Portal de Periódicos da CAPES

Microbial Metabolism of Inflammatory Bowel Disease Drugs: Current Evidence and Clinical Implementations

2021; Elsevier BV; Volume: 162; Issue: 1 Linguagem: Inglês

10.1053/j.gastro.2021.07.050

1528-0012

Heike E. F. Becker, John Penders, Daisy Jonkers,

Biosimilars and Bioanalytical Methods

Immune-suppressive drugs are the most important treatment regimens for inflammatory bowel disease (IBD).1Torres J. Bonovas S. Doherty G. et al.ECCO guidelines on therapeutics in Crohn’s disease: medical treatment.J Crohns Colitis. 2020; 14: 4-22Crossref PubMed Scopus (433) Google Scholar,2Harbord M. Eliakim R. Bettenworth D. et al.Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 2: current management.J Crohn’s Colitis. 2017; 11: 769-784Crossref PubMed Scopus (741) Google Scholar However, a substantial number of IBD patients experience drug nonresponse or toxicity, leading to prolonged active inflammation and a higher risk of a worse long-term outcome, decreased quality of life, and high healthcare costs. Drug nonresponse varies among patients and between different drugs (with prevalence up to 50%) and remains largely unpredictable.3Li J. Wang F. Zhang H.J. et al.Corticosteroid therapy in ulcerative colitis: clinical response and predictors.World J Gastroenterol. 2015; 21: 3005-3012Crossref PubMed Scopus (17) Google Scholar, 4Timmer A. Patton P. Chande N. et al.Azathioprine and 6-mercaptopurine for maintenance of remission in ulcerative colitis.Cochrane Database Syst Rev. 2016; 5: CD000478Google Scholar, 5Mcdonald J.W.D. Wang Y. Tsoulis D.J. et al.Methotrexate for induction of remission in refractory Crohn’s disease.Cochrane Database Syst Rev. 2014; 8: CD003459Google Scholar Although some mechanisms, such as antibody formation against biologics,6Ben-Horin S. Chowers Y. Review article: loss of response to anti-TNF treatments in Crohn’s disease.Aliment Pharmacol Ther. 2011; 33: 987-995Crossref PubMed Scopus (434) Google Scholar are well known, most cases of nonresponse remain far from understood. To fill this knowledge gap, research is increasingly focusing on the role of intestinal microbiota in drug nonresponse or toxicity. Accumulating evidence shows that several medical drugs can have reciprocal interactions with intestinal microbiota, meaning that microbes are able to metabolize drugs, whereas drugs can alter the microbial composition by growth inhibition or stimulation and thereby can influence host (patho-)physiology. A classic example of microbiota-induced drug nonresponse is the inactivation of digoxin by Eggertella lenta strains expressing cardiac glycoside reductase, which can reduce digoxin into the less effective dihydrodigoxin.7Haiser H.J. Gootenberg D.B. Chatman K. et al.Predicting and manipulating cardiac drug inactivation by the human gut bacterium Eggerthella lenta.Science. 2013; 341: 295-298Crossref PubMed Scopus (392) Google Scholar Next to drug inactivation, microbial drug metabolism can also lead to drug activation (such as bacterial azoreduction of sulfasalazine resulting in active mesalazine)8Peppercorn M.A. Goldman P. The role of intestinal bacteria in the metabolism of salicylazosulfapyridine.J Pharmacol Exp Ther. 1972; 181: 555-562PubMed Google Scholar or toxicity (such as bacterial β-glucuronidase leading to toxic irinotecan reactivation).9Wallace B.D. Wang H. Lane K.T. et al.Alleviating cancer drug toxicity by inhibiting a bacterial enzyme.Science. 2010; 330: 831-835Crossref PubMed Scopus (636) Google Scholar In addition, IBD drugs have been shown to be metabolized by intestinal microbes, leading to less or more active metabolites.10Zimmermann M. Zimmermann-Kogadeeva M. Wegmann R. et al.Mapping human microbiome drug metabolism by gut bacteria and their genes.Nature. 2019; 570: 462-467Crossref PubMed Scopus (423) Google Scholar, 11Liu F. Ma R. Riordan S.M. et al.Azathioprine, mercaptopurine, and 5-aminosalicylic acid affect the growth of IBD-associated Campylobacter species and other enteric microbes.Front Microbiol. 