Immunogenetic Study in Chinese Population with Ankylosing Spondylitis: Are There Specific Genes Recently Disclosed?

Purpose. Ankylosing spondylitis (AS) is a systemic, autoimmune disease resulting in the destruction of the affected joints. Over the past 5 years, several new genes or genetic regions associated with AS have been identified in the Chinese population. This paper aims to discuss the major findings and related potential mechanisms of these studies in our population. Recent Findings. In recent years, due to the rapid advances in computational genetics and technology, there has been an increasing list of well-validated genes or genetic regions associated with AS susceptibility. So far, several genes or genetic regions have now been reported in the Han ethnic Chinese population, containing the major histocompatibility complex (MHC), ERAP1, IL-23R, 12q12, 2p15, 5q14.3, and so on. Different hypotheses for disease mechanisms have been investigated on the basis of the functional studies of these genes or genetic regions. Summary. This paper tries to summarize the association of several candidate genes with risk for AS in the Han ethnic Chinese population and aims to identify the novel inflammatory pathways and provide potential strategies for better therapies.


Introduction
Ankylosing spondylitis (AS) is a chronic, systemic, in�ammatory disease that is characterized by the in�ammation of the axial skeleton, the peripheral joints, the attachments of ligaments, and entheses. e main clinical feature of this disease is in�ammatory low back pain, and over time some patients develop spinal immobility and ankylosis. AS appears to have multifactorial etiologies that contain underlying genetic susceptibility and environmental factors.
In the early 1970s, the discovery of the association between HLA-B27 and AS in a sense proved that the familial aggregation of AS was genetically determined [1]. Later, more and more researches proved that HLA-B27 was not the only genetic determinant for AS. As a result of the development of genetic methodologies, from family studies with micro-satellite marker to the genome-wide association study (GWAS), the genes or genetic regions associated with AS were investigated separately. Aer that, many researchers investigated speci�c variants for particular genetic loci by a variety of approaches such as RFLP-PCR and real-time genotyping PCR. On one hand, these researchers in turn have enabled the replication of GWAS results; on the other hand, they have made those associated polymorphisms with AS to be con�rmed in different populations. e aim of this paper is to discuss these �ndings in the Han ethnic Chinese population and what these �ndings provide about the pathogenesis of AS. Table 1 shows the genes and gene regions we mentioned later.

Epidemiology
e prevalence of AS is 0.2-0.54% among Han ethnic Chinese population which accounts for 92% of the whole population [2,3], and it is similar to the prevalence in Europe and America [4,5]. e clinical manifestations, severity, and risk to get the disease can vary by ethnicity, geography, and sex, with a prevalence that is higher in men during their young ages and lower in some populations such as Japan [4]. Twin studies by Brown et al. have shown that the monozygotic twin recurrence rate is 63%, compared with the dizygotic recurrence rate of 12.5% [6]. e recurrence risks in different degrees of relatives in Europeans reported by Brown were �rst degree relatives 8.2% (441/5390), second degree relatives 1.0% (8/834), and third degree relatives 0.7% (7/997) [7]. While, in Chinese population, the recurrence risks reported by Lin were �rst degree relatives 3.84% (341/8869), second degree relatives 0.87% (234/26902), and third degree relatives 0.315% (127/40258) [8]. All these data con�rm that familial aggregation in AS is related to genetic factors rather than environmental ones.

