AAV Diffuses across Zona Pellucida for Effortless Gene Delivery to Fertilized Eggs

Abstract

Gene delivery to fertilized eggs is often the first step in creation of transgenic animals, CRISPR knock-out, or early developmental studies. The zona pellucida, a hardened glycoprotein matrix surrounding the mammalian fertilized eggs, often complicates gene delivery by forming a barrier against transfection reagents and viruses. High efficiency techniques to perforate or penetrate the zona allow for access and gene delivery to fertilized eggs. However, these techniques often rely on highly skilled technologists, are costly, and require specialized equipment for micromanipulation, laser perforation, or electroporation. Here, we report that adenoassociated viruses (AAVs) with serotypes 1 or DJ can efficiently diffuse across the zona to deliver genes without any manipulations to fertilized eggs. We observe lowered rates of embryo development after treatment of embryos with all AAV serotypes. However, we were able to reduce adverse effects on embryo development by exposing embryos to AAVs at later stages of in vitro development. AAVs have low immune response and do not incorporate into their host chromosomes to cause insertional mutations. Hence, AAVs can serve as a highly effective tool for transient delivery of genes to fertilized mammalian eggs.

Introduction

Gene delivery to fertilized eggs is a powerful technique for studying the early development of mammalian embryos. Researchers overexpress genes encoding proteins or interfering RNAs at the single cell stage to follow the activity of genes throughout the development as cells divide and differentiate to form adult organisms. Modulation of protein levels and resulting phenotypic changes have provided a wealth of information about the in vivo function of genes to researchers [1]. Furthermore, transient delivery of CRISPR-Cas9 components or recombinases such as Cre, Dre, or Flp, to early stage embryos has been used to remodel the genome by mutating, deleting, activating or silencing genes.

Introduction of exogenous genes onto mammalian fertilized eggs are technically achieved via pronuclear microinjection of genes into fertilized eggs, electroporation, or physical and chemical removal of the zona pellucida (ZP) for gene delivery [2,3,4,5,6]. Fertilized mammalian eggs are enveloped in a glycoprotein matrix ZP that protects the embryonic cells prior to blastocyst formation and implantation in uterus [7,8]. Chemical removal of the ZP results in damage and aberrant development of mammalian embryos [9]. The barrier ZP prevents RNA or DNA fragments, transfection reagents, and viruses to reach mammalian fertilized eggs. Currently, most gene delivery protocols rely on puncturing or disruption of the mammalian ZP [3,4,5,10]. Since the ZP plays a crucial role in protection and proper development of fertilized eggs, disruptions often result in damage [11,12]. Nevertheless, physical micromanipulation and laser perforation of ZPs are routinely used for gene delivery to mammalian fertilized eggs [3,4]. Although highly efficient, physical manipulation of mammalian fertilized eggs require specialized equipment and skilled technologist. Here, we demonstrate that adeno-associated viruses (AAVs) diffuse across the mouse ZP for efficient delivery of genetic material without any physical or chemical manipulation of fertilized eggs. AAVs are capable of simultaneous delivery of multiple genes to a large number of fertilized eggs in culture.

Recombinant adenoassociated viruses (AAVs) are highly efficient vehicles for delivering genes in vivo and in vitro. AAVs are single stranded DNA viruses that belong to the family of Parvoviridae viruses (16, 17). The viral particle is approximately 20 nm in diameter and can carry a genomic load of up to 4.5 kb. The genomic load in recombinant AAVs can accommodate promoters, genes, selection markers, or fluorescent moieties. Natural and designer capsid variations in AAVs offer specificity in transduction, allow for the genetic manipulation of a broad range of host cells, and elicit low immune response [13,14]. Genes delivered by AAVs do not incorporate into host chromosomes to cause insertional mutations, nonetheless persist as concatemeric circles for lasting expression in some tissue [15]. Therefore, AAVs are among favorites for gene delivery in research and are the most popular viral gene delivery system explored in clinical trials [16,17].

