Genetic variations in methotrexate metabolic pathway genes influence methotrexate responses in rheumatoid arthritis patients in Malaysia

Methotrexate (MTX) is the most widely used disease-modifying anti-rheumatic drug (DMARD) for rheumatoid arthritis (RA). Many studies have attempted to understand the genetic risk factors that affect the therapeutic outcomes in RA patients treated with MTX. Unlike other studies that focus on the populations of Caucasians, Indian and east Asian countries, this study investigated the impacts of six single nucleotide polymorphisms (SNPs) that are hypothesized to affect the outcomes of MTX treatment in Malaysian RA patients. A total of 647 RA patients from three ethnicities (NMalay = 153; NChinese = 326; NIndian = 168) who received MTX monotherapy (minimum 15 mg per week) were sampled from three hospitals in Malaysia. SNPs were genotyped in patients using TaqMan real-time PCR assay. Data obtained were statistically analysed for the association between SNPs and MTX efficacy and toxicity. Analysis of all 647 RA patients indicated that none of the SNPs has influence on either MTX efficacy or MTX toxicity according to the Chi-square test and binary logistic regression. However, stratification by self-identified ancestries revealed that two out of six SNPs, ATIC C347G (rs2372536) (OR 0.5478, 95% CI 0.3396–0.8835, p = 0.01321) and ATIC T675C (rs4673993) (OR 0.5247, 95% CI 0.3248–0.8478, p = 0.008111), were significantly associated with MTX adequate response in RA patients with Malay ancestry (p < 0.05). As for the MTX toxicity, no significant association was identified for any SNPs selected in this study. Taken all together, ATIC C347G and ATIC T675C can be further evaluated on their impact in MTX efficacy using larger ancestry-specific cohort, and also incorporating high-order gene–gene and gene–environment interactions.


Introduction
Rheumatoid arthritis (RA) is an autoimmune disorder which abnormally attacks normal joints and results in in ammation. Most epidemiological studies of RA have been done in Western countries, showing a prevalence of RA in the range of 0.5-1.0% in the USA and northern European countries 1 . In Malaysia, 0.5% of the population is affected by RA 2 . It should be noted that there are 3 main ancestries in Malaysia, being 69.8% Malays, 22.4% Chinese and 6.8% Indian as according to the 2021 press release by the Department of Statistics Malaysia 3 .
Currently, the most commonly prescribed medication for RA is disease-modifying anti-rheumatoid drugs (DMARDs) including conventional synthetic DMARDs (csDMARDS), targeted synthetic DMARDs (tsDMARDS) and biologic DMARDs (bDMARDs). Other prescribed drugs include non-steroidal anti-in ammatory drugs (NSAIDs) and glucocorticoids 4 . Previous csDMARDs such as anti-malarial drugs (chloroquine and hydroxychloroquine) have been in use since the 1950s 5 . Since methotrexate (MTX) was readapted in the late 1980s, it has become the most widely used csDMARD 6 .
MTX is a folate anti-metabolite that suppresses disease activity and reduces joint pain. A low dose (15 mg -25 mg) of MTX per week is prescribed to patients through either subcutaneous or oral administration for at least three months and this drug has been proven to be an effective DMARD for RA 7,8 . However, about 30-50% of RA patients do not respond to MTX and thus ruling out MTX as a treatment option for these non-responders [9][10][11][12][13] . Moreover, up to 35% RA patients are forced to discontinue MTX due to the adverse drug effects including stomatitis, gastrointestinal upset, headache or minor central nervous system disturbance, hair loss, ulcers, liver toxicity, pancytopenia and pneumonitis 6,9−13 . Despite these limitations, MTX is still the gold standard for the treatment of RA and in Malaysia, the usage of MTX has increased 6-fold from 1997 to 2007 14 . Malaysia Clinical Practice Guidelines on RA recommends that patients are to take either an alternative medication or an increased dosage when MTX therapy shows signs of failure in e cacy or side effects 15 . Hence, anti-RA medicine can become unexpectedly lengthy, costly and ranges from not effective to partially effective in meeting treatment expectation.
