The mutagenicity of isoniazid in Salmonella and its effects on DNA repair and synthesis in human fibroblasts

What are the origins of growing microbial resistance? Both Lamarck and Darwin were right

INTRODUCTION

Microorganisms of clinical importance frequently develop resistance to drug therapy, now a growing problem. The experience with Mycobacterium tuberculosis is a representative example of increasing multi-drug resistance. To avoid reaching a crisis in which patients could be left without adequate treatment, a new strategy is needed. Anti-microbial therapy has historically targeted the mechanisms rather than origin of drug resistance, thus allowing microorganisms to adapt and survive.

AREAS COVERED

This contribution analyses the historical development (1943-2020) of the evolution of multi-drug resistance by M. tuberculosis strains in light of Darwin's and Lamarck's theories of evolution.

EXPERT OPINION

Regarding the molecular origin of microbial drug resistance, genetic mutations and epigenetic modifications are known to participate. The analysis of the history of drug resistance by M. tuberculosis evidences a gradual development of resistance to some antibiotics, undoubtedly due to random mutations together with natural selection based on environmental pressures (e.g., antibiotics), representing Darwin's idea. More rapid adaptation of M. tuberculosis to new antibiotic treatments has also occurred, probably because of heritable acquired characteristics, evidencing Lamarck's proposal. Therefore, microbial infections should be treated with an antibiotic producing null or low mutagenic activity along with a resistance inhibitor, preferably in a single medication.

Mutagenic Activity of p-Toluenesulfonhydrazide

Toluenesulfonhydrazide (TSH) is a high volume production chemical for which there is relatively little toxicological data. In this study, the mutagenic activity of TSH was determined in the Salmonella/mammalian microsome assay and the in vitro chromosomal aberration assay using Chinese hamster ovary cells. TSH induced gene mutations both with and without metabolic activation in the Salmonella/mammalian microsome assay but that it did not induce chromosomal aberrations in Chinese hamster ovary cells. The results of this study indicate that TSH is an in vitro mutagen and should be assessed for in vivo mutagenicity.

Detection of oxidative mutagenesis by isoniazid and other hydrazine derivatives in Escherichia coli WP2 tester strain IC203, deficient in OxyR: strong protective effects of rat liver S9

Strain IC203, deficient in the OxyR function, was sensitive to both cytotoxic and mutagenic effects of isoniazid (INH) whereas its parent, WP2 uvrA/pKM101, was resistant to these effects. Four other hydrazine compounds, hydrazine hydrate (HZH), phenylhydrazine (PHZ), hydralazine (HLZ) and nialamide (NLD), were mutagenic in WP2 uvrA/pKM101. Increases in mutagenicity were observed in IC203 for HZH and PHZ but not for HLZ and NLD. Growth inhibition zones by HZH, PHZ and NLD were larger in IC203 than in WP2 uvrA/pKM101. The enhancements in the effects of INH, HZH and PHZ in IC203 with respect to its oxyR+ parent are considered to be caused by the production of reactive oxygen species. This is consistent with its inhibition in IC203 by S9 from liver of uninduced rats, probably through the action of catalase. Mutagenicities of INH, PHZ and HLZ were low in strains IC204, a derivative of WP2 uvrA carrying a deletion of the umuDC genes, and IC206, a derivative of IC204 deficient in the MutY glycosylase. In these strains, HZH and NLD induced a high level of revertants which carry suppressor mutations resulting exclusively from G:C-A:T transitions, thus suggesting a direct reaction of the two hydrazines with cytosine.

Formation and persistence of isoniazid-DNA adducts in mouse tissues

Further confirmation is provided to support the identity of the product formed by the in vitro reaction of isoniazid (INH) with cytosine as 4-deamino-4-isoniazidocytosine (INH-cytosine). Use of INH-treated mice, in which the tissue DNA is prelabelled by the neonatal administration of [3H]-deoxycytidine, has revealed the presence of two other DNA products in addition to INH-cytosine. Tissue differences in the persistence of these three DNA products suggest the presence of repair reactions for certain adducts. These processes and the effects of hepatotoxicity lead to a selective retention of adducts in the DNA of lung which is the target tissue for INH-carcinogenicity in mice. INH-modified DNA templates are weakly promutagenic during in vitro DNA synthesis. The implications of these observations for the role of INH as an initiating agent and for the species differences in its carcinogenicity are discussed.

