MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xiu, M.
Right arrow Articles by Yee, A. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xiu, M.
Right arrow Articles by Yee, A. S.
Molecular and Cellular Biology, December 2003, p. 8890-8901, Vol. 23, No. 23
0270-7306/03/$08.00+0     DOI: 10.1128/MCB.23.23.8890-8901.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

The Transcriptional Repressor HBP1 Is a Target of the p38 Mitogen-Activated Protein Kinase Pathway in Cell Cycle Regulation

Mei Xiu,1,2 Jiyoung Kim,1,2 Ellen Sampson,1 Chun-Yin Huang,1,2 Roger J. Davis,3 K. Eric Paulson,1,2 and Amy S. Yee1*

Department of Biochemistry, School of Medicine,1 School of Nutrition Science and Policy, Tufts University, Boston, Massachusetts 02111,2 Howard Hughes Medical Institute and Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, Massachusetts 016053

Received 9 May 2003/ Returned for modification 2 June 2003/ Accepted 26 August 2003


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The p38 mitogen-activated protein (MAP) kinase signaling pathway participates in both apoptosis and G1 arrest. In contrast to the established role in apoptosis, the documented induction of G1 arrest by activation of the p38 MAP kinase pathway has attracted recent attention with reports of substrates that are linked to cell cycle regulation. Here, we identify the high-mobility group box protein HBP1 transcriptional repressor as a new substrate for p38 MAP kinase. Our previous work had shown that HBP1 inhibits G1 progression in cell and animal models, and thus indicated that HBP1 could be a relevant substrate for p38 MAP kinase in cell cycle regulation. In the present work, a p38 MAP kinase docking site (amino acids [aa] 81 to 125) and a p38 MAP kinase phosphorylation site (serine 401) were identified in the HBP1 protein. Furthermore, the docking and phosphorylation sites on HBP1 were specific for p38 MAP kinase. In defining the role of p38 MAP kinase regulation, the inhibition of p38 MAP kinase activity was shown to decrease HBP1 protein levels by triggering protein instability, as manifested by a decrease in protein half-life. Consistently, a decrease in protein levels was accompanied by a decrease in overall DNA binding activity. A mutation of the p38 MAP kinase phosphorylation site at aa 401 [(S-A)401HBP1] also triggered HBP1 protein instability. While protein stability was compromised by mutation, the specific activities of (S-A)401HBP1 and of wild-type HBP1 appeared comparable for transcriptional repression. This comparison of transcription-specific activity highlighted that p38 MAP kinase regulated HBP1 protein levels but not the intrinsic activity for DNA binding or for transcriptional repression. Finally, p38 MAP kinase-mediated regulation of the HBP1 protein also contributed to the regulation of G1 progression. Together, our work supports a molecular framework in which p38 MAP kinase activity contributes to cell cycle inhibition by increasing HBP1 and other G1 inhibitory factors by regulating protein stability.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The mitogen-activated protein (MAP) kinase cascades transduce many extracellular signals to regulate diverse intracellular processes, including differentiation, survival, and growth (for reviews, see references 4, 21, and 33). The physiological function of the stress-activated MAP kinase pathways JNK and p38 has been a subject of extensive investigation. Many studies have clearly shown that activation of the JNK and/or p38 MAP kinase pathways was both necessary and sufficient for stress-induced apoptosis (reviewed in reference 12). For example, the role of neuron-specific JNK3 in excitatory stimuli-induced apoptosis was confirmed by gene knockout, in which JNK3-/- mice have reduced neuronal apoptosis in response to a glutamate receptor agonist (35). Similarly, analysis of cytochrome c-mediated apoptotic signals demonstrated a requirement for JNK (29). Finally, knockout of the JNK scaffold protein JIP1 reduced excitatory stimuli-induced neuronal apoptosis (32). However, other evidence suggests that activation of the stress-activated pathways may not always lead to apoptosis (8, 9, 17, 20). In fact, in some cases JNK activity is required for survival (7). Thus, the full function of the stress-activated pathways is still unclear and likely to be dependent on the specific signal and cellular context.

The role of the stress-activated p38 MAP kinase pathway in nonapoptotic contexts is not well understood but is clearly important for diverse effects in cellular regulation. The role of p38 MAP kinase activity in the context of cell cycle progression and inhibition has been documented (6, 19, 30), but the mechanisms remain unclear. An early study demonstrated that activated p38 MAP kinase inhibits the cyclin D1 promoter (14). Recent studies have highlighted intriguing substrates for p38 MAP kinase in cell cycle arrest. The phosphorylation of the cyclin D1 protein by p38 MAP kinase targets the protein for ubiquitin-dependent degradation, contributing to a G1 arrest (3). In an alternative mechanism for G1 arrest, p38 MAP kinase phosphorylates the G1 inhibitor p21CIP, with phosphorylation stabilizing the p21CIP protein (10). Another recent paper has reported that p53, with dual roles in apoptosis and in G1 arrest, is regulated by p38 MAP kinase phosphorylation (11). Thus, a working model is that the p38 MAP kinase pathway may coordinate aspects of cell cycle progression by controlling the levels of important G1 regulatory proteins.

In the present work, we isolated the high-mobility group (HMG) box protein HBP1 transcriptional repressor in a yeast two-hybrid screen for new p38 MAP kinase substrates in G1 regulation. For this work, the most relevant observation is that HBP1 consistently inhibits G1 progression in cell and animal models (27, 28). HBP1 was originally identified as an interactor with the retinoblastoma protein (RB) and p130 growth and tumor suppressor genes (13, 28). HBP1 is an HMG-box transcriptional repressor that regulates the expression of several genes involved in cell cycle regulation, including cyclin D1, N-MYC, and others (15, 16, 25) (28). HBP1 levels are induced in the terminal differentiation program of many tissues and appear to regulate differentiation in an RB-dependent manner (26, 28). These results suggest that HBP1 may function as a proliferation barrier in terminally differentiated tissues (27). HBP1 was also identified as an inhibitor of the Wnt oncogenic pathway (25). Impaired genetic and biochemical regulation of the Wnt pathway (at APC, ß-catenin, and axin) are linked to colorectal, hepatocellular, and numerous other cancers. Together, these observations predict a role for HBP1 in tumor suppression by inhibiting G1 progression and/or major oncogenic signaling pathways.

