Mycophenolic acid (MPA) is an immunosuppressive agent available either as an ester prodrug or as a sodium salt. Mycophenolate mofetil (MMF) is the 2-morpholinoethyl ester prodrug of MPA formulated to improve its bioavailability [Article:2308896]; [Article:1346731]. Mycophenolate sodium is a delayed release formulation that delivers MPA in the small intestine without being released into the stomach. MPA is indicated as prophylactic agent in patients receiving allogeneic renal, cardiac or hepatic transplants. It is a noncompetitive, selective and reversible inhibitor of inosine monophosphate dehydrogenase (IMPDH). IMPDH is an important rate-limiting enzyme involved in purine synthesis, which converts inosine monophosphate (IMP) to guanosine monophosphate (GMP) [Article:10878285]; [Article:8681386].
Following oral administration, the enteric-coated mycophenolate sodium is mainly absorbed in the small intestine where it is easily dissolved in the neutral pH of intestine. The enteric-coated tablet thus has an advantage over the ester prodrug (MMF) in that it is less likely to cause gastric disturbances in patients [Article:14974945]; [Article:17373988]. The oral bioavailability of MPA after MMF administration ranges from 80.7% to 94% [Article:17201457]; [Article:9646007]. The mean elimination half-life of the drug is 9 to 17 hrs. In blood, MPA is up to 97% albumin-bound [Article:17201457]; [Article:16269051].
Orally administered MMF undergoes rapid presystemic bioactivation to mycophenolic acid by carboxylesterases (CES), mainly CES-1 and CES-2 [Article:11950776]; [Article:20823294]. A review by Bullingham et al. reports that MMF is absent in plasma from healthy volunteers and renal transplant patients after administration of MMF, which hints towards rapid conversion of MMF to MPA [Article:9646007]. Within the intestine, MMF gets hydrolyzed to MPA, N-(2-carboxymethyl)- morpholine, N-(2-hydroxyethyl)-morpholine and the N-oxide of N-(2-hydroxyethyl)-morpholine [Article:11950776]; [Article:20823294] by CES-2 (see Figure).
MMF that escapes initial intestinal hydrolysis enters into the liver via the portal vein and gets converted to MPA in the hepatocytes. CES-1 and CES-2 are both expressed in liver; however only CES-2 is expressed in the intestine. The in-vitro hydrolysis rate of MMF is greater in human liver microsomes than it is in the human intestine microsomes [Article:20823294].
Phase II glucuronidation of MPA is a major metabolic pathway mediated by intestinal and liver UDP glucuronosyl transferases (UGTs). Evaluation of heterologously expressed UGTs for MPA 7-O-glucuronidation in liver, kidney, and intestine microsomes demonstrated that MPA glucuronidation occurs primarily in liver but also to some extent in intestine and kidney [Article:15258099]. Primary enzymes involved in MPA glucoronidation were UGT1A8 and UGT1A9 with minor roles for UGT1A1, 1A7 and 1A10. MPA-7-O-glucuronide (MPAG) is mainly excreted in urine by active tubular secretion and glomerular filtration. UGT1A8 and UGT1A10 are however expressed only extra-hepatically and thus are responsible for MPA metabolism in gastrointestinal tract [Article:15258099], [Article:15205394], [Article:10688250], [Article:9647757], [Article:15470161], [Article:11226133]. UGT1A9 plays a predominant role in hepatic MPA metabolism [Article:15258099], [Article:15470161].
Another metabolite of MPA is its acyl glucuronide form, Ac-MPAG, (generated by UGT2B7) which has comparable potency to MPA [Article:15470161], [Article:12021631], [Article:12548138], [Article:10053049], [Article:16790554]. Ac-MPAG induces cytokine release from mononuclear leukocytes, a likely cause of MMF side effects [Article:10751588]. Ac-MPAG is a minor metabolite and although it works through the same inhibition mechanism as MPA, it was found to be a weaker inhibitor of rhIMPDH II as compared to MPA, suggesting that it would not be pharmacologically active and might not be contributing to MPA's effect [Article:19299544].
The metabolite 6-O-desmethyl-MPA (DM-MPA) is formed by hepatic cytochrome P450 (CYP) enzymes mainly CYP3A4, CYP3A5 and to a lesser extent by CYP2C8 [Article:15570183]. It undergoes further conjugation to form two glucuronides that constitute very minute fractions of MPA. Their structural identification was not possible but they are theoretically assumed to be 4-O-phenyl and 6-O-phenyl glucuronides of DM-MPA [Article:15570183].
