Association of Glutathione Transferase M1, T1, P1 and A1 Gene Polymorphism and Susceptibility to IgA Vasculitis

Endothelial cell injury is a hallmark of IgA vasculitis (IgAV), possibly associated with various factors, including oxidative stress. Certain single nucleotide polymorphisms (SNPs) of glutathione S-transferases (GST) genes have been shown to increase susceptibility to oxidative stress. The objective of our study was to evaluate the gene polymorphisms of GSTM1, GSTT1, GSTP1, and GSTA1 in patients with IgAV. DNA was extracted from the blood of 124 children with IgAV and 168 age-matched healthy controls. A higher frequency of the GSTM1 null genotype was observed in patients with gastrointestinal (GI) system involvement compared to those without GI system involvement (51.5% vs. 28.6%, p = 0.011). Additionally, the GSTM1 null genotype was less prevalent (30.8% vs. 69.2%, p = 0.032), while the GSTP1 Val/Val genotype was significantly more prevalent in patients who developed urogenital complications (scrotal swelling) during the course of the disease (60% vs. 40%, p = 0.039). This study is the first to suggest an association between GSTM1 and GSTP1 polymorphisms and various phenotypes observed during the clinical course of IgAV in the pediatric population. However, it was performed on a national and likely single ethnic cohort, too small for definitive conclusions, so larger studies are needed to confirm this association.


Introduction
Immunoglobulin A vasculitis (IgAV), previously known as Henoch-Schönlein purpura, is a small-blood vessel vasculitis occurring predominantly in childhood [1,2].Its most prominent sign is skin manifestations in the form of non-thrombocytopenic palpable purpura or petechiae usually distributed on the lower extremities.Although skin involvement is the most typical feature of the disease, IgAV is often associated with systemic manifestations, including gastrointestinal pain and bleeding, arthralgia, and/or arthritis and glomerulonephritis [2,3].The etiology of IgAV remains to be clearly defined but is thought to be multifactorial, with genetic, environmental, and antigenic components [4].
Glutathione S-transferases (GSTs) are a superfamily of enzymes belonging to phase II enzymes, which catalyze the conjugation of many endogenous and exogenous electrophilic compounds to glutathione [5].They are involved in the detoxification of xenobiotics and the protection of oxidative damage.In humans, GSTs are expressed in cells of all organs and tissues in addition to the cellular specific distribution of individual isoforms (e.g., GSTP in biliary tract cells).Although different isoforms can be differently expressed in the same tissue depending on tissue exposure to endobiotics and xenobiotics, high intracellular concentrations in combination with their cell-specific distribution enables the use of GSTs as biomarkers of the expected damage of a certain cell type [6,7].
Class mu genes are approximately 5 kb in length, and in humans all five genes are clustered together on chromosome 1p13.3[8].Three alleles of GSTM1 class have been identified so far: GSTM1*A, GSTM1*B and GSTM1*0.The GSTM1*0/GSTM1*0 (GSTM1-null) genotype results in the deletion of the entire gene and therefore a complete lack of enzyme activity.Since no evidence of a functional difference between GSTM1*A and GSTM1*B has been reported, these two alleles are considered as a single functional phenotype.
GSTP1 catalyzes the binding of cisplatin to glutathione so it can be converted to a water-soluble form and, hence, can be easily excreted [9].The gene coding for the GSTP1 enzyme is found on chromosome 11q13 and is polymorphic.A non-synonymous A/G single nucleotide polymorphism at nucleotide position 313 in exon 5 results in the replacement of adenine to guanine, which results in the substitution of the amino acid isoleucine to valine at position 105.This substitution in the GSTP1 protein encoded results in decreased enzyme activity, thermal stability, and substrate specificity [10].
Because of their well-known xenobiotic metabolizing activity, genetic variations in GST genes have gained much attention recently.Any change in the expression of GST protein level might possibly influence an individual's susceptibility to carcinogens and various inflammatory and chronic diseases, including rheumatic diseases [11][12][13][14].
Increased neutrophil and monocyte functions such as chemotaxis, phagocytosis, and the production of reactive oxygen species (ROSs), and an imbalance between the oxidant and antioxidant system are suggested to contribute to the pathogenesis of IgAV as well [15,16].The defense mechanism against ROS is complex and may involve several enzymes including GSTs.
We have recently demonstrated that increased erythrocyte glutathione S-transferase (e-GST) activity may serve as a subtle indicator of kidney function impairment in children with IgAV [17].At disease onset, e-GST activity was significantly higher in patients who developed nephritis and remained higher after three and six months since disease onset.
Considering the above, the aim of our study was to assess the gene polymorphisms of GSTM1, GSTT1, GSTP1 and GSTA1 in patients with IgAV and to investigate their possible role as a genetic component of this disease.