Because of its known role in G1 regulation, HBP1 is an attractive target for p38 MAP kinase and cell cycle regulation, since both HBP1 and p38 MAP kinase share functional similarities in the control of G1 progression (6, 10, 14, 19, 27, 28). In this paper, we present evidence that HBP1 is a functionally relevant substrate for p38 MAP kinase in cell cycle control. The apparent functional consequence of p38 MAP kinase activity is an increase in HBP1 protein stability. Conversely, the inhibition of p38 MAP kinase activity leads to a decrease in HBP1 protein levels and the elimination of HBP1-mediated G1 inhibition. Together with other data, these results suggest a molecular framework by which the p38 MAP kinase pathway may coordinate a network of HBP1 and other G1 regulatory proteins by controlling protein stability.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Two-hybrid screening. The p38 MAP kinase bait for the yeast two-hybrid system was based on the pAS2 yeast expression vector and was a derivative of CMV-Flag-p38 (23). Specific p38-interacting proteins were isolated in a two-hybrid screen with a cDNA library constructed in {lambda}ACTII from mouse liver. A true positive was defined as a cDNA encoding a prey protein that was able to interact with the original bait (p38{alpha}) but not with irrelevant proteins (e.g., SNF1, a gift of S. Elledge, Baylor College, Houston, Tex.). Of the 46 positive clones obtained, clones 15 and 27 contained nearly full-length mouse HBP1, as determined by DNA sequencing.

Animals. Male, young adult (approximately 12 weeks old) C57 BL6/J mice (Charles River Laboratory) were randomly divided into three groups. These three groups included mice which were (i) fed the control diet, (ii) fed the control diet and fasted for 36 h, and (iii) fed the control diet, fasted for 36 h, and then refed a high-carbohydrate-no-fat diet. Water was available ad libitum. All groups were maintained on the control diet for at least 7 days before the onset of the experiment. Mice were sacrificed at the various time points shown in the figures, at which time livers were removed as indicated below. The control and high-carbohydrate-no-fat diets were isocaloric modifications of the AIN-76 diet (2) and composed as follows. The control diet was composed of casein (200 g/kg of body weight), DL-methionine (3 g/kg), cornstarch (145 g/kg), sucrose (484 g/kg), cellulose (70 g/k), safflower oil (50 g/kg), AIN salt mix (35 g/kg), AIN vitamin mix (10 g/kg), and choline bitartrate (2 g/kg). The high-carbohydrate-no-fat diet modified the control diet amounts of cornstarch (173 g/kg), sucrose (577 g/kg), and safflower oil (0 g/kg). Mice expressing a liver-specific transthyretin promoter and MKK3 transgene, in which the MKK3 contains a substitution mutation in the kinase active site (serine-to-alanine, amino acid [aa]183, termed MKK3ala) (24) were prepared as described (C.-Y. Huang, L.-R. Contois, M. Xiu, C. Tuzon, A. S. Yee, R. J. Davis, and K. E. Paulson, submitted for publication). A direct PCR assay was used for the mouse screening. Primers designed against the TTR promoter and MKK3ala cDNA are 5'-CTTTTTGCACCATCCACCTTTC-3' and 5'-GTTTCTGTCTCCAATGGTGATG-3', respectively. The HBP1 primer is 5'-CACTTTGAACAGCCTGAAG-3'. Briefly, a portion of the mouse tail was digested with 2.5 µl of protease K in 50 µl of digestion buffer (50 mM KCl, 1.5 mM MgCl2, 10 mM Tris-Cl [pH 8.5], 0.01% gelatin, 0.45% NP-40, 0.45% Tween 20). The digestion mixture was then incubated at 55°C overnight, followed by inactivation for 15 min at 95°C. The PCR mixture was comprised of 1 µl of supernatant from the digestion solution and 49 µl of PCR mixture (5 µl of 10x PCR buffer, 1µ of 10 mM dNTP, 1.5 µl of 100 mM MgCl2, 1.5 µl of 10 pM TTR primer, 1.5 µl of 10 pM MKK3ala or HBP1 primer, 0.5 µl of Taq DNA polymerase, 38 µl of H2O). PCRs were performed under the following conditions: 95°C for 5 min, 94°C for 30 s, 51°C for 30 s, and 72°C for 2 min; this cycle was repeated for 35 cycles. The PCR products were analyzed on a 1% agarose gel. In addition, a pair of primers that were designed against cyclin D1 was used as control.

Plasmids. Plasmid pEFBOSHAHBP1 wild-type (WT) and mutant constructs have been published previously (25, 28). Plasmid pEFBOSmycHBP1 was also constructed by PCR to add the MYC epitope. Plasmid pEFBOS-HA(S-A401)HBP1 was constructed by site-directed mutagenesis (Stratagene) to obtain a point mutation that gave an amino acid change from serine to alanine at position 401 of HBP1. The CMVflagMKK6-glu, CMV-Flag-p38 (dominant negative), and CMVflagMKK3ala were described previously (23, 24).

Cell culture and transfections. C2C12 muscle cells were maintained in Dulbecco's modified Eagle's medium (DMEM; GIBCO) supplemented with 20% fetal bovine serum (FBS) and grown in 5% CO2. 293T cells were maintained in DMEM supplemented with 10% FBS and grown in 5% CO2. All media were supplemented with glutamine, penicillin, and streptomycin. Transfections were performed by the calcium phosphate precipitation method. Treatments with 10 µM SB203580 or 10 µM MG132 were performed 24 h after transfection, and cells were harvested at various time points by using the cell lysis buffer provided by the manufacturer (Cell Signaling, Beverly, Mass.)

The 4XJ reporter is specific for HBP1 transcriptional repression (37; S. P. Berasi, M. Xiu, A. S. Yee, and K. E. Paulson, unpublished data). Multiple copies of the high-affinity HBP1 DNA binding site [(TTCATTCATTCA)4] were inserted upstream of the pGL3-control luciferase reporter (Promega). This reporter contains the enhancer and promoter of simian virus 40 and provides sufficient background promoter activity for detection of repression upon insertion of the high-affinity HBP1 DNA binding sites.

Liver protein extracts. Nontransgenic and transgenic C57BL6/J (25-g) mice were anesthetized at the time points indicated. The animals were perfused with normal saline, and livers were then removed. A total of 0.5 g of liver tissue was finely minced and extracted in 5 volumes of WCE buffer (25 mM HEPES [pH 7.5], 0.3 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.1% Triton X-100, 20 mM ß-glycerophosphate, 0.1 mM orthovanadate, 0.5 mM dithiothreitol, 0.4 mM phenylmethylsulfonyl fluoride [PMSF], 1 µg of leupeptin/ml, 1 µg of pepstatin/ml) for 30 min at 4°C. The resulting liver extracts were then centrifuged at 100,000 x g at 4°C for 30 min, and the cleared lysate was frozen at -80°C (18).