Genetic Mode
For nearly 40 years, AS has been extensively considered as a multifactorial genetic disease with a well-accredited knowledge of linkage to HLA-B27. e frequency of HLA-B27 positive status in the Han ethnic Chinese population ranged from 3.6% to 5.7% [2]. Yet, only 1-5% of these B27 positive individuals develop AS. Actually, HLA-B27 can explain no more than 30% of the overall genetic susceptibility of AS [9]. And this result suggested the contribution of additional genes or genetic regions. Meanwhile, more and more investigations supported the presence of non-MHC genetic susceptibility in this disease. Twelve years ago, Brown et al. showed us that the likely genetic model of AS was an oligogenic model with multiplicative interaction between loci, according to the study of the sibling recurrent risk [7]. Subsequently, Gu et al. initially proposed an autosomal dominant inheritance mode in a multigenerational large pedigree study of AS in the Han ethnic Chinese population based on not only pedigree investigation and segregation analysis, but also parametric linkage analysis with genome scans and �ne mapping [10]. And for the �rst time, they reported that the incomplete penetrance of this Chinese Han pedigree was 0.54 [10]. However, it is a pity that the pathogenic genes involved have not been isolated until now.
No matter which kind of inheritance mode it is, the ultimate proof of these theories is to isolate the pathogenic genes and clarify the mechanisms.

Susceptibility Loci
Although HLA-B27 is almost essential for the inheritance of AS within families, epidemiological and genetic mode studies have suggested that there are other genetic risk factors involved. Brown et al. �rstly undertook a genomewide linkage screen in AS patients with highly polymorphic microsatellite markers and reported the susceptibility loci containing the MHC region and non-MHC region [11]. Subsequently, with genome-wide meta-analysis through aggregate analysis of four AS genome-wide scan studies, Huang and Gu managed to screen out several AS-linked loci, containing 6p22.3-p21.1, 16q, 3p, 10q, 2p, 17p, and 2q [12]. Later, by the study of the large Chinese AS families, Gu et al. validated one candidate non-MHC region for AS on 2q36.1-2q36.3 of family A and B [10]. Aerward, with the development of computational genetics and technology, especially the genome-wide association study (GWAS), some susceptibility loci such as MHC region were subsequently con�rmed, and new associations of AS at 2p15 and 21q22 have been identi�ed in a European population [13]. More recently, another GWAS in a population of Chinese descent replicated the previous associations at 2p15 and identi�ed two additional susceptibility loci at 5q14.3 and 12q12 [14]. e associated SNP rs10865331 (2p15), located 99 kb upstream of B3GNT2 and 182 kb downstream of TMEM17, was found in both European and Chinese populations. However, the mechanism and function of this region still need to be explored. e associated SNP rs4552569 at 5q14.3 is located between the hyaluronan and proteoglycan link protein 1 (HAPLN1) and EGF-like repeats and discoidin I-like domains 3(EDIL3) genes. It has been reported that HAPLN1 is associated with spinal osteophyte formation and disc degeneration in Japanese women [15], while EDIL3 regulates the bone formation by way of inhibiting the Wnt signaling [16]. e association at rs17095830 of chromosome 12q12 is within an intron of anoctamin 6 (ANO6), which encodes a multipass transmembrane protein belonging to the anoctamin family. ANO6 has the Ca-dependent phospholipid scramblase activity in various biological systems and may participate in the osteoclast formation in bone [17,18]. In the future, the validation of 5q14.3 and 12q12 in other populations and the function and mechanism investigations especially on cartilage development and bone formation should be carried on.

