Skip to main content
BMC Biotechnology logoLink to BMC Biotechnology
. 2025 Aug 12;25:83. doi: 10.1186/s12896-025-01021-0

Optimization of in vivo delivery methods and their applications in seminiferous tubules of mice

Xinyuan Dai 1,✉, Fangzhu Wang 1,2, Siting Wang 1
PMCID: PMC12341149  PMID: 40796825

Abstract

This study aims to explore and optimize the conditions for in vivo transfection and gene editing, emphasizing their potential applications in the treatment of infertility in mice. Our findings indicate that physical transfection can be effective in the dynamic fluid environment of mouse seminiferous tubules, with electroporation achieving transfection in multilayered cell tissues. Using the mTmG fluorescence reporter system, we visually assessed the efficiency of electroporation-based transfection and observed stable gene editing outcomes across different individuals. Additionally, we achieved effective transfection of germ cells in vivo for the clinical application of gene tools. We further investigated the impact of various delivery methods and molecular methods on transfection efficacy, revealing that RNP technology is adaptable and efficient in vivo, particularly in the context of treating hereditary diseases. We attempted to leverage gene editing techniques to address spermatogenesis blockage at different stages in Ythdc2-KO and CK137956-KO mice. While we did not succeed in rescuing spermatogenic blockage in Ythdc2 KO mice, the treatment of CK137956 KO mice yielded significant physiological responses. These results could be beneficial for the optimization of in vivo gene editing technologies for clinical applications.

Graphical Abstract

graphic file with name 12896_2025_1021_Figa_HTML.jpg

Supplementary Information

The online version contains supplementary material available at 10.1186/s12896-025-01021-0.

Keywords: Gene therapy, Meiosis, Male infertility, CRISPR/Cas9

Introduction

Gene editing has emerged as a widely utilized technique in both scientific research and clinical applications [1, 2]. DNA mutations within the human genome are closely linked to various diseases and phenotypes in individuals [3]. The goal of gene editing in medical therapies is to modify disease phenotypes on the basis of the genetic information of pathogenic genes. Currently, the primary approaches to gene therapy include cell therapy and in vivo gene editing [4, 5]. In cell therapy, various gene editing methods are employed to treat target cells phenotypically or to enhance their therapeutic functions. The appropriately modified cells are then transplanted into the patient’s body to achieve the desired therapeutic outcome. However, the necessity for cell implantation currently limits this approach to certain diseases.

The effective delivery methods for in vivo gene editing fall into three main categories: viral delivery, material encapsulation, and physical transfection [5–8]. While viral delivery systems can accommodate a broad range of therapeutic applications, they still face challenges concerning safety, immunogenicity, cost, and lengthy synthesis times [9]. Material encapsulation methods often suffer from low efficiency, high costs, and potential cytotoxicity. Physical transfection is characterized by rapid application and good safety profiles; however, it can be technically challenging and may yield inconsistent transfection efficiency. Although there have been advancements in gene therapy utilizing physical transfection, comprehensive evaluations demonstrating superior efficiency or optimization of specific treatment conditions are lacking.

Infertility represents a significant reproductive health issue, with the prevalence of affected individuals steadily increasing in recent years [10, 11]. Male infertility can result from various factors, including hypogonadism, erectile dysfunction, and ejaculatory disorders [12]. The incidence of male patients experiencing azoospermia due to spermatogenesis arrest is also increasing. These individuals often face difficulties in achieving biological offspring through existing assisted reproductive technologies. Numerous genes responsible for spermatogenesis arrest have been identified, and mutations in these genes can serve as pathogenic loci leading to male infertility [11, 13].

In our study, we investigated the transfection efficiency and conditions associated with in vivo electroporation in mouse testes and subsequently optimized these conditions, resulting in a significant increase in transfection efficiency. We examined the transfection efficiency using four different delivery formats: plasmid, protein, ribonucleoprotein (RNP), and RNA. Once stable editing efficiency was achieved, we proceeded to repair infertility in Ythdc2−/− mice with spermatogenesis arrest at the spermatocyte stage and in CK137956−/− mice with arrest at the round spermatid stage. After repair, spermatogenesis was not restored in Ythdc2−/− mice, whereas CK137956−/− mice presented advanced spermatogenesis, with sperm detected in both the testes and the epididymal caput. This study presents an organized approach to optimizing treatment conditions and summarizes suitable gene delivery methods for electroporation transfection.

Methods

Animals

All animals included in this study were bred and housed at the Animal Center of Nanjing Medical University under specific pathogen-free (SPF) conditions and maintained on a 12-hour light/dark cycle. All animal experiments were conducted in accordance with the guidelines approved by the Animal Protection and Use Committee of Nanjing Medical University. The animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC), Approval No. IACUC-2,406,089.

Mouse seminiferous tubule transfection by electroporation

The mice were anaesthetized via intraperitoneal injection of lidocaine until they exhibited no discernible response to external stimuli and their vital signs stabilized. Following disinfection with 75% ethanol, a 1 cm incision was made to access the abdominal cavity, and sterile filter paper was placed around the incision site. The testes and surrounding fat pads were gently extracted from the abdominal cavity and positioned on filter paper, with normal saline applied to maintain moisture. The efferent ductules of the testes were clamped with forceps, and a glass needle was inserted for injection purposes. Electrode forceps connected to the ECM 830 square wave electroporation device were applied on both sides of the testes to administer electroporation stimulation, consisting of 8 pulses at a rate of 50 ms per pulse.

Development of CRISPR/Cas9 knockout mice

We designed two specific guide RNAs (sgRNAs) targeting exon 3 of CK137956 and exon 5 of Ythdc2. Following in vitro transcription, these sgRNAs were coinjected with spCas9 mRNA into fertilized mouse eggs. The injected fertilized eggs were then incubated and implanted into the uteri of female mice to generate offspring. These offspring were subsequently backcrossed with wild-type C57BL/6J mice to establish stable knockout mouse strains.

