Comment on “Increased MKK4 Abundance with Replicative Senescence Is Linked to the Joint Reduction of Multiple MicroRNAs”

Science Signaling  20 Jul 2010:
Vol. 3, Issue 131, pp. lc1
DOI: 10.1126/scisignal.3131lc1


Marasa et al. (Research Article, 27 October 2009, DOI: 10.1126/scisignal.2000442) reported that the human kinase p38–regulated/activated protein kinase (PRAK) was phosphorylated on residue Ser93 in senescent cells. We have been unable to detect phosphorylation at this site with the antibody that they used, and the commercial supplier of this antibody has discontinued its availability, which casts doubt on whether this residue of PRAK is phosphorylated.

The classical mammalian mitogen-activated protein kinase (MAPK) pathways operate through three successive phosphorylation events mediated by a MAPK kinase kinase (MAPKKK), a MAPK kinase (MAPKK), and a MAPK. The latter phosphorylates various substrates, including other protein kinases. MAPK signaling pathways are involved in many cellular processes, including senescence. MKK4 is a MAPKK that can activate its downstream substrates, the MAPK family members JNK and p38 (1). Marasa et al. showed that MKK4 abundance is controlled by the action of four microRNAs (miR-15b, miR-24, miR25, and miR-141) and that the abundance of these miRNAs decreases during replicative senescence and coincides with increased MKK4 abundance (2). The MKK4 substrate p38 and its downstream target p38–regulated/activated protein kinase (PRAK, which is known in mouse as MK5), are essential for Ras-induced senescence (3, 4). Therefore, in senescent human cells the authors monitored phosphorylation of p38 at Thr180/Tyr182 and PRAK at Ser93 with phosphorylation site-specific antibodies. Phosphorylation of PRAK was detected with a commercially available antibody (Santa Cruz Biotechnology, Santa Cruz, California; #sc-130212). The authors reported increased phosphorylation of PRAK in senescent human fetal lung diploid fibroblast WI-38 cells, whereas they found that PRAK phosphorylation was reduced in WI-38 cells transfected with the microRNAs that targeted MKK4 as compared with that of cells transfected with control small interfering RNA. From their analysis, Marasa et al. concluded that the MKK4-p38-PRAK signaling pathway participated in senescence.

In vitro studies had previously shown that p38 and two other MAPKs, JNK and ERK2, phosphorylated and activated MK5 and PRAK (5, 6), with p38MAPK, JNK2, and ERK2 phosphorylating Thr182 of human PRAK and p38 weakly phosphorylating Ser212. In vitro analysis of a Ser93-to-Ala mutant PRAK demonstrated that this residue is not a phosphoacceptor site for p38, JNK2, or ERK2 and that this mutation did not inhibit the basal or p38-induced in vitro kinase activity of PRAK (5). Because of this evidence indicating that Ser93 in PRAK is not a substrate for MAPK phosphorylation, we decided to test the commercially available antibody used by Marasa et al. (Santa Cruz Biotechnology phosphoSer93 PRAK antibody), which was sold as an antibody that recognizes phosphoSer93 in MK5 or PRAK.

Human PRAK is reported to be a ~54 kD protein (5), and the sequence surrounding Ser93 in mouse and human PRAK is conserved (Fig. 1). Thus, we would expect that the phosphoSer93 PRAK antibody would recognize Ser93-phosphorylated forms of either species. We transfected human embryonic kidney (HEK) 293 cells with wild-type MK5, a Ser93→Ala (S93A) MK5 mutant, which cannot be phosphorylated at this site, or a Ser93→Asp (S93D) MK5 mutant, which mimics the phosphorylated form, tagged with enhanced green fluorescence protein (EGFP) and Western blotted the cell lysate with this antibody (Fig. 2A). We were unable to detect immunoreactivity with the Ser93-phosphorylation site-specific antibody at the expected position of endogenous PRAK (~54 kD) or at the position of the EGFP-MK5 fusion protein (~81 kD).

Fig. 1

Alignment of human PRAK and the mouse homolog MK5. Ser93 is indicated by shading.

Fig. 2

Testing of the Santa Cruz Biotechnology phosphoSer93 PRAK antibody. (A) HEK293 cells were transfected with EGFP fused to the indicated forms of MK5, and proteins were isolated and analyzed with either the Santa Cruz Biotechnology phosphoSer93 PRAK antibody (top) or with an antibody that recognizes PRAK but that is not phosphorylation–site specific (Santa Cruz Biotechnology; A-7, #sc-46667) (bottom). The arrows indicate the expected position of endogenous PRAK (54 kD) and the EGFP-MK5 fusion proteins (81 kD). M, protein molecular mass marker (in kilodaltons). (B) HEK293 cells were transfected with EGFP-MK5 and either constitutively active JNK1, JNK2, MKK4, or MKK6. Cells were also either transfected with p38 or treated with sodium arsenite (250 μM for 30 min). Phosphorylation of Ser93 PRAK (top) and PRAK and EGFP-MK5 abundance (bottom) were monitored as described in (A).

Because it was possible that, in these proliferating cells and under conditions in which MAPK signaling is not stimulated, PRAK may not be phosphorylated, we also tested whether this Ser93-phosphorylation site-specific antibody would detect a protein at the expected molecular weight for PRAK (or the MK5 fusion proteins) under conditions in which JNK or p38 signaling was stimulated. Thus, we cotransfected HEK293 cells with EGFP-MK5 and either constitutively active JNK1 (7), constitutively active JNK2 (7), MKK4 (8), or constitutively active MKK6 and p38 (9). In addition, we treated one set of cells transfected with EGFP-MK5 with sodium arsenite, an activator of the p38 pathway, which induces phosphorylation of PRAK at Thr182 (5). We were unable to detect signals corresponding to the phospho-Ser93 forms of endogenous PRAK (~54 kD) or ectopically expressed EGFP-MK5 (~81 kD) (Fig. 2B). Therefore, neither increased MKK4 abundance, nor a stimulus known to activate MKK4 (sodium arsenite), nor overexpression of constitutively active forms of the protein kinases downstream of MKK4 (JNK and p38) resulted in detectable phosphorylation of MK5 at Ser93.

Taken together, our results call into question the credibility of the phosphoSer93 PRAK antibody. When we contacted Santa Cruz Biotechnology, they informed us that they removed this product from their product list.


View Abstract

Navigate This Article