In vivo targeted and deterministic single-cell malignant transformation
- Laboratoire de Physique de l’Ecole Normale Supérieure LPENS, ENS, PSL Research University, CNRS, Sorbonne Université, Université de Paris, France
- Inovarion, France
- High Throughput qPCR Core Facility of the ENS, Ecole Normale Supérieure, PSL Research University, IBENS, France
- Dept. Chemistry and Biochemistry, UCLA, United States
Figures
Figure 1 with 6 supplements
Optogenetic setup for single-cell induction.
(A) Photocontrol of a protein fused to an estrogen receptor (ERT) is achieved by releasing the protein from its complex with cytoplasmic chaperones (CC), upon uncaging of caged cyclofen (cCYC). (B) The transgenic zebrafish line engineered to express an oncogene (KRASG12V) upon photoactivation of a CRE-recombinase fused to ERT (CRE-ERT) (as shown in A). (C) The mRNAs of Ventx-GR and mRFP (used as a marker) are injected at the one-cell stage. (D) At 1 day post-fertilization (dpf) the embryos are mounted in channels in an agarose gel and incubated for 45 min in cCYC on a microscope stage. (E) They are washed and illuminated at 405 nm on a microscope to uncage cCYC close to the otic vesicle. A diaphragm (DIA) defines an illumination zone of ~80 μm diameter (see G). Excitation (EX), dichroic mirrors (DM), and emission (EM) filters allow for visualization of Eos and mTFP. The cell in which KRASG12V has been induced is observed within ~1 hr by the fluorescence of mTFP (see G). (F) The embryos are transferred into individual wells, incubated overnight in dexamethasone (DEX), washed at 1 day post-induction (dpi) and monitored over the next 5 days. (G) Lateral view of zebrafish at ~1 hr post-activation displays a single induced cell (top: blue spot shown by arrowhead in mTFP channel) in the illumination region (middle: Eos channel) in the vicinity of the otic vesicle (white arrow) and bottom: merger of both channels. Body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are shown.
Figure 1—figure supplement 1
Characterization of the double transgenic line Tg(β-actin:loxP-EOS-stop-loxP-KRASG12V-T2A-H2B-mTFP; ubi:Cre-ERT; myl7:EGFP).
(A) A transgenic fish at 6 day post-fertilization (dpf) displays green EOS fluorescence throughout the body and a strong green fluorescence in the heart due to the additive expression of EGFP under a heart-specific (myl7) promotor, red arrowhead. (B) Upon global illumination (at 1 dpf) with an ~365 nm UV lamp in presence of caged cyclofen (cCYC), cyclofen is uncaged which releases CRE-ERT from its complex with cytoplasmic chaperones and results in floxing of EOS and ubiquitous expression of KRASG12V and H2B-mTFP. The embryos thus display blue fluorescence in the nuclei shown here at 1 hr post-induction (hpi). (C) Stable and ubiquitous expression of the blue H2B-mTFP fluorescent protein in zebrafish late embryos (3 dpf) and larvae (6 dpf). Note that in all pictures the asterisk (*) indicates the eye. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are shown.
Figure 1—figure supplement 2
Transient activation of Ventx reprogramming factor.
(A) Top: At one-cell stage Ventx-GR mRNA is injected together with mRFP mRNA (used as a tracer), which expression throughout the embryo (here shown at 1 day post-fertilization [dpf]) is a proxy for Ventx-GR expression. (B) Bottom: Immunofluorescence analysis show that Ventx protein can be visualized in fixed embryos with the help of an antibody (Ab) directed against the HA tag linking Ventx and GR (and visualized by a second Ab linked to a green fluorescent marker). Upon addition of dexamethasone (DEX) at 1 dpf embryos, Ventx is released from its complex with cytoplasmic chaperones and diffuses into the cell nucleus, resulting in a pointillist image of the nuclei (middle). At 2 dpf, after washing out DEX, Ventx protein is no longer detected and the pointillist image is lost (right). Note that in all pictures the asterisk (*) indicates the eye. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral; l: left; r: right) are shown.
