B7-H4 Polymorphism Influences the Prevalence of Diabetes Mellitus and Pro-Atherogenic Dyslipidemia in Patients with Psoriasis

Background: The co-inhibitory molecule B7-H4 is located in the genomic regions associated with type 1 diabetes (T1D) susceptibility. However, the correlation of B7-H4 with glycometabolism and dyslipidemia has never been studied. Objective: To explore the influence of B7-H4 polymorphism on the prevalence of diabetes mellitus (DM) and dyslipidemia in psoriasis. Methods: In this single-center cross-sectional study, we recruited 265 psoriatic patients receiving methotrexate (MTX) treatment. Thirteen single-nucleotide polymorphisms (SNPs) in B7-H4 were genotyped. Serum levels of total cholesterol (TC), triglycerides (TG), lipoprotein (a) (LP(a)), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein (LDL), apolipoprotein A1 (ApoA1), and apolipoprotein B (ApoB) were measured at baseline and week 12. Results: The GG genotype carriers of rs12025144 in B7-H4 had a higher prevalence of DM (57.14% vs. 17.71% vs. 18.67%, p = 0.0018), and had a poorer response to MTX in diabetic patients (p < 0.05), compared with AA or AG genotype carriers. The AG genotype of rs2066398 was associated with higher levels of pro-atherogenic lipids. MTX significantly downregulated the level of anti-atherogenic lipid ApoA1 in AA genotype carriers of rs2066398. Conclusions: The genotypes rs12025144 and rs2066398 in B7-H4 were correlated with a higher prevalence of DM and dyslipidemia in psoriasis, respectively.


Introduction
Psoriasis is a chronic, inflammatory skin disease mediated by interactions between keratinocytes, dendritic cells, and distinct T cells (Th1 and primarily Th17 cells) [1]. B7-H4 is a newly discovered B7 family member that negatively regulates T-cell immunity, and it is located in the genomic regions associated with type 1 diabetes (T1D). Moreover, B7-H4 has been shown to protect islet β cells against attack by Th1 cells [2], Th17 cells [3], and self-reactive CD8+ T cells [4] by co-localizing with insulin on islet β cells [5]. Although B7-H4 negatively regulates Th1 and Th17 cell responses, the role of B7-H4 in psoriasis and its comorbidities have yet to be elucidated.
Diabetes mellitus (DM) is a category of metabolic illnesses defined by chronic hyperglycemia resulting from an absolute or relative insulin shortage and insulin resistance (IR). Psoriatic patients with type 2 diabetes (T2D) had a significantly higher IR than T2D patients without psoriasis. However, psoriatic patients with T1D may have relatively preserved pancreatic β cell function due to their significantly higher levels of C-peptide in both the fasted state and after the glucose challenge [6]. Epidemiological studies report that psoriasis is associated with increased incidence rates of new-onset DM, and the severity of psoriasis is correlated with the risk of T2D [7]. Notably, psoriasis and T2D share genetic and pathological similarities as well as common targets, such as CDKAL1, PSORS2, PSORS3, and PSORS4, which are psoriasis susceptibility genes that are also associated with T2D [8].
B7-H4 is expressed on islet β cells and co-localizes with insulin; however, the effect of B7-H4 on the secretion and action of insulin has not yet been explored. In this study, we analyzed the association of the genetic variation of 13 single nucleotide polymorphisms (SNPs) in B7-H4 with the prevalence of DM and with lipid profiles in 265 psoriatic patients.

Patients and Study Design
A total of 265 psoriatic patients aged ≥18 years were recruited at the Department of Dermatology at Huashan Hospital, Fudan University, between February 2015 and August 2019. All participants received oral methotrexate (MTX) therapy for 12 weeks. The medical ethics committee of Huashan Hospital authorized the protocol (approval number: MTX201501), and all patients provided informed written consent. The diagnoses were based on typical clinical features and/or histopathological criteria. Excluded were those who had received systemic therapies (acitretin, cyclosporin, glucocorticoids) for arthritis or psoriasis within the previous month. The topical treatments were discontinued for a minimum of 1 week prior to the study's onset. The therapeutic regimen followed the European guidelines on contraindications and restrictions on MTX. None of the patients used lipid-lowering drugs.

Treatment
The first oral dosage of MTX ranged from 7.5 to 10 mg once weekly. The dose was increased by 2.5 mg every 2-4 weeks to a maximum of 15 mg weekly, depending on the patient's clinical response, side effects, and hematology/chemistry test results. If liver enzymes were >2-and <3-fold higher than normal, the MTX dose was lowered by 2.5 mg per week and administered once 2-4 weeks later. MTX medication was terminated if liver enzyme levels were raised by more than 3-fold [9].