RNase protection assay. Total RNA was isolated from frozen mouse liver tissues with Trizol reagent (Sigma) according to the manufacturer's directions. RNase protection assays were performed with an RPA III kit (Ambion). Briefly, 30 µg of total RNA was hybridized to a 32P-labeled antisense murine HBP1 probe and then treated with RNase (Ambion), followed by analysis on a 6% polyacrylamide-8 M urea gel. An 18S RNA protection assay was also performed as an internal control for loaded RNA, with a 32P-labeled antisense probe for 18S RNA. HBP1 mRNA levels were quantified with a PhosphorImager (Molecular Dynamics) and normalized for 18S RNA.

MAP kinase assays. p38 MAP kinase activity was measured with a nonradioactive IP-kinase assay and the p38 MAP Kinase Assay kit as described by the manufacturer (Cell Signaling). In addition, p38 MAP kinase, JNK, or ERK phosphorylation of substrate with [32P]ATP was performed as described previously (18).

Western blot analysis. Whole-cell extracts (100 µg of protein) were diluted and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and enhanced chemiluminescence Western blots (NEN). The primary antibodies for p38, ERK, and JNK were purchased from Santa Cruz Biotechnology. The anti-flag monoclonal antibody and antihemagglutinin (anti-HA) polyclonal antibodies were purchased from Sigma and Covance Research Products, respectively. The polyclonal rabbit antibody to HBP1 was previously described (27).

In vitro binding assays. MC 1061 bacterial cells were transformed with pGEX2THBP1, various HBP1 mutants, or pGEX2T-p38. Glutathione S-transferase (GST) fusion proteins were affinity purified from bacterial culture according to a Pharmacia Biotech protocol and as previously described (34). Cellular extracts transfected with CMVflagp38 were lysed in NETN buffer (20 mM Tris [ph8.0], 1mM EDTA, 0.5% NP-40 [wt/vol], 100 mM NaCl) that was supplemented with 0.4 mM PMSF, 1 mM dithiothreitol, 1 µg of leupeptin/ml, 1 µg of pepstatin/ml, 0.1 mM orthovandate, and 20 mM ß-glycerophosphate. The resulting extracts were added to various GST fusion proteins that were bound to glutathione beads (Pharmacia) and incubated at 4°C for 2 h with gentle agitation. The bound proteins were washed with NETN buffer and analyzed by SDS-PAGE. Western blotting was performed with the indicated antibodies (see figure legends) to detect bound proteins. The expression of GST protein was monitored by SDS-PAGE.

Coimmunoprecipitation. Cell extracts were prepared by lysing transfected 293T cells in NETN lysis buffer that was supplemented with protease and phosphatase inhibitors (see above). Cell lysates were precleared with protein A-Sepharose beads (Repligen) before antibodies were added. Anti-myc (or other) antibody and beads were added to the precleared cell lysates and incubated at 4°C for 6 to 8 h with gentle agitation. Following three washes with NETN buffer, the beads were boiled in SDS sample buffer for 5 min and analyzed by SDS-PAGE. Western blot analysis was then performed with the indicated antibodies as described in the figure legends.

EMSA. An electrophoretic mobility shift assay (EMSA) was performed by incubating 0.5 ng of the 32P-labeled oligonucleotide containing the HBP1 high-affinity binding sequence (37) and 2 µg of poly(dI-dC) reaction mixture with 4 µg of cell extract for 30 min at room temperature. The mixture was then loaded on a 6% polyacrylamide-0.25 x TBE gel and run at 350 V for 2.5 h at 4°C. As a specificity control, 50-fold excess unlabeled competitor was used in the HBP1 binding reaction.

Protein half-life measurements. A conventional pulse-chase protocol was used. After 24 h of transfection, 293T cells were washed with phosphate-buffered saline (PBS). Cells were then pulse labeled with 35S protein-labeling mixture (NEN) in Hanks' salt (GIBCO) for 1 h. Complete DMEM was used for the chase portion. Labeled cells were harvested at the specified time points, extracts were prepared, and 100 µg of protein was immunoprecipitated with anti-HA antibody and protein A-Sepharose. The resulting pellets were analyzed by SDS-PAGE and phosphorimaging..

BrdU incorporation. The percentage of NIH 3T3 cells in S phase was detected by bromodeoxyuridine (BrdU) incorporation. The day before transfection, cells were plated so that they were 90 to 95% confluent on the day of transfection. One hour before transfection, cells were changed to Opti-MEM I reduced serum medium (Gibco) without serum. For a p100 plate, NIH 3T3 cells were transfected using Lipofectamine 2000 (LF2000) reagent (LF2000; Invitrogen). A total of 5 µg of pEF-BOS-HA-HBP1 and 2 µg of pEGFP (green fluorescent protein [GFP] expression plasmid) were diluted into 1.5 ml of Opti-MEM without serum. Ninety microliters of (LF2000) reagent was diluted into another 1.5 ml of Opti-MEM and incubated for 5 min at room temperature. Both diluted DNA and diluted LF2000 were combined and incubated at room temperature for another 20 min. The transfection mixture was then added to the cells and incubated overnight. The following day, transfected cells were replated onto coverslips in DMEM with 10% FBS, incubated for 4 h, and then treated with or without 10 µM SB 203580 for 4 h in complete DMEM (Gibco) with 10% FBS. In control cells, dimethylsulfoxide was used as the vehicle. Cells then were incubated with 10 µM BrdU (Roche) for 1 h. Cells on coverslips were then washed with PBS, fixed with 70% ethanol with 15 mM glycine for 20 min in -20°C, and rehydrated in PBS. Primary and secondary antibodies were appropriately diluted in DMEM containing 2% calf serum. The fixed cells on coverslips were incubated for 1 h with anti-BrdU primary antibody (1/100 dilution in PBS; Roche), washed six times for 5 min in PBS, and then incubated for 30 min with secondary anti-mouse tetramethyl rhodamine isothiocyanate (1/1,000 dilution in PBS; Jackson ImmunoResearch) at 37°C for direct staining. Fixed cells were washed six times for 5 min in PBS. For visualization of all nuclei in a field, coverslips were then stained with Hoechst dye (25 µg/ml) in DMEM with 2% calf serum for 3 min at room temperature. All coverslips were mounted on slides and examined by fluorescence microscopy with appropriate filters. Cotransfected GFPs were used as markers to detect transfected cells.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
HBP1 is a p38 MAP kinase interacting protein. We used a yeast two-hybrid screen to clone proteins that interact with p38{alpha} MAP kinase, with the kinase-defective form of p38{alpha} as the bait protein. Of 46 positive clones, HBP1 constituted one group of clones that was represented twice. As described in the introduction, HBP1 is an HMG box transcriptional repressor that suppresses G1 progression (27, 28). Given the functional similarity between HBP1 and p38 MAP kinase in regulating G1 progression, we focused our efforts on understanding the physical and functional relationship between HBP1 and p38 MAP kinase.

