To learn about protein function in complex systems biologists need tools to visualize endogenous proteins. A powerful approach to achieve this is the introduction of genetic tags to proteins of interest, fluorescent or otherwise. Endogenous protein labeling is highly advantageous compared to other approaches. First, it removes the dependence on specific antibodies which require extensive validation. Second, introducing fluorescent tags circumvents difficulties of antibodies physically accessing their targets, such as core components of the postsynaptic density (PSD), in which the antibody epitopes can be sterically occluded by other proteins. Third, genetic tagging with fluorescent proteins is compatible with longitudinal imaging in live tissue in vitro and in vivo. Finally, tagging endogenous proteins prevents potential confounds of protein mis-localization following overexpression and preserves the natural transcriptional and posttranscriptional regulation of the gene. The discovery of CRISPR/Cas9 changed the gene editing field (Cho et al., 2013; Cong et al., 2013; Doudna and Charpentier, 2014; Hsu et al., 2014; Jinek et al., 2012; Mali et al., 2013; Sander and Joung, 2014), making genome editing much easier, faster, and cheaper. Transgenic knock-in organisms can be generated to tag specific proteins in every cell. CRISPR/Cas9 technology has facilitated this approach; however, it is still relatively time-consuming and considered low throughput, especially in rodents. In addition, having knock-in (KI) in all cells in an organism can be problematic in dense tissues. Therefore, methods to enable genome editing in single or a relatively sparse population of cells are highly desirable. Such CRISPR/Cas9-based genome editing strategies traditionally rely on homology-dependent recombination (HDR) (Lombardo et al., 2011; Wang et al., 2013; Yang et al., 2013); however, HDR does not work efficiently in post-mitotic cells, such as neurons (Mao et al., 2008; Nishiyama et al., 2017; Orthwein et al., 2015; Saleh-Gohari and Helleday, 2004). Consequently, in post-mitotic cells researchers have adopted methods based on non-homologous end joining (NHEJ) (Bétermier et al., 2014; Cox et al., 2015; Gao et al., 2019; Lieber, 2010; Suzuki et al., 2016; Willems et al., 2020). Such approaches have achieved higher KI efficiency in neurons compared to HDR-based methods. However, these approaches directly edit the coding sequence of the target gene, possibly causing (insertion or deletion) INDEL mutations, which commonly occur at CRISPR/Cas9 mediated editing sites (Suzuki et al., 2016). This is especially a concern if the tag is not introduced at the C-terminal portion of the protein. Furthermore, even when the KI proceeds as planned, the deletion or addition of some amino acids flanking the tag is unavoidable, and the exact site of KI is still dependent on a (protospacer-adjacent motif) PAM and reliable guide to be present at the right place.

To address these problems, we designed a new strategy – Targeted Knock-In with Two (TKIT) guides – to enable precise genome editing by utilizing two guide RNAs located in non-coding regions upstream and downstream of the coding sequence to be edited. This strategy protects from INDEL mutations within the coding region and enables absolute control over the sequence surrounding the KI site. In addition, since the coding region is not being targeted directly, we have fewer limitations during the selection of potential guide RNAs, increasing the availability of high-score guides and lowering the likelihood of off-target editing.

In this study, we have used TKIT to KI Super Ecliptic pHluorin (SEP), a pH dependent GFP variant (Lin and Huganir, 2007; Miesenböck et al., 1998), and other protein tags to the N-terminus of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl D-aspartic acid (NMDA) receptor subunits, which form glutamate receptors responsible for the majority of excitatory transmission in the brain. We show that TKIT enables efficient KI in primary cultured neurons from mouse and rat and that TKIT can be used for genome editing in vivo with in utero electroporation or through injection of adeno-associated viral (AAV) vectors in adult mice. Thus, TKIT provides an efficient, easy to use, and readily scalable approach for the tagging of endogenous proteins to facilitate their study in intact biological systems.

Homology Independent Target Integration (HITI) based KI strategies have been used to tag endogenous proteins in postmitotic cells such as neurons (Gao et al., 2019; Suzuki et al., 2016; Willems et al., 2020); however, these approaches are sensitive to INDEL mutations as the coding sequence is directly edited. This is particularly problematic when labeling the N-terminal region of proteins, where the introduction of INDELs will be most damaging. Additionally, in the vast majority of cases, a Cas9 cutting site is not present at exactly the desired site of insertion. Therefore, the tag cannot always be precisely targeted, and deletion or addition of extra base pairs is unavoidable even when editing occurs as planned and without generation of INDELs. Further, the choice of guides with high cutting efficiency and good off-target scores near the desired insertion site is often very limited. In an attempt to overcome these issues – which are common to all currently used NHEJ-based CRISPR KI methods – we designed a modified CRISPR/Cas9-based KI strategy. Our approach utilizes two guide RNAs that cut the genomic DNA flanking the coding region of interest and includes a replacement donor DNA fragment encoding the excised coding sequence with the addition of a fluorescent or small (i.e., Myc) protein tag. We named this new CRISPR/Cas9 based knock-in strategy TKIT guides.

Genome editing of coding sequence by targeting non-coding regions

As a proof of principle, we sought to KI SEP, a pH sensitive GFP variant that is used to visualize cell surface proteins, to the N-terminus of the AMPA receptor subunit GluA2, encoded by the Gria2 gene. We designed guides to cut within the 5’-UTR (untranslated region) and intron 1–2 of Gria2, choosing guides located ~100 bp away from splice junctions so as not to affect mRNA processing (Figure 1A). We constructed a donor DNA fragment containing the endogenous Gria2 sequence, with the addition of a SEP-tag behind the signal peptide, and the same two guides and PAM sequences at the ends. The two guides within the donor have opposite locations and flipped sequences compared to the genomic DNA (Figure 1A, Figure 1—figure supplement 1). We added this switch-and-flip design to promote forward insertion of the donor DNA, which occurs randomly through the process of NHEJ. If the donor inserts into the genome in the reverse orientation (and without INDELs) the guides and PAMs remain intact so they will be cut again by Cas9, enabling another chance at insertion in the correct orientation. The cycle will continue until the donor is inserted in the correct orientation, or until INDELs destroy the guides and PAM sequence (Figure 1B, Figure 1—figure supplement 1).