MPAG and AcMPAG, and not MPA, are substrates for Organic Anion Transport Polypeptides (OATPs). Using OATP transfected human embryonic kidney (HEK) cells it was observed that cells expressing OATP1B3 and OATP1B1 significantly accumulated MPAG [Article:21142914], [Article:17906856], [Article:19890249]. MPAG and Ac-MPAG get excreted in bile via canalicular Multidrug Resistance-associated Protein 2 (MRP2) encoded by the gene ABCC2. Biliary excretion of MPAG is possibly also related to the observed GI toxicity due to MPA. A study in Eisai hyperbilirubinemic rats (EHBRs), which are compromised in MRP2 due to presence of mutations, showed rapid clearance from plasma but limited biliary excretion of MPAG; suggesting its transport via Mrp2 on the bile canalicular membrane [Article:14978191]. Role of P-gp (MDR1) in oral absorption of MPA has been suggested by studies performed in MDR1a/1b
/ mice [Article:18586494]. MPAG excreted in bile undergoes de-glucuronidation by bacterial enzymes in the gastrointestinal tract, forming MPA, which again gets recycled [Article:17699403]. Pharmacokinetic studies show appearance of a secondary peak of MPA after 6 to 12 hours of oral administration indicating this enterohepatic circulation [Article:17699403]. MPA is primarily excreted in urine as MPAG (87%) metabolite and in negligible amounts as MPA (<1%). Using HEK293 cells for uptake experiments Uwai et al. suggested involvement of human renal organic anion transporters hOAT3 (SLC22A8), and hOAT1 (SLC22A6) [Article:17462604]. These results primarily supported role of hOAT3 in cellular uptake and renal tubular secretion of MPAG. Another study using Xenopus oocytes also suggested role of hOAT1 and hOAT3 in renal secretion of MPA and its metabolites [Article:17526543]. Since MRP2 is also expressed in renal proximal tubule brush border membrane, its can potentially play a role in renal transport of MPAG.
Two major pathways of purine synthesis exist: the de novo and salvage pathways. In the de novo synthesis, which operates in the lymphocytes [Article:10878285], [Article:16251851], [Article:415850], [Article:8516949], the first step is conversion of 5-ribose phosphate to 5-phosphoribosyl-1-pyrophosphate (PRPP). Next is the incorporation of a purine ring on the ribose phosphate, which involves a series of steps. This includes conversion of PRPP to inosine monophosphate (IMP). IMP then gets dehydrogenated to xanthine monophosphate (XMP) by IMPDH and further to guanosine monophosphate by GMP synthase. GMP gets converted to GTP and dGTP, which are involved in RNA and DNA synthesis, respectively. Conversion of IMP to XMP is the rate-limiting step in purine synthesis and is targeted by MPA (see Figure 1).
MPA has several mechanisms of actions that are related. The basic mechanism of action of MPA is the selective inhibition of T-lymphocyte proliferation at S phase. This is by selective inhibition of IMPDH thus depleting guanosine pool in the cell. Thymus and spleen have greater amounts of IMPDH in lymphocytes leading to greater cytostatic effect in these tissues as compared to other tissues [Article:10878285], [Article:16251851]. Of the two IMPDH isoforms, IMPDH1 is expressed in most cell types while IMPDH2 is expressed in activated lymphocytes [Article:1969416]. MPA inhibits IMPDH2 up to 4-5 fold more than IMPDH1 resulting in the more potent cytostatic effects of MPA on lymphocytes than on other cells [Article:10878285], [Article:415850].
Furthermore, reduction of GTP production decreases the expression of adhesion molecules that are responsible for recruiting monocytes and lymphocytes to sites of inflammation and graft rejection [Article:1826793]. Thus the goal of MPA treatment is to reduce allograft rejection by acting as an immunosuppressant [Article:10878285], [Article:8681386], [Article:16251851], [Article:415850], [Article:8516949], [Article:8322167].
Genetic variants within genes involved in MPA uptake and metabolism, and in its targets have been reported to affect MPA pharmacokinetics and response in patients undergoing transplantation. Some of the most significant studies reporting polymorphisms (SNPs) within UGT1A9, UGT2B7, SLCO1B1, SLCO1B3 and IMPDH are summarized below.