The interaction of HBP1 and p38{alpha} was first investigated in mammalian cells. 293T cells were transfected with constructs that expressed p38{alpha} containing an N-terminal FLAG epitope (FLAG-DNp38{alpha}) and HBP1 containing an N-terminal HA epitope (HA-HBP1), either together or separately. The respective cell extracts were immunoprecipitated with anti-HA antibody, and immunoblots (IB) were performed with anti-FLAG antibody to detect any interacting p38{alpha} (Fig. 1A). When FLAG-DNp38{alpha} was coexpressed with HA-HBP1 (lane 4), HA-HBP1 was specifically detected in the FLAG {alpha}-p38 immunoprecipitate, indicating a physical association. No interaction was detected under several control immunoprecipitations (IP; lanes 1, 2, 3, and 5). IB showed that the p38 MAP kinase and HBP1 proteins were equivalently expressed in the relevant extracts. The FLAG-DNp38{alpha} epitope is a dominant-negative p38 MAP kinase; efforts to detect an interaction with WT p38 MAP kinase in mammalian cells were repeatedly unsuccessful. The basis is unclear and is under investigation. Nonetheless, this result suggested that p38{alpha} and HBP1 could interact within cells under certain conditions and verified the initial data from the yeast two-hybrid screening.




View larger version (47K):
[in this window]
[in a new window]
  		FIG. 1. HBP1 interacts with p38 in mammalian cells. (A) Interaction of HBP1 with p38 in vivo. Cells were transiently transfected with CMV-FLAG-DNp38, pEFBOS-HA-HBP1, or both, as indicated in the figure. Lanes 1, 2, and 3 represent control lanes in which no plasmid, HA-HBP1 alone, or FLAG-p38 alone was expressed, respectively. Lanes 4 and 5 represent extracts from cells expressing both FLAG-DNp38 and HA-HBP1. IP with anti-FLAG or normal immunoglobulin G (IgG), respectively, is shown in lanes 4 and 5. Any p38-associated HBP1 was then detected by subsequent IB with anti-HA. A specific interaction was indicated by the presence of HA-HBP1 in lane 4 but not lane 5. The expression of the respective proteins in the input extracts for HBP1 (Input extract, {alpha}HA) and for p38 (Input extract, {alpha}FLAG) is shown. (B) Schematic diagram of GST-HBP1 constructs. The indicated GST-HBP1 proteins were used to identify the p38 MAP kinase-interacting domain on HBP1 (shown in panel C). Interacting constructs GST N393, GST(135-283) and GST(81-152) were used to define the potential interaction domain as the region between aa 125 and 165. This domain was confirmed by using the GST construct GST(125-165), which was sufficient to interact with p38 MAP kinase. Symbols indicate the absence (-) or presence (+) of a p38 MAP kinase interaction with the indicated HBP1 construct and summarize the primary data shown in panel C. (C) Identification of the p38 MAP kinase interaction domain on HBP1. The binding of p38{alpha} to HBP1 was measured in vitro by incubating the indicated GST-HBP1 mutants with extracts from cells expressing FLAG-p38{alpha}. The binding of p38{alpha} was detected by IB analysis with {alpha}FLAG. Symbols indicate the presence (+) or absence (-) of the p38 protein extract in the experimental lane. Based on the data shown in panel C, the p38 MAP kinase interacting region on HBP1 was defined as a region between aa 135 and 152 of HBP1.

 
The region of HBP1 that interacts with p38{alpha} was next defined in vitro. A battery of GST fusion proteins that span the HBP1 coding region was incubated with extracts from cells that expressed FLAG-p38{alpha} (Fig. 1B). The HBP1 mutants that interacted with the p38{alpha} protein were detected by IB with anti-FLAG antibody. As shown in Fig. 1C and summarized in Fig. 1B, the p38{alpha} interacting region on HBP1 was defined as aa 135 to 152. Furthermore, deletion of this region from the full-length protein (GST HBP{Delta}p38) abolished p38{alpha} interaction in vitro (Fig. 1B and C). Together, these results define a p38 MAP kinase docking site within the HBP1 protein.

p38 MAP kinase activity regulates HBP1 protein levels. While p38 MAP kinase did specifically interact with HBP1, the functional impact of p38 MAP kinase on the HBP1 protein itself or its known activities was next investigated. The experiments shown in Fig. 2A and B used 293T cells because of a high endogenous p38 MAP kinase activity (Fig. 2B). This activity allowed the use of dominant-negative and chemical reagents to block p38 MAP kinase activity. As shown in Fig. 2A, the HBP1 protein levels were reduced when coexpressed with dominant-negative MKK3 (MKK3ala), which resulted in the inhibition of p38 MAP kinase activity (23). Repeated attempts with reagents that activate p38 MAP kinase (e.g., MKK6-glu) (24) were either confusing or uninformative. Because the endogenous p38 MAP kinase activity is already high in 293T cells, the expression of MKK6-glu either gave little additional p38 MAP kinase activation over a high baseline or gave significant apoptosis (data not shown). We found that cleaner comparisons could be achieved by using dominant-negative (e.g., MKK3ala) and/or chemical (e.g., SB203580) means to block endogenous p38 MAP kinase activity.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 2. HBP1 protein level and DNA binding activity are dependent on p38 MAP kinase activity in cultured cells. (A) p38 MAP kinase regulation of HBP1. The effects of the p38 MAP kinase on HBP1 protein levels and DNA binding activity were measured. HA-HBP1 was cotransfected with FLAG-MKK3ala, which is a dominant-negative MKK3 and which should inhibit the activation of the p38 MAP kinase activity (24). As shown in the upper two panels, the coexpression of HA-HBP1 and FLAG-MKK3ala resulted in reduced expression levels of the HBP1 protein (second panel, right lane) compared to levels in the control (second panel, left lane). The expression of FLAG-MKK3ala (MKK3ala) was measured by Western blotting with anti-FLAG antibody (IB:FLAG). Similarly, HBP1 expression was detected by Western blotting with anti-HA antibody (IB:HA).In the third panel, HBP1 DNA binding activity was detected by EMSA with a 32P-labeled high-affinity HBP1 DNA binding site probe. Specificity was established by competition with unlabeled high-affinity site HBP1 oligonucleotides. Symbols indicate the absence (-) or presence (+) of a 50-fold molar excess of cold competitor DNA. In the fourth panel, a control SP1 EMSA is shown to demonstrate that expression of MKK3ala did not affect the activity of an unrelated transcription factor. It is important to note that the decreased level of HBP1 DNA binding activity (third panel) likely reflects decreased HBP1 protein levels (as shown in the upper two panels) and is not a result of a change in the intrinsic HBP1 activity (see Results and Fig. 6B). (B) Endogenous HBP1 protein levels in 293T cells are regulated by p38 MAP kinase activity. 293T cells were treated with the p38 MAP kinase-specific inhibitor SB203580 (10 µM) for 24 or 48 h. Cell extracts were prepared asdescribed in Materials and Methods, and p38 MAP kinase activity was measured. As shown, 293T cells have high endogenous p38 MAP kinase activity, which can be inhibited by SB203580 treatment. In the same extracts, endogenous HBP1 protein levels were measured by Western blotting (IB:HBP1) with an anti-HBP1 antibody. As a control for total protein levels, the level of the p38 MAP kinase protein was invariant (IB: p38).

