Integrin Subunit Alpha M, ITGAM Nonsynonymous SNP Is Associated with Knee Osteoarthritis among Thais: A Case-Control Study

Knee osteoarthritis (OA), which is one of the most common degenerative joint diseases, presents a multifactorial etiology, involving multiple causative factors including genetic and environmental determinants. Four human neutrophil antigen (HNA) systems can be determined using each HNA allele by single-nucleotide polymorphisms (SNPs). However, there are no data on HNA polymorphisms and knee OA in Thailand, so we investigated the association of HNA SNPs and knee OA in the Thai population. In a case-control study, detection of HNA-1, -3, -4, and -5 alleles by polymerase chain reaction with sequence-specific priming (PCR-SSP) was performed in participants with and without symptomatic knee OA. Logistic regression models were used to estimate the odds ratio (OR) and 95% confidence interval (CI) between cases and controls. Among 200 participants, 117 (58.5%) had knee OA; 83 (41.5%) did not and were included as controls in this study. An integrin subunit alpha M (ITGAM) nonsynonymous SNP, rs1143679, was markedly associated with symptomatic knee OA. The ITGAM*01*01 genotype was identified as an important increased risk factor for knee OA (adjusted OR = 5.645, 95% CI = 1.799–17.711, p = 0.003). These findings may contribute to our understanding of the application prospects for therapeutic approaches to knee OA.


Introduction
Osteoarthritis (OA) is the most common degenerative joint disease and can affect both cartilage and subchondral bones. OA in joints is manifested by progressive articular cartilage destruction, subchondral bone thickening, osteophyte formation, synovial inflammation, ligament and meniscus degeneration, and capsular hypertrophy [1][2][3]. The areas most frequently affected are the knees, hips, feet, ankles, distal and proximal interphalangeal joints, first carpometacarpal joint, and lower spine [4]. The main clinical presentations are pain and stiffness of the joints [5]. Pain tolerance is unique to each person and decreases with age [6], resulting in a delayed OA diagnosis among athletes and young people [5]. The diagnostic process begins with an interview and physical examination and includes a combination of radiologic imaging and laboratory testing. The main purposes of OA treatment are to relieve pain and disability and restore functionality; however, joint replacement surgery is considered for patients with severe pain and disability from large-joint OA [5].
Knee OA is a common form of arthritis among the elderly [7]. The prevalence of knee OA varies depending on the defined symptomatic or radiographic findings and characteristics of the study population, such as age and sex. There is a 10% prevalence of symptomatic knee OA in men and 18% in women among adults aged > 60 years [8].

Study Design, Participants, and Periods
This case-control study involved individuals residing in Pathumtani Province, Thailand, and was conducted from May 2021 to March 2022. A total of 200 recruited participants met the following inclusion criteria: age 45 years or older, able to walk independently, and possessing the linguistic and cognitive ability to understand and fill out the informed consent. Participants were excluded if they were diagnosed with a neurologic condition or experienced traumatic events secondary to surgery that would affect lower limb function. Cases and controls were grouped by an orthopedic surgeon based on a review of all clinical findings (pain, morning stiffness, crepitus on motion, bony tenderness, and enlargement and palpable warmth of synovium) and radiological findings of both knees. Standard anteroposterior (AP) and lateral weight-bearing knee radiographs were performed with participants standing on both legs and were confirmed by diagnostic radiology with grades according to the Kellgren and Lawrence scale (KL; grade 0 to 4) [23]. Knee OA was diagnosed based on the presence of either clinical symptoms and physical findings along with radiographic findings, or only clinical symptoms and physical findings, while the control group had no existing medical diagnosis of knee OA. The complete workflow of the study is provided in Figure 1.

HNA Allele Detection by Polymerase Chain Reaction with Sequence-Specific Primer (PCR-SSP)
The individual genotyping of HNA-1, -3, -4, and -5 by PCR-SSP, sequences of the primer combinations used in primer mixtures, detected alleles, SNPs, and the product size of each mixture are shown in Table 2. PCR conditions were similar to those in a related report with some modifications [19]. For each PCR reaction, the mixture consisted of 5 µL of 2× PCR reaction mixture (GoTaq ® Hot Start Colorless Master Mix, Promega, Madison, WI, USA), 1 µL (50-100 ng/µL) of genomic DNA, 1 µL of 5 µmol/L HNA-specific sense primer, 1 µL of 5 µmol/L HNA-specific antisense primer, and 2 µL of 2.5 µmol/L HGH primers (sense and antisense), running as internal control at a final volume of 10 µL. PCR was performed in a T100 thermal cycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA). PCR conditions were as follows: initial denaturation at 95 • C for 10 min; 10 cycles of denaturation at 95 • C for 30 s, annealing at 64 • C for 40 s, and extension at 72 • C for 45 s; 20 cycles of denaturation at 95 • C for 30 s, annealing at 61 • C for 30 s, and extension at 72 • C for 45 s; and final extension at 72 • C for 5 min. PCR products were separated on a 1.5% agarose gel containing SYBR Safe DNA Gel Stain (Invitrogen, Paisley, UK), electrophoresed in 1× Tris borate ethylenediamine-tetraacetate (TBE) buffer at 100 volts, and visualized under a blue light transilluminator. Regarding the FCGR3B deficiency and FCGR3B*null confirmation by PCR-SSP, the primers and PCR conditions were identical to those in a related report [24].