Results
A total number of 292 individuals (124 IgAV patients and 168 controls) were genotyped for the GSTM1 and GSTT1, 289 individuals (121 IgAV patients and 168 controls) were genotyped for the GSTA1 and 288 individuals (120 IgAV patients and 168 controls) for the GSTP1.All IgAV patients had skin manifestations which were severe in 6 (4.8%) patients.Arthralgia or arthritis was diagnosed in 85.5% IgAV patients, gastrointestinal (GI) involvement occurred in 41% IgAV cases, and IgA vasculitis nephritis (IgAVN) in 30.6% cases.The demographics and clinical and laboratory parameters are summarized in Table 1.
Patients with IgA vasculitis were grouped according to the dominant clinical phenotype, i.e., according to the dominant organ involvement, into the following subgroups: patients with extended purpura, patients with severe skin manifestations (bullae, ulcerations, necroses), patients with arthralgias/arthritis, patients with nephritis, patients with GI manifestations (abdominal pain, bleeding from the GI tract), patients with intestinal perforation, and patients with swelling of the scrotum (Tables 2 and 3).Data are presented as a whole number (%); # applicable only for boys; a chi-square test; b Fisher exact test; * p < 0.05.
A higher frequency of the GSTM1 null genotype was observed in IgAV patients with GI system involvement in comparison to patients without GI system involvement (51.5% vs. 28.6%,p = 0.011).The GSTP1 Val/Val genotype appeared significantly more often in patients who developed urogenital complications (scrotal swelling) within the disease course (60% vs. 40%, p = 0.039).The same IgAV type, with scrotal swelling, shows significant differences for the GSTM1 null genotype (30.8% vs. 69.2%,p = 0.032).It should be emphasized that none of the patients with scrotal edema were suffering from IgAV or nephrotic syndrome that could potentially cause scrotal swelling.
Regarding the GSTT1 and GSTA1 genotypes, we have not observed any difference in distribution in IgAV patients regarding their clinical phenotypes (Supplement Tables S1 and S2).
Distribution of the GSTM1, GSTT1, GSTP1 and GSTA1 gene polymorphisms among IgAV patients and the control group are summarized in Table 4.By the term present genotype, we considered the presence of at least one allele in the case of GSTM1 or GSTT1.Among the 124 IgAV patients, 68 (54.8%) were found to have a GSTM1-null (homozygous deletion for GSTM1) genotype; the remaining 56 (45.2%) individuals were positive for GSTM1 (had at least one functional GSTM1 allele).The proportion of control group individuals with a GSTM1 null genotype was 86 (51.2%).No significant differences were found between the IgAV group (54.8%) and the control group (51.2%) for the GSTM1 null mutation (Table 4).
The GSTT1 null genotype was detected in 34 (27.4%)IgAV patients and 34 (20.2%)controls.No statistically significant differences were observed between patients and the control group.
The frequency of the recessive alleles T/T of the GSTA1 genotype was 17.4% in IgAV patients and 12.5% in controls, without significant differences between the two groups.The difference in GSTA1 genotypes between the two groups was not observed using codominant, dominant, recessive, overdominant, and log-additive genetic models.
We performed logistic regression analysis to understand the association between the clinical features of IgAV and investigated GST gene polymorphisms (Tables 5 and 6).The selected variables were included in the GSTA1 multinomial regression model: age at diagnosis, gender, a rash extended above the waist, severe skin changes (bullae, ulcerations, and necrotic lesions), arthritis and/or arthralgias, gastrointestinal involvement, nephritis, occult hemorrhage, number of relapses.Orchitis was a significant predictor between the C/C and T/T genotypes.For each unit increase in orchitis, the relative risk of orchitis being associated with C/C versus T/T genotype was 0.096 times higher, which was highly significant (p < 0.01).For heterozygous C/T vs. homozygous C/C genotype, the relative risk was 0.235 times higher, but the coefficient is not significant (p > 0.05).
Multinomial regression could not be generated for the GSTP1 variable due to an insufficient number of cases with the GSTP1 wild-type Val/Val genotype [18].
We further investigated the potential association between polymorphisms in the analyzed GST genes and the response to treatment as well as the prognosis of IgAV.The analysis revealed a very weak association between the examined polymorphisms, treatment response, and prognosis.Subsequent multivariate logistic regression analysis indicated that none of the polymorphisms significantly influenced the treatment response or prognosis of IgAV (Supplement Tables S3-S5).Legend: CI: confidence interval; # applicable only for boys.Note: * p < 0.05.