 
Consistent with the correlation of reduced HBP1 protein levels and p38 MAP kinase activity, we reasoned that p38 MAP kinase activity should alter overall HBP1 DNA binding activity. To address this question, we examined HBP1 DNA binding activity in the same extracts from 293T cells that expressed HBP1 with or without dominant interfering MKK3ala in greater than 90% of the transfected cell population. As shown in Fig. 2A, levels of HBP1 DNA binding were correspondingly reduced in the presence of MKK3ala and were consistent with the observed reduction in HBP1 protein levels. It should be noted that changes in HBP1 DNA binding could result from changes in overall protein expression and/or in intrinsic activity. As shown below (see Fig. 6B), the latter appears independent of p38 MAP kinase regulation. The activity of an unrelated transcription factor (SP1) was not affected by inhibiting p38 MAP kinase activity, which suggests that p38 MAP kinase activity does have some specificity for HBP1. Figure 2A highlights the specificity of the p38 MAP kinase in regulating HBP1 protein levels.



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 6. Phosphorylation of HBP1-S401 regulates HBP1 protein accumulation but not transcriptional repressor specific activity. (A) Loss of phosphorylation on serine 401 inhibits p38 MAP kinase-mediated protein stabilization. To address the role of inhibiting p38 MAP kinase activity on HBP1 protein levels, a mutation in the putative p38 MAP kinase phosphorylation site at serine 401 [denoted HA-(S-A)401HBP1] was tested for overall expression levels and for regulation by the p38 Map kinase. Either HA-(S-A)401HBP1 or HA-HBP1 was coexpressed with FLAG-MKK3ala. The levels of the respective HBP1 constructs were detected by anti-HA immunoblotting (bottom panel). The levels of FLAG-MKK3ala were detected by anti-FLAG immunoblotting (top panel). Compared to protein levels of WT HA-HBP1, mutation of the p38 phosphorylation site [HA-(S-A)401HBP1] resulted in significantly lower protein levels that were unaffected by expression of MKK3ala. (B) Phosphorylation on serine 401 does not affect specific activity in transcriptional repression. To compare specific activity of the WT HBP1 and (S-A)401HBP1 mutant, equivalent expression was required, despite a large difference in protein stability (see panel A). Empirical conditions were identified in which the (S-A)401HBP1 protein was expressed at a similar level as full-length WT-HBP1 protein by adding significantly more DNA during transfection. Both constructs are HA-tagged, allowing for a clean comparison of expression levels. An HBP1-specific luciferase reporter (4XJ; see Materials and Methods) was used to detect specific HBP1 repressor function. When WT-HBP1 and (S-A)401HBP1 were expressed at the same protein level, as defined by HA-Western blotting, both had equally strong repression on an HBP1-specific reporter. These results suggest that serine 401 phosphorylation of HBP1 does not grossly affect its intrinsic specific activity for transcriptional repression.

 
Next, we examined the endogenous HBP1 protein levels in 293T cells in the presence and absence of p38 MAP kinase activity. In 293T cells, the endogenous p38 MAP kinase activity is robust, as demonstrated by an IP kinase assay (Fig. 2B). Inhibition of the endogenous p38 MAP kinase activity by the inhibitor SB203580 (10 µM) resulted in a marked reduction in the levels of the endogenous HBP1 protein at either 24 or 48 h of treatment (Fig. 2B). Together, the data represented in Fig. 2A and B indicated that HBP1 protein levels in cells declined in the absence of p38 MAP kinase activity.

Similar results were obtained in a unique in vivo system in which transgenic mice express HBP1 in the liver (27). In experiments that are analogous to the experiments depicted in Fig. 2A and B, p38 MAP kinase activity was either activated or inhibited in the liver. Both transgenic and endogenous HBP1 protein levels were examined. Using a previous observation, we noted that p38 MAP kinase is specifically responsive to metabolic function in the liver (18; Huang et al., submitted). Because of the normal metabolic regulation of p38 MAP kinase activity in mouse liver, the consequences of p38 MAP kinase pathway stimulation on HBP1 levels were more informative because there was no detectable apoptosis to confuse experimental interpretations. The endogenous p38 MAP kinase activity in mouse liver could be manipulated by fasting (inhibition of p38 MAP kinase activity) (Fig. 3A) or by fasting and then refeeding the mouse (activation of p38 MAP kinase activity) (Fig. 3A). As shown in Fig. 3B, when p38 MAP kinase was active, HBP1 protein levels were correspondingly elevated. Similarly, when p38 MAP kinase activity was low, HBP1 protein levels were correspondingly reduced. Finally, to eliminate the confusing possibilities arising from dietary manipulation and to directly address the role of p38 MAP kinase activity, the HBP1 protein levels in doubly transgenic mice that expressed both HBP1 and MKK3ala in the mouse liver were examined. In single or doubly transgenic mice that express the dominant-negative MKK3ala, the p38 MAP kinase activation in response to the fasting and refeeding protocol is markedly reduced (Fig. 3B) and indicates a true blockade of activation. While HBP1 protein levels increased in response to p38 activation (Fig. 3A), there were low levels of HBP1 protein in the double transgenic mice that express both HBP1 and MKK3ala. As in the cell-based models, transgenic mice expressing MKK6-glu that would provide for robust p38 MAP kinase activation have never been obtained, despite repeated attempts (data not shown). Still, the results depicted in Fig. 2 and 3 from both cell-based and animal models demonstrate that a blockade of p38 MAP kinase activity correlates with reduced HBP1 protein levels.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 3. HBP1 protein level and DNA binding activity is dependent on p38 MAP kinase activity in transgenic (Tg) mice. (A) Tg HBP1 protein levels are regulated by p38 MAP kinase activity. In Tg HBP1 mouse liver (27), HBP1 protein levels covaried with p38 MAP kinase activity (B) during the fasting and refeeding treatment as determined by Western blotting for HBP1 and the p38 MAP kinase activity assay described previously. However, in HBP1/MKK3ala double Tg mouse liver, this pattern was diminished by inhibition of p38 MAP kinase activity. (B) Tg dominant-negative MKK3 (MKK3ala) expression in the liver blocks p38 MAP kinase activation. WT and MKK3ala Tg mice were fasted (FAST) to inhibit p38 MAP kinase activity and fasted and then refed (F/R) to stimulate p38 MAP kinase activity. p38 MAP kinase activity activated in liver extracts after refeeding WT mice was measured by IP kinase assay ([9.2 ± 1.7]-fold activation; n = 6). Neither JNK nor ERK is activated by fasting or by fasting and refeeding (data not shown). p38 MAP kinase activity is weakly activated in liver extracts from MKK3ala Tg mice ([2.2 ± 0.6]-fold activation; n = 6). HBP1 DNA binding activity in the liver is activated in a p38 MAP kinase-dependent manner. Liver extracts from WT and MKK3ala Tg mice were used in EMSA with an HBP1 high-affinity binding site oligonucleotide (Berasi et al., unpublished) (see Materials and Methods). Extracts from WT mice show activation of DNA binding in F/R mouse liver, concomitant with p38 MAP kinase activation. MKK3ala expression in the liver results in much weaker activation of HBP1 DNA binding. Symbols indicate the absence (-) or presence (+) of 50-fold excess cold competitor.