MHC and HLA-B27
e major histocompatibility complex (MHC) region is a large genomic region located on chromosome 6, which is associated with disease susceptibility in most autoimmune diseases. It covers 3.6 megabases and comprises 224 known genes. Because of high allelic diversity and extreme linkage disequilibrium of MHC, the studies of MHC genes and the mechanisms that affect diseases are difficult [19]. GWAS of AS in Europe by Reveille et al. showed �ve highly signi�cant associations across the MHC region (rs7743761, rs2596501, rs3915971, rs2516509, and rs1265112) [13]. Our GWAS analysis results replicated these SNPs except the rs7743761 in the Han ethnic Chinese population [14]. In addition, we reported another strong association at rs13202464, which was consistent with another GWAS analysis of AS in individuals of European ancestry [20].
In humans, the MHC is called the human leukocyte antigen (HLA) system, in which HLA-B27 has excessive association with AS. According to the data published in the international ImMunoGeneTics database (IMGT, release 2.24.0), 104 alleles exist, encoding 85 different proteins. All of the subtypes of HLA-B27 are ancestrally related to B * 2705. Several studies have compared the intensity of the association with B * 2704 and B * 2705, the predominant subtypes in the Han ethnic Chinese population, and found that B * 2704 was more strongly associated with AS than B * 2705 [21][22][23]. Similarly, it has been found that B * 2704 might be more strongly associated with AS than other subtypes present in Taiwanese Chinese [24]. For the Western European Caucasians, the main subtypes of B * 2702 and B * 2705 are equally associated with AS [22]. Other B27 subtypes are uncommon and mostly discovered in only single individuals or families. As a rare Asian subtype, B * 2715 was reported in single individuals and families [25]. It is deserved to be mentioned that B * 2715 was observed in AS group only and slightly more in juvenile AS group than in adult-onset AS group in our Chinese Han population study [26]. According to the above studies, it is obvious that the different B27 subtypes provide different strengths to AS, which is related to different ethnicity and geography. e different strengths associated with different B27 subtypes provide a useful method to validate the potential mechanisms of association with HLA-B27 in AS patients.
Interestingly, we have found that the association at rs13202464 remained genome-wide signi�cant when the risk effect of HLA-B27 was controlled [14]. at is, there are possible other risk variants beyond HLA-B27 within the MHC region that are expressed by the rs13202464 [14]. Further studies are badly needed to discover the other possible MHC regions beyond B27 and investigate the mechanisms within different subtypes of B27 and the additional risk variants beyond B27.

ERAP1
In 2007, the association of ERAP1 (the gene for endoplasmic reticulum aminopeptidase 1 (ERAP-1)) with AS was �rst identi�ed by the Wellcome Trust Case-Control Consortium and Australo-Anglo-American Spondyloarthritis Consortium (WTCCC/TASC) in European Caucasian population [27]. And then, the association with AS has been found in other Caucasian populations [28,29] as well as in Chinese population [30]. Our studies demonstrated two novel SNPs (rs27434 and rs27529) in ERAP1 other than that reported previously [14]. e crystal structures in open and closed states of human ERAP1 were �rstly reported by �ochan et al. [31]. ERAP1 is a zinc metallopeptidase with typical H-E-X-X-H-(X)(18)-E zinc binding and G-A-M-E-N motifs characteristic for members of the gluzincin protease family [31].
ERAP1 has three known physiological functions. First, it is involved in trimming peptides to optimal length for perfect binding and presentation by MHC class I molecules within the endoplasmic reticulum [32]. It is known to us that HLA-B27, the most important molecules in AS, belongs to MHC Class I molecules. Urgent study has investigated that the polymorphisms of ERAP1 only affect AS risk in HLA-B27 positive individuals [20]. e crystal structure study of a �528R mutant associated with AS shows signi�cantly altered peptide processing characteristics [31]. It is proved that ERAP1 represents functionally signi�cant interaction with speci�c subtypes of HLA-B27, containing B * 2704 and B * 2705 [33]. It is supposed that the mechanism explaining the association of HLA-B27 with AS may be related to the aberrant processing of antigenic peptides presentation associated with ERAP1. e interaction between HLA-B27 and ERAP1 should be informative suggesting the potential mechanisms. Perhaps the genetic model of AS is digenic for some individuals or families. Second, ERAP1 can promote the shedding of the proin�ammatory cytokine receptors such as IL-1RII, type I TNF receptor, and the IL-6R in vitro [34]. Due to the cleavage, the in�ammatory cytokines are downregulated. However, it is observed that ERAP1 does not have a major in�uence on cytokine receptor trimming in mice [20]. Whether the cleavage plays a role in the AS patient is still unknown. ird, ERAP1 is involved in the activation of macrophages induced by lipopolysaccharide (LPS) and interferon (IFN)- [35]. It is reported that the bacteria may be involved in the development of AS, but the mechanisms are not clari�ed. Researches focused on the relationship between the LPS of bacteria and ERAP1 may be informative for the mechanism investigation.
Another report in the Spanish population showed that ERAP1 gene not only is associated with genetic predisposition to AS, but also in�uences the functional severity of the disease in a Spanish population [36]. e association of ERAP1 with AS in Chinese population was replicated by many researchers; however, the relation with disease severity and the potential B27-linked and -unlinked mechanisms still need to be explored.