Assessment of Lipid Profiles and Disease Characteristics
Two licensed dermatologists used the psoriasis area severity index (PASI) and body surface area (BSA) score to assess the severity and extent of psoriasis. Standard laboratory procedures at Huashan Hospital were used to evaluate lipid profiles at baseline and after 12 weeks of MTX therapy, as well as fasting blood glucose at baseline. The following were recorded: sex, age, age at disease onset, smoking habits, alcohol intake, hypertension, diabetes, height, weight, and body mass index (BMI).
The SequenomMassARRAY Assay Design 3.0 software was used to design the PCR parameters and detection primers. The PCR products were subsequently used as templates for locus-specific single-base extension reactions. The resulting products were desalted and transferred to a 384-element SpectroCHIP array (Sequenom Inc., San Diego, CA, USA). MALDI-TOF MS (Sequenom Inc.) was used for allele detection. The mass spectrograms were analyzed using MassARRAY software (Sequenom Inc.). We performed quality control of SNPs and samples at a call rate of 99.7%, and we analyzed the distribution of 13 SNPs in psoriasis patients with the Hardy-Weinberg equilibrium (p > 0.05).

Statistical Analysis
The data are expressed as mean ± standard deviation (SD). As applicable, statistical analyses were conducted using the Mann-Whitney U test, unpaired t-test, unpaired t-test with Welch correction, paired t-test, χ 2 test, or Fisher's exact test. After adjusting for sex, age, age at disease onset, disease duration, height, weight, BMI, hypertension, diabetes, smoking, and baseline PASI score, a stepwise multiple regression analysis was performed. GraphPad Prism version 8 (Software Inc., La Jolla, CA, USA) and SPSS Statistics version 23.0 (IBM Inc., Chicago, IL, USA) software were utilized for data analysis. A p-value less than 0.05 was regarded as statistically significant.

Association between GG Genotype of rs12025144 in B7-H4 and Prevalence of DM in Psoriatic Patients
The SNP rs12025144 in B7-H4 was successfully genotyped in 264 of 265 psoriatic patients. Patients' clinical characteristics and lipid levels are summarized in Table 1. The prevalence of DM in GG genotype carriers of rs12025144 was significantly higher than in AA and AG genotype carriers (57.14% vs. 17.71% vs. 18.67%, p = 0.0018). Furthermore, GG genotype carriers of rs12025144 had an older age, older age at disease onset, higher weight and BMI, shorter disease duration, and lower prevalence of hypertension compared to AA and AG genotype carriers; however, no statistical difference was found due to the small sample size of patients with the GG genotype. Univariate regression analysis demonstrated that age (p = 0.000), age at disease onset (p = 0.001), disease duration (p = 0.011), weight (p = 0.015), BMI (p = 0.004), and hypertension (p = 0.001) were positively correlated with the prevalence of DM in psoriatic patients, as well as with the genotypes of rs12025144, rs12034996, and rs10047089. Only age (p = 0.000), BMI (p = 0.004), and the rs12025144 genotype (p = 0.004) were significantly correlated with DM in psoriatic patients in multivariate analysis ( Table 2).

Diabetic Patients with GG Genotype of rs12025144 in B7-H4 Had a Poorer Response to MTX Relative to Those with AA and AG Genotypes in Psoriasis
Among the 264 psoriatic patients were 53 diabetic patients (8 patients with GG genotype and 45 patients with AA or AG genotypes). The differences were compared between the clinical characteristics and the lipid profiles between psoriatic patients with and without DM and diabetic patients with different genotypes of rs12025144 ( , and a lower prevalence of hypertension (25.00% vs. 64.44%, p = 0.0544) than those with the AA or AG genotypes, but no statistically significant differences were found due to the small sample size of diabetic patients with the GG genotype.

MTX Significantly Downregulated ApoA1 in AA Genotype Carriers of rs2066398
We analyzed the effect of a 12-week MTX treatment on the lipid profiles according to the genotype of rs2066398 in 263 psoriatic patients. As shown in Table 5, the levels of TC (p < 0.0001), LDL (p = 0.0025), ApoB (p < 0.0001), and Lp(a) (p < 0.0001) were significantly downregulated in 263 patients. However, the significant downregulation of MTX on the level of ApoA1 (1.05 ± 0.17 vs. 1.03 ± 0.16 g/L, p = 0.0131) was observed in the AA-but not the AG-genotype carriers of rs2066398.

