measuring these parameters directly in living cells improves upon previous measurements in other settings (Lee et al., 2003; Tan et al., 2012), although ultimately such biochemical and biophysical approaches will also need to be compared.

With our approach we were able to obtain these specific parameters for endogenous CTNNB1, in a bound and unbound form, in both the cytoplasm and the nucleus of living human cells for the first time – which is of major importance for the field, as also remarked by reviewer 3. For all other parameters, however, we currently depend on other studies (such as (Lee et al., 2003; Tan et al., 2012)), to supplement our own data. Additional measurements of endogenous concentrations of other components in the future, will indeed help to compare cell-type specific behaviors and will also be needed for the continuous improvement of computational models.

We carefully checked all of the parameters we provided in our first submission and we could not detect and missing information. We apologize for any confusion that may have arisen with respect to details on the concentration of destruction complex components, but we can alleviate this concern: The reviewer is correct that these details were not provided, but this is because in our model we simply do not include individual destruction complex components or destruction complex assembly (in contrast to (Lee et al., 2003)). The reason for this decision was two-fold: First, we did not know the actual biological parameters of the individual components of the destruction complex in HAP1 cells (as Reviewer 1 also points out) and we wanted to minimize the number of assumptions we had to make based on other studies. Second, we realized we could simplify our computational model by only considering the fully formed destruction complex and solely estimating its concentration based on our own data obtained for the cytoplasmic complex as well as on the association of CTNNB1 K1=k1/k2 in our model is equal to K8=120 in the model of (Lee et al., 2003). We have now clarified this in the model description on page 54 lines 1056-1065:

“Our reaction (1) corresponds closely to step 8 in Lee et al. therefore, we use the value of the dissociation constant 𝐾₈ = 120 nM from Lee et al. for our dissociation constant $K_{1} = \frac{K_{2}}{K_{1}} .$ The concentration of the destruction complex is obtained from equation (1) under equilibrium conditions using equations (6), (8), (10), (11) and (14).

−𝑘₁𝑥₁𝑥₂ + 𝑘₂𝑥₃ + 𝑏 = 0

The value of 𝑏 is assumed to be small compared to the two other terms, so we calculate the concentration of the destruction complex as: $K_{1} = K_{1} \frac{x_{3}}{x_{1}} = 120 \frac{62.5}{91} = 82.4$ nM. It was then verified in our interactive app that this value for the destruction complex is indeed consistent with the equilibrium conditions without WNT stimulation.”

For our purposes (i.e. describing the behavior of CTNNB1) this simplified computational model turned out to be sufficient. In the future, individual steps of destruction complex assembly and disassembly could be added to address different questions, but we think this will only be informative if actual biophysical parameters on complex assembly and complex component concentrations are obtained that are relevant for the system under study.

In this new submission, we now also include the (slightly updated) computational model as an interactive app (https://wntlab.shinyapps.io/WNT_minimal_model/) that allows users to change certain parameters of the model, to explore how this affects the levels of different species of CTNNB1. We hope this will make the computational model more accessible for non-computational biologists, who can intuitively “play with” the model. This might already provide readers with some insights into the regulatory nodes in their own specific experiments, similar to the new data we have included in the new Figure 8 of our manuscript. Here we strengthen the link between our data and the model by showing that the regulatory nodes responsible for nuclear retention (k8/k9) and nuclear shuttling (k6/k7) remain equally important during physiological (WNT3A stimulation) and oncogenic (S45F mutant) WNT/CTNNB1 signaling. In addition, we provide the raw scripts (https://osf.io/jx29z/ (WNT_minimal_model_v2.3.R)) so anyone wishing to adapt or add to any of the parameters or steps currently present can do so.

Reviewer #2:

The manuscript by S.M.A. de Man et al., presents a study on the cellular response to Wnt activation and on the intracellular kinetics of β catenin (CTNNB1). The authors have developed cell lines expressing GFP reporters of CTNNB1 using CRISPR CAS9. They present different convincing controls on the specificity of the reporter and decided to analyze the temporal behavior of the best reacting clone. Then, they investigate the temporal evolution of fluorescent signal in the cell cytoplasm and nucleus upon Wnt signaling activation. They quantify the kinetics of the relocalization of CTNNB1 from the cytoplasm to the nucleus upon different strength of activation of the Wnt signaling and GSK3 inhibition. Using FCS, they identify that a dual diffusion model fit better the experimental data than a classical single diffusion model, suggesting the presence of complexes of different sizes. They measure the diffusion parameters and concentrations of the complexes in the nucleus and in the cytoplasm.

Using a dynamical model, the authors reveal that, to recapitulate the experimental observations, the regulation of CTNNB1 upon Wnt signaling has to be controlled at three levels, the destruction complex, the nuclear transport and the binding affinity to the chromatin.

Overall, the study is solid, presenting novel information on the kinetics of CTNNB1 during Wnt signaling. The results are consistent with the classical view on the regulation of β catenin during Wht signaling. I believe it is of interest for a general cell and developmental biology readership such as the one of eLife. I have few comments essentially on the methodology.

We thank reviewer 2 for their concise summary and positive assessment of our work. We will answer their comments in detail below. Please note that we have numbered them for clarity.

– The authors have designed a new cell line allowing for tracing the kinetics of betacatenin over time following Wnt signaling activation. They follow the relative changes in concentration in the nucleus and cytoplasm upon activation of Wnt signaling. Normalized changes render difficult to evaluate if the difference in the increase in the cytoplasm and the nucleus is due to a higher increase in the nucleus or simply due the absence of betacatenin in the nucleus at the onset of the process therefore enhancing the quantification. A non-normalized plot showing the increase in grey levels in the nucleus and cytoplasm should be added to complement the quantification and identify the differences between nuclear and cytoplasmic β catenin. It would also help the reader to compare with the results of concentrations extracted from the FCS.

We agree that the relation between the nuclear and cytoplasmic levels are relevant for the reader, since it allows discrimination between nuclear exclusion in the absence of WNT3A stimulation and/or nuclear enrichment in the presence of WNT3A stimulation. Note that in our first submission these data were already included in Figure 4E (now Figure 3D), which depicts a ratio of the raw (i.e. non-normalized) intensity values, with a nuclear/cytoplasmic ratio <1 indicating nuclear exclusion and a ratio >1 indicating nuclear enrichment. This indeed shows that before stimulation there is nuclear exclusion of CTNNB1 while upon treatment nuclear and cytoplasmic levels are very similar and perhaps only slightly enriched in the nucleus. The latter is in agreement with our FCS data, which also show a nuclear/cytoplasmic ratio greater than 1 following WNT3A stimulation or CHIR99021 treatment (new Figure 8I-J).

The remainder of the live-cell imaging analysis (Figures 4C-E, now Figure 3B-D) required normalization to be able to combine different biological experiments in a single graph. This was not true for the ratio in Figure 4E (now Figure 3D), since by dividing the two intensities to obtain the nuclear-cytoplasmic ratio, the ratio is independent of the raw values and can therefore be compared across different experiments without further normalization.

As per Reviewer 2’s suggestion we have now also added an additional Figure 3 supplement 3 with non-normalized (i.e. raw intensity) values for the cytoplasm and the nucleus from a single biological experiment, so the reader can more directly appreciate the nuclear exclusion at the beginning of the experiment, which may be more intuitive than the ratio depicted in Figure 4E (now Figure 3D).

– The response in figure 4 upon Wnt signaling activation and GSK3 inhibition are different (with the absence of a plateau in the case of GSK3 inhibition). The explanation of this difference is unclear as it is. I would suggest the authors to details a bit more their thoughts on the reason of the difference.

Could this simply be that Wnt activation clusters just a subset of GSK3 at the membrane and that inhibition can reach higher level of depletion of GSK3 in the cytoplasm?

In our first submission we explained the different response upon WTN3A stimulation and CHIR99021 treatment in relation to AXIN2-mediated feedback: “The fact that intracellular SGFP2-CTNNB1 levels in the 8 μM CHIR99021 condition continued to accumulate, suggests that negative feedback, for example through AXIN2 (Lustig et al., 2002), is overridden under these circumstances”. As AXIN2 mediated feedback is relevant to restore destruction complex functionality, it would not play a role in circumstances where functionality is already blocked by other mechanisms, such as direct inhibition of GSK3 activity. Therefore, we deem this a very likely mechanism for the observed discrepancy between the WNT3A and CHIR99021 treatment.

