Humoral immunity is an immune mechanism by which B lymphocytes produce antibodies to achieve the purpose of protection. The key process of humoral immunity is the production of highly effective and short-lived effector B lymphocytes, which secrete antibodies to remove antigens. B lymphocytes, also called “B cells” could generate nearly an infinite number of antibodies with diverse binding sites through combinatorial rearranging of large numbers of duplicated genes that encode antibody specificity, by modifying the joining sites using exonucleases and terminal deoxynucleotidyl transferase (TdT) and then targeting these rearrangements for SHM using nucleases like activation-induced cytidine deaminase (AID) (1). There was a specific B cell receptor (BCR) on the surface of each B cell, which directly binds the antigen to humoral immunity. When an antigen invades the body, only B cells with complementary BCR can combine with the antigen, then receive selective stimulation, and then clone on a large scale (clonal selection theory) (2). Classical immunoglobulin has a Y-shaped structure which consists of two identical IgL and two identical IgH (3). The variable region of IgH and Igκ/Igλ constitute the epitopes of immunoglobulin, which directly determines the specificity and affinity of the antibody to antigen by binding with antigen. The constant region mainly regulates immune cells and effector molecules or cells. The physicochemical and structural properties of CDRs between Igκ and Igλ were quite different, but there was no significant difference in their pairing with IgH (4).

In response to a variety of pathogenic microorganisms in the environment, the body has evolved a variety of mechanisms, such as V(D)J recombination, gene conversion (GCV), SHM and class switch recombination (CSR), resulting in immunoglobulin diversity (5). In addition to CSR targeting constant regions, the other four mechanisms enrich the types of antibodies by increasing the diversity of variable regions. Recombination-activating genes 1/2 (RAG 1/2) and high-mobility group (HMG) protein initiate V(D)J recombination, specifically recognize recombination signal sequence (RSS) on the flank of the recombinant gene and form RAG complex, and then cut the RSS, repair the DNA double strand break (DSB) and complete rearrangement (6–9).

The schematic structure of the genomic organization and diversity of immunoglobulin in some ruminants had been reported. In 2016, Dr. Ma completed the genome organization structure of bovine IgH. The IgH consists of 43 VH genes, 23 DH genes and 12 JH genes. The bovine IgH had an ultra-long CDR3H not found in other species, encoding up to 65 amino acids (AA) (10, 11). Ultra-long CDR3H had previously been found in IgNAR of sharks and was thought to exist only in cartilaginous fish. But then longer ultra-long CDR3H was found in cattle, which presents a unique ‘stalk and knob’ structure in the crystal of ultra-long CDR3H antibodies. Different paired combinations of cysteine (Cys) in CDR3H form disulfide bonds, thus forming a variety of different spatial configurations of antigen binding sites under the same CDR3H sequence (12, 13). Goat IgH contains eight VH genes, three DH genes and six JH genes, and only three VH genes and three DH genes and two JH genes were involved in the VDJ recombination. The composition of IgH was similar between sheep and goats, but the number of germ line genes is much less than that of cattle. In bovids, due to the low level of κ gene expression in cattle, there were few studies on the κ chain in cattle, Dr. Qin and Dr. Yu completed the study on the structure and expression diversity of immunoglobulin κ gene in sheep and goat in 2015 and 2020, respectively (14, 15). The sheep λ chain consists of 32 Vλ genes, three Jλ genes and two λ genes, and only seven Vλ genes have potential functions. The goat λ chain consists of 35 Vλ genes, three Jλ genes and three λ genes, just like sheep, there are only seven Vλ genes with potential function. The number of V, D and J genes germline genes in cattle was slightly higher than that in sheep and goats, but far lower than that in humans and mice.

Known as the boat of the plateau, yak live on the Qinghai-Tibet plateau at an altitude of 3 km above sea level and showed satisfactory cold and disease resistance. Therefore, the study of the structure and diversity mechanism of the yak immunoglobulin gene could be of academic and practical significance. Although there was a breakthrough in the structure of immunoglobulin in cattle and sheep, the insufficient basic knowledge of yak immune function hindered our ability to breed and maintain optimal yak health. Therefore, the study of the immunoglobulin gene map and diversity mechanism of yak is of great significance for disease resistance breeding, disease prevention and vaccine production of yak.

All experimental procedures were carried out accordance according to the Regulations on the Administration of Affairs Concerning Experimental Animals approved by the State Council of the People’s Republic of China. The research was approved by the Institutional Animal Care and Use Committee of Northwest A&F University.