Available literature has converged to hypothesize that the variability in MTX e cacy and toxicity is due to a dysregulation in the MTX pathway ( Fig. 1) [16][17][18] . As multiple enzymes mediate the metabolism of MTX, it is conceivable that the alterations of enzymes' availability and activity have a direct impact on MTX treatment. Our study focused on the potential dysregulation of 4 key enzymes: folylpolyglutamate synthase (FPGS), γglutamyl hydrolase (GGH), aminoimidazole-4-carboxamide ribonucleotide (AICAR) transformylase, and inosine triphosphate pyrophosphatase (ITPA). FPGS is an important enzyme responsible for converting MTX into a range of polyglutamate forms -methotrexate polyglutamate (MTX PG) and is therefore important to allow the retention of MTX bioavailable in the cell. The therapeutic effects of MTX in RA patients rely on its conversion to MTX PG 18 . The conversion of MTX to MTX PG by FPGS can be reversed by GGH 16 (Fig. 1). GGH removes the polyglutamates from MTX PG by performing serial trimming on the long chain MTX PG to yield MTX which is able to be exported from the cell. AICAR transformylase (encoded by ATIC gene). AICAR transformylase converts AICAR to formyl-AICAR (FAICAR), playing a role in the de novo purine synthesis (Fig. 1). MTX is able to block AICAR transformylase and hence causes an increase of intracellular level of AICAR. High level of AICAR inhibits the adenosine deaminase (ADA) and AMP deaminase, resulting in accumulation of intracellular adenosines 17 . These adenosines are then transported out of the cell into the extracellular compartment where they bind to adenosine receptors, notably the A2a and A3 receptors that activate the adenosine signaling pathway 19 . Active adenosine signaling can lead to multiple anti-in ammatory mechanisms, one of which is the inhibition of NF-κB signaling in broblast-like synoviocytes at the synovial joints, thus suppressing the cell proliferation and the secretion of in ammatory factors that causes dysregulated angiogenesis in RA. Other anti-in ammatory mechanisms include the inhibition of proin ammatory activities in neutrophils, macrophages, T-cells and endothelial cells [20][21][22] . ITPA is also played an important role in the adenosine signaling pathway. The presence of C94A (rs1127354; chr20:3213196) mutation may modulate the ability of ITPA to convert inosine triphosphate (ITP) to inosine monophosphate (IMP) required for de novo purine synthesis which will affect the amount of intracellular adenosines which can be exported for binding to adenosine receptors 19 , leading to multiple anti-in ammatory mechanisms.
Since FPGS, GGH, AICAR and ITPA are known to determine the rate limiting steps in MTX metabolism, the effects of genetic variations such as single nucleotide polymorphisms (SNPs) could modulate the pharmacokinetic properties of these enzymes 17,23−26 . In our study, six SNPs were selected for the genotyping analysis as based on their previously reported association with MTX treatment outcomes. These SNPs include FPGS A1994G (rs10106; chr9:127813796), GGH C452T (rs11545078; chr8:63026205), GGH C401T (rs3758149; chr8:63039169), ATIC C347G (rs2372536; chr2:215325297), ATIC T675C (rs4673993; chr2:215347616) and ITPA C94A (rs1127354; chr20:3213196). In addition, these candidate gene association studies were previously and mainly conducted in the Caucasian population and no similar study has been carried out for Malaysian RA patients. Therefore, the novelty of this study is to reassess the possible association of the SNPs with MTX treatment outcome in a Malaysian population. We hypothesize that speci c SNP changes that can alter gene function are able to explain the variability observed in MTX e cacy and toxicity. Thus, this study aimed to genotype six SNPs of the candidate genes (FPGS, GGH, ATIC and ITPA) in MTX metabolic pathway and determine their association with MTX therapeutic outcomes in Malaysian RA patients.

Characterization of the studied population
Demographics. This study recruited 647 RA patients from Sunway Medical Centre (n=297), Hospital Tuanku Ja'afar Seremban (n=304) and Hospital Selayang (n=96) ( Table 1). The number of Chinese RA patients, most of which were recruited from Sunway Medical Centre, was two times higher than Malay or Indian RA patients. Furthermore, female RA patients (88.7%) outnumbered male RA patients (11.3%) in this study. This sex-imbalance is consistent with the current literature which have shown a higher number of female RA patients.
MTX E cacy and Toxicity. Based on our criteria for categorization for MTX e cacy, we obtained a total of 252 adequate responders (ARs) and 352 inadequate responders (IRs): 58% of RA patients did not respond well to MTX ( Table 1). As for MTX toxicity, we identi ed 448 non-adverse drug reaction (Non-ADR) and 199 adverse drug reaction (ADR) patients: 1 in 3 RA patients developed at least one type of side effects during the MTX treatment.
