A zebrafish larval model for drug-induced hepatic Injury due to first-line antituberculosis drugs and possible prevention/treatment with N-acetyl-cysteine

Date
2022-11
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Background Tuberculosis (TB) is ranked as the second most deadly infectious disease worldwide. First-line TB medication is associated with the development of drug-induced liver injury (DILI). The hepatotoxicity from this combination therapy is mostly due to pyrazinamide (PZA), isoniazid (INH) and rifampicin (RIF), however, no information has been reported on the hepatotoxicity potential of ethambutol (EMB). DILI typically occurs within the initial few weeks of the intensive phase of therapy. The only way to treat DILI is by stopping the medication and considering a liver transplant in the case of liver failure. Halting treatment can lead to the development of multi-drug resistant TB (MDR-TB). Therefore, close monitoring during therapy is required. Moreover, a preventative intervention needs to be put in place in order to prevent TB-DILI from developing in the first place. Acetaminophen (APAP) has been shown to cause DILI due to the accumulation of the drug’s toxic metabolites in the liver following overdose of the drug. N- acetylcysteine (NAC) is an effective treatment and works by replenishing cellular glutathione in hepatocytes, thereby preventing liver injury from progressing to liver failure. It is therefore hypothesized that NAC may be able to reverse liver injury due to first line TB medication since it has been shown to be highly effective in the treatment of DILI associated with APAP, and has in fact become the standard of care. Previous studies have reported the potential of NAC in treating TB-dug associated DILI. However, this needs to be confirmed through further studies. Therefore, new models are needed for predicting which therapeutic compounds could cause DILI in humans, and new markers and mediators of DILI need to be identified. The physiological and metabolic processes in zebrafish (Danio rerio) are similar to that of humans and the transparency of the zebrafish larvae makes this vertebrate a suitable model for studying TB-DILI mechanism and treatment. This study therefore aims to develop a zebrafish larval model (<5dpf) for DILI due to TB drugs, using APAP as a positive control that is known to cause DILI and to investigate the potential of NAC in preventing liver injury. Methods High-performance liquid chromatography (HPLC) analysis of INH, PZA and RIF was performed using previously developed methods, while liquid chromatography tandem mass spectrometry (LC-MS/MS) using a previously developed method was used for the evaluation of EMB. An HPLC method for the quantitation of APAP was developed and partially validated. The method used a Shimadzu HPLC system coupled with a PDA detector. Successful separation was achieved by isocratic elution on a reverse-phase Venusil XBP C18 (4.6 X 100 mm, 5 μm) column using a mobile phase consisting of 0.1% formic acid in water and acetonitrile (82:18; A:B, v:v) at 0.650 ml/min flow rate, detection wavelength of 247 nm, column oven temperature of 28°C and injection volume of 10 μl. The chromatographic retention time was consistent at 3.26 min. The calibration curve covered a range of 3.13 to 200 μg/ml with a quadratic regression weighted 1/c (c: concentration). These analytical methods were used to determine the solubility and stability of these drugs in E3 medium at 28°C for 3 days. Thereafter, dose response experiments were performed for all drugs to obtain the dose (s) that resulted in liver steatosis as an indicator of DILI using Oil red O positive liver stains and EthoVision movement tracking software (for NAC) as endpoints. Optimisation of the doses and the evaluation of the ability of NAC to prevent or reverse TB drug-associated DILI was attempted in zebrafish larvae. For NAC, APAP, INH and PZA, previously reported zebrafish larval doses for each drug were used a as reference for the study. However, for rifampicin (RIF) and EMB, the doses were adjusted according to the human dose. Results Solubility and stability data showed that APAP, INH, PZA and EMB were soluble in E3 embryo water and stable at 28˚C for 3 days. However, RIF remained insoluble and unstable in E3 medium at 28°C after 24 hours, even with the addition of ascorbic acid at 20 μg/ml. Doses of 1 mM, 7 mM, 10 μM, 1.2 mM and 0.5 mM were selected as the doses associated with DILI for INH, PZA, RIF, EMB and APAP, respectively. EthoVision data showed that 12 μM and 16 μM NAC resulted in irritation and toxicity, respectively. However, 8 μM NAC was shown to be consistent with the negative control and was therefore selected as the safe dose for NAC. Due to time constraints and difficulties in breeding, NAC was not evaluated for its ability to prevent or reverse DILI due to first-line TB drugs and this will be elucidated in future. Conclusion A time efficient, economical zebrafish larval model for DILI was successfully developed. Current data illustrates that this model is able to accurately simulate known toxicity effects of first-line antituberculosis drugs on the liver. Furthermore, this model can be used to evaluate potential treatment interventions and prophylaxis for the reversal and/or prevention of DILI.
