Pharmacokinetic Herb-Drug Interactions involving african traditional medicines - fingerprint analysis and in vitro metabolism studies

Thesis (PhD)--Stellenbosch University, 2018.Introduction Traditional, complementary and alternative medicines have been used to treat various health conditions. The use of such medicines among HIV/AIDS and TB patients in sub- Saharan Africa has increased considerably, and within this context, questions have been raised about the medical use of herbs or extracts as treatment alternatives without adequate clinical testing and without monitoring of adverse effects once on the market. Furthermore, potential herb-drug interactions (HDI) have been predicted based on the pharmacological and pharmacokinetic properties of prescription medications and the phytoconstituents within the herbs. This study investigated the potential of six popular African herbs consumed by HIV/AIDS and TB patients, viz., Withania somnifera, Glycyrrhiza glabra, Astragalus membranaceus, Inula helenium, Althaea officinalis and Ocimum basilicum, to inhibit the cytochrome P450 enzyme CYP2B6 and the esterase-mediated metabolism pathway of rifampicin and their ability to induce CYP3A4 and CYP2B6. The study was undertaken in four phases: (1) qualitative assessment of various classes of phytocompounds present in each herbal extract by biochemical phytoprofiling, (2) study of the potential inhibitory effects of the extracts on cytochrome CYP2B6 and on the metabolism pathway of rifampicin to 25-O-desacetyl rifampicin by in vitro assays using human liver microsomes (HLM), (3) analysis of the potential inducing effects of the herbal extracts on mRNA expression of CYP2B6 and CYP3A4 in HepG2 cell lines, using reverse transcription polymerase chain reaction (RT-PCR) and agarose gel electrophoresis (AGE), and (4) fingerprint analysis, identification and relative quantification of the major phytoconstituents present in each extract and prediction of compounds which may cause HDI by liquid chromatography-mass spectrometry coupled with photo diode array detection (LC-MS/PDA). Methods Dried roots of Withania somnifera, Glycyrrhiza glabra, Astragalus membranaceus, Inula helenium, Althaea officinalis and dried leaves and inflorescence of Ocimum basilicum were obtained from Pharma Germania, South Africa. Aqueous, methanol, ethanol and ethyl acetate extracts were prepared and analysed using biochemical tests to identify the presence of various classes of phytocompounds. HLM assays were conducted to evaluate the inhibitory potential of each extract on the CYP2B6-mediated metabolism of efavirenz to 8-Hydroxy efavirenz, and the biotransformation of rifampicin to 25-O-desacetyl rifampicin. The protocol included the incubation of the herbal extract, HLM, co-factors and substrates in phosphate buffer for 30 min at 37 oC, termination of reaction and HPLC analysis of the supernatant from the centrifuged assay sample (at 245 nm for efavirenz and its metabolite, and 254 nm for rifampicin and its metabolite). The half-maximal inhibitory concentrations (IC50) for the active extracts were calculated based on the percentage of remaining activity relative to the control. Time-dependent inhibition (TDI) IC50 fold-shift was evaluated using 30 min pre-incubation with NADPH, followed by incubation with substrate in buffer for another 30 min, using six concentrations (1-200 g/mL) of the herbal extract. CYP2B6 and CYP3A4 mRNA expression assays were conducted for measuring the induction potential of the extracts, where the 50% cytotoxic concentration (CTC50) of all herbal extracts was determined by screening them 1000.00- 31.25 g/mL) against HepG2 cells. The HepG2 cells were incubated for 24 h with the CTC50 concentration, for each herb. This was followed by extraction and purification of total mRNA and its expression through RT-PCR, followed by AGE. Relative sample expression levels were calculated and represented as fold-response levels of induction relative to a cell control (using rifampicin and dexamethasone as