The alternative explanation provided by the reviewer is interesting, but less likely in our opinion: In the case of incomplete GSK3 inhibition by WNT3A, we would expect to observe a difference in the kinetics of CTNNB1 accumulation or a more apparent difference in the overall levels of CTNNB1 between these two conditions. However, it is possible that other feedback mechanisms, such as the removal of the WNT/receptor complex through endocytosis, does limit WNT3A-mediated GSK3 inhibition at later time points, which would be another explanation for the fact that WNT3A and CHIR99021 would start to differ at this time (>4h). Additionally, non-biological explanations may also play a role (e.g. the stability and denaturation of the purified WNT3A protein).

Please note that in our new submission, we have divided the WNT3A and CHIR99021 livecell imaging data over different figures (Figure 3 for WNT3A, Figure 8 supplement 1A- for CHIR99021) to accommodate other changes made to the manuscript. However, our observations and conclusions remain the same, with addition of receptor feedback now added within the text (page 26, lines 480-485:” However, in contrast to what is observed for WNT3A treatment, no plateau was reached at the highest concentration of CHIR99021 (8 µM). The fact that intracellular SGFP2-CTNNB1 levels in the 8 μM CHIR99021 condition continued to accumulate, suggests that negative feedback, for example through AXIN2 (Lustig et al., 2002) or through internalization of receptor complexes (Agajanian et al., 2019), is overridden under these circumstances”)

– How GSK3 inhibition treatment affects the FCS measurements, particularly concentrations and different complexes compositions? The differences with Wnt3 activation could provide additional information on the nature of the identified complexes.

At Reviewer 2’s request (and as also suggested by Reviewer 1, see our reply on page 7-8 above), we have now also performed FCS experiments upon CHIR99021 treatment (new Figure 7). Interestingly, we find that under these circumstances a large portion of CTNNB1 is also retained in a large cytoplasmic complex. The same is true in cells in which we introduced an S45F mutation in CTNNB1, resulting in constitutive activation of the pathway. Moreover, we find that the speed of this cytoplasmic complex is similarly increased as observed with WNT3A treatment, both upon mutational (S45F) or pharmacological (CHIR99021) inhibition of GSK3. Based on these additional experiments, we can now confidently conclude that the size of this cytoplasmic complex is therefore related to the phosphorylation status of CTNNB1: (page 24, lines 450-456): “Similar to the situation detected under physiological conditions (Figure 5D), we find a large fraction of SGFP2-CTNNB1^S45F to reside in a cytoplasmic complex (Figure 7A). As observed for physiological stimulation with WNT3A (Figure 5E), the speed of this complex is increased in SGFP2-CTNNB1^S45F compared to unstimulated HAP1^SGFP2-CTNNB1 cells (Figure 7B). We find similar behavior when we block the GSK3 mediated phosphorylation of wild-type CTNNB1 using CHIR99021 (Figure 7C-D). The reduction in cytoplasmic complex size therefore must occur downstream of CTNNB1 phosphorylation.”.

To our opinion, this leaves only limited room for speculation as to its precise identity, as also mentioned in the discussion: (page 33, lines 627-636): “One interesting possibility, therefore, is that phosphorylated CTNNB1 is required for coupling the destruction complex to the ubiquitination and proteasome machinery. In fact, although not explicitly mentioned in the main text, supplementary table 1 of Li et al., 2012 shows that in HEK293 cells, which harbor no mutation in the core components of the WNT pathway, CTNNB1 was found to interact with subunits of the proteasome, whereas in the S45F-CTNNB1 mutant cell line Ls174T these interactions were not detected. In conclusion, although we do not directly determine its identity, our measured biophysical parameters of the cytoplasmic CTNNB1 complex are consistent with it representing a large, multivalent destruction complex that is coupled to the proteasome as long as CTNNB1 is being phosphorylated.”.

Interestingly, it were thus not the differences but rather the similarities observed between treatments and CTNNB1 activation status that provided additional information on the nature of the identified complexes and we thank Reviewer 1 for this suggestion.

– The dynamical model presented in the paper shows a non-monotonous change in the concentration of betacatenin in the cytoplasm after activation. This seems to be due to the kinetics of nuclear transport and does not seems to be present in the experimental observations. Can the authors comment on this point? Is there a way by modulating parameters associated to transport to suppress this discrepancy?

We thank the Reviewer for their detailed interest in this behavior of the model, which is explained by the way in which WNT/CTNNB1 signaling is ‘switched on’ computationally.

Originally, we modelled WNT addition by simultaneously and immediately changing all parameter values from the “off” state to those we determined for the equilibrium situation in the “on” state. As can be seen in Author response image 6, in this old situation the direct change of k8/k9 and k6/k7 causes an immediate drop in the cytoplasmic concentration (panel A) and an immediate jump in the nuclear concentration (panel C), which does not correspond with our live-cell imaging findings. This is also very clearly seen in the nuclear cytoplasmic ratio (panel E).

We reasoned that inside the cell, all of these parameters would gradually change over time. We therefore added a term to the equations (called “steep” in the new interactive model available via https://wntlab.shinyapps.io/WNT_minimal_model/) for both k6/k7 and k9/k8 as well as for k5/k4 to represent this gradual change. We fitted this term to match the timing of the change in nuclear/cytoplasmic ratio we observed with live-cell imaging. The resulting curves (panel B,D,F Author response image 6) are a better approximation of the experimental observations and we explain the inclusion of this term on page 56 lines 1101-1107: “It seems likely that such changes do not occur instantaneously. In our model we therefore allow a gradual rise in 𝑘₅/𝑘₄ and a gradual transition of the ratios of 𝑘₆/𝑘₇ and 𝑘₉/𝑘₈ from WNT ‘off’ to the WNT ‘on’. In our model this is included by setting a parameter (“Steep”) that indicates the time after application of WNT the transition from WNT ‘off’ parameter values to WNT ‘on’ parameter values is complete. The value that gives a good approximation of the experimentally observed concentration curves is Steep = 150 minutes (Figure 6 panels B-F).”

However, the change in the model is indeed still not constant as keenly spotted by reviewer 2. Further tweaking of delays in the activation of k9/k8 or k6/k7 and the rise of k5/k4 (the same term is now used for all) could perhaps further smoothen this line. However, we don’t think this will provide new biological information on the underlying regulatory mechanisms of WNT signaling and have therefore decided not to pursue this further at the moment.

Author response image 6

Download asset Open asset

Comparison of old model without steep (A,C,E) and new model with steep (B,D,F).

In the old model the change in k6/k7 and k9/k8 is immediate at the addition of WNT3A. This results in a sudden loss of cytoplasmic CTNNB1 (A) and gain in the nucleus (C) as is also apparent from the nuclear-cytoplasmic ratio (E). Our live imaging experiment shows that the changes. This excluded the slow adjustment of the k6/k7 and k8/k9 (called steep in the newest version of the model). Without this adjustment the parameters in the nucleus adjust immediately resulting in an immediate change in the nuclear-cytoplasmic ratio (C) and the TCF-bound fraction (E). This also leads to a dip in the cytoplasmic concentrations of CTNNB1 (B) and immediate increase in cytoplasmic concentrations (C).

– Finally, the model is consistent with the experimental observations but the authors di not checked with any type of perturbation how the model would compare with the experiments. For instance, how does the model would compare with experiments in the case of GSK3 inhibition, or when nuclear transport is affected. Adding a perturbation case would significantly strengthen the connection between model and experiment and the message of the manuscript.

We thank the reviewer for this excellent suggestion. We have now added and extensively analyzed a perturbation case in our new Figure 8.

The mechanism of nuclear translocation of CTNNB1 remains to be resolved. In fact, CTNNB1 does not contain nuclear import or export signals and can translocate independently of classical importin and exporter pathways (Fagotto et al., 1998; Wiechens and Fagotto, 2001; Yokoya et al., 1999). Hence, we could not specifically perturb nuclear/cytoplasmic shuttling (which represents the first critical node in our model, namely k6/k7). Similarly, we empirically found that we could not fully perturb nuclear retention (which represents the second critical node in our model, namely k8/k9) with the small molecule inhibitors currently available (e.g., ICG-001). Therefore, we decided to perturb the third critical node in our model, namely destruction complex activity (k3), via two independent means of blocking the GSK3-mediated inhibition of CTNNB1 (CHIR99021 treatment and S45F mutation).

Using this perturbation case, we were able to establish that upon destruction complex inactivation we still need similar changes in k6/k7 and k8/k9 as in the case of WNT3A stimulation to match the predictions of the model with our experimental observations. These findings underscore the critical nature of all three regulatory nodes in both physiological and oncogenic conditions. We think that these data indeed strengthen the link between our experimental and computational findings and we hope the reviewer agrees.