Discussion
Endothelial cell injury is a characteristic of IgAV, potentially linked to diverse factors like immune complex deposition, complement activation, inflammatory elements, and oxidative stress [19].Although the roles of immune complexes, complement, and various pro-inflammatory cytokines are relatively well-established in IgAV, there is limited data on the impact of oxidative stress on the disease's pathogenesis, specifically in relation to endothelial dysfunction [20].Oxidative stress might contribute to the pathophysiology of autoimmune diseases, either by activating transcription factors involved in cytokine production or by engaging in oxidative post-translational modifications of proteins [21].
Members of the GST enzyme superfamily display robust antioxidant capabilities against reactive oxygen species and peroxides.Nevertheless, the ability to trigger the GST response to oxidative stress appears to be genetically influenced.Specifically, nearly all members of the GST family demonstrate genetic polymorphism, leading to either a complete absence or reduction of enzyme activity [22].
Within the GST genes, GSTM1, GSTT1, GSTA1 and GSTP1 harbor well-defined single nucleotide polymorphisms (SNPs).Homozygous deletion polymorphisms of the GSTM1 or GSTT1 gene, as well as the GSTP1 Ile105 SNP and GSTA1 T/T have been demonstrated to abolish enzyme activity and elevate susceptibility to oxidative stress.Consequently, the insufficiency in GST enzyme activity resulting from genetic polymorphisms of these genes could pose a risk factor for the development of various diseases, including glaucoma, chronic obstructive pulmonary disease, psychosis, and cancer [23][24][25][26].To the best of our knowledge, the role of GST polymorphisms in the development of IgAV has not been investigated so far.
We recently published encouraging results related to the possibility of using e-GST activity as a sensitive marker of kidney function impairment in children with IgAV [17].
Namely, e-GST activity in IgAVN patients was significantly higher than in patients without nephritis and controls and remained significantly higher during the 6-month follow-up period.This prompted an investigation into whether differences exist in the distribution of specific GST genotypes among IgAV patients in relation to clinical phenotypes, as well as an assessment of the association between GST genotypes and the risk of developing IgAV and IgAVN.
Our study revealed that GST polymorphisms do not exhibit a heightened inclination toward developing IgAV but might be associated with an elevated occurrence of gastrointestinal and urogenital manifestations.This was primarily observed for the GSTM1 null genotype while the GSTP1 Val/Val (GSTP1-low activity) genotype and GSTA1 C/C genotype (wild type) showed certain tendencies that may be indicative.
GSTM1 gene deletion frequencies can vary according to the ethnicity of the studied population with implications for achieving the goal of precision/personalized medicine in clinical practice.The GSTM1 null genotypes have extensively been studied in various human populations [27].The GSTM1 null phenotype frequency found in the Croatian sample of healthy individuals (51.19%) is in line with the frequency observed for the European population (Austria 57.3%, France 49.0%, Italy 49.2%, Spain 55.3%) reported by other authors [28].The best studied polymorphism, GSTP1 Ile105Val, resulting from the substitution of the 105 Ile codon for the Val codon, most probably leads to changes in the substrate-specific thermostability and affects the conjugation activity.When compared with the GSTP1 Ile/Ile variant, the GSTP1 Val/Val variant was associated with reduced detoxification of epoxy diols of certain polycyclic aromatic hydrocarbons [29][30][31].The frequencies of heterozygotes Ile/Val (43.45%) and homozygotes Val/Val (5.36%) in the Croatian sample of healthy individuals are in concordance with observed frequencies in the European population.The GSTP1 Val/Val genotype is uncommon and exists in 5% of Caucasians [32].In the population of white Europeans, the frequency of heterozygotes Ile/Val was between 38 and 45.7% [33].
