Intraplacental injection of AAV9-CMV-iCre results in the widespread transduction of multiple organs in double-reporter mouse embryos

Abstract

Adeno-associated virus serotype 9 (AAV9) has become a popular tool for gene transfer because of its ability to cross the blood-brain barrier and efficiently transduce genetic material into a variety of cell types. The study utilized GRR (Green-to-Red Reporter) mouse embryos, in which the expression of iCre results in the disappearance of Green Fluorescent Protein (GFP) expression and the detection of Discosoma sp. Red Fluorescent Protein (DsRed) expression by intraplacental injection. Our results demonstrate that AAV9-CMV-iCre can transduce multiple organs in embryos at developmental stages E9.5–E11.5, including the liver, heart, brain, thymus, and intestine. These findings suggest that intraplacental injection of AAV9-CMV-iCre is a viable method for the widespread transduction of GRR mouse embryos.

Introduction

Approximately 350 million people worldwide suffer from various monogenic disorders, and most of these individuals begin to develop symptoms from childhood [1]. Mutations in more than 4,300 genes were found to be associated with 7,300 phenotypes, according to the latest version of the OMIM database [2]. Gene therapy has the potential to provide an effective, one-time treatment for those affected by such conditions, including yet-to-be-born individuals [3]. This growing promise has been driven by advances in precise editing tools that enable more effective genetic intervention [4].

One of the major rationales for in utero therapy is the genetic correction of cells affected by mutations before the monogenic disease progresses [5,6,7]. Indeed, intervention at earlier developmental stages may offer opportunities to correct genetic mutations affecting organ development-particularly in the cases of congenital heart disorders (CHDs), as well as immunological, metabolic, and neurodevelopmental conditions with early developmental onsets [8,9,10,11]. Nevertheless, before these approaches can be applied to human patients, it is crucial to conduct extensive proof-of-concept studies to assess the safety and feasibility of such interventions. Murine models can provide valuable insight into all above mentioned. In particular, the developmental period of E9.5–E11.5 is equivalent to the late first trimester and early second trimester of human pregnancy [12]. However, little attention to these stages has been paid to our knowledge.

The placenta is one of the target organs for human in utero gene therapy, but it can also be the site for direct access to the fetus, especially during earlier development [13,14,15]. Intraplacental injection as a route of delivery was first established in 1979 as a method for hematopoietic stem cell transplantation (HSCs) (Fig. 1) [16]. Utilizing this method, our group previously reported the successful rescue of HSC-lacking mice by allogeneic transplantation [17]. Few papers reporting viral vector delivery by intraplacental injection have been previously published [18, 19]. Türkay and Woo independently demonstrated the successful delivery of lentivirus and adenovirus to mouse embryos at stages E9.5–E12.5. Specific transduction in the heart was observed, and this route of injection was described as systemic However, to the best of our knowledge, no further studies have until now been conducted.

Fig. 1.

Intraplacental injection allows cargo delivery into stage E9.5 embryos. (A) Diagram of intraplacental injection at E9.5. (B) Photo of surgical procedure. Numbers indicate the individual fetus. White arrowhead indicates the site of injection in placenta (C) Photos of Day 9 embryo injected with saline and Indian ink respectively. Ink can be observed in the embryonic heart and, allantois (white arrowheads) and blood vessels.

Adeno-Associated viruses (AAVs) are single-stranded icosahedral viruses, characterized by long-term expression, low immunogenicity, and high engineering capacity to be widely used for gene experiments, including one for editing [20]. Among the variety of serotypes, AAV serotype 9 (AAV9) has shown promising results for systemic gene transfer and is characterized by broad transduction to a variety of cell types and the ability to penetrate the blood-brain barrier (BBB) [21]. However, there is a need for investigating the transduction efficiency of AAV9 in specific cell types and organs to further optimize its use in gene therapy.

In this study, we report the proof-of-concept study of AAV9 delivery by intraplacental route using Cre-recombination in double-reporter mice for identifying target organs. Our results demonstrated the transduction to multiple organs suggesting a promising future for the intraplacental route for AAV9 delivery.