 
Similar to our observations in cells, the changes in HBP1 protein levels were also reflected in changes in the level of overall DNA binding activity in the mouse. As described above, the identical fasting and refeeding protocol was used to manipulate the p38 MAP kinase activity in the liver. As shown previously in Fig. 3A, p38 MAP kinase activity was increased in liver extracts of refed nontransgenic mice with an accompanying increase in HBP1 protein levels. Correspondingly, endogenous HBP1 DNA binding activity was substantially activated in extracts from the liver when p38 MAP kinase activity was elevated. To verify that p38 MAP kinase activity was required, we examined HBP1 DNA binding activity from liver extracts of transgenic mice that expressed MKK3ala, which have markedly reduced p38 MAP kinase activity, as shown in Fig. 3B. In the presence of MKK3ala expression, a decrease in HBP1 DNA binding activity (Fig. 3B) also correlated with a decrease in HBP1 protein levels (Fig. 3A).

While changes in HBP1 DNA binding could result from changes in overall protein expression and/or in intrinsic activity, the data represented in Fig. 3 (see also Fig. 6) argue that the decrease in HBP1 protein levels upon inhibiting the p38 MAP kinase explains the overall decrease in DNA binding activity. In the absence of p38 MAP kinase activity in cells and in the mouse, the levels of HBP1 protein were markedly reduced with a corresponding decrease in DNA binding activity. These different analyses indicate that p38 MAP kinase activation is necessary for regulating HBP1 protein levels in cells and tissues. The impact of p38 MAP kinase on the specific transcriptional repressor activity of HBP1 will be addressed below (see Fig. 6).

p38 MAP kinase regulates HBP1 protein half-life. We first examined the mechanism for p38 MAP kinase regulation of HBP1 protein levels through analysis of HBP1 mRNA levels. However, the experiments depicted in Fig. 2 and 3, in which HBP1 mRNA and protein was expressed by three entirely different means (endogenous HBP1 promoter, transgenic TTR promoter, and exogenous EF-BOS expression vectors), already argue that p38 MAP kinase probably did not regulate HBP1 protein levels through the promoter. Nevertheless, we measured endogenous mouse liver HBP1 mRNA in response to changes in p38 MAP kinase activity. As shown in Fig. 4A, HBP1 mRNA may even increase slightly when p38 MAP kinase is active (fasted and refed mice), compared to when p38 MAP kinase is inactive (fasted mice). However, in numerous experiments measuring HBP1 mRNA levels, the change in HBP1 mRNA was <=30% and certainly did not correlate with the observed decrease in HBP1 protein when p38 MAP kinase activity was blocked. Finally, the activity of an EF-BOS-luciferase control reporter showed no change in the absence or presence of either MKK3ala or of SB203580 (data not shown). The EF-BOS vector contains the promoter for the EF1{alpha} translation elongation factor and is necessary for all HBP1 expression experiments. Because SB203580 does not affect the EF1{alpha} promoter activity in the EF-BOS-luciferase vector, any difference in transfected HBP1 protein levels cannot result from effects on the expression vector.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 4. HBP1 protein half-life, but not mRNA, is affected by p38 MAP kinase activity. (A) HBP1 mRNA expression level was not significantly affected by p38 MAP kinase activity. The level of HBP1 mRNA during fasting and refeeding (F/R) treatment was scored by RNase protection assay with total RNA that was isolated from liver tissue. 18S RNA was used as an RNA loading control. The results were quantified with a PhosphorImager and normalized for 18S RNA. These data are an average of two experiments that did not vary more than 10%. F, fasted mice. (B) p38 MAP kinase inhibition decreases HBP1 protein stability. The HA-HBP1 expression construct was transfected into a single 150-mm2 plate of 293T cells, which was then distributed to multiple smaller dishes to ensure uniform expression of HA-HBP1 across all plates. Cells were placed in methionine-free media, treated with 10 µM SB203580 or with vehicle, pulsed with [35S]methionine for 15 min and then chased with excess unlabeled methionine-supplemented media. Cells were harvested at the indicated timepoints. 35S-labeled HBP1 was immunoprecipitated with anti-HA antibody. The labeled HBP1 protein was quantitated by phosphorimager analysis after SDS-10% PAGE (bottom panel). The decay of HBP1 followed the expected first order kinetics. The slope of the decay line was calculated by standard linear regression, and the protein half-life was determined accordingly (top panel) (see Results). In the presence of SB203580, the HBP1 half-life was 0.61 ± 0.13 h, compared to 1.47 ± 0.14 h for untreated cells (P < 0.002).