While there have not been comparable studies in IgAV to date, it is noteworthy that in one of the limited studies investigating the impact of oxidative stress on the development of IgAV, heightened DNA oxidation was observed in patients with gastrointestinal involvement during the active phase of IgAV [19].This could be attributed to the susceptibility of gastrointestinal cells to oxidative damage or the efficient repair mechanisms of these cells.Gastrointestinal involvement in IgAV may resemble the early phases of inflammatory bowel disease (IBD), both clinically and histologically [34].Studies investigating oxidative DNA damage in IBD have demonstrated elevated levels of 8-hydroxydeoxyguanosine, a marker of oxidative DNA damage, in the affected patients.In these investigations, the levels of 8-hydroxydeoxyguanosine remained unaffected by systemic inflammation or disease activity.The findings suggest an enhanced vulnerability of gastrointestinal cells to DNA oxidation in patients with IBD, mirroring our observations in patients with gastrointestinal involvement in IgAV [34][35][36].
Concerning the link between GST polymorphisms and the urogenital system, some research shown that variations in GSTM1, GSTT1, and GSTP1 are linked to male infertility [37][38][39].This is attributed to the primary role of the enzymes encoded by GSTs in mitigating reactive oxygen species.Consequently, these enzymes are more likely to influence the quantity, survival, and/or activity of produced sperm.
When considering GSTA1 polymorphisms, those linked to a reduced expression of the GSTA1 enzyme may result in an accumulation of carcinogens in the body, thereby elevating cancer risk [40].It has been observed that the expression of this enzyme rises under conditions of oxidative stress and proliferative hyperplasia related to chronic inflammation [41], but decreases in prostate intraepithelial tumors [42].This supports the hypothesis that GSTA1 could play a role in the development of prostate cancer.Nonetheless, the role of this enzyme has not been examined in IgAV.
Our study does not provide an explanation for why GST polymorphisms demonstrated an influence solely on gastrointestinal and urogenital manifestations in IgAV, without affecting other organ systems like kidneys, skin, or joints.Nevertheless, the finding regarding the absence of a link between GST polymorphisms and kidney involvement in IgAV is particularly surprising.This is because certain studies have indicated that the GSTM1-null genotype is correlated with a higher risk of developing end-stage renal disease and an increased susceptibility to oxidative stress in dialysis patients [43].Conversely, certain studies have not identified associations between the activity e-GST (enzymes encoded by GST genes) and markers of systemic inflammation or kidney function [6,17,44].
Another potential link between GST gene polymorphisms and IgAV is illustrated through the case of cyclophosphamide, which is also utilized in the treatment of severe nephritis in individuals with IgAV.Indeed, studies have demonstrated a significant association between the allelic variation GSTP1 Ile105Val and disease remissions dependent on cyclophosphamide treatment in patients with various autoimmune diseases, including vasculitis.Individuals carrying the Ile105Val SNP in at least one copy exhibited a notably higher response rate to the treatment [45].This specific variant of GSTP1, characterized by a lower conjugation capacity, leads to an extended and higher therapeutic dose of cyclophosphamide, suggesting that the reduced activity of this GSTP1 variant may underlie more effective disease treatment.In our study, we did not observe an association between polymorphisms in the analyzed GST genes and the response to treatment as well as the prognosis of IgAV.Additionally, the patients in this study were administered alternative immunosuppressants rather than cyclophosphamide.
We need to recognize several limitations of this case-control study.The relatively small sample size could significantly reduce the statistical power to investigate genuine associations, so for any potential associations further larger studies are needed.The lack of an association between the investigated GST polymorphisms and susceptibility to IgAV was unexpected.The absence of interdependence cannot exclude the possibility of an association in another population or different ethnic group since genetic association studies are susceptible to population-specific genotype effects.Furthermore, we were not able to perform functional analyses, i.e., measuring markers of oxidative stress.Finally, we did not examine gene-gene or gene-environment interactions.