Materials and Methods

Animal experiments

C57BL/6N-Gt(ROSA)26Sor<tm1(CAG-EGFP/DsRed)Utr>/Rbrc (GRR) mice were generated at the Laboratory Animal Resource Center at the University of Tsukuba [22]. All animals were maintained under specific pathogen-free conditions at the University of Tsukuba Laboratory Animal Resource Center. Mouse experiments were performed in compliance with relevant Japanese and institutional laws and guidelines and were approved by the Animal Ethics Committee of the University of Tsukuba. Intraplacental injections were performed as previously described [17].

Plasmid cloning

pAAV-CMV-iCre plasmids were generated by cloning the iCre gene, which was amplified from a pBS-GSGP2A-iCre-rGpA plasmid (a kind gift from Professor Seiya Mizuno, University of Tsukuba), into the pAAV-CMV backbone (Takara, Shiga, Japan) using the In-Fusion Cloning Kit (Takara). The pHelper plasmid was purchased as a part of the AAVPro helper-free system (Takara). The pAAV2/9 plasmid was a gift from James M. Wilson (Addgene plasmid #112865).

AAV production

AAVs were generated by triple plasmid transfection of 293T cells with polyethyleneimine (PEI MAX; Polysciences, Valley RD, PA, USA). AAV isolation and purification were performed using a previously published two-step purification method, with minor modifications [23]. In particular, the clarification step by PVDF filter was replaced with centrifugation at 10,000 G for 10 min, to reduce the viral loss.

Fluorescence macroanalysis

Mice were humanely sacrificed at the postnatal age of 14 days (P14), perfused with PBS, and their organs were dissected for immediate imaging. A fluorescence stereomicroscope (MZFLIII; Leica, Nussloch, Germany) was used for ex vivo analysis of green fluorescent protein (GFP) and DsRed expression in the organs of AAV9-CMV-iCre injected mice.

Cryosectioning analysis

The dissected organs were fixed in fresh PBS containing 4% paraformaldehyde (PFA) or Mildform (FUJIFILM Wako, Osaka, Japan) overnight at 4°C. The tissues were then washed with PBS, immersed in 30% sucrose, and embedded in an optimal cutting temperature (OCT) compound. The sections were cut into 5 µm sections. After washing with PBS, the sections were mounted with Fluoromount (Diagnostic BioSystems, Pleasanton, CA, USA). Nuclei were counterstained with Hoechst 33,342 (Thermo Fisher Scientific, Waltham, MA, USA).

The percentage of DsRed-positive area was determined by analyzing sections at five random sites from the same organ and calculating using the formula given below:

DsRed+ area (%) = DsRed+ area / (DsRed+ area + GFP+ area) × 100

Results

Viability of fetuses following intraplacental injection of AAV9-iCre

Previously, intraplacental injection has been performed at the embryonic stage E11.5 [17]. However, in this study, we performed intraplacental injection at E9.5, allowing earlier gene delivery to the developing embryo (Figs. 1A and B). Initially, prior to virus injection, we tested the feasibility of injection using Indian ink. Our results show that intraplacental injection at E9.5 allows for efficient delivery, as evidenced by the successful delivery of Indian ink to the embryonic heart, allantois, and blood vessels (Fig. 1C).

Next, we incorporated AAV9-CMV-iCre into intraplacental injections at E9.5 and E11.5, respectively (Table 1). Specifically, 1 µl of 1 × 10¹⁰ genome copies (GC)/µl of AAV9-CMV-iCre were injected into E9.5 embryos, while E11.5 embryos received 1 µl of 2 × 10¹⁰ GC/µl of the virus. Of the 22 injected pups at E9.5 obtained from 3 dams, 15 survived until the analysis (survival rate of 68%). Furthermore, 4 dams were used for E11.5 injections. 32 embryos were injected and 10 of the 13 fetuses from 2 dams survived for 2 weeks (77% survival). Two other dams ate their pups during the weaning, leaving one survivor from each batch.

