Analysis of Gene Expression of Human Leukocyte Antigen B27 in Patients with Ankylosing Spondylitis and Correlation with Related Indicators

The aim of this study was to investigate the potential correlation between the human leukocyte antigen (HLA)-B*27 subtypes in patients with ankylosing spondylitis (AS) and the clinical features of AS. The prevalence of HLA-B*27 subtypes was investigated in 132 healthy donors recruited from our center and in 143 patients with AS. We investigated the differences in HLA-B27 subtype status for the patients and healthy donors by using a PCR-SSP (polymerase chain reaction with sequence-specic primer) method. The clinical outcomes of the patients, including lower back pain, uveitis, peripheral arthritis, stiffness, joints, erythrocyte sedimentation rate (ESR), Anti-streptolysin O (ASO) and C-reactive protein (CRP), were recorded. The male-to-female ratio was 2.5, and the mean age at diagnosis was 29.3 years. HLA-B*27 positivity was detected in 111 patients (77.6%) of patients with AS. The rate of HLA-B*27 positivity was signicantly higher in the AS group than patients without AS. The subtypes observed in patients with AS were HLA-B*2704 (55.9%, 62/111), HLA-B*2705 (39.6%, 44/111), HLA-B*2702 (1.80%, 2/111), HLA-B*2707 (0.90%, 1/111), and HLA-B*2704/05(1.80%, 2/111). The main genotypes were HLA-B*2704 and HLA-B*2705. There were no signicant differences in clinical manifestations. Multivariate analysis showed statistically signicant differences in sex (P=0.01), disease duration (P=0.023), hematocrit (P=0.01), and CRP (P=0.01) between patients in the HLA-B*27(+) AS group and the HLA-B*27(-) AS group. In addition, HLA-B*15, HLA-B*40, HLA-B*13, and HLA-B*46 were the major alleles with associated HLA-B types in patients with HLA-B*27(+) AS. Therefore, we conclude that HLA-B*27 is highly correlated with AS and can be used as an early predictor of AS. In addition, sex, disease duration, ESR, and CRP were signicantly different in the HLA-B*27(+) AS group and HLA-B*27(-) AS group. Finally, our study also found a correlation between HLA-B*15, HLA-B*40, HLA-B*13, and HLA-B*46 subtypes and the development of AS.


Introduction
Ankylosing spondylitis (AS) is a chronic progressive autoimmune disease of the spine and sacroiliac joints and is a common orthopedic disease; the incidence is 0.2-0.3% in China. The main clinical signs of the disease are sacroiliac arthritis and spondylitis, which lead to ankylosis and the eventual loss of spinal mobility. Peripheral arthritis, lower back pain, stiffness, and sacroiliac arthritis are also major symptoms in patients with AS. The disease can also be associated with extra-skeletal manifestations such as anterior uveitis and psoriasis. Therefore, early detection and timely treatment are particularly critical [1] .
Various hypotheses have been reported to explain the relationship between human leukocyte antigen B27 (HLA-B*27) and AS. The arthritic peptide hypothesis states that T-cell receptors can recognize complexes of foreign or unknown MHC autologous peptides and when bound together [2] . The molecular mimicry hypothesis suggests a cross-reactivity between antigens of the associated bacteria and the HLA-B27 molecule [3,4] , whereas a third hypothesis is based on the aberrant folding of HLA-B27, which induces the disorder of intracellular tra c of this molecule [5] . However, none of these hypotheses has fully elucidated the pathogenicity mechanism of HLA-B27 in AS. The etiology and pathogenesis of AS are still not well understood, and it is generally believed that it they are closely related to genetics and infection.
Human leukocyte antigen B27 (HLA-B*27) is the gene that is reported to be most closely associated with AS, with over 155 subtypes identi ed to date [6] . However, the prevalence of HLA-B*27 varies across racial and ethnic populations, possibly as a result of the different genetic interactions and geographic origins [7][8][9][10] . In general, HLA-B*2705 is the most common subtype in all populations. The subtype HLA-B*2702 is present in Asia, North Africa, Europe, and certain Mediterranean countries. It is reported to be the most common subtype in Caucasian and Arab populations [4] . HLA-B*2704 is the most common subtype in Chinese and Japanese populations [11] . It has been shown that patients with AS carrying the HLA-B*2704 gene are more likely to develop uveitis than patients with AS carrying the HLA-B*2705 gene. Recently, it has been found that some alleles at the B locus, other than HLA-B*27, are also closely associated with AS; for example, HLA-B*40 and HLA-B*51 increase the risk of developing AS, whereas HLA-B*07 has been reported to protect against AS.
Anti-streptolysin O (ASO) is an antibody produced by streptococcal infection and is associated with immune diseases, such as rheumatism and nephritis, and leukocytes, C-reactive protein (CRP), and sedimentation are all indicators of in ammation. In this study, the above indicators were tested to nd associations with AS and to achieve early diagnosis and early treatment of AS.
In this study, we will investigate the correlation between HLA-B subtypes and the incidence and clinical symptoms of AS, as well as the differences in clinical symptoms between patients with AS carrying the HLA-B*2704 and those carrying the HLA-B*2705 gene, and to explore the correlation between AS and B locus alleles other than HLA-B*27. In addition, the relationships between AS and indicators such as ASO, leukocytes, CRP, and hematocrit were also explored through multifactorial and univariate analyses.

