Pigs are one of the most important livestock animals globally, and their economic value extends beyond being a source of meat for human consumption to include considerable application in the biotechnology industry (Thornton, 2010). Recently, pigs have been utilized as model organisms in in vitro or in vivo gene editing (GE) research, and studies on livestock disease resistance (Whitworth et al., 2022). Among the numerous diseases affecting pigs, porcine reproductive and respiratory syndrome (PRRS), stands out as a major concern, both economically and in terms of animal health (Whitworth et al., 2022). PRRS is generally caused by the PRRS virus (PRRSV), which primarily targets the respiratory and reproductive system in swine. The virus infects pigs via respiratory transmission, entering host cells through specific receptors, including CD163, a scavenger receptor expressed on pulmonary alveolar macrophages (PAMs) (Calvert et al., 2007). Upon binding to CD163, the virus enters the cell, where it replicates and subsequently spreads, triggering a complex immune response that often leads to persistent infection (Van Gorp et al., 2008). Therefore, studies on PRRS using GE strategies are essential for enhancing disease resistance in pigs.

Over the last decade, substantial progress has been made in using GE technologies, including the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) system, to enhance desirable traits in pigs—such as disease resistance and reproductive performance—and to produce transgenic and knockout models (Gaj et al., 2013). CRISPR/Cas9 system has revolutionized the GE tool, providing a precise and efficient method to edit targeted genomic sites, making it possible to introduce favorable traits while eliminating unwanted characteristics (Singh and Ali, 2021). This technology has been instrumental in enhancing disease resistance, such as developing pig resistant to PRRS using CD163 gene deletion, that has a devastating economic impact on the swine industry (Whitworth et al., 2016). Moreover, intriguingly, pigs that express Cas9 endonuclease on their own are being generated to overcome the limitations of traditional transgenic animal generation processes (Wang et al., 2017). This advancement offers the advantage of directly verifying the function of various genes in living pigs. Such technological progress is expected to drastically reduce the time and cost required to identify and validate genes suitable for producing various transgenic pigs.

In addition to CRISPR/Cas9, adeno-associated viruses (AAVs) have become effective tools for delivering key CRISPR/Cas9 components, such as the Cas9 protein and single guide RNA, in pigs (Senís et al., 2014). AAVs, which are small, non-enveloped DNA viruses from the Parvoviridae family, are advantageous due to their low immunogenicity and their ability to target specific tissues, making them ideal for in vivo GE applications (Asmamaw Mengstie, 2022). AAVs are classified into distinct serotypes based on differences in their capsid proteins, including virus proteins viral protein 1 (VP1), VP2, and VP3. These capsid proteins are crucial in defining viral tropism, as they determine the interactions between AAVs and their target cells, thereby influencing tissue and cell specificity (Gao et al., 2004; Zincarelli et al., 2008; Mingozzi and High, 2011).

Several studies have aimed to identify tissue-specific AAV serotypes in pigs. For instance, AAV2H22 has successfully transduced porcine airway epithelial cells in vivo (Steines et al., 2016; Cooney et al., 2019). AAV4 is associated with respiratory diseases, such as cystic fibrosis, in pigs (Chen et al., 2021). AAV1 has been reported to be involved in efficient gene transfer to pig myocardium (Su et al., 2008). Additionally, in a study on transgene expression in porcine hepatocytes, AAVrh20 was identified as one of the most efficient capsids, demonstrating high transduction efficiency (Willimann et al., 2024). Furthermore, AAV2.8 effectively delivered RNAi-therapeutics targeting vascular endothelial growth factor A to reduce choroidal neovascularization (Haldrup et al., 2024).

However, AAV serotypes related to lung tissue-derived cells have not been fully screened in pigs. Based on tropisms and infection mechanisms of PRRS virus within the respiratory system (Calvert et al., 2007; Zhang et al., 2022), we compared the transduction efficiencies of various AAV serotypes in PAMs and porcine lung epithelial cells (pLCs) derived from the Cas9-expressing pig. For both types of porcine respiratory cells, our findings revealed extremely high tropism for the AAV2 serotype. This research aims to identify optimal vectors that can advance pulmonary GE strategies and immune modulation.

Cell culture

The pLCs utilized in this study were isolated from a Cas9-expressing pig (Kim et al., 2024); a breed derived from native Korean Black pigs (Cho et al., 2019) and maintained under 5% CO₂ in PneumaCult^TM-Ex Plus basal medium (StemCell Technologies Inc., Vancouver, BC, Canada, #05041) supplemented with 50× PneumaCult^TM Ex-Plus supplement (STEMCELL Technologies Inc., Vancouver, BC, Canada, #05042) in a humidified incubator at 37℃.