In this report, we describe yet another robust application for AAVs – transient gene delivery to mouse fertilized eggs. We demonstrate that AAVs cross the mouse ZP without any manipulation and serotypes 1 and DJ are capable of infecting embryonic cells. Serotype DJ was created by DNA family shuffling and merging of multiple natural AAV capsid isolates [18]. AAVs with the DJ serotype have a robust tropism. Researchers have created several synthetic, chimera, and/or mutagenized serotypes to alter AAV tropism. According to Yoon, et.al., AAV6 is the most robust serotype for gene delivery to mouse embryos [19]. Others report that both AAV1 and AAV6 can efficiently transduce mouse embryos [20,21,22]. We compared AAVs with several commonly used serotypes that are often parental serotypes for synthetic, chimera, or mutagenized serotype variants to test gene delivery to embryos. Among the tested serotypes, AAV1 was the least damaging and the most efficient for gene delivery to mouse embryos in our experiments. AAV1 offers a simplified method for genetic modification of mouse embryos. AAV-DJ is extremely efficient in gene delivery to embryos (100%) but is partially toxic and hinders embryo development. We observe lowered rates of embryo development after treatment of embryos with all AAV serotypes. However, we were able to reduce adverse effects on embryo development by exposing embryos to AAVs at later stages of in vitro development. Transduction with AAVs at later stages may improves the rate of embryo development for other serotypes and variants as well.

Materials and Methods

Reagents

M2 and KSOM media (Millipore); Hyaluronidase (Sigma); DMEM (Gibco); Polyethylenimine MAX (PEI “MAX”, Polysciences, Warrington, PA, USA) stock solution was prepared at 16 mg/ml in sterile water and the pH was adjusted to 4.5 with sodium hydroxide; Turbonuclease (Eton Bioscience, San Diego, CA, USA).

Plasmids

Plasmids used for production of AAVs were transfer plasmids pOTTC407 - pAAV EF1a eGFP (Addgene Cat# 60058) and pAAV-EF1a-tdTomato-WPRE-pGHpA (Addgene Cat# 67527); rep/cap plasmids pAAV2/1 (Cell Biolabs Inc., Cat# VPK-421), pAAV2/2 (Cell Biolabs Inc., Cat# VPK-422), pAAV2/5 (Cell Biolabs Inc., Cat# VPK-425), pAAV2/6 (Cell Biolabs Inc., Cat# VPK-426), pAAV2/8 (Cell Biolabs Inc., Cat# VPK-428), or pAAV2/9 (UPenn Vector Core); pHelper plasmid containing adenovirus E2A, E4 and VA genes (Cell Biolabs Inc., Part No. 340202).

Animal studies and treatment

Animals used in this study were purchased from Jackson Laboratories, USA. All animal procedures complied with the NIH/NIEHS animal care guidelines and were approved by the Animal Care and Use Committee (ACUC) at the NIH/NIEHS, animal Protocol # 2012–0004.

Harvest and in vitro culture of mouse fertilized eggs

Female C57BL/6J mice (stock # 000664) strains were mated and checked the following morning for successful mating indicated by the presence of vaginal plugs. Mated females were humanely euthanized via CO₂ inhalation followed by cervical dislocation and vital organ transection. Their ova were collected and placed in Dulbecco’s Modified Eagle Medium (DMEM) media. Fertilized eggs were isolated 12–20 hours after mating by tearing of collected ova with a syringe needle and transferring them to sterile M2 media. The collected fertilized eggs were separated from cumulus cells by treatment with hyaluronic acid in M2 media, washed, and cultured in KSOM media at 37°C, 5% CO₂, 5% O₂ and 90% N₂.

