HNE––signaling pathways leading to its elimination

Abstract

The oxidation of polyunsaturated fatty acids results in the production of HNE, which can react through both non-enzymatic and enzyme catalyzed reactions to modify a number of cellular components, including proteins and DNA. Multiple pathways for its enzyme catalyzed elimination include oxidation of the aldehyde to a carboxylic acid, reduction of the aldehyde to an alcohol, and conjugation of the carbon–carbon double bond to glutathione (GSH). Interestingly, the enzymes that result in HNE elimination are induced by HNE itself although the chemical mechanism for signaling is not well understood. One of the striking effects of HNE is that after a transient decrease in GSH, synthesis of GSH is elevated through induction of glutamate cysteine ligase (GCL), which catalyzes the first step in de novo synthesis of GSH. GCL has two subunits, which are transcriptionally regulated by a wide variety of agents, including oxidants and electrophiles, such as HNE, which elevates both. The transcriptional regulation of GCL has been the subject of many investigations yielding a complex picture in which the pathways for up-regulation of the subunits appear to be independent and vary with inducing agent and cell type. We have found that in human bronchial epithelial cells, HNE acts through AP-1 activation with signaling through the JNK pathway, and that neither the ERK nor p38^MAPK pathways is involved. With these results we review what is currently known about the signaling mechanisms for removal of HNE, focusing principally on conjugation mechanisms involving GSH.

Introduction

HNE is an α,β-unsaturated aldehyde that is formed from the reaction of reactive oxygen species with n-6 polyunsaturated fatty acids in cellular membranes during inflammation and exposure to oxidative air pollutants, such as nitrogen dioxide and ozone (Hamilton et al., 1996; Robison et al., 1995). HNE is capable of undergoing Michael addition at the carbon–carbon double bond. This reaction with glutathione (GSH) can occur non-enzymatically, but the rate is greatly facilitated by the glutathione S-transferase (GST) subclasses that have relative specificity for alkenals. Expression of these enzymes, GSTA4-4 and GST5.8, is regulated by their substrates, including HNE (Tjalkens et al., 1998; Cheng et al., 2001). Exposure of rat L2 cells (Liu et al., 1998) and human HBE1 cells (Dickinson et al., 2002) to HNE causes an increase in GSH biosynthesis. Whether HNE affects GSH synthesis by initially depleting GSH via conjugation (Hartley et al., 1995; Tjalkens et al., 1999), through direct interaction with kinases such as JNK (Parola et al., 1998) or by signaling upstream via caspase activation (Chiarpotto et al., 1999), remains controversial.

GSH is synthesized de novo through the sequential action of two ATP-dependent enzymes. The first is glutamate cysteine ligase (GCL) which is composed of two subunits, catalytic (GCLC) and modulatory (GCLM). The second enzyme is glutathione synthase (GS). The first enzyme combines glutamate and cysteine to form γ-glutamylcysteine, then glycine is added to form GSH. The two GCL subunits and the GS enzyme are encoded by separate genes (Gclc, Gclm, and Gs, respectively). Regulation of the Gcl genes in humans in response to stress is the subject of much investigation, and many laboratories have reported the induction of one or both of these genes. These genes can be differentially regulated at transcriptional, post-transcriptional and even post-translational levels (for recent reviews see Dahl and Mulcahy (2001) and Soltaninassab et al. (2000)).

Human lung epithelial cells are routinely exposed to NO₂, ozone, and small particulates; these molecules and species can damage cellular membranes, resulting in the formation of HNE. HBE1 cells are transformed human bronchial epithelial cells, which retain many of the characteristics of normal human bronchial epithelial cells.

We demonstrated that in HBE1 cells exposure to HNE resulted in both time- and dose-dependent changes in the intracellular concentration of reduced GSH, with no change in the GSSG content. At 3 h after initiation of the exposure, GSH content was either unchanged or decreased depending upon HNE concentration. At 30 μM HNE, the initial decline in GSH content was >70%. All concentrations of HNE used (5–30 μM) caused an increase in GSH during the next 9 h in a time and concentration dependent manner; maximum GSH content was observed with 10 μM HNE at 12 h. By 24 h, GSH content had begun to decline, but was still elevated versus the initial content.

