The article highlighted in this issue is “Vomitoxin-Induced Cyclooxygenase-2 Gene Expression in Macrophages Mediated by Activation of ERK and p38 but Not JNK Mitogen-Activated Protein Kinases,” by Yuseok Moon and James J. Pestka (pp. 373–382).
It is increasingly apparent that many xenobiotics function by modulating growth factor-mediated signal transduction pathways. Thus, in many instances chemical-induced toxicity may be the result of disrupting cell signaling. Inhibiting intracellular signaling modifies, in turn, processes such as cell survival, growth and differentiation. Furthermore, xenobiotics can also activate these signaling pathways. Recent research has focused on signaling via the mitogen-activated protein (MAP) kinase pathways as targets for chemical toxicity. These studies have revealed that the MAP kinases, which include the extracellular signal-regulated kinase (ERK) 1 and 2, and the stress-activated protein kinases including c-jun NH2-terminal kinase (SAPK1/JNK) and p38 kinase (SAPK2), are activated by many xenobiotics. Several of these toxins, which include trichothecene mycotoxins, ultraviolet light, certain tumor promoters, antibiotics, and pathogen-and plant-derived toxins, effectively target the ribosome, and through a process referred to as “ribotoxic stress,” modulate MAP kinase activity. These xenobiotics appear to act by altering the peptidyltransferase activity of the 28S rRNA in actively translating eukaryotic ribosomes. Elucidating the mechanisms by which ribosomal dysfunction mediates alterations in MAP kinase activity represents a major challenge critical for determining the mechanism of action of xenobiotics that act as ribotoxic stressors.
In response to chemical toxicity, cells in target tissues undergo numerous changes including alterations in the expression of genes associated with proliferation and differentiation, and/or apoptosis. These processes are controlled by cues received from circulating factors such as growth factors and cytokines as well as the extracellular matrix. The subsequent induction of signaling is known to occur via MAP kinase cascades including the activity of ERK1/2, and the stress-activated protein kinases JNK and p38 kinase (for reviews see Chang and Karin, 2001; Enslen and Davis, 2001; van Drogen and Peter, 2002). Indeed, MAP kinase signaling is activated following exposure to diverse xenobiotics including ultraviolet light, acetaminophen, ozone, carbon tetrachloride and endotoxin (Mendelson et al., 1996; Bae et al., 2001; Laskin and Laskin, 2001; Sunil et al., 2002). These toxicants can stimulate MAP kinase signaling by directly modulating a broad spectrum of physiological stimuli including the extracellular matrix and/or appropriate growth factors and cytokines. For example, TNF-α and interleukin-1, mediators that are induced by these xenobiotics, are known to be potent activators of MAP kinase signaling. Alternatively, MAP kinase signaling in cells can be activated by various forms of environmental stress including heat shock and hyperosmotic stress, as well as oxidative and nitrosative stress (Heck, 2001; Martindale and Holbrook, 2002). Carbon tetrachloride is an example of a toxin that induces oxidative stress in the liver, a process that may be directly responsible for activation of MAP kinase signaling (Mendelson et al., 1996).
Ribotoxic Stress as a Mechanism Regulating the Activation of MAP Kinases
An emerging concept important for understanding the mechanism by which various xenobiotics activate MAP kinase signaling is the ribotoxic stress response (Iordanov et al., 1997). This activity is a cellular reaction to toxicants that perturb the functioning of the 3′-end of the large 28S ribosomal RNA. During protein synthesis, this region of the ribosome functions in aminoacyl-tRNA binding, peptidyltransferase activity, and ribosomal translocation (Uptain et al., 1997). Depending on the cell type, toxicant-induced disruption of this activity results in activation of JNK and p38 kinase and/or alterations in ERK1/2 signaling (Iordanov et al., 1997, 1998, 2002). In most cases, active ribosomes appear to be required as mediators of this signaling response and many of the inducers of the ribotoxic stress response at least partially inhibit protein synthesis.