Figure 1—figure supplement 3
Synergy between a reprogramming factor and KRASG12V oncogene efficiently induces tumors in zebrafish larvae.
(A) Images of control zebrafish larvae at 6 days post-fertilization (dpf): without any treatment (left), with transient incubation in caged cyclofen (cCYC) (without UV, middle) and with incubation in dexamethasone (DEX)+cCYC (without UV, right). Note that zebrafish does show neither morphological anomalies nor mortality. (B) Global photoactivation (via cCYC uncaging with 5 min UV illumination and Cre-ERT activation) of KRASG12V (blue nuclei) in 1 dpf zebrafish without (left) or with subsequent transient activation (right) of Ventx-GR by DEX (described in Figure 2). These larvae (labeled as KR+VX) develop hyperplasic tissues within 6–9 dpf (dorsal view, red arrows indicate hyperplasia). Global photoactivation (via cCYC uncaging and Cre-ERT activation) of KRASG12V (blue nuclei) in 1 dpf zebrafish with subsequent transient activation of NANOG-GR or POU5/OCT4-GR by DEX (setup described in A). These larvae (labeled as KR+NANOG or KR+POU5/OCT4) develop hyperplasic tissues within 6–9 dpf (dorsal view, red arrows indicate hyperplasia). (C) Quantification of zebrafish developing abnormal hyperplasia upon the indicated treatments. Note that only the synergy between reprogramming factors (i.e. VENTX, NANOG, or POU5/OCT4) and the KRASG12V oncogene reproducibly and efficiently induce tumors in zebrafish larvae. The total number of zebrafish larvae analyzed are indicated above the bars. Φ indicates no event (0%). Note that in all pictures the asterisk (*) indicates the eye. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral; l: left; r: right) are shown.
Figure 1—figure supplement 4
Synergy between the Ventx reprogramming factor and the KRASG12V oncogene induces tumors in zebrafish larvae.
(A) Phosphorylated ERK (pERK) immunofluorescence at 6 days post-fertilization (dpf) in nonactivated larvae (Ctrl), in larvea in which only KRASG12V was activated (KR) and in larvae in which both KRASG12V and VENTX were activated globally (KR+VX). Notice in the last one the strong pERK activation in the hyperplasic abdominal outgrowth. The digestive tract is shown between the red arrows. (B) Confocal microscopy of KR+VX zebrafish larva at 6 dpf (ventrolateral view) displays the immunofluorescence of active pERK (in green and indicated by red arrowheads) in the abdomen of a DAPI labeled zebrafish larva. (C) Histopathological analysis by hematoxylin and eosin (H&E) staining of hyperplasic tissues in zebrafish larvae (6 dpf) upon activation of KRASG12V and Ventx (KR+VX) is compared with normal tissues (Ctrl). Larvae overexpressing KR+VX develop hyperplasic and dysplastic cancer-like features and tissue outgrowth in the gut/intestine (red arrow), pancreas (blue arrow), and liver. (D) Such cancer-like features specifically and exclusively reduce the survival rate to 20% 2 weeks after induction.
Figure 1—figure supplement 5
KRASG12V plus Ventx induces a cancer-like gene expression signature.
Both KRASG12V and Ventx-GR (KR+VX) were activated globally in zebrafish at 1 day post-fertilization (dpf). The larvae (n=24) were collected at 6 dpf and processed for reverse transcriptase quantitative PCR (RT-qPCR) analysis. Compared to the controls (Ctrl; Ctrl+ cCYC; Ctrl+cCYC+DEX) and larvae in which only KRASG12V (+cCYC+UV; +cCYC+UV +DEX) or Ventx (Ventx+DEX) were activated, the co-activation of KRASG12V+Ventx (+cCYC+UV+VentX+DEX) displayed significant changes in gene expression, with overexpression of genes involved in pluripotency/reprogramming (nanog, pou5f3/oct4, lin28) and stemness (aldh1a, tert) and underexpression of oncosuppressor genes (foxo3, lats1,spop) and genes involved in epigenetic memory (tet3, dnmt). Raw data can be found in the file: Figure 1—figure supplement 5—source data 1.