Importantly, these perturbation experiments also revealed that the changes in nuclear retention and nuclear/cytoplasmic shuttling are directly correlated with the CTNNB1 phosphorylation status, which opens up the exciting possibility that the N-terminal posttranslational modification status of CTNNB1 plays an active and important role in regulating its subcellular distribution (see page 36 lines 690-695: “The fact that direct inhibition of GSK3 mediated phosphorylation of CTNNB1 results in the same behavior, indicates that the phosphorylation status of CTNNB1 plays a critical role. This further emphasizes the importance of posttranslational modifications and conformational changes in CTNNB1 for its subcellular localization and function (Gottardi and Gumbiner, 2004; Sayat et al., 2008; Valenta et al., 2012; van der Wal and van Amerongen, 2020; Wu et al., 2008).”).

Reviewer #3:

1) As I state below the paper is carefully done using a difficult and sophisticated biophysical technique, FCS to assess the changes in β catenin diffusion within the cell following Wnt signaling. So it passes the test on being an original piece of work executed well. However what has been learned is quite limited. A few interactions, such as the slow diffusion in the cytoplasm can be interpreted several ways. It is very helpful to have concentrations in the nucleus and cytoplasm for β catenin for future modeling. They could have tried to use single cross correlation with labeled APC or axin or the proteasome to derive more important information about the path through the destruction sequence. But that may be too hard to ask for at this stage. They could have combined their measurements with appropriate mutants or knockouts. I come down close to the line, high on the importance of the problem and the methods and execution; lower on the current take home lesson.

We thank the reviewer for their careful assessment of our work and we are happy that they consider our work to be original, well executed and high on importance. At the same time, we also understand that the reviewer might have been left wondering what has been learned in terms of take-home lesson. We feel that in our first submission these take-home messages may have gotten lost amidst the complex biophysical details.

Here we will touch upon several points and suggestions made by Reviewer 3. For clarity we have subdivided them and will address them in this order.

A) New take-home messages from the revised manuscript

B) Fluorescence cross-correlation spectroscopy

C) S45F mutant and CHIR99021 perturbation case

A. New take-home messages from the revised manuscript: We recognize that our previous version of the manuscript did not allow us draw strong conclusions about the nature and identity of the complexes we identified, but we were careful not to overinterpret our results. We are confident that we have remedied this situation in our new submission. In particularly, we have included two new figures (Figure 7 and Figure 8) based on two distinct perturbation cases (CHIR99021 treatment and S45F mutation of CTNNB1, explained in more detail below). The combined live-cell imaging, FCS measurements and NandB analyses now allow us to summarize the novel findings and take-home messages of our study as follows, as also highlighted on page 1 and 2 of this rebuttal:

2. Reviewer 3 acknowledges the relevance of obtaining biophysical data for modelling purposes. In fact, we think that these particular points of our study should not be underestimated: We use our newly acquired biophysical parameters to build a computational model that predicts the levels and subcellular distribution of CTNNB1 across the different cytoplasmic and nuclear pools. This wholistic picture reveals that, in addition to destruction complex inactivation, nuclear shuttling and nuclear retention are critical regulatory nodes within the pathway in both physiological and oncogenic settings.

3. Our data challenge current views of the core WNT/CTNNB1 signaling mechanism (i.e. turnover of CTNNB1 by the destruction complex) in three ways.

3.1 First, we show that upon pathway 35-40% of cytoplasmic CTNNB1 remains bound within a complex, regardless of whether signaling is activated via WNT3A (physiological stimulation), GSK3 inhibition (hyperactivation) or a CTNNB1^S45F point mutation (oncogenic activation). These data challenge (and essentially overthrow) a long-standing dogma in the field, namely that CTNNB1 accumulates as a free and monomeric protein in the cytoplasm upon WNT stimulation.

3.2 Second, we show that the cytoplasmic CTNNB1 complex undergoes a major reduction in size depending on the phosphorylation status of CTNNB1. The nature of destruction complex inactivation has been debated for almost a decade. Our data are consistent with a model in which the destruction complex is bound to the ubiquitination and proteasome machinery when CTNNB1 is phosphorylated, but in which this association is lost when CTNNB1 phosphorylation is blocked.

Third, despite the large size of this cytoplasmic complex (compatible with a multivalent destruction complex) our data do not support highoccupancy binding of CTNNB1 under physiological conditions. However, occupancy does increase when a phospho-dead form (CTNNB1^S45F) is present. This has major impacts on how we conceptualize the workings of the CTNNB1 destruction machinery – especially in the context of cancer, since mutations in CTNNB1 (affecting occupancy) may have different biochemical consequences than mutations in APC (affecting multimerization and valency of the destruction complex itself).

We don’t want to presume that our current study is the be-all and end-all to the WNT/CTNNB1 signaling mechanism, but we are confident that it offers solid evidence to challenge different aspects of the textbook model and, in doing so, provides a significant conceptual advance as well as multiple leads for new and testable hypotheses.

B. Fluorescence cross-correlation spectroscopy: We also share the reviewer’s enthusiasm and interest in determining the precise nature of the slow-moving CTNNB1 complexes. As also suggested by Reviewer 3 (and reviewer 1, see page 7 for their comment and page 10 for our reply), we have begun the generation of double-tagged cell lines to further pinpoint the composition of the cytoplasmic complex. While we do already have a cell line in which we tagged GSK3B with a red fluorophore (mScarlet-i) in addition to our SGFP2CTNNB1 allele (Author response image 7), our genome editing approach has not yet been successful for other destruction complex components and we suspect that low expression levels are a bottleneck. Unfortunately, this is especially true for AXIN1, which would be the most relevant candidate to tag in order to confidently identify the destruction complex. C.

Author response image 7

Download asset Open asset

C. S45F mutant and CHIR99021 perturbation case: At Reviewer 3’s suggestion we have also added FCS and NandB data obtained under mutant conditions. For this, we introduced an S45F mutation in CTNNB1 resulting in constitutive activation of the pathway. This mutation is one of the earliest identified oncogenic mutations in CTNNB1 (Morin et al., 1997), and it removes the CSNK1 priming phosphorylation site on CTNNB1 that is needed for sequential phosphorylation by GSK3, and thus blocks its proteasomal degradation (Amit et al., 2002; Liu et al., 2002). In addition, we also obtained new FCS and NandB data after CHIR99021 treatment as suggested by Reviewer 1 and 2 (see page 7 and 21 for their comments). Interestingly, we find that under both these circumstances a large portion of CTNNB1 is also retained in a large cytoplasmic complex (new Figure 7). Moreover, we find that the speed of this cytoplasmic complex is similarly increased as observed with WNT3A treatment, both upon mutational (S45F) or pharmacological (CHIR99021) inhibition of GSK3.

These additional FCS measurements, have given us more information on the identity of this cytoplasmic complex that further challenge the textbook model for WNT/CTNNB1 signaling: The fact that the increase in speed (i.e. reduction in size) of this cytoplasmic complex is consistent between WNT3A stimulation, CHIR99021 treatment and S45F mutational activation of the pathway, allows us to pinpoint that the size of this complex is correlated to the phosphorylation status of CTNNB1. We think that these biophysical parameters we measure are in agreement with a model in which the destruction complex is bound to the ubiquitination and proteasome machinery when CTNNB1 is being phosphorylated, but loses this association when CTNNB1 phosphorylation is blocked. See (page 33, lines 627-636): “One interesting possibility, therefore, is that phosphorylated CTNNB1 is required for coupling the destruction complex to the ubiquitination and proteasome machinery. In fact, although not explicitly mentioned in the main text, supplementary table 1 of Li et al., 2012 shows that in HEK293 cells, which harbor no mutation in the core components of the WNT pathway, CTNNB1 was found to interact with subunits of the proteasome, whereas in the S45F-CTNNB1 mutant cell line Ls174T these interactions were not detected. In conclusion, although we do not directly determine its identity, our measured biophysical parameters of the cytoplasmic CTNNB1 complex are consistent with it representing a large, multivalent destruction complex that is coupled to the proteasome as long as CTNNB1 is being phosphorylated.”.

Of note, we used the same perturbation to determine the effect of an oncogenic CTNNB1 mutation on brightness (and thus occupancy) of the cytoplasmic complex. While our data are not compatible with the prominent presence of a high occupancy complex (i.e. more than 1 or 2 – possibly 3 molecules of SGFP2-CTNNB1) under physiological conditions (quite contrary to what would be expected from a large multivalent destruction complex, as appears to be supported by many current reports in the WNT field), in the case of S45F mutation we do find an effect on the brightness, suggesting that in the presence of this oncogenic mutation the cytoplasmic complex does contain multiple SGFP2-CTNNB1 molecules.