Statistical Analysis
Statistical analysis was performed using Microsoft Excel and SPSS version 25.0 (SPSS Inc., Chicago, IL, USA). Normally distributed continuous variables were described in terms of the mean (standard deviation, SD), and non-normal variables as the median (interquartile range, IQR). Categorical variables were summarized and expressed as absolute frequencies and percentages. As the outcome variable (knee OA) was binary, the t-test and Mann-Whitney U test were used to compare continuous variables for parametric and nonparametric tests. The multicollinearity of all explanatory variables was checked before multivariable logistic regression analysis to exclude highly interdependent variables using the variance inflation factor (VIF). Multivariable logistic regression was applied to all of the data, including continuous and categorical explanatory variables. The cut-off for the continuous predictor variable was derived by using the ROC curve for the age and BMI variables. Each HNA-1, -3, -4, and -5 genotype variable was dummy-coded into two variables representing each genotype (1 = the person possesses it, 0 = otherwise). In addition, associations were estimated by calculating the odds ratio (OR) with a 95% confidence interval (CI). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were also calculated. Differences between KL grades were assessed using Pearson's chi-square (χ 2 ) test. A p-value less than 0.05 (two-tailed) was considered statistically significant.

Description of Participants
In the 11-month recruitment period, among 200 participants we identified 117 (58.5%) as having knee OA and the remaining 83 (41.5%) as controls. Tabulated data on cases and controls by sex, age, weight, height, body mass index (BMI), and blood type are presented in Table 3. Of the 117 cases, 89 (76.1%) were women and 28 (23.9%) were men, with a median age of 62 years (IQR 53.5, 67.5). The vast majority of cases in this age group were in the 61-to-70-year-old group, comprising 39.4% of the population older than 45 years. Consequently, the proportion of women and the median age in the case group were significantly higher than those in the control group (p < 0.05). Median height in the case group was 155 cm (IQR 150, 162), which was significantly shorter compared to the control group (p = 0.001). On the other hand, no differences were found between the two groups with respect to weight, BMI, and blood type.
When predictor variables were evaluated for collinearity, weight, height, and BMI were found to be highly correlated (VIF = 160.12, 50.23, and 111.70, respectively). Due to the significant collinearity of these variables, only BMI was retained for subsequent data analysis due to its clinical relevance in previous studies of knee OA [8]. Multivariable logistic regression was performed to derive adjusted odds of each significant variable derived from univariable analysis. The variables of sex, age, HNA-4 genotype, and BMI of the participants were retained in the model. Additionally, HNA-1, -3, and-5 genotypes were included based on clinical adjustment and literature review. The cut-off for age was 56.5 years, with 68.3% sensitivity and 53.0% specificity; thus, an age of 57 years was chosen as the rounded cut-off for the study. Additionally, the cut-off for BMI of 25.0, with a strength of 56.63% sensitivity and 54.70% specificity, was hence chosen as the cut-off for the further analysis. We performed a series of models (see Supplemental Tables S1-S15) with different genotypes and selected the final model, as shown in Table 5. The results of multivariable analysis show that three variables had a significant relationship to knee OA (p-value < 0.05). The most dominant factor based on the largest adjusted OR value was ITGAM*01*01 (5.645), and the smallest was age ≥ 57 years (2.485). Women have a 3.727 times greater chance of developing knee OA than men, after controlling for other variables (   ; alleles of sample 001, lanes 1A to 5B: HNA-1b/1b, HNA-3a/3b, HNA-4a/4b, and HNA-5a/5b; sample 002, lanes 1A to 5B: HNA-1a/1a, HNA-3a/3b, HNA-4a/4b, and HNA-5a/5b.