It is critical to understand the natural immune mechanisms that fight infection and how vaccination, biosafety, nutrition, livestock and management practices could be used to maintain and enhance immune protection (43). The sequence and structure of IgH constant genes in yak were similar to those in bovids (11). Interestingly, two μ genes with high similarity were found in the genome of cattle, but only one μ gene had been found in the genome of other bovids (yak, sheep and goat) (11, 24, 44, 45). Three JH genes were found in the yak genome, but none of them were transcribed. The three JH genes did not have typical ‘WGXG’ and ‘TVSS’ AA, which might be pseudogenes (45). The results showed that more than 95% of recombinant sequences in yak expressed bovine JH10 gene and 4% of recombinant sequences expressed bovine JH6 gene. By comparing JH gene sequences of yak and cattle, it was found that yak JH1 gene was highly similar to bovine JH1 gene, yak JH2 gene was highly similar to bovine JH5 gene, and yak JH3 gene was similar to bovine JH12 gene. Therefore, we speculated that the sequence between bovine JH5-JH6-μ1-…-JH7- JH8- JH9- JH10- JH11 might be missing in yak genome. The IgH span of yak was 475 kb, that of cattle was 678 kb, and that of JH5-JH6-μ1-…-JH7-JH8-JH9-JH10-JH11 was about 200 kb, which was consistent with the difference between the length of cattle and yak IgH. In the study of VDJ recombination of cattle, JH10 was recombined with the μ2 gene, and JH6 was recombined with μ1 gene. For these reasons, some JH genes and one μ gene might be missing in the yak genome. There was also a strong preference for yak DH genes in VDJ recombination. DH 28 and DH 34 were mainly used in VDJ recombination, which had obvious (YG)n AA characteristics, which was consistent with the characteristics of bovine DH8 gene (11). There were 42 germline VH genes in the yak genome, only 9 VH genes with potential function, all of which belong to clan II. The genes in clan II were about 90% homologous. Yak V gene contributed less to VDJ recombinant diversity, which was similar to horse. There were 54 species VH genes in horse, only VH4 and VH5 (clan II) were mainly used in VDJ recombination. Figure 1E showed that horse VH clan II and Yak VH clan II belong to the same branch (26). Since the VDJ recombination of bovine IgH was not reported. The three VH genes involved in VDJ recombination in goat belong to the clan I, but the VH genes used in goat were more abundant and less conservative than in other species, so the VH genes in goats contribute more to the diversity of VDJ recombination in bovids (24, 45). In summary, recombinant diversity of VDJ recombination in bovids contributes limited immunoglobulin diversity compared to human.

The previous researchers did not study the diversity of Igκ expression because the expression level of Igκ was extremely low. In this study, the Igκ of yak was amplified successfully, and the structure and diversity of Igκ were analyzed. The Igκ of yak had a typical structure of the genomic organization, containing only one κ gene, 5 Jκ genes and 23 Vκ genes. The κ genes were conserved in most mammals and were highly similar in sequence and number, but rabbits had two κ genes in particular. Yak, mouse, rabbit, horse and pig had five Jκ genes, while goat and sheep had only three Jκ genes and one was 19 bp irregular RSS (20, 25, 46, 47). Jκ1-4 was used in yak and horse VJ recombination and was similar in type and frequency of Jκ used (26). In this study, high-throughput sequencing found that six Vκ genes and four Jκ genes were used in VJ recombination, and Vκ16 and Jκ3 with extremely low frequency (less than 1%) were also found. The abundance of VJ recombination of Igκ in yak was higher than that in sheep and goats. Previous studies have found that Igλ of all species followed the rule of (Jλ-Cλ)n. The structure of the genomic organization of Igλ was Vλn-(Jλ-Cλ)₇-Vλn in horse, Vλn-(Jλ-Cλ)₄ in cattle, Vλn-(Cλ-Jλ)₃-λ4 in pigs, Vn-(Jλ-Cλ)₂, in sheep and Vλn- (Jλ-Cλ)₃ in goat (15, 26). Interestingly, we found that yak Igλ did not follow such a rule, and the structure of the genomic organization of yak Igλ was Jλ1-Jλ2-λ1-λ2-Jλ3-Jλ4-Jλ5-λ3-Jλ6-λ4-Jλ7-λ5-Jλ8-λ6-Jλ9-λ7-λ8. 11 Vλ genes and four Jλ genes (Jλ1, Jλ4, Jλ6, Jλ7) were used in VJ recombination of Equine Igλ (25, 26). Cattle mainly expresses Vλ1a, Vλ1b, Jλ2 and Jλ3. Pig only Jλ2-λ2 and Jλ3-λ3 were used in recombination (47, 48). Only the combination of Jλ1-λ1 was found in the recombination sequence for sheep (14). 11 Vλ genes and four Jλ genes were used for VJ recombination of Equine Igλ (25, 26). The bovine Vλ1a, Vλ1b, Jλ2 and Jλ3 gene were mainly expressed. Jλ2-Cλ2 and Jλ3-Cλ3 have been used for recombination in pigs. Only Jλ1-Cλ1 combinations were found in the recombination sequence of sheep. Six Vλ genes and four Jλ genes were involved in the VJ recombination of yak (27, 29). The abundance of yak VJ recombination was close to that of horse but higher than that of goat, sheep and cattle.

