p38 MAPK Signalling Is Not Sufficient to Induce γ-Globin Gene Expression

Abstract 1083

Several recent publications have implicated p38 MAPK signalling in the pharmacologic induction of γ-globin gene expression. These conclusions are primarily based on demonstrations of increased p38 phosphorylation (P-p38) by inducing compounds and inhibition of induction by the p38 inhibitor SB203580 (SB). Based on these studies and work from our own lab, we proposed that p38 signalling might be the key mechanism underlying γ-globin gene induction by most active agents. We have performed a series of experiments in K562 cells designed to critically test this hypothesis and to better define the role of p38 signalling in γ-globin gene induction. We first performed dose-response experiments with sodium butyrate (NaB), an agent that induces γ-globin mRNA in vitro and in vivo and has been proposed to work through p38 signalling. If p38 signalling is inducing γ-globin gene expression, then p38 phosphorylation should correlate with increased γ-globin mRNA levels. We found 3–3.5-fold induction of steady-state γ-globin mRNA at doses ranging from 0.25 to 2mM NaB. SB suppressed induction at all dose levels. However, P-p38 was increased only at doses below 1mM and was actually suppressed compared to untreated controls at 2mM NaB. We next examined the effects of heat shock and UV and X-irradiation, all well-known activators of p38 signalling. Here too we found specific stress “doses” that did not increase P-p38 but still strongly induced γ-globin mRNA and were inhibited by SB. These results demonstrated a lack of correlation between p38 phosphorylation and γ-globin induction but for every experimental condition tested, SB inhibits basal or induced γ-globin mRNA levels. To more closely examine this discrepancy, we transiently expressed a constitutively active MKK6 mutant (MKK6-Glu; the direct upstream activator of p38), to mimic transient increases in P-p38 seen with drugs and stresses. While MKK6-Glu expression elevated p38 phosphorylation to levels seen with NaB and physical stresses, it did not increase basal γ-globin mRNA levels. Furthermore, increased p38 activation by MKK6-Glu did not enhance NaB-induced γ-globin expression. We next used p38 siRNA in combination with NaB and heat shock conditions that induce γ-globin mRNA. This allowed us to decrease downstream p38 signalling (as judged by Hsp27 phosphorylation) to levels equivalent to those achieved with SB. While SB still completely inhibited induction, there was only a partial decrease in NaB induction and no effect on heat shock induction with the p38 knockdown. Our results show that p38 phosphorylation does not strictly correlate with γ-globin induction suggesting that it is not required for at least some degree of γ-globin induction by NaB and stress. The fact that we saw no γ-globin mRNA induction with targeted activation of p38 by MKK6-Glu suggests that p38 activation is not sufficient for induction. Our results also suggest that SB inhibits γ-globin mRNA induction by a mechanism other than direct p38 inhibition. Upon investigating potential mechanisms, we observed SB to promote ERK phosphorylation, suggesting that a modulation of ERK signalling or p38/ERK cross-talk may be involved. Additionally, we find that SB treatment appears to decrease steady-state mRNA levels but not nascent γ-globin mRNA, suggesting a role in transcript processing or stability.