An increase in the intracellular content of GSH is a balance between loss through use, primarily as a reductant in glutathione peroxidase and GST reactions, and replenishment through de novo synthesis; profound increases, such as those seen with HNE in HBE1 cells, typify increased production. As GCL is feed-back inhibited by GSH, to some extent a decrease in GSH content will result in increased activity of the pre-existing GCL, leading to a transient increase in production. Prolonged increases generally require synthesis of new GCL subunits. We determined that along with the increased intracellular GSH content following HNE exposure in HBE1 cells, the content of both GCLC and GCLM increased. Increased content of GCL subunits generally results from increased protein synthesis resulting from increased gene expression. As predicted, HNE increased the steady-state content of both Gclc and Gclm mRNA species (Dickinson et al., 2002).

Increased steady-state mRNA content is usually due to increased gene expression, although notably, rat lung epithelial cells exposed to HNE also showed increased mRNA stability (Liu et al., 1998). Gene expression is enhanced by the binding of transcription factor complexes to cis-elements in the gene’s promoter region; these regions have been cloned, sequenced, and characterized in humans (Mulcahy and Gipp, 1995; Moinova and Mulcahy, 1998). The most prominent candidates for enhancing Gclc and Gclm expression are TRE (AP-1 binding), EpRE (electrophile response element, sometimes called ARE, or antioxidant response element) and NF-κB (which is absent in Gclm) sites. In HBE1 cells we demonstrated using the electrophoretic mobility shift assay that exposure to HNE led to increased DNA binding activity to TRE sequence probes, but not EpRE or NF-κB. This data suggested that activation of AP-1 binding led to the HNE-induced increase in GSH content.

AP-1 is formed by dimerization of Jun family proteins with other Jun family members, Fos family members, or other proteins, such as ATF-2. The composition of the AP-1 dimer will affect its binding affinity, and activity. Phosphorylation of Jun family members is controlled through the JNK pathway, while synthesis of Fos family members is largely controlled through the ERK pathway. Nonetheless, all four major MAPK pathways may contribute to increased AP-1 activation depending upon cell type and stimulant. Previous work in rat L2 cells demonstrated that both the ERK pathway, and by deductive logic the JNK pathway, were involved in HNE-mediated signaling for GSH synthesis (Liu et al., 2001). Using pharmacological inhibitors of the ERK and p38^MAPK signaling pathways, we investigated and determined that neither have a role in HNE-mediated signaling for GSH synthesis in HBE1 cells. In contrast, an intracellularly delivered peptide containing a JNK binding domain of c-Jun diminished the increase in AP-1 binding activity in response to HNE, and abrogated the HNE-mediated increase in steady-state Gclc and Gclm mRNA content, consistent with a role for the JNK pathway.

HNE has been demonstrated to activate the JNK pathway (Parola et al., 1998; Camandola et al., 1997; Uchida et al., 1999), which in turn activates AP-1 through phosphorylation of Jun family members, c-Jun and JunB. We determined that HNE exposure increased the content of phosphorylated c-Jun, allowing for increased transcriptional activity of AP-1 DNA binding complexes. Moreover, this increase in phosphorylated c-Jun can only be due to increased JNK activity (the only kinase known to phosphorylate c-Jun), thus providing a direct in vivo demonstration of JNK activation by HNE in this system.

It has been demonstrated that initial depletion of cellular reduced GSH in response to HNE exposure corresponded to the production of the GSH–HNE conjugate (glutathionyl 4-hydroxynonanal, GS-HNE) (Hartley et al., 1995). Removal of HNE by conjugation with GSH via GST activity is probably the major pathway of removal in this system based on the initial decrease in GSH content upon HNE exposure; however this remains to be determined empirically. The GS-HNE conjugate can be further reduced by aldose reductase (ALR) to form glutathionyl 1,4-dihydroxynonene (GS-DHN) (Srivastava et al., 2000). The activity of ALR and the steady-state concentration of mRNA have both been shown to increase in response to HNE (Spycher et al., 1996; Spycher et al., 1997).

Oxidation of the aldehyde to form a carboxylic acid (4-hydroxy-2-nonenoic acid, HNA) has been shown to occur, and is mediated by the enzyme aldehyde dehydrogenase (ALDH). While the promoter sequence of this gene remains unknown, ALDH has been reported to be induced by HNE (Luckey and Petersen, 2001). HNA can serve as a substrate for the Cyp4A enzymes, marking it for removal from the cell. Another major pathway for removal of 4HNE is its conversion to 4-hydroxynonenol by an aldehyde reductase (Spycher et al., 1996) that is also inducible by 4HNE (Spycher et al., 1997). The relative contributions of each of these pathways in the removal of 4HNE has been reported for aortic endothelial cells (Srivastava et al., 2001) and isolated perfused rat heart (Srivastava et al., 1998), and depends on cell-type. The signaling mechanisms inducing each of these pathways remains to be critically examined in any one cell type.