Figure 1 with 1 supplement see all

Download asset Open asset

To test TKIT we transfected primary mouse cortical cultures at DIV7-9 with plasmids containing: (1) both guide RNAs along with SpCas9, (2) SEP-GluA2 donor DNA fragment, and (3) mCherry as a cell morphology marker (Figure 1A,C). At DIV14-16 we live-imaged transfected neurons and found neurons with punctate SEP signal concentrated in dendritic spines, consistent with the labeling of AMPARs (Figure 1E). We used immunofluorescence with antibodies against GFP and the C-terminus of GluA2 to show that the GFP signal was co-localized with GluA2 (Figure 1F–G).

To confirm that SEP KI to the Gria2 gene still results in normal RNA splicing, we extracted bulk mRNA from DIV19 neurons electroporated with TKIT components, conducted reverse transcription to generate cDNA, and performed PCR and Sanger sequencing. Sequencing results showed that the splice junctions between exon1 and exon2 were intact, indicating that SEP KI did not affect Gria2 mRNA splicing (Figure 1D, Figure 2—figure supplement 1). Immunostaining of GluA2 in primary cultured neurons with TKIT mediated SEP-GluA2 KI showed that the expression level of SEP-GluA2 is comparable with GluA2 in non-transfected surrounding neurons (Figure 2—figure supplement 1). KI was dependent on co-transfection of the TKIT donor and the Cas9 with two guides constructs, as transfection of the TKIT donor alone did not lead to any detectable KI (Figure 2—figure supplement 1). These data indicate that TKIT enables accurate genomic editing in neurons, with broad potential for future applications in neurons and other post-mitotic cells.

Efficient targeting of synaptic proteins with TKIT

NMDA and AMPA receptors mediate the majority of excitatory synaptic transmission in the mammalian CNS and are critical for the tuning of synaptic strength. In particular, the trafficking of AMPA receptors is known to underlie synaptic plasticity, supporting learning, and memory (Diering and Huganir, 2018). To date, the vast majority of AMPA receptor imaging studies have relied on the overexpression of (often SEP) tagged receptors (Ashby et al., 2006; Kopec et al., 2006; Lin and Huganir, 2007; Thomas et al., 2008). While these studies have helped us learn about the regulation of AMPA receptors in vitro, and more recently, in vivo (Roth et al., 2020; Tan et al., 2020; Zhang et al., 2015), there is always the concern that overexpression could result in aberrant and non-physiological receptor regulation. Therefore, tools to tag endogenous AMPARs will be highly useful.

We sought to tag a variety of synaptic proteins, including AMPA and NMDA receptor subunits, extending our proof of principle experiments showing SEP KI to GluA2. We designed TKIT constructs for SEP KI to the N-terminus of AMPAR subunits GluA1, GluA2, and GluA3, as well as the most abundant NMDAR subunits (NR1, NR2A, and NR2B) and the beta2 subunit of the gamma-aminobutyric acid (GABA) receptor. In addition, we wanted to demonstrate that TKIT also works for cytosolic proteins. To do this we targeted gephyrin, a postsynaptic scaffold protein that is localized to inhibitory synapses, as a candidate for GFP KI. Since previous reports tagged the N-terminus of gephyrin (Lardi-Studler et al., 2007; Tyagarajan et al., 2011), we also targeted GFP KI to the N-terminus of gephyrin. To target this set of synaptic proteins, guide cutting sites were located either in introns flanking the target exon, or in the 5’UTR of the targeted exon and the downstream intron (Figure 2—figure supplement 2).

As before, primary mouse cultured neurons were transfected with TKIT constructs at DIV7-9, and confocal images were captured between DIV14 and 16 following immunofluorescence staining. The SEP-GluA1-3 signal in neurons, visualized with GFP immunostaining, showed clear puncta at dendritic spines (Figure 2A–C). As expected, GFP-gephyrin (Figure 2D) and SEP-Beta2 (Figure 2—figure supplement 3) showed punctate signals along dendrites.

Figure 2 with 6 supplements see all

Download asset Open asset

Figure 2—source data 1

Efficiency counts to assess TKIT knockin of SEP-GluA1, SEP-GluA2, SEP-GluA3, and GFP-Gephyrin in mouse neurons.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig2-data1-v1.xlsx

Download elife-65202-fig2-data1-v1.xlsx

To calculate the efficiency of TKIT KI we counted the number of KI cells as a percentage of all transfected neurons based on the mCherry cell fill. The efficiency for SEP KI to AMPA receptor subunits GluA1, GluA2, and GluA3 was 21.9%, 38.9%, and 20.5%, respectively (Figure 2E–G). The efficiency for Beta2 KI was 27.2% (Figure 2—figure supplement 3) and 41.7% for GFP-gephyrin KI (Figure 2H). Targeting NMDA receptors resulted in similar spine enrichment for SEP-NR1, SEP-NR2A, and SEP-NR2B (Figure 3A–C) with efficiencies of 15.2%, 16.7%, and 27.5%, respectively (Figure 3D–F).