Porcine alveolar macrophages 3D4/31 (PAMs 3D4/31) was obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA, #CRL-2844) and maintained under 5% CO₂ in Roswell Park Memorial Institute 1640 (RPMI-1640) medium (+ L-Glutamine, + 25 mM HEPES) (Caisson Labs, Smithfield, UT, USA, #RPL09) supplemented with 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA, #16000-044) and 1× penicillin-streptomycin (PS; Gibco, Grand Island, NY, USA, #10378-016) in a humidified incubator at 37℃. Dissociation of the pLCs and 3D4/31 PAMs from the culture ware was performed in the incubator using 0.25% trypsin-EDTA (Gibco, Grand Island, NY, USA, #25200-056) over 5 min, with the resulting cells washed once with phosphate-buffered saline without calcium magnesium (PBS; Caisson Labs, Smithfield, UT, USA, #PBL01).

Porcine kidney-15 (PK-15) cells and porcine ear fibroblasts were maintained under 5% CO₂ in Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, Grand Island, NY, USA, #10569-010) supplemented with 15% FBS, 1% minimum essential medium non-essential amino acids (MEM NEAA; Gibco, Grand Island, NY, USA, #11140-050), 1× Antibiotic-Antimycotic (Gibco, Grand Island, NY, USA, #15240-096), 0.1× Low Serum Growth Supplement (LSGS; Gibco, Grand Island, NY, USA, #S-003-10), and 0.001× beta-mercaptoethanol in a humidified incubator at 37℃.

Immortalization of pLCs using lentivirus particles

Immortalization was achieved by infecting primary pLCs at passage 4 according to the manufacturer’s instructions. Prior to infection, 0.2 mL of lentivirus suspension was placed in each well and diluted with complete medium containing hTERT-pCMV Hygro vector (GeneCopoeia Inc., Rockville, MD, USA, #LP-728-100) and polybrene transfection reagent (EMD Millipore Corp., Burlington, MA, USA, #TR-1003-G) to obtain a final concentration of 5 μg/mL. The original culture medium containing the pLCs was then removed and the virus-containing medium added to the wells. The control was produced by adding 0.2 mL of complete medium with polybrene to pLCs. Cells were then maintained under 5% CO₂ in the humidified incubator at 37℃ for 18 h and the medium, with or without virus, was aspirated and replaced with fresh complete medium (0.5 mL) without polybrene and left for 24 h. To select stably transduced pLCs, the medium was replaced with fresh complete medium containing 150 μg/mL Hygromycin B (Gibco, Grand Island, NY, USA, #10687010) every 3 days. Cells were then split at a 1:10 ratio by trypsinization and re-seeded onto culture dishes until hygromycin-resistant colonies became visible. Immortalized pLCs (pLCs^Imt) were harvested at passage 10 and analyzed for the transient expression of human telomerase reverse transcriptase (hTERT) using real-time quantitative polymerase chain reaction (qPCR). Primary pLCs from passage 4 were used as internal hTERT gene controls. The primers used are listed in Table 1.

Table 1. Primers used for real-time quantitative PCR analysis.

Symbol	Gene name	Primer sequence (5’ → 3’)	Accession no.	Tm (℃)	Product size (bp)
eGFP	Enhanced green fluorescent protein	Fw: AAG CAG AAG AAC GGC ATC AA	X96418.1	55.5	97
		Rv: GGG GGT GTT CTG CTG GTA GT		61.0
hTERT	Human telomerase reverse transcriptase	Fw: GCC GAG ACC AAG CAC TTC CTC GAC T	NM_001193376.3	63.8	111
		Rv: GCA ACT TGC TCC AGA CAC TCT TCC G		61.5
KRT5	Keratin 5	Fw: GCA GAT TGA GTG GAG AAG GTG TT	XM_021067000.1	58.0	67
		Rv: CCA GAG GAG AGG GTG TTT GTG		59.1
FOXJ1	Forkhead box protein 1	Fw: CGC CAC AAC CTG TCT CTG AA	XM_003357959.4	58.8	68
		Rv: CCC TTG CCC GGC TCA T		60.5
SYP	Synaptophysin	Fw: CCT CAT CGG CTG AAT TCT TGG	XM_003135078.5	54.7	61
		Rv: GCC CCC ATG GAG TAG AGG AA		60.4
Scgb1a1	Secretoglobin family 1A member 1	Fw: CAG AGG TCT GCC CGA GCT T	XM_003353832.3	61.6	60
		Rv: TGG CAA GTG TGC CCT TGA		58.7
PDPN	Podoplanin	Fw: TGC CAG TTC TGC TCT TCG TTT	XM_005665017.3	57.9	66
		Rv: CCG TGC TGG CTC CTT CTG		60.8
SFTPB	Surfactant protein-B	Fw: ACG CCC CAA CCG ATG AC	NM_001102679.1	60.1	54
		Rv: TGA GGA TGC TGG CGA TGT C		59.0
GAPDH	Glyceraldehyde-3-phosphate dehydrogenase	Fw: ATG TTC CAG TAT GAT TCC ACC C	AF017079	55.2	132
		Rv: ACC AGC ATC ACC CCA TTT G		56.8