AAV production and transduction of mouse fertilized eggs and embryos

AAVs were produced according to published protocols [23,24]. Briefly, 15-cm dishes were seeded with HEK293-AAV cells at 6 × 10⁶ cells. Cells were incubated at 37°C in 5% CO2 for 24 hours before transfection with AAV transfer, rep/cap and pHelper plasmids (33.3 ug of each per dish) using 12.5 μl of PEI stock (16mg/ml) per dish. Next day, culture media was changed to DMEM alone. 120 hours post-transfection, AAVs were collected, treated with turbonuclease and concentrated by ultracentrifugation. Virus was aliquoted and stored at −80°C. The AAV genome copies and titers were determined by qPCR as previously described [25]. AAV samples were serially diluted from 10–2 to 10–8-fold, and the standard curve was generated by diluting the AAV plasmid containing the ITR2 sequence (from 105 to 32 copies per 5 μl). The Q-PCR was performed using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) and a LightCycler 96 System (Roche, Indianapolis, IN, USA) according to manufacturer’s protocols. The ITR forward primer (5’ GGAACCCCTAGTGATGGAGTT 3’) and the ITR reverse primer (5’ CGGCCTCAGTGAGCGA 3’) were added to the reaction with final concentration of 300 nM in 25 μl of total volume (5 μl of sample or standard curve plus 20 μl of primers and SYBR mix), and under the following PCR conditions: 95°C for 10 minutes, 95°C for 10s, 60°C for 60s, for 45 cycles. The virus titer was calculated by LightCycler 96 SW1.1 software using the parallel standard curve in the reaction, and the titer was given in genome copy/ml (GC/ml). All AAVs had titers of 1e12–1e13 genomic copies per ml. For transduction, AAVs with serotype 1, 2, 5, 8, and 9 were diluted 1:5 and AAV-DJ was diluted 1:10. 1 ul of each diluted AAV was added to KSOM drop seeded with fertilized eggs.

Results

To test AAV transduction, C57BL/6J mouse fertilized eggs were harvested on gestation day E 0.5 and cultured according to standard published protocols [3]. The cumulus cells were removed by incubation with 0.1% hyaluronidase in M2 media and triturated. Harvested fertilized eggs were transferred to KSOM media and incubated under 5% CO₂, 5% O₂, 90% N₂, at 37°C for development. We have previously shown that cultured mouse embryos can be kept in KSOM media at 37°C, 5% CO₂, 5% O₂ and 90% N₂ for up to 4 days and re-implanted in pseudopregnant mice for successful generation of transgenic mice (3). Embryos were incubated for 2 hours after harvesting to recover from the procedure. 1 ul of various diluted AAV serotypes were added to fertilized eggs or embryos in 50ul of KSOM media for transduction. There was no need to transfer fertilized eggs to alternate media for micromanipulation for transduction.

Transduction of mouse embryonic cells with AAVs

To test gene delivery, we used AAV transfer vectors that expressed enhanced green fluorescent protein (GFP) under the control of a constitutive elongation factor 1a (EF1a) promoter. AAVs were packaged with serotypes 1, 2, 5, 8, 9, and DJ. Fertilized mouse eggs were incubated for two hours after harvesting. For transduction, AAVs with serotype 1, 2, 5, 8, and 9 were diluted 1:5 and AAV-DJ was diluted 1:10 in KSOM. 1 uL of each diluted AAV was pipetted directly into the KSOM media of the dishes. Fertilized eggs or embryos were cultured for 4 days to form blastocysts. The embryo development and fluorescent expression was monitored daily. Embryos transduced with either AAV-EF1a-GFP serotypes 1 or DJ expressed green fluorescence 48 hours post transduction (Fig 1). We observed continued GFP expression in formed blastocysts. The GFP expression among embryonic cells of a single blastocyst varied indicating differences in the EF1a expression among differentiating cells and/or a varying number of AAV copies per cell. Number of cultured mouse fertilized eggs per KSOM drop had no effect on gene delivery. AAV serotypes 1 and DJ crossed the zona pellucida (ZP) of mouse fertilized eggs efficiently and transduced embryonic cells without microinjection or micromanipulation - for en masse gene delivery (Fig 1).

Fig 1. En Masse gene delivery to mouse fertilized eggs with AAVs.

Development of C57Bl/6J embryos from harvest (Day1) until the time of blastocyst formation (Day 5) in (a) AAV serotype 1 and (b) AAV serotype DJ. The embryos were transduced with 1 uL of diluted AAV1-EF1a-GFP (10¹² GC/ml) in 50uL of KSOM drops. The presence of fluorescence was detected from day 3 after infection and persisted until the blastocyst formation. The data is representative of three independent experiments.