Genes encoding MPA metabolizing enzymes
UGT1A9 is highly expressed in liver and is the major enzyme involved in MPA glucouronidation to MPAG [Article:16790554]. Evaluation of genetic variation in both UGT1A8 (*2 and*3) and UGT1A9 (*2 and *3) on MPAG formation identified UGT1A8*3 to have a significantly altered glucourodination. However as UGT1A8 is extrahepatic, the role of UGT1A8*3 in gastroinstestinal tract has been implicated [Article:15258099], [Article:16790554]. UGT1A9*3 was associated with lower clearance and could have potential influence on inter-individual variation in the metabolism of MPA. However since both these SNPs (UGT1A8*3 and UGT1A9*3) occur with a minimum allele frequency of <5% their clinical impact is limited. Another in-vitro study in human liver microsomes identified novel polymorphisms within UGT1A9 and demonstrated high inter-individual variability in UGT1A9 expression. Several SNPs in the UGT1A9 promoter region were found to be significantly associated with UGT1A9 levels. These include -2152C>T (rs17868320) (p=0.0004), -275T>A (rs6714486) (p=0.0006), -440C>T (rs2741045) / -331T>C (rs2741046) (p= 0.046) and -665C>T (p=0.042) [Article:15284532]. Follow up studies identified significant impact of promoter SNPs (rs2741045/[rs2741046|/rsid/rs2741046]) in UGT1A9 on MPA pharmacokinetics in renal transplant patients [Article:17924828]. Presence of these UGT1A9 promoter variants was associated with greater MPA exposure however MPAG levels were shown to be associated with renal function [Article:17924828]. UGT1A9 promoter SNPs (-275T>A/-2152C>T; rs17868320/[rs6714486|/rsid/rs6714486]) (both occur in LD) have been associated with low MPA exposure in renal allograft recipients [Article:16198654], as well as in healthy volunteers [Article:17339869]. In healthy volunteers, the presence UGT1A9*3 was associated with higher exposure of MPA and AcMAPG [Article:17339869]. Pazik et al. also reported that presence of UGT1A9 promoter SNPs (-2152T/-275A) was associated with increased risk of rejection in Polish kidney allograft recipients [Article:22210424]. Another study in 338 renal transplant patients further confirmed the association of -275T>A/-2152C>T with lower MPA exposure in patients receiving tacrolimus in addition to corticosteroids and MMF as part of the immunosuppressive therapy [Article:19494809]. Additionally UGT1A9*3 was associated with higher MPA exposure when MMF was given in combination with tacrolimus or cyclosporine [Article:19494809]. These results were also confirmed in another study in stable renal transplant patients where promoter SNPs -275T>A/-2152C>T were associated with lower MPA exposure and additionally the carriers of these SNPs had higher incidence of gastrointestinal side effects [Article:19715905]. Although the promoter SNPs -275T>A and -2152C>T are not found in Asian population, Zhang et al. reported that in Chinese renal transplant patients another promoter SNP UGT1A9*1b (deletion of T at -118: dT9/10; rs3832043), which has been shown to enhance glucouronidation both in vitro and in vivo [Article:15284532], [Article:15115919] was associated with increased enterohepatic circulation of MPA [Article:18946804]. Dose-adjusted AUC6-12, which is indicative of enterohepatic circulation, was higher in patients carrying at least one allele without T deletion (-118dT10/10 and 9/10) as compared to patients with T deletion [Article:18946804], [Article:23052409]. The overall role of UGT1A9 promoter SNPs has been explored for association with MPA pharmacokinetics as well as with the risk of rejection, although the studies have demonstrated consistent results in transplant patients with respect to pharmacokinetics of MPA, more prospective studies are required to establish the benefit of genotyping in predicting the risk of rejection to further improve MMF therapy.
UGT1A8 and UGT2B7 polymorphisms:
Sequencing of first exon on UGT1A8 in 254 Caucasians and 41 African Americans identified 8 nonsynonymous SNPs. Stable expression of these variants in embryonic kidney cell lines identified potential impact of UGT1A8*3 (C277Y; rs17863762), *5 (G173A240), *7 (A231T), *8 (S43L; rs372427845), and *9 (N53G) proteins on formation of MPAG and AcMPAG, indicating importance of these amino acids for enzymatic activity [Article:16790554]. For UGT2B7 variants the impact was minimal. In Japanese renal transplant patients none of the variants of either UGT1A8 or UGT2B7 had an influence on MPA plasma concentration [Article:17211619]. Gastrointestinal side effects are a major problem in patients receiving MPA therapy, and evaluation of UGT2B7 802C>T variant showed association of this SNP with lesser side effects (as compared to the wild type) when measured on Gastrointestinal Symptom Rating Scale (GSRS) (p=0.009) [Article:19730281]. In another study focused on investigating the role of SNPs within UGT1A8, UGT1A9, UGT1A7, ABCC2 and UGT2B7 on gastrointestinal adverse events, it was observed that patients with UGT1A1*1/*1 genotype (non-carriers of the variant UGT1A8*2 allele) had higher risk of diarrhea as compared to carriers of UGT1A8*2 allele (with UGT1A1*1/*2 or *2/*2 genotypes) [Article:20565459]. Further since patients with either combination of MMF with tacrolimus or sirolimus had higher risk of diarrhea as compared to patients receiving combination with cyclosporine, analysis according to co-treatment with other immunosuppressive agent showed significant association of UGT1A8*2 with higher risk of diarrhea in patients receiving MMF with cyclosporine [Article:20565459].