2017; 8: 1-12PubMed Google Scholar, 12Oancea I. Movva R. Das I. et al.Colonic microbiota can promote rapid local improvement of murine colitis by thioguanine independently of T lymphocytes and host metabolism.Gut. 2017; 66: 59-69Crossref PubMed Scopus (48) Google Scholar, 13Letertre M.P.M. Munjoma N. Wolfer K. et al.A two-way interaction between methotrexate and the gut microbiota of male Sprague-Dawley rats.J Proteome Res. 2020; 19: 3326-3339Crossref PubMed Scopus (24) Google Scholar Considering the repeatedly reported microbiota perturbations in IBD patients and the interindividual differences in the microbial genetic content (metagenome) that dramatically exceed differences in the human genome, it is likely that the microbiota may be involved in interindividual variations in treatment (non)response and toxicity. The identification of microbial genes or pathways involved in IBD drug metabolism could therefore open new avenues to understand and prevent nonresponse and toxicity. In addition, microbiota perturbations could potentially even be restored to improve treatment response and reduce microbiota-induced toxicity. The aim of this commentary is first to provide insights into the current evidence on microbiota-mediated IBD drug metabolism and second to provide recommendations on future studies and techniques to facilitate timely clinical implementation. Major categories of IBD drugs include corticosteroids, immunomodulators, and biologics, with variable evidence regarding the involvement of human microbiome metabolism of these drugs. To date, the most detailed mechanistic insight has been acquired for thiopurines, which are recommended by the European Crohn’s and Colitis Organisation and American Gastroenterological Association guidelines and are still widely used for remission maintenance as well as for combination therapy with biologics.1Torres J. Bonovas S. Doherty G. et al.ECCO guidelines on therapeutics in Crohn’s disease: medical treatment.J Crohns Colitis. 2020; 14: 4-22Crossref PubMed Scopus (433) Google Scholar,14Feuerstein J.D. Ho E.Y. Shmidt E. et al.AGA clinical practice guidelines on the medical management of moderate to severe luminal and perianal fistulizing Crohn’s disease.Gastroenterology. 2021; 160: 2496-2508Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar,15Burisch J. Kiudelis G. Kupcinskas L. et al.Natural disease course of Crohn’s disease during the first 5 years after diagnosis in a European population-based inception cohort: an Epi-IBD study.Gut. 2019; 68: 423-433Crossref PubMed Scopus (127) Google Scholar They can be prescribed as 3 different prodrugs, namely azathioprine, 6-mercaptopurine, or 6-thioguanine. After liver metabolism, the active metabolites 6-thioguanine nucleotides (6-TGN) can, among other mechanisms, inhibit RAC1 and downstream target genes, leading to T-lymphocyte apoptosis.16Derijks L.J.J. Wong D.R. Hommes D.W. et al.Clinical pharmacokinetic and pharmacodynamic considerations in the treatment of inflammatory bowel disease.Clin Pharmacokinet. 2018; 57: 1075-1106Crossref PubMed Scopus (14) Google Scholar In silico screening of 5 intestinal bacterial strains against public genome and protein databases revealed that all strains contain at least 2 of 4 enzymes involved in thiopurine pro-rug metabolism. For example, Bacteroides vulgatus ATCC 8482 and Escherichia coli K12 contain the enzymes required for the metabolism of azathioprine to 6-TGN, whereas Bacteroides fragilis ATCC 25285 lacks glutathione S-transferase, catalyzing the conversion of azathioprine to 6-mercaptopurine.11Liu F. Ma R. Riordan S.M. et al.Azathioprine, mercaptopurine, and 5-aminosalicylic acid affect the growth of IBD-associated Campylobacter species and other enteric microbes.Front Microbiol. 2017; 8: 1-12PubMed Google Scholar Furthermore, a study in mice lacking hypoxanthine guanine phosphoribosyl transferase, which catalyzes 6-thioguanine into 6-TGN, has demonstrated that dextran sodium sulfate-induced colitis could effectively be ameliorated by exclusive microbial 6-thioguanine metabolism.12Oancea I. Movva R. Das I. et al.Colonic microbiota can promote rapid local improvement of murine colitis by thioguanine independently of T lymphocytes and host metabolism.Gut. 2017; 66: 59-69Crossref PubMed Scopus (48) Google Scholar This alternative, local treatment route may be highly interesting for further studies, because avoiding systemic thiopurine exposure likely prevents the numerous well-described side effects. A study comparing germ-free and recolonized rats further confirmed gut microbial impact on intestinal and hepatic levels of phase 2 xenobiotic-metabolizing enzymes. For example, the presence of a murine or humanized microbiota led to lower levels of glutathione S-transferase in colonic tissue.17Meinl W. Sczesny S. Brigelius-flohe R. et al.Impact of gut microbiota on intestinal and hepatic levels of phase 2 xenobiotic-metabolizing enzymes in the rat.Drug Metab Dispos. 