-
Figure 1—figure supplement 5—source data 1
Reverse transcriptase quantitative PCR (RT-qPCR) raw data.
mRNA was extracted from whole single (or groups of a few) embryos and analyzed on a Biomark HD system as described in Materials and methods.
- https://cdn.elifesciences.org/articles/97650/elife-97650-fig1-figsupp5-data1-v1.xlsx
-
Figure 1—figure supplement 5—source data 2
Data file of the heatmap shown in Figure 1—figure supplement 5.
- https://cdn.elifesciences.org/articles/97650/elife-97650-fig1-figsupp5-data2-v1.xlsx
Figure 1—figure supplement 6
KRASG12V plus Ventx induces a cancer-like gene expression signature.
Some of the data shown in Figure 5 are here displayed with their error bars. Compared to the controls (Ctrl in light gray; Ctrl+cCYC in gray; Ctrl+cCYC+DEX in dark gray) and larvae in which only KRASG12V (KRASG12V in light blue; KRASG12V+DEX in blue) or Ventx (VENTX in green) were activated, the co-activation of KRASG12V+VENTX (in red) displayed significant changes in gene expression. (A) Pluripotency/reprogramming factors like nanog, pou5f3/oct4, and lin28, as well as stemness markers like aldh1a2 and telomerase (tert) are significantly upregulated (p<0.05, compared to Ctrl) by KR+VX but not by KRASG12V (KR) or Ventx (VX) alone. (B) Conversely, onco-suppressors such as foxo3, lats1, and spop are significantly downregulated (p<0.05) by KR+VX but not by KRASG12V (KR) or Ventx (VX) alone when compared to control. For all qPCR graphs, error bars represent s.e.m. values. For statistical analyses, samples were compared with the respective control by Unpaired Student’s t-test. *p<0.05, **p<0.005. ***p<0.0001.
Figure 2 with 4 supplements
Malignant transformation of a single cell triggering carcinogenesis in vivo.
(A) At 1 day post-fertilization (dpf), a single cell in a zebrafish brain was photoinduced to express the oncogene KRASG12V, identified (white arrowhead) within ~1 hr by the blue fluorescent H2B-mTFP. Membrane-bound mRFP is used as tracer. Transient (24 hr) dexamethasone (DEX) activation of Ventx is done following photoactivation. The probability p of inducing one or more (blue fluorescent) cells is shown on the right panel, together with the Poisson distribution (red curve) expected for the independent induction of k cells (error bars are statistical errors on the mean: , where N is the total number of observed embryos). (B) At 1 day post-induction (1 dpi), the activated cell may have divided (middle panel) giving rise to two mTFP+ (blue fluorescent) cells (white arrowhead) or may not have divided (left panel). The probability of observing k blue fluorescent cells at 1 dpi is shown on the right panel. (C) At 3 dpi the original cell expanded clonally (white arrowheads) by short-range dispersal within the brain. At 4 dpi the brain has been colonized (middle panel) by the progeny of the activated cell that display tumor growth as well as dispersal in the head or entering into the cardiovascular system (red arrowhead). At 5 dpi (right panel), a tumor mass is formed. (D) Confocal microscopy of a larval head displaying tumors (white arrowheads) and dispersal in the trunk (red arrowheads). (E) Histopathological sections of larval brain (dorsal view). Dashed lines (1–5) indicate sections shown in (F). Hematoxylin and eosin (H&E) staining of brain at 5 dpi of KR+VX-induced larva (depigmented) is compared to normal brain (Ctrl, melanocytes in brown). At 5 dpi, the optic tectum (red arrows), the tegmentum (black arrow), and hypothalamus (red arrowhead) are infiltrated by a dysplastic tumor, progeny of the initial induced single cell. An asterisk (*) indicates the eye and a white arrow the otic vesicle. Scale bars and the body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are shown. T=telencephalon; M=mesencephalon; C=cerebellum; Mc =melanocytes (yellow arrows).
Figure 2—figure supplement 1
Photoactivation of one or more cells by caged cyclofen (cCYC) uncaging.