Table 1. Injection record of AAV9-CMV-iCre into Green-to-Red Reporter (GRR) mice.

Dam #	Stage injected	Fetuses injected	Number of surviving fetuses at 2 week old	Titer of AAV9-CMV-iCre
1	E9.5	6	5	1*1010 GC/μl
2		8	5
3		8	6
4	E11.5	8	5	2*1010 GC/μl
5		9	1
6		10	1
7		5	5

AAV9-iCre injected mice demonstrate systemic transduction

To investigate the transduction efficiency, we injected AAV9-CMV-iCre into double-reporter GRR mouse embryos. If a virus infects a given cell, the native GFP is replaced by DsRed, in a process mediated by Cre-recombinase activity (Fig. 2A). To determine the location of virus transduction, we performed ex vivo macroscopic analysis of the organs (Fig. 2B). We designated as “strong” those organs where DsRed was strongly expressed in most parts of the organ, “moderate” those with strong DsRed expression in some parts of the organs, and “weak” those in which slight DSRed expression was observed anywhere in the organ. The representative data are shown in Fig. 3.Ex vivo analysis revealed that AAV9-CMV-iCre can transduce multiple organs in GRR mouse embryos. In E9.5-injected pups, DsRed signal was observed in all organs analyzed, including the liver (11/15, 73.3%), heart (13/15, 86.6%), brain (11/15, 73.3%), thymus (11/15, 73.3%), and intestine (15/15, 100%) (Fig. 2B, left panel). Fewer DsRed-positive organs were observed in E11.5-injected pups. The liver (10/13, 77%) had the strongest presence of DsRed, while intestinal expression was observed in 46% of pups analyzed, and DsRed-positive signals were observed in the heart, brain, and lungs of 38% of pups. Only one E11.5-injected pup exhibited broad transduction, similar to the E9.5-injected pups. Collectively, these results suggest that intraplacental injection of AAV9-CMV-iCre into E9.5–E11.5 embryos can result in the transduction of multiple organs.

Fig. 2.

Summary of AAV9-CMV-iCre injections. (A) Graphical diagrams of principle demonstrated by the GRR gene structure and Cre-mediated recombination (top) and experimental flow (bottom) of AAV9-CMV-iCre injections in GRR mouse. (B) Summary of ex vivo observation of DsRed fluorescence resulting from injections at E9.5 and E11.5 stages. Each column represents an individual pup. The order of pups has been presented according to the levels of DsRed intensity observed.

Fig. 3.

Representative photos of organs used for the ex vivo macroanalysis of AAV9-CMV-iCre-injected GRR embryos. Scale bar 2 mm. DsRed was designated as “strong” when DsRed was strongly expressed throughout the tissue, “moderate” when partial expression was observed, and “weak” when only slight expression was observed. Scale bar 2mm.

DsRed fluorescence was observed in the thymus of E9.5-injected mice

AAV9-CMV-iCre injection resulted in a relatively high level of thymus transduction in embryos at E9.5. Among the pups, 67.7% exhibited DsRed fluorescence in the thymus. Of these, one pup showed strong DsRed fluorescence, whereas six pups showed moderate fluorescence (Figs. 2B and 3). Analysis of thymus sections from nine pups revealed an average DsRed-positive area of 25% (Figs. 4A and B). In two pups, the DsRed-positive area revealed high transduction rates of 33% and 38%.

Fig. 4.

DsRed transduction observed in thymus in E9.5 injections. (A) Representative photos of thymus sections of 2-week-old mice injected with AAV9-CMV-iCre at E9.5 stage. Scale bar 50 µm. (B) DsRed-positive area calculation in thymus. Each dot represents an individual sample.

We conclude that intraplacental injection of E9.5 embryos with AAV9 resulted in a relatively high level of thymus transduction. Therefore, we suggest that this method could be a valuable tool for ectopic gene expression or gene-editing approaches targeting cells in the thymus.