Determination of the activity of 16 hydrazine derivatives in the bioluminescence test for genotoxic agents

The activity of 16 hydrazine derivatives was determined in the bioluminescence test for genotoxic agents (BLT). Hydrazine compounds that were shown to exert mutagenic activity in the Ames test were also active in the BLT. Isoniazid and p-tolylhydrazine which reacted as weak mutagens in the Ames test were highly active in the BLT.

Evaluation of the DNA repair host-mediated assay. II. Presence of genotoxic factors in various organs of mice treated with chemotherapeutic agents

The DNA repair host-mediated assay was further calibrated by testing 7 chemotherapeutic agents known to possess carcinogenic activity, namely bleomycin (BLM), cis-diamminedichloroplatinum-II (cis-Pt), cyclophosphamide (CP), diethylstilboestrol (DES), isonicotinic acid hydrazide (isoniazid, INH), natulan (NAT) and mitomycin C (MMC). Differential survival of wild-type and uvrB/recA E. coli strains served as a measure of genotoxic activity. In in vitro assays, BLM, cis-Pt and MMC exhibited high genotoxic activity. The other 4 compounds had no measurable effect on the survival of the two strains, either with or without mouse liver preparations. In the host-mediated assays BLM, cis-Pt, MMC and also NAT induced strong killing of the DNA repair-deficient bacteria recovered from liver, spleen, lungs, kidneys and the blood of treated mice compared to the wild-type strain. The results are not indicative of large organ-specific differences in genotoxically active amounts of the drugs immediately after their application to the host animals. CP, INH and DES did not show geneotix activity in these assays even at very high exposure levels. To compare the genetic endpoint measured in the DNA repair assays, i.e. induction of repairable DNA damage, with the induction of gene mutations, the ability of the 7 drugs to induce valine-resistant (VALr) mutants in E. coli was measured in host-mediated assays under identical treatment conditions. INH showed considerable mutagenic activity in E. coli cells recovered from liver and spleen, while BLM and MMC induced a 3-4-fold increase in VALr mutants above spontaneous levels. The other compounds showed no mutagenic activity under these in vivo conditions. From these results it can be concluded that the type of primary DNA lesions produced by these chemotherapeutic agents (cross-links by MMC and cis-Pt, and strand breaks by BLM and possibly by NAT; base alkylation by INH) appears to determine whether a compound will be highly positive in the DNA repair assay as in the case of BLM, cis-Pt, MMC and NAT, and less effective in inducing mutations under similar conditions, or whether the opposite will occur, as in the case of INH; DES and CP probably do not interact sufficiently with bacterial DNA to show an effect in either of the genetic endpoints; and the present DNA repair host-mediated assay may represent a sensitive, rapid and economic method for monitoring genotoxic factors in various organs of experimental animals which have been treated with cytostatic drugs.

Genetic effects of isoniazid and the relationship to in vivo and in vitro biotransformation

The mutagenic activity of isoniazid, N-acetyl-isoniazid and hydrazine dihydrochloride was investigated in S. typhimurium. Isoniazid was found to possess a weak mutagenic activity only in repair-deficient strains TA1535 and TA100 as well as in the plasmid-containing strain TA92 (10-30 mg/plate) in the Ames test without metabolic activation. Addition of microsomal enzymes by S9 mix decreased this direct mutagenic activity. In contrast, preincubation of isoniazid with crude liver homogenate from mice, rats or Syrian golden hamsters for 4 h prior to plating with bacteria liberated a mutagenic compound which is equally active in both repair-deficient and repair wild-type strains (0.5-5 mg/plate). This activation pathway is independent of NADPH, is heat-sensitive and is operative only in a total liver homogenate in suspension. The highest capacity for mutagenic activation was achieved with liver homogenate from hamsters, followed by that from mice and rats. Furthermore, this mutagenic activation is paralleled by formation of hydrazine, as demonstrated in colorimetric measurements with p-dimethylaminobenzaldehyde. N-Acetyl-isoniazid is without mutagenic activity under similar conditions, and liberation of hydrazine was never detected. This means that, besides having a weak direct genetic activity, isoniazid is a promutagen, and formation of hydrazine is the first step in metabolic activation. It is concluded that the genotoxic properties of isoniazid in mammals are primarily determined by the pharmacokinetic behavior of the ultimate reactive metabolite. This result must be taken into consideration in risk assessment performed for mutagenic and carcinogenic properties of isoniazid in man.