Yak IgH, Igλ and Igκ showed a significant preference for J gene selection when V(D)J was recombined, and the IgH also had obvious preference for DH genes. Studies have shown that a single DH gene fragment could form its own CDR3H pedigree, which was sufficient to meet the needs of B cell development and immune function. However, multiple DH gene fragments are still necessary for full immune function. This study also found that the retention and development of CDR3H in specific species has its own characteristics in terms of length and AA utilization. CDR3H includes 3’ end of VH gene, DH gene, 5’ end of JH gene, V-D and D-J junction sequences (N nucleotides and P nucleotides). The random deletion of the 3’ end of V gene and the 5’ end of J gene, and the random insertion of N and P nucleotides were counted in this study (49). The random deletion of V gene mainly concentrated in 0 - 3 bp, with the largest deletion of 8 bp. The random deletion of J gene was mainly 0 bp, 3 bp, 4 bp and 6 bp, and the longest deletion was 12 bp. Compared with human (6.5 ± 1.7 bp), mouse (5.8 ± 1.7 bp) and cattle (8.8 ± 3.7 bp), the contribution of VH gene fragment to CDR3H length was smaller in yak (5.5 ± 2.1 bp) (11, 20, 50, 51). The contribution of Yak JH gene to CDR3H length was 7.6 ± 3.7 bp, and that of cattle was 15.1 ± 4.0 bp. The most notable and significant difference is the contribution length of DH to CDR3H. The DH contribution length of human and mouse was 14.3 ± 5.5 bp and 10.8 ± 4.7 bp, respectively. The contribution of bovine μ1 and μ2 to CDR3H length was 27.5 ± 8.5 bp and 43.1 ± 25.3 bp, respectively (52). The length of DH contribution of yak was 39.9 ± 13.7 bp, which was between bovine μ1 and μ2. Previous studies have found that cattle have an ultra-long CDR3H, and the longest CDR3H reaching 65 AA, which is not available in other species. By cloning yak IgH, we found the average length of CDR3H in yak was 58.0 ± 15.3 bp. Unfortunately, the longest CDR3 fragment did not match the DH expressed in the genome, which might be due to the missing of some fragment in genome splicing or the large variation of SHM in the CDR3H region. It is also possible that D-D fusion exists (53, 54). It was important since D-D fusions often result in long (at least 24 AA) and ultralong CDR3Hs (55). However, there was no evidence to prove whether D-D fusion really exists and the mechanism of D-D fusion, so D-D fusion was not further analyzed in this study (56). Previous studies had shown that ultra-long CDR3H was considered to be a unique and diverse form of bovine immunoglobulin production, but in this study, we confirmed that ultra-long CDR3H also exists in yak, and CDR3H includes V gene and was also realized by ultra-long DH. Dr. Ma found that the bovine DH gene of ultra-long CDR3H had a ‘YxYx’ feature, which also appeared in the ultra-long CDR3H sequence of yak (11). CDR3H was located in the center of the antigen binding site and plays a key role in determining the specificity and affinity of antibodies (43). DH gene was the main component of CDR3H and largely determines the length and AA composition of CDR3H. As the core element of VDJ recombination, DH gene seems to have different adaptations during the evolution of different animals. As the core element and contributor of CDR3H, DH genes showed large sequence variation among species and were less conserved than VH and JH genes. Thus, the DH gene in a particular species may be the key to producing more protective antibodies and fewer autoreactive antibodies in that species. The antibody of ultra-long CDR3H had a special structure and the ability to bind to special antigens. Additionally, ultra-long CDR3H resulted in “microfolds” allowing bovine IgH to bind antigens that would otherwise be inaccessible (12). The number of Yak VH, DH and JH genes was limited, and it was thought that IgH diversity was achieved through frequent recombination and endogenous mutations in the CDR3H region (57). It was still interesting to know exactly what advantages ultra-long CDRH3 confers on immunoglobulin function. For cattle and yaks, ultra-long CDR3H might have developed as another mechanism for maximizing diversification in the face of limited combinations of V(D)J. Alternatively, the ultra-long CDR3H might have evolved to optimize binding to antigens from rumen microbes or certain bovine specific pathogens which were not typically encountered by other vertebrates (12, 58, 59).