bp, base pair; eGFP, enhanced green fluorescent protein; FOXJ1, forkhead box protein 1; Fw, forward; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; hTERT, human telomerase reverse transcriptase; KRT5, keratin 5; PDPN, podoplanin; Rv, reverse; Scgb1a1, secretoglobin family 1A member 1; SFTPB, surfactant protein-B; SYP, synaptophysin..

Single cell sorting by FACS

All procedures were conducted by the manufacturer’s instructions. Briefly, whole pLCs^Imt were index-sorted on an Invitrogen^TM Bigfoot Spectral Cell Sorter (Thermo Fisher Scientific Inc., Waltham, MA, USA, #PL00304) operated by users experienced in single cell technologies. The instrument was used in a setup with a 70 μm nozzle and regular quality checks were done by running a Bigfoot^TM Calibration Beads (Thermo Fisher Scientific Inc., Waltham, MA, USA, #PL00287). Best drop delay was determined using the Bigfoot^TM Calibration Beads (Thermo Fisher Scientific Inc.) just before each experiments. Cells were hydrodynamically focused within a fluidic stream where single cells become encapsulated in individual droplets. To optimize the positioning of single sorted cells in the center of each well of the 96-well plate, “Run Image Alignment” was confirmed before every sorted plate. Next, single cells were sorted directly into 96-well clear bottom TC Surface plate (Thermo Fisher Scientific Inc., Waltham, MA, USA, #165306) pre-filled with PneumaCult^TM-Ex Plus basal medium supplemented with 50× PneumaCult^TM Ex-Plus supplement (1 cell per well). In each plate, three wells were left without sorted pLCs^Imt as the negative controls. The single-cell sorted pLC^Imt was maintained immediately after FACS sorting under 5% CO₂ in a humidified incubator at 37℃. Single clones including #2, 3, 5, 6, 8, 10, 11, 16, 18, 19, 20, C2, C4, D1, and D4 were used for analyzing the expression level of the epithelial specific markers.

AAVs and in vitro transduction

All AAV serotypes (including AAV1-9, 6.2, rh10, DJ, DJ/8, PHP.eB, PHP.S, 2-retro, 2-QuadYF, and 2.7m8) were purchased from VectorBuilder (VectorBuilder Inc., Chicago, IL, USA, #PANEL-AAVSP01-28), containing the enhanced green fluorescent protein (eGFP) reporter gene-driven cytomegalovirus promoter.

For in vitro transduction, pLCs^Imt and PAMs 3D4/31 were seeded in 24-well plates (Thermo Fisher Scientific Inc., Waltham, MA, USA, #142475) with the appropriate growth medium under 5% CO₂ in the humidified incubator, which was maintained at 37℃. The medium was aspirated when the cells reached approximately 80% confluence, and transduction performed. To maximize the transduction efficiency, 250 μL of the AAV-containing growth medium was added to cells with the designated multiplicity of infection (MOI). The 24-well plate was then gently swirled to mix and cover the cells and the results incubated under 5% CO₂ in the humidified incubator at 37℃ for 18 h. The AAV-containing medium was then safely removed, replaced with 500 μL fresh growth medium, and the results incubated overnight in the incubator.