Immunofluorescence or histochemical staining

Tissues used for immunofluorescence or histochemical staining were fixed in 4% paraformaldehyde (PFA) or modified Davidson’s fluid (mDF) for more than 24 h, followed by paraffin embedding at 60 °C. After solidification, the tissues were sectioned at a thickness of 5 μm. The sections were then rehydrated via a graded ethanol series and subjected to antigen retrieval with sodium citrate solution. After antigen retrieval, the sections were blocked with sheep serum for at least 2 h.

The sections were incubated with primary antibody overnight at 4 °C, followed by three gentle washes with PBS buffer (15 min each). The secondary antibody was then applied, and the samples were incubated at room temperature for 2 h. After another round of washing, the slides were mounted with 50% glycerol.

Endogenous catalase activity was quenched by incubating the slides in 3% H₂O₂ prior to antigen retrieval. Following secondary antibody incubation, the chromogenic solution was applied, developed for 3–4 min, and then washed away. Finally, the nuclei were counterstained with hematoxylin, and the slides were sealed with dehydrated resin.

Cell culture and in vitro editing efficiency evaluation

In vitro experiments were conducted using 293FT cells and mTmG knock-in mouse embryonic fibroblasts (MEFs) for editing tests. All the cell lines were cultured in high-glutamine DMEM (Gibco) supplemented with 10% fetal bovine serum and 1x penicillin‒streptomycin. The cultured cells were electroporated via the Lonza Nucleofector. Drug screening was performed 24 h postelectroporation.

Forty-eight hours after editing, the editing status of the cells was assessed. Fluorescence observations were carried out using mTmG-MEFs. 293FT cells were digested, centrifuged, and resuspended in an alkaline solution. Following lysis at 65 °C, target DNA fragments were extracted for PCR amplification. The PCR products were then analysed via Sanger sequencing and TIDE analysis, or they were annealed and digested with T7 endonuclease I for gel electrophoresis.

Flow cytometry

The cultured cells were digested with trypsin, centrifuged, resuspended in phosphate-buffered saline (PBS), and filtered through a 400-mesh filter before analysis. For mouse testicular tissue, the capsule was dissected, and the tissue was cut into small pieces. These pieces were then added to serum-free DMEM containing collagenase and incubated in a 37 °C water bath for 10 min. The tissue was gently shaken to facilitate the dispersion of fine tubules, after which the supernatant was discarded. Trypsin was added, and the mixture was incubated in a 37 °C water bath for an additional 10 min. The cells were subsequently gently pipetted to achieve further dispersion, centrifuged, resuspended in PBS, and passed through a 400-mesh filter before analysis.

CASA

After the mice were sacrificed, the epididymal tail was excised and cut longitudinally. The mixture was placed in prewarmed modified human tubal fluid (mHTF) containing 10% fetal bovine serum and incubated in a 37 °C water bath for 5 min. A drop of the supernatant was then collected and analysed via a prewarmed computer-assisted sperm analysis (CASA) instrument.

Transcription in vitro by the T7 enzyme

The target sgRNA and T7 promoter were coamplified via high-fidelity enzymes, and the resulting fragment was purified. In an RNA-free environment, transcription was performed via the mMESSAGE mMACHINE™ T7 kit at 37 °C for 2 h. After confirming the presence of the target band via gel electrophoresis, the product was purified via a filter column and eluted in RNA-free double-distilled water.

For mRNA transcription, the target fragment was cloned between the T7 promoter and the PmeI restriction site on a plasmid. Prior to transcription, the 3’ end of the coding sequence was cleaved via the PmeI endonuclease. The gel-purified fragment was then transcribed at 37°C for 2 hours via the mMESSAGE mMACHINE™ T7 ULTRA Kit in an RNA-free environment, with a poly(A) tail added to the 3’ end of the mRNA. After the correct product was verified through gel electrophoresis, the mRNA was purified via a filter column and eluted in RNA-free double-distilled water.

Statistical analysis

All the experiments were independently replicated multiple times (n ≥ 3), and the experimental data were statistically analysed via ordinary one-way ANOVA via GraphPad Prism 9.0.

Results

In vivo conditions for transfection with the EGFP-N1 plasmid

To establish a direct and effective method for evaluating in vivo transfection efficiency, we utilized the EGFP N1 plasmid as a positive control in the transfection experiments. The EGFP N1 plasmid was dissolved in the SE Cell Line 4D-Nucleofector™ X Kit S and injected into the seminiferous tubules of the testis via microinjection, targeting germ cells for transfection (Fig. 1A, Su Fig. 1A). We injected EGFP N1 in an electroporation system at a final concentration of 1 µg/µL into the testes of 3- to 5-week-old mice and performed square-wave-assisted electroporation at three voltage settings: 40 V, 65 V, and 75 V, n = 3 for each condition. The transfected testes displayed notable fluorescence when examined under a fluorescence stereomicroscope. Testes at 40 V presented localized and faint fluorescence, whereas those at 65 V presented uniform, widespread fluorescence, both during in vivo development and in vitro culture. Conversely, testes at 75 V presented concentrated fluorescence, particularly along the seminiferous tubules (Fig. 1B, Su Fig. 1B).

Fig. 1.

Fig. 1

In situ transfection strategy for EGFP-N1. A. In situ EGFP-N1 transfection schedule. B. EGFP-N1-transfected testis whole mounts under conditions; scale bar = 50 μm. C. EGFP-N1-transfected testes positive cell counting data and whole-mount slices that were analysed via one-way ANOVA, n = 3 for each condition, P = 0.0015. Scale bar = 200 μm. D. EGFP-N1-transfected testis immunofluorescence stain, anti-GFP & DDX4. Scale bar = 50 μm. E. EGFP-N1-transfected testis IHC, anti-GFP. Scale bar = 50 μm

Through Hoechst nuclear staining of frozen sections, we observed that the EGFP fluorescence was confined to the lumen of the seminiferous tubules (Fig. 1B). On this basis, we analysed frozen sections of the testes and evaluated the transfection efficiency by calculating the ratio of EGFP-positive cells to the total area of the testicular cross section, n = 3 for each condition. This analysis, which was conducted via ordinary one-way ANOVA, revealed a significant increase in the transfection efficiency at 65 V (P = 0.0015) compared with the other voltage conditions (Fig. 1C; Table 1).

Table 1.