2) The support for the somewhat limited conclusions is strong as it is.

We are happy that Reviewer 3 appreciates our thoroughness. We hope that in our answer under Major comment #1 above we have also convinced the reviewer that the conclusions, novel insights and conceptual advances presented in this work are equally strong as their experimental support and relevant for not just the WNT signaling field, but also a broader scientific community.

3) There are some technical issues. There is some concern with the FCS data. Itself. Figure 5F and 5G are of some concern. The curve doesn't drop to 1 at long correlation time (>100ms) and there are big fluctuations in the region of short correlation times (<0.1 ms). This could be due to the very long time course (120s) used in the experiment. Have the authors tried to image the same spot multiple times in short interval (etc 10s), or try to analyze 10s sub-trace of the original long trace to see if the conclusions hold? This type of error could influence the calculation of the diffusion coefficient of complexes of CNNTB1. They also affect the quantification of concentration. In line 352-353 the authors mentioned the nuclear concentration of CNNTB1 increases 2.1 fold based on FCS measurement, which is smaller than the fluorescent intensity change. Is this the result of errors such as this?

The reviewer touches upon several aspects of the FCS technique, which remains challenging in a complex environment such as a living mammalian cell, for example due to heterogeneous autofluorescence, cell movement etc. The reviewer refers to “errors”, but we think these points rather refer to inherent properties of this technique. For this reason, we have devoted an entire section in the discussion to discuss the challenges and opportunities of this technique (updated and modified in this submission).

We do thank the reviewer for all their questions and suggestions, because we do agree that they are relevant and important to consider. To clearly address them we have organized them as follows and will address them in this order:

A) Concerns relating to the offset of the autocorrelation curve

B) Concerns relating to large fluctuation at short correlation times

C) Concerns relating to duration of the FCS acquisition

D) The discrepancy between fluorescence and concentration

A. Offset of the autocorrelation curve: The reviewer expects the autocorrelation curve to drop to 1 after 100ms. First, please note that the decay seen between 10ms and 1000ms is still part of the slow diffusion component. In these experiments, long correlation times therefore are >1000ms, rather than 100ms. Second, there is indeed a small offset in the autocorrelation curve at these long correlation timepoints (>1000 ms). However, this offset is approximately 0.002 in the specific example depicted in Figure 5F-G (now Figure 4FG). If we look at all of our measurements, the offset of the curves is generally between 0.000 and 0.004, which is not expected to have a big influence on either the particle number or the diffusion times. We also show another curve without any offset from one of our datasets (Author response image 8), to show the difference with the autocorrelation curve shown in the manuscript that was chosen without cherry picking.

Author response image 8

Download asset Open asset

Autocorrelation curves from the same experiment with a small offset of 0.

002 (A) as original example in Figure 5G (now Figure 4G) or without offset (B).

For our analysis we removed any measurements with significant intensity drift Material and Methods, page 48, line 951-954: “Intensity traces with significant photobleaching, cell movement or focal drift were excluded from further analysis (see supplementary file 1 – tab FCS measurements and fitting). From other traces a portion of the trace with minimal (less than 10%) intensity drift or bleaching was selected to generate autocorrelation curve (AC).” This also means that even though our measurements are 120 seconds in total, analysis is sometimes performed on shorter portions of the trace. In the analyzed portion of the trace some minor movement, drift or bleaching (i.e. less than 10% per minute) might still be present and we think that this could be responsible for causing a minor offset of the autocorrelation curve. We therefore used a model that includes a term describing this offset (G∞), (Materials and methods page 48, line 956-957: “G∞ accounts for offset in the AC for example by intensity drift”). This ensures that the amplitude (used to fit the number of particles for the slow and fast fraction of SGFP2-CTNNB1) is not affected by the small offset in the autocorrelation curve. Therefore, the concentrations and diffusion kinetics from our analysis, should not be affected. We hope this addresses the reviewer’s concern.

B. Large fluctuations at short correlation times: The reviewer also wonders about the “big fluctuations in the region of short correlation times (<0.1 ms)”. At these short correlation times we observe two phenomena: first we have afterpulsing, which is a APD (avalanche photodiode detector) artefact that is specific for every detector, and second, we are dealing with the photophysical properties of the specific fluorophore (e.g. fast blinking). Indeed, the noise around this part of the curve seems quite large. However, this is quite normal for autocorrelation curves from a single APD detector, due to the statistics (i.e. the number of events measured by the APD) associate with these very short time intervals due to the pulsed laser excitation and gated detection (which results in the detection of only 1 event per 50ns).

When we look at the fitting of these two events, described by F_trip and τ_trip(see Material and methods page 48 line 958-961: “F_trip and τ_trip describe the fraction of molecules in the dark state and the relaxation of this dark state respectively. Of note, in this case, F_trip and τ_trip account both for blinking of the fluorescent molecules and for the afterpulsing artefact of the APD”.) We see that the residual noise of the experimental curve is fitted nicely around the analysis model as can also be judged by the residual plot in Figure 4G, indicating that the fitted parameters are reliable. In addition, the typical times that are fitted as τ_trip are in the 1-50 µs range, whereas our first diffusion time is significantly removed at around 1 ms. therefore, the fitting of the particle numbers and diffusion times is not easily influenced by the noise at these very early timescales. We would welcome any additional questions or remarks if the reviewer’s concern remains.

C. Duration of the FCS acquisition: Although there are people in the field that prefer multiple short measurements, our own experience, with our experimental setup, is that at these low endogenous concentrations the autocorrelation curves we obtain with these very short traces are very noisy. In our hands, one single measurement with more photons tends to give more determined parameters than statistically combining multiple short measurements.

However, as per the reviewer’s suggestion we have compared the analysis of a large portion of the trace (90s) with shorter, 10s sub-traces. As can be seen in Author response image 9, the results of the 10s traces are indeed noisier, which results in less well-defined confidence intervals of the individually fitted parameters. This, in turn, contributes to the variation between the estimated particle numbers (N) and fraction of slow CTNNB1 (F2) for the different 10s measurements. The analysis of a larger portion of the FCS measurement results in more determined parameters, and also corresponds well to the average of the10s traces. Therefore, we expect that general application of shorter analysis intervals, will yield the same conclusions about concentration and diffusion coefficients as we get from the current analysis strategy. Thus, we have no reason to believe that our data are more error-prone than if we were to perform shorter measurements.

Author response image 9

Download asset Open asset

Comparison of analysis of short and long FCS traces.

Autocorrelation curves were made from nine 10s portion of the measurement (trace), or a single 90s trace from the same FCS measurement of SGFP2-CTNNB1 and subsequently fitted with the same fitting model. The graphs depict the fitted values (dots) and confidence intervals determined by the goodness of fit in the FCS analysis software (lines) for the particle number (N, panel A) and _{the fraction of slow component} (F2, panel B) of the 10s and 90s.

D. Discrepancy between fluorescence and concentration: As for the specific example that Reviewer 3 refers to in line 352-353 page 14-15 lines 276-278 in our new submission:

“This increase was smaller than expected from fluorescence intensity measurements (Figure 3B), for which we currently have no explanation (Figure 5 supplement 1).”: We have, of course, been puzzled ourselves by the difference between the FCS measurement and the fluorescent intensity change and therefore described this discrepancy for clarity. We had also prepared a Figure supplement that addresses this issue, but by mistake this was not include (although it was available at BioRXIV). We have now corrected this mistake and included this supplement in our new submission. As outlined in the figure supplement, we have verified that the discrepancy is not related to bleaching or quenching. As detailed above, we also do not think that any of the reviewer’s listed concerns (the offset of the autocorrelation curve at long correlations times, the big fluctuations in the region of short correlation times or the time intervals used for FCS measurements) is responsible for this discrepancy. However, if the reviewer would have further suggestions, we would be happy to look into these, as we currently do not have an explanation ourselves.

In summary, we appreciate the critical feedback from the reviewer on this challenging biophysical technique and we hope to have alleviated their concerns.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Essential Revisions:

1) For free β-catenin, the authors have calculated the theoretical diffusion coefficient based on the measured diffusion coefficient of SGFP2 and the size difference between them. But it is possible that β-catenin will interact with other proteins in a specific or non-specific manner, which will make the β-catenin diffuse substantially slower than SGFP2. The free β-catenin diffusion coefficient will affect the fitted value of the slow diffusion component. I wonder if the authors could fit the autocorrelation curve directly with 2 component model without assuming a free β-catenin diffusion coefficient. On the other hand, if a predetermined free β-catenin diffusion coefficient is preferred to fit the autocorrelation curves, maybe it is better to make a ΔARM-β-catenin and use the diffusion coefficient of that protein as the diffusion coefficient of the fast component, or measure the diffusion coefficient of SGFP2-β-catenin directly in the cell extract. In addition, some of the free receptors have a surface diffusion coefficient in the range of 0.1 um2/s. Have the authors considered the possibility that the β-catenin complexes bind some large structure inside the cell, like ER membrane, showing a very slow diffusion coefficient?

