Lessons Learned from Monitoring Adeno-Associated Virus-9 Neutralizing Antibody (AAV9-NAb) Seroconversion in a Cohort of Cynomolgus Macaques (Macaca fascicularis)
The purpose of this study was to track seroconversion in a cohort of 12, pair-housed, macaques that were previously screened negative for adeno-associated virus-9 neutralizing antibody (AAV9-NAb). Over a 6-month period, specific biosecurity strategies were implemented with the intention of understanding if following defined protocols could play a role in preservation of AAV9-NAb negative status. AAV9-NAb-negative animals were selected for shipment to the facility approximately 2 months after the initial screening. After arrival, animals were paired and housed in a single room with a dedicated housing corridor, cage wash, and equipment. They were then screened for AAV9-NAb status monthly by 2 different labs to confirm results and ascertain potential for variation of results. Upon initial screening at the facility (within one week of arrival), 2 of the 12 NHPs that were seronegative before shipment had seroconverted to AAV9-NAb positive status. The positive animals and their negative partners were moved to a different room but remained within the same isolated corridor with the same biosecurity practices. Serum that was taken on a monthly basis was also used to screen other AAVs. At the end of the 6-month period, AAV9 NAb status did not change from the time the animals were initially screened on site until the end of the study. Paired animals that were cohoused in the same cage with a positive partner did not seroconvert. Although a control group was not used to validate that biosecurity practices played a role in mitigating seroconversion, unpublished data from a facility employing less restricted biosecurity strategies suggest that the seroconversion process involves a more intricate process.
Introduction
The establishment of SPF macaque colonies that exclude specific viruses is a common and well-documented practice.1–4 The purpose of excluding a particular agent may be related to the health effects of the pathogen on animals, its zoonotic potential, or its impact on research. AAV9 is particularly relevant because it is one of the most common adeno-associated viruses (AAVs) studied in gene therapy.5 Establishing and maintaining an SPF colony that excludes AAV9 would help increase confidence in serostatus before and during a study. In turn, this reduces the need to recruit larger cohorts of animals to ensure seronegative status, thereby mitigating preshipment stressors associated with sedation and blood collection for AAV testing in a large number of animals and allowing the maintenance of bonded social pairs. However, the formation and maintenance of SPF laboratory animals can be a complex and challenging undertaking. This is especially true in cases where the biology and transmission of the excluded agent are unknown or difficult to control. The source of environmental exposure to wild-type AAV is often unclear.
Genetic therapies based on AAV are currently approved to treat retinal, neuromuscular, and clotting disorders.5 AAV drug development requires preclinical studies in relevant animal models to evaluate the safety and efficacy of the therapeutic vector. Among the animal models, mice offer the benefit of an inbred clonal population to study drug pharmacokinetics and pharmacodynamics (PK/PD).6 Importantly, laboratory mice lack preexisting AAV immunity from an environmental infection that can impact drug PK/PD. This contrasts with nonhuman primates (NHPs) that can harbor preexisting anti-AAV humoral and cellular immunity that impacts safety, biodistribution, and gene transfer of AAV vectors.7 For instance, low levels of anti-AAV antibodies can bind and neutralize vector reducing liver gene transfer in macaques.8,9 Moreover, antibody-bound vector was shown to be routed to secondary lymphoid organs, which can potentially increase the risk of immunogenicity.8 Cytotoxic T cells, elicited from a prior infection or vector administration, can still target capsid antigens presented on Class I HLA molecules and eliminate cells that take up the vector.10 The findings of reduced gene transfer in seropositive animals were later confirmed in human clinical trials where patients received intravenous vector administration.11 These findings have led to the exclusion of seropositive subjects from receiving AAV gene therapy.12,13
To accurately model the clinical outcome, only seronegative animals are enrolled in AAV gene therapy studies.13,14 While inbred laboratory mice are seronegative, NHPs and farm animals can be seropositive from environmental exposure to the virus.15 Since only seronegative animals are included in preclinical AAV gene therapy studies,16,17 the process involves first screening a large group of animals at the vendor source to identify seronegative animals. These seronegative animals are then procured, shipped, and quarantined before study start, which could further increase the risk of environmental exposure to AAV. Unexpected seroconversion before study start has resulted in reduced efficacy in NHPs that otherwise demonstrated adequate transduction of an AAVLK03.GFP vector8 and as a result requires increased screening and recruitment of a larger cohort of seronegative animals in prospective macaque AAV gene therapy studies.1–4 The current study was used to better understand the AAV9-NAb and other AAV-NAb seroconversion patterns in a cohort of cynomolgus macaques. The impact of biosecurity practices on seroconversion over a period of 6 months was also compared with unpublished data from another facility using less restricted biosecurity practices.