EGFP-N1 transfected testis positive efficiency counting data

Condition Signals Area(mm^2) Signal/area
EGFP N1 40v-1 124 3.718113 33.35025052
40v-2 3 0.569014 5.272278011
40v-3 51 2.596721 19.64015387
65v-1 130 1.8013 72.17009937
65v-2 133 1.995745 66.64178039
65v-3 158 1.828081 86.42943064
75v-1 58 1.234 47.00162075
75v-2 25 3.117269 8.019840444
75v-3 64 1.052251 60.82199019

To characterize the transfected cells further, we costained the EGFP signal with germ cell markers such as DDX4, SCP3, and PLZF via immunofluorescence. The poststaining results revealed abundant colocalization of the transfected plasmid with DDX4-positive cells, whereas fewer colocalized signals were observed with SCP3 and PLZF (Fig. 1D, Su Fig. 1D). Additionally, immunohistochemical staining revealed that more transfection signals were localized within Sertoli cells, indicating that further optimization of the transfection conditions is needed (Fig. 1E).

In addition to physical transfection methods, we explored liposome-based transfection. We premixed an equivalent amount of plasmid with Lipofectamine 2000 and injected this mixture into the mouse testes (n = 3); however, no positive signals were detected (Su Fig. 1C). Given the substantial volume of fluid flow in the seminiferous tubule lumen and the surrounding Sertoli cell matrix, which limits liposome adhesion, we opted to pursue the electroporation method for further experiments [14].

Gene editing verification system by mTmG

Following our investigation of the transfection efficiency under various conditions and statistical analyses, we further explored the in vivo editing efficiency. For this purpose, we employed a dual-fluorescence knock-in (KI) system, Membrane-tdTomato-Membrane-GFP (mTmG), as the reporting system for gene editing [15]. This system emits red fluorescence from tdTomato in the absence of editing. Upon targeting the 5’ end of tdTomato and the 3’ end of the terminator with the CRISPR system, nonhomologous end joining occurs at the end of the CAG promoter and at the beginning of EGFP, enabling normal expression of EGFP (Fig. 2A).

Fig. 2.

Fig. 2

Exploration of in vivo gene editing efficiency in mTmG knock-in mice. A. mTmG-edited strategy involving CRISPR spCas9. B. mTmG MEF edited by CRISPR spCas9 RNP. Scale bar = 50 μm. C. mTmG Testis frozen section edited by CRISPR spCas9 RNP. Scale bar = 50 μm. D. Transfected mTmG testis positive efficiency counting and whole mount slices were analysed via ordinary one-way ANOVA, n = 3 for each condition, P = 0.0003, scale bar = 200 μm. E. mTmG-transfected testes were subjected to immunohistochemistry and immunofluorescence staining with anti-GFP and anti-DDX4 antibodies. Scale bar = 50 μm

We first assessed the knockout effect of the CRISPR system in mTmG knock-in mouse embryonic fibroblasts (MEFs). As expected, untransfected MEFs presented red fluorescence, whereas a substantial number of the transfected cells presented only green fluorescence (Fig. 2B). After confirming the feasibility of this system for evaluating editing efficiency, we proceeded to transfect the CRISPR editing system into the testes of mTmG mice. To control expression levels and ensure the safety of in vivo editing, we utilized the RNA-binding protein (RNP) form of CRISPR for transfection and editing [16–18].

Fourteen days posttransfection, the testes of mTmG mice were frozen, sectioned, and subjected to Hoechst staining for observation. Green fluorescent germ cells were identified in the lumen (Fig. 2C). We further evaluated the transfection efficiency by counting the positive cells and calculating the ratio of positive cells to the cross-sectional area of the testis, n = 3 for each condition (Fig. 2D; Table 2). Compared with that under the other voltages, the editing efficiency under the 65 V electroporation conditions was significantly greater, with no positively edited cells detected at 40 V. Immunohistochemical (IHC) staining revealed a substantial number of seminiferous tubules with edited lumens, with positive cells clustered in these regions (Fig. 2E).

Table 2.

Cas9 RNP transfected mTmG testis positive efficiency counting data

Condition Signals Area(mm^2) Signal/area
mTmG Edited 40v-1 0 4.146934 0
40v-2 0 3.992099 0
40v-3 0 1.984087 0
65v-1 202 2.213197 91.27065109
65v-2 91 1.244884 73.09918033
65v-3 59 0.890316 66.26860575
75v-1 19 1.288786 14.74255617
75v-2 53 1.293164 40.98474749
75v-3 138 4.665034 29.58177797

The costaining of GFP with germ cell markers revealed positive cells at various stages of spermatogenesis (Fig. 2F, Su Fig. 2A). Given that the positive signal was localized to the cell membrane, we further stained the digested testicular cells. Verification of the staining results revealed that many DDX4-positive germ cells expressed EGFP fluorescence (Su Fig. 2B).

Edited mTmG mouse prognoses

To evaluate the sustained effects and subsequent development of physical transfection-based gene editing in live mice, we analysed the subjects six weeks posttransfection. Through anti-GFP immunohistochemistry and immunofluorescence, we observed that the testes of mTmG mice still contained EGFP-positive cells six weeks after transfection, with EGFP-positive sperm detected in the caput. These findings indicate that the gene-edited germ cells produced via this method can generate sperm, confirming the long-lasting effects of the transfection (Fig. 3A, B).

Fig. 3.

Fig. 3

Edited mTmG mouse prognoses. A. Immunohistochemistry of mTmG-transfected testes and caput after 6 weeks and of mTmG-NGN3-Cre testicles. Scale bar = 50 μm. B. Immunofluorescence staining of the mTmG-transfected caput after 6 weeks. anti-GFP. Scale bar = 50 μm. C. Mmunohistochemistry of mTmG-transfected cauda after 6 weeks. Scale bar = 50 μm. D. CASA of the mTmG-transfected caput after 6 weeks, as analysed by an unpaired one-way ANOVA, n = 3, P = 0.5568. E. Pups of mTmG-transfected mouse offspring, n = 3. F. Fertility statistics of wild-type mice and mTmG-transfected mice analysed via an unpaired one-way ANOVA, n = 3, P = 0.7893

Additionally, we crossed NGN3-Cre KI mice or AMH-Cre KI mice with mTmG mice to obtain control groups expressing mG specifically in germ cells or Sertoli cells, respectively. We found that mG expression in germ cells was significantly stronger, whereas AMH-Cre-guided mG expression was weak. Both fluorescence observations and immunohistochemical analyses corroborated this finding, suggesting that the GFP signal we detected originated primarily from edited germ cells (Fig. 3A, Su Fig. 3A, B).