Transduction efficiency

To elucidate the transduction efficiency, pLCs^Imt and PAMs 3D4/31 were seeded at a concentration of 5.0 × 10⁴ cells per well in 24-well plates and incubated with 20K MOI of AAVs for 18 h. The AAV-containing medium was then replaced with fresh complete medium. Three days post-transduction, pLCs^Imt and PAMs 3D4/31 cells were stained with 1 μg/mL of Hoechst 33342 (Invitrogen Inc., Waltham, MA, USA, #H3570) in the growth medium for 5 min at room temperature. The staining medium was then removed and the cells washed once with DPBS containing 1× PS to remove any residual Hoechst dye. The eGFP expression levels were observed in both cell types three days following AAV transduction. A laser-scanning confocal microscope (Nikon AX, Nikon, Tokyo, Japan) was then used to visualize and photograph three random microscopic fields, with the blue-fluorescent nuclei followed by the green and red fluorescent signals. The images obtained were analyzed using ImageJ software and the total cell numbers elucidated by counting the blue-fluorescent nuclei, which was performed independently by two co-workers to ensure accuracy. The blue fluorescent images were then merged with the green (eGFP-positive cells) and red fluorescent images (tdTomato; Cas9) (Kim et al., 2024), and the transduction efficiency of the AAVs expressed as the percentage of green fluorescent cells relative to the total cell count. Data were presented as mean ± standard error of the mean.

Real-time qPCR

To assess the expression level of the eGFP mRNA (transduction), total RNA was isolated from pLCs^Imt and PAMs 3D4/31 five days following transduction of the AAVs using TRIzol^TM Regeant (Invitrogen Inc.) and the Direct-zol^TM RNA MiniPrep kit (Zymo Research Corp.) according to the manufacturer’s instructions. To further determine expression levels of epithelial specific markers such as keratin 5, forkhead box protein 1, synaptophysin, secretoglobin family 1A member 1, podoplanin, and surfactant protein-B, total RNA was isolated from single-cell sorted pLCs^Imt. Whole pLCs^Imt and porcine ear fibroblasts were used as positive and negative controls, respectively.

The quality and concentration of the extracted RNA were assessed using a NanoDrop^® ND-1000 spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE, USA) and the RNA stored at -80℃ until use. Synthesis of cDNA from the extracted total RNA (100 ng) was performed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems Inc., Waltham, MA, USA, #4368814), with reverse transcription performed at 37℃ for 120 min according to the manufacturer’s instructions. Specific primers were designed using Primer Express 3.01 software (Applied Biosystems Inc., Waltham, MA, USA). Details of the primer information can be seen in Table 1. Real-time qPCR was performed using the SYBR Green Master mix (Applied Biosystems Inc., Foster City, CA, USA, #4309155) and a StepOnePlus Real-Time PCR system (Applied Biosystems Inc., Foster City, CA, USA) under the following thermal-cycling procedure: 95℃ for 10 min, 40 cycles at 95℃ for 15 s, and 60℃ for 1 min (1.6℃/s ramp rate) according to the manufacturer’s instructions (Applied Biosystems Inc., Foster City, CA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. Gene expression was quantified via relative quantification (RQ) using the using ΔΔCt method and the data presented as the RQ. All experiments were performed at least in triplicate.

Statistical analysis

All data were analyzed for statistical significance using one-way ANOVA followed by post hoc Tukey’s HSD multiple range test for three or more groups using Statistical Products and Service Solutions (SPSS; International Business Machines SPSS Corp., Armonk, NY, USA) software (ver. 27). Error bars in the data for eGFP mRNA expression and the percentage of eGFP-positive cells represent the minimum and maximum RQ expression levels around the mean. Different letters denote significant differences between the groups. Student’s t-test was performed to compare the differences between two independent groups, allowing analysis of the hTERT expression levels. A statistical p-value of < 0.05 was considered significant in this study. All experiments were performed in triplicate.

AAV serotypes show different transduction efficiencies in immortalized porcine lung epithelial cells

The pLCs^Imt utilized in this study were generated via a meticulous process which involved the isolation and culture of primary pLCs (Supplementary Fig. 1A, left panel) from porcine lung tissue that intrinsically expresses Cas9 followed by immortalization via transduction with the hTERT gene. The resulting pLCs^Imt (Supplementary Fig. 1A, middle panel) exhibited high expression of the hTERT gene at passage 10 or beyond. Before the transduction of AAV serotypes in pLCs^Imt, we analyzed the expression related to the 6 epithelial specific markers (keratin 5, forkhead box protein 1, synaptophysin, secretoglobin family 1A member 1, podoplanin, and surfactant protein-B) in single-cell sorted pLCs^Imt whether they exhibit the characteristics of epithelial cells. As shown in Supplementary Fig. 2, while individual pLC^Imt sorted as single cell displayed relative differences in the expression of genes, it was confirmed that pLCs^Imt exhibit distinct epithelial cell-specific characteristics, compared to porcine ear fibroblasts. However, we used the whole pLCs^Imt, not the single-cell sorted pLCs^Imt, to investigate the AAV serotype transduction efficiency in pLCs^Imt.