2009; 37: 1179-1186Crossref PubMed Scopus (77) Google Scholar Another study in mice showed that Lactobacillus rhamnosus OLL2838 could reduce glutathione S-transferase levels and activity during gliadin-induced enteropathy.18Ogita T. Bergamo P. Maurano F. et al.Immunobiology modulatory activity of Lactobacillus rhamnosus OLL2838 in a mouse model of intestinal immunopathology.Immunobiology. 2015; 220: 701-710Crossref PubMed Scopus (11) Google Scholar This suggests that the intestinal microbiota can be involved in modulating the sites of thiopurine metabolism. Together, these studies demonstrate that the available microbial gene pool can contribute to the success of bacterial thiopurine metabolism. The generation of TGN by colonic bacteria might on one hand locally alleviate inflammation by affecting either the mucosal epithelium or immune cells, whereas microbial thiopurine conversion and sequestering could on the other hand reduce bioavailability to the host tissue, thereby contributing to thiopurine nonresponse. We highly recommend that future studies further explore to what extent microbiota-mediated thiopurine metabolism indeed plays a role in treatment (non)response. A few studies addressed the interaction between intestinal microbes and biologics, such as antibodies against tumor necrosis factor-α. Although they are administered parenterally, interaction with bacteria or their products may occur at the intestinal inflammation interphase with a leaky mucosal barrier. An in vitro study has shown that anti-tumor necrosis factor immunoglobulin 1-type infliximab could be cleaved by the immunoglobulin-degrading enzyme from the pathogen Streptococcus pyogenes (IdeS). Subsequent incubation with complement factor C1q and the natural killer cell receptor FCγRIIIA showed decreased binding with cleaved infliximab IgGs leading to reduced FCγRIIIA-mediated natural killer cell response.19Deveuve Q. Lajoie L. Barrault B. et al.The proteolytic cleavage of therapeutic monoclonal antibody hinge region: more than a matter of subclass.Front Immunol. 2020; 11: 1-12Crossref PubMed Scopus (7) Google Scholar Although S. pyogenes is a skin pathogenic commensal and typically does not reside in the intestine, similar mechanisms should be worth investigating in intestinal commensals. Furthermore, proteins from other bacterial taxa have been shown to exert nonimmune binding on the Fc or Fab region from IgGs, such as the immunoglobulin-binding protein family from E. coli. Immunoglobulin-binding proteins can be expressed on the surface of bacteria or can be excreted and can be smaller than 20 kDa.20Sidorin E. Solov’eva T. IgG-binding proteins of bacteria.Biochemistry. 2011; 76: 295-308PubMed Google Scholar Corticosteroids, including prednisolone and budesonide, are recommended for remission induction and are generally administered orally or rectally.1Torres J. Bonovas S. Doherty G. et al.ECCO guidelines on therapeutics in Crohn’s disease: medical treatment.J Crohns Colitis. 2020; 14: 4-22Crossref PubMed Scopus (433) Google Scholar,14Feuerstein J.D. Ho E.Y. Shmidt E. et al.AGA clinical practice guidelines on the medical management of moderate to severe luminal and perianal fistulizing Crohn’s disease.Gastroenterology. 2021; 160: 2496-2508Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar So far, there is only limited evidence on microbial metabolic mechanisms. A fecal culture study showed complete degradation of prednisolone and budesonide after a few hours,21Yadav V. Gaisford S. Merchant H.A. et al.Colonic bacterial metabolism of corticosteroids.Int J Pharmacol. 2013; 457: 268-274Crossref Scopus (41) Google Scholar and another in vitro study could detect 11β-hydroxy-1,4-androstene-3,17-dione as a product of prednisolone metabolism by Clostridium scindens.22Ly L.K. Rowles J.L. Paul H.M. et al.Bacterial steroid-17,20-desmolase is a taxonomically rare enzymatic pathway that converts prednisone to 1,4-androstanediene-3,11,17-trione, a metabolite that causes proliferation of prostate cancer cells.J Steroid Biochem Mol Biol. 2020; 199: 105567Crossref PubMed Scopus (14) Google Scholar Microbial corticosteroid metabolism before drug action is therefore likely, but the clinical relevance remains to be further explored. Other IBD drugs, which are typically administered orally and therefore may also interact with the intestinal microbiota, are acetylsalicylate mesalazine, calcineurin inhibitors tacrolimus and cyclosporine A, and JAK inhibitors, such as tofacitinib. However, studies investigating drug–microbiota interactions for these drugs in IBD patients are hardly available. So far, a combination of in silico, in vitro, and in vivo study designs have been applied to study