We further searched for EGFP-positive sperm in the cauda of the mice but did not identify any gene-edited positive sperm (Fig. 3C). The absence of detectable fluorescence in the edited testes under a stereomicroscope led us to consider the possibility that low editing efficiency contributed to the failure to detect mature gene-edited sperm (Su Fig. 3A).

Moreover, we investigated whether the cytotoxicity of the injected reagent system, electrical stimulation-induced damage to the blood‒testis barrier, or inflammation adversely affected the health and prognosis of the mice. We performed computer-assisted sperm analysis (CASA) on three mice six weeks after transfection (n = 3), and the results revealed no statistically significant difference in the proportion of motile cells between the wild-type and transfected mice, suggesting that this transfection method had no significant negative effect on the health of the mice (Fig. 3D). Additionally, we mated three transfected mice with female mice to obtain pups, and the number of pups born was not significantly different from that of wild-type mice, n = 3 per group (Fig. 3E, F).

In vivo delivery method optimization

To further enhance editing efficiency, we conducted a transfection in 16-day-old mice. The spermatogenic epithelium of these mice has fewer cell layers, facilitating closer contact between the cells and the electroporation solution. Spermatogonial stem cells, located at the basal layer of the spermatogenic epithelium, benefit from this reduced thickness, making them more accessible for transfection. Additionally, at fewer days, spermatogonial stem cells are more concentrated in the lumen of the seminiferous tubules [19–21].

For 16-day-old mice, we adjusted the electroporation strategy, as the voltage used in cell electroporation is closely related to the distance between electrodes [22–25]. When the electrode spacing is decreased, the voltage must be reduced to prevent cellular damage. Given the smaller testicular diameter of 16-day-old mice, we opted to transfect them at 40 V to minimize congestion and avoid damaging cells and the blood‒testis barrier during in vivo transfection. Immunohistochemical staining revealed that, in contrast to our experiments involving 3- to 5-week-old mice, 40 V successfully delivered CRISPR RNP for editing, resulting in a significantly greater number of cells in the edited lumen compared with our previous methods (Fig. 4A).

Fig. 4.

Fig. 4

In vivo delivery method optimization. A. Immunohistochemistry of mTmG-transfected testes under normal conditions and modified conditions; scale bar = 50 μm. B. mTmG-RNP testis FANC, analysed by an unpaired t test with one-way ANOVA, n = 3, P = 0.0335. The difference was statistically significant. C. Self-expression fluorescence of mTmG-transfected testes under normal conditions and modified conditions. D. Immunohistochemistry of mTmG-transfected testes generated via multiple delivery methods. E. Immunohistochemistry of mTmG-transfected epididymis cells subjected to modified conditions; scale bar = 100 μm

Using flow cytometry for fluorescence cell statistics, we found that the number of transfected positive cells with the improved transfection method was significantly greater than that generated by the previous method (n = 3 per group) (Fig. 4B, Su Fig. 4C). Observations of testis fluorescence indicated that the previous method failed to produce detectable signals under a stereomicroscope, whereas the testes from mice transfected with our improved method displayed prominent transfection-positive tubules (Fig. 4C).

In addition to the RNP form, we explored other delivery methods, including the mRNA and plasmid forms of the CRISPR system, and compared these methods to the RNP editing method. For the plasmid delivery experiments, we utilized Pst1374 NLS-spCas9 and PGl3 2U6 sgRNA plasmids at a 2:1 ratio, which were prepared in Lonza electroporation solution at a final concentration of 3 µg/µL. For the mRNA delivery experiment, 12 µg of spCas9 mRNA and 12 µg of sgRNA were dissolved in Lonza electroporation solution to a total volume of 20 µL. For the RNP delivery experiment, we dissolved 32 µg of purified spCas9 protein and 12 µg of sgRNA in Lonza electroporation solution, resulting in a total volume of 20 µL.

Posttransfection, immunohistochemical staining revealed that the plasmid form of the Cas9 system was ineffective for editing in the testis, whereas the mRNA form produced only a limited number of effectively edited cells within a single lumen, leading to low transfection efficiency (Fig. 4D). Despite several in vitro experiments indicating that the efficiency of plasmid editing is significantly greater than that of RNP, we speculated that the molecular form influences in vivo delivery efficiency [26–28].

We further purified EGFP and examined the transfected testicular slices. Through fluorescence and immunohistochemical observations, we noted that, compared with EGFP N1 transfection, electroporation of EGFP resulted in more uniform signal localization, and the transfection area within a single seminiferous tubule lumen was larger (Su Figs. 1E and 4B).

After enhancing editing efficiency, we sought to observe edited positive sperm in the epididymides of the mice. However, we did not detect any positive sperm in the cauda, and low-magnification observations revealed GFP-positive signals in the epithelial lumen of the epididymis, particularly in the cauda and its surrounding areas. Unlike in the caput, we detected abundant signals indicative of edited sperm. As the lumen expands, the detection of positive sperm becomes increasingly challenging near the cauda (Fig. 4E).

Ythdc2-/- mouse infertility rescue

After evaluating the effects of in vivo gene editing, we aimed to utilize in vivo transfection to repair the infertility phenotype. Given that meiosis is a critical process in spermatogenesis, with the transcriptional products during the spermatocyte stage supporting the division of spermatogenic cells and the specialization of spermatids, we constructed Ythdc2 knockout (KO) mice in which spermatogenesis is arrested at the spermatocyte stage [19, 29]. This model allowed us to observe the potential repair effects of in vivo transfection on infertility, which is blocked at the early stages of meiosis.