Understanding seroconversion under specific biosecurity measures could help clarify the risk of AAV environmental exposure and seroconversion among animals held at the study site. Results of this study along with future studies using a control group could lead to a better understanding of the potential for establishing an AAV9 SPF colony. AAV9 was chosen because this is a common AAV studied in gene therapy.5 Banked serum was used to evaluate seroconversion against a larger panel of vectors, including binding antibodies (BAbs) to understand cross-reactivity.
Materials and Methods
Animals.
Six male and 6 female (age: 2 to 3 year; weight: 2.5 to 4.0 kg) Mauritian-sourced cynomolgus macaques (Macaca fascicularis) were used for this study based on results of health screening and suitability for the study. Animals prescreened for NAbs were procured from the animal supplier, shipped, and quarantined before study start (Figure 1). Animal demographics, including identification, sex, age, partnership, initial AAV9-NAb status, and housing room location, are detailed in Table 1. All animal procedures were approved by the Pfizer Pearl River IACUC. Animals were maintained in full accordance with the Guide for the Care and Use of Laboratory Animals (8th ed)18 and the Animal Welfare Act and Regulations19 in an indoor, AAALAC-accredited facility. Macaques were pair-housed in compatible same sex pairs within standard stainless-steel cages (Lab Products, Aberdeen, MD). Diet consisted of a commercial monkey chow (TEKLAD 2050 Primate Diet; Inotiv, West Lafayette, IN) fed twice daily, and treated municipal water was provided ad libitum. Fresh produce and foraging items were provided daily. Rooms were maintained at 21 ± 2 °C with relative humidity between 30% and 70% and 10 to 15 room air changes hourly. Lighting was provided on a 12:12-h light:dark cycle (lights on:off, 6 am:6 pm).
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-24-163
| Pairs/NHP ID | Sex | Age (y) | Weight (kg) | Initial on-site AAV9-NAb status | Housing room assignment after initial screening |
|---|---|---|---|---|---|
| 1 | |||||
| 1A | F | 2 | 2.5 | + | A→B |
| 1B | F | 2 | 2.8 | − | A→B |
| 2 | |||||
| 2A | M | 3 | 2.8 | + | A→B |
| 2B | M | 3 | 3.1 | − | A→B |
| 3 | |||||
| 3A | M | 2 | 3.4 | − | A |
| 3B | M | 2 | 3.7 | − | A |
| 4 | |||||
| 4A | M | 3 | 3.7 | − | A |
| 4B | M | 3 | 4 | − | A |
| 5 | |||||
| 5A | F | 3 | 2.5 | − | A |
| 5B | F | 3 | 2.5 | − | A |
| 6 | |||||
| 6A | F | 2 | 2.6 | − | A |
| 6B | F | 2 | 2.5 | − | A |
Abbreviations: F, female; M, male.
Before arrival, all animals had been socially housed at the supplier with no separation from other facility animals, with staff using standard personal protective equipment (face mask, eye protection, gloves, and gown). A total of 42 NHPs were screened in the original cohort, of which 50% were AAV9NAb positive. The 12 animals selected for shipment were deemed AAV9-NAb seronegative based on screening by the supplier at 8 week preshipment. Upon arrival, animals followed standard facility-based NHP quarantine requirements (approximately 8 weeks) in conjunction with initiation of study sampling. During quarantine, all NHPs underwent physical examination, intradermal tuberculin testing (3 tests at least 2 weeks apart), complete blood count (CBC), serum chemistry, fecal screening (Salmonella, Shigella, Yersinia, and Campylobacter [Campylobacter jejuni and Campylobacter coli]), and viral screening for agents described below. All NHPs were negative for Tuberculosis, Salmonella, Shigella, and Yersinia. Physical examination and CBC, and serum chemistry were unremarkable. The serology results were negative for simian immunodeficiency virus, simian T lymphotropic virus, macacine herpes virus, simian virus 40, and simian retrovirus. All animals were positive for lymphocryptovirus, cytomegalovirus, and macaque rhadinovirus and had protective titers for measles virus due to previous vaccination. Nine of the 12 animals were positive for simian foamy virus.