drug–microbiota interactions for different drug classes and different diseases. Although a combination of different experimental designs is fundamental to gain a detailed understanding covering clinical outcomes and causal relationships, the heterogeneity of study designs and methods used so far makes it challenging to accumulate consistent findings and draw firm conclusions. Each study design has advantages and disadvantages (Table 1). For example, in vitro studies can contribute to unraveling causality and provide mechanistic insights into drug metabolization and responsible genes. However, the physiologic situation is poorly represented, which in turn is a strength of in vivo studies.Table 1Pros and Cons of Established Techniques Applied in Research on Drug–Microbiota InteractionsTechniqueUseProConIn vivoHuman longitudinal fecal samplingPhysiologic situation; can be related to treatment outcomesHigh interindividual variationsMice intestinal or fecal samplingComplete organ system; possibilities of host and microbiota modulationLimited translation to patient situation and individual differencesEx vivoComplex fermentation systemsComplex and adjustable model with controlled conditionsLargely loss of microenvironment and original composition, loss of host factorsFecal culturesLarge-scale screening of drug interactions and patient-specific microbiotaRapid loss of microenvironment and original composition, loss of host factorsIn vitroMock community culturePartially resembling microbial interactionsLack of microenvironment and host factorsSingle microbial cultureHigh internal validity, easy to modulateLack of microenvironment, microbial interactions, and host factorsIn silicoDatabase-driven associationsLarge-scale prediction of microbiota functions or drug responseLack of physiologic validation Open table in a new tab To date, comprehensive databases on microbial genes encoding drug-metabolizing enzymes are still lacking. Such databases are a prerequisite to generate adequate prediction models for drug–microbiota interactions similar to what has already been successfully implemented for human genes (ie, pharmacogenomics).23Freshour S.L. Kiwala S. Cotto K.C. et al.Integration of the drug–gene interaction database (DGIdb 4.0) with open crowdsource efforts.Nucleic Acids Res. 2021; 49: D1144-D1151Crossref PubMed Scopus (157) Google Scholar To ensure that the field of drug–microbiome research, or pharmacomicrobiomics, provides timely solid findings that can lead to creating such databases and ultimately improve treatment outcomes, we advocate several aspects to consider for future research: 1. Expand research on microbiota-related drug nonresponse. Several studies in various diseases found associations between microbiota composition and drug nonresponse. Increasing the number of studies in IBD is needed to create solid hypothesis that can then be further investigated to prove causality and reveal underlying mechanisms. We also recommend future clinical trials or cohort studies on IBD drug efficacy, including combination therapy, to include longitudinal microbiome analyses to relate microbiota composition and function with treatment outcomes and adverse effects. Study populations need to be carefully selected when investigating drug–microbiota interactions in vivo or ex vivo, because disease- and subject-specific microbiota compositions and function may influence the type of microbial drug metabolism. For this reason, Crohn’s disease and ulcerative colitis should be studied separately, and different phenotypes should be considered (Figure 1). Furthermore, specific drug formulations should be studied separately instead of combining them into drug classes. Different types of administration, such as oral or intravenous, should further be considered. 2. Apply knowledge on microbial physiology and advanced research techniques. To aid in identifying potentially unintended drug–microbiota interactions contributing to drug nonresponse or toxicity, research may focus on highly conserved pathways, which can be functionally comparable between humans and microbes. Subsequently, the suspected mechanisms should be confirmed in vitro and in vivo. Innovative techniques can further aid in combining mechanistic understanding with clinically relevant outcomes and may be superior to rather time-consuming combinations of established techniques (Figure 1). For instance, integrated cross-kingdom in vivo transcriptomics and proteomics24Westermann A.J. Vogel J. Cross-species RNA-seq for deciphering host–microbe interactions.Nat Rev Genet. 