The Patients Selection and Study Design
A case-control study included 124 children with IgAV from three Croatian tertiary centers for pediatric rheumatology in the period between 2019 and 2022.Patients were enrolled in the study upon satisfying the criteria set by the European League Against Rheumatism (EULAR)/Pediatric Rheumatology International Trials Organization (PRINTO)/Pediatric Rheumatology European Society (PRES) [46].Clinical data and laboratory parameters were collected from medical records.Four IgAV patients were excluded from GSTP1 statistical analysis due to amplification failure.A total of 168 age-and sex-matched individuals without any history of autoimmune diseases were selected as a control group.All the participants and/or their parents or their legal guardians provided written informed consent and the study was carried out with the principles of the Declaration of Helsinki.

GST Genotyping
3 mL of peripheral venous blood was collected from participants into sterile EDTAcontaining tubes.Immediately after collection, the whole blood was stored at +4 • C until use.Whole EDTA blood was used for genomic DNA extraction using QIAamp DNA Mini Kit kit (Qiagen, Inc., Chatsworth, CA, USA) according to the manufacturer's instructions.For assessing the presence of amplified PCR products of GSTM1 and GSTT1 multiplex PCR was performed.The housekeeping gene CYP1A1 was applied as an internal control [47].Primers used were: GSTM1 forward: 5 ′ -GAACTCCCTGAAAAGCTAAAGC-3 ′ and GSTM1 reverse: 5 ′ -GTTGGGCTCAAATATA CGGTGG-3 ′ .Exon 7 of the CYP1A1 gene was co-amplified and used as an internal control using the following primers: CYP1A1 forward: 5 ′ -GAACTGCCACTT CAGCTGTCT-3 ′ and CYP1A1 reverse: 5 ′ -CAGCTGCATTTGGAAGTGCTC-3 ′ .The primers used were GSTT1-forward: 5 ′ -TTCCTTACTGGTCCTCACATCTC-3 ′ and GSTT1-reverse: 5 ′ -TCACGGGATCAT GGCCAGCA-3 ′ [47].The assay does not distinguish heterozygous from homozygous wild type genotypes.Therefore, it notes only the presence (at least one allele present, homozygote or heterozygote-GSTM1-active genotype, GSTT1-active genotype) or the absence (complete deletion of both alleles, homozygote-GSTM1-null genotype, GSTT1-null genotype) of the specific genotype.PCR was carried out in DNA thermal cycler (Mastercycler PCR, Eppendorf, Hamburg, Germany).The presence of 215 bp band indicated the presence of amplified GSTM1 gene.The presence of 480 bp band indicated the presence of the amplified GSTT1 gene.PCR products were separated on a 2.5% agarose gel at 150 V for 10 min and stained with ethidium bromide.Single nucleotide polymorphisms of GSTP1 were analyzed using the PCR-RFLP method by Harries et al. [48].The GSTA1 polymorphism was determined by PCR-restriction fragment length polymorphism according to Ping et al. [49].The primers used were GSTA1 forward: 5 ′ -TGT TGA TTG TTT GCC TGA AAT T-3 ′ and GSTA1 reverse: 5 ′ -GTT AAA CGC TGT CAC CCG TCC T-3 ′ .The PCR was carried out in PCR thermal cycler (Eppendorf Mastercycler, Hamburg, Germany).PCR products were separated on a 2.5% agarose gel at 150 V for 10 min and stained with ethidium bromide.The presence of 176 bp band indicated the presence of the amplified GSTP1 gene.The presence of 481 bp band indicated the presence of the amplified GSTA1 gene.The amplification products (20 µL) were digested by 2U of restriction endonuclease Alw26I at 37 • C for 16 h.The restriction products were separated on a 2.5% agarose gel.The restriction site resulting in two fragments (91 bp and 85 bp) indicated a GSTP1 Val/Val (GSTP1-low activity) genotype.The presence of three fragments (176 bp, 91 bp and 85 bp) was indicative for GSTP1 Ile/Val genotype, while the presence of one 176 bp band indicated a GSTP1 Ile/Ile genotype (referent genotype).For the GSTA1 gene, the restriction site resulting in two fragments (385 bp and 96 bp) indicated a GSTA1 T/T (GSTA1-low activity) genotype, while the presence of one 481 bp band indicated a GSTA1 C/C genotype (referent genotype).The presence of three fragments (481 bp, 385 bp and 96 bp) was indicative of a GSTA1 C/T genotype.