Among 199 RA patients who experienced ADRs, 43 patients showed severe side effects and their MTX therapy were immediately ceased (Table 1). These 43 patients were excluded from the MTX e cacy analysis but included in the MTX toxicity analysis.

Differences of allelic and genotype frequencies among 3 ethnic groups
The TaqMan SNP genotyping assay was performed on all study samples for the following SNPs: FPGS A1994G (rs10106), GGH C452T (rs11545078), GGH C401T (rs3758149), ATIC C347G (rs2372536), ATIC T675C (rs4673993), and ITPA C94A (rs1127354). 5% of the samples were randomly chosen for each SNPs and then veri ed by Sanger sequencing. The sequencing results con rmed the accuarry of the TaqMan SNP genotyping assay results. The allelic frequencies and genotype counts for each SNPs in Malay, Chinese and Indian RA patients are shown in Table 2. The minor allele frequency (MAF) of all six SNPs, except ITPA C94A (rs1127354), showed signi cant variation among the RA patients for the three ethnic groups. The MAFs of ITPA C94A (rs1127354) in Malay, Chinese and Indian patients are 0.15, 0.16 and 0.14, respectively (p> 0.05; solid-line box in Table 2). In addition, genotype counts for the six SNPs were compared among the three ethnic groups using chi-square test. The results revealed that except GGH C452T (rs11545078) and ITPA C94A (rs1127354), other four SNPs signi cantly differ in genotype frequencies among the Malay, Chinese and Indian RA patients (dashed line boxes in Table 2).
Association of six metabolic SNPs with MTX e cacy and toxicity in three ancestry-speci c RA patients When the association study of SNPs with the MTX treatment was carried out using the entire cohort (n=647), there was no signi cant difference between ARs and IRs as well as Non-ADR groups and ADR groups in the allelic association tests. Logistic regression was then performed to test the standard models of disease penetrance (dominant, recessive, additive) for the interaction of six SNPs with MTX e cacy and toxicity in the cohort of 647 RA patients. The forest plot (Fig. 2) for the association between SNPs and the MTX e cacy and toxicity was performed using R package ggplot2 27,28 indicated no signi cant association between the SNPs with either MTX e cacy or MTX toxicity (Supplementary Table S2 and Supplementary Table S3, respectively).
We then strati ed the study cohort into three ancestry-speci c groups, Malay, Chinese and Indian. The numbers of AR and IR of the strati ed groups with their respective genotypes are presented in Supplementary Table S4. Using these data, signi cant differences were observed in ATIC C347G (rs2372536) (OR=0.5478, 95%CI=0.3396-0.8835, p = 0.01321) and ATIC T675C (rs4673993) (OR=0.5247, 95%CI=0.3248-0.8478, p = 0.008111) (Supplementary Table  S6). Based on the effect sizes obtained from our analyses, the risk of patients becoming IR was reduced by ATIC C347G (rs2372536) and ATIC T675C (rs4673993) for approximately 55% and 57%, respectively. When the inheritance models were applied to the ancestry-speci c strati cation, it could be inferred that: (i) ATIC C347G (rs2372536) was associated with AR in Malay RA patients under dominant and additive models; (ii) the minor allele of ATIC T675C (rs4673993) under three genetic models (dominant, recessive and additive) may predict a higher success rate in MTX treatment among Malay RA patients (Supplementary Table S6). All positive results for the association between SNPs and MTX e cacy are as shown in the forest plot (Fig. 3). All the six SNPs were not signi cantly associated with the MTX e cacy in either Chinese or Indian RA patients. Furthermore, there was no signi cant association of all six SNPs with MTX toxicity in the three ancestry-speci c groups (Supplementary Table   S5 and Supplementary Table S7).

Discussion
Majority of the available drugs used for the treatment of RA were clinically evaluated in European ancestries, this raises a concern about their e cacy and toxicity in other ancestry groups globally 29 . Asian populations especially those in South East Asia were considerably under-represented in pharmacogenomic and pharmacogenetic studies of RA 29.30 . Hence, the present study evaluated the outcomes of MTX treatment in three major ancestry groups in Malaysia and their association with 6 SNPs from the enzymes involved in MTX metabolism. Comparing with the studies conducted by geographical locations, our study attempted to delineate ancestry speci c risk factors that would increase the precision of the proposed association.