The transduction efficiency of the 18 studied AAV serotypes (AAV1-9, 6.2, rh10, DJ, DJ/8, PHP.eB, PHP.S, 2-retro, 2-QuadYF, and 2.7m8) on the pLCs^Imt with distinct capsid sequences was systematically compared, for which cells were infected with each AAV serotype and the visual transduction efficiency assessed three days post-infection using fluorescence microscopy. As shown in Fig. 1A, highly eGFP expression appeared three days post-infection for AAV2, AAV-DJ, and AAV2.7m8 as compared to the control, with weak green fluorescence observed for AAV1, AAV3, AAV5, AAV6, AAV6.2, AAV7, AAV8, and AAV9. AAV2-QuadYF was weakly expressed as compared to AAV2, AAV-DJ, and AAV2.7m8; however, no green fluorescence was observed in pLCs^Imt infected with AAV4, AAV-rh10, AAV-DJ/8, AAV-PHP.eB, AAV-PHP.S, and AAV2-retro three days post-infection. The highest percentage of 52.46 ± 2.4% was observed for the AAV2 serotype in terms of eGFP expression in the pLCs^Imt, with all other serotypes demonstrating much lower results in comparison, such as AAV-DJ at 22.11 ± 3.73% and AAV2.7m8 at 18.31 ± 2.14% (Fig. 1B).

Figure 1.Transduction efficiency of AAV serotypes containing eGFP on day 3 post-infection in pLCs^Imt. (A) Representative eGFP (green) fluorescence images in pLCs^Imt three days after transduction of various AAV vectors (AAV1–9, 6.2, rh10, DJ, DJ/8, PHP.eB, PHP.S, 2-retro, 2-QuadYF, and 2.7m8). Individual AAV serotypes were infected at MOIs of 20 K for 18 h. The mock was not treated with the AAV vector and served as the negative control. Nuclei were stained with Hoechst 33342 (blue). Scale bars indicate 250 μm. (B) Quantification of eGFP-positive cells in different AAV serotype-infected pLCs^Imt. Data are presented as a mean ± standard error. Two independent co-workers counted. Different letters denote significant differences within each group, ^a-dp < 0.05 (n = 3). (C) Comparison with the mRNA expression level of eGFP in AAV serotype infected-pLCs^Imt. Relative quantification (RQ) of the gene expression was calculated using the ΔΔCt method, with data presented as RQ. Error bars represent the RQ of the minimum and maximum expression levels around the mean. All of quantification were normalized to the internal housekeeping gene GAPDH. Different letters denote significant differences within each group, ^a-kp < 0.05. All experiments were performed in triplicate. AAV, adeno-associated virus; eGFP, enhanced green fluorescence protein; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MOI, multiplicity of infection; pLCs^Imt, immortalized porcine lung epithelial cells; RQ, relative quantification.

To further ascertain the transduction efficiency of the individual AAV serotypes, the mRNA expression levels of eGFP (all AAV vectors used in this study were eGFP-tagged) were genetically assessed using the pLCs^Imt (Fig. 1C). As seen in Fig. 1C, compared to the control group, the AAV2 serotype exhibited 27,215.58-fold higher eGFP expression levels, and although AAV-DJ (6,557.42-fold) and AAV2.7m8 (7,908.83-fold) showed high eGFP expression in pLCs^Imt, the AAV2 serotype demonstrated the highest transduction efficiency of the 18 serotypes evaluated. These results indicate that AAV2 is highly efficient for the transduction of pLCs^Imt and that AAV serotypes are cell type-specific.