This comment concerns two different points, which we will therefore address separately:

1) the free CTNNB1 diffusion coefficient and (2) the diffusion coefficient of the slower CTNNB1 pool.

1. Indeed, we made multiple assumptions to calculate the theoretical diffusion coefficient of free SGFP2-CTNNB1 (as detailed in lines 1097-1103 of the Materials and methods: “As the temperature (T), dynamic viscosity (η) and Boltzmann’s constant (kB) are equal between SGFP2 and SGFP2‑CTNNB1 measurements, the expected difference in diffusion speed is only caused by the radius (r) of the diffusing molecule assuming a spherical protein.”, already present in the previous version). In reality, some of these theoretical assumptions (such as a perfect spherical nature of the particle) will not be met completely. This is an inherent limitation of the approach (and, perhaps, quantitative biology in general). Therefore, the question as to what will happen if we do not fix the first (fast) component is valid. In fact, it had already occurred to us while we were performing our measurements. At the time, we noticed that fitting our two-component model without a fixed first component resulted in similar diffusion coefficients for the first (fast) component, but we had not formally analyzed this. As suggested by the reviewer, we have now reanalyzed the data from Figures 4 and 5. Specifically, we have fitted these measurements with a two-diffusion component model in which both the first diffusion coefficient and the second diffusion coefficient were unfixed. The first diffusion coefficient fitted by this model is highly similar (median 14.8 μm²/s) to the fixed value we determined theoretically for free SGFP2-CTNNB1 (14.9 μm²/s).

This analysis is included in the revised manuscript (new Figure 4 supplement 1).

As for the reviewer’s concern that “The free β-catenin diffusion coefficient will affect the fitted value of the slow diffusion component”, we have verified that the other parameters (concentration, second diffusion coefficient and slow fraction) are also not substantially different (nor any better) from those obtained using the model with the fixed first component (Author response image 10, top row shows the results using the unfixed two-component fitting model and bottom row shows the results with our first-fixed two-component fitting model). Therefore, our previous interpretations and conclusions remain unaltered.

Author response image 10

Download asset Open asset

SGFP2-CTNNB1 parameters fitted by a two-diffusion component model where T1 is free (A-C) or fixed at 14.9 um2/s (D-F).

(A and D) Concentration of SGFP2-CTNNB1 (B and E) Fraction of slow SGFP2-CTNNB1 (C and F) Speed of slow SGFP2-CTNNB1.

We do want to point out that every degree of freedom in the fitting model, will increase the number of possible solutions that can fit the raw curve, leading to less well-defined parameters, especially in more noisy data. Therefore, we still prefer to fix the first component and we think this is justified as we have now shown that the data do not change.

We appreciate the additional suggestions from the reviewer to determine the free diffusion coefficient of monomeric SGFP2-CTNNB1. With both suggested alternatives, CTNNB1 could retain some interactions with other proteins, potentially generating multiple diffusing species (i.e. monomeric and complexes). In addition, the cellular context (i.e. cytoskeletal meshwork and crowding) might be quite different in a cellular extract, which would result in significant differences in the free diffusion coefficient measured ex cellulo as opposed to in cellulo. It is our honest opinion that none of these suggested alternatives is necessarily better, and each control will have different pros and cons. We chose our current control (SGFP2 transfected in HAP1 cells) because it allows us to measure the same fluorophore, which is by definition monomeric, in the same cellular (i.e. biologically relevant) context as our SGFP2-CTNNB1.

Taking everything together, we think the solid fit of our model to our data (Figure 4), and our reanalysis showing similar results when using unbiased fitting (new Figure 4 supplement 1) support the fact that our use of the estimated theoretical diffusion coefficient for free monomeric SGFP2-CTNNB1 is a valid simplification.

This point is also addressed in the main text of the revised manuscript, which now reads as follows:

Line 263-272: “We therefore tested the fit of a two-component model. To this end, we deduced the theoretical diffusion speed of monomeric, unbound SGFP2‑CTNNB1 to be 14.9 µm2/s. This theoretical speed was confirmed by fitting an unbiased two-component model to our experimental data (Figure 4 supplement 1). To limit variability due to noise in the measurements, we proceeded with the two component model in which the first diffusion component was fixed to the theoretically determined diffusion speed of monomeric SGFP2‑CTNNB1 (14.9 µm2/s) and with the second diffusion component limited to slower speeds compatible with point-FCS imaging (see Materials and methods for details). This model provided good fits for our autocorrelation curves obtained in both cytoplasmic and nuclear point FCS measurements (Figure 4G).”

Line 1119-1126: “To validate the obtained first diffusion coefficient of 14.9 µm2s^-1 for SGFP2‑CTNNB1, the data were tested with an unfixed two-component model where both the first and the second diffusion coefficient were fitted (shown in Figure 4 supplement 1). The following limits were set; G∞ [0.5-1.5], N [0.001, 500], τtrip [1*10-6-0.05 ms], τdiff1 [0.5-10 ms], τdiff2 [10-150 ms]. This resulted in a median diffusion time for the first component of 14.8 µm2s^-1 (Figure 4 supplement 1), which was in line with our calculated diffusion coefficient of 14.9 µm2s^-1. All analyses were performed with the two-component model with the fixed first component to reduce variability.”

2. Regarding the speed of the slower SGFP2-CTNNB1 pool: the reviewer is correct that our slow diffusion coefficient is not far removed from those measured for certain membrane associated proteins. Typically, photobleaching is a significant issue in point FCS for proteins stably located at membrane, which we did not observe in our measurements (see lines 1062-1063 and 1075-1076 for how we control for this, already present in the previous version). That being said, we did consider the possibility that the slow diffusion coefficient could be explained by dynamic association with cellular structures. In fact, in our first submitted version of this manuscript (28-05-2020-RA-BX-eLife-59433), we actually addressed this issue specifically, but in streamlining the manuscript for our new submission (11-01-2021-RA-eLife-66440), this detail got lost. In the current revision, we have reintroduced this possibility in the section on “Challenges and opportunities for fluorescence fluctuation spectroscopy techniques”. (Line 818-828: “Moreover, the measured diffusion coefficients do not reveal the identity of the complexes. Previous studies have shown that a significant pool of CTNNB1 is associated with destruction complex components in presence and absence of WNT signaling (Gerlach et al., 2014; Kitazawa et al., 2017; Li et al., 2012), and it is therefore likely that at least part of the slow fraction of CTNNB1 we measure does indeed represent this destruction complex bound pool. As we discuss above, association and dissociation of the destruction complex with the proteasome offers one potential explanation for the different diffusion coefficients measured in the cytoplasm in WNT ‘ON’ and ‘OFF’ conditions. However, in both conditions other processes, such as transient association with intracellular structures (e.g. vesicular membranes or cytoskeletal components), could contribute to the diffusion coefficients we observe.)”

If the reviewer has any additional concerns or knowledge specifically concerning ER-binding, of which we are not currently aware, we would be happy to include any literature that has described this.

2) The model that the author proposed has some logical flaws. In line 1042, equation (9), the authors write the equation w(k4 χ2 – k5 x4) as the mathematical form of Wnt activation, where w is applied to both k4 χ2 term and the k5 x4 term. The χ2 term represents the active destruction complex, and the x4 term represents the inactive destruction complex. This means Wnt will regulate both forms of the destruction complex the same way. And the ratio of χ2/x4 will not change after Wnt pathway activation. This way to model is logically flawed. The authors do show a response of the model after Wnt treatment, but this is because the destruction complex also has a substrate-bound fraction: x3. This fraction is not controlled directly by Wnt input.

We respectfully disagree that our model contains some logical flaws and we would like to refute this remark as follows:

The reviewer correctly notes that in equation (9)

$\frac{d x_{4}}{dt} = w (k_{4} x_{2} - k_{5} x_{4})$ the parameter w multiplies both the terms $k_{4} x_{2}$ (representing the free/active destruction complex, DC) and $k_{5} x_{4}$ (representing the inactivated destruction complex, DC*). Note that these two terms are also present in equation (7), which describes the change in the concentration of DC (χ2).