2021; 22: 361-378Crossref PubMed Scopus (21) Google Scholar accompanied by luminal and systemic drug metabolomics could provide detailed information about the triad of host–microbiota–drug interactions. Thereby, drug-related bacterial enzymatic activity and key bacterial taxa involved in a certain drug–microbiota interaction can be linked to detailed outcomes on luminal and systemic drug metabolites. Valid results can subsequently be used to build a database on drug–microbiota interactions. In addition, mechanistic insights connected with the patient’s physiologic response can directly be related to treatment response, toxicity, and other emerging relevant outcomes. Another innovative technique is the recently established gut-on-a-chip model, which allows a variety of manipulations and the co-culture of host tissue and intestinal microbes.25Ashammakhi N. Nasiri R. Barros NR de et al.Gut-on-a-chip: current progress and future opportunities.Biomaterials. 2020; 255: 1-19Crossref Scopus (69) Google Scholar Such an experimental setup can be used to study microbial enzymatic pathways involved in drug metabolism, using for instance microbial gain- or loss-of-function genetic screens, alongside with the impact of drug metabolites on host-derived tissue. It can be made host-specific by including intestinal microbiota donations and patient-derived stem cells. 3. Set required steps for timely clinical implementation. So far, the available evidence is not sufficient to implement findings into IBD clinical practice. However, emerging solid data can further help to improve precision medicine for IBD patients. Knowledge about microbiota composition and/or function could be used as a noninvasive screening tool for drug response or to adjust for site-specific drug action to reduce systemic side effects. Selected microbial consortia may increase, for instance, local activation of thiopurine prodrugs. Bacteriophages or enzyme inhibitors may be used to reduce certain taxa or inhibit specific microbial activity, such as β-glucuronidase inhibitors, to alleviate irinotecan toxicity.9Wallace B.D. Wang H. Lane K.T. et al.Alleviating cancer drug toxicity by inhibiting a bacterial enzyme.Science. 2010; 330: 831-835Crossref PubMed Scopus (636) Google Scholar Innovative approaches of engineered microorganisms with certain metabolic function or targeted delivery can also provide future applications. Promising examples include ammonium capture and L-arginine production by engineered E. coli Nissle 1917 to reduce hyperammonemia an treat urea cycle disorders,26Kurtz C.B. Millet Y.A. Puurunen M.K. et al.An engineered E. coli Nissle improves hyperammonemia and survival in mice and shows dose-dependent exposure in healthy humans.Sci Transl Med. 2019; 11: eaau7975Crossref PubMed Scopus (158) Google Scholar inflammation or tumor sensing with targeted drug release,27Yu X. Lin C. Yu J. et al.Bioengineered Escherichia coli Nissle 1917 for tumour-targeting therapy.Microb Biotechnol. 2020; 13: 629-636Crossref PubMed Scopus (37) Google Scholar or the delivery of nanofibers to improve mucosal barrier function in IBD.28Praveschotinunt P. Duraj-Thatte A.M. Gelfat I. et al.Engineered E. coli Nissle 1917 for the delivery of matrix-tethered therapeutic domains to the gut.Nat Commun. 2019; 10: 5580Crossref PubMed Scopus (119) Google Scholar Continuative research needs to increase knowledge on clinical efficacy and the safety of administering genetically engineered microorganisms. Certainly, patients would benefit from such low-invasive approaches, improving treatment outcomes and reducing side effects and complication risk while limiting additional risks. In general, the elaborated concepts and recommendations also apply at least in part to other diseases and medical drugs. Translated concepts should therefore be explored and integrated in future research on, for instance, other gastrointestinal or immune-mediated diseases. In conclusion, drug–microbiota–host interactions are relevant to consider in IBD treatment. Today, the evidence is too limited for direct clinical implementation but bears numerous promising leads. Applying purposive study designs and advanced research techniques to build solid databases on disease-related microbial drug metabolizing enzymes, new findings may be implemented in a timely fashion to facilitate precision medicine, thereby improving treatment outcomes and patient quality of life. Author names in bold designate shared co-first authorship. Heike Becker was funded by the NUTRIM Graduate Program Grant from Maastricht University.