Discussion
Our results demonstrated that the prevalence of DM in psoriatic patients with the GG genotype of rs12025144 in B7-H4 was significantly higher than those with either the AA or AG genotype (57.14% vs. 17.71% vs. 18.67%, p = 0.0018). Moreover, GG genotype carriers had older age, older age at disease onset, shorter disease duration, and a higher weight and BMI relative to AA and AG genotypes carriers, but no statistically significant differences were found due to the limited sample size. According to a meta-analysis, the prevalence of DM in Afghanistan was 12.13%, and univariate meta-regression revealed that the prevalence of DM rose with age, hypertension, and obesity [10]. Other studies have also found a link between DM and the severity of psoriasis [11]. Although severe forms of psoriasis are more prevalent in males than in females, our study demonstrated that there were more women, but not more severe psoriasis, among GG genotype carriers. Our results align with those of a previous report, in which higher BMI and weight gain were found to be risk factors for incident psoriasis in older women [12,13]. Therefore, we infer that metabolic abnormalities in patients with the GG genotype may occur before the onset of psoriasis.
In addition, we further analyzed the differences in clinical characteristics in psoriatic patients with DM between the AA/AG and GG genotypes. A significantly lower mean PASI improvement at 12 weeks was observed in GG than in AA or AG genotype carriers (43.75% vs. 67.07%, p < 0.05), which was consistent with our previous finding that higher BMI was associated with a poorer response to MTX [14]. The GG genotype carriers with DM had a shorter disease duration and less hypertension than AA and AG genotype carriers with DM, which differed from the finding that psoriatic patients with DM had a significantly higher prevalence rate of hypertension (58.49% vs. 35.38%, p = 0.0028) and longer disease duration (15.93 ± 12.45 vs. 12.41 ± 9.97 years, p = 0.0299) compared with those without DM. Furthermore, significantly increased levels of the pro-atherogenic lipids TC, TG, LDL, and ApoB were found. Another study also reported higher levels of TC and LDL in psoriatic patients with DM than in those without DM [15].
The human B7-H4 gene (VTCN1) is found on chromosome 1p13.1 [16], which is among the genomic regions linked to T1D susceptibility (insulin-dependent diabetes gene: Idd10) [17,18]. T1D is characterized by T-cell responses directed against insulin-secreting pancreatic β cells. The reduction of membrane-tethered VTCN1 (mVTCN1) on islet cells and islet-resident macrophages has been reported to be correlated with the elevated serum soluble VTCN1 (sVTCN1) levels seen in T1D patients. Moreover, a gradual loss of mVTCN1 on the islet cells and islet-resident macrophages was mediated by metalloproteinase nardilysin (NRD1), and sVTCN1 was considerably less effective than mVTCN1 at inhibiting T-cell proliferation [19]. In addition, B7-H4 was co-localized with insulin on islet β cells [5], indicating that the B7-H4 pathway might play an important role in β cell function maintenance and insulin secretion. Therefore, we inferred that the increased rate of DM in GG genotype carriers of rs12025144 could be mediated by the insulin-dependent diabetes gene; however, its exact role requires further exploration.
Our study also found that the AG genotype of rs2066398 in B7-H4 was significantly associated with higher levels of pro-atherogenic lipids following multiple regression analysis. MTX significantly decreased the levels of pro-atherogenic lipids in all psoriatic patients, which corroborates the findings of a previous report [20]. However, lowering anti-atherogenic lipid levels using MTX was controversial. Our study showed that the significant downregulation induced by MTX on ApoA1 levels was only observed in AA genotype carriers of rs2066398. Another study also reported that MTX slightly decreased HDL levels [21]. According to reports, MTX has an anti-lipolytic effect mediated by the adenosine release [22]. Ex vivo epididymal adipose tissue from MTX-treated mice secreted less glycerol when cultured [22]. In addition, MTX promoted glucose uptake and lipid oxidation in skeletal muscle via the AMP-activated kinase (AMPK) activation [23]. Our previous study also identified that MTX could restore the immunosuppressive function of Tregs by upregulating CD73, thereby activating AMPK and inhibiting the mTOR pathway [24]. AMPK is a central regulator of energy metabolism, directly connecting MTX and energy metabolism.
Our study was limited by the lack of consensus on the classification of DM, as the registries do not hold data on, for example, HbA1C (glycosylated hemoglobin) or C-peptide levels.
In conclusion, this study firstly explores the association between B7-H4 SNPs and the prevalence of DM and pro-atherogenic dyslipidemia in psoriatic patients. The GG genotype of rs12025144 was significantly correlated with a higher prevalence of DM in psoriatic patients. The genetic background of abnormal lipid metabolism in GG genotype carriers of rs12025144 facilitated the induction of psoriasis. They differed from patients with psoriasis-induced DM in clinical characteristics as they had shorter psoriasis disease duration and a lower prevalence of hypertension. The AG genotype of rs2066398 was significantly correlated with increased levels of pro-atherogenic lipids, whereas the AA genotype of rs2066398 was correlated with the significant MTX-induced reduction in the level of the anti-atherogenic lipid ApoA1.

Conclusions
Briefly, B7-H4 may play a crucial role in abnormal glucose and lipid metabolism in psoriatic patients by regulating islet β cell function and insulin secretion.

Institutional Review Board Statement:
The study was reviewed and approved by Huashan Hospital, Fudan University institutional review board (approval #MTX201501).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors have no conflict of interest to declare.  T1D  Type 1 diabetes  T2D  Type 2 diabetes  VTCN1 V-set domain containing T cell activation inhibitor 1