AAV serotypes show comparable transduction efficiency in PAMs 3D4/31 to pLCs^Imt

To evaluate and compare the PAMs 3D4/31 (Supplementary Fig. 1A, right panel), the cells were transduced with each AAV serotype. The results were then compared with those observed for pLCs^Imt. Similar to pLCs^Imt, PAMs 3D4/31 were infected with all 18 of the AAV serotypes and the transduction efficiency evaluated. As shown in Fig. 2A, the results showed high eGFP expression for AAV2, AAV-DJ, and AAV2.7m8 as compared to the control, with weak green fluorescence observed in other serotypes, such as AAV1, AAV6, AAV6.2, and AAV2-QuadYF, and almost no eGFP expression for AAV3 and AAV5. No green fluorescence was observed in PAMs 3D4/31 infected with AAV4, AAV7, AAV8, AAV9, AAV-rh10, AAV-DJ/8, AAV-PHP.eB, AAV-PHP.S, and AAV2-retro. In particular, quantification of the eGFP expression for the AAV2 and AAV-DJ serotypes in PAMs 3D4/31 showed higher percentages of 64.08 ± 2.4% and 52.81 ± 4.73% as compared to the other AAV serotypes (Fig. 2B).

Figure 2.Transduction efficiency of individual AAV serotypes containing eGFP on day 3 post-infection in PAMs 3D4/31. (A) Representative eGFP (green) fluorescence images in PAMs 3D4/31 three days after transduction of various AAV vectors. Individual AAV serotypes were infected at MOIs of 20 K for 18 h. The mock was not treated with the AAV vector and served as the negative control. Nuclei were stained with Hoechst 33342 (blue). Scale bars indicate 250 μm. (B) Quantification of eGFP-positive cells in different AAV serotype-infected PAMs 3D4/31. Data were presented as a mean ± standard error. Two independent co-workers counted. Different letters denote significant differences within each group, ^a-fp < 0.05 (n = 3). (C) Comparison with the mRNA expression level of eGFP in AAV serotype infected-PAMs 3D4/31. RQ of the gene expression was calculated using the ΔΔCt method, with data presented as RQ. Error bars represent the RQ for the minimum and maximum expression levels around the mean. All of quantification were normalized to the internal housekeeping gene GAPDH. Different letters denote significant differences within each group, ^a-ip < 0.05. All experiments were performed in triplicate. PAMs 3D4/31, pulmonary alveolar macrophages 3D4/31.

To further determine the transduction efficiency of the various AAV serotypes, the mRNA expression levels of eGFP were also genetically assessed in PAMs 3D4/31 (Fig. 2C). As shown in Fig. 2C, compared with the control, the AAV2 serotype exhibited 1,951.98-fold higher eGFP expression levels, with the AAV-DJ serotype (1,921.74-fold) showing marginally lower eGFP expression than the AAV2 serotype, and the next highest eGFP expression levels exhibited by AAV2-QuadYF (1,341.23-fold) and AAV2.7m8 (680.99-fold). Similar to pLCs^Imt, AAV2 demonstrated the highest transduction efficiency for PAMs 3D4/31.

Transduction efficiency of individual AAV serotypes in porcine-derived cell types

Comprehensive analysis of the serotype transduction efficiencies for the various AAV serotypes in pLCs^Imt and PAMs 3D4/31 unequivocally demonstrated the highest transduction efficiency for the AAV2 serotype across both cell types (Fig. 3A, i-vi), with the AAV-DJ and AAV2.7m8 serotypes manifested sequentially lower yet substantial transduction efficiencies. In contrast, analogous transduction experiments in PK-15 cells resulted in a disparate serotype preference (Supplementary Fig. 3A and 3B), with AAV-DJ emerging as the serotype with paramount transduction efficiency, surpassing both AAV2 and AAV2.7m8. Notably, AAV2, which has previously shown superior efficacy in lung cells (Fig. 3A, i and iv), exhibited the lowest transduction efficiency among the three serotypes in renal cells. These findings indicate a marked cellular specificity for the AAV serotype transduction efficiencies, which underscores the differential affinity and transduction potential of the distinct AAV serotypes in various cellular environments. Specifically, the data showed pronounced tropism for AAV2 in both pLCs^Imt and PAMs 3D4/31, while the AAV-DJ serotype demonstrated a superior propensity for transduction in porcine renal cells. This cellular specificity is indicative of inherent differences in the cellular receptor availability, intracellular trafficking mechanisms, and nuclear entry pathways among the cell types examined (Fig. 3B).

Figure 3.Comparison of distinct AAV serotypes in vitro transduction efficiency in porcine. (A) Representative images showing significant eGFP expression (green) in both the pLCs^Imt and PAMs 3D4/31 transduced with AAV2, AAV-DJ, and AAV2.7m8 serotypes. Scale bars indicate 250 μm. (B) Schematic illustration demonstrating the cellular specificity of the various AAV serotypes assessed in this study. Individual AAV serotypes show different cellular tropism in the porcine model.

