Comparative analysis of swine leukocyte antigen gene diversity in European farmed pigs

Summary In Europe, swine represent economically important farm animals and furthermore have become a preferred preclinical large animal model for biomedical studies, transplantation and regenerative medicine research. The need for typing of the swine leukocyte antigen (SLA) is increasing with the expanded use of pigs as models for human diseases and organ‐transplantation experiments and their use in infection studies and for design of veterinary vaccines. In this study, we characterised the SLA class I (SLA‐1, SLA‐2, SLA‐3) and class II (DRB1, DQB1, DQA) genes of 549 farmed pigs representing nine commercial pig lines by low‐resolution (Lr) SLA haplotyping. In total, 50 class I and 37 class II haplotypes were identified in the studied cohort. The most common SLA class I haplotypes Lr‐04.0 (SLA‐1*04XX‐SLA‐3*04XX(04:04)‐SLA‐2*04XX) and Lr‐32.0 (SLA‐1*07XX‐SLA‐3*04XX(04:04)‐SLA‐2*02XX) occurred at frequencies of 11.02 and 8.20% respectively. For SLA class II, the most prevalent haplotypes Lr‐0.15b (DRB1*04XX(04:05/04:06)‐DQB1*02XX(02:02)‐DQA*02XX) and Lr‐0.12 (DRB1*06XX‐DQB1*07XX‐DQA*01XX) occurred at frequencies of 14.37 and 12.46% respectively. Meanwhile, our laboratory has contributed to several vaccine correlation studies (e.g. Porcine Reproductive and Respiratory Syndrome Virus, Classical Swine Fever Virus, Foot‐and‐Mouth Disease Virus and Swine Influenza A Virus) elucidating the immunodominance in the T‐cell response with antigen specificity dependent on certain SLA‐I and SLA‐II haplotypes. Moreover, these SLA–immune response correlations could facilitate tailored vaccine development, as SLA‐I Lr‐04.0 and Lr‐32.0 as well as SLA‐II Lr‐0.15b and Lr‐0.12 are highly abundant haplotypes in European farmed pigs.