IL-23R
Polymorphisms in IL-23R (interleukin-23 receptor) have been recently shown to have a signi�cant association with AS in European Caucasian population [27]. And the most strongly associated IL-23R SNPs in recent genome-wide association study were rs11209026 and rs 11465817, respectively, falling in exon9 and intron 9 of the gene [13]. Nevertheless, there was a distinguished inconsistency among the results in the Chinese Han population. Some recent studies have shown that the SNPs of IL-23R are susceptible to AS in Chinese Han population [37], while other studies have shown con�icting results [29,38]. However, the genetic �ndings just provide us with a novel insight into the etiology of AS. In the future, the broader validation in different populations and speci�c mechanisms exploration should be carried on.
IL-23R belongs to the hemopoietin receptor family as a subunit of the receptor for IL-23 and is involved in the production and differentiation of T helper 17(17) cells. 17 cells are identi�ed as a subset of the CD4� T lymphocytes secreting high levels of the proin�ammatory cytokine IL-17 upon stimulation [39]. In addition, related studies of Chinese population reported that the IL-23 level in the supernatants of cultured peripheral blood mononuclear cells (PBMCs) and the expression of IL-23p19 mRNA in PBMCs in patients with AS were signi�cantly higher than that in healthy control group [40]. ese investigations indirectly revealed that the polymorphisms of IL-23R were associated with AS. Interestingly, the variation of IL-23R has also been demonstrated in in�ammatory bowel disease and psoriasis [41,42], providing more or less part of the explanation as the accepted close connection between these diseases. According to all of these �ndings, the treatments targeting IL-23 may prove effective and need to be explored.

Other Non-MHC Regions
Except all the non-MHC regions mentioned before, there are other regions such as anthrax toxin receptor 2 (ANTXR2), runt-related transcription factor 3 (RUNX3), interleukin-12B (IL12B), and tumor necrosis factor receptor superfamily, member 1A (TNFRSF1A) proved to be associated with AS de�nitively in European population [13]. e association between AS and TNFRSF1A in the Han Chinese population was con�rmed by Davidson et al. several months ago [43]. At the same time, the association of IL-12B genetic polymorphism with the susceptibility of AS was proved in the Taiwanese Chinese population [44]. However, the reports about the association of RUNX3 with AS in Han Chinese population have not been found yet. Besides, based on the speci�c genetic loci studies, many genes such as Fc receptorlike molecule 4 (FCRL4) [45] and Fc gamma receptor IIB (FCGR2B) [46] in Han ethnic Chinese population and matrix metallopeptidase 3 (MMP-3) [47], TIMP metallopeptidase inhibitor 1 (TIMP-1) [47], and ORAI calcium releaseactivated calcium modulator 1 (ORAI1) [48] in Taiwanese Chinese population were reported to be associated with AS development.

Conclusions
e development of genomic technologies such as GWAS and candidate gene studies has helped to understand the involvement of some genes in the etiology of AS. ese �ndings strongly proved the presence of non-MHC genes involved in AS which was consistent with the prior results. However, a report showed that the familial aggregation was related to HLA-B27 other than the non-MHC susceptibility loci described recently [49]. e reason for the phenomenon needs to be explored in the future. If this can be validated in many populations, the results may be helpful for understanding the different pathogenesis of these genes. e researches concerning the genetics of AS are still in early stage. With the larger sample sizes and more new technology, more and more genes associated with AS will be identi�ed. ese new �ndings provide a useful platform for hypothesis-driven research into AS pathogenesis and make it possible to develop new treatments which can inhibit damage to bone structure and radiographic progression in AS.