The MAF of ATIC T675C (rs4673993) recorded in Genome Aggregation Database (gnomAD) and 1000 Genomes is 0.3251 and 0.2855, respectively 31,32 . In our study, the allele frequency of ATIC T675C (rs4673993) for the overall cohort is 0.4 and it is 0.44, 0.31 and 0.54, respectively, in Malay, Chinese and Indian populations in Malaysia ( Table  2). By comparing the allele frequency between our study and the public database, we noticed that our population is carrying a higher allele frequency of ATIC T675C (rs4673993). When comparing allele frequencies by ethnicity within our study cohort, there was a signi cant difference between Malay, Chinses and Indian for this SNP.
Apparently, the allele frequency of ATIC T675C (rs4673993) in Indian and Malay subjects were signi cantly higher than that observed in Chinese subject and in the public databases.
Interestingly, our study suggested that the Malay RA patients with ATIC T675C (rs4673993) have a better treatment outcome upon MTX monotherapy. In other words, this minor allele was associated with an increased remission rate ≥ Current RA literature consistently highlights the hypothesis that the anti-in ammatory action of MTX is achieved through the indirect inhibition of AICAR. The ATIC T675C (rs4673993) SNP is positioned in the intronic region of ATIC. To our knowledge, there are no functional studies on this particular SNP in ATIC activity. Nevertheless, the intronic SNP either interferes the transcriptional regulation of the coding-enzyme or is in linkage disequilibrium (LD) with another coding SNP 23,36−38 . In the present study, since similar effect size of ATIC T675C (rs4673993) and ATIC C347G (rs2372536) was observed, both SNPs can be in LD. Nevertheless, this observation need further validation, since the current sample size is too small to perform a LD test and the lack of a reference panel for ATIC T675C (rs4673993) and ATIC C347G (rs2372536) in Malay patients.
Similar to ATIC T675C (rs4673993), the allele frequency of ATIC C347G (rs2372536) in Malay, Chinese and Indian populations is 0.45, 0.31 and 0.54, respectively; and the allele frequency of ATIC C347G (rs2372536) for the entire study cohort is 0.40 (  40 studied 422 Caucasian RA patients in Poland who were treated with MTX therapy and found that GG minor genotype signi cantly exhibited a good response to MTX. However, the lack of association between rs2372536 polymorphism and the clinical response to MTX was also reported in some studies 41,42 . Recently, two meta-analyses were performed to investigate the association between ATIC C347G (rs2372536) and MTX response 43,44 . The rst meta-analysis was based on ve studies of 1056 RA patients in which 722 were MTX responders and 334 were non-responders. This analysis found the difference of ATIC C347G (rs2372536) between Caucasions (Spain, Slovenia and Netherlands) and Asians (India), being associated with non-responsiveness to MTX treatment in Caucasians but not associated in Asians 43 . The second meta-analysis combined two European (Spain and Netherlands), one East Asian (Japan) and two South Asian (India) studies with 458 MTX responders and 398 non-responders in total 44 . When combining ve studies, ATIC C347G (rs2372536) demonstrated a signi cant association with non-responsiveness of MTX under the dominant and codominant models. Yet, geographical strati cation showed that the association of ATIC C347G with MTX response was still observed in Europeans in pre-allele, dominant and codominant models but not in South Asian populations 44 .
Despite all studies above demonstrated a signi cant association between ATIC variants and MTX e cacy, the results were raher inconsistent. Common factors for inconsistency such as small sample size and insu cient statistical power, study design, medication dosage, grouping criteria, and patient condition could cause limitations in the association study. Moreover, gene-gene interactions within folate and adenosine biosynthesis pathways may complicate the association study between SNPs and MTX treatment outcomes 45 . In fact, RA has complex inheritance patterns and no single genetic variant has a decisive role in MTX e cacy or MTX toxicity in the treatment of RA. By using the MDR (Multifactor Dimensionality Reduction) method, a cohort of 255 RA patients treated with MTX in the USA was evaluated with the e cacy of MTX treatment, and the results showed that 53% MTX responders was associated with high-order interactions among SNPs in ITPA (C94A), RFC1 (G80A), and ATIC (C347G) genes 45 . Upon excluding the predisposing genotype combinations, a 3.8-fold lower e cacy was observed 45 . Later, the same researchers extended their study of gene-gene interactions using ITPA (C94A), RFC1 (G80A), and ATIC (C347G) to another 3 RA cohorts (USA, Dutch and Swedish) 46 . Both USA and Dutch cohorts (n=435) con rmed a predisposing genetic attribute signi cantly associated with methotrexate response [odds ratio (OR)=2.9, 95% con dence interval (CI): 1.9-4.2; P<0.001]. Although the association of combined SNPs with MTX responsiveness in the Swedish cohort (n=530) could not be determined, the association was observed after the non-genetic factors, age, sex and anti-citrullinated protein antibody (ACPA) status were included in MDR analysis 46 . Thus, individual variants of ATIC may not play a direct role in MTX e cacy, future studies shall map the ATIC variants to drug response as based on the detection of nonlinear multigene interactions, this may improve the accuracy of predicting the MTX e cacy. In addition, other non-genetic covariates should be considered because the association study between genetic variants and MTX e cacy sometimes seems oversimpli ed understanding the MTX response in RA.