Figure 3 with 1 supplement see all

Download asset Open asset

Figure 3—source data 1

Efficiency counts to assess TKIT knockin of SEP-NR1, SEP-NR2A, and SEP-NR2B in mouse neurons.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig3-data1-v1.xlsx

Download elife-65202-fig3-data1-v1.xlsx

In addition to SEP, we used TKIT to KI the larger fluorescent protein tdTomato to Gephyrin (Figure 2—figure supplement 4) and GluA2 (Figure 2—figure supplement 5) as well as the smaller tag Myc to GluA1 and GluA2 (Figure 2—figure supplement 5). We validated the fidelity of KI by showing colocalization of GFP and the endogenous proteins for SEP-GluA1, SEP-GluA2, and SEP-GluA3 (Figure 2—figure supplement 6), and SEP-NR1, SEP-NR2A, and SEP-NR2B (Figure 3—figure supplement 1) with immunofluorescence. Correlation analysis showed strong correlation between the intensities of anti-GFP signal and anti-GluA1-3 signal in TKIT-mediated SEP KI neurons (Figure 2—figure supplement 6). We further validated the correct subcellular localization of proteins tagged with TKIT. We verified the synaptic localization of SEP-GluA1-3 with immunofluorescence using antibodies against GFP and PSD95 (Figure 4A–C).

Figure 4 with 3 supplements see all

Download asset Open asset

Figure 4—source data 1

Immunofluorescence data for correlation of PSD95 signal with SEP-GluA1, SEP-GluA2, and SEP-GluA3.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig4-data1-v1.xlsx

Download elife-65202-fig4-data1-v1.xlsx

We observed a strong correlation between the anti-GFP signal and anti-PSD95 signal in SEP KI neurons (Figure 4D–F). We found that GFP- or tdTomato-gephyrin signal colocalized with gephyrin and the GABA receptor Beta2 subunit, used as a marker for inhibitory synapses (Figure 2—figure supplement 4). Sequence analysis of reverse-transcribed cDNA from mRNA extracted from neuronal cultures electroporated with each TKIT construct showed intact splicing of all tagged proteins in this study (Figure 4—figure supplement 1). TKIT-mediated KI of various tags to endogenous proteins will aid the characterization and functional study of proteins avoiding potential artifacts of protein overexpression.

Because guides flank a critical portion of the gene being edited, it is possible that a portion of exon 1 is excised and repaired without the addition of the donor DNA fragment. This would lead to the generation of a knock-out (KO) allele for the target gene. To assess the KI/wild-type (WT)/KO possibility in the Gria2 KI experiments in neuronal cultures described in Figure 1, we designed primer-pairs flanking the two guides KI site and performed genomic PCR. Genomic DNA samples were extracted from neuron culture electroporated with TKIT constructs. The primer-pair can pick up three fragments (KI/WT/KO bands). DNA electrophoresis of the PCR products showed clear KI and WT bands, but no detectable KO band (Figure 4—figure supplement 2). In addition, we transfected primary neuron cultures with SpCas9 and two guides, SEP-GluA2 donor DNA and mCherry as a morphology marker at DIV 7–9 and then used immunofluorescence to detect GFP and GluA2 at DIV 14–16. GluA2 was not detectable in 12.4% of the mCherry-positive cells, indicating knock-out of GluA2 in a small subset of cells that failed to knock-in the donor DNA (Figure 4—figure supplement 2). To test the possibility that a subset of neurons naturally lacked GluA2 expression, independent of manipulation, we assessed the levels of GluA2 in mouse cortical neurons. We found that 1.2% of cells had undetectable GluA2 (Figure 4—figure supplement 3). Therefore, we think that undetectable GluA2 observed in some cells transfected with TKIT plasmids is an unfortunate, but not completely unexpected, result of gene knock-out through double strand break generation without donor insertion.

Endogenous KI of SEP-GluA2 in vivo

After demonstrating efficient KI in primary cultured neurons, we sought to test the utility of TKIT to introduce a SEP-tag to the N-terminus of GluA2 in vivo. Plasmids containing SEP-GluA2 donor DNA and complementary two guides with spCas9 were transfected into layer 2/3 pyramidal neurons of somatosensory cortex using in utero electroporation of mouse embryos at embryonic day 15 (E15; Figure 5A). Cranial windows were then implanted over the barrel cortex of adult animals. After 2–3 weeks of recovery in vivo two-photon images were acquired (Figure 5B). Punctate SEP-GluA2 could be clearly observed along the dendrites of layer 2/3 pyramidal neurons (Figure 5C and D, Figure 5—video 1), similar to previously studies showing the over-expressed SEP-GluA1 and SEP-GluA2 (Suresh and Dunaevsky, 2017; Zhang et al., 2015). We were even able to observe SEP signal in mice that were more than 1-year old, indicating the long-term stability of the knock-in tag. Therefore, using TKIT we have been able to tag, and subsequently visualize, endogenous AMPAR subunits in vivo. This approach will allow the longitudinal tracking of endogenous AMPAR plasticity in behaving animals.

Figure 5 with 1 supplement see all

Download asset Open asset

Figure 5—source data 1

Immunohistochemistry data for SEP-GluA2 knockin efficiency calculation with dual virus approach.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig5-data1-v1.xlsx

Download elife-65202-fig5-data1-v1.xlsx

Cell-type specific TKIT in vivo

To gain cell-type specificity of TKIT KI we bred heterozygous Camk2a^Cre or homozygous Pvalb^Cre mice with homozygous lox-stop-lox conditional Cas9 animals to generate offspring with Cas9 expression in forebrain CamK2a positive neurons or Pvalb (PV) positive interneurons (Camk2a^Cre/wt:Cas9^fl/wt or Pvalb^Cre/wt:Cas9^fl/wt). We injected high titer AAV-TKIT-SEP-GluA1 into the dorsal hippocampus of adult Camk2a^Cre/wt:Cas9^fl/wt and Pvalb^Cre/wt:Cas9^fl/wt animals (Figure 6A,C). After 2–3 weeks we observed robust KI of SEP-GluA1 in the dorsal hippocampus of the injected hemisphere. Super-resolution confocal (Airyscan, see Materials and methods) images were acquired of CA1 hippocampal neurons (Figure 6B,D). Here, endogenous SEP signal could be seen concentrated in putative dendritic spines of excitatory neurons (Camk2a^Cre/wt:Cas9^fl/wt animals) and along the smooth dendrites of PV neurons (Pvalb^Cre/wt:Cas9^fl/wt animals).