Materials And Methods
This study was approved by the ethics review board of the Tianjin First Central Hospital with Nankai Medical University. Written informed consent was provided by each enrolled patient. The study was performed in accordance with the Declaration of Helsinki and its recommendations.

Selection of subjects
This study was conducted in the division of orthopedics and immunology department at a government hospital. Between January 2020 and October 2020, 143 patients with AS were randomly selected were recruited for the purpose of HLA typing; another 132 patients without AS were selected over the same consultation period.
Medical charts were reviewed and the data of demography, lower back pain, uveitis, peripheral arthritis, and laboratory indicators, including erythrocyte sedimentation rate (ESR), anti-streptolysin O (ASO), and CRP, were collected. All data were obtained at diagnosis.
The inclusion criteria for patients with AS were: 1) history of pain in the lower back and lumbar region, morning stiffness for more than 3 months, symptom relief after activity, no improvement with rest; 2) manifestation of restricted movement in the lateral exion direction and anterior-posterior direction of the lumbar spine; 3) thoracic mobility lower than normal people of the same age and sex; 4) imaging showing bilateral grade II or higher sacroiliac arthritis or unilateral grade III or higher sacroiliac arthritis.
Those who met the above imaging manifestations in 4 and any of the clinical manifestations of criteria 1-3 were enrolled.
Non-AS inclusion criteria: simple bone and joint disease, no abnormal liver or kidney function, abnormal hematopoietic function, no neurological disease, malignancy, and other immune system diseases.

Detection of HLA-B*27 subtypes
The HLA-B27 genotyping assay was performed by a PCR-SSP method. For DNA extraction (QIAGEN Nucleic acid extraction kit; HLA-B27 genotyping kit (Tianjin Xiu peng Biological), 200 µL of EDTAanticoagulated blood was mixed thoroughly with 100 µL of template DNA in accordance with the operating instructions, at a concentration of 20-100 ng/μL, a purity (A260/A280 value) of 1.6-2.0, and stored at -20℃. For PCR ampli cation, the DNA template and working solution was mixed thoroughly and added to each well. The ampli cation was conducted in accordance with the operating instructions, and a 2% agarose gel was loaded with 5 µL of the ampli cation product and 2 µL of loading buffer. Sample electrophoresis was performed at 140 V for 10 min, and the gel was transferred to an imaging system to observe the results of electrophoresis.
CRP and ASO assays were performed by transmission turbidimetry with direct serum aspiration and sample addition, and analyzed by an automatic immunoassay analyzer. The ESR assay was performed using the sedimentation method with sodium citrate anticoagulant.

Statistical analysis
All statistical analyses were performed using Statistical Package for the Social Sciences (SPSS, Chicago, Illinois, USA; version 20.0). The results of the descriptive analysis were presented as the mean ± standard deviation if a normal distribution was present and, if not, as the median and interquartile range. For pairwise comparisons, an independent t-test was used for normally distributed numerical variables, and chi-square test or Fisher's exact test was used, when appropriate, for categorical variables. Two-sided p values of less than 0.05 were considered statistically signi cant.