We employed the CRISPR system with zygote injection to create infertile mouse lines featuring a stop codon mutation and a 1 bp insertion frameshift mutation in exon 5 (Fig. 5A). sgRNAs were designed for mice with frameshift mutations guided by spCas9, and the target fragment was inserted into a T vector before being cotransferred with the editing system into 293FT cells for editing. The editing efficiency of the Cas9 system was verified via T7 endonuclease I digestion, which yielded relatively good results (Fig. 5B). We subsequently used a double sgRNA system for further experiments.

Fig. 5.

Fig. 5

Ythdc2−/− mouse infertility rescue. A. Ythdc2−/− mouse CRISPR knockout and rescue schedule. B. Ythdc2−/− CRISPR cleavage efficiency test with the T7en1 enzyme. C. Ythdc2−/− mice were subjected to EGFP N1 testis immunofluorescence staining and anti-GFP and anti-PLZF antibodies. Scale bar = 50 μm. D. Ythdc2−/− mice were subjected to EGFP N1 testis immunohistochemistry with an anti-GFP antibody. Scale bar = 200 μm. E. Ythdc2−/− mouse testes were transfected with the CRISPR RNP rescue system, n = 3. Scale bar = 50 μm. F. Ythdc2−/− mouse testes were transfected with the piggyBac-Ythdc2 rescue system; n = 3; scale bar = 50 μm. G. Ythdc2−/− and mTmG knock-in double-genotyping mouse lineage schedule. H. IHC of Ythdc2−/− and mTmG-KO double-genotyping testes transfected with mTmG-targeted Cas9 RNP, anti-GFP, n = 3, scale bar = 50 μm. I. IHC of Ythdc2−/− mouse testes transfected with the piggyBac EGFP and Ythdc2 rescue system, anti-GFP, n = 3; scale bar = 50 μm

To assess the effect of transfection on Ythdc2−/− mouse testes, we used EGFP N1. Posttransfection, the testes exhibited robust fluorescence, showing marked colocalization with the PLZF spermatogonial stem cell marker. Analysis of the entire testicular section revealed that the positively transfected cells were widely distributed (Fig. 5C and D, Su Fig. 5A). Following confirmation of the transfection effect, we proceeded to transfect the CRISPR rescue system into the testes of infertile mice and observed testicular development after 21 days, n = 3 per group. Despite multiple attempts, no cells that had developed beyond the spermatocyte stage were found in the injected mice (Fig. 5E). We also attempted to use the PiggyBac vector to transfer the Ythdc2 expression system, but again, no cells developed into round spermatids after 21 days (Fig. 5F).

We tested the editing efficiency of the stop codon mutation line and performed injections for repair purposes; however, we did not observe any infertility repair in the injected mice (Su Fig. 5B and C). To explore further the reasons for the inability of these mice to repair infertility, we crossed Ythdc2−/− mice with mTmG mice to obtain mTmG knock-in Ythdc2−/− mice, which were then transfected with CRISPR RNPs targeting mTmG, n = 3 per group (Fig. 5G). After 14 days, we detected clusters of mTmG-edited positive cells, indicating that spermatogenic cells in Ythdc2−/− mice could be effectively edited (Fig. 5H).

We also employed the PiggyBac vector to concurrently introduce the Ythdc2 expression system and the EGFP expression system (n = 3). Staining revealed that EGFP was normally expressed; however, spermatogenesis was not promoted (Fig. 5I). Considering the potential impact of transposons on genomic stability during spermatogenesis [30]. We tested a pCMV vector backbone for delivering the Ythdc2 and EGFP expression systems. Although we observed high levels of EGFP expression in germ cells four days posttransfection, the expression levels were markedly reduced eight days later, and spermatogenesis was not promoted (Su Fig. 5D).

Finally, we investigated the use of the plasmid form of the CRISPR system to repair infertility in Ythdc2−/− mice (n = 3), but no therapeutic effect was observed (Su Fig. 5E). We ruled out potential obstacles related to transfection, gene expression, and editing in Ythdc2−/− mice. Given that multiple methods fail to restore fertility, we speculate that the infertility observed in Ythdc2−/− mice may be challenging to rescue [31, 32].

CK137956-/- mouse infertility rescue

To evaluate the efficacy of gene therapy in addressing late-stage sperm development arrest and to investigate treatment outcomes in mice with spermatogenesis blockage at different stages, we conducted infertility treatment on CK137956−/− mice with sperm deformation issues [33]. We initially generated CK137956 exon 3-inserted 1 bp mutant mice through zygote injection via the CRISPR system and assessed the rescue editing efficiency in the 293FT cell line via T7 endonuclease I (Fig. 6A, B). A dual sgRNA strategy was used for injection into infertile CK137956−/− mice, n = 3 per group. Six weeks later, we examined the injected mice and observed that the testes on the injected side were larger than those on the noninjected side, with varying degrees of enlargement among individuals (Fig. 6D, Su Fig. 6A).

Fig. 6.

Fig. 6

CK137956−/− mouse infertility rescue. A. CRISPR-mediated knockout of CK137956 in a−/− mouse and rescue schedule. B. CK137956−/− CRISPR cleavage efficiency test with the T7en1 enzyme. C. CK137956−/− mouse testes and epididymides were transfected via the CRISPR RNP rescue system; scale bar = 50 μm. The block indicates a pathological cell pellet, the arrowhead indicates sperm, and the white arrowhead indicates blocked round sperm, n = 3. D. CK137956−/− mice were transfected with the CRISPR RNP rescue system; the arrowhead indicates that the testicle was rescued, n = 3. E, ICSI cells of CK137956−/− CRISPR rescue sperm; scale bar = 50 μm

Histological examination with HE staining revealed that spermatogenic tubules affected by blockage of sperm development in the testes were affected, resulting in sperm production. In untreated tubules, abnormal cell clusters resembling those found in CK137956−/− mice persisted. The analysis of sections revealed considerable variability in the sperm count and the number of spermatogenic tubules capable of producing sperm across individuals. Both deformed sperm and abnormal round sperm cells following spermatogenesis blockage were observed in the caput (Fig. 6C, Su Fig. 6A). Upon assessing sperm maturation, we detected abnormal round sperm cells in the cauda of the mice but no mature sperm, which is consistent with the results from the mTmG experiment; no edited positive cells were identified in the cauda (Su Fig. 6C).