Genetically engineered pigs hold significant promise for application in both agriculture and biotechnological research. Nevertheless, their production remains challenging, costly, and time-intensive, primarily due to the lack of authentic germline-competent pluripotent stem cells (Fan and Lai, 2013). Due to these limitations, living Cas9-expressing GE pigs attracts considerable interest and substantially facilitate in vivo functional investigation of genes related disease resistance, such as PRRS, in agricultural and livestock industry (Whitworth et al., 2016; Wang et al., 2017).

AAV vectors have been rigorously investigated for their potential use in GE and biotechnological researches (Bijlani et al., 2022). Owing to their versatility, non-pathogenic characteristics, remarkable efficiency in gene delivery, and sustained persistence, AAVs have been projected to be extensively utilized across a broad range of fields (Asmamaw Mengstie, 2022; Issa et al., 2023), indicating their higher value. To date, 12 distinct AAV serotypes and over 100 natural variants have been identified (Balakrishnan and Jayandharan, 2014). Numerous AAV serotypes and variants exist, each exhibiting distinct transduction efficiencies depending on the specific cell or tissue type (Srivastava, 2016; Wiley et al., 2018; Issa et al., 2023). Several researches showed the importance of selecting the optimal AAV serotype for efficient gene delivery, particularly when targeting specific tissue including the lung (Liu et al., 2020; Yoon et al., 2021). Therefore, selecting an appropriate AAV serotype is crucial for effective AAV-based gene delivery. Our analysis of AAV recombinant serotypes using the porcine lung cells-derived from the Cas9-expressing pig (Kim et al., 2024) showed that AAV2 has the higher and superior tropism for pLCs^Imt. In addition to pLCs^Imt, AAV2 showed substantially higher transduction efficiency in PAMs 3D4/31. These findings indicate the specificity and superior efficiency of AAV2 across both cell types compared to other serotypes tested, thereby underscoring the importance of selecting the optimal AAV serotype for efficient gene delivery.

The AAV2 is capable of infiltrating and transducing various cellular phenotypes including lung across interspecies boundaries (Halbert et al., 2000; Chen et al., 2005). AAV2 has reported that it has also remarkable efficiency in neurons, hepatocytes, and mesenchymal stromal cells compared to other serotypes, such as serotypes 1, 3, 4, 5, 6, and 8 (Chng et al., 2007; Stender et al., 2007; Srivastava, 2016; Logan et al., 2017). In our study, AAV2 showed the highest transduction efficiency in both pLCs^Imt and PAMs 3D4/31, with eGFP expression levels of 52.46 ± 2.4% in pLCs^Imt and 64.08 ± 2.4% in PAMs 3D4/31 compared the others. The results firmly underscore its potential as a promising vector for the effective gene delivery in porcine lung tissues. Meanwhile, AAV-DJ and AAV2.7m8 also exhibited significant transduction capabilities in both cell types in this study, although their efficiencies were lower than that of AAV2. These results are consistent with Lisowski et al. (2015), who reported broad tropism for AAV-DJ across various tissues. AAV2.7m8, a variant of AAV2 modified to enhance its tropism (Bennett et al., 2020), also performed well, suggesting that engineered variants such as AAV2.7m8 hold promise for applications where high transduction efficiency is specifically required (Mendell et al., 2021). Interestingly, AAV-DJ exhibited relatively higher transduction efficiency than both AAV2 and AAV2.7m8 in PK-15 cells, indicating that the sensitivity of AAV varies in a cell-type-specific manner (Grimm et al., 2008; Issa et al., 2023).

In contrast, the other serotypes such as AAV4, AAV-rh10, AAV-DJ/8, AAV-PHP.eB, AAV-PHP.S, and AAV2-retro exhibited minimal to no transduction in both lung cell types in this study. In case of AAV4, although it has been reported potential as a vector for lung tissue-specific gene delivery (Issa et al., 2023), it has been reported to exhibit low overall expression levels in murine lung tissue (Zincarelli et al., 2008). In fact, AAV4 has primarily showed higher transduction efficiency in retinal cells in rodent, canine, and non-human primate (Weber et al., 2003). Furthermore, AAV-DJ/8 has been reported that it showed diverse tropisms and proclivity for liver, spleen, and neural cells (Grimm et al., 2008). In case of AAV-PHP.eB and -PHP.S, they display a predilection for the central nervous system (Chan et al., 2017). Similar with AAV-PHP.eB and -PHP.S, AAV2-retro demonstrates specific tropism for spinal neurons (Wang et al., 2018). Previous studies have also highlighted the variable efficiency of these serotypes in different tissues. These findings underscore the importance of appropriate AAV serotype selection according to target cell types once more (Su et al., 2008; Haldrup et al., 2024).

