A gene expression signature for oxidant stress/reactive metabolites in rat liver

Abstract

Formation of free radicals and other reactive molecules is responsible for the adverse effects produced by a number of hepatotoxic compounds. cDNA microarray technology was used to compare transcriptional profiles elicited by training and testing sets of 15 oxidant stressors/reactive metabolite treatments to those produced by approximately 85 other paradigm compounds (mostly hepatotoxicants) to determine a shared signature profile for oxidant stress-associated hepatotoxicity. Initially, 100 genes were chosen that responded significantly different to oxidant stressors/reactive metabolites (OS/RM) compared to other samples in the database, then a 25-gene subset was selected by multivariate analysis. Many of the selected genes (e.g., aflatoxin aldehyde reductase, diaphorase, epoxide hydrolase, heme oxgenase and several glutathione transferases) are well-characterized oxidant stress/Nrf-2-responsive genes. Less than 10 other compounds co-cluster with our training and testing set compounds and these are known to generate OS/RMs as part of their mechanisms of toxicity. Using OS/RM signature gene sets, compounds previously associated with macrophage activation formed a distinct cluster separate from OS/RM and other compounds. A 69-gene set was chosen to maximally separate compounds in control, macrophage activator, peroxisome proliferator and OS/RM classes. The ease with which these ‘oxidative stressor’ classes can be separated indicates a role for microarray technology in early prediction and classification of hepatotoxicants. The ability to rapidly screen the oxidant stress potential of compounds may aid in avoidance of some idiosyncratic drug reactions as well as overtly toxic compounds.

Introduction

Oxidative stress refers to the generation of reactive oxygen species (ROS, such as superoxide, hydroxy radicals and hydrogen peroxide) and to the adaptive cellular responses to reactive oxygen species (e.g., depletion of anti-oxidant stores and activation and induction of protective and repair enzymes) [1]. Several groups have shown that most reactive electrophiles, including many hepatotoxic drug metabolites, produce effects similar, if not identical, to reactive oxygen species [2], [3]. Cells appear capable of handling low doses of OS/RMs, while higher doses overwhelm protective capacity and lead to damage or death of the cells.

A common property of ROS and reactive compounds/metabolites involves inducing the expression of protective enzymes. A common site in the promoters of these oxidant stress-responsive genes has been termed the anti-oxidant response element (ARE) or electrophilic response element (EpRE) [4]. Induction of oxidant stress-responsive genes is largely dependent on binding of the transcription factor Nrf-2 to the ARE/EpRE. Under resting conditions Nrf-2 appears to be sequestered by sulfhydryl bonding to Keapl; oxidation of the sulfhydryl bonds (and activation of certain kinases) releases Nrf-2 to bind ARE/EpRE [4]. Although the binding of many other transcription factors to promoter response elements is affected by cellular redox status and may contribute to gene expression, Nrf-2 binding to ARE/EpRE appears critical for induction of well-characterized oxidant stress-responsive genes, such as superoxide dismutase, catalase, glutathione synthases and transferases, and NADPH:quinone oxidoreductases [5]. Reactive metabolites formed in the liver are generally strong inducers of phase II/conjugation enzymes (glutathione transferases, UDP glucuronyltransferases, sulfotransferases), which comprise a subset of the Nrf-2-inducible genes.

The relevance of oxidative stress in drug safety evaluation of prospective pharmaceuticals is unclear. While the present study focused on high doses of hepatotoxicants, there is increasing evidence that low doses of toxicants particularly oxidant stressors, may have beneficial adaptive or hormetic effects. Mild oxidant stressors that induce protective enzymes but minimal cell damage have been touted as chemopreventive (anticancer) agents [6], [7]. Effects of sulforaphane and 1,2-dithiole-3-thione, two chemopreventive compounds derived from broccoli, have been studied in Nrf-2-knockout and control mice using microarray technology [8], [9]. A comprehensive list of oxidant stress- and Nrf-2-responsive genes has been compiled for mouse intestine [8] and mouse liver [9]. However, compounds that induce oxidative stress or give rise to reactive metabolites frequently are not overtly toxic, particularly in the rat.

Conversely, many innocuous or protective Nrf-2 activating compounds, such as many anti-oxidants, abruptly become hepatotoxic at high doses [10], [11], [12]. Compounds that potently induce oxidant stress-responsive genes may not be good candidates for development into therapeutic drugs, since these compounds or their metabolites are clearly reactive inside cells. Oral administration of such compounds often leads to oxidative stress and covalent binding to proteins in the gut and liver after absorption. There is evidence that some human populations may be deficient at handling oxidative stress/reactive metabolites [13], [14], [15], [16].

In the present study, a training/testing set of OS/RMs, cDNA microarrays and commercially available clustering and gene selection algorithms were used to determine a transcriptional profile for OS/RMs. A goal of our toxicogenomics program is to be able to detect as many classes of toxicants as possible before their pathologies are manifest and prior to gene changes becoming secondary to the pathology; hence, the focus was on changes that occur 24 h after administration of compound. Using a broad coverage of compounds and a ‘guilt by association’ approach, the OS/RM potential of approximately 100 toxicants was characterized in rat liver.

A number of our prototypical compounds are well-characterized oxidant stressors or are documented to give rise to reactive molecules and metabolites. Reactive electrophilic compounds or metabolites that conjugate with and deplete cellular glutathione levels comprise many of the best characterized hepatotoxicants, such as bromobenzene [2], [17], precocene I [18], pulegone [19] and hexachlorocyclohexane γ (lindane, [20]). Redox cycling and generation of reactive oxygen species have been postulated to explain the toxicities of NSAIDs [21], [22], estrogens [23], [24], troglitazone [25], aniline (in the spleen, [26]), paraquat (in the lung, [27]) and doxorubicin (in the heart, [28]). Phase II enzyme induction, in particular glutathione transferase induction, has been presented as a hallmark of OS/RMs, such as the hepatotoxins 4-methylthiazole [29] and trans-anethole [30]. Even many relatively innocuous compounds, which possess protectant anti-oxidant activity at low doses, become strong oxidant stressors/reactive electrophiles (OS/REs) and hepatotoxicants at high doses; these compounds include butylated hydroxytoluene [31], tannic acid [32], disulfiram (or rather its reactive metabolite diethyldithiocarbamate) [33] and piperonyl butoxide [34]. Additionally, OS/REs with planar structures can intercalate into DNA, and if the cells survive the resultant damage, carcinogenesis can occur. Redox cycling and reactive electrophiles appear central to the effects of carcinogens and antineoplastic drugs [35]. Macrophage and neutrophil activators, such as LPS, zymosan and concanavalin A are often viewed as profound oxidative stressors [36], [37]; however, cytokines (such as TNFa) and the extra-parenchymal derived superoxide, nitric oxide and peroxynitrate result in hepatic gene expression responses distinct from those produced by intracellular generated reactive molecules [38]. Similarly, peroxisome proliferators also induce a specialized type of oxidative stress that is easily distinguished from oxidative stress caused directly by reactive compounds or their metabolites [25], [38], [39].