We performed intracytoplasmic sperm injection (ICSI) via edited sperm cells to successfully develop embryos to the two-cell stage. We implanted 7 or 8 embryos into each of three recipient female mice and simultaneously implanted 6 or 7 wild-type embryos into each female to increase the implantation rate (Fig. 6E). To trace the origin of the embryos, we bred CK137956 mutation heterozygous mice with mTmG mice to generate homozygous knock-in mice for treatment (Su Fig. 6B). After the three female mice became pregnant, eight offspring were produced. However, genotyping via PCR revealed that no offspring from edited or repaired sperm were generated (Su Fig. 6E).

In addition, we attempted to perform dual-phenotype gene editing in vivo in mTmG KI and CK137956 KO double-genotype mice (n = 3). Poststaining, we identified mTmG-edited positive cell clusters; however, their numbers were low, and no tubules treated with spermatogenesis blockade were detected. Moreover, no sperm were found in the epididymis (Su Fig. 6D). These findings suggest ongoing challenges associated with conducting dual-gene editing experiments in vivo.

We also explored the repair of infertility in CK137956 mice via the plasmid form of the CRISPR system (n = 3). These results are consistent with those of the mTmG experiment, as the plasmid form of CRISPR failed to remedy infertility in these mice (Su Fig. 6F). Our data suggest that, compared with nucleic acid forms, RNP delivery methods exhibit greater preference and adaptability for in vivo applications.

Discussion

Currently, there is a growing body of research on gene editing in eukaryotic cells [2, 5, 34, 35]. However, in vivo transfection presents greater difficulties and challenges than experiments conducted at the cellular level [6]. While numerous studies have been dedicated to optimizing editing efficiency and immunogenicity through the development of gene tools, few have addressed the optimization of transfection conditions and surgical methods in vivo [36–38]. Furthermore, there is a lack of assessment and control of the transfection efficiency. Therefore, we aimed to explore, optimize, and evaluate the conditions for in vivo transfection and gene editing. Compared with viral transfection or material-coated transfection, traditional physical transfection methods demonstrate better efficiency and ease of operation while also minimizing immune responses, inflammatory reactions, and potential bleeding complications [6, 7, 39].

Our research results indicate that although the seminiferous tubules of mice exist in a dynamic liquid environment, physical transfection can still achieve effective results. Unlike Lip2000, which demonstrates effective transfection at the cellular level, achieving effective transfection in vivo remains challenging. Our data further suggest that electroporation can be used to transfect non-single-layered cell tissues effectively in vivo. The mTmG transfection results indicate that the fluorescence staining of PLZF, along with the persistence of positive germ cells six weeks posttransfection, demonstrate that electroporation can successfully target spermatogonial stem cells located deep within the seminiferous epithelium and shielded by supporting cells. The ability to transfect tissues comprising multiple layers of cells has significant implications for the clinical application of gene tools, suggesting that electroporation may also facilitate the transfection of solid organs and tumors, similar to biological methods such as viral transfection.

A persistent challenge in in vivo transfection is the lack of effective methodologies to evaluate the transfection efficiency and editing efficacy of targeted cell populations. Currently, editing efficiency following the transfection of all tissue cells is primarily assessed through high-throughput sequencing (HTS) or a combination of transcriptome sequencing and bioinformatics analysis [40]. We can directly observe and compare transfection effects under various conditions and analyse the distribution of transfection effects and tissue localization through the mTmG fluorescence reporter system in conjunction with flow cytometry and fluorescence staining, especially in cases where the proportion of transfected cells in vivo is rare. In contrast to the anticipated results, the mTmG experiment revealed that gene editing efficiency following electroporation transfection is stable across different individuals and highly reproducible. By transfecting 16-day-old mice—characterized by thinner seminiferous epithelium and a higher concentration of stem cells—and controlling the voltage, we significantly enhanced the transfection efficiency. This suggests the need for dual optimization of gene editing tools and surgical techniques that consider the characteristics of the transfected tissue to achieve clinical objectives.

Moreover, we found that, in addition to the delivery method, the molecular form of the delivered molecules plays an important role in the transfection efficacy. Some previous studies have indicated that plasmids often outperform mRNAs and, in some cases, proteins in cellular transfection attempts [28]. However, our in vivo transfection study yielded different results. RNP transfection demonstrated effective editing, with successful transfection distributed within the area affected by the injection of the transfection system. In contrast, a mixed system of mRNA and sgRNA succeeded in transfecting only one tubule per testis cross-section, whereas CRISPR plasmids failed to achieve effective editing in vivo. This diverges from earlier findings, suggesting that the differences in delivery forms may have an impact on the treatment of specific diseases.

A comparison of the EGFP- and EGFP-N1 plasmid-transfected testicular sections revealed that the protein exhibited a uniform distribution within spermatogenic epithelial cells after electroporation. Conversely, the seminiferous tubules effectively transfected with the EGFP N1 plasmid were localized primarily to the shallow region at the center of the electrical stimulation contact surface, and the number of these tubules was relatively small, similar to the localization of CRISPR RNA system editing, with positive results.

In our study, we also sought to explore the potential of this method for treating infertility. We performed transfection injections into Ythdc2 KO and CK137956 KO mice at 16 days post birth. Our experimental design aimed to observe the outcomes in mice with spermatogenesis arrested at different stages following in vivo transfection treatment. Ythdc2-knockout mice exhibit a block of spermatocytes at the pachytene stage. After the EGFP N1 plasmid was transfected, EGFP-positive cells were distributed in the spermatogonia at the base of the Ythdc2-knockout mouse spermatogenic epithelium, suggesting that germ cell transfection is effective. The mTmG reporter system demonstrated that effective editing can be accomplished before the spermatocyte block stage. Afert cell-level Ythdc2 mutation side editing efficiency was confirmed; despite our efforts to treat Ythdc2−/− mice with various gene editing treatments, we were unable to rescue the spermatogenic block. To explore the factors that contribute to suboptimal treatment, we attempted to compensate for this deficiency by inserting the promoter and coding sequence of Ythdc2 from the mouse genome into PiggyBac vectors and nongenome-integrating vectors for transfection into spermatogenic cells, but we still failed to achieve rescue. Given that Ythdc2 is a reader of m6A RNA modifications and that the Ythdc2 protein plays a regulatory role in multiple stages of spermatogenesis at the transcriptome level [41], considering the multiple undesired therapeutic effects of Ythdc2-KO mice, including gene editing and gene addition, we speculate that current treatment methods may not effectively restore spermatogenesis.