Figure 6

Download asset Open asset

Next, we tested if the TKIT-SEP-GluA2 virus could be coupled with two-photon imaging in vivo, as we demonstrated previously with in utero electroporation. To test this, the somatosensory cortex of Camk2a^Cre/wt:Cas9^fl/wt mice was injected with high titer TKIT-SEP-GluA2 AAV. Immediately after virus injection a cranial window was placed above the injection site (Figure 6E). After 2 weeks, animals were imaged with two-photon microscopy. As with in utero electroporation, we could clearly observe cells with SEP-GluA2 KI in vivo (Figure 6F). Given the higher infection rate of TKIT-SEP-GluA2 AAV compared to in utero electroporation, we were able to visualize a higher density of SEP-GluA2 expressing neurons which will allow the longitudinal study of synaptic surface GluA2 on multiple neurons within one field of view, with KI sparse enough to distinguish individual synaptic puncta and neuronal processes. These data demonstrate the utility of TKIT to KI fluorescent tags to endogenous proteins in a cell-type specific manner in adult mice following injection of a single AAV, enabling their subsequent visualization in vivo. Recent reports show that fragments of AAVs can integrate into double stranded DNA breaks induced by CRISPR/Cas9 (Hanlon et al., 2019; Wolter et al., 2020). Presently, we do not have data to confirm if this is occurring in our experiments with TKIT-mediated KI of SEP-GluA1 and SEP-GluA2 with AAVs. Integration of AAV fragments into the 5’UTR of GluA1 or GluA2 could disrupt function, potentially contributing to the decreased efficiency observed in vivo (Figure 5H) compared to in vitro (Figure 2F).

TKIT of SEP AMPA receptor subunits in rat neurons

After successfully using TKIT in mouse neurons in vitro and in vivo, we wanted to confirm that this approach will generalize to other species. Rats are commonly used in biomedical research, yet there are far fewer transgenic rat models compared to mice. We re-designed TKIT guides and donors to be complementary to the rat genome, to enable tagging of rat GluA1 and GluA2 with SEP. We transfected these TKIT constructs into rat hippocampal neuron culture at DIV7-8 and used GFP immunohistochemistry to detect KI cells at DIV14-15 (Figure 7A,B, Figure 7—figure supplement 1). As expected, we observed green puncta at dendritic spines, which co-localized with the synaptic marker PSD95 (Figure 7—figure supplement 1). This demonstrates that TKIT can also mediate editing of the rat genome, with efficiencies roughly equivalent to those observed in mouse neurons (Figure 7C,D).

Figure 7 with 1 supplement see all

Download asset Open asset

Figure 7—source data 1

Efficiency counts to assess TKIT knockin of SEP-GluA1 and SEP-GluA2 in rat neurons.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig7-data1-v1.xlsx

Download elife-65202-fig7-data1-v1.xlsx

Figure 7—source data 2

Fluorescence recovery after photobleaching data to assess SEP-GluA1 mobility in rat neurons.

Source Data for figure supplements.

https://cdn.elifesciences.org/articles/65202/elife-65202-fig7-data2-v1.xlsx

Download elife-65202-fig7-data2-v1.xlsx

Strategies to efficiently, and reliably, edit genomic DNA in post mitotic cells, such as neurons, are lacking behind methods based on HDR that are routinely applied to dividing cells (Wang et al., 2013; Yang et al., 2013). Existing methods, based on NHEJ, have shown promise for enabling KI of tags (fluorescent proteins/small epitopes) to endogenous proteins in neurons (Gao et al., 2019; Suzuki et al., 2016); however, such methods are sensitive to INDEL mutations during the KI process. This culminates in reduced KI efficiency and generally restricts the locations where tags can be inserted within the target protein, typically limited to the C-terminal region. Although recent studies have shown that residual HDR activity or introducing exogenous recombinase in neurons is capable of inducing KI (Cai et al., 2019; Nishiyama et al., 2017), the efficiency of such methods remains reasonably low, prompting us to develop alternative strategies for KI to neurons.

Here, we introduce a CRISPR/Cas9 based method termed TKIT guides to improve the efficiency and reliability of KI to post-mitotic cells and dramatically increasing the flexibility of where tags can be inserted to target proteins. Using TKIT, a tag can be inserted into any user-defined location without undesired amino acid loss or addition. By using a pair of guide RNAs instead of a single guide RNA, TKIT utilizes the use of Cas9 cutting sites that lie in non-coding regions of the gene of interest (such as 5’-UTR and introns). This approach is significantly less sensitive to INDEL mutations since no coding regions of the gene are cut. Even if INDELs are introduced around the cut sites, it is highly likely that the mRNA and subsequent protein are left intact. In addition, because non-coding DNA regions are targeted, and guides do not have to be close to the desired insertion site, there is a greatly increased pool of potential guide RNAs to choose from, enabling high-ranked guides with high on-target score and low off-target probability to be picked. Finally, the switch-and-flip design built into the donor DNA sequence helps to bias the insertion of donor DNA into the correct orientation, further increasing KI efficiency.