Initiation of the ribotoxic stress response has been described for a number of different classes of xenobiotics which appear to have in common the ability to bind to or damage a specific region at the 3′-end of the 28S ribosomal RNA (Iordanov et al., 1997). Initially characterized for anisomycin, a potent activator of JNK that inhibits peptidyltransferase activity, this response is also observed in cells treated with aminohexose pyrimidine nucleoside antibiotics. These agents bind to the same region of the 28S ribosomal RNA. Additional xenobiotics including ricin A chain and α-sarcin, which damage specific sites in a conserved loop (the R/S loop) in the 28S ribosomal RNA, induce a ribotoxic stress response (Iordanov et al., 1997). Ricin A chain causes a depurination in the R/S loop while α-sarcin causes RNA cleavage (Iordanov et al., 1997). A similar robotoxic stress is observed in cells following treatment with E. coli-derived Shiga toxin, a toxin which also possesses 28S ribosomal RNA N-glycosidase activity (Kojima et al., 2000). MAP kinase activation by these ribotoxic stressors was suppressed by inhibitors that disrupt the functional activity of ribosomes including emetine which arrests cells in the elongation cycle (Iordanov et al., 1997). This has led to the hypothesis that, at the time of exposure to agents that induce the ribotoxic stress response, functionally active ribosomes are required (Iordanov et al., 1997) although this model has recently been questioned (see further below).
A variety of trichothecene mycotoxins have been reported to be effective triggers of the ribotoxic stress response including T2-triol, nivalenol and scirpentriol (Shifrin and Anderson, 1999). These xenobiotics also bind to the 28S ribosomal RNA peptidlytransferase site and activate JNK and p38 kinase (Shifrin and Anderson, 1999). However, not all trichothecenes activating the MAP kinases are effective inhibitors of protein synthesis suggesting that this is not a requirement for initiating ribotoxic stress. Indeed, Shifrin and Anderson (1999) suggested unique structural elements of trichothecenes could be responsible for initiating translational arrest and JNK/p38 activation. This may be due to distinct sites on the ribosome for these functional activities. These investigators also provided evidence to suggest that the ribotoxic stress response does not always require active translation. This was based on the findings that translation inhibitors which are effective in blocking JNK/p38 kinase activation induced by ribotoxic stressors such as anisomycin, failed to inhibit activation induced by T2-triol. Moreover, anisomycin was found to be active in the presence of trichothecenes that are effective inhibitors of peptidyltransferase activity but do not induce the ribotoxic stress response. Clearly, further studies are required to better define the precise requirements for initiating ribotoxic stress by xenobiotics.
Of interest are recent reports demonstrating that ribotoxic stress may also underlie activation of MAP kinases by ultraviolet light radiation (Iordanov et al., 1998, 2002; Iordanov and Magun, 1999). In many cell types including keratinocytes, ultraviolet light, in particular, UVC (200–290 nm) and UVB (290–320 nm) rapidly activate JNK and p38 MAP kinase. As observed with many other ribotoxic stressors, this response is associated with inhibition of protein synthesis and is dependent on the presence of active ribosomes during the ultraviolet light exposure. Thus, emetine, an inhibitor of translation, abrogates ultraviolet light-induced activation of the stress kinases. Importantly, both UVC and UVB light are able to cause nucleotide- and site-specific damage to the 3′-end of 28S ribosomal RNA, an effect expected to impair peptidyl transferase activity (Iordonov et al., 1998). Thus, similar to pathogen- and plant-derived ribotoxic stressors, ultraviolet light is also able to damage a critical region in the ribosomal RNA that is important for initiating MAP kinase activation. It should be noted that in many cell types ultraviolet light also activates ERK1/2. This appears to occur by a process that does not utilize the ribotoxic stress response (Iordonov et al., 1998). However, it has recently been reported that primary cultures of human keratinocytes consitutively express elevated levels of ERK1/2 kinase and transiently down-regulate this enzyme activity following UVB light exposure (Iordonov et al., 2002). This was found to be due to ribotoxic stress-induced uncoupling of ras from ERK1/2 in these cells. These data indicate that ribotoxic stress can alter ERK1/2 signaling cascades.