positive controls). LC-MS/PDA was used to identify and relatively quantify the potential phytochemical constituents in each extract. Results O.basilicum, G.glabra I.helenium and A.membranaceus contained most of the relevant groups of phytocompounds such as flavonoids, phenols, alkaloids, glycosides and terpenoids based on the biochemical qualitative analysis. The aqueous and methanolic extracts of O.basilicum showed reversible and time-dependent inhibition of CYP2B6 (TDI IC50s 33.35 g/mL, 4.93 g/mL, IC50 shift-fold >1.5 for both extracts), while the methanolic and ethanolic extracts inhibited the formation of 25-O-desacetyl rifampicin (IC50s 31 g/mL, 8.94 g/mL). The methanolic extract of O.basilicum showed the highest TDI with a 7.4-fold increase in the IC50. All extracts of I.helenium inhibited CYP2B6 (IC50s 63 g/mL, 89.43 g/mL) and rifampicin metabolism (IC50s 42.79 g/mL, 18.58 g/mL, 62.10 g/mL); the aqueous extract showed the highest TDI with a 3-fold increase in the IC50. Only the methanolic and ethyl acetate extracts of W.somnifera inhibited CYP2B6 (IC50 79.16 g/mL, 57.96 g/mL). TDI was mainly observed between the herbal extracts and CYP2B6. The ethanolic and methanolic extracts of A.officinalis induced CYP3A4, with 48%-foldresponse shift compared to the cell control (no inducer). The ethanolic extract of O.basilicum and G.glabra caused moderate induction of CYP3A4 and CYP2B6. The aqueous extract of A.membranaceus showed moderate and equal induction of CYP3A4 and CYP2B6 (36% fold-response increase). All extracts exhibited less than 2-fold induction (200%) response, in contrast to the positive controls, rifampicin and dexamethasone. Major phytocompounds detected in the LC-MS/PDA analysis of the extracts included flavonoids, phenols, glycosides, saponins and terpenoids. Relative amounts of the identified compounds were determined by comparison to standard calibrators, quercetin and gallic acid (expressed in mg/L equivalent units). Phenols such as rosmarinic acid (approximately 2298 mg/L in the aqueous extract) and caftaric acid were found in O.basilicum extracts along with the flavonoids salvigenin, rutin and isoquercetin and other compounds such as linalool, hydroxyjasmonic acid and eucommiol. The aqueous extract of I.helenium contained mostly polyphenols such as chlorogenic and caffeoylquinic acids, whereas other solvent extracts contained the sesquiterpenoid tanacetol A, helenin (isoalantolactone) and macrophyllilactone B. The methanolic and ethyl acetate extracts of W.somnifera comprised of withaperuvin, isopelletierine, salvigenin, withanolides and withaferin A (approximately 1117 mg/L in the ethyl acetate extract), and the extracts of A.membranaceus contained mainly calycosin, formononetin, astragalosides I and IV. The extracts of A.officinalis mainly comprised of quinic acid and altheahexacosanyl lactone derivatives (approximately 3874 mg/L in the methanol extract), d-galacturonic acid monohydrate, phloretin along with the fatty acid trihydroxy-octadecenoic acid, and the G.glabra extract mainly consisted of glabridin, liquiritin apioside, liquiritigenin and licochalcones. Conclusion A.membranaceus, I.helenium, O.basilicum and W.somnifera have been shown to contain astragalosides, alantolactones, phenolic acids and withanolides that may inhibit drug metabolising enzymes such as CYP2B6, CYP3A4 and the enzymes responsible for metabolism of rifampicin (including B-esterases). A.officinalis and G.glabra can cause moderate induction of CYP3A4 due to the higher concentration of lactones and chalcones present in their extracts. These in vitro findings may be relevant for clinical use of these herbs together with conventional medicines metabolised by these enzymes, if sufficient hepatic concentrations are attained in humans. The results of this study will help to guide planning and designing of clinical trials to confirm the potential relevance of HDI in patients.Inleiding Tradisionele, aanvullende en alternatiewe medisyne word gebruik om verskeie gesondheidstoestande te