Statistical Analysis
Numerical data were described as mean and standard deviation (SD) or median [interquartile range (IQR)], while categorical variables were presented as whole numbers (N) with percentages (%).The normality of the data distribution was assessed using the Shapiro-Wilk test.Differences in the categorical variables between groups were assessed using χ 2 and Fisher exact test and among the numerical ones using the Mann-Whitney U test.Binomial and multinomial logistic regression analyses were performed to investigate the association between the GSTM1, GSTT1, GSTP1, and GSTA1 gene polymorphisms in several IgAV and control group phenotypes.Kendall's tau correlation coefficient was used to investigate the correlation between gene polymorphisms and treatment (glucocorticoids, immunosuppressants).Binomial and multinomial logistic regression models were generated in R ver 4.2.3.Binomial logistic regression models were generated using the generalized linear models (glm) function embedded within the R base packages.The multinomial logistic regression model was generated using the net [50] package.Relative risks between the GSTA1 phenotypes were calculated using the Stargazer R package [51].The level of statistical significance was determined with p < 0.05.

Conclusions
In conclusion, the etiology of IgAV is intricate, encompassing genetic and environmental factors.Our data suggest that the GSTM1, GSTT1, GSTA1 and GSTP1 genotypes are not associated with susceptibility to IgAV in the population of Croatia nor with the response to treatment or the prognosis of the disease.However, when considering IgAV type with

Table 1 .
Demographic data, clinical features and laboratory findings in patients with IgAV.

Table 2 .
The distribution of particular genotypes for the GSTM1 polymorphism in IgAV patients regarding clinical phenotypes.
Data are presented as a whole number (%); # applicable only for boys; a chi-square test; b Fisher exact test; * p < 0.05.

Table 3 .
The distribution of particular genotypes for the GSTP1 polymorphism in IgAV patients regarding clinical phenotypes.

Table 4 .
GST genotypes and risk of developing IgAV.

Table 5 .
Binomial logistic regression analysis of association between GSTM1 and GSTT1 polymorphisms and the clinical features of IgAV.

Table 6 .
Multinomial logistic regression for between different GSTA1 polymorphisms and the clinical features of IgAV.The wild-type genotype (C/C) was selected as the reference category.