The reviewer is not correct, however, that the ratio of χ2/x₄ will not change after Wnt pathway activation, because both x₄ and w itself are equal to zero in the absence of WNT (as was already mentioned in the methods section, paragraph ‘model description’). We suspect that this misunderstanding stems from the fact that in our previous submission we did not place sufficient emphasis on the fact that the term w only comes into play when the WNT pathway is switched on and we apologize for any confusion this may have caused. To remedy this, we now make explicit mention of the term w in Table 4, Table 5 and Figure 6 in addition to the methods section (lines 1172-1174). It is also addressed in the main text as follows:

Line 409-411: “When WNT is present in the system, we describe the inactivation of the destruction complex (‘DC*’) by DVL through the parameter $w$ (see Materials and methods section model description).”

Put differently, our minimal model assumes that WNT acts by reducing the concentration of the active destruction complex (DC in Figure 6A, χ2 in Table 4). We do this by opening up a pathway through which the active destruction complex is removed only in the presence of WNT (WNT OFF, w=0; WNT ON, w=1). In the model, this pathway is represented by equation (9), which describes the increase in the concentration of the inactivated destruction complex (DC* in Figure 6A, x₄ in Table 4) and through the same terms in equation (7), which describes the reduction in the concentration of the active destruction complex (χ2). The term k₄x₂ thus describes the appearance of inactivated destruction complex (DC*, x₄) at the cost of the free destruction complex (DC, χ2).

Regarding x₃, the reviewer correctly notes that in our model WNT does not directly control destruction complex bound to CTNNB1 (CB*-DC in Figure 6A, x₃ in Table 4). A similar indirect mode of action of WNT treatment is described in the much more elaborate model from the Kirschner group (Lee et al., 2003), where the action of WNT is simulated by the formation of activated Dishevelled, which reduces the concentration of the active form of the destruction complex, resulting in an increase in CTNNB1 levels. Thus, in both our model and in the model from the Kirschner group, the concentration of the free/active destruction complex is reduced, thereby increasing CTNNB1 levels.

It is important to note that the removal of free/active destruction complex (i.e. an increase in DC* at the expense of DC) in and by itself does not necessarily lead to an increase in CTNNB1. In our model, and also indirectly in the Kirschner group model, this increase is a consequence of (i) the constant rate at which the destruction complex is released upon CTNNB1 degradation (i.e. k₃ is not affected by WNT) and (ii) our experimental observation that the DC-bound phosphorylated CTNNB1 (CB*-DB in Figure 6A, x₃ in Table 4) has the same concentration in the absence and presence of WNT 62.5 nM, as per the median concentration of the slow component in our FCS measurements

3) It's better to use SGFP2 dimer and trimer as control instead of EGFP, in Figure 5H and 7E/F. The dimer and trimer constructs are important control for the average size and the percentage of bright fluorescent proteins will vary for different fluorescent proteins. I have no idea how SGFP2 compares to EGFP.

We understand this point from the reviewer’s perspective. We insufficiently realized that the similarities between SGFP2 and EGFP would not be obvious to all and we apologize for not having been more explicit about the properties of SGFP2. As we will explain below, it is highly comparable to EGFP and therefore the use of EGFP controls is justified.

The superGFP2 (SGFP2) variant was developed by colleagues in our institute, who are experts in fluorescent protein biology. It is based on the EGFP template but has been optimized for maturation in mammalian cells (Kremers et al., 2007). SGFP2 and EGFP have the exact same size (239 amino acids) and only differ by 4 amino acids (Author response image 11A). As described by Kremers et al., the spectral and biochemical characteristics of EGFP and SGFP2 are highly similar with a slightly lower extinction coefficient (-27%) and a slightly higher quantum yield (+7%) for SGFP2. When other factors are also taken into consideration (including the improved maturation and a lower fraction of dark state kinetics for SGFP2) this results in a slightly lower molecular brightness (-5%) of monomeric SGFP2 in comparison to monomeric EGFP (as can be judged from Figure 5H and Figure 7E/F).

To further illustrate this, we have performed additional NandB measurements of the two monomeric proteins in HeLa cells (Author response image 11B). These data confirm that for cellular measurements any differences between EGFP and SGFP2 are well within the 95% confidence range and thus too small to have any significant impact on our measurements and interpretations. Based on the above, we believe our controls in Figure 5H and 7E/F are beyond dispute.

Author response image 11

Download asset Open asset

Differences between EGFP and SGFP2.

(A) protein alignment of EGFP (top) and SGFP2 (bottom) showing the 4 amino acid difference. (B) Independent measurements in an additional mammalian cell line of monomeric EGFP and SGFP2 brightness. The same control constructs, measured in HAP1 cells, were already included in figures 5 and 7. The molecular brightness of sGFP2 is slightly lower, but well within the 95% confidence interval of the biological measurements. Any differences between the proteins are thus too small to be significant.

Please also note that because the maturation of SGFP2 is improved compared to EGFP (which was an important reason behind our choice to tag CTNNB1 with SGFP2 in the first place), this should only aid the detection of SGFP2 dimers, trimers or higher order oligomers in our SGFP2-CTNNB1 cells. Put differently: It is highly unlikely that we miss them. The fact that we only detect an increase in molecular brightness in the S45F mutant cell line, thus further strengthens our conclusion that few if any cytoplasmic complexes contain multiple CGFP2-CTNNB1 molecules under physiological conditions.

We have added the following text to the Materials and methods section to explain our choice for SGFP2 considering its fluorescent properties and the relation between EGFP and SGFP2 in the NandB controls:

Line 886-887: “SGFP2 was chosen for its favorable brightness, maturation and photo-stability (Kremers et al., 2007).”

Line 1130-1136: “SGFP2 and EGFP are highly similar in sequence (with only 4 amino acid changes) and in spectral and biochemical characteristics. SGFP2 has a slightly higher quantum yield (+7%), lower extinction coefficient (-27%), and enhanced protein expression and maturation compared to EGFP (Kremers et al., 2007). The resulting brightness of monomeric SGFP2 in comparison to monomeric EGFP is slightly lower (-5%). In cellular measurements this difference is within the biological and technical variation and therefore SGFP2 and EGFP controls can be considered synonymous in these experiments.”

4) The data on the concentrations of β-catenin in the cytoplasm and nucleus before and after Wnt signaling and after inhibition of GSK3 are a major advance. However, I share the opinion of previous reviewer 2 (comment 1) that too much attention is paid to relative differences and not enough to absolute differences. There is strong evidence in the literature that one function of the destruction complex is to retain β-catenin in the cytoplasm (e.g. https://pubmed.ncbi.nlm.nih.gov/21471006/), and thus reduce the amount that can enter the nucleus. Their quantitative imaging (e.g. Figure 3A) reveals this quite clearly and combined with their FCS data suggest that the cytoplasmic pool includes this "bound" β-catenin and that this bound pool remains after Wnt stimulation. I would suggest making this clearer and putting it in the context of earlier work on cytoplasmic retention.

As was already mentioned in the manuscript, we use our measurements and absolute differences to determine the appropriate biophysical parameters for our computational model.

We have now further emphasized this point in the discussion:

Line 602-603: “Importantly, this allowed us to measure hitherto unknown biophysical parameters of WNT/CTNNB1 in individual living human cells for the first time.”

We also thank the reviewer for their additional suggestions to incorporate these absolute values in different contexts, which we have done in our revision throughout:

First, we realize that we did not include absolute numbers and values for our S45F mutant experiments to the same extent as for wildtype CTNNB1 in the presence and absence of WNT. This has been remedied in the Results section as follows:

For the differences between physiological (WNT3A) and oncogenic (S45F) WNT/CTNNB1 signaling, we have more clearly emphasized the absolute concentrations in the Results section. Lines “514-525: With FCS and NandB we quantified the accumulation of CTNNB1 levels of mutant SGFP2‑CTNNB1^S45F (Figure 8B, Figure 8 supplement 2A) and wild-type SGFP2‑CTNNB1 upon CHIR99021 treatment (Figure 8 supplement 2B-C). Both exceeded the levels observed with physiological WNT3A stimulation (Figure 3A-C, Figure 5A). Specifically, the absolute concentration of SGFP2‑CTNNB1^S45F in the cytoplasm (median 351 nM, 95% CI 276-412) exceeded that of SGFP2‑CTNNB1 in WNT3A treated cells (median 221 nM, 95% CI 144-250). In the nucleus, the concentration of SGFP2‑CTNNB1^S45F reached 429 nM (median, 95% CI 387-481), as opposed to 240 nM (median, 95% CI 217-325) for SGFP2‑CTNNB1 in WNT3A treated cells as a result of losing its priming phosphorylation site. While this further increase in concentration is evident, it should be noted that in both the cytoplasm and the nucleus CTNNB1 levels thus rise less than 2-fold in an oncogenic setting compared to WNT3A treatment.”