Genes encoding transporters
SLCO1B1 and SLCO1B3 polymorphisms:
SLCOs are genes that encode for Organic Anion Transporter Polypeptides (OATPs). SLCO1B1 and SLCO1B3 are the genes implicated in uptake of MPA and its glucuronide metabolite MPAG into hepatocytes. Picard et al. evaluated the role of OATP1A2, OATP1B1, and OATP1B3 in MPA and MPAG uptake using human embryonic kidney cells (HEK cells) and reported that cells expressing OATP1B3 and OATP1B1 accumulated more MPAG [Article:19890249]. Further they observed significant association of SLCO1B3 SNP 334T>G/699G>A with MPA and MPAG pharmacokinetics in renal transplant patients receiving tacrolimus or sirolimus in combination with MMF but not in patients receiving cyclosporine in combination with MMF [Article:19890249]. Although in limited number of healthy Chinese subjects (n=42), dose adjusted MPA AUC4-12 was lower in adults carrying SLCO1B3 334T allele as compared to those with 334G allele; a similar association was not found in renal transplant patients co-administered with prednisone. Two of the important covariates associated with MPA and MPAG levels in this study were weight and concomitant steroid use indicating that these covariates should not be ignored when screening for genetic associations [Article:22227166]. In contrast to the previous study, results from Japanese renal transplant patients demonstrate significant association of SLCO1B3 334GG genotype with higher MPAG AUC0-12 as compared to 334T genotype (P = 0.027). This implies that the ratio of MPAG/MPA levels were higher in GG homozygotes as compared to TT homozygotes, which is in accordance to report by Picard et al. [Article:19890249]. With respect to OATP1B1 there are limited reports on polymorphisms, with one study by Miura et al., showing higher dose adjusted MPAG exposure in SLCO1B1*1/*1 carriers as compared to carriers with the SLCO1B1*15 allele (P=0.002) [Article:17906856] and another by Michelon et al. reporting reduced MPA transport associated with the SLCO1B1*5 allele (p<0.002) [Article:21142914].
MPAG is excreted in bile primarily by MRP2 (ABCC2) and this transport is essential for the enterohepatic circulation. Naesens et al. evaluated 7 SNPs in ABCC2 for impact on MPA PK and observed that promoter SNPs, rs717620 (-24C>T) and rs3740066 (-3972C>T) protect renal transplant recipients from a decrease in MPA exposure that is associated with liver dysfunction [Article:17060857]. Additionally they observed that -24C>T SNP was associated with lower MPA clearance at steady-state conditions. All the patients in this study received tacrolimus and corticosteroid in combination with MPA [Article:17060857]. In contrast to results mentioned above, among Chinese renal transplant recipients, ABCC2 -24C>T was not significantly associated with MPA, MPAG or Ac-MPAG exposure levels. These patients received cyclosporine, which is speculated to have masked the effect of the SNP [Article:18946804]. However another SNP, ABCC2 1249G>A was found to be associated with higher Ac-MPAG levels as compared to wild type (p = 0.016) [Article:18946804]. The combined effect of polymorphisms in SLCO1B3 and ABCC2 is reported by Miura et al., where the clearance of MPA in recipients with both the SLCO1B3 334T>G and the ABCC2 -24C>T variant genotypes (334GG and -24TT) was significantly lower than in those with both the SLCO1B3 (334T>G) TT and ABCC2 (-24C>T) CC genotypes [Article:17906856]. Although results indicated above show possible interaction of genotype with the combination drug given along with MPA, recent report suggests that MRP2 mediated transport of MPA is not influenced by cyclosporine, tacrolimus or sirolimus [Article:23339626].