AICAR transformylase contains two domains which are MGS (methylglyoxal synthetase) like domain and AICAR binding domain 47 . ATIC C347G (rs2372536) causes the substitution of threonine (Thr) with serine (Ser) at position 116. Thr116 lies in the binding pocket of MGS-like domain and is the rst residue of α8 helix which likely serves as a N-cap residue stabilizing the helix by interacting with the amide groups from the main chain. We proposed that the side-chain hydroxyl group of Thr116 forms hydrogen bonds with the amide groups of Val117 and Glu118 (green arrowhead in Fig. 4), while its main-chain carboxyl group forms hydrogen bond with the amide group from Glu119 (blue arrowhead in Fig. 4). The methyl group of Thr116 might stabilize the hydrogen bond between Thr116 and the main chain. As Thr116 is substituted with serine, the methyl group can be removed and this results in a more exible C-N rotation. In other words, Ser116 causes the rearrangement of the protein structure at N-cap and thus, potentially affects the substrate-binding a nity and AICAR transformylase enzyme activity. This explains why the RA patients with the minor allele of ATIC C347G (rs2372536) might have a phenotypic change in response to MTX.

Conclusion
Present study suggested that ATIC C347G (rs2372536) and ATIC T675C (rs4673993) could in uence the response to MTX monotherapy in Malay patients with RA, while the other four SNPs failed to demonstrate their associations with the reduction of disease activity following the MTX monotherapy. ATIC C347G (rs2372536) and ATIC T675C (rs4673993) are not the only ancestry-speci c SNPs since any variations appear in the genes of MTX metabolic pathway are potentially able to affect the effectiveness of MTX treatment. As more Malay-speci c SNPs can be revealed, the prediction of poor response would enable patient to be placed on alternative drugs, while those with predicted good response could proceed with MTX treatment.
As for the future recommendation, ancestry speci c signal of ATIC should be validated in a larger replication cohort of a similar ancestry group pro le to reduce the Type II error rate of MTX treatment response. Both ATIC C347G (rs2372536) and ATIC T675C (rs4673993) warrant an in-depth investigation, especially in the Malay RA patients in Malaysia.

Study subjects
RA patients were recruited at Sunway Medical Centre (Selangor, Malaysia), Hospital Tuanku Ja'afar Seremban (Seremban, Malaysia) and Hospital Selayang (Selangor, Malaysia) from December 2016 to May 2019. The study was performed in accordance with the principles stated in the Declaration of Helsinki. Prior to starting the study, the ethical approval was obtained from the Sunway University Research Ethics Committee (SUNREC 2017/066), the Sunway Medical Centre Independent Ethics Committee (007/2016/ER), and the Medical Research Ethics Committee of Ministry of Health Malaysia (NMRR-17-2901-38245(IIR). RA patients enrolled in this study had ful lled ACR-EULAR (2010) response criteria and satisfy the inclusion criteria: (i) must be at least 18 years old, (ii) are Malaysian Malay, Chinese or Indian origin, (iii) have been treated with 15 mg MTX or more per week for at least 3 months and (iv) have been followed up 6 months since MTX treatment. The self-declared ancestry of a patient was decided based on both parents being of the same ancestry as the patient. Patients of non-Malaysian origin were excluded from this study. All recruited RA patients were subsequently subjected to two independent association studies on the potential pharmacokinetic impact of SNPs with MTX e cacy and with MTX toxicity. MTX E cacy. RA patients were DMARD naive at the time of MTX commencement and they were categorized into adequate responder (AR) and inadequate responder (IR). Adequate responders were interpreted as patients who are in clinical remission or have achieved low disease activity as de ned by Disease Activity Score-28 (DAS28CRP) for at least 6 months. On the other hand, inadequate responders have the same RA treatment as adequate responders but failed to achieve clinical remission or low disease activity as de ned by DAS28CRP but at present been treated with other DMARDs, mono, duo (excluding MTX and HCQ group) or triple therapy, either csDMARDs or tsDMARDs or bDMARDs. Hence, patients may or may not be on MTX at the point of recruitment.