At 1 day post-fertilization (dpf), an area of diameter ~80 μm in the vicinity of the otic vesicle (white arrow) was illuminated for 7 min at 405 nm, uncaging cCYC and resulting in one or a few cells in the illuminated region to express the oncogene kRASG12V and its blue fluorescent marker (H2B-mTFP). In these conditions single-cell activation represents about 50% of the cases (see Figure 2).
Figure 2—figure supplement 2
Single-cell activation of the KRASG12V oncogene is not sufficient to initiate carcinogenesis.
(A) At 1 day post-fertilization (dpf), a single cell in the brain was photo-induced to express the oncogene kRasG12V and identified (white arrowhead) within ~30 min by the blue fluorescence of the expression marker (H2b-mTFP). The otic vesicle (indicated by white arrow) is used as a spatial reference. At 5 days post-induction (5 dpi), the activated cell has disappeared. Note in the graph that, whereas KRASG12V plus VENTX (KR+VX) efficiently induces cancer (100%), KRASG12V alone (KR) is not sufficient (0%, indicated by Φ). (B) Expression of kRasG12V was activated in 1 dpf embryos (without subsequent activation of Ventx). The cells of the larvae were dissociated and isolated cells (kRas expressing, H2b-mTFP+ blue cells) were transplanted (≈ 1 cell per host) at 2 dpf in a Nacre (mitf -/-) zebrafish line for a better tracking of the transplanted cells. The transplanted H2b-mTFP+ blue cell (red arrow) can be visualized as early as 3 hr post-transplantation (3 hpt) in the yolk of the host. At 6 dpt the blue cell has disappeared in host Nacre zebrafish larvae. Note in the graph that, whereas the transplanted cell experiencing KRASG12V plus VENTX (KR+VX) activation efficiently give rise to cancer (60%), KRASG12V alone (KR) is not sufficient to initiate cancer in host zebrafish (0%, indicated by Φ). In all figures, the otic vesicle (white arrow) is indicated as well as the eye (white asterisk: *). Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral; l: left; r: right) are shown.
Figure 2—figure supplement 3
Tracking of the clonal expansion of a photoinduced cell in one embryo over 7 days.
(A) At 1 day post-fertilization (dpf), a single cell (white arrowhead) in the vicinity of the otic vesicle (white arrow) was photo-induced to express the oncogene KRASG12V, identified within ~1 hr (1 hr post-induction [hpi]) by the fluorescent H2B-mTFP marker. Transient (24 hr) dexamethasone (DEX) activation of Ventx-GR was done following photoactivation resulting in malignant transformation of the induced cell and clonal expansion. At 3 day post-induction (dpi) a few cells (white arrowheads) progeny of the induced one are observed in the vicinity of the otic vesicle (white arrow). At 7 dpi, a tumor mass is observed in the brain (white arrowheads) and some of the cells (B) have metastasized (white arrowheads) to colonize new tissues as the proctodeum, close to the ventral fin (yellow arrows). Body axes are shown (a: anterior; p: posterior; d: dorsal; v: ventral; l: left; r: right).
Figure 2—figure supplement 4
Metastatic spreading following activation of KRASG12V and Ventx (KR+VX).
(A) Schematic representation and confocal microscopy of zebrafish head showing the early malignant cells (H2B-mTFP+ blue cells; white arrowheads) in the brain. Green EOS fluorescence has been used to visualize the whole zebrafish head. (B) Short-range dispersal in the brain of the early progeny of the initial malignant cell (H2B-mTFP+ blue cells; red arrowheads). An asterisk (*) indicates the eye and a white arrow the otic vesicle. (C) Schematic representation of the zebrafish larva trunk (left, dotted square), where H2B-mTFP+ cells (white arrowheads in center panel) can localize in KR+VX larvae, both as isolated cells (as in the dotted square a, 44.5 µm of mean distance) and as small clusters (as in the dotted square β, r=22.7 µm). (D) Schematic representation of zebrafish larva digestive tract (left, dotted square) and confocal microscopy showing H2b-mTFP+ cells (white arrowheads, center and right panel) localized in the gut. Green EOS fluorescence has been used to visualize the digestive tract. Red arrows indicate the bulb, the green arrows point to the mid-gut, the blue arrows the distal-gut, and the blue asterisk (*) indicates to the gut lumen. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are shown.
