The 3‘-untranslated region of the Ste20-like kinase SLK regulates SLK expression

【摘要】  Ste20-like kinase, SLK, a germinal center kinase found in kidney epithelial cells, signals to promote apoptosis. Expression of SLK mRNA and protein and kinase activity are increased during kidney development and recovery from ischemic acute renal failure. The 3'-untranslated region (3'-UTR) of SLK mRNA contains multiple adenine and uridine-rich elements, suggesting that 3'-UTR may regulate mRNA stability. This was confirmed in COS cell transient transfection studies, which showed that expression of the SLK open-reading frame plus 3'-UTR mRNA was reduced by 35% relative to the open-reading frame alone. To further characterize the SLK-3'-UTR, this nucleotide sequence was subcloned downstream of enhanced green fluorescent protein (EGFP) cDNA. In COS, 293T, and glomerular epithelial cells, expression of EGFP mRNA and protein was markedly reduced in the presence of the SLK-3'-UTR. After transfection and subsequent addition of actinomycin D, EGFP mRNA remained stable in cells for at least 6 h, whereas EGFP-SLK-3'-UTR mRNA decayed with a half-life of 4 h. A region containing five AUUUA motifs within the SLK-3'-UTR destabilized EGFP mRNA. Deletion of this region from the SLK-3'-UTR, in part, restored mRNA stability. By UV cross-linking and SDS-PAGE, the SLK-3'-UTR bound to protein(s) of 30 kDa in extracts of COS cells, glomerular epithelial cells, and kidney. Cotransfection of HuR (a RNA binding protein of 30 kDa) increased the steady-state mRNA level of EGFP-SLK-3'-UTR but not EGFP. Thus the SLK-3'-UTR may interact with kidney RNA-binding proteins to regulate expression of SLK mRNA during kidney development and after ischemic injury.

【关键词】  ischemiareperfusion injury kidney development protein kinase RNA binding proteins

KIDNEY DEVELOPMENT IS DEPENDENT on the action of growth factors, growth factor receptors, and interaction of cells with extracellular matrices ( 33 ). During development, there is proliferation of renal cells, as well as apoptosis ( 16, 31 ), and tight regulation of cell growth and apoptosis is essential for formation of normal renal anatomy and cell differentiation. In the mature kidney, interruption of blood flow leads to ischemia, a major cause of acute renal failure ( 12, 21, 35 ). Ischemia typically injures renal tubular epithelial cells and also glomerular cells and is characterized by ATP depletion and other metabolic derangements. Restoration of blood flow (reperfusion) is associated with production of reactive oxygen species. Ischemia-reperfusion may lead to apoptosis or necrosis of kidney cells; however, the cells that survive the insult may undergo a process resembling development, which includes dedifferentiation, reentry into the cell cycle, and proliferation to replace the dead cells.

The germinal center kinases (GCKs) comprise a family of protein kinases that are homologous to Ste20 of Saccharomyces cerevisiae ( 10, 17 - 19 ). The GCK family has been classified into eight groups. The group I GCKs are perhaps best understood; they interact with MEK kinase-1, and they activate the JNK pathway but not extracellular signal-regulated or p38 pathways ( 17 ). The GCKs in groups II-VIII (originally classified together as group II GCKs) are with one exception expressed ubiquitously ( 10 ), and some of these GCKs can be activated in vivo by various stresses, e.g., heat shock, arsenite, staurosporine, ischemic injury, or ATP depletion ( 25, 26 ). Most of the group II-VIII GCKs do not fit into the well-defined MAPK pathways. These kinases do not activate ERK, JNK, or p38 pathways, although there are a few exceptions ( 10, 14, 29, 30 ).

During screening of rat developing kidney mRNA for protein tyrosine kinases ( 15 ), we discovered an mRNA consistent with the group V GCK: Ste20-like kinase, SLK (or SK2 in the rat). Further studies showed that expression of SLK mRNA and protein and kinase activity were increased in rat fetal kidney homogenates, compared with adult control kidneys ( 9 ). Moreover, in adult kidneys subjected to ischemia-reperfusion injury, SLK mRNA and protein expression and kinase activity were increased, compared with untreated contralateral control kidneys. SLK was localized mainly in the cytoplasm of tubular epithelial cells in fetal and adult kidneys, and there was also some expression in developing and mature glomerular epithelial cells (GECs, or podocytes). Thus SLK is a renal epithelial protein kinase whose expression and activity are increased during development and recovery from acute renal failure, where injured tubular epithelial cells may regenerate by recapitulating developmental processes ( 12 ). Localization in GECs also suggests a possible role in glomerulogenesis or glomerular injury.

We and others have begun to characterize regulatory and functional aspects of SLK. Incubation of cultured kidney epithelial cells with serum, a source of growth factors, increased SLK activity ( 9 ). In cultured cells, chemical anoxia and reexposure to glucose (anoxia/recovery) recapitulates the ATP depletion and production of reactive oxygen species seen during ischemia-reperfusion in vivo. Exposure of cultured kidney epithelial cells to anoxia/recovery stimulated SLK activity ( 9 ). Overexpression of SLK in epithelial cells attenuated increases in cell number under normal culture conditions, as a result of apoptosis, and in the setting of anoxia-exacerbated cell death ( 9, 14 ). In other cell lines, transient overexpression of SLK also induced apoptosis ( 29, 30 ), but a recent study has also shown a requirement for SLK for progression through the cell cycle ( 24 ). Epidermal growth factor, anisomycin, and hyperosmolality did not stimulate SLK activity in transfected COS-7 cells, and transfection of constitutively active Ras, Rac, or cell division cycle-42 also did not enhance SLK activity ( 36 ).