Increasing the amount of virus used for gene delivery adversely affected the development of fertilized eggs, especially the DJ serotype. The expression of GFP was evident at the morula stage in cultured embryos of the inbred strain C57BL/6J. The number of fertilized eggs used and the expression of fluorescence in blastocysts are summarized in Table 1. AAV with serotype DJ significantly reduced the number of blastocysts developed from mouse fertilized eggs. The amount of AAV-DJ-eGFP was diluted 20-fold, but the virus still remained harmful to normal embryo development, but all developed embryos expressed GFP (data not shown). The AAV with serotype 1 yielded the greatest number of fluorescent blastocysts with minimal damage to blastocyst formation. Therefore, AAV1 is the best candidate for gene delivery to mouse fertilized eggs among tested serotypes in our experiments. There are hundreds of AAV serotypes available to researchers. Development of mouse blastocyst into pups and the length of expression will be dependent on the selected AAV serotype, target tissue, and the nature of the delivered genes; and are therefore beyond the scope of this study. For future experimentation, designer and chimeric serotypes should be assessed individually at several dilutions to determine ideal concentrations to foster embryo viability and to establish gene delivery efficiencies.

Table I.

AAV serotypes and gene delivery to mouse fertilized eggs

AAV serotype	# of Starting Fertilized Eggs	# of Blastocyst	# of Fluorescent Blastocyst	%Blastocyst	% Fluorescent Blastocyst
No virus	70	47	0	67	0
AAV 1	97	51	41	53	80
AAV 2	72	27	5	38	19
AAV 5	69	31	0	45	0
AAV 8	51	17	2	33	12
AAV 9	71	25	4	35	16
AAV DJ	71	10	10	14	100

AAV infection of fertilized eggs reduced the rate of embryonic development to blastocyst. Embryonic transduction results for various AAV serotypes are listed. Compared to other serotypes, AAV1 has the least impact on blastocyst development (AAV1 vs. no virus group, p < 0.05; AAV2 or AAV5 vs. no virus group, p < 0.005; AAV8, AAV9 or AAV DJ vs. no virus group, p < 0.0001, one-tailed T-test statistical analysis).

Developmental stage of embryonic cells differentially affects transduction of AAV serotypes

We also compared effects of mouse embryo transduction with AAV1 and DJ at 1-cell, 2-cell and 4-cell stages of development. Transduction with AAV1 or AAV-DJ at 2-cell or 4-cell stage improved the blastocyst formation as compared for transduction at the 1-cell stage but decreased the number of fluorescent blastocysts (Table 2). The rate of blastocyst formation for AAV-DJ was improved from 14% to 41% and 50% when transduction was delayed to 2-cell and 4-cell, respectively. Therefore, AAV-DJ serotype is less harmful to embryo development once introduced at later stages of development but retains its high transduction rate. Based on our results, transduction with either serotype resulted in gene expression after one day, which makes AAVs an ideal tool for gene delivery to study early embryonic development.

Table 2.

AAV Transduction at different embryonic developmental stages

AAV serotype	# of Starting Fertilized Eggs	# of Blastocyst	# of Fluorescent Blastocyst	%Blastocyst	% Fluorescent Blastocyst
AAV 1 (2-cell)	63	40	23	64	58
AAV 1 (4-cell)	73	48	21	66	44
AAV DJ (2-cell)	69	28	17	41	61
AAV DJ (4-cell)	68	34	28	50	82

Transduction of embryos at the 2-cell or 4-cell stage improved the rate of embryonic development into blastocyst significantly (AAV DJ vs. AAV DJ at 2-cell stage, p < 0.005; vs. AAV DJ at 4-cell stage, p < 0.0001, one-tailed T-test statistical analysis]. However, delayed transduction resulted in lowered GFP expression and increased mosaicism in embryonic cells.

Concurrent gene delivery with AAVs

To demonstrate concurrent delivery of genes, an AAV transfer vector expressing dTomamto under the control of a constitutive elongation factor 1a (EF1a) was prepared. Both AAVs were packaged with serotype 1. A 1 ul mixture of AAV1-EF1a-GFP and AAV1-EF1a-tdTomato was added to harvested fertilized eggs for transduction. As shown in Fig 2, both AAVs transduced and expressed GFP and tdTomato in developed blastocysts. Within each embryo, the expression pattern for each AAV differed and demonstrated independent transduction of cells.

Fig 2. Concurrent AAV Gene Delivery to Mouse Fertilized Eggs.

C57Bl/6J fertilized eggs were exposed to 2ul of AAV mixture: AAV1-EF1a-eGFP and AAV1-EF1a-tdTomato (1e12–1e13 GC/ml) and allowed to develop for 4 days into blastocysts. Embryos expressed both eGFP and tdTomato fluorescence.