behandel. Die gebruik van sodanige medisyne vir MIV/vigs- en TB-pasiënte in Afrika suid van die Sahara het aanmerklik toegeneem, en in hierdie konteks word vrae gevra oor die medisinale gebruik van kruie of ekstrakte as behandelingalternatiewe sonder voldoende kliniese toetsing en sonder monitering van negatiewe gevolge nadat dit op die mark verskyn het. Moontlike kruiemedikasieinteraksies (KMIs) word voorts voorspel op grond van die farmakologiese en farmakokinetiese eienskappe van voorskrifmedikasie en die fitobestanddele in die kruie. Hierdie studie was ʼn ondersoek na die potensiaal van ses gewilde Afrika-kruie wat deur MIV/vigs- en TB-pasiënte gebruik word, naamlik Withania somnifera, Glycyrrhiza glabra, Astragalus membranaceus, Inula helenium, Althaea officinalis en Ocimum basilicum, om die sitochroom P450-ensiem CYP2B6 en die esterase-bemiddelde metabolismebaan van rifampisien en hul vermoë om CYP3A4 en CYP2B6 te induseer, te inhibeer. Die studie is in vier fases onderneem: (1) kwalitatiewe assessering van verskillende klasse van fitoverbindings teenwoordig in elke kruie-ekstrak deur biochemiese fitoprofilering, (2) bestudering van die potensiële inhiberende gevolge van die ekstrakte op sitochroom CYP2B6 en op die metabolismebaan van rifampisien tot 25- O-desasetiel-rifampisien deur in vitro-ontledings met gebruik van menslewermikrosome (MLM), (3) ontleding van die potensiële induserende uitwerkings van die kruie-ekstrakte op mRNA-uitdrukking van CYP2B6 en CYP3A4 in HepG2-sellyne met gebruik van omgekeerde transkripsie polimerasiekettingreaksie (RT-PCR) en agarose-jelelektroforese (AGE), en (4) vingerafdrukontleding, -identifisering en relatiewe versyfering van die vernaamste fitobestanddele teenwoordig in elke ekstrak en voorspelling van verbindings wat KMI‟s kan veroorsaak deur vloeibare chromatografiemassa- spektrometrie gepaard met fotodiode-rangskikkingwaarneming (LC-MS/PDA). Metodes Droë wortels van Withania somnifera, Glycyrrhiza glabra, Astragalus membranaceus, Inula helenium en Althaea officinalis en droë blare en infloressensie van Ocimum basilicum is van Pharma Germania, Suid-Afrika, verkry. Water-, metanol-, etanol- en etielasetaat-ekstrakte is voorberei en ontleed met behulp van biochemiese toetse om die teenwoordigheid van verskillende klasse fitoverbindings te identifiseer. MLM-ontledings is uitgevoer om die inhiberende potensiaal van elke ekstrak op die CYP2B6-bemidelde metabolisme van efavirenz tot 8-hidroksie-efavirenz en die biotransformasie van rifampisien tot 25-O-desasetiel-rifampisien te evalueer. Die protokol het die inkubasie van die kruie-ekstrak, MLM, kofaktore en substrate in fosfaatbuffer vir 30 minute teen 37 oC, beëindiging van reaksie en hoëdruk- vloeibare chromatografie-ontleding van die bodrywende stof van die gesentrifugeerde toetsmonster (teen 245 nm vir efavirenz en sy metaboliet, en 254 nm vir rifampisien en sy metaboliet) ingesluit. Die half-maksimale inhiberende konsentrasies (IC50) vir die aktiewe ekstrakte is bereken op grond van die persentasie oorblywende aktiwiteit in verhouding tot die kontrole. Tydafhanklike inhibisie (TAI) IC50-verskuiwingsvou is geëvalueer deur 30 minute pre-inkubasie met NADPH, gevolg deur inkubasie met substraat in buffer vir ʼn verdere 30 minute met die gebruik van ses konsentrasies (1-200 g/mL) van die kruie-ekstrak. CYP2B6- en CYP3A4 mRNA-uitdrukkingontledings is uitgevoer om die induksiepotensiaal van die ekstrakte te meet, waar die 50% sitotoksiese konsentrasie (CTC50) van alle kruieekstrakte bepaal is deur hulle teen HepG2-selle te toets (1000.00-31.25 g/mL). Die HepG2-selle is vir 24 uur met die CTC50-konsentrasie vir elke krui geïnkubeer. Dit is gevolg deur ekstraksie en suiwering van totale mRNA en sy uitdrukking deur RT-PCR, gevolg deur AGE. Relatiewe monsteruitdrukkingvlakke is bereken en as vouresponsvlakke van induksie in verhouding tot ʼn selkontrole voorgestel (met gebruik van rifampisien en deksametasoon as positiewe kontroles). LC-MS/PDA is gebruik om die potensiële fitochemiese