Studies addressing the regulatory mechanisms of SLK activation have shown that SLK underwent dimerization via the COOH-terminal domain and that dimerization enhanced SLK activity ( 14 ). Phosphorylation and dephosphorylation of SLK were associated with changes in SLK activity in some but not all studies ( 13, 14, 29, 30 ). In kidney epithelial cells, SLK induced phosphorylation of apoptosis signal-regulating kinase-1 (ASK1) and increased ASK1 activity, indicating that ASK1 is a substrate of SLK. Moreover, SLK stimulated phosphorylation of p38 MAPK via ASK1 but not JNK or ERK. The proapoptotic actions of SLK involved release of cytochrome c and stimulation of caspase-9, and apoptosis was reduced significantly with p38 and caspase-9 inhibitors. Finally, attenuation of the protective aspects of the endoplasmic reticulum stress response by SLK may contribute to its proapoptotic effect ( 14 ).

Studies performed so far have revealed certain aspects of the regulation and signaling by GCKs, but much of their regulation is still poorly understood ( 10, 17 ). By analogy to some non-GCKs ( 4, 23, 32, 37, 38 ), our results suggest that SLK may be, in part, regulated via changes in mRNA and/or protein expression. An increase in expression may then facilitate oligomerization and enhance kinase activity. In support of this view, the 3'-untranslated region (3'-UTR) of SLK ( 29, 36 ) contains multiple adenine and uridine-rich elements (AREs) that may regulate mRNA stability ( 1, 3 ). Thus it is reasonable to propose that, because of the presence of multiple AREs, SLK mRNA may be unstable and that stabilization of mRNA may increase expression. We demonstrate that SLK-3'-UTR can destabilize the mRNA of SLK and of an enhanced green fluorescence protein (EGFP) reporter. A substantial amount, but not all of the destabilizing activity, was found in a single region of five AUUUA motifs.

MATERIALS AND METHODS

Materials. Tissue culture reagents were obtained from Invitrogen Life Technologies (Burlington, ON). Electrophoresis and immunoblotting reagents were from Bio-Rad Laboratories (Mississauga, ON). [ - 32 P]dCTP (3,000 Ci/mmol) was purchased from Perkin-Elmer Canada (Woodbridge, ON). [ - 32 P]CTP (800 Ci/mmol) was from Amersham Biosciences (Baie d?Urfé, QC). The pEGFP-C2 vector was from BD Biosciences (Mississauga, ON). Pwo DNA polymerase and Fugene-6 transfection reagent were purchased from Roche Diagnostics (Laval, QC). Restriction enzymes and other molecular biology reagents were from Invitrogen, New England Biolabs (Mississauga, ON), or Ambion, (Austin, TX). Mouse anti-GFP antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse anti-FLAG antibody was from Sigma-Aldrich Canada (Mississauga, ON). The cDNA of full-length human SLK plus the 3'-UTR was kindly provided by Dr. T. Nagase (Kazusa DNA Research Institute). The HuR-C-FLAG cDNA expression construct (pcDNA3-HuR-C-FLAG) was a generous gift from Dr. I. Gallouzi (McGill University).

Plasmid construction. The SLK open-reading frame (ORFs; nucleotides 233-3844) ( 36 ) and the 3'-UTR (nucleotides 3845-5794) or the SLK ORF alone were subcloned into the expression vector pcDNA3.1/Myc-His(-), as described previously ( 14 ) ( Fig. 1; constructs 1 and 2 ). Full-length or fragments of the SLK-3'-UTR were subcloned into the pEGFP-C2 vector ( Fig. 1; construct 3 ) at the 3' end of EGFP cDNA. Full-length SLK-3'-UTR consisted of a 2.1-kbp cDNA, starting at nucleotide 3628 (near the 3' end of the SLK ORF), and containing the stop codon of SLK plus the entire SLK-3'-UTR ( Fig. 1; construct 4 ). Regions of the SLK-3'-UTR containing various AREs (see Fig. 4 A, below) were amplified by PCR (Pwo polymerase) ( 8, 15 ), using primers terminating in restriction sites. The PCR products were then digested with restriction enzymes and subcloned into pEGFP-C2 ( Fig. 1; construct 5 ). The nucleotide numbers and PCR primer sequences are listed in Table 1. To construct SLK-3'-UTR- (see Fig. 5 A, below), a 1.7-kbp DNA fragment (nucleotides 3628-5303 of SLK-3'-UTR) was subcloned into the pEGFP-C2 vector. The polyadenylation tails of bovine growth hormone or simian virus 40 ( Fig. 1 ) were components of the cloning vectors.

Fig. 1. cDNA constructs (see MATERIALS AND METHODS for details). BGH, bovine growth hormone; EGFP, enhanced green fluorescent protein; ORF, open-reading frame; pA, poly A; P CMV, cytomegalovirus promoter; SV40, simian virus 40; 3'-UTR, 3'-untranslated region.

Fig. 2. 3'-UTR of SLK destabilizes SLK mRNA. A : COS cells were transiently transfected with cDNAs encoding the SLK ORF ( construct 1 in Fig. 1 ) or the SLK ORF plus the full-length 3'-UTR. Cells were cotransfected with EGFP. Expression of SLK mRNA was measured by real-time RT-PCR (at 48 h) and was normalized for EGFP mRNA. * P < 0.025 SLK vs. SLK-3'-UTR (3 experiments). B - E : SLK-3'-UTR destabilizes EGFP mRNA and protein expression. The full-length SLK-3'-UTR was cloned into the pEGFP-C2 vector at the 3' end of EGFP cDNA (EGFP-SLK-3'-UTR). COS cells, 293T cells, and glomerular epithelial cells (GECs) were transiently transfected with pEGFP-SLK-3'-UTR or with pEGFP (empty vector). After 48 h, expression of EGFP mRNA was monitored by Northern blotting ( B, top; duplicate lanes are presented for COS cells). B, bottom : ethidium bromide-stained RNA, showing equal loading. Note that EGFP-SLK-3'-UTR mRNA is greater in size than EGFP. EGFP protein expression was assessed by immunoblotting with anti-GFP antibody ( C ) or fluorescence microscopy ( D ). The lower molecular weight band in 293T cells (EGFP lane) is probably nonspecific. E : COS cells were cotransfected with a cDNA encoding cytosolic phospholipase A 2 (cPLA 2 ) and pEGFP-SLK-3'-UTR or pEGFP (empty vector). Transfection of pEGFP-SLK-3'-UTR had no effect on the expression of cPLA 2 (representative immunoblots).