Upon completion of quarantine, animals remained in the same location for continued monthly study sampling. Biosecurity measures described below were maintained from arrival throughout the study.
Biosecurity.
Study animals were housed remotely from other facility NHPs in a dedicated area, which included dedicated equipment and cagewash (Tecniplast 900XL Chamber Washer; Tecniplast USA, West Chester, PA). The study animals and colony animals were located in the same building, on the same floor of the vivarium in 2 separate hallways, located approximately 350 feet apart. The study hallway had no other traffic flow other than what was specifically required for their care. Before animal arrival, all housing room(s), caging, and handling equipment were sanitized with vaporized hydrogen peroxide (BioQuell Z Mobile Hydrogen Peroxide Vapor Generator; Ecolab, Horsham, PA). Whenever possible, dedicated personnel were used to care for this cohort. If dedicated personnel were unavailable, these study rooms were entered first before entering other facility NHP rooms. Standard facility personal protective equipment (PPE) for entry into NHP rooms included a Tyvek jumpsuit, 2 pairs of shoe covers, a hairnet, an N95 facemask (without valves), goggles, and 2 pairs of gloves.
Upon arrival, all animals were pair-housed within one room (study room A). After the initial AAV9-NAb screening, any NHPs found to be AAV9-NAb positive were moved to a separate room (study room B) for the remainder of the study along with their partners (regardless of AAV9-NAb status). Traffic flow between the 2 study rooms required entry into study room A before entry into study room B. Separate equipment was maintained for each room to avoid cross-contamination.
During the study, as per standard facility standard operating procedures, all cages, room walls, and drainage troughs were cleaned in place daily with hot water. Floors were cleaned and disinfected daily with Peroxigard Ready to Use One-Step Disinfectant Cleaner and Deodorizer (Virox Technologies, Oakville, Ontario, Canada). Cages were changed a minimum of once every 14 days or more frequently as needed. Once per month, NHP rooms (that is, walls, doors, ceilings, fixtures) were cleaned and fully disinfected with the aforementioned product. To reenter study room A, staff were required to don new PPE including shoe covers, Tyvek, gloves, head cover, N95 respirator, and goggles. They entered both rooms twice per day to observe animal well-being. There was dedicated animal handling equipment in each room. Upon the first AAV9-NAb screening on-site, 2 animals were seropositive (1A and 2A); these animals and their partners were moved to an identically treated separate room (study room B). The remainder of the seronegative cohort was cared for first, with the seropositive animals and their partners cared for last. A complete change of personal protective equipment was performed between rooms. Staff were not allowed to return to the room housing AAVnAb-9 sero-positive nonhuman primates.
Study samples.
Blood was collected for study at 7 days postarrival and at monthly intervals thereafter. At each time point, each macaque was lightly anesthetized with ketamine 5 to 10 mg/kg + dexmedetomidine, 0.01 to 0.03 mg/kg IM. Blood samples were routinely collected from the femoral vein. Anesthesia was then reversed with antisedan 0.15 mg/kg IM, and all animals recovered from the procedures uneventfully.
Serum.
Serum was aliquoted into 0.5-mL sterile screw-top vials using aseptic handling techniques. Aliquoting procedures were performed in a Class II/A2 biosafety cabinet to reduce the risk of environmental contamination. Several aliquots from each animal were shipped overnight on dry ice to the diagnostic laboratories for AAV9-NAb screening. Remaining serum aliquots were banked and placed in a −80 °C freezer for later analysis. All serum samples used for assays were heat-inactivated at 56 °C for 30 min before use.
AAV vectors.
The AAV9vector expressing luciferase used in the NAb and BAb assay was obtained from the Viral Vector Core at University of Massachusetts, Medical School, Worcester, MA.
Neutralizing antibody assay.
Serum was diluted serially from an initial 1:5 dilution and incubated with AAV vectors expressing luciferase for one hour before being added to HEK293T cells. 24 hours later, cells were assayed for luciferase expression using the Bright-Glo Luciferase Assay System (Promega Corporation, Madison, WI). Samples were considered NAb positive when serum diluted at least 5-fold and admixed with AAV9 vectors resulted in >50% reduction in transgene expression.