Conversely, the treatment of CK137956 mice was relatively successful. CK137956-knockout mice exhibit a block of spermatocytes at the round spermatozoa stage. The injected testes exhibited significant enlargement, with sperm detected in both the testis and caput, although no sperm were found in the cauda. Unlike the mTmG situation, CK137956 rescue showed marked individual variability. Compared with that on the contralateral side, the enlargement of the injected testes was inconsistent. Histological analysis revealed variability in the number of tubules producing sperm and the quantity of sperm within those tubules, with no sperm found in the cauda. Similarly, in the mTmG experiment, markers using EGFP from edited cells revealed a rich presence of EGFP-positive cells in the tubule epithelium near the cauda region of the epididymis, where edited positive sperm were scarcely detected. In contrast, the epithelial tubules at the starting end, devoid of EGFP, contained a considerably high number of EGFP-positive sperm. Further research on the mechanism by which edited sperm do not mature in the cauda could be beneficial for improving therapeutic methods and evaluating long-term therapeutic outcomes.

Numerous aspects of this study merit further exploration. Although we achieved relatively high transfection efficiency and gained insights into the influence of specific conditions, there remains potential for improvement in in vivo transfection methodologies. While we have investigated conditions related to DNA double-strand break (DSB)-associated in vivo gene editing, methods for homologous recombination and point editing still require additional research [38, 42, 43]. In the in vivo context, more complex editing techniques need further investigation and validation, along with the design of more suitable reporter system mouse models. Furthermore, compared with other transfection methods, the electroporation method lacks specificity for cell types and needs to be optimized with both transfection methods and gene vehicles.

Conclusion

In this study, we optimized an in vivo gene editing delivery method using electroporation and CRISPR/Cas9-based technologies to treat male infertility caused by spermatogenesis blockage at different stages. Our findings suggest the efficacy of electroporation as an effective delivery method, enabling high transfection efficiency in the testes, even for non-single-layer tissues such as seminiferous tubules. Effective transfection was observed in spermatogonial stem cells, with persistent effects even at six weeks posttransfection.

Gene therapy with CRISPR/Cas9 resulted in the restoration of sperm production in CK137956−/− mice, indicating the potential of this approach for treating infertility in cases of late-stage spermatogenesis blockade. However, the inability to rescue spermatogenesis in Ythdc2−/− mice suggests that different pathogenic mutation genes may present different challenges in gene therapy.

While promising, the study also identifies several areas requiring further optimization, such as base editing for gene therapy and the evaluation of long-term therapeutic outcomes. Future research should focus on improving the efficiency and precision of gene editing tools, developing more suitable reporter systems, and exploring the impact of specific electroporation methods to selectively target different cell populations in spermatogenesis. These advancements will be beneficial for translating gene therapy into clinical applications for male infertility treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 (28.5MB, docx)
Supplementary Material 2 (871.7KB, zip)

Acknowledgements

We are grateful to the experimental platform of Nanjing Medical University and the Animal Experiment Center of Nanjing Medical University for their support.

Author contributions

WF processd plasmid construction and protein purification. WS processed the animal experience. DX processed the research design and all the experiments and was a major contributor to the writing of the manuscript. All the authors read and approved the final manuscript.

Funding

We are grateful to the Bin Shen research group of Nanjing Medical University for their financial assistance.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC), Approval No. IACUC-2406089.

Consent for publication

Not applicable.