Fig. 3. SLK-3'-UTR accelerates decay of EGFP mRNA. Multiple plates of 293T cells were transiently transfected with pEGFP-SLK-3'-UTR or pEGFP (empty vector). After 48 h, actinomycin D (10 µg/ml) was added, and RNA was extracted over the subsequent 6 h. Expression of EGFP mRNA was monitored by Northern blotting. A : representative blot. B : densitometric quantification of 3 experiments.

Fig. 4. Mapping of SLK adenine and uridine-rich elements (AREs) that destabilize EGFP mRNA. A : 4 EGFP cDNA constructs, each containing 2-5 AUUUA motifs from SLK-3'-UTR, were assembled with PCR (3'-UTR1-4). Each AUUUA is denoted by an asterisk. COS cells were transiently transfected with these cDNAs, as well as with those encoding EGFP plus the entire SLK-3'-UTR (3'-UTR-C) or EGFP alone (empty vector). RNA and protein were extracted after 48 h, and expression of EGFP was monitored by Northern blotting ( B and C ) or immunoblotting ( D and E ). C and E : representative blots (lanes in C are in duplicate). B and D : densitometric quantification of 4 and 3 experiments, respectively. In constructs containing either the full SLK-3'-UTR (3'-UTR-C) or the distal region with five AUUUAs (3'-UTR-2), levels of EGFP mRNA and protein were reduced significantly relative to EGFP alone (empty). In B, * P < 0.001, ** P < 0.005. In D, * P < 0.0001, ** P < 0.05.

Table 1. Regions of the SLK-3'-UTR containing AUUUA motifs (also see Fig. 4A )

Fig. 5. A region of the SLK-3'-UTR reduces EGFP protein expression markedly but has a modest affect on mRNA. A portion of the full-length SLK-3'-UTR (3'-UTR-C) ranging from just 5' of the five AUUUA motifs to the 3' end was deleted to produce SLK-3'-UTR- ( A ). COS cells were transiently transfected with cDNAs encoding EGFP-SLK-3'-UTR, EGFP-SLK-3'-UTR-C, or EGFP alone (empty vector). RNA and protein were extracted after 48 h, and expression of EGFP was monitored by Northern blotting ( B and C ) or immunoblotting ( D and E; lanes on gels are in duplicate). mRNA expression was lower in EGFP-SLK-3'-UTR- transfections than in EGFP (empty vector) ( C : * P < 0.02) but greater than in EGFP-SLK-3'-UTR-C (** P < 0.003, 4 experiments). In contrast, protein expression was markedly reduced in both EGFP-SLK-3'-UTR and EGFP-SLK-3'-UTR-C transfections, compared with EGFP (empty vector). In E, * P < 0.005 (3 experiments).

Cell culture and transfection. Monkey kidney COS-1 and human embryonic kidney 293T cells were cultured in DMEM-10% FCS. Rat GEC culture and characterization have been published previously ( 5, 8 ). GECs were cultured in K1 medium, and studies were done with cells between passages 8 and 60. Cells were transiently transfected with plasmid DNAs by the Fugene-6 reagent, according to the manufacturer?s instructions ( 8 ). Generally, 1 x 10 5 cells in 35-mm wells were transfected with 1 µg of plasmid DNA. Transfection efficiency 50%, in keeping with results provided by the manufacturer of the Fugene-6 transfection reagent.

Northern blotting and immunoblotting. All studies involving animals were approved by the McGill University Animal Care Committee. Preparation of RNA or protein from rat kidneys and cultured cells was described previously ( 9 ). To prepare the probe for Northern blotting, a cDNA encoding the ORF of EGFP was excised with appropriate restriction enzymes and was labeled with [ - 32 P]dCTP, as described previously ( 8, 9 ). RNA gel electrophoresis and hybridization were performed as described previously ( 8, 9 ). Methods for immunoblotting were published previously ( 8, 9 ). Densitometry of multiple Northern blots or immunoblots was performed with the use of NIH Image software. Preliminary studies demonstrated that there was a linear relationship between densitometric measurements and the amounts of protein loaded onto gels.

Quantification of SLK mRNA by real-time RT-PCR. COS cells were transiently transfected with cDNAs encoding the SLK ORF ( construct 1 in Fig. 1 ) or the SLK ORK plus the full-length 3'-UTR ( construct 2 in Fig. 1 ). Cells were cotransfected with a cDNA encoding EGFP. RNA was extracted 48 h after transfection, and reverse transcription into single-stranded cDNA was performed with random primers (Invitrogen) and the Powerscript kit (Clontech, Palo Alto, CA). Expression was measured by real-time RT-PCR. PCR primers for SLK (forward, 5'-CAGCAGCAGAAGGCAGAGTT-3'; reverse, 5'-CGTTCTTCGAGCTCCCAAATT-3') and for EGFP (forward, 5'-GTCCAGGAGCGAACCATCT-3'; reverse, 5'-ATGCCCTTCAGCTCGATGC-3') were designed with the use of Primer Express software (Applied Biosystems). Real-time PCR reactions were performed using the default PCR cycle on an ABI Prism 7900 HT sequence detection system (Applied Biosystems), and amplified DNA was detected by SYBR green incorporation. Dissociation curve analyses were performed to confirm specificity of the SYBR green signals in each experiment. Quantification of relative amounts of SLK was carried out by Sequence Detection Systems Software 2.0 (Applied Biosystems) and the comparative standard curve method. SLK values were normalized to EGFP mRNA levels in the same sample.