Binding antibody assay (IgM and IgG).
For the IgG and IgM assay, the 96-well ELISA plates were coated with 2E9 particles/well of AAV9-luciferase vector. The plates were incubated overnight at 4 °C. The Next day, the plates were blocked with 5% nonfat milk and incubated for 2 hours at room temperature (RT). During this incubation period, the serum samples were prepared at 1:100 for IgG and 1:500 for IgM assays. The serum samples were then added to the respective plates and incubated for 1 hour at RT. The plates were then washed and incubated with the detection antibody Goat Anti-Monkey IgG HRP for IgG (catalog no. 4700-05; SouthernBiotech, Birmingham, AL) at a 1:5000 dilution and Goat anti-Monkey IgM (catalog no. NBP1-73726; Novus Biologicals, Centennial, CO) at a 1:10,000 dilution for an hour at RT. Then add TMB ELISA substrate for the development of a deep blue color in the wells. Once the color developed, the reaction was stopped with ELISA Stop Solution. The plates were then read in SpectraMax i3 spectrophotometer (Molecular Devices, Inc., San Jose, CA) at an OD of 450 nm (OD450).
Results
Seroconversion among previously naïve animals.
Among the seropositive animals, 1A was weakly positive with a NAb titer of 5, which was close to the negative titer cutoff (less than 5; Figure 2A). NAb titers did not vary in this animal, and the levels were consistent for the period of 6 months. In contrast, animal 2A had a titer of greater than 40, the highest reported titer in the assay, throughout the study. Since greater than 40 was the highest titer evaluated in the assay, it is not certain if the NAb titers fluctuated during study. Nevertheless, the markedly higher NAb titer in this animal was clearly indicative of a recent AAV infection with AAV9 or a closely related AAV serotype.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-24-163
An orthogonal ELISA-based BAb assay that measured both anti-AAV IgG and IgM binding antibodies was used. Animal 2A had elevated AAV9 IgG binding antibodies throughout the study, which was consistent with the presence of high levels of NAbs to AAV9 (Figure 2B). IgG levels in animal 1A were lower with a brief spike (1.5-fold) at the 3-month time point. Surprisingly, the highest IgG levels were observed in animal 5A for the duration of the study. IgG levels trended lower for the first 2 months before rebounding and were still increasing at the conclusion of the study. Even though animal 5A had the highest serum IgGs, the antibodies did not neutralize the AAV9 vector and may represent nonneutralizing binding antibodies.
The IgM levels were not remarkably different in any of the AAV9 NAb-positive (2A and 1A) or IgG-positive (5A, 2A, and 1A) animals (Figure 2C). Instead, IgM levels were highly elevated in animal 1B at the one-month time point, which then continued to decline for the duration of the study. There was no concomitant increase in IgG levels in this animal as is expected of a canonical viral infection where increases in IgM levels precede IgG activation. Similarly, there were no elevated AAV9 IgMs in animal 5A before the spike in IgG levels as is expected during a typical viral infection with IgM levels increasing before the isotype switch to IgGs.
Seroconversion risk against multiple AAVs.
The initial observation of seroconversion was based on the presence of anti-AAV9 NAbs or BAbs. Serum was surveyed for NAbs against a panel of 6 other AAVs belonging to diverse clades: AAV1 and AAV6 (clade A), AAV2 (clade B), AAV8 (clade E), AAV3B, and AAV5 (not assigned to any clade).
The results from this survey demonstrated that the AAV9 NAb seropositive animal, 2A, also had cross-reactive NAbs to AAV8 throughout the study, albeit at a lower level. The animal also briefly became seropositive for AAV3B at the 4-month time point. Most interestingly, animal 2B, which was cohoused with 2A, did not have either AAV9 or AAV8 NAbs. Instead, this animal (2B) was seropositive for AAV1, AAV3B, and AAV6 (Figure 3A, C, and E), although AAV6 NAbs were only detected at months 2 to 5. The higher anti-AAV3B NAb titer (greater than 40) suggests that the animal may have been infected with AAV3B or a similar virus. The brief spike in AAV3B antibodies in the cage mate 2A at the 4- months mark may suggest the possibility of transmission between the animals, although it cannot be ruled out that there is the possibility of antibody cross-reactivity. Among the other animals, only 5B was AAV3B positive. With the exception of NAbs to AAV6, which were not detected at months 1 and 6, NAbs to other capsids were detected throughout the study at a consistent level.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-24-163
Temporal changes in IgG and IgM seropositivity.