 
A second possible mechanism is that p38 MAP kinase activity might regulate HBP1 protein stability. The results shown in Fig. 2 and 3 indicated that blocking p38 MAP kinase activity reduced HBP1 protein levels and suggested a reduction in protein half-life. A standard pulse-chase experiment was used to measure HBP1 protein half-life in the presence or absence of the p38 MAP kinase inhibitor SB203580 (Fig. 4B). The following kinetic equation for a first-order decay process was used: ln[(HBP)t/(HBP)0] = -kt, where k is the first order decay constant and t is time. (HBP)t and (HBP)0 represent the concentration of HBP1 at time equal t and 0, respectively. Thus, k is the slope of a graph of ln[(HBP)t/(HBP)0] versus t. The half-life is the time to reach 50% of (HBP)0. The kinetic constant and half-life of the HBP1 protein in the presence and absence of p38 MAP kinase activity was determined by linear regression analysis. As expected, the decay of the labeled HBP1 protein had first-order kinetics upon chase with cold methionine. As shown in Fig. 4B, in 293T cells with constitutive p38 MAP kinase activity, HBP1 protein has an apparent half-life of ~84 min. Upon treatment with the p38 MAP kinase inhibitor SB203580 at 10 µM, the apparent half-life of HBP1 declined to ~36 min. These results indicate that p38 MAP kinase activity regulates HBP1 protein half-life and contributes to overall regulation of the HBP1 protein levels.

The p38 MAP kinase-mediated phosphorylation of HBP1 (serine 401) confers protein stability but does not regulate transcription-specific activity. The data suggest that p38 MAP kinase might regulate HBP1 protein stability through phosphorylation. While p38 MAP kinase clearly binds to HBP1, we next determined whether HBP1 was a substrate for p38 MAP kinase by defining the relevant phosphorylation site(s). With GST-HBP1 fusion proteins (Fig. 1B) and in vitro p38 MAP kinase assays, the full-length HBP1 protein was phosphorylated by p38 MAP kinase (Fig. 5A). Furthermore, only HBP1 mutant proteins that contained the HMG box domain were specifically phosphorylated by p38 MAP kinase (Fig. 5A). The GST fusion protein sequence from aa 81 to 152 [GST(81-152)], GST(135-283), and GST(1-393) (N393) are not phosphorylated. Similarly, neither the JNK nor ERK Map kinases interacted with nor phosphorylated any HBP1 protein (data not shown), highlighting specificity for the p38 MAP kinase.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 5. HBP1 is specifically phosphorylated by p38 MAP kinase in vitro. (A) In vitro mapping of p38 MAP kinase (MAPK) phosphorylation. GST-HBP1 fusion proteins were used as substrates for IP kinase assays with p38 MAP kinase. The positive control for p38 MAP kinase is denoted GST-ATF-2 and represents a GST fusion protein with 100 aa of the N terminus. As shown, HBP1 is specifically phosphorylated by p38 MAP kinase. Only the full-length protein (FL) and the HMG box proteins were phosphorylated. The GST(81-152), GST(135-283), and GST(1-393) (N393) fusion proteins were not phosphorylated. These data localize the phosphorylation site(s) to the region from aa 396 to 513. (B) Homology with other p38 MAP kinase substrates and consequences of mutation at serine 401. The sequence of HBP1 from aa 396 to 409 has strong homology to the known p38 MAP kinase substrate CHOP (31) with a potential site at serine 401. Therefore, serine 401 within the GST-HMG box construct was mutated to alanine, creating HBP1(S-A), as indicated by an asterisk. (C) To determine the consequences of serine 401 mutation by p38 MAP kinase phosphorylation, the p38 MAP kinase activity was immunoprecipitated from cells that were treated with UV with {alpha}-p38 (denoted +p38). The control was a normal immunoglobulin G IP (denoted -p38). The control (-p38) and test (+p38) immunoprecipitates were used in kinase reactions with the indicated substrates. The bottom and top panels show a Coomassie blue-stained SDS-10%PAGE and the resulting autoradiogram, respectively. The position of the substrate is indicated by an arrow. ATF-2 is a GST fusion protein with the 100 aa of the N-terminus of ATF and is a positive control for p38 kinase activity. FL-HBP1(WT) represents a GST fusion protein with full-length HBP1. FL-HBP1(S-A)401 represents a GST fusion protein with an alanine substitution at serine 401 in the context of the FL-HBP1. HMG-HBP1 (WT) represents a GST fusion protein from aa 393 to 513 which contains the HMG box region. HMG-HBP1 (S-A)401 represents a GST fusion protein from aa 393 to 513, which contains the HMG box region and an alanine substitution at serine 401. As shown, mutation of serine 401 eliminates phosphorylation by the p38 MAP kinase.

 
The data represented in Fig. 5A suggest that a potential p38 MAP kinase phosphorylation site exists within the HMG box region. The sequence from aa 396 to 409 has strong homology to the known p38 MAP kinase site in the CHOP protein (31) (Fig. 5B) and with a potential phosphorylation site at serine 401. Therefore, we mutated serine 401 to alanine within the contexts of the full-length HBP1 protein [FL-HBP1(S-A)401] and of the HMG box proteins [HMG-HBP1 (S-A)401]. As shown in Fig. 5C, a mutation at serine 401 abolished p38 MAP kinase-mediated phosphorylation of either the FL-HBP or of the shortened HMG box construct. The data shown in Fig. 5 indicate that serine 401 is a potential p38 MAP kinase phosphorylation site on HBP1.

Based on the data in which p38 MAP kinase activity regulates HBP1 protein stability (Fig. 2 to 4), the loss of the phosphorylation site at serine 401 (Fig. 5A) might have two functional consequences. An (S-A)401HBP1 mutant should have low stability and should be refractory to p38 MAP kinase regulation. Therefore, we created an (S-A)401 mutation within the FL-HBP1 protein. As shown in Fig. 6A, the protein levels of the (S-A)401 HBP1 mutant were nearly undetectable, when compared to WT HBP1 at equivalent transfection. The protein levels of (S-A)401HBP1 were unchanged in the presence of MKK3ala and were refractory to p38 MAP kinase activity status. By contrast, MKK3ala-mediated inhibition of p38 MAP kinase activity resulted in decreased HBP1 protein levels (Fig. 2A). Thus, mutation of the p38 MAP kinase site in HBP1 reduced protein stability and abolished the consequences of p38 MAP kinase regulation.

Mutation of serine 401 triggered HBP1 protein instability but apparently did not change its specific activity for transcriptional repression. The repression activity of WT HBP1 and (S-A)401HBP1 was compared with a specific HBP1 reporter construct (termed 4XJ). This reporter construct has high-affinity HBP1 sites that were fused upstream of an activated promoter, providing an excellent tool to probe HBP1-specific repression (37; Berasi et al., unpublished). To directly compare the specific activity of HBP1 that was either refractory or sensitive to p38 MAP kinase activity, equal expression of the otherwise unstable (S-A)401HBP1 and WT HBP1 was required and achieved by varying the amount of transfected DNA [increasing (S-A)401HBP1 DNA and decreasing WT HBP1 DNA]. As shown in Fig. 6B, both WT HBP1 and (S-A)401HBP1 had equivalent repressor activities on the HBP1-specific reporter 4XJ at equivalent protein levels. Thus, while (S-A)401HBP1 was apparently unstable, this mutant retained full transcriptional repressor activity, with apparent specific activity that was comparable to WT HBP1. This observation suggests that the primary mechanism for p38 MAP kinase regulation of HBP1 is the stabilization of protein levels rather than regulation of intrinsic repressor activity.