Liver transduction varies depending on the stage of the injection

Macroscopic analysis revealed that the liver is a site of significant transduction (Fig. 2B). In E9.5-injected pups, one strong, one moderate, and nine weak DsRed fluorescence signals were observed in the liver. In contrast, the E11.5-injected pups exhibited three strong, three moderate, and five weak DsRed fluorescence signals.

Cryosection analysis of the liver showed average DsRed-positive areas of 3.1% in E9.5-injected pups and 14.5% in E11.5-injected pups. The highest level of transduction (73%) was observed in an E11.5-injected pup, with most cells being hepatocytes (Fig. 5B).

Fig. 5.

DsRed fluorescence in liver of E9.5- and E11.5-injected mice. (A) Representative photos of liver sections of 2-week-old mice injected with AAV9-CMV-iCre at E9.5 stage. Scale bar 50 µm. (B) DsRed-positive area calculation in liver of E9.5 and E11.5 injected mice. Each dot represents an individual sample.

These findings suggest that the liver can specifically be targeted at later gestational stages, which is consistent with the results of previous studies. Developmentally, at the E9.5 stage, liver progenitor cells begin to differentiate, and umbilical veins have not yet given rise to the hepatic vein branches [24]. This may be the reason for the lower rates of liver transduction in E9.5-injected mice compared to E11.5-injected mice.

Evaluation of transduction to reproductive organs

One of the major safety concerns regarding gene therapy is the transduction of germline cells [4,5,6,7]. Macroscopic analysis revealed the presence of Dsred expression in the reproductive organs (3/15, 20%) from E9.5 injected mice (Fig. 2B). To identify transduced cells, ovary cryosectioning was performed (Fig. 6). Results indicate the presence of Dsred positive cells in ovary region, but is unclear if this indicates the transmission to the germline. Further evaluation needs to be conducted to address this concern.

Fig. 6.

DsRed fluorescence in the ovaries of E9.5 injected mice. Scale bar 100 µm in 4× magnification photos and 50 µm in 40× magnification photos.

Discussion

In utero gene therapy holds great promise because of its potential advantages compared to later-stage genetic intervention, which include decreased immune response, smaller body size at the time of injection, accessible stem cells, a permeable blood-brain barrier, and most significantly, the opportunity to intervene before the onset of disease [5,6,7]. To progress towards clinical translation, a comprehensive evaluation within and across animal models is essential. This study is part of a preliminary investigation aimed at confirming the feasibility of therapeutic strategies that can be translated to human subjects.

Several groups have previously demonstrated positive examples of phenotypic rescue in rare genetic disease mouse models by in utero gene editing via different routes of injection. To rescue lysosomal storage disease mouse models, base-editing vectors were introduced via the vitelline vein at E14.5 [25, 26]. Similarly, the cardiac hypertrophy model was corrected using base editing at E16.5 [27]. To rescue lethal interstitial lung disease model mice, adenovirus packaged with spCas9 was injected into the amniotic cavity of E16.5 embryos [28]. In the Angelman syndrome mouse model, cerebral injections of fetuses at E16.5 with AAV9-Cas9 were performed, resulting in phenotypic restoration [29]. Together, these studies demonstrated proof-of-concept rescues of disease models but focused on mid-to-late gestation stages of murine fetuses, comparable to the late second and third trimesters in human pregnancy.

Our study investigated the potential of AAV9-based recombinase-based techniques for prenatal gene editing experiments as early as E9.5, and our findings suggest promising implications for both basic and translational research. On the other hand, our study yielded troubling findings regarding transduction in reproductive organs. To mitigate this concern, alternative strategies can be employed, such as utilizing tissue-specific promoters to limit the transgene expression in germline cells or using vectors with more restricted tropism. These approaches may prove effective in minimizing the risk of unintended transgene expression in reproductive tissues. To expand these results, we plan to evaluate alternative delivery vectors, including adenovirus, lipid nanoparticles (LNPs), and virus-like particles (VLPs) [30,31,32]. In addition, we intend to conduct proof-of-concept rescue experiments using suitable animal models of the disease. The results of these experiments will provide further insights into the potential of early-stage prenatal gene therapy.