The IgG and IgM responses to the 6 additional AAVs were next examined. Animal 5A, which had increasingly high anti-AAV9 IgG levels starting from 3 months, displayed similar kinetics of increase in IgG titer against all other capsids except AAV2 (Figure 4A–F). The lack of binding to AAV2 was unexpected based on its similarity to other capsids (AAV1 and AAV6) that bind IgG in this animal. The presence of cross-reactive binding antibodies to multiple capsids was also in discordance with the NAb assay, which did not demonstrate transduction inhibition of any capsid. Similarly, animal 4B had increasing IgG titers against AAV1, AAV2, and AAV6 starting at month 5 but did not demonstrate neutralization of these capsids.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-24-163
Most of the animals with an IgG-positive response had evidence of anticapsid IgM antibodies. Only animal 1B, which had anti-AAV9 IgM antibodies, also had cross-reactive antibodies that recognized all capsids (Figure 5A–F). The kinetics of IgM decline were consistent with the results observed against AAV9 with IgM levels starting off at a higher level and then rapidly declining within 30 days before following a slower taper. This animal also did not have IgG antibodies against any other capsid.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-24-163
Historical unpublished data.
In place of a control group, unpublished data were used from a Charles River Laboratory (CRL; Wilmington, MA) assessment of 100 Asian-sourced cynomolgus macaques, housed in separate rooms. These animals were arbitrarily assessed based on serum availability and opportunity to test. They were mixed with cohorts from other suppliers. These NHPs were not always kept with the same roommates and occasionally were grouped with cohorts from other suppliers. Sera were collected from NHPs at 3 time points starting with all 100 animals at the start of the study (0 month). Before the next collections, some animals were released for studies, so sera from only the remaining 87 at 4 months and 65 at 7 months postarrival were collected. Of the 65 NHPs remaining at 7 months, the sera from 20 NHPs, 12 males and 8 females, were tested by AAV2, AAV8, AAV9, and AAVrh74 NAb screening assays at sample dilutions of 1/10, 1/20, and 1/40. While the majority of NHPs showed no significant change in AAV NAb titer over the course of the study, a small number of animals showed an increase in NAb titer greater than 4-fold in 4 to 7 months. In addition, CRL did not begin their assessment with seronegative animals, so these numbers do not necessarily indicate seroconversion. Simply, these data show that animals were seronegative at a certain time point and then seroconverted at later time points.
In a second assessment at CRL, 300 male and 300 female cynomolgus macaques from a single supplier were screened on arrival (week 0), and 76 of them were negative for AAV8 NAbs. Another round of AAV8 NAb screening was performed at week 8 with 13/76 sera showing seroconversion. In addition, these 13 seroconverted NHPs (week 0 compared with 8) were screened by AAV2 and AAV9 NAb assays showing a reduced rate of seroconversion for these 2 serotypes indicating seroconversion for different serotypes was not similar in individual NHPs. Approximately 17.1% showed a change in AAV-NAb status over 8 weeks.
Discussion
Reduced AAV gene transduction upon systemic vector administration to seropositive patients has led to the adoption of vigorous assays to evaluate and exclude seropositive subjects from receiving AAV gene therapy.2 Hence, modeling AAV human gene transfer in animals relies on the selection of seronegative animals for study inclusion.20 Serostatus is determined before animal procurement from the vendor and again before study initiation. Since typical studies are 2 to 6 months in duration, a 6-month testing period was chosen. A limitation of the study was the lack of a control group to understand if specific biosecurity procedures might mitigate seroconversion. Further studies comparing biosecurity practices, run concurrently, would be needed to understand if animals housed using these practices might allow them to maintain a negative status. Nevertheless, the logistics of obtaining serum, running the diagnostic immune assays, and reporting results require approximately 2 weeks. The 2-week window between the last serum collection for evaluating NAb antibodies and the administration of the AAV drug substance is a potential risk as well as staff compliance with PPE practices when animals can be exposed to environmental AAV and develop immunity while awaiting study start (unpublished CRL data).