We demonstrate that TKIT can be used to tag different AMPA and NMDA receptor subunits, as well as the cytosolic protein gephyrin, with efficiencies of up to 41.7% (GFP-gephyrin). The KI efficiency of SEP-GluA2 reached 16.1% in vivo with a two-virus system. This is greater than the efficiency achieved with the original demonstration of HITI (Suzuki et al., 2016; Nishiyama et al., 2017). Furthermore, because of the increased protection from INDEL mutations TKIT is particularly useful for tagging proteins at the N-terminus. Exploiting this, we were able to successfully KI SEP to the N-terminus of AMPA and NMDA receptor subunits, positioning the SEP-tag on the extracellular portion of the receptor so the pH sensitivity of SEP (Miesenböck et al., 1998) can be utilized to visualize the surface pool of receptors.

While TKIT is resistant to INDELs in the coding region of the gene being edited, the method is not immune to other caveats. For example, if the donor DNA (with the tag/label) is not present to replace the excised genomic DNA then a KO allele could be generated. For GluA2 tagging with SEP we observed KO of GluA2 in 12.4% of transfected cells. Therefore, it will be important to consider that KI neurons are likely surrounded by a small fraction of neurons that are heterozygous or homozygous KO’s for the gene of interest.

We have demonstrated that TKIT works in vivo following in utero electroporation, as well as in adult mice following AAV injection. We show that AMPAR subunits can be tagged in wild-type mice with a dual-AAV system, as well as in a cell-type specific manner using transgenic mice. In proof of principle experiments, we showed that TKIT KI of SEP to GluA2 is compatible with two-photon imaging in vivo, enabling us to visualize the surface expression of endogenous GluA2-containing AMPARs in live animals for the first time. This opens up possibilities for tagging and studying endogenous proteins (such as AMPARs) in wild-type or disease model mice (dual virus approach), or in a cell-type-specific manner in Cre-driver mouse lines. While we observed a lower efficiency of KI in vivo (16.1%) compared to in vitro (38.9%) with SEP-GluA2, it is likely that we are under-representing the in vivo efficiency, as a proportion of the cells that were transduced with SpCas9 might have failed to receive the TKIT donor DNA.

Tagging endogenous proteins to study their function and subcellular localization is preferential to exogenous protein overexpression. Therefore, the adoption of endogenous KI strategies such as TKIT will dramatically aid the study of proteins expressed at physiological levels. A potential application of TKIT is its combination with modern proximity labeling techniques such as APEX, BioID, and TurboID (Branon et al., 2018; Rhee et al., 2013; Roux et al., 2012). Those techniques crucially rely on proper subcellular targeting of fusion proteins. Tagging endogenous proteins with proximity labeling enzymes will reduce the confound of nonspecific subcellular protein localization associated with protein overexpression and increase the signal to noise ratio of ‘hits’ as there will be lower background from mis-localized proteins. However, the feasibility of coupling TKIT with proximity labeling remains to be tested and will likely depend upon the transfection or transduction efficiency, the KI efficiency for the target protein, as well as the absolute abundance of the target protein being tagged. In addition, the flexibility with where tags can be introduced will be of great value as it will allow known protein interaction motifs (i.e., C-terminal PDZ ligands) to be avoided when tagging target proteins, further preventing mis-localization of the protein being studied.

While we detect faithful mRNA splicing of GluA2 following TKIT SEP KI, it remains a potential concern that targeting non-coding regions could have unforeseen effects on mRNA processing. A greater understanding of the biology underlying mRNA processing will facilitate the design of guides to avoid such effects. Given the high flexibility in guide RNA choice, this concern will be easily overcome, as long as genomic DNA cutting sites can be identified that leave mRNA processing unaffected.

In summary, we demonstrate that TKIT is an efficient, reliable, and easy to implement method for tagging endogenous proteins in vitro and in vivo. TKIT can edit mouse and rat genomic DNA, with promised utility in all model organisms. Although, to date, we have only used TKIT to KI tags to proteins, a potential application to clinical therapy to correct causative mutations in genetic disorders could be possible.

Source data files have been provided for all figures. Due to the size of the raw image data these will be made available upon request.

1. 1. Arendt KL
   2. Royo M
   3. Fernández-Monreal M
   4. Knafo S
   5. Petrok CN
   6. Martens JR
   7. Esteban JA

   (2010) PIP3 controls synaptic function by maintaining AMPA receptor clustering at the postsynaptic membrane

   Nature Neuroscience 13:36–44.

   https://doi.org/10.1038/nn.2462

   • PubMed
   • Google Scholar
2. 1. Ashby MC
   2. Maier SR
   3. Nishimune A
   4. Henley JM

   (2006) Lateral diffusion drives constitutive exchange of AMPA receptors at dendritic spines and is regulated by spine morphology

   Journal of Neuroscience 26:7046–7055.

   https://doi.org/10.1523/JNEUROSCI.1235-06.2006

   • PubMed
   • Google Scholar
3. 1. Bétermier M
   2. Bertrand P
   3. Lopez BS

   (2014) Is non-homologous end-joining really an inherently error-prone process?

   PLOS Genetics 10:e1004086.

   https://doi.org/10.1371/journal.pgen.1004086

   • PubMed
   • Google Scholar
4. 1. Branon TC
   2. Bosch JA
   3. Sanchez AD
   4. Udeshi ND
   5. Svinkina T
   6. Carr SA
   7. Feldman JL
   8. Perrimon N
   9. Ting AY

   (2018) Efficient proximity labeling in living cells and organisms with TurboID

   Nature Biotechnology 36:880–887.

   https://doi.org/10.1038/nbt.4201

   • PubMed
   • Google Scholar
5. 1. Cai Y
   2. Cheng T
   3. Yao Y
   4. Li X
   5. Ma Y
   6. Li L
   7. Zhao H
   8. Bao J
   9. Zhang M
   10. Qiu Z
   11. Xue T

   (2019) In vivo genome editing rescues photoreceptor degeneration via a Cas9/RecA-mediated homology-directed repair pathway

   Science Advances 5:eaav3335.