General clinical features of patients with AS and healthy controls in the Chinese population
The mean age of the 143 patients was 38.5±14.5 years, the mean age at disease diagnosis was 29.3±10.2 years, with a mean disease duration of 9.3±8.0 years. The male-to-female ratio was 2.49, with 102 (77.6%) male patients. With respect to the frequency of symptoms, 56.6% had in ammatory back pain, 71.3% had stiffness, 62.9% had sacroiliac joints, 7.0% had uveitis, and 18.    (Table 4).

Discussion
Ankylosing spondylitis (AS) is a chronic progressive immune disorder with the involvement of the medial joints as the main manifestation; it occurs mostly in young and middle-aged men. The disease has a slow course with no obvious symptoms in the early stages, and when patients develop clinical manifestations or histopathology, the disease has often reached an intermediate to advanced stage, with the presence of irreversible damage owing to the formation of connective tissue and spinal ankylosis, with varying degrees of dysfunction, which can lead to disability and affecting the quality of life of patients [12,13] . Therefore, early the detection and early treatment of AS are particularly critical. In recent years, although AS has been well studied, the pathogenesis and principles of the disease are still less clear, although it is generally believed that it is closely related to genetics and infection. In this study, 71.3% (102/143) of the patients were men and 28.7% (41/143) were women, with a male-to-female ratio of 2.49, which was consistent with previous reports in the literature [14] . Some studies have shown that there is a stronger association with family genetic history for patients with a lower age of onset [15,16] . An association of early disease onset with a high familial history of HLA-B27-positive patients have also been reported [17] . This pattern was not found in this study, but in terms of patient age and age at rst diagnosis, the AS-positive group was younger than the AS-negative group, but there was no statistically signi cant difference between them. In terms of family history, the genetic incidence was 21.6% (24/111) in the AS-positive group compared with 18.8% (6/32) in the AS0-negative group, which was not a statistically signi cant difference (p=0.725). However, the disease duration was 8.8 ± 7.9 years in the ASpositive group and 10.8 ± 8.5 years in the AS-negative group, which con rmed to be a statistically signi cant difference through the multifactorial analysis (P=0.023).
HLA-B*27 is an allele at the B locus of the MHC-1 class of molecules on the short arm of human chromosome 6 and consists of an alpha chain with a molecular weight of 4.4 × 10 4 Da and a beta chain with a molecular weight of 1.2 × 10 4 Da [18] . The α chain includes α1, α2, and α3 chains, and the β chain is bound by β2 macroglobulin; together, they form the HLA-B*27 molecule. In 1973, British scientists rst reported that AS was closely associated with the HLA-B*27 gene, with signi cant familial aggregation. Although research into AS has progressed in recent years, HLA-B*27 remains the most strongly associated gene with AS identi ed to date. One study showed that about approximately 90% of patients in AS carry the HLA-B*27 antigen, whereas only 4-9% of the healthy population are positive for HLA-B*27 [19] . In this study, the positive percentage of B27 in the AS group was 77.6%, which is low and might relate to the inclusion criteria; some early-stage patients positive for HLA-B*27 but without symptoms on imaging were not included, whereas a total of 111 of 133 patients positive for HLA-B*27 were diagnosed with AS (83.5%), indicating that HLA-B*27 was very useful for the diagnosis of AS, especially in early diagnosis.
The current methods for detecting HLA-B*27 are micro lymphocytotoxic assay, ow cytometry, and PCR-SSP gene assay. The micro lymphocytotoxic assay is a classic method for the detection of HLA-B*27, but it is in uenced by cell purity, complement difference, monoclonal antibody potency, and operator experience, which can lead to false negatives. In recent years, the PCR-SSP method has been applied to the detection of HLA-B*27. This is because under speci c reaction conditions, the rst base at the 3'-end of the primer is complementary to the allele-speci c base, and the speci c primer ampli es only the allele that matches it under strict experimental operations and in molecular biology laboratories, to avoid the The difference in the polymorphisms of HLA-B*27 isoforms in the development of HLA-B*27-related diseases leading to the associated symptoms shows their different involvement in the development of the disease [21,22] . Uveitis is considered to be the most common extra-articular manifestation of AS, with a distribution of 5.3-33.2% in different races [14] , and some studies showed that HLA-B*2705-positive patients were