Seroconversion was observed among animals against multiple AAVs using different assays. Interestingly, NAb and IgG binding antibodies correlated in only one of the seropositive animals. This study not only details the risk of seroconversion upon environmental exposure over time but also underpins the variability in anti-AAV antibodies that has broader implications on vector transduction and subject enrollment.
The 2 animals with positive AAV9 NAb findings in the study may suggest an environmental exposure that preceded study start and which elicited stable long long-lasting NAbs. The observation of multiple NAbs to multiple AAVs is consistent with previous observations of cross-reactivity between AAV clades from a natural infection in humans and animals.7
With respect to temporal changes in IgG and IgM seropositivity, it is possible that the binding antibody levels were not sufficiently high to neutralize the vector. Alternatively, the binding may have been at a location that was unlikely to impact vector uptake and/or transduction. It is also possible that the NAb assay was not sensitive enough to detect neutralization. The starting sample dilution was 1:5, and a lower dilution may have detected neutralization. However, this laboratory and others have observed subjects having binding but nonneutralizing antibodies for reasons which are not entirely clear. These antibodies do not block vector uptake in vivo and may instead increase vector transduction of certain tissues.
Despite these heightened biosecurity measures, seroconversion occurred among animals using 2 different assays over the period of 6 months. In the CRL data, of the 76 seronegative animals, 63 remained seronegative despite no extra precautions. Though uncertain, the possibility exists that animals may seroconvert despite precautions. Alternatively, a biosecurity breach may have occurred in transit or at the originating facility, since animals were seronegative before shipment, and, based on seroconversion rates, it would take several weeks for Ab to be detected once animals are exposed.15,21,22 In positive (AAV9) animals, there was a spike in IgG, but not IgM. This might indicate that seroconversion is not related to an acute infection but possibly to a fluctuation in antibody levels, considered a bystander effect. Serum interrogation using the NAb assay demonstrated that animal 2A was AAV9 seropositive throughout the study even though the animal was seronegative before procurement. It is conceivable that this animal had an environmental exposure to AAV9 or a similar capsid before study start. Animal 2A was a male that was cohoused with male animal 2B. Interestingly, animal 2B remained AAV9 negative for the study duration. However, animal 2B was NAb seropositive for both AAV3b and AAV1, albeit the titers against AAV3b were significantly higher (>2 serial dilutions). It remains unclear why NAbs against different AAV serotypes are elicited among some cohoused animals. It is possible that animals 2A and 2B were independently infected by AAVs that were similar to AAV9 or AAV3B, respectively. The cohousing of these seropositive animals was purely coincidental. Alternatively, these animals may have been exposed to the same AAV that nevertheless elicited NAb that cross-reacted against AAV9 in one animal and against AAV3B/AAV1 in another animal.
Animal 2A also had higher IgG binding antibodies against AAV9 throughout the study duration, which was consistent with the presence of NAb. Surprisingly, the cage mate (2B) did not have elevated IgGs against AAV3B/AAV1. It is possible that the sensitivity of the assay used to evaluate IgG against these capsids was not sufficiently high. Serum samples were diluted 100-fold before use in the assay to detect IgG binding antibodies, and a lower dilution may have increased the sensitivity to detect IgGs against these capsids.
Animal 5A was unusual in having IgG binding antibodies against several AAVs that started rising between 3 and 4 weeks after study start. Similarly, animal 4B also had mild elevations in IgG titers after 3 weeks. Despite the similar temporal increases in anti-AAV IgGs against the same capsid (AAV1, AAV6, and AAV8), these animals differed by gender and were housed separately; and neither animal had NAbs against any AAV that was tested in the study.
IgM responses were unchanged in the 2 animals that demonstrated an increase in IgG levels after 3 weeks. This was unexpected given the known kinetics of antiviral immunity where IgM responses peak before an increase in IgGs.9 A similar perplexing finding was the increased IgM levels in animal 1B at study start against all AAVs tested and which then declined by 3 weeks without a concomitant increase in IgG levels against any capsid. The reason for this discrepancy is unclear. Nevertheless, AAV vectors are known to induce tolerance, and several reports have demonstrated compromised T- and B-cell responses following vector administration.23 It is possible that these differences in AAV immunity following a natural infection may vary due to induction of tolerance by the virus. If a subject is truly naïve, it is important to consider why it would have no evidence of humoral or cellular immunity to the vector. The temporal changes in IgG and IgM titers suggest that these levels may be more subject to changes compared with NAb titers, which appear to be more durable. Further studies are warranted to better understand environmental AAV exposure and its impact on host immunity.