bestanddele in elke ekstrak te identifiseer en te relatief kwantifiseer. Resultate O.basilicum, G.glabra, I.helenium en A.membranaceus het die meeste van die toepaslike groepe fitobestanddele soos flavonoïde, fenole, alkaloïde, glikosides en terpenoïede op grond van die biochemiese kwalitatiewe ontleding bevat. Die water- en metanoliese ekstrakte van O.basilicum het omkeerbare en tydafhanklike inhibisie van CYP2B6 getoon (TAI IC50s 33.35 g/mL, 4.93 g/mL, IC50-verskuiwingsvou 1.5 vir albei ekstrakte), terwyl die metanoliese en etanoliese ekstrakte die vorming van 25-O-desasetielrifampisien geïnhibeer het (IC50s 31 g/mL, 8.94 g/mL). Die metanoliese ekstrak van O.basilicum het die hoogste TAI getoon met ʼn 7.4-voudige toename in die IC50. Alle ekstrakte van I.helenium het CYP2B6 (IC50s 63 g/mL, 89.43 g/mL) en rifampisienmetabolisme (IC50s 42.79 g/mL, 18.58 g/mL, 62.10 g/mL) geïnhibeer; die waterekstrak het die hoogste TAI getoon met ʼn drievoudige toename in die IC50. Slegs die metanoliese en etielasetaatekstrakte van W.somnifera het CYP2B6 (IC50 79.16 g/mL, 57.96 g/mL) geïnhibeer. TAI is hoofsaaklik tussen die kruie-ekstrakte en CYP2B6 waargeneem. Die etanoliese en metanoliese ekstrakte van A.officinalis het CYP3A4 geïnduseer, met 48%- vouresponsverskuiwing in vergelyking met die selkontrole (geen induseerder nie). Die etanoliese ekstrak van O.basilicum en G.glabra het matige induksie van CYP3A4 en CYP2B6 veroorsaak. Die waterekstrak van A.membranaceus het matige en gelyke induksie van CYP3A4 en CYP2B6 (36%-vouresponstoename) getoon. Alle ekstrakte het minder as tweevoudige induksierespons (200%) getoon in vergelyking met die positiewe kontroles, rifampisien en deksametasoon. Vername fitobestanddele waargeneem in die LC-MS/PDA-ontleding van die ekstrakte het flavonoïde, fenole, glikosides, saponiene en terpenoïede ingesluit. Relatiewe hoeveelhede van die geïdentifiseerde verbindings is bepaal in vergelyking met standaard kalibreerders, kwersetien en galliensuur (uitgedruk in mg/L ekwivalente eenhede). Fenole soos rosmariensuur (ongeveer 2298 mg/L in die waterekstrak) en kaftariensuur is gevind in O.basilicum-ekstrakte tesame met die flavonoïde salvigenien, rutien en isokwersetrien en ander verbindings soos linaloöl, hidroksiejasmoniese suur en eukommiol. Die waterekstrak van I.helenium het meestal polifenole soos chlorogeniese en kaffeoïelkiniensure bevat, terwyl die ander oplosmiddelekstrakte die seskwiterpenoïed tanasetol A, helenien (isoalantolaktoon) en makrofillielaktoon B bevat het. Die metanoliese en etielasetaat-ekstrakte van W.somnifera het bestaan uit witaperuvien, isopelletiërien, salvigenien, witanolides en witaferien A (ongeveer 1117 mg/L in die etielasetaatekstrak), en die ekstrakte van A.membranaceus het hoofsaaklik kalikosien, formononetien en astragalosiede I en IV bevat. Die ekstrakte van A.officinalis het hoofsaaklik bestaan uit kiniensuur en alteaheksakosaniel-laktoon-derivate (ongeveer 3874 mg/L in die metanolekstrak), dgalakturonsuur-monohidraat en floretien tesame met die vetsuur trihidroksieoktadesenoësuur, en die G.glabra-ekstrak het hoofsaaklik bestaan uit glabridien, liwiritinapiosiede, likwiritigenien en likochalkone. Gevolgtrekking Daar is gevind dat A.membranaceus, I.helenium, O.basilicum en W.somnifera astragalosiede, alantolaktone, fenoliese sure en witanolides bevat wat middelmetaboliserende ensieme soos CYP2B6, CYP3A4 en die ensieme verantwoordelik vir metabolisme van rifampisien (insluitende B-esterases) kan inhibeer. A.officinalis en G.glabra kan matige induksie van CYP3A4 veroorsaak weens die hoër konsentrasie laktone en chalkone teenwoordig in hul ekstrakte. Hierdie in vitro-bevindinge kan relevant wees vir kliniese gebruik van hierdie kruie tesame met konvensionele medikasie wat deur hierdie ensieme gemetaboliseer is indien voldoende hepatiese konsentrasies in mense verkry word. Die resultate van hierdie studie sal help om die beplanning en ontwerp van kliniese toetse te rig om die potensiële relevansie van KMI‟s onder pasiënte te bevestig