Frequency of Kellgren and Lawrence (KL) Grades for Pairs of Knees
All participants underwent radiographic examination of the bilateral knees using AP and lateral views with weight-bearing. Figure 3 shows sample images of knee joints classified as KL grade 0 to 3 in this study. All 83 controls were classified as bilateral KL grade 0. Among the 117 cases of symptomatic knee OA, 115 were diagnosed with clinical and radiographic changes, and two were diagnosed as early knee OA with clinical relevance only concerning both knees (bilateral KL grade 0). Table 6 shows the frequencies of KL grades for the left and right knees in the case group. Based on the most observed radiographic changes, 58 OA knees (49.6%, 95% CI 40.2-58.9) were classified as bilateral KL grade 1. Among the others, 23 cases (19.7%, 95% CI 12.9-28.0) were classified as bilateral KL grade 2, 13 cases (11.1%, 95%CI 6.1-18.3) as unilateral KL grade 1, six cases (5.1%, 95%CI 1.9-10.8) as unilateral KL grade 2, and five cases (4.3%, 95%CI 1.4-9.7) as bilateral KL grade 3. However, unilateral or bilateral KL grade 4 was not found in this study.

Discussion and Conclusions
The data from this study demonstrate that knee OA symptoms, including pain, stiffness, crepitus, tenderness, and enlargement and palpable warmth of the joint, were experienced by 117 participants (58.5%), whereas only 39 participants (19.5%) had knee OA (KL grade ≥ 2) observed radiographically on X-rays. Given the discrepancy between symptomatology and radiographic findings, individuals who are diagnosed with early knee OA can exhibit the symptoms without radiographic changes. No cases of X-ray findings without any symptoms were noted in this study. However, severe symptoms of knee  The distribution of cases in each KL grade and for sex, age, and ITGAM genotype is summarized in Table 7. Individuals were classified by their most severe KL grade for analysis. The differences in sex, age, and ITGAM genotype between groups were not significant (p > 0.05). KL grades for women, those 57 years old or above, and those with ITGAM*01*01 genotype did not tend to be worse than the grades for men, those < 57 years old, and those with the ITGAM*01*02 genotype.