For additional parameters we have similarly added more quantitative description of the data in the main text. For example, Line 535-538: “The fraction of bound SGFP2‑CTNNB1 in the nucleus was comparable between our HAP1^{SGFP2‑CTNNB1(S45F)} mutant cell line (median 0.29, 95% CI 0.27-0.33) (Figure 8E), 8 μM CHIR99021 (median 0.38, 95% CI 0.29-0.46) (Figure 8F) and WNT3A (median 0.32, 95% CI 0.30-0.34) (Figure 5D) treated wild-type HAP1^{SGFP2‑CTNNB1} cells.”

Second, in the discussion we put our FCS measurements in proper context of the existing literature on cytoplasmic retention:

Lines 668-677: “The substantial fraction of slow diffusing CTNNB1 that remains upon physiological and oncogenic stimulation of the pathway, is also consistent with the previously proposed role for cytoplasmic retention by the destruction complex (Krieghoff et al., 2006; Roberts et al., 2011; Yamulla et al., 2014). These studies also show that both nuclear and cytoplasmic retention have an important role in determining the subcellular distribution of CTNNB1. Moreover, cytoplasmic retention of CTNNB1 can contribute to reducing downstream pathway activation even in the presence of oncogenic APC mutations (Kohler et al., 2008; Li et al., 2012; Schneikert et al., 2007; Yang et al., 2006), which further highlights the importance of this process.”

5) One model currently in the field is that the key regulated step in β-catenin destruction is not its phosphorylation but instead its transfer to the E3 ligase, an event that may be controlled by GSK3 phosphorylation of APC (e.g. https://pubmed.ncbi.nlm.nih.gov/22682247/;. https://pubmed.ncbi.nlm.nih.gov/26393419/; https://pubmed.ncbi.nlm.nih.gov/32129710/). In talking about this, the authors state: "Our data clearly show that a substantial fraction of CTNNB1 in the cytoplasm remains bound upon pathway stimulation (Figure 5D). This is not predicted by any of the above-mentioned models and challenges the long-held view that mainly monomeric CTNNB1 accumulates." However, in my mind the retention of high MW complexes after Wnt signaling seems like it is consistent with the model in the manuscripts cited above. Further, while model cartoons (I use them myself) indicate accumulation of monomeric β-catenin in the cytoplasm, in fact what seems more likely is that monomeric β-catenin that accumulates after Wnt signaling rapidly enters the nucleus and is "trapped" as part of high MW transcriptional complexes like those their data appear to document, and thus monomeric cytoplasmic levels may be relatively low. I think a more nuanced description of these data would be useful.

We have rewritten this part of the discussion to be a bit more nuanced as per the reviewer’s suggestion. Lines 663-668: “Our data clearly show that a substantial fraction of CTNNB1 in the cytoplasm remains bound upon pathway stimulation (Figure 5D). This is not captured by textbook models and cartoons that typically still depict the view that mainly monomeric CTNNB1 accumulates. It is, however, consistent with the notion of the destruction complex remaining intact and transfer to the E3 ligase, rather than phosphorylation and release of CTNNB1 itself, being the key controlled step (Li et al., 2012; Pronobis et al., 2015; Schaefer et al., 2020).”

6) One major controversy in the field is the nature of the active destruction complex and its multimeric state. Some data (e.g. https://pubmed.ncbi.nlm.nih.gov/26124443/; https://pubmed.ncbi.nlm.nih.gov/26393419/; https://pubmed.ncbi.nlm.nih.gov/29641560/) suggest large multimeric assemblies while in vitro studies have suggested much smaller assemblies (e.g. https://pubmed.ncbi.nlm.nih.gov/32297861/). It was quite surprising that the authors did not review this literature-in fact the latter paper isn't even in the reference list. Their data, while not definitive, certainly has something to say about this discussion. That having been said, however, I do NOT agree with Reviewer 1 that extending this work to visualize tagged destruction complex components is essential. Of course, this is a very important next step, but this manuscript is already rich in novel data. The reduction in "size" of the potential destruction complexes after Wnt treatment is also consistent with recent observations in Drosophila (https://pubmed.ncbi.nlm.nih.gov/29641560/)

We thank the reviewers for their support of our work. We too look forward to future follow up studies and extension of this work by ourselves and others. Indeed, the multimerization state of the active destruction complex is one of the main outstanding questions in the field. We have included the suggested references in our revised discussion to better reflect the current state of the literature:

Line 678-686: “We show that the cytoplasmic CTNNB1 complex in WNT3A or CHIR99021 treated cells as well as in S45F mutant cells has an over 3-fold increased mobility compared to control cells. Therefore, while the diffusion coefficient is still very low (indicating that the remaining complex is still very large), this implies it is a vastly different complex than that observed in the absence of WNT stimulation. The precise nature of these complexes remains unknown, but could be consistent with a reduced destruction complex size after WNT treatment, as also recently observed in Drosophila for AXIN complexes (Schaefer et al., 2018), or with the formation of inactivated destruction complexes (‘transducer complexes’) in response to WNT/CTNNB1 pathway activation (Hagemann et al., 2014; Lybrand et al., 2019).”

Line 631-641 “In addition, our NandB data indicate that most of the slow diffusing CTNNB1 complexes contain one or very few SGFP2‑CTNNB1 molecules in either the absence or presence of WNT3A stimulation. In view of the above-mentioned destruction complex oligomerization and its presumed multivalency this finding was quite unexpected. Several mechanisms could explain this apparent discrepancy. On the one hand, destruction complex multimerization at endogenous levels might be more subtle than previously thought. For example, quantification of AXIN polymerization in vitro showed that even at exceedingly high concentration (24μM), AXIN polymers typically contained only 8 molecules (Kan et al., 2020). On the other hand, even if the multivalent destruction complex offers multiple CTNNB1 binding sites, occupancy at any one time might be low, due to the continuous and high turnover of CTNNB1.”

7) I think Reviewer 2 comment 2 (previous submission) and the authors response should be more clearly laid out in the manuscript. While many of us said for years that Wnt signaling "turns the destruction complex off", it is now abundantly clear that it only "turns the destruction complex down". For example, in both Drosophila and in cultured mammalian cells complete loss of GSK3 or APC leads to much higher levels of β-catenin than are seen after Wnt signaling.

We appreciate this suggestion and we have added a section in the discussion (Differences and similarities in physiological and oncogenic WNT signaling) to specifically address the differences between oncogenic and physiological signaling.

Line 762-78 “As discussed above, several behaviors of CTNNB1 are conserved between the different modes of stimulation. For instance, WNT3A treatment, GSK3 inhibition and the oncogenic S45F mutation all result in (1) increased overall levels of CTNNB1 (2) a substantial fraction of CTNNB1 in a faster, albeit still very large complex in the cytoplasm (3) increased nuclear accumulation of CTNNB1 and (4) increased retention of CTNNB1 in nuclear complexes. Our computational modelling further confirms that in addition to regulation of CTNNB1 turnover – either by removal of activated destruction complex or through inhibition of phosphorylation and ubiquitination -, nuclear shuttling and nuclear retention are equally important regulatory nodes in oncogenic (CHIR99021 treatment or S45F mutation) and physiological (WNT3A stimulation) signaling.

However, the absolute levels of CTNNB1 and the resulting transcriptional activation are distinct in each of these conditions: Cells treated with a GSK3 inhibitor continue to accumulate CTNNB1 after 4 hours, when WNT3A treated cells reach a plateau. The latter is likely due to the fact that negative feedback mechanisms kick in, such as reconstitution of the destruction complex by AXIN2 or internalization of WNT-bound receptor complexes (Agajanian et al., 2019; Lustig et al., 2002), both of which function upstream of GSK3. Alternatively, it could reflect the notion that physiological WNT signaling does not turn the destruction complex off completely, but rather “turns it down”, as our NandB data support the fact that under physiological conditions the destruction complex itself provides a surplus reservoir of CTNNB1 binding sites that may only become occupied when WNT signaling is hyperactivated. As a combined result, GSK3 inhibition or S45F mutation of CTNNB1 can result in higher total intracellular levels of CTNNB1. Indeed, concentrations of SGFP2‑CTNNB1 in S45F mutated cells exceed those in WNT3A treated cells in both the cytoplasm (1.6-fold) and the nucleus (1.8-fold). This subtle increase in CTNNB1 levels is likely amplified at the transcriptional level (Jacobsen et al., 2016), consistent with the well-known fact that constitutive activation of the pathway through different mechanisms, including APC mutation, results in higher pathway activation than physiological stimuli.”