The porcine major histocompatibility complex (MHC) harbours the highly polymorphic swine leukocyte antigen (SLA) class I and II gene clusters encoding glycoproteins which present antigenic peptides to T cells that are required to stimulate the adaptive immune response (Lunney et al. 2009;Hammer et al. 2020;Kamal et al. 2020). As pathogen effects on SLA gene expression drive swine immune responses, the SLA complex plays a key role for swine models in biomedical research (reviewed in Hammer et al. 2020). Associations of SLA class I and/or class II genes or haplotypes with differences in swine vaccine and disease responses are well documented (reviewed in Lunney et al. 2009). In vaccine research, either genetically defined pig lines (e.g., Babraham pigs) or outbred pig lines are used (Tungatt et al. 2018;De Le on et al. 2020). As well as using SLA-typed animals in vaccine research, pigs are often used to develop disease models and for basic research studying allogeneic and xenogeneic transplantation (reviewed in Ladowski et al. 2019;Hammer et al. 2020;Ladowski et al. 2021). To understand and control SLA complexity, mainly miniature swine models are used to establish SLA-inbred/-defined pig lines (reviewed in Hammer et al. 2020;Ladowski et al. 2021). In contrast, in vascularised composite allograft transplantation or for end-stage renal disease, porcine transplantation models have been established with SLAmismatched outbred pigs (I. Arenas Hoyos et al. and M. Jensen-Waern et al. unpublished data).
Here we propose two underlying rationales for conducting SLA haplotyping-assisted animal trials in vaccine and transplantation research: (i) SLA typing of the resource population enables directed mating of founder animals based on their SLAbackground (Fig. S1): and (ii) the designation of SLA-defined study groups achieves an experimental advantage of preselecting animals expressing certain SLA phenotypes and thus enhancing the understanding of experimental outcomes (Fig. S1). As a prerequisite for transplantation and vaccine research, our laboratory provides information about the MHC background usingy high-throughput low-resolution (Lr) SLA haplotyping in swine specifying SLA gene-specific allele groups (reviewed in Hammer et al. 2020).
We have contributed to several correlation studies addressing vaccine design against Porcine Reproductive and Respiratory Syndrome Virus (PRRSV), Classical Swine Fever Virus, Foot-and-Mouth Disease Virus (FMDV) and Swine Influenza A Virus (FLUAVsw) by SLA haplotyping outbred pigs. Furthermore, our laboratory is involved in studies with minipigs for various purposes in transplantation research (Fig. 1, Table S1). In this study, we present comprehensive data about SLA alleles and low-resolution haplotypes and their prevalence in nine commercial European pig populations.
A total of 549 farmed pigs ( Fig. 1, Table S1) representing nine commercial pig lines were genotyped for their SLA class I and II haplotypes by running low-resolution PCR screening assays on Peripheral blood mononuclear cell (PBMC)-or whole blood-derived genomic DNA. Therefore, genomic DNA was isolated from 5 9 10 6 porcine PBMCs or 200 µl whole blood using commercial kits following the manufacturer's instructions (DNeasy Blood and Tissue Kit, Qiagen; E.Z.N.A. â Blood and Tissue DNA Kit, Omega Biotek, Inc.). SLA class I (SLA-I) and SLA class II (SLA-II) lowresolution haplotypes (Lr-Hp) were identified by a PCRbased typing assay to define the animals' MHC backgrounds on the allele-group level. SLA typing was performed by PCR with the complete set of typing primers specific for the allele groups of three SLA-I loci (SLA-1, SLA-2 and SLA-3) and three SLA-II loci (DRB1, DQB1 and DQA) (Table S2; Ho et al. 2009aHo et al. , 2010Essler et al. 2013;Gimsa et al. 2017). The criteria and nomenclature used for SLA-I and SLA-II haplotyping were based on those proposed by the SLA Nomenclature Committee (Ho et al. 2009b andreviewed in Hammer et al. 2020). Interpretation of the results was deduced from the presence of allele-specific PCR products of the expected size in each lane. Low-resolution SLA-I and -II haplotypes were assigned based on comparison with previously published haplotypes (Ho et al. 2009a(Ho et al. , 2010Gao et al. 2017, reviewed in Hammer et al. 2020) and unpublished breed-or farm-specific haplotypes (C.-S. Ho et al. unpublished data).
The studied cohort of 549 farmed pigs representing nine commercial pig lines comprised 50 SLA-I Lr-Hp, including three potential novel allele-group combinations (Lr-01.0/ 04.0, Lr-V.0, Lr-Y1.0) ( Table 1)    published SLA-typing studies, Lr-04.0 was also found in the pig populations (i) of studies from the Kansas State University (KSU, PRRSV study, unknown breed raised in the USA), (ii) of studies with Porcine Circo Virus (PCV, pigs with susceptibility to subgroups of PCV type 2, unknown breed raised in the USA), (iii) of the Big Pig group (Large White/Landrace crosses raised in the USA) and (iv) of Yorkshire pigs of Canadian origin (Ho et al. 2009a;Gao et al. 2017). In contrast, Lr-32.0 was observed only in the pig groups Big Pig and Landrace of Canadian origin (Ho et al. 2009a;Gao et al. 2017). Lr-22.0 and Lr-01.0 were shared with KSU, PCV and Big Pig, and the latter with the Yorkshire only (Ho et al. 2009a;Gao et al. 2017). Lr-59.0 was only found within the PCV group, Lr-43.0 was found in the KSU group and Lr-37.0 was shared in Yorkshire pigs, but Lr-24.0 did not occur in any of these five studied cohorts (Ho et al. 2009a;Gao et al. 2017).