These studies were undertaken to understand the risk of seroconversion upon implementing animal husbandry protocols to minimize the risk of environmental AAV exposure. Nevertheless, 50% of the animals were seropositive against at least one of the AAV capsids during the 6-months study based on the different assays used to evaluate seropositivity. However, only one animal (1/12) had consistently high NAbs for the duration of the study. While it is uncertain whether the measures that were implemented reduced environmental exposure to AAV, the seroconversion rate based on NAb seropositivity was low and not very different from previous experience. There are few other published reports looking at the environmental exposure of cynomolgus macaques to AAV. A previous study15 evaluated NAb against AAVs in captive chimpanzees over a period of several years. This study noted yearly fluctuations in NAb titer against several AAVs. Another study24 examined 190 cynomolgus macaque sera for NAb against various serotypes of AAV. The incidence of animals with negative NAb status against AAV2, 8, and 9 was substantially higher in animals housed in a “super-SPF” colony relative to colonies with animals positive for these agents. This colony was negative for simian retrovirus D, Epstein-Barr virus, and cytomegalovirus.24 To our knowledge, this report is the first to detail the more acute flux in antibody titers from a natural infection even when animals were isolated and maintained under measures to reduce environmental AAV exposure. The risk of AAV transmission in studies appears to be consistent with the CRL data where no special practices were in place and animals were comingled. It is possible that staff compliance with PPE may be a risk factor with respect to maintaining the AAV status of an SPF colony such as this. Since productive AAV transmission would also require coinfection with a helper virus, it is conceivable that immunity to the helper virus, as an example antiadenovirus immunity, in some animals may have prevented successful AAV infection.25 The studies presented here also suggest that fluctuations in anti-AAV antibody titers require repeated sampling to confirm serostatus before excluding subjects, especially when titers are at the threshold of seropositivity. Finally, the findings also have implications for transmission of AAV infection between seropositive and seronegative human or animal subjects living in close proximity.
Schematic Showing the Process Involved With Screening, Procuring, Shipment, and Quarantine of Animals Before Study Start
Neutralizing and Binding Antibody Profiles in Cynomolgus Macaques. Serum from 12 cynomolgus macaques was evaluated for neutralizing (NAb) and binding antibodies (IgG and IgM) against AAV9 every month over a period of 6 months. (A) NAb titer changes plotted as a reciprocal of serum dilution. (B and C) Absorbance values at OD450 for IgG and IgM antibodies that bind to immobilized AAV9 capsids using a modified ELISA assay. The x-axes are time in months.
Neutralizing Antibody Titer Changes in Cynomolgus Macaques Against a Panel of AAV Capsids. Serum from 12 cynomolgus macaques was evaluated for neutralizing (NAb) every month over a period of 6 months against 6 different capsids ([A] AAV1, [B] AAV2, [C] AAV3b, [D] AAV5, [E] AAV6, and [F] AAV8). NAb titers are plotted as a reciprocal serum dilution. The x-axes are time in months.
Anti-AAV IgG Binding Antibody Levels in Cynomolgus Macaques. Serum from 12 cynomolgus macaques was evaluated for IgG binding antibodies over a period of 6 months against 6 different capsids ([A] AAV1, [B] AAV2, [C] AAV3b, [D] AAV5, [E] AAV6, and [F] AAV8). IgG levels are plotted as absorbance values at OD450 for antibodies that bound immobilized AAV capsids using a modified ELISA assay. The x-axes are time in months.
Anti-AAV IgM Binding Antibody Levels in Cynomolgus Macaques. Serum from 12 cynomolgus macaques was evaluated for IgG binding antibodies over a period of 6 months against 6 different capsids ([A] AAV1, [B] AAV2, [C] AAV3b, [D] AAV5, [E] AAV6, and [F] AAV8). IgM levels are plotted as absorbance values at OD450 for antibodies that bound immobilized AAV capsids using a modified ELISA assay. The x-axes are time in months.
Contributor Notes