8) The fact that their Number and Brightness data suggest few destruction complexes bind multiple β-catenins is very surprising. It doesn't fit my preconceptions and that is OK! I like the idea that usually throughput is rapid enough that multiple β-catenins don't accumulate (waterwheel model). It is surprising that Wnt signaling doesn't alter this but those are the data.

We appreciate this positive feedback and we still share the reviewers’ surprise. We appreciate their support for our waterwheel model and have incorporated this idea more clearly in the main text.

Line 639-645 “On the other hand, even if the multivalent destruction complex offers multiple CTNNB1 binding sites, occupancy at any one time might be low, due to the continuous and high turnover of CTNNB1. In this respect, the CTNNB1 bindings sites in the destruction complex could be envisioned to act similar to the wooden vanes in the paddle wheel of an old-fashioned watermill: like the water in the analogous example, CTNNB1 would be continuously scooped up (for phosphorylation) and dropped off (for degradation).”

https://doi.org/10.7554/eLife.66440.sa2

Article and author information

Author details

Saskia MA de Man

Developmental, Stem Cell and Cancer Biology, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands

Contribution
Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Writing - original draft, Project administration

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-0906-5276
Gooitzen Zwanenburg

Biosystems Data Analysis, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands

Contribution
Conceptualization, Data curation, Software, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

For correspondence
Gooitzen.zwanenburg@gmail.com

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-2745-0285
Tanne van der Wal

Developmental, Stem Cell and Cancer Biology, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands

Contribution
Investigation, Project administration, Writing - review and editing

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-3042-6948
Mark A Hink
1. Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands
2. van Leeuwenhoek Centre for Advanced Microscopy, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands
Contribution
Resources, Software, Supervision, Validation, Writing - review and editing

Contributed equally with
Renée van Amerongen

For correspondence
m.a.hink@uva.nl

Competing interests
No competing interests declared
Renée van Amerongen

Developmental, Stem Cell and Cancer Biology, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, Netherlands

Contribution
Conceptualization, Software, Supervision, Funding acquisition, Validation, Methodology, Writing - original draft, Project administration, Writing - review and editing

Contributed equally with
Mark A Hink

For correspondence
r.vanamerongen@uva.nl

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-8808-2092

Funding

University of Amsterdam (MacGillavry fellowship)

R van Amerongen

KWF Kankerbestrijding (ANW 2013-6057)

R van Amerongen

KWF Kankerbestrijding (2015-8014)

R van Amerongen

Nederlandse Organisatie voor Wetenschappelijk Onderzoek (VIDI 864.13.002)

R van Amerongen

Nederlandse Organisatie voor Wetenschappelijk Onderzoek (OCENW.KLEIN.169)

R van Amerongen

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank the van Leeuwenhoek Centre for Advanced Microscopy (LCAM, Section Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam) for the use of their facilities and LCAM staff for sharing their expertise and providing technical support, Gonzalo Congrains Sotomayor for support with FACS sorting, Jasmijn Span for cloning the color variants of the repair plasmid that have been deposited at Addgene and Sanne Lith for gRNA cloning as students in our lab, Marten Postma and Joachim Goedhart for assistance with data handling and analysis, Dorus Gadella for carefully reading the manuscript, and all colleagues for stimulating discussions and suggestions.

Senior Editor

Aleksandra M Walczak, École Normale Supérieure, France

Reviewing Editor

Felix Campelo, The Barcelona Institute of Science and Technology, Spain

Reviewer

Mark Peifer, University of North Carolina at Chapel Hill, United States

Publication history

Received: January 11, 2021
Accepted: June 29, 2021
Accepted Manuscript published: June 30, 2021 (version 1)
Version of Record published: August 5, 2021 (version 2)

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

3,059

Page views
381

Downloads
14

Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Mendeley

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

Saskia MA de Man
Gooitzen Zwanenburg
Tanne van der Wal
Mark A Hink
Renée van Amerongen

(2021)

Quantitative live-cell imaging and computational modeling shed new light on endogenous WNT/CTNNB1 signaling dynamics

eLife 10:e66440.

https://doi.org/10.7554/eLife.66440

Categories and tags

Research organism

Human

Share this article

Cite this article

Generation of HAP1SGFP2-CTNNB1 cell lines.

Functional validation of three independent HAP1SGFP2-CTNNB1 clones.

Figure 2—source data 1

Live imaging of HAP1SGFP2-CTNNB1.

Representative video of confocal time-lapse experiments, showing SGFP2-CTNNB1 (left, green), SiR-DNA staining (middle, magenta), and transmission image (right, gray) after treatment with vehicle control (BSA).

Representative video of confocal time-lapse experiments, showing SGFP2-CTNNB1 (left, green), SiR-DNA staining (middle, magenta), and transmission image (right, gray) after treatment with 100 ng/ml WNT3A.

Representative video of confocal time-lapse experiments, showing SGFP2-CTNNB1 (left, green), SiR-DNA staining (middle, magenta), and transmission image (right, gray) after treatment 8 µM CHIR99021.

Video showing the quantification of the normalized intensity of SGFP2-CTNNB1 in the cytoplasm of time-lapse microscopy series (from Figure 4 and Videos 1–3) at each time point showing all individual cells from three biological experiments.

Video showing the quantification of of the normalized intensity of SGFP2-CTNNB1 in the nucleus of time-lapse microscopy series (from Figure 4 and Videos 1–3) at each time point showing all individual cells from three biological experiments.

Video showing the quantification of the nuclear-cytoplasmic ratio of SGFP2-CTNNB1, calculated from raw intensity values underlying Videos 4 and 5.

Two diffusion-component fit-model for SGFP2-CTNNB1 FCS measurements.

Figure 4—source data 1

Abundance and mobility of SGFP2-CTNNB1 molecules in living cells after 4 hr WNT3A treatment or control.

Figure 5—source data 1

Total number of SGFP2-CTNNB1 molecules in the confocal volume and corresponding calculated concentrations obtained from FCS measurements in n=3 independent experiments.

Number of SGFP2-CTNNB1 molecules in the confocal volume and corresponding calculated concentration of SGFP2-CTNNB1 molecules with the fast or slow diffusion coefficient obtained from FCS measurements in n=3 independent experiments.

Brightness of SGFP2-compared relative to EGFP-monomer and -dimer controls in n=2 independent experiments. N is the number of analyzed cells.

Computational model of WNT/CTNNB1 based on FCS concentrations for free and complexed CTNNB1 (Tables 1–2).

Variables Minimal Model of WNT signaling.

Equilibrium conditions for the Minimal Model of WNT signaling.

Cytoplasmic complex characteristics in absence of SGFP2-CTNNB1 N-terminal phosphorylation.

In silico and experimental perturbation of WNT signaling.

primers/oligonucleotides used in this study.

Boxplots and individual datapoints from a single biological experiment (that is part of Figure 3 previously Figure 4) at 0 and 4 hours after treatment.

Cells were divided into 3 size categories (small: <500 pixels, middle: 500-1500 pixels, large: >1500 pixels).

These images show a double-tagged line that has endogenously tagged SGFP2-CTNNB1 (left panel, green) and endogenously tagged mScI-GSK3B (middle panel, magenta).

Figure displaying the relation between the particle number (x-axis) and the normalized brightness (y-axis).

This table describes different scenarios of how different brightness species would impact the average apparent brightness.

Waterwheel model of CTNNB1 turnover.

Comparison of old model without steep (A,C,E) and new model with steep (B,D,F).

These images show a double-tagged line that has endogenously tagged SGFP2CTNNB1 (left panel, green) and endogenously tagged mScI-GSK3B (middle panel, magenta).

Autocorrelation curves from the same experiment with a small offset of 0.

Comparison of analysis of short and long FCS traces.

SGFP2-CTNNB1 parameters fitted by a two-diffusion component model where T1 is free (A-C) or fixed at 14.9 um2/s (D-F).

Differences between EGFP and SGFP2.

Author details

Saskia MA de Man

Contribution

Competing interests

Gooitzen Zwanenburg

Contribution

For correspondence

Competing interests

Tanne van der Wal

Contribution

Competing interests

Mark A Hink

Contribution

Contributed equally with

For correspondence

Competing interests

Renée van Amerongen

Contribution

Contributed equally with

For correspondence

Competing interests

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

Categories and tags

Research organism

Further reading

Generation of HAP1^SGFP2-CTNNB1 cell lines.

Functional validation of three independent HAP1^SGFP2-CTNNB1 clones.