Characterization of RNA binding proteins by UV cross-linking and SDS-PAGE. To prepare a template for in vitro RNA transcription, a 500-bp DNA fragment covering the 3' end of the SLK-3'-UTR and containing five AUUUA motifs was subcloned into the pBluescript SK + vector adjacent to the T7 promoter ( Fig. 6 A, below). The template was linearized with the Not I restriction enzyme. Radiolabeled and "cold" (unlabeled) probes were synthesized by using an in vitro transcription kit from Ambion (MAXIscript), including T7 RNA polymerase and [ - 32 P]CTP, according to the manufacturer?s instructions. Unincorporated nucleotides were removed by Chroma Spin-100 columns (BD Biosciences). The radiolabeled probe (1.5 x 10 5 cpm) was incubated with cytosolic extracts ( 9 ) of kidneys or cells (30 µg protein) for 20 min at 22°C in buffer containing 100 mM HEPES, 15 mM MgCl 2, 200 mM KCl, 5 mM DTT, 5 mg/ml heparin, and 25% glycerol. To verify the specificity of the binding, a 100-fold excess of "cold" probe was included in the reaction (competitor). Free, unbound 32 P-labeled RNA was removed by incubating the mixture with RNase T1 (100 U, 15 min, 22°C) and RNase A (2 µg, 30 min, 37°C). Samples were then subjected to UV cross-linking. Protein-RNA complexes were resolved by SDS-PAGE and subjected to autoradiography ( 6, 11 ).

Fig. 6. Identification of a SLK-3'-UTR binding protein by UV cross-linking and SDS-PAGE. A radiolabeled RNA probe ( A ), which contained 500 nucleotides of the 3' end of the SLK-3'-UTR (including 5 AUUUAs), was incubated with cytosolic extracts for 20 min at 22°C. To verify the specificity of the binding, a 100-fold excess of "cold" probe was included in the reaction (competitor). After digestion of free probe, samples were then subjected to UV cross-linking, and protein-RNA complexes were resolved by SDS-PAGE. Representative autoradiograms are presented. The probe bound to a protein of 30 kDa (arrows) in cytosolic extracts of GECs ( B ), COS cells ( C ), and embryonic day 21 fetal and adult kidney ( D ).

Statistics. Data are presented as means ± SE. The t -test statistic was used to determine significant differences between two groups. One-way ANOVA was used to determine significant differences among groups. Where significant differences were found, individual comparisons were made between groups using the t -test and adjusting the critical value according to the Bonferroni method.

RESULTS

The 3'-UTR of SLK destabilizes SLK mRNA. In an earlier study ( 9 ), our group demonstrated that levels of SLK mRNA were increased in developing rat kidney on embryonic days 17, 19, and 21, compared with adult control kidneys. Furthermore, SLK mRNA was increased in adult rat kidneys after unilateral ischemia (induced by clamping of the renal artery for 30 min) and after 12, 24, or 48 h of reperfusion, compared with contralateral control kidneys. In addition to these increases in mRNA, there were analogous increases in SLK protein expression and kinase activity in rat fetal kidneys and in adult kidneys after unilateral ischemia-reperfusion injury, compared with contralateral control kidneys or sham-operated controls ( 9 ). Results of these studies suggested that SLK activity may be, at least in part, regulated via changes in mRNA and protein expression. Analysis of the 3'-UTR of human SLK revealed that the 3'-UTR contains multiple AREs, including at least nine AUUUA motifs ( 36 ), which might regulate mRNA stability ( 1, 3 ). Thus it was reasonable to propose that SLK mRNA may be generally unstable and that the stabilization of mRNA would increase expression. To test this hypothesis, we first carried out transient transfection studies in COS cells, which showed that expression of the SLK ORF plus 3'-UTR mRNA was reduced by 35% relative to the ORF alone ( Fig. 2 A ).

To further characterize the effect of the SLK-3'-UTR on mRNA stability, the full-length 3'-UTR was cloned into the pEGFP-C2 vector at the 3' end of EGFP cDNA. COS cells, 293T cells, and GECs were then transiently transfected with pEGFP-SLK-3'-UTR or with pEGFP (empty vector). The advantage of this approach is that both constructs are under control of the same promoter, which facilitates sufficient expression levels. The EGFP reporter is not endogenously expressed in cells, and the regulatory properties of SLK-3'-UTR are considered in isolation from the full-length SLK. EGFP mRNA and/or protein were evident mainly in the transfection with pEGFP (empty vector), although small amounts of mRNA and protein were detected in some EGFP-SLK-3'-UTR transfections ( Fig. 2, B - D ). (Because of the very low transfection efficiency of GECs, we were able to examine only levels of EGFP protein.) This result suggests that the SLK-3'-UTR destabilizes EGFP mRNA and prevents protein expression and that the effect is similar in multiple cell lines. Cotransfection of pEGFP-SLK-3'-UTR with a cDNA encoding cytosolic phospholipase A 2 had no effect on the expression of the cytosolic phospholipase A 2 protein ( Fig. 2 E ), confirming that the effect of the SLK-3'-UTR was specific to EGFP and not due to a broad reduction in mRNA or protein expression.

To confirm that EGFP mRNA levels actually reflect altered stability, mRNA was measured after treatment of 293T cells 48 h posttransfection with actinomycin D, which arrests de novo mRNA transcription ( 7, 11 ). Thus the level of remaining mRNA at different time points after transcriptional block is an indicator of the mRNA stability. RNA was quantified at serial time points to determine the half-life. SLK-3'-UTR accelerated decay of EGFP mRNA with a half-life of 4 h, whereas in the absence of the 3'-UTR, EGFP mRNA was very stable ( Fig. 3 ). Because of the marked destabilizing effect of the 3'-UTR, we were not able to achieve comparable starting levels of mRNAs (i.e., before decay was monitored), and EGFP mRNA was present in excess, compared with EGFP-SLK-3'-UTR. Possibly, decay of EGFP may emerge after a number of hours, but we were not able to continue the experiment beyond 6 h because actinomycin D becomes cytotoxic at later time points. Thus one has to be cautious with the interpretation of comparative half-lives.