Genes encoding drug target
IMPDH1 and IMPDH2 are the targets of MPA and are responsible for the suppression of lymphocyte proliferation. Influence of IMPDH activity on MPA response has been reported by several studies (reviewed by Glander et al. [Article:21889500]). SNPs in these genes might affect the immunosuppressive response, efficacy of MPA, and therefore, acute rejection in transplant patients. However the reports of association of SNPs with response and toxicity are not consistent and differ among studies. Evaluation of eight IMPDH2 SNPs in de novo kidney transplant patients identified only one intronic SNP, 3757T>C (rs11706052) and the variant allele © was found to be associated with increased IMPDH activity (p=0.04) [Article:19617864]. Another study found an association between the rs11706052 SNP and reduced anti-proliferative activity of the MPA, but not with any increase in the IMPDH activity [Article:19770842]. A study in a cohort of Polish renal transplant patients observed IMPDH2 rs11706052 to be associated with higher lymphocyte count (p=0.04 at week 4 and p =0.068 at week 8) and also decreased lymphopenia (p=0.032) in kidney allograft recipients, however no association with acute rejection, neutropenia or GI disturbance was observed [Article:21996196].
With respect to IMPHD1, two intronic SNPs, rs2278293 and rs2278294, have been shown to be significantly associated the incidence of biopsy-proven acute rejection in the first year post-transplantation [Article:17851563]. In renal transplant patients (n=456) IMPDH1, rs2278294 SNP was significantly associated with lower risk of rejection and a higher risk of leucopenia over the first year post-transplantation [Article:20679962]. In Japanese renal transplant patients IMPDH1 variants rs2278293 and rs2278294 were associated with development of subclinical acute rejection (which was dependent on night time or day time high MPA exposure), thus warranting further evaluation of these SNPs along with therapeutic monitoring of MPA in transplant patients [Article:20136638]. Study focused on evaluation of phased haplotypes inferred from 5 IMPDH1 SNPs (rs2288553, rs2288549, rs2278293, rs2278294, and rs2228075) showed that carriers of the most common haplotype experienced GI intolerance as compared to non-carriers in pediatric heart transplant patients [Article:20649757].
Cao et al. studied effects of IMPDH polymorphisms in 240 hematopoietic stem cell transplant patients and reported that for rs2278294 genetic variation, recipients with the GG genotype had a significantly higher incidence of acute graft-versus-host disease than in recipients with the GA or AA genotype (p=0.002) [Article:21745452].
Overall, SNPs in IMPDH1 and 2 have shown some significant association with MPA pharmacokinetics and pharmacodynamics, but the limited reports with inconsistent evaluation warrants a systematic evaluation of variants in these genes in a bigger cohort of patients.
Mycophenolic acid is an immunosuppressant used to prevent rejection in organ transplantation. Inter-individual variability in the MPA exposure levels within and among populations has been observed. One of the reasons for this variability could be the inter-individual variability in MPA pharmacokinetics and pharmacodynamics. MPA is extensively metabolized by several hepatic as well as extra-hepatic UGTs and is a substrate for several influx and efflux transporters. Thus SNPs within the key genes involved in MPA pharmacokinetics (UGTs, SLCO1Bs, ABCC2 etc.) and pharmacodynamics (IMPDHs) could have impact on the MPA drug exposure and outcome. Here we summarize studies reporting the clinical association of genetic variants in these genes with MPA exposure and risk of rejection or GI toxicity (one of the major toxicities associated with MPA). Further since MPA is given in combination with other immune-suppressants such as tacrolimus or cyclosporine, impact of the combination drug used, or use of steroids have been shown to influence the association with SNPs. Thus, in conclusion, it is crucial to comprehend MPA PK/PD pathways and various polymorphisms within the enzymes involved, as well a consideration of other co-variates (combination drug, weight etc.) to improve therapeutic benefit of MPA.
M. Whirl-Carrillo, E.M. McDonagh, J. M. Hebert, L. Gong, K. Sangkuhl, C.F. Thorn, R.B. Altman and T.E. Klein. "Pharmacogenomics Knowledge for Personalized Medicine" Clinical Pharmacology & Therapeutics (2012) 92(4): 414-417. Full text
If you would like to reproduce this PharmGKB pathway diagram, please acknowledge the copyright to PharmGKB and state that permission has been given by PharmGKB and Stanford University. Also, please send a brief email to firstname.lastname@example.org to inform us of which pathway diagram you are using and for what purpose.
Entities in the Pathway
Drugs/Drug Classes (2)
Relationships in the Pathway
Download data in TSV format. Other formats are available on the Downloads/LinkOuts tab.