AFRIKAANS OPSOMMING: Agtergrond Tuberkulose (TB) word as die tweede dodelikste infeksiesiekte ter wêreld beskou. Eerstelinie- TB-medisyne word met die ontwikkeling van middelgeïnduseerde lewerskade (DILI) verbind. Die hepatotoksisiteit van hierdie kombinasiebehandeling is hoofsaaklik te wyte aan pirasinamied (PZA), isoniasied (INH) en rifampisien (RIF), hoewel geen inligting nog oor die hepatotoksisiteitspotensiaal van etambutol (EMB) aangemeld is nie. DILI kom gewoonlik binne die eerste paar weke van die intensiewe behandelingsfase voor. Die enigste manier om DILI te behandel is om die medisyne te staak en, in geval van lewerversaking, ’n leweroorplanting te oorweeg. Die staking van behandeling kan multimiddelweerstandige TB (MDR-TB) tot gevolg hê. Daarom word sorgvuldige monitering gedurende behandeling vereis. Boonop moet voorkomingsmaatreëls getref word om te probeer keer dat TB-DILI in die eerste instansie ontwikkel. Asetaminofeen (APAP) is bekend daarvoor dat dit DILI veroorsaak weens die opeenhoping van die toksiese metaboliete van dié middel in die lewer ná ’n oordosis. N- asetielsisteïen (NAC) is ’n doeltreffende behandeling wat sellulêre glutatioon in hepatosiete vervang en sodoende keer dat lewerskade in lewerversaking omsit. Daarom is die hipotese dat NAC dalk lewerskade as gevolg van eerstelinie-TB-medisyne kan omkeer omdat dit hoogs doeltreffend is in die behandeling van APAP-verwante DILI, en trouens die standaardbehandeling daarvoor geword het. Vorige studies het aangedui dat NAC potensiaal toon vir die behandeling van TB-middelverwante DILI. Nietemin moet dít deur verdere navorsing bevestig word. Daarom word nuwe modelle vereis om te voorspel watter terapeutiese verbindings DILI by mense kan veroorsaak, en moet nuwe DILI-merkers en - mediators geïdentifiseer word. Die fisiologiese en metaboliese prosesse by die visspesie bontrok (Danio rerio, of “zebra fish”) is soortgelyk aan dié by mense, en die deursigtigheid van die bontroklarwes maak dié werweldier ’n geskikte model vir die studie van die TB-DILI- meganisme en -behandeling. Die doel van hierdie studie was dus om ’n bontroklarwemodel (<5 dpf) te ontwikkel vir TB-middelverwante DILI, met APAP as ’n positiewe kontrole wat bekend is daarvoor dat dit DILI veroorsaak, en om die potensiaal van NAC vir die voorkoming van lewerskade te evalueer. Metodes Hoëprestasievloeistofchromatografie (HPLC) met behulp van voorheen ontwikkelde metodes en APAP is gebruik om eerstelinie-TB-middels te ontleed [EMB is met behulp van vloeistofchromatografie-massaspektrometrie (LC-MS) ontleed] ten einde oplosbaarheid en stabiliteit in E3-medium by 28 °C oor drie dae te bepaal. APAP is op ’n Shimadzu HPLC-stelsel in samehang met ’n PDA-opspoorder ontleed. Suksesvolle skeiding is verkry met behulp van isokratiese eluering op ’n omgekeerdefase- Venusil XBP C18- (4.6 X 100 mm, 5 μm) kolom, met ’n mobiele fase wat uit 0.1% mieresuur in water en asetonitriel bestaan (82:18; A:B, v:v), met ’n vloeitempo van 0.650 ml/min, ’n opsporingsgolflengte van 247 nm, ’n kolom- oondtemperatuur van 28 °C en ’n inspuitingsvolume van 10 μl. Die chromatografiese retensietyd was deurgaans 3.26 min. Die kalibreerkromme het oor ’n bestek van 3.13 tot 200 μg/ml gestrek, met ’n lineêre regressie met ’n gewigstoekenning van 1/c (c: konsentrasie). Daarna is dosisreaksieproewe met alle middels uitgevoer om te bepaal watter dosis(se) tot lewersteatose, as ’n aanwyser van DILI, lei. Hiervoor is ORO- (“oil red O”-)positiewe lewerkleuring en EthoVision-nasporingsagteware (vir NAC) as eindpunte gebruik. Bontroklarwes is gebruik vir dosisoptimalisering en die evaluering van die vermoë van NAC om TB-middelverwante DILI te voorkom of om te keer. Vir NAC, APAP, INH en PZA is voorheen aangemelde bontroklarwedosisse van elke middel as verwysingspunt vir die studie gebruik. Vir rifampisien (RIF) en EMB is die dosisse egter na gelang van die menslike dosis aangepas. Resultate Oplosbaarheid- en stabiliteitsdata toon dat APAP, INH, PZA en EMB oplosbaar en stabiel is in E3-embriowater by 28 ˚C oor drie dae. Nietemin bly RIF onoplosbaar en onstabiel in E3- embriowater by 28 °C na 24 uur, selfs al word askorbiensuur in ’n konsentrasie van 20 μg/mL bygevoeg. Deur ORO-positiewe kleuring van die lewer van bontroklarwes is dosisse van onderskeidelik 1 mM, 7 mM, 10 μM, 1.2 mM en 0.5 mM gekies as die dosisse INH, PZA, RIF, EMB en APAP wat met DILI verband hou. EthoVision-data toon dat onderskeidelik 12 μM en 16 μM NAC tot irritasie en toksisiteit lei. Die 8 μM NAC-dosis het egter met die negatiewe kontrole ooreengestem. Daarom is 8 μM NAC gekies as die veilige dosis om te evalueer in watter mate NAC TB-middelverwante lewerskade by bontroklarwes kan voorkom/omkeer. Weens beperkte tyd en uitdagings met teling is die vermoë van NAC om eerstelinie-TB- middelverwante DILI te voorkom of om te keer, nié geëvalueer nie. Dít sal in toekomstige navorsing bestudeer word. Gevolgtrekking ’n Tyddoeltreffende, ekonomiese bontroklarwemodel vir DILI is suksesvol ontwikkel. Huidige data toon dat die model bekende toksisiteitsuitwerkings van eerstelinie-TB-middels op die lewer akkuraat simuleer. Boonop kan die model gebruik word om moontlike behandelingsintervensies en profilakses vir die omkering en/of voorkoming van DILI te evalueer.
Description
Thesis (MSc) -- Stellenbosch University, 2022.
Keywords
Drugs -- Side effects, Antitubercular agents, Hepatotoxicology, Diseases -- Animal models, UCTD
Citation