Discussion and Conclusions
The data from this study demonstrate that knee OA symptoms, including pain, stiffness, crepitus, tenderness, and enlargement and palpable warmth of the joint, were experienced by 117 participants (58.5%), whereas only 39 participants (19.5%) had knee OA (KL grade ≥ 2) observed radiographically on X-rays. Given the discrepancy between symptomatology and radiographic findings, individuals who are diagnosed with early knee OA can exhibit the symptoms without radiographic changes. No cases of X-ray findings without any symptoms were noted in this study. However, severe symptoms of knee pain have been related to advanced radiographic findings of structural pathology in different ethnic populations [25]. In addition, being female and age 57 years or above were strongly associated with the diagnosis of knee OA among these Thai participants. Women were more likely to develop higher severity than men [26]; it can be hypothesized that women may be particularly sensitive due to hormonal changes occurring across the menopausal transition [27]. The related data support the observation that women who supplemented with estrogen were 15% less likely to need total knee replacement than those who were not taking the hormone [27]. Age has been shown to be one of the strongest risk factors for OA of all joints [7,8], similar to the result of our study in which the median age of participants with knee OA was significantly higher compared to the control group. The effects of exposure to various risks factors and biologic changes of the elderly may cause more severe joint damage than would be experienced by younger adults. However, many recent studies have demonstrated that being overweight or obese, as screened by BMI, had been found to be the greatest potent risk factor for developing knee OA [28][29][30]. Such modifiable risk factors can be controlled, and their effect can be reduced by making behavioral changes.
Knee OA has a multifactorial etiology and can be considered to involve a relationship between modifiable and non-modifiable factors. Not only sex and age, but also genetic factors are among the non-modifiable risk factors that influenced knee OA. GWAS studies have revealed the correlated SNPs mentioned above, which showed a statistically significant association with knee OA in different populations [11][12][13][14][15][16], but the impact of HNA SNPs associated with this disease remains unknown. It was found that peripheral macrophages and neutrophils were found in synovium and synovium fluid, clearly demonstrating the involvement of neutrophils in the sterile inflammatory process and progression of knee OA. An α4β7 integrin plays an important role in macrophage infiltration in OA synovium [31]. However, the mechanism of neutrophil migration into OA synovium remains unclear. In the current study, the variant rs1143679 was the new locus identified as being associated with knee OA among Thai participants. We observed a directional concordance of risk between the ITGAM*01 (HNA-4a) allele and knee OA, while the antithetical ITGAM*02 (HNA-4b) allele was shown to be protective. This intergenic SNP is encoded by the ITGAM gene on chromosome 16p11.2 and is expressed on the CD11b/α M subunit of the α M β 2integrin (CD11b/CD18, Mac-1, CR3). In terms of functionality, the β 2 integrin is rapidly activated in leukocytes and enables adhesion of cells to counter-receptors, transmigration, phagocytosis, and oxidative burst [32,33]. HNA-4a and HNA-4b alleles differ in an ITGAM*230G>A, leading to an Arg61His substitution in the mature protein [17]. The mechanism by which changes in the polymorphism of those amino acids may lead to increased or reduced risk of knee OA remains unclear at present. It could be speculated that HNA-4 might play a role in neutrophil migration into the OA synovium and mediate OA pathogenesis. This notion warrants further experimental validation. However, strong evidence indicates the important role of integrin dysfunction in affecting OA cartilage, subchondral bone and synovium [34]. Generally, an imbalance in the anabolic and catabolic activity of the chondrocytes, especially in articular cartilage, can cause changes in the components of the extracellular matrix (ECM) in OA, as progressive factors. The integrin-mediated signaling pathway plays a key role in catabolic activity for joint destruction [35]. Likewise, excessive mechanical load signals from the ECM can trigger integrins, which work together to encourage the progression of cartilage matrix destruction in OA [36]. The interaction between integrins and cytokines secreted in the ECM, such as insulin-like growth factor and transforming growth factor beta, can also promote OA progression [37]. Changes in integrins, as mentioned above, are a key source of pathologic changes critical for the cartilage, subchondral bone, and synovium in OA [34].
Of the ITGAM genotypes observed in this study, the ITGAM*01*01 genotype showed the strongest correlation in knee OA cases, with a calculated OR up to 5.645 compared to the controls, as mentioned above. The ITGAM*01*02 genotype has been shown to have a statistically reduced risk of symptomatic knee OA development. The five cases were all women (mean age: 65.8 years) who received a diagnosis of pre-/early knee OA with bilateral KL grade 1. It has been acknowledged that knee OA is a multifactorial degenerative disease, so potentially the non-modifiable risk factors of sex and age could affect the diagnosis [8]. Having the ITGAM*01*02 genotype could also mean a reduced risk for knee OA and delayed disease progression. However, the occurrence of the ITGAM*02*02 genotype is very rare in Thailand and other populations [17,18,38], thus the uninformative nature of this genotyping result cannot confirm or indicate whether an individual either has a reduced risk for or is protected from developing knee OA.
Among the 117 symptomatic knee OA cases in this study, the prevalence of unilateral and bilateral KL ≥ 2 was 3.0 and 16.5%, respectively (Table 6), whereas a high prevalence was found of 12.5 and 34.1% in Beijing, and 15.2 and 19.7% in Framingham studies [38]. The high prevalence of bilateral knee OA may indicate that environmental or ethnic (genetic) factors are among the essential factors in the development of knee OA; however, we did not find an association between knee OA severity (by KL grade) and significant variables (female, ≥57 years old, and ITGAM*01*01 genotype) in this study. Among those, the knee OA severity was almost classified as an early stage of the disease (KL grades 1 and 2). From the results, the present finding of HNA SNP (rs1143679) may be closely involved in the etiology and pathogenesis of knee OA, but not in the disease activity. Moreover, severe radiographic knee OA (KL grade 4) involving large osteophytes, marked narrowing of joint space, severe sclerosis, and definite deformity of bone ends was not observed in this study. Because individuals with KL grade 4 knee OA experience great pain and discomfort when they walk or move the joints, they cannot visit research sites by themselves and were therefore excluded from this study. Among patients with KL grade 4 (advanced stage and severe OA), SNP rs1143679 should be analyzed for further study to better assess disease activity.
The major limitation of this study was the heterogenous nature of the disease in a small study cohort. The findings should be replicated in a larger sample population, in order to generalize the results observed in the current investigation and find new associations of variants, if any, in the future. In addition, further studies of the condition in different ethnic populations are needed to better characterize the risk profiles of knee OA based on genetic variants. Relatedness between individuals can lead to spurious associations between genotype and phenotype; this could not be accounted for in our model. Furthermore, participants in the case and control groups may have had different recollection regarding exposure and reverse causality, leading to a unique source of bias in case-control studies.
The case-control classification should be more specific regarding the defined criteria of the dependent variable. The KL criteria are widely used because the prevalence of radiographic OA [39], defined as grade 2 or higher, is higher than that of symptomatic OA. In addition, clinical manifestations such as knee pain can be caused by other processes not related to knee OA.
In conclusion, our analysis of individual risk factors for symptomatic knee OA is the first such study in the Thai population and confirms that the SNP rs1143679 is associated with symptomatic knee OA. These results indicate that the ITGAM*01*01 genotype is a novel genetic risk factor in knee OA among Thais. These findings may contribute to understanding the beneficial application prospects for therapeutic approaches to knee OA.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cimb45050265/s1. Tables S1-S15: The modeling analysis 1 to 15 of logistic regression.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The datasets and materials used and/or analyzed during the current study are available from the corresponding author on reasonable request.