MTX Toxicity. Patients were categorized into two groups for potential toxicity and side effect: non-adverse drug reaction (Non-ADR) group and adverse drug reaction (ADR) group. The categorization was marked based on whether they have experienced drug intolerance during the MTX treatment. All side effects were recorded from the start of the MTX treatment until the withdrawal due to adverse drug reactions.
Blood sample and clinical data collections A total of 647 RA patients were involved in this study after their informed consent was obtained. A total of 5 ml of the whole blood sample was collected in ethylenediaminetetraacetic acid (EDTA) tube from individual patients by venepuncture during patients' regular visits at the hospitals. Besides, clinicopathological and demographic data were also extracted from patients' clinical records and linked to deidenti ed patient blood samples collected for this study.

TaqMan® SNP Genotyping Assays
Genomic DNA from patients was isolated from patients' whole blood samples. Brie y, buffy coat was obtained from the blood sample by centrifugation. Genomic DNA was then extracted from the buffy coat by using QIAamp DNA Blood Mini Kit (Qiagen, Germany) according to the manufacturer's instructions. SNP genotyping of FPGS A1994G (rs10106), GGH C452T (rs11545078), GGH C401T (rs3758149), ATIC C347G (rs2372536), ATIC T675C (rs4673993) and ITPA C94A (rs1127354) were performed by using TaqMan® SNP Genotyping Assays (Thermo Fisher, USA) according to manufacturer's instructions. The genotype data of each participant were analyzed using an online software named "Genotyping V4.2" (Thermo Fisher Connect TM ). A total of 5% of the samples (n=33) for each respective SNP were randomly selected for PCR ampli cation (AmpliTaq Gold™ 360 Master Mix, Thermo Fisher Scienti c, USA) (Supplementary Table S1) and subsequently for the genotype veri cation by Sanger sequencing (1st BASE Pt Ltd). Sequencing results were curated with SnapGene V4.3.10 (from GSL Biotech; available at https://snapgene.com).

Statistical analysis
Genotype and allele frequency of all the selected six SNPs were calculated. A chi-square independence test was performed to test the association between SNPs and ethnic groups. Chi-square test and binary logistic regression were performed to investigate the association between SNPs and MTX e cacy and MTX toxicity (PLINK V1.09) 48 . Effect sizes of potential associations were calculated as odds ratio (OR) and 95% con dence intervals (CI) as a measure of the association between the categorical variables. A p-value of < 0.05 was considered to be statistically signi cant.
Declarations Tables   Table 1. Characteristics of the patients enrolled in this study. A total of 647 RA patients were strati ed into three ethnic groups, Malay, Chinese and Indian. a Data are presented in number (percentage) or mean (standard deviation) unless otherwise indicated; b Rheumatoid factor; c Anti-cyclic citrullinated peptide; d Adequate responder; e Inadequate responder; f Non-adverse drug reaction; g Adverse drug reaction.  Figure 1 Cellular pathway of MTX -uptake, transport, conversion to polyglutamate forms and downstream effects. MTX is absorbed through active transport mediated by solute carrier family 19 member 1 (SLC19A1). Inside the cell, MTX is converted to active methotrexate polyglutamates (MTX PGs) by folylpolyglutamate synthase (FPGS) and this process can be reversed by γ glutamyl hydro¬lase (GGH). MTX PGs directly inhibit dihydrofolate reductase (DHFR), aminoimidazole-4-carboxamide ribonucleotide (AICAR) transformylase which is coded by ATIC and thymidylate synthetase (TYMS). Proteins highlighted in blue are encoded by the genes chosen for genotyping in this study. Red diamond ( ) indicates the inhibitory activity and black arrowhead (▼) indicates the directional ow of reaction.   Tertiary structure of AICAR transformylase complexed with antifolate (aminoimidazole 4-carboxamide ribonucleotide) was retrieved from Protein Data Bank (https://www.rcsb.org) (PDB ID: 5UZ0)49. Thr116 (or T116) is in the MGS domain. The green arrowhead indicates the side-chain hydroxyl group of Thr116 that forms hydrogen bonds with the amide group from Val117 and Glu118. The blue arrowhead is where the main-chain carboxyl group of Thr116 forms hydrogen bond with the amide group of Glu119.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementaryTableS1S7Shaetal.docx