Characterization of the AREs that destabilize EGFP mRNA. Four cDNAs containing EGFP plus regions of the SLK-3'-UTR, each containing two to five AUUUA motifs, were assembled by PCR ( Fig. 4 A; see MATERIALS AND METHODS ). COS cells were transiently transfected with these cDNAs, as well as with cDNAs encoding EGFP plus the full-length SLK-3'-UTR ( Fig. 4 A, construct 3'-UTR-C) or EGFP (empty vector). A single region of five AUUUAs ( Fig. 4 A, construct 3'-UTR-2) destabilized EGFP mRNA and induced an analogous effect on the expression of EGFP protein ( Fig. 4, B - E ). Although this destabilizing effect on mRNA was substantial, the effect of the full-length SLK-3'-UTR appeared to be greater ( Fig. 4 B ). Subgroups of two or three AUUUAs from this same region ( Fig. 4 A, constructs 3'-UTR-3 and -4) were, however, not effective in reducing EGFP mRNA and protein ( Fig. 4, B - E ). Furthermore, another region of the 3'-UTR containing three AUUUAs ( Fig. 4 A, construct 3'-UTR-1) did not show a significant destabilizing effect ( Fig. 4, B - E ). Interestingly, although these three AUUUAs had no effect on mRNA and protein levels, the EGFP protein appeared to be slightly smaller in size ( Fig. 4 E ). Further studies will be required to determine the significance of this observation.

Because the construct SLK-3'-UTR-2 (containing five AUUUAs) destabilized EGFP mRNA significantly, it may be expected that deletion of this region from the full-length SLK-3'-UTR would abrogate the destabilizing effect. EGFP mRNA expression ( Fig. 5, B and C ) was significantly greater with the EGFP-SLK-3'-UTR deletion construct ( Fig. 5 A, construct 3'-UTR- ) than with the EGFP-SLK full-length 3'-UTR ( Fig. 5A, construct 3'-UTR-C). Nevertheless, EGFP-SLK-3'-UTR- mRNA expression was somewhat lower compared with EGFP (empty vector) ( Fig. 5, B and C ). This result suggests that, although the five AUUUAs in SLK-3'-UTR-2 have a robust effect on destabilizing mRNA, other regions of the 3'-UTR appear to have additive effects. In contrast to mRNA, protein expression was markedly reduced in both EGFP-SLK-3'-UTR- and EGFP-SLK full-length 3'-UTR-C transfections, compared with EGFP (empty vector) ( Fig. 5, D and E ).

An 30-kDa protein binds to the SLK-3'-UTR. 3'-UTRs can mediate their regulatory effects through direct interaction of AREs with various ARE-binding proteins (AREBPs) ( 1, 3 ). To determine whether proteins could bind to the SLK-3'-UTR, we employed a radiolabeled RNA probe, which contained five AUUUA motifs near the 3' end of the SLK-3'-UTR ( Fig. 6 A ). Using UV cross-linking and SDS-PAGE, we determined that this region of the SLK-3'-UTR bound to protein(s) of 30 kDa in cytosolic extracts of GECs, COS cells, and fetal and adult kidneys ( Fig. 6, B - D ). Specificity of the binding was verified by showing that a 100-fold excess of "cold" probe displaced the bound probe significantly ( Fig. 6, B - D ).

The size of the SLK-3'-UTR AREBP ( Fig. 6, B - D ) was in keeping with the size of HuR, a protein that has been shown to bind a number of RNAs, and is expressed ubiquitously ( 1, 3 ). To determine whether HuR can interact with the SLK-3'-UTR, we cotransfected COS cells with EGFP or EGFP-SLK-3'-UTR, with or without FLAG-HuR cDNAs. On SDS-PAGE, FLAG-HuR ( Fig. 7 A ) migrated at a similar position as the AREBP in Fig. 6. As expected, transfection of FLAG-HuR had no effect on EGFP mRNA. In contrast, HuR increased the expression of EGFP-SLK-3'-UTR mRNA ( Fig. 7, A and B ), implying that binding of HuR stabilized the EGFP-SLK-3'-UTR mRNA. To verify the cellular localization of FLAG-HuR, we isolated nuclei from COS cells and immunoblotted cell fractions with antibodies to FLAG and to other proteins to verify the purity of the preparations ( 20 ). FLAG-HuR was distributed in nuclear fractions and in postnuclear supernatants, which primarily contain cytosol ( Fig. 7 C ).

Fig. 7. HuR stabilizes EGFP-SLK-3'-UTR mRNA. COS cells were transfected with EGFP alone or EGFP-SLK-3'-UTR cDNA, together with FLAG-HuR cDNA or irrelevant control plasmid (Ctrl). RNA or protein was extracted after 48 h, and expression of EGFP was monitored by Northern blotting. FLAG-HuR expression was monitored by immunoblotting ( A, bottom ). Transfection of FLAG-HuR had no effect on EGFP mRNA but increased the expression of EGFP-SLK-3'-UTR mRNA ( A : representative Northern blot; B : densitometric quantification of 3 experiments, * P < 0.01 HuR vs. Ctrl EGFP-SLK-3'-UTR and P < 0.01 EGFP-SLK-3'-UTR+HuR vs. EGFP + Ctrl). C : FLAG-HuR is found in the nuclear fraction (Nuc) and postnuclear supernatant (Sup) of transfected COS cells. Cyclooxygenase-1 (COX-1), a protein associated with the nuclear envelope, is detected only in the nuclear fraction. The p42 and p44 isoforms of ERK (predominantly a cytosolic protein in resting cells) are present mainly in the postnuclear supernatant, which is primarily cytosol (representative immunoblots).