SHM was an important mechanism of antibody diversity after V(D)J recombination. The AA changes caused by the mutation could improve the affinity of antibodies (5, 60). The SHM of yak IgH had a preference, and the mutation frequency from A to G and G to A was the highest, followed by the mutation between A and T, which was the same as that of cattle, mice, zebrafish, etc., which indicates that the mutation pattern of SHM is evolutionarily conservative (11, 37). However, it is worth noting that in addition to CDR1 and CDR2, high-frequency SHM loci also exist on FR2, possibly because some germ line VH genes exist in unsplicing genome fragments, resulting in an incomplete template library. In addition, studies have shown that SHM is produced in order to increase the diversity of fetal bovine antibodies, and it has also been reported that the point mutations produced by SHM were mainly concentrated at 100-200 bp away from the gene transcription start site, which might be the reason why SHM was not concentrated in CDRs regions in yak (61, 62). The SHM pattern of the yak was different from sheep and goats. The average SHM of the VH gene, Vλ gene and Vκ gene was 8.26%, 6.61% and 16.75%. The SHM frequencies of two horses used in previous study were 8.35% and 7.79% for the VH, 10.24% and 10.27% for the Vλ, and 4.17% and 5.51% for the Vκ (26). The SHM frequencies of the yak VH gene were 7.1%, 10.2% and 7.3%, Vλ gene SHM frequencies were 6.92%, 8.26% and 7.58%, and Vκ gene SHM frequencies were 2.24%, 2.13% and 2.85%, respectively. The SHM frequency of yak H chain and λ chain was similar to that of other large livestock, but the SHM frequency of κ clan was much lower than that of other large livestock. SHM was an irreversible process, and the accumulated SHM were all favorable mutations, and SHM frequency was higher in species with more perfect germinal centers.

In this study, the diversity of yak IgH was analyzed by high-throughput sequencing - PE300, which greatly increased the number of reads analyzed (nλ=16464, nκ=3939), and some low-frequency recombinant sequences of VJ were found, such as Vλ25-Jλ3, Vλ28-Jλ3, Vλ36-Jλ3 and Vλ40-Jλ3. PE300 could avoid the omission of low-frequency expression genes due to the small number of samples. The differences between the findings of existing studies on Igκ in pigs might be influenced by the fact that too few items were sequenced. (47, 63). PE300 was not used for the IgH diversity study in this study, because yak had extremely long CDR3 (59, 64, 65).

By comparing the production mechanism of immunoglobulin diversity in different species, it was found that humans and mice had rich V(D)J recombination diversity, and they could produce rich immunoglobulin types through recombination to resist the invasion of different antigens (66). For species with less V(D)J recombination types, such as rabbit, it compensated for this limitation through GCV of different somatic cells and high levels of SHM, resulting in a more diverse antibody lineage than human or mouse. Structurally, rabbit antibodies rely more on IgL for antigen specificity, using longer CDR3L rings and interdomain disulfide bonds (67). In camels, B cells produce heavy chain only antibodies (HCAbs) in addition to regular antibodies with a IgH/IgL pairing configuration. These HCAbs contained some unusual structural features that further diversify the camel’s antibody library (68). In this study, by analyzing the structure and expression diversity of immunoglobulin gene in the yak, it was found that yak compensated for the restriction of little V(D)J recombination by ultra-long CDR3H and abundant SHM, which was consistent with the previous production mode of immunoglobulin diversity in cattle. However, the mechanism that relies on ultra-long CDR3H to increase immunoglobulin diversity had not been found in other species. Therefore, we hypothesized that the ultra-long CDR3H was the unique mechanism of the rich immunoglobulin diversity acquired during the evolution of the bovid. It was critical to understand the natural immune mechanisms that fight infection and how vaccination, biosafety, nutrition, livestock and management practices could be used to maintain and enhance immune protection. It was hoped that the study on yak immunoglobulin could provide basic theoretical knowledge for yak breeding and immunity research.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, PRJNA805364.

The animal study was reviewed and approved by Institutional Animal Care and Use Committee of Northwest A&F University.

HZ, MW, and XT analyzed the data; MW, XT, WZ, and QL wrote the manuscript; HZ, MW, WZ, QL, and XY carried out the experiment; XY and XS reviewed and edited the manuscript; HZ designed the experiment. All authors contributed to the interpretation of the results and writing of the manuscript. All authors contributed to the article and approved the submitted version.

Project supported by National Natural Science Foundation of China (31972557).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.