2
The alphabetical suffix in haplotype designations was used to differentiate between closely related haplotypes (i.e. haplotypes with identical lowresolution group specificities, but different allele specificities).
5 Not yet confirmed haplotype.
In veterinary vaccine design, the characterisation of the peptide-binding specificity of SLA-I and SLA-II molecules is pivotal to understanding adaptive immune responses of swine towards infectious pathogens (reviewed in Hammer et al. 2020). Herein we briefly discuss key findings on the correlation of SLA haplotypes and immune responses for the animals enrolled in this study. Immunity against the PRRSV is not well understood, although there is evidence suggesting that virus-specific T-cell IFN-c responses play an important role. It was demonstrated that PRRSV-vaccinated and challenged pigs carrying SLA-I haplotype Lr-01.0/04.0 or Lr-59.0 and SLA-II haplotype Lr-0.27 showed significant IFN-c responses, pointing towards a positive correlation of SLA haplotype and T-cell response (Burgara-Estrella et al. 2013). Another PRRSV study suggested that the antigenic region NSP5 156-167 could be restricted by the SLA-I haplotype Lr-22.0, meaning that a T cell will only respond to this particular antigen when it is bound to either SLA-1*08XX, SLA-3*06:01 or SLA-2*12XX. Additionally, pigs demonstrating CD4 + T cell responses to the antigenic peptide M 29-43 were haploidentical, sharing both SLA-II haplotypes Lr-0.01 and Lr-0.15b. This combination appearing exclusively in these animals suggests restriction by one of these two haplotypes (Mokhtar et al. , 2016. A proteome-wide screening revealed immunodominance in the CD8 T-cell response against Classical Swine Fever Virus with antigen specificity dependent on SLA-I haplotypes. The variability in the antigen-specificity of these immunodominant CD8 T-cell responses was confirmed to be associated with the expression of distinct SLA-I haplotypes. Moreover, recognition of NS2 1223-1230STVTGIFL (Lr-22.0) and NS3 1902-1912 VEYSFIFLDEY (Lr-01.0) by a larger group of C-strain vaccinated animals showed that these peptides could be restricted by additional haplotypes (Franzoni et al. 2013).
In the analysis of FLUAVsw, the porcine T-cell response has been poorly characterised to date. In a cohort of 40 outbred pigs, Talker and co-workers showed that animals with a strong expansion of Ki-67 + CD8b + T cells and the highest frequencies of FLUAVsw-specific cytokine-producing CD4 + T cells were homozygous for the SLA-I haplotype Lr-01.0 and for the SLA-DQA locus (DQA*02XX) (Talker et al. 2015(Talker et al. , 2016. In 2018, Schwartz and co-workers fully characterised the SLA background of the inbred Babraham pigs at a high-resolution level: SLA-1*14:02-SLA-3*04XX-SLA-2*11:04 and DRB1*05:01-DQB1*08:01/02-DQA*01:03. Based on this SLA-defined pig model, it was then possible to develop a toolset that included the identification of novel immunodominant FLUAVsw-derived T-cell epitopes (Schwartz et al. 2018;Tungatt et al. 2018).
Previous studies showed the promising potential of dendrimer peptides as vaccine candidates against FMDV. Several B-cell epitope dendrimers, harbouring a major FMDV antigenic B-cell site in VP1 protein that is covalently linked to heterotypic T-cell epitopes from 3A and/or 3D proteins, elicited consistent levels of neutralising antibodies and IFN-c-producing cells in pigs (De Le on et al. 2020). Robust correlations of certain SLA haplotypes (Lr-22.0, Lr-59.0, Lr-0.15b, Lr-0.24 and Lr-0.27) with antibody titres and IFN-c-producing cells support the contribution of SLA class-II restricted T-cells to the magnitude of the T-cell response and to the antibody response evoked by the B 2 T dendrimers, being of potential value for peptide vaccine design against FMDV (De Le on et al. 2020). In addition, Patch and colleagues used inbred minipigs to show that FMDV infection results in induction of cytotoxic T cell responses that are classically antigen specific and MHC restricted (Patch et al. 2014). Following on, these investigators used SLA-1*04:01 and SLA-2*04:01 class I tetramers to show that, upon vaccination with replication defective adenovirus 5 vectors expressing the FMDV P1 protein, T cell specificities expand with each vaccine boost (Pedersen et al. 2016).
of Animal Sciences, Wageningen University), Artur Summerfield (Institute of Virology and Immunology, Vetsuisse-Faculty, University of Bern) and Simon Graham (The Pirbright Institute). This work is part of a startup project financially supported by Profile Line 2 'Infection and prevention' from the University of Veterinary Medicine Vienna, Austria. aerosol delivery of vaccine antigen or virus in the Babraham inbred pig. PLoS Pathogens 14, e1007017. eCollection 2018 May.

Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Table S1 Detailed list of European farmed pigs incorporated in the present study.
Table S2 Plate layout of the PCR primer panel for genotyping swine leukocyte antigen class I (a) and class II (b) alleles.

Figure S1
Two basics concepts for swine leukocyte antigen haplotyping-assisted animal trials in vaccine and transplantation research. Figure S2 Frequency of swine leukocyte antigen class I (a) and class II (b) low-resolution haplotypes identified in 549 and 341 European farmed pigs by PCR screening assays respectively. Figure S3 Swine leukocyte antigen class I (a) and class II (b) low-resolution haplotype diversity in nine and seven European commercial pig populations respectively.