DISCUSSION

SLK is a renal epithelial protein kinase, whose expression and activity are increased during development, as well as during recovery from acute renal failure, where injured tubular epithelial cells may regenerate by recapitulating developmental processes ( 9 ). Earlier studies demonstrated that the action of SLK in renal cells is proapoptotic ( 9, 14 ). Understanding the regulation of protein kinase activity associated with cell proliferation and apoptosis is important because tight regulation of cell growth and apoptosis is essential for formation of normal renal anatomy and cell differentiation in both kidney development and recovery from injury. Analysis of the 3'-UTR of SLK in both humans and mice revealed that both 3'-UTRs are extensive, with long stretches of the cDNA sequence being 80-90% homologous, and the 3'-UTRs contain AREs ( 29, 36 ). The observed changes in SLK mRNA in the kidney, together with the structural features of the SLK-3'-UTR, suggested that SLK activity may be regulated via mRNA stability ( 1, 3 ). The present study demonstrates that expression of the SLK ORF plus 3'-UTR mRNA was reduced significantly, compared with the ORF alone ( Fig. 2 ).

To further characterize the role of the SLK-3'-UTR in regulating mRNA stability, we employed an EGFP reporter with and without the full-length 3'-UTR fused to the 3' end of EGFP cDNA ( 7, 11 ). EGFP is a protein that is not expressed endogenously; however, after transfection, levels are very stable in cultured cells ( 2 ). Thus this model facilitates sufficient expression levels of EGFP mRNA and/or protein reporter, and the regulatory properties of the SLK-3'-UTR are considered in isolation from the full-length SLK. Our results show that the SLK-3'-UTR induced a marked reduction in the expression of EGFP mRNA and/or protein in multiple kidney cell types ( Fig. 2 ). Experiments that addressed mRNA levels after treatment with actinomycin D (which arrests de novo mRNA transcription) confirmed that SLK-3'-UTR destabilizes mRNA ( Fig. 3 ). To our knowledge, this is the first example that the 3'-UTR of a MAPK family member can regulate mRNA stability. It should be noted that, although SLK-3'-UTR reduced the level of SLK mRNA by 35% ( Fig. 2A ), the effect of the SLK-3'-UTR on EGFP mRNA appeared to be substantially greater, i.e., 60-70% ( Figs. 4 B and 5 C ). This result suggests that aspects of the SLK ORF in addition to the 3'-UTR may be involved in regulation of mRNA levels. Further study will be required to characterize these features. Alternatively, COS cells also contain endogenous SLK mRNA, which may have reduced the precision when comparing levels of mRNA originating from transfected cDNAs.

Because the full-length SLK-3'-UTR had a marked destabilizing effect on EGFP mRNA, studies were carried out to define the relevant region(s) of the 3'-UTR more precisely. We focused on AREs made up of AUUUA motifs, which have been associated with binding of proteins that modulate mRNA stability ( 1, 3 ). Most of the destabilizing activity was found within a region of the 3'-UTR consisting of <200 nucleotides, which contained five AUUUA motifs ( Fig. 4 ). Deletion of this region from the full-length SLK-3'-UTR stabilized EGFP mRNA significantly, but not entirely, suggesting that, although this region of five AUUUAs has a marked destabilizing effect, other regions of the 3'-UTR have additional effects. Actually, the SLK-3'-UTR contains another region of 135 nucleotides, containing three AUUUA motifs. This region did not destabilize EGFP mRNA independently ( Fig. 4 ), but perhaps it can take on a more important role when the region of five AUUUA motifs is deleted. Alternatively, AREs that do not contain AUUUA motifs can also be involved in modulation of mRNA levels ( 1 ).

Because the region of the SLK-3'-UTR containing five AUUUA motifs was important in regulating mRNA stability, this region of RNA was used to determine whether putative RNA binding proteins were present in cultured cell and kidney extracts. In an in vitro assay, a protein or proteins of 30 kDa in cytosolic extracts of cells and kidneys bound to the SLK-3'-UTR ( Fig. 6 ). The size of the AREBP(s) was in keeping with the size of HuR ( Fig. 7 ). Transfection of EGFP-SLK-3'-UTR, together with FLAG-HuR cDNA, increased the expression of EGFP-SLK-3'-UTR mRNA, whereas transfection of EGFP together with FLAG-HuR had no effect on EGFP mRNA ( Fig. 7 ). This result is consistent with the view that HuR can interact with the SLK-3'-UTR and stabilize the EGFP-SLK-3'-UTR mRNA. HuR is a protein that has been shown to bind a substantial number of RNAs and is expressed ubiquitously ( 1 ). Generally, HuR binding has been reported to stabilize mRNAs, and our result is therefore consistent with these earlier observations. HuR is able to interact with AUUUA motifs, as well as other types of AREs. Furthermore, ischemia has been reported to promote binding of HuR to the 3'-UTR of the vascular endothelial growth factor and increase its expression ( 34 ). However, because the SLK-3'-UTR reduced EGFP expression in cultured cell lines, most likely there must have been other AREBPs that bound to the SLK-3'-UTR in these cells and facilitated degradation of EGFP mRNA. For example, multiple AREBPs have been shown to interact with the 3'-UTR of cyclooxygenase-2, some of which stabilize and others that degrade mRNA ( 6 ). Thus one needs to consider the possibility that the band identified by UV cross-linking and SDS-PAGE ( Fig. 6 ) may be composed of more than one AREBP, each of approximately similar size. For example, AREBPs of similar size to HuR, but with mRNA destabilizing properties, could include certain isoforms of AUF1, CBF-A, hnRNP A1, AU-A, and AU-B ( 1, 3, 6 ). Alternatively, the RNA probe that was utilized for the binding studies in vitro may not have been capable of identifying all AREBPs relevant to the full-length SLK-3'-UTR. Further studies involving additional RNA probes, cell lines, and subcellular fractions will be required to identify other putative AREBPs. It should also be noted that the enhanced expression of SLK mRNA observed in the developing kidney and in kidneys postischemic injury is not reliably reproduced by stimulating cultured cells with growth factors or subjecting cells to chemical anoxia and glucose reexposure. Probably, the cultured cell lines employed so far are missing certain factors that are present in vivo. Further studies will be required to develop appropriate models to study the interactions of the SLK-3'-UTR with AREBPs.