Discussion

In the present study, we have identified AAV serotypes that penetrate the intact zona to transduce C57BL/6J mouse embryonic cells. The AAV gene delivery is less invasive to fertilized eggs, can be used to transduce fertilized eggs in bulk, takes less time, and removes the technical limitations imposed by micromanipulation. Moreover, AAVs transduce embryonic cells without minimizing the zona protection of developing embryos. Therefore, AAVs have the potential to serve as a highly promising tool for transient delivery of genes used to monitor or modify mammalian fertilized eggs during development - for example: sensors and actuators such as GCaMP6f [26] and PHluorin [27]; knock-down components of CRISPR-Cas9; or recombinases such as Cre or Flp.

We tested AAVs with serotypes 1, 2, 5, 8, 9, and DJ and observed that AAV1 and AAV-DJ are capable of crossing the zona pellucida (ZP) surrounding the mouse fertilized eggs and transduce embryonic cells without microinjection or micromanipulation. AAV-DJ has a very robust tropism, readily available to researchers, and was not previously shown to transduces embryonic cells. Despite its very efficient transduction, AAV-DJ was harmful to embryo development at 1-cell stage. When delivered at the later stages of development, AAV-DJ was less harmful and delivered a rapid and robust expression. Other groups have reported that AAV serotypes 1 and 6 are capable of transducing mouse embryonic cells [19,20,21,22]. AAV-DJ was not included in the list of tested serotypes and AAV1 was out performed by other serotypes. We hypothesized that the AAV transfer vectors, delivered cargos, and the viral packaging methods may be responsible for these variations in performance. Since AAVs of various serotypes have the same size, we hypothesize that all AAV serotypes cross the mouse ZP, and the differences in transduction results from the interaction of AAV capsid proteins with surface protein composition of embryonic cells. As cells divide, differentiate, and modulate surface protein expression, the AAV serotype transduction efficiency may change. An increasing number of AAV serotype that have been isolated from nature, in addition to libraries of chimeric and designer capsid coats, have broadened the specificity and tropism of AAVs [28,29,30]. Therefore, a myriad of AAV serotypes are available for screening and use in developmental studies.

Genes delivered by recombinant AAVs rarely integrate and are typically lost through cell division due to deletion of rep and cap genes [31,32]. Although, persistent expression of delivered genes are promoted by the presence of Internal Terminal Repeats (ITR) and formation of AAV genome concatemers in the host nucleus. Therefore, AAVs are effective and ideal vectors for short-term expression of genes such as recombinases and CRISPR-Cas9 components used for creation of animal models and in vivo gene editing. We were able to transduce multiple mouse fertilized eggs in a drop of KSOM media simultaneously. Genes delivered by AAVs, such as CRISPR-Cas9 components, are rapidly expressed within days of delivery and eventually, as embryonic cells divide, dissipate and disappear [19].

Using AAV vectors for gene delivery has its limitations. The genetic load carried by each virus is limited to 4.5 kilobases and too much AAV is damaging to developing embryos. We also noticed mosaicism in the developed blastocysts as evident by the variation in the GFP fluorescence intensity. This phenomenon could be due to the variation in the number of AAV copies per cell and can be partially remedied by limiting the length of embryo incubation with the AAVs. In our experiments, we allowed embryos to be exposed to AAVs for the entire length of in vitro incubation to maximize expression. The variation in GFP expression may also be due to transcriptional regulation as embryonic cells differentiate within the blastocyst. We aimed to minimize the transcriptional variation among differentiating embryonic cells by using a ubiquitously expressed promoter, EF1a.

We anticipate that facilitated gene delivery to mouse fertilized eggs by AAVs may also apply to fertilized eggs of other species with barriers similar to the mouse zona. However, other AAV serotypes must be screened for their efficiencies in gene delivery to embryonic cells of other mammalian species.

Highlights.

• Adeno-associated viruses (AAVs) with serotypes 1 or DJ can efficiently diffuse across the zona to deliver genes to mouse embryos.

• AAV-DJ transduces embryonic cells efficiently but causes lowered rates of embryo development.

• The adverse effects of AAV-DJ treatment on embryo development was reduced by exposing embryos to AAV-DJ at later stages of in vitro development.

• AAVs have low immune response, do not incorporate into their host chromosomes to cause insertional mutations, and hence can serve as a highly effective tool for transient delivery of genes to fertilized mammalian eggs.