   https://doi.org/10.1126/sciadv.aav3335

   • PubMed
   • Google Scholar
6. 1. Cho SW
   2. Kim S
   3. Kim JM
   4. Kim JS

   (2013) Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease

   Nature Biotechnology 31:230–232.

   https://doi.org/10.1038/nbt.2507

   • PubMed
   • Google Scholar
7. 1. Cong L
   2. Ran FA
   3. Cox D
   4. Lin S
   5. Barretto R
   6. Habib N
   7. Hsu PD
   8. Wu X
   9. Jiang W
   10. Marraffini LA
   11. Zhang F

   (2013) Multiplex genome engineering using CRISPR/Cas systems

   Science 339:819–823.

   https://doi.org/10.1126/science.1231143

   • PubMed
   • Google Scholar
8. 1. Cox DB
   2. Platt RJ
   3. Zhang F

   (2015) Therapeutic genome editing: prospects and challenges

   Nature Medicine 21:121–131.

   https://doi.org/10.1038/nm.3793

   • PubMed
   • Google Scholar
9. 1. Diering GH
   2. Huganir RL

   (2018) The AMPA receptor code of synaptic plasticity

   Neuron 100:314–329.

   https://doi.org/10.1016/j.neuron.2018.10.018

   • PubMed
   • Google Scholar
10. 1. Doudna JA
    2. Charpentier E

    (2014) Genome editing. The new frontier of genome engineering with CRISPR-Cas9

    Science 346:1258096.

    https://doi.org/10.1126/science.1258096

    • PubMed
    • Google Scholar
11. 1. Frischknecht R
    2. Heine M
    3. Perrais D
    4. Seidenbecher CI
    5. Choquet D
    6. Gundelfinger ED

    (2009) Brain extracellular matrix affects AMPA receptor lateral mobility and short-term synaptic plasticity

    Nature Neuroscience 12:897–904.

    https://doi.org/10.1038/nn.2338

    • PubMed
    • Google Scholar
12. 1. Gao Y
    2. Hisey E
    3. Bradshaw TWA
    4. Erata E
    5. Brown WE
    6. Courtland JL
    7. Uezu A
    8. Xiang Y
    9. Diao Y
    10. Soderling SH

    (2019) Plug-and-Play protein modification using Homology-Independent universal genome engineering

    Neuron 103:583–597.

    https://doi.org/10.1016/j.neuron.2019.05.047

    • PubMed
    • Google Scholar
13. 1. Hanlon KS
    2. Kleinstiver BP
    3. Garcia SP
    4. Zaborowski MP
    5. Volak A
    6. Spirig SE
    7. Muller A
    8. Sousa AA
    9. Tsai SQ
    10. Bengtsson NE
    11. Lööv C
    12. Ingelsson M
    13. Chamberlain JS
    14. Corey DP
    15. Aryee MJ
    16. Joung JK
    17. Breakefield XO
    18. Maguire CA
    19. György B

    (2019) High levels of AAV vector integration into CRISPR-induced DNA breaks

    Nature Communications 10:4439.

    https://doi.org/10.1038/s41467-019-12449-2

    • PubMed
    • Google Scholar
14. 1. Heine M
    2. Groc L
    3. Frischknecht R
    4. Béïque JC
    5. Lounis B
    6. Rumbaugh G
    7. Huganir RL
    8. Cognet L
    9. Choquet D

    (2008) Surface mobility of postsynaptic AMPARs tunes synaptic transmission

    Science 320:201–205.

    https://doi.org/10.1126/science.1152089

    • PubMed
    • Google Scholar
15. 1. Hsu PD
    2. Lander ES
    3. Zhang F

    (2014) Development and applications of CRISPR-Cas9 for genome engineering

    Cell 157:1262–1278.

    https://doi.org/10.1016/j.cell.2014.05.010

    • PubMed
    • Google Scholar
16. 1. Jinek M
    2. Chylinski K
    3. Fonfara I
    4. Hauer M
    5. Doudna JA
    6. Charpentier E

    (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity

    Science 337:816–821.

    https://doi.org/10.1126/science.1225829

    • PubMed
    • Google Scholar
17. 1. Kerr JM
    2. Blanpied TA

    (2012) Subsynaptic AMPA receptor distribution is acutely regulated by actin-driven reorganization of the postsynaptic density

    Journal of Neuroscience 32:658–673.

    https://doi.org/10.1523/JNEUROSCI.2927-11.2012

    • PubMed
    • Google Scholar
18. 1. Kopec CD
    2. Li B
    3. Wei W
    4. Boehm J
    5. Malinow R

    (2006) Glutamate receptor exocytosis and spine enlargement during chemically induced long-term potentiation

    Journal of Neuroscience 26:2000–2009.

    https://doi.org/10.1523/JNEUROSCI.3918-05.2006

    • PubMed
    • Google Scholar
19. 1. Lardi-Studler B
    2. Smolinsky B
    3. Petitjean CM
    4. Koenig F
    5. Sidler C
    6. Meier JC
    7. Fritschy JM
    8. Schwarz G

    (2007) Vertebrate-specific sequences in the gephyrin E-domain regulate cytosolic aggregation and postsynaptic clustering

    Journal of Cell Science 120:1371–1382.

    https://doi.org/10.1242/jcs.003905

    • PubMed
    • Google Scholar
20. 1. Lieber MR

    (2010) The mechanism of double-strand DNA break repair by the nonhomologous DNA end-joining pathway

    Annual Review of Biochemistry 79:181–211.

    https://doi.org/10.1146/annurev.biochem.052308.093131

    • PubMed
    • Google Scholar
21. 1. Lin DT
    2. Huganir RL

    (2007) PICK1 and phosphorylation of the glutamate receptor 2 (GluR2) AMPA receptor subunit regulates GluR2 recycling after NMDA receptor-induced internalization

    Journal of Neuroscience 27:13903–13908.