The effects of the SLK-3'-UTR on changes in EGFP protein generally paralleled effects on mRNA, suggesting that changes in protein levels were a direct consequence of changes in mRNA. Thus the full-length SLK-3'-UTR reduced EGFP mRNA and protein markedly, and the region of five AUUUA motifs also had a significant reducing effect on both ( Figs. 2 and 4 ). Interestingly, deletion of this region from the full-length SLK-3'-UTR (SLK-3'-UTR- ), although stabilizing EGFP mRNA, did not enhance protein expression ( Fig. 5 ). In certain mRNAs, the 3'-UTRs are involved in the regulation of translation ( 22 ), and our result suggests that SLK-3'-UTR- may have reduced translation of EGFP. The functional relevance of this observation remains to be determined, as deletion of a portion of the SLK-3'-UTR may not be a physiologically relevant situation. Nevertheless, one can speculate that, under certain circumstances, a RNA binding protein may bind to the 3' end of SLK-3'-UTR and perhaps block the functional effects of this region, thereby unmasking regulatory effects mediated by the 5' end of the SLK-3'-UTR. It should also be noted that the 135-nucleotide region of the SLK-3'-UTR containing three AUUUA motifs appeared to induce expression of a shorter form of EGFP protein, even though the total amount of protein was not affected ( Fig. 4 E ). This result may be due to a change in the EGFP translation start site, supporting a role for SLK-3'-UTR in regulating aspects of translation independently of mRNA stability. One cannot exclude the possibility that the shorter form of EGFP represents a degradation product, but this seems unlikely because the shorter protein was not found in transfections of any other region of SLK-3'-UTR, and all incubations were carried out under the same experimental conditions ( Fig. 4 E ).

On the basis of the present study, it is reasonable to propose that changes in SLK protein expression are generally secondary to levels of SLK mRNA. Our group ( 14 ) has shown previously that SLK protein may homodimerize via the COOH-terminal domain and that homodimerization may enhance kinase activity ( 14 ). Increased levels of SLK protein may facilitate homodimerization, thereby leading to increased activity and enhanced downstream proapoptotic signaling in the developing kidney or after ischemic injury. Regulation of kinase activity via changes in mRNA or protein levels has been reported for some other proteins kinases ( 4, 23, 32, 37, 38 ). These changes in expression have been due to changes in transcription or protein degradation, and in some cases the mechanisms have not been defined. The regulation of SLK activity is complex, and changes in kinase activity can also occur without changes in expression. For example, in some but not all experimental systems, an increase in SLK phosphorylation was reported to be associated with an increase in activity ( 13, 14, 29, 30 ). Similar to other GCKs, SLK protein contains four potential Pro/Glu/Ser/Thr-rich (PEST) sequences ( 36 ). The PEST motif is related to rapid degradation of proteins ( 27, 28 ); consequently, kinase activity may be downregulated, in part by a short half-life due to the PEST domain. Further studies will be required to address the complex regulation of SLK activity and its role in renal pathophysiology.

GRANTS

This work was supported by research grants from the Canadian Institutes of Health Research and the Kidney Foundation of Canada. A. V. Cybulsky and T. Takano hold scholarships from the Fonds de la Recherche en Santé du Québec.

ACKNOWLEDGMENTS

We thank Drs. Moshe Szyf (McGill University) and Phil Marsden (University of Toronto) for helpful discussion and Mian Chen and Julie Guillemette for expert technical assistance.

【参考文献】
  Barreau C, Paillard L, Osborne HB. AU-rich elements and associated factors: are there unifying principles? Nucleic Acids Res 33: 7138-7150, 2005.

Bence NF, Sampat RM, Kopito RR. Impairment of the ubiquitin-proteasome system by protein aggregation. Science 292: 1552-1555, 2001.

Bevilacqua A, Ceriani MC, Capaccioli S, Nicolin A. Posttranscriptional regulation of gene expression by degradation of messenger RNAs. J Cell Physiol 195: 356-372, 2003.

Buse P, Tran SH, Luther E, Phu PT, Aponte GW, Firestone GL. Cell cycle and hormonal control of nuclear-cytoplasmic localization of the serum- and glucocorticoid-inducible protein kinase, Sgk, in mammary tumor cells. A novel convergence point of anti-proliferative and proliferative cell signaling pathways. J Biol Chem 274: 7253-7263, 1999.

Coers W, Reivinen J, Miettinen A, Huitema S, Vos JT, Salant DJ, Weening JJ. Characterization of a rat glomerular visceral epithelial cell line. Exp Nephrol 4: 184-192, 1996.

Cok SJ, Acton SJ, Sexton AE, Morrison AR. Identification of RNA-binding proteins in RAW 264.7 cells that recognize a lipopolysaccharide-responsive element in the 3'-untranslated region of the murine cyclooxygenase-2 mRNA. J Biol Chem 279: 8196-8205, 2004.

Cok SJ, Morrison AR. The 3'-untranslated region of murine cyclooxygenase-2 contains multiple regulatory elements that alter message stability and translational efficiency. J Biol Chem 276: 23179-23185, 2001.

Cybulsky AV, Takano T, Papillon J, Bijian K. Role of the endoplasmic reticulum unfolded protein response in glomerular epithelial cell injury. J Biol Chem 280: 24396-24403, 2005.

Cybulsky AV, Takano T, Papillon J, Khadir A, Bijian K, Chien CC, Alpers CE, Rabb H. Renal expression and activity of the germinal center kinase SK2. Am J Physiol Renal Physiol 286: F16-F25, 2004.