The tumor promoter palytoxin, a non-peptide marine toxin also appears to function as a ribotoxic stressor, but acts by a distinct mechanism not involving direct binding to ribosomes. However, like other ribotoxic stressors, palytoxin stimulates the JNK and p38 kinase while inhibiting protein synthesis in a process requiring actively functioning ribosomes at the time of exposure to the tumor promoter (Iordanov and Magun, 1998). Palytoxin appears to induce ribotoxic stress by augmenting potassium efflux from cells, an effect mediated by toxin binding to the Na+/K+ATPase in the plasma membrane (Iordanov and Magun, 1998). Presumably, the 3′-end of the 28S ribosomal RNA possesses a potassium-sensitive site where lowered intracellular concentrations of the cation perturb its functional activity and thus mimic direct acting agents capable of inducing a ribotoxic stress response. This may be due to alterations in the protein binding or conformational changes in the ribosome.
It is well recognized that the expression of specific genes can occur as a consequence of MAP kinase activation. The most notable example is the transcriptional activation of the immediate early genes including c-fos and c-jun. Many additional genes induced by ribotoxic stressors have been identified using microarray techniques (Rolli-Derkinderen and Gaestel, 2000) and differential display reverse transcription-PCR (Kojima et al., 2000). Of particular interest is the finding that verotoxin (Shiga toxin) effectively induces mitogen-activated protein kinase phosphatase 1, a dual-specificity MAP kinase phosphatase (Kojima et al., 2000). This enzyme can inactivate ERK1/2 as well as JNK and p38 kinase (Enslen and Davis, 2001). Kojima et al. (2000) suggested that this may act as a feedback loop, balancing the increase in activity of the MAP kinases and potentially limiting the responses of cells to ribotoxic stressors. This type of complex feedback network may be important in preventing toxicity and/or initiating repair.
As indicated above, a variety of trichothecene mycotoxins function by inducing ribotoxic stress. In this issue, Moon and Pestka (2002) report that vomitoxin, a common trichthecene agricultural contaminant produced by Fusarium graminearum and Fusarium culmorum, is a potent inducer of JNK and p38 kinase as well as ERK1/2 in a mouse macrophage cell line. Activation of p38 and JNK is consistent with ribotoxic stress. As described above, in response to ultraviolet light, depending on the cell type, cellular ribotoxic stress can also modulate ERK1/2 signaling. This may also be the case for vomitoxin, although it can not be ruled out that this toxin activates ERK1/2 by a ribotoxic-independent mechanism. Moon and Pestka (2002) also showed that, as a consequence of MAP kinase activation, vomitoxin induced expression of cyclooxygenase-2 (COX-2), the rate limiting enzyme in the inducible synthesis of prostaglandin endoperoxides. These lipid mediators are critical components of the inflammatory response and have been found to mediate toxicity. In these studies, increased expression of COX-2 protein was due to increased synthesis of COX-2 mRNA as well as stabilization of the message. Both processes are regulated by MAP kinase activity. Although induction of COX-2 appears to be mediated by activation of ERK1/2 and p38, but not JNK, the ribotoxic stress response appeared to be a likely mediator of this process.
Several ribosome-directed toxicants have the capacity to damage 28S ribosomal RNA and/or interfere with its functioning, thus compromising protein synthesis. This can lead to what has been called “ribotoxic stress,” a response that stimulates MAP kinase signaling. This process can lead to the expression of genes important in cellular homeostasis as well as in the control of cell survival, proliferation and differentiation. Ribotoxic stress should be considered a potential mechanism of toxicity with chemical agents that have the capacity to inhibit protein synthesis as well as stimulate MAP kinase signaling. That the ribotoxic stress pathway is a highly specific event is indicated by the finding that not all protein synthesis inhibitors have the capacity to activate the MAP kinase pathways. Ribotoxic stressors appear to be restricted to those toxicants that interact with or damage the R/S loop near the 3′-end of the 28S rRNA. A major question remains as to the precise linkage between ribosomal RNA damage and induction of MAP kinase signaling cascades. Iordanov et al. (1997) hypothesized that there must be as yet unidentified intermediate signal transduction steps, presumably mediated by proteins, between toxicant-damaged 28S rRNA and the MAP kinases. Since active ribosomes are required, it was suggested that this activity was restricted to selected stages of the ribosomal cycle. A number of proteins bind to the site where toxicant damage occurs in the ribosome, in particular, various elongation factors; some of these, when modified, can suppress MAP kinase activation. It is not yet known whether the elongation factors themselves, or other related ribosomal binding factors, interact with the MAP kinase signaling cascades to mediate the toxicant-induced ribotoxic stress response.