    https://doi.org/10.1523/JNEUROSCI.1750-07.2007

    • PubMed
    • Google Scholar
22. 1. Lombardo A
    2. Cesana D
    3. Genovese P
    4. Di Stefano B
    5. Provasi E
    6. Colombo DF
    7. Neri M
    8. Magnani Z
    9. Cantore A
    10. Lo Riso P
    11. Damo M
    12. Pello OM
    13. Holmes MC
    14. Gregory PD
    15. Gritti A
    16. Broccoli V
    17. Bonini C
    18. Naldini L

    (2011) Site-specific integration and tailoring of cassette design for sustainable gene transfer

    Nature Methods 8:861–869.

    https://doi.org/10.1038/nmeth.1674

    • PubMed
    • Google Scholar
23. 1. Makino H
    2. Malinow R

    (2009) AMPA receptor incorporation into synapses during LTP: the role of lateral movement and exocytosis

    Neuron 64:381–390.

    https://doi.org/10.1016/j.neuron.2009.08.035

    • PubMed
    • Google Scholar
24. 1. Mali P
    2. Yang L
    3. Esvelt KM
    4. Aach J
    5. Guell M
    6. DiCarlo JE
    7. Norville JE
    8. Church GM

    (2013) RNA-guided human genome engineering via Cas9

    Science 339:823–826.

    https://doi.org/10.1126/science.1232033

    • PubMed
    • Google Scholar
25. 1. Mao Z
    2. Bozzella M
    3. Seluanov A
    4. Gorbunova V

    (2008) DNA repair by nonhomologous end joining and homologous recombination during cell cycle in human cells

    Cell Cycle 7:2902–2906.

    https://doi.org/10.4161/cc.7.18.6679

    • PubMed
    • Google Scholar
26. 1. Miesenböck G
    2. De Angelis DA
    3. Rothman JE

    (1998) Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins

    Nature 394:192–195.

    https://doi.org/10.1038/28190

    • PubMed
    • Google Scholar
27. 1. Nishiyama J
    2. Mikuni T
    3. Yasuda R

    (2017) Virus-Mediated genome editing via Homology-Directed repair in Mitotic and postmitotic cells in mammalian brain

    Neuron 96:755–768.

    https://doi.org/10.1016/j.neuron.2017.10.004

    • PubMed
    • Google Scholar
28. 1. Orthwein A
    2. Noordermeer SM
    3. Wilson MD
    4. Landry S
    5. Enchev RI
    6. Sherker A
    7. Munro M
    8. Pinder J
    9. Salsman J
    10. Dellaire G
    11. Xia B
    12. Peter M
    13. Durocher D

    (2015) A mechanism for the suppression of homologous recombination in G1 cells

    Nature 528:422–426.

    https://doi.org/10.1038/nature16142

    • PubMed
    • Google Scholar
29. 1. Rhee HW
    2. Zou P
    3. Udeshi ND
    4. Martell JD
    5. Mootha VK
    6. Carr SA
    7. Ting AY

    (2013) Proteomic mapping of mitochondria in living cells via spatially restricted enzymatic tagging

    Science 339:1328–1331.

    https://doi.org/10.1126/science.1230593

    • PubMed
    • Google Scholar
30. 1. Roth RH
    2. Zhang Y
    3. Huganir RL

    (2017) Dynamic imaging of AMPA receptor trafficking in vitro and in vivo

    Current Opinion in Neurobiology 45:51–58.

    https://doi.org/10.1016/j.conb.2017.03.008

    • PubMed
    • Google Scholar
31. 1. Roth RH
    2. Cudmore RH
    3. Tan HL
    4. Hong I
    5. Zhang Y
    6. Huganir RL

    (2020) Cortical synaptic AMPA receptor plasticity during motor learning

    Neuron 105:895–908.

    https://doi.org/10.1016/j.neuron.2019.12.005

    • PubMed
    • Google Scholar
32. 1. Roux KJ
    2. Kim DI
    3. Raida M
    4. Burke B

    (2012) A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells

    Journal of Cell Biology 196:801–810.

    https://doi.org/10.1083/jcb.201112098

    • Google Scholar
33. 1. Saito T
    2. Nakatsuji N

    (2001) Efficient gene transfer into the embryonic mouse brain using in vivo electroporation

    Developmental Biology 240:237–246.

    https://doi.org/10.1006/dbio.2001.0439

    • PubMed
    • Google Scholar
34. 1. Saleh-Gohari N
    2. Helleday T

    (2004) Conservative homologous recombination preferentially repairs DNA double-strand breaks in the S phase of the cell cycle in human cells

    Nucleic Acids Research 32:3683–3688.

    https://doi.org/10.1093/nar/gkh703

    • PubMed
    • Google Scholar
35. 1. Sander JD
    2. Joung JK

    (2014) CRISPR-Cas systems for editing, regulating and targeting genomes

    Nature Biotechnology 32:347–355.

    https://doi.org/10.1038/nbt.2842

    • PubMed
    • Google Scholar
36. 1. Schindelin J
    2. Arganda-Carreras I
    3. Frise E
    4. Kaynig V
    5. Longair M
    6. Pietzsch T
    7. Preibisch S
    8. Rueden C
    9. Saalfeld S
    10. Schmid B
    11. Tinevez JY
    12. White DJ
    13. Hartenstein V
    14. Eliceiri K
    15. Tomancak P
    16. Cardona A

    (2012) Fiji: an open-source platform for biological-image analysis

    Nature Methods 9:676–682.

    https://doi.org/10.1038/nmeth.2019

    • PubMed
    • Google Scholar
37. 1. Suresh A
    2. Dunaevsky A

    (2017) Relationship between synaptic AMPAR and spine dynamics: impairments in the FXS mouse

    Cerebral Cortex 27:4244–4256.