Dan I, Watanabe NM, Kusumi A. The Ste20 group kinases as regulators of MAP kinase cascades. Trends Cell Biol 11: 220-230, 2001.

Detich N, Ramchandani S, Szyf M. A conserved 3'-untranslated element mediates growth regulation of DNA methyltransferase 1 and inhibits its transforming activity. J Biol Chem 276: 24881-24890, 2001.

Devarajan P. Update on mechanisms of ischemic acute kidney injury. J Am Soc Nephrol 17: 1503-1520, 2006.

Ellinger-Ziegelbauer H, Karasuyama H, Yamada E, Tsujikawa K, Todokoro K, Nishida E. Ste20-like kinase (SLK), a regulatory kinase for polo-like kinase (Plk) during the G2/M transition in somatic cells. Genes Cells 5: 491-498, 2000.

Hao W, Takano T, Guillemette J, Papillon J, Ren G, Cybulsky AV. Induction of apoptosis by the Ste20-like kinase SLK, a germinal center kinase that activates apoptosis signal-regulating kinase and p38. J Biol Chem 281: 3075-3084, 2006.

Kee N, McTavish AJ, Papillon J, Cybulsky AV. Receptor protein tyrosine kinases in perinatal developing rat kidney. Kidney Int 52: 309-317, 1997.

Koseki C, Herzlinger D, Al-Awqati Q. Apoptosis in metanephric development. J Cell Biol 119: 1327-1333, 1992.

Kyriakis JM. Signaling by the germinal center kinase family of protein kinases. J Biol Chem 274: 5259-5262, 1999.

Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol Rev 81: 807-869, 2001.

Kyriakis JM, Avruch J. Protein kinase cascades activated by stress and inflammatory cytokines. Bioessays 18: 567-577, 1996.

Liu J, Takano T, Papillon J, Khadir A, Cybulsky AV. Cytosolic phospholipase A2-alpha associates with plasma membrane, endoplasmic reticulum and nuclear membrane in glomerular epithelial cells. Biochem J 353: 79-90, 2001.

Liu S, Humes HD. Cellular and molecular aspects of renal repair in acute renal failure. Curr Opin Nephrol Hypertens 2: 618-624, 1993.

Mazumder B, Seshadri V, Fox PL. Translational control by the 3'-UTR: the ends specify the means. Trends Biochem Sci 28: 91-98, 2003.

Monfar M, Blenis J. Inhibition of p70/p85 S6 kinase activities in T cells by dexamethasone. Mol Endocrinol 10: 1107-1115, 1996.

O?Reilly PG, Wagner S, Franks DJ, Cailliau K, Browaeys E, Dissous C, Sabourin LA. The Ste20-like kinase SLK is required for cell cycle progression through G2. J Biol Chem 280: 42383-42390, 2005.

Pombo CM, Bonventre JV, Molnar A, Kyriakis J, Force T. Activation of a human Ste20-like kinase by oxidant stress defines a novel stress response pathway. EMBO J 15: 4537-4546, 1996.

Pombo CM, Tsujita T, Kyriakis JM, Bonventre JV, Force T. Activation of the Ste20-like oxidant stress response kinase-1 during the initial stages of chemical anoxia-induced necrotic cell death. Requirement for dual inputs of oxidant stress and increased cytosolic [Ca 2+ ]. J Biol Chem 272: 29372-29379, 1997.

Rechsteiner M, Rogers SW. PEST sequences and regulation by proteolysis. Trends Biochem Sci 21: 267-271, 1996.

Rogers S, Wells R, Rechsteiner M. Amino acid sequences common to rapidly degraded proteins: the PEST hypothesis. Science 234: 364-368, 1986.

Sabourin LA, Rudnicki MA. Induction of apoptosis by SLK, a Ste20-related kinase. Oncogene 18: 7566-7575, 1999.

Sabourin LA, Tamai K, Seale P, Wagner J, Rudnicki MA. Caspase 3 cleavage of the Ste20-related kinase SLK releases and activates an apoptosis-inducing kinase domain and an actin-disassembling region. Mol Cell Biol 20: 684-696, 2000.

Savill J. Regulation of glomerular cell number by apoptosis. Kidney Int 56: 1216-1222, 1999.

Sivaraman VS, Wang H, Nuovo GJ, Malbon CC. Hyperexpression of mitogen-activated protein kinase in human breast cancer. J Clin Invest 99: 1478-1483, 1997.

Sorokin L, Ekblom P. Development of tubular and glomerular cells of the kidney. Kidney Int 41: 657-664, 1992.

Tang K, Breen EC, Wagner PD. Hu protein R-mediated posttranscriptional regulation of VEGF expression in rat gastrocnemius muscle. Am J Physiol Heart Circ Physiol 283: H1497-H1504, 2002.

Thadhani R, Pascual M, Bonventre JV. Acute renal failure. N Engl J Med 334: 1448-1460, 1996.

Yamada E, Tsujikawa K, Itoh S, Kameda Y, Kohama Y, Yamamoto H. Molecular cloning and characterization of a novel human STE20-like kinase, hSLK. Biochim Biophys Acta 1495: 250-262, 2000.

Yeagley D, Moll J, Vinson CA, Quinn PG. Characterization of elements mediating regulation of phosphoenolpyruvate carboxykinase gene transcription by protein kinase A and insulin. Identification of a distinct complex formed in cells that mediate insulin inhibition. J Biol Chem 275: 17814-17820, 2000.

Zhou R, Snyder PM. Nedd4-2 phosphorylation induces serum and glucocorticoid-regulated kinase (SGK) ubiquitination and degradation. J Biol Chem 280: 4518-4523, 2005.

作者单位：1 Department of Medicine, McGill University Health Centre, Montreal, Quebec, and 2 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto General Research Institute, Toronto, Ontario, Canada