more closely associated with uveitis than HLAB*2704 [23,24] . Uveitis accounted for 9.0% (10/111) of patients in this study, but the positivity rate was not statistically signi cant for either (p=0.732). Peripheral arthritis was found in 35 cases in this study, accounting for 31.5%, which was consistent with previous studies [25] . There were 82 cases (73.9%) of sacroiliac arthritis and 65 cases (58.6%) of lower back pain. These clinical manifestations were not statistically signi cantly different from the various subtypes of B27, which was consistent with some studies [26] . There were two cases of pure B27:04/05 in this study, which were not counted owing to the small number of cases, but some scholars found no signi cant difference in clinical presentation and imaging progression between pure and heterozygous HLA-B27 [27,28] .
Environmental factors are also important in uences in the development of AS; among which the in uence of infection cannot be ignored [29] . Some studies have proposed a close relationship between streptococcal infection and the development of arthritis. Anti-streptolysin O (anti-streptolysin O ASO) is an antibody produced by streptococcal infection, a positive ASO indicates that the body has streptococcal infection, and the antibody potency reaches a 3-6-week peak [30] . Streptococcal infection stimulates the body's immune response, induces the production of multiple antibodies, and activates T and B lymphocytes, promoting the release of various in ammatory factors, mediating the aggravation of damage to joints, and therefore participating in the development of AS. CRP is an acute temporal reactive protein, a non-speci c indicator of in ammation, and hematocrit (ESR) is also an indicator of in ammation. In the present study, a univariate analysis showed statistically signi cant differences in CRP (P=0.001) and ASO (P=0.027), but no statistically signi cant ESR (P=0.188) in both groups, whereas a multifactorial analysis showed statistically signi cant ESR (P=0.01) and CRP (P=0.01), but no statistically signi cant ASO (P=0.16), indicating that these in ammatory indicators need to be analyzed together to aid the diagnosis of AS.
Recent studies on the B locus allele of HLA-B*27 in patients with AS have emerged; the association of HLA-B*40 with AS has now been convincingly demonstrated now in East Asian studies [31,32] . Other studies have found that 18.2% of patients with AS carry HLA-B*227/B40 and that HLA-B*240 increases the susceptibility of HLA-B*27 to AS [33,34] . In this study, only 105 patients positive for HLA-B*2704 and HLA-B*2705 were tested. In total, 24 cases of HLA-B*40 (22.9% 24/105) were detected, including nine cases of HLA-B*4001, ten cases of HLA-B*4002, and ve cases of HLA-B*4006. The frequency was generally consistent with previous studies. A total of 21 cases of HLA-B*15 (20.0%, 21/105) was detected, including 9 cases of HLA-B*1501,9 cases of HLA-B*1502 and 3 cases of HLA-B*1511. Some studies have shown that B15 positivity increases the risk of peripheral joint involvement [33] , and HLA-B*07 and HLA-B*51 alleles have been shown to have a protective in uence on the risk of developing AS in several Mediterranean populations [35,36] . The number of B7 and B51 cases in this study was low, with only one HLA-B*7 and four HLA-B*51 cases. Other B loci found in higher numbers in this study were HLA-B*13 (17.1% 18/105), HLA-B*46 (11.4% 12/105), and HLA-B*58 (8.6% 9/105). In terms of diagnostic value, further research is needed regarding the signi cance of heterozygosity for HLA-B loci owing to the small number of patients.
Our study has some limitations. First, as it is a retrospective study, some clinical manifestations may have been missed owing to the speci c patients; secondly, the number of cases in both groups was small and there was some bias in the results; and nally, owing to the reagents, only B27-positive patients were tested for the B locus allele, whereas the control group was only B27-negative and no B locus testing was performed, preventing some corroboration of the results.
In conclusion, our study showed that in addition to the HLA-B*27 and HLA-B*27 subtypes, the allelic typing of HLA-B*27, and patients family genetic background was closely associated with AS. This paper may be somewhat different from the real-word situation owing to specimen size and patient enrollment. Moreover, some results are not su ciently clear and require further study.

Declarations
Funding: This work was supported by no speci c funding. The fund-ing sources had no role in the study design, data collection, analysis or interpretation, or the writing of this manuscript.
Con ict of interest The authors declare that they have no competing interests.
Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent: Informed consent was obtained from all individual participants included in the study.