Figure 3
Metastatic cells following single-cell malignant transformation.
(A) A zebrafish larva, in which a single cell in the brain was photoinduced (at 1 day post-fertilization [dpf]) to express the oncogene KRASG12V by the blue fluorescence of the expression marker (H2b-mTFP) as in Figure 2, shows that the progeny of the cell of origin of cancer give rise to a tumor mass in the brain (white arrowhead), as well as to migrating metastatic-like cells that disseminate in the whole organism (white arrowhead), some localizing in proximity of arterial branchial arches (indicated by ζ and white arrowhead), trunk, and tail fin. (B) H2b-mTFP positive cells can migrate far from the site of induction (brain) and colonize ectopic tissues located in the heart and (C) the bottom of otic vesicle (in proximity of the primary head sinus, designated by α), the digestive tract (designated by β and γ), a feature characteristic of metastatic cancer cells. Note that no anesthetics (e.g. tricaine) or mounting media (low melting point agarose or methylcellulose) were used to block live zebrafish, during both monitoring and imaging performed on live zebrafish. An asterisk (*) indicates the eye and a white arrow the otic vesicle. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are indicated.
Figure 4
Transplantation of single cell(s) from hyperplasic tissue reveals cancer-initiating potential.
(A) Following KRASG12V plus Ventx activation, hyperplasic tissues (blue fluorescent) are observed at 6 days post-fertilization (dpf), top figure. The cells of the hyperplasia were dissociated and isolated blue (KRAS expressing) cells were transplanted (≈1 cell per host) at 2 dpf in a Nacre (mitf -/-) zebrafish line for a better tracking of the transplanted cells. The transplanted H2b-mTFP positive (blue) cell (red arrow) can be visualized as early as 3 hr post-transplantation (3 hpt) in the yolk of the host, close to the duct of Cuvier. White arrow indicates the head/eye (lateral view). (B) At 3 dpt the blue cell(s) from the hyperplasic tissue of the donor have colonized the host Nacre zebrafish larvae. Tumors in the brain, digestive tract, and intestine are observed and characterized by the blue fluorescence of the donor KRAS expressing cells (red arrows; n=31 out of 52 host individuals). In the bottom, immunofluorescence (IF) analysis of representative host zebrafish larvae with specific high level of phosphorylated ERK activity (pERK, red arrows) in the brain, intestine, and digestive tract. (C) A high number of exogenous blue fluorescent cells are here observed to migrate in the tail (red arrows). These observations indicate that the transplanted founder cell has both migratory, colonizing behavior, as well as survival growth advantage in the host to form tumors, and thus to re-initiate carcinogenesis. Scale bars and body axes (a: anterior; p: posterior; d: dorsal; v: ventral) are shown; T=telencephalon; M=mesencephalon; C=cerebellum; N=notochord; VF = ventral fin.
Figure 5
A two-step ‘Vogelgram’ model of deterministic and irreversible single-cell malignant transformation in vivo.
Within a healthy tissue, a normal cell (in light gray) acquires (step 1) genetic mutation(s) in driver oncogene(s) (Mut-Drivers such as kRASG12V). Such a mutant cell (in blue, referred to as a preprocancer [Brash, 2015] cell) can (step 2) aberrantly activate the expression of Epi-Drivers involved in pluripotency/reprogramming (e.g. VENTX/NANOG, POU5/OCT4) thus undergoing deterministic and irreversible malignant cell transformation (red cell, the cell of origin of cancer); (step 3) in situ short-range dispersal of the early malignant cells and (step 4) further progression to cancer mass and to the appearance of metastatic cells. Inversely, the mutant (preprocancer) cell (in blue) can maintain its physiological functions and be eventually eliminated from the healthy tissue.
Additional files
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)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
In vivo targeted and deterministic single-cell malignant transformation
eLife 13:RP97650.
https://doi.org/10.7554/eLife.97650.3