    https://doi.org/10.1093/cercor/bhx128

    • PubMed
    • Google Scholar
38. 1. Suzuki K
    2. Tsunekawa Y
    3. Hernandez-Benitez R
    4. Wu J
    5. Zhu J
    6. Kim EJ
    7. Hatanaka F
    8. Yamamoto M
    9. Araoka T
    10. Li Z
    11. Kurita M
    12. Hishida T
    13. Li M
    14. Aizawa E
    15. Guo S
    16. Chen S
    17. Goebl A
    18. Soligalla RD
    19. Qu J
    20. Jiang T
    21. Fu X
    22. Jafari M
    23. Esteban CR
    24. Berggren WT
    25. Lajara J
    26. Nuñez-Delicado E
    27. Guillen P
    28. Campistol JM
    29. Matsuzaki F
    30. Liu GH
    31. Magistretti P
    32. Zhang K
    33. Callaway EM
    34. Zhang K
    35. Belmonte JC

    (2016) In vivo genome editing via CRISPR/Cas9 mediated homology-independent targeted integration

    Nature 540:144–149.

    https://doi.org/10.1038/nature20565

    • PubMed
    • Google Scholar
39. 1. Tan HL
    2. Roth RH
    3. Graves AR
    4. Cudmore RH
    5. Huganir RL

    (2020) Lamina-specific AMPA receptor dynamics following visual deprivation in vivo

    eLife 9:e52420.

    https://doi.org/10.7554/eLife.52420

    • PubMed
    • Google Scholar
40. 1. Thomas GM
    2. Lin DT
    3. Nuriya M
    4. Huganir RL

    (2008) Rapid and bi-directional regulation of AMPA receptor phosphorylation and trafficking by JNK

    The EMBO Journal 27:361–372.

    https://doi.org/10.1038/sj.emboj.7601969

    • PubMed
    • Google Scholar
41. 1. Tyagarajan SK
    2. Ghosh H
    3. Yévenes GE
    4. Nikonenko I
    5. Ebeling C
    6. Schwerdel C
    7. Sidler C
    8. Zeilhofer HU
    9. Gerrits B
    10. Muller D
    11. Fritschy JM

    (2011) Regulation of GABAergic synapse formation and plasticity by GSK3β-dependent phosphorylation of gephyrin

    PNAS 108:379–384.

    https://doi.org/10.1073/pnas.1011824108

    • PubMed
    • Google Scholar
42. 1. Wang H
    2. Yang H
    3. Shivalila CS
    4. Dawlaty MM
    5. Cheng AW
    6. Zhang F
    7. Jaenisch R

    (2013) One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering

    Cell 153:910–918.

    https://doi.org/10.1016/j.cell.2013.04.025

    • PubMed
    • Google Scholar
43. 1. Willems J
    2. de Jong APH
    3. Scheefhals N
    4. Mertens E
    5. Catsburg LAE
    6. Poorthuis RB
    7. de Winter F
    8. Verhaagen J
    9. Meye FJ
    10. MacGillavry HD

    (2020) ORANGE: a CRISPR/Cas9-based genome editing toolbox for epitope tagging of endogenous proteins in neurons

    PLOS Biology 18:e3000665.

    https://doi.org/10.1371/journal.pbio.3000665

    • PubMed
    • Google Scholar
44. 1. Wolter JM
    2. Mao H
    3. Fragola G
    4. Simon JM
    5. Krantz JL
    6. Bazick HO
    7. Oztemiz B
    8. Stein JL
    9. Zylka MJ

    (2020) Cas9 gene therapy for angelman syndrome traps Ube3a-ATS long non-coding RNA

    Nature 587:281–284.

    https://doi.org/10.1038/s41586-020-2835-2

    • Google Scholar
45. 1. Yang H
    2. Wang H
    3. Shivalila CS
    4. Cheng AW
    5. Shi L
    6. Jaenisch R

    (2013) One-step generation of mice carrying reporter and conditional alleles by CRISPR/Cas-mediated genome engineering

    Cell 154:1370–1379.

    https://doi.org/10.1016/j.cell.2013.08.022

    • PubMed
    • Google Scholar
46. 1. Zhang Y
    2. Cudmore RH
    3. Lin DT
    4. Linden DJ
    5. Huganir RL

    (2015) Visualization of NMDA receptor-dependent AMPA receptor synaptic plasticity in vivo

    Nature Neuroscience 18:402–407.

    https://doi.org/10.1038/nn.3936

    • PubMed
    • Google Scholar

Author details

1. Huaqiang Fang

   1. Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, United States
   2. Kavli Neuroscience Discovery Institute, Johns Hopkins University, Baltimore, United States
   3. PKU-Nanjing Institute of Translational Medicine, Nanjing, China
   4. Research Unit of Mitochondria in Brain Diseases, Chinese Academy of Medical Sciences, Nanjing, China

   Present address

   1. PKU-Nanjing Institute of Translational Medicine, Nanjing, China
   2. Research Unit of Mitochondria in Brain Diseases, Chinese Academy of Medical Sciences, Nanjing, China

   Contribution

   Conceptualization, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Contributed equally with

   Alexei M Bygrave

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2375-859X

2. Alexei M Bygrave

   Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, United States

   Contribution

   Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Contributed equally with

   Huaqiang Fang

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2291-923X

3. Richard H Roth

   Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, United States

   Contribution

   Validation, Investigation, Visualization, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6855-999X

4. Richard C Johnson

   Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, United States

   Contribution

   Validation, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

5. Richard L Huganir

   1. Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, United States
   2. Kavli Neuroscience Discovery Institute, Johns Hopkins University, Baltimore, United States

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Project administration, Writing - review and editing

   For correspondence

   rhuganir@jhmi.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9783-5183

• 19,967

  views

• 1,722

  downloads

• 34

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.