We next investigated how p38 MAP kinase phosphorylation of HBP1 might increase HBP1 stability. The turnover of many cell cycle regulators is regulated by ubiquitin-dependent proteolysis (reviewed in reference 36). Given our results, we predicted that the absence of p38 MAP kinase phosphorylation on HBP1 might trigger degradation by the ubiquitin-dependent pathway and that p38 MAP kinase phosphorylation might protect HBP1 from ubiquitin-dependent degradation. Thus, the p38 MAP kinase phosphorylation state might dictate the HBP1 sensitivity to ubiquitin-mediated degradation. We hypothesized that, in cells with high basal p38 MAP kinase activity, WT HBP1 cells should be insensitive to inhibitors of the ubiquitin degradation pathway (e.g., MG132). By contrast, the (S-A)401HBP1 mutant should be sensitive to inhibitors of the ubiquitin degradation pathway, since this mutant cannot be phosphorylated and protected by p38 MAP kinase phosphorylation.

To test our hypothesis, 293T cells with high basal p38 MAP kinase activity (Fig. 2B) were transfected with WT HPB1 and treated with either 10 µM SB203580 (to inhibit p38 MAP kinase activity) or 10 µM MG132 (to inhibit the 26S proteosome and ubiquitin-mediated degradation) or with both. As shown in Fig. 7, MG132 had no effect on WT HBP1 protein levels in the presence of constitutive p38 MAP kinase activity. Importantly, HBP1 levels declined with SB203580 treatment (Fig. 2), but cotreatment with SB203580 and MG132 restored HBP1 levels. MG132 blocked ubiquitin-mediated degradation and prevented the reduction of HBP1 levels upon p38 MAP kinase inhibition by SB203580. This result suggests that degradation of the HBP1 protein occurred via the ubiquitin pathway but that p38 MAP kinase phosphorylation protected HBP1 from degradation.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 7. A role for p38 MAP kinase phosphorylation and proteosome-mediated degradation in HBP1 protein stability. Both WT HBP1 and (S-A)401HBP1 were transfected into 293T cells that have constitutive p38 MAP kinase activity. The cells were treated with either 10 µM SB203580 (SB), 10 µM MG132 (M), or both. WT HBP1 protein levels decreased upon treatment with SB203580, while treatment with MG132 had no effect. However, MG132 treatment of SB203580-treated cells blocked the SB203580-induced degradation of WT HBP1. (S-A)401HBP1 protein levels were low and were not affected by treatment with SB203580. However, treatment of (S-A)401HBP1 with MG132 increased (S-A)401HBP1 protein levels. C, control.

 
The identical experiment was performed with the (S-A)401HBP1 mutant. As shown in Fig. 7, the basal (S-A)401HBP1 levels were reduced, as expected. Unlike the levels of WT HBP1, (S-A)401HBP1 levels increased with MG132 treatment but were refractory to treatment with SB203580. As described above and in Fig. 5B and C, (S-A)401HBP1 lacks a critical serine and is refractory to p38 MAP kinase regulation. This observation indicates that the ubiquitin-proteosome pathway is likely responsible for HBP1 degradation. Together, the data in Fig. 6 and 7 indicate that p38 MAP kinase phosphorylates HBP1 at serine 401 and that phosphorylation may protect against ubiquitin-mediated degradation.

p38 MAP kinase regulates HBP1-dependent cell cycle arrest. Because HBP1 is a G1 regulatory protein, the next set of experiments addressed the relationship of p38 MAP kinase and HBP1 in the context of cell cycle regulation. HBP1 was expressed in NIH 3T3 cells (along with the cell marker GFP) in the presence or absence of the p38 MAP kinase inhibitor SB203580. The percentage of transfected cells in S phase was scored by BrdU incorporation in GFP-positive cells. To maximize the detection of HBP1-mediated G1 inhibitory activity, the transfected cells were first starved by serum withdrawal and then restimulated by serum addition. This protocol provided the maximal levels of G1 inhibition by HBP1 (Fig. 8) (28) and thus provided the best scenario to detect the consequences of p38 MAP kinase activity on HBP1. Because of the experimental design and the strong cell cycle induction by serum, the consequences of treatment with the p38 MAP kinase inhibitor SB203580 alone could not be clearly assessed, although other studies have shown that p38 MAP kinase inhibition can drive S phase under different conditions (1). As shown in Fig. 8, subsequent treatment with the p38 MAP kinase inhibitor SB203580 relieved the HBP1-imposed G1 block. Given different criteria of HBP1 function and the data shown in Fig. 1 to 7, our results are consistent with a model in which p38 MAP kinase may regulate the cell cycle by controlling the levels of HBP1 and other G1 regulatory proteins (see Discussion and Fig. 9).



View larger version (48K):
[in this window]
[in a new window]
  		FIG. 8. Inhibition of p38 MAP kinase activity rescues an HBP1-induced block in S phase. HBP1 was expressed in the NIH 3T3 cells in the presence and absence of p38 activation by treatment with or without 10 µM SB203580. Coexpression of GFP was used to mark the transfected cells. In each experiment, the percentage of S-phase cells was scored by BrdU incorporation and expressed as a fraction of total GFP-positive and transfected cells. In each experiment, 300 to 400 cells were scored to determine the percentage of cells in S phase. To maximize the detection of HBP1-mediated G1 inhibitory activity, the cells were first starved by serum withdrawal for 18 h and then restimulated by serum addition. The percentages of control and HBP1-expressing transfected cells that are in S phase is shown. Each experiment was repeated three times in which values differed by <10%.

 


View larger version (14K):
[in this window]
[in a new window]
  		FIG. 9. Model of p38 MAP kinase regulation of G1 progression. Upon appropriate stimuli, the p38 MAP kinase phosphorylates several substrates involved in either activation of the cell cycle (cyclin D1) or inhibition of the cell cycle (p21CIP and HBP1). Phosphorylation of cyclin D1 results in proteosome-mediated degradation and decreased protein levels (3). By contrast, phosphorylation of p21CIP (10) and HBP1 (this work) result in protection from proteosome-mediated degradation and increased protein levels. These observations suggest that the p38 MAP kinase may control a network of G1 regulatory proteins