In this way, the transduction efficiency observed in tissues does not always translate reliably to isolated cell types due to the factors such as the local microenvironment, receptor expression levels, and intracellular tracking pathways, which can vary between tissues and isolated cells (Zengel et al., 2023). While our study only focused on the transduction efficiencies in Cas9-expressing pig-derived lung cells (Kim et al., 2024), further validation using isolated and purified cell types could enhance our understanding of species- and cell type-specific transduction mechanisms of AAVs (Qi et al., 2013).

However, one of the limitations of this study is the lack of assessment of the effect of pre-existing neutralizing antibodies (NAbs) on the infection efficiency of the tested AAV serotypes. Neutralizing antibodies are immune system proteins that specifically recognize and bind to viral particles, effectively preventing them from infecting cells. Pre-existing NAbs can significantly hinder the efficacy of AAV-mediated gene delivery, as they can neutralize the viral vector before it reaches the target cells (Kotterman et al., 2015; Gorovits et al., 2020). Studies have shown that NAb prevalence against AAV2 is notably high in both humans and pigs, potentially limiting its effectiveness in clinical applications (Calcedo et al., 2011; Dai et al., 2022). NAbs can vary significantly across populations and species, leading to variability in vector efficacy. This variation complicates the translation of preclinical findings to clinical settings and impacts applications in livestock health and productivity. Therefore, future research should include a comprehensive evaluation of NAbs to better predict and enhance the effectiveness of AAV-based gene therapies in both human and livestock contexts. Specifically, understanding the role of NAbs in livestock is crucial to ensure the success of gene delivery methods aimed at improving productivity, disease resistance, and overall animal health. Strategies to circumvent the challenges posed by NAbs could include engineering novel capsid variants that evade recognition by existing antibodies, using empty capsids as decoys to absorb NAbs, or employing immunosuppressive regimens to temporarily reduce antibody levels during vector administration.

Our study focuses on the application of in vivo GE technologies at enhancing disease resistance, such as PRRS, in pigs. To achieve this, we utilized the lung cells derived from Cas9-expressing pig models to identify the optimal and efficient AAV vector carrying CD163 guide RNA. In this process, we found that AAV2 exhibited considerably higher transduction efficiency in lung epithelial cells. Despite the many promising suggestions for AAV delivery systems into the lung, there are still challenges that require practical consideration. These findings from our attempt would help in applying AAV-based in vivo GE for PRRS in the future. The current study is meaningful in that AAVs with high transduction efficiencies for transgene delivery into the lung in Cas9-expressing pig were compared and determined among various AAV serotypes.

Our data indicate that AAV2 was identified as the most efficient serotype for transducing pLCs^Imt and PAMs 3D4/31, highlighting its potential for pulmonary GE application. Pigs have been used as a useful model and important resources in livestock industry as well as biotechnological research (Bähr and Wolf, 2012; Hryhorowicz et al., 2017; Wells et al., 2017; Clauss et al., 2019). Besides, ensuring the safety of GE is crucial, particularly minimizing immune responses and off-target effects on organs by selecting the efficient vector in AAV-mediated GE application. These findings emphasize the importance of serotype selection in gene delivery and provide a foundation for further research aimed at optimizing AAV vectors for in vivo GE application in livestock industry. The demonstrated cellular specificity of AAV transduction also calls for continued investigation of the factors influencing vector performance in different cellular contexts, paving the way for tailored GE approaches.

Supplementary material can be found via https://doi.org/10.12750/JARB.39.4.254

This study was supported by 2024 the RDA Fellowship Program of the National Institute of Animal Science, Rural Development Administration, Republic of Korea.

Conceptualization, W.S.J. and S.K.; methodology, W.S.J. and S.K.; investigation, W.S.J., A.C., and S.K.; data curation, W.S.J., A.C., and S.K.; writing-original draft preparation, W.S.J. and S.K.; writing-review and editing, W.S.J., J-Y.L., H.L., J.N., S.L., K.O., J.G.Y., and S.K.; project administration, S.K.; funding acquisition, S.K.

This research was supported by the Cooperative Research Program for Agriculture Science and Technology Development (Project No. PJ015712) of Rural Development Administration, Republic of Korea.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

No potential conflict of interest relevant to this article was reported.