Language editing

We are grateful to the language editing platform of Rubriq.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Pacesa M, Pelea O, Jinek M. Past, present, and future of CRISPR genome editing technologies. 2024. [DOI] [PubMed]
  • 2.Derry WB. CRISPR: development of a technology and its applications. FEBS J. 2021;288(2):358–9. [DOI] [PubMed] [Google Scholar]
  • 3.Claussnitzer M, Cho JH, Collins R, et al. A brief history of human disease genetics. Nature. 2020;577(7789):179–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Chavez M, Chen X, Finn PB, Qi LS. Advances in CRISPR therapeutics. Nat Rev Nephrol. 2023;19(1):9–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Komor AC, Badran AH, Liu DR. CRISPR-Based technologies for the manipulation of eukaryotic genomes. Cell. 2017;168(1–2):20–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Raguram A, Banskota S, Liu DR. Therapeutic in vivo delivery of gene editing agents. Cell. 2022;185(15):2806–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Yin H, Kauffman KJ, Anderson DG. Delivery technologies for genome editing. Nat Rev Drug Discov. 2017;16(6):387–99. [DOI] [PubMed] [Google Scholar]
  • 8.Madigan V, Zhang F, Dahlman JE. Drug delivery systems for CRISPR-based genome editors. Nat Rev Drug Discov. 2023;22(11):875–94. [DOI] [PubMed] [Google Scholar]
  • 9.van Haasteren J, Li J, Scheideler OJ, Murthy N, Schaffer DV. The delivery challenge: fulfilling the promise of therapeutic genome editing. Nat Biotechnol. 2020;38(7):845–55. [DOI] [PubMed] [Google Scholar]
  • 10.Tharakan T, Bettocchi C, Carvalho J, et al. Male sexual and reproductive Health-Does the urologist have a role in addressing gender inequality in life expectancy?? Eur Urol Focus. 2020;6(4):791–800. [DOI] [PubMed] [Google Scholar]
  • 11.Sang Q, Ray PF, Wang L. Understanding the genetics of human infertility. Science. 2023;380(6641):158–63. [DOI] [PubMed] [Google Scholar]
  • 12.Xavier MJ, Salas-Huetos A, Oud MS, Aston KI, Veltman JA. Disease gene discovery in male infertility: past, present and future. Hum Genet. 2021;140(1):7–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ferlin A, Dipresa S, Delbarba A, et al. Contemporary genetics-based diagnostics of male infertility. Expert Rev Mol Diagn. 2019;19(7):623–33. [DOI] [PubMed] [Google Scholar]
  • 14.Waites GM, Gladwell RT. Physiological significance of fluid secretion in the testis and blood-testis barrier. Physiol Rev. 1982;62(2):624–71. [DOI] [PubMed] [Google Scholar]
  • 15.Muzumdar MD, Tasic B, Miyamichi K, Li L, Luo L. A global double-fluorescent Cre reporter mouse. Genesis. 2007;45(9):593–605. [DOI] [PubMed] [Google Scholar]
  • 16.Burger A, Lindsay H, Felker A, et al. Maximizing mutagenesis with solubilized CRISPR-Cas9 ribonucleoprotein complexes. Development. 2016;143(11):2025–37. [DOI] [PubMed] [Google Scholar]
  • 17.Jinek M, Jiang F, Taylor DW, et al. Structures of Cas9 endonucleases reveal RNA-mediated conformational activation. Science. 2014;343(6176):1247997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kim S, Kim D, Cho SW, Kim J, Kim J. Highly efficient RNA-guided genome editing in human cells via delivery of purified Cas9 ribonucleoproteins. Genome Res. 2014;24(6):1012–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chen Y, Zheng Y, Gao Y, et al. Single-cell RNA-seq uncovers dynamic processes and critical regulators in mouse spermatogenesis. Cell Res. 2018;28(9):879–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ernst C, Eling N, Martinez-Jimenez CP, Marioni JC, Odom DT. Staged developmental mapping and X chromosome transcriptional dynamics during mouse spermatogenesis. Nat Commun. 2019;10(1):1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bellve AR, Cavicchia JC, Millette CF, et al. Spermatogenic cells of the prepuberal mouse. Isolation and morphological characterization. J Cell Biol. 1977;74(1):68–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Tsong TY. Electroporation of cell membranes. Biophys J. 1991;60(2):297–306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Chen C, Smye SW, Robinson MP, Evans JA. Membrane electroporation theories: a review. Med Biol Eng Comput. 2006;44(1–2):5–14. [DOI] [PubMed] [Google Scholar]
  • 24.Andre F, Mir LM. DNA electrotransfer: its principles and an updated review of its therapeutic applications. Gene Ther. 2004;11(Suppl 1):S33–42. [DOI] [PubMed] [Google Scholar]
  • 25.Kotnik T, Rems L, Tarek M, Miklavcic D. Membrane electroporation and electropermeabilization: mechanisms and models. Annu Rev Biophys. 2019;48:63–91. [DOI] [PubMed] [Google Scholar]
  • 26.Li T, Yang Y, Qi H, et al. CRISPR/Cas9 therapeutics: progress and prospects. Signal Transduct Target Ther. 2023;8(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Lin Y, Wagner E, Lachelt U. Non-viral delivery of the crispr/cas system: DNA versus RNA versus RNP. Biomater Sci. 2022;10(5):1166–92. [DOI] [PubMed] [Google Scholar]
  • 28.Carroll D. Genome engineering with targetable nucleases. Annu Rev Biochem. 2014;83:409–39. [DOI] [PubMed] [Google Scholar]
  • 29.Hsu PJ, Zhu Y, Ma H, et al. Ythdc2 is an N(6)-methyladenosine binding protein that regulates mammalian spermatogenesis. Cell Res. 2017;27(9):1115–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Wang X, Ramat A, Simonelig M, Liu M. Emerging roles and functional mechanisms of PIWI-interacting RNAs. Nat Rev Mol Cell Biol. 2023;24(2):123–41. [DOI] [PubMed] [Google Scholar]
  • 31.Bailey AS, Fuller MT. YTHDC2 serves a distinct late role in spermatocytes during germ cell differentiation. Proc Natl Acad Sci U S A. 2024;121(42):e2309548121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Wojtas MN, Pandey RR, Mendel M, et al. Regulation of m(6)A transcripts by the 3’-->5’ RNA helicase YTHDC2 is essential for a successful meiotic program in the mammalian germline. Mol Cell. 2017;68(2):374–e38712. [DOI] [PubMed] [Google Scholar]
  • 33.Chen Y, Zhang X, Jiang J, et al. Regulation of Miwi-mediated mRNA stabilization by Ck137956/Tssa is essential for male fertility. BMC Biol. 2023;21(1):89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Doudna JA. The promise and challenge of therapeutic genome editing. Nature. 2020;578(7794):229–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ran FA, Hsu PD, Wright J, et al. Genome engineering using the CRISPR-Cas9 system. Nat Protoc. 2013;8(11):2281–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Knott GJ, Doudna JA. CRISPR-Cas guides the future of genetic engineering. Science. 2018;361(6405):866–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Jiang T, Zhang X, Weng Z, Xue W. Deletion and replacement of long genomic sequences using prime editing. Nat Biotechnol. 2022;40(2):227–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Matsoukas IG, Commentary. Programmable base editing of A.T to G.C in genomic DNA without DNA cleavage. Front Genet. 2018;9:21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Taha EA, Lee J, Hotta A. Delivery of CRISPR-Cas tools for in vivo genome editing therapy: trends and challenges. J Control Release. 2022;342:345–61. [DOI] [PubMed] [Google Scholar]
  • 40.Behr M, Zhou J, Xu B, Zhang H. In vivo delivery of CRISPR-Cas9 therapeutics: progress and challenges. Acta Pharm Sin B. 2021;11(8):2150–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Shi H, Wei J, He C. Where, when, and how: Context-Dependent functions of RNA methylation writers, readers, and erasers. Mol Cell. 2019;74(4):640–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Zhao D, Li J, Li S, et al. Glycosylase base editors enable C-to-A and C-to-G base changes. Nat Biotechnol. 2021;39(1):35–40. [DOI] [PubMed] [Google Scholar]
  • 43.Gapinske M, Luu A, Winter J, et al. CRISPR-SKIP: programmable gene splicing with single base editors. Genome Biol. 2018;19(1):107. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (28.5MB, docx)
Supplementary Material 2 (871.7KB, zip)

Data Availability Statement

No datasets were generated or analysed during the current study.


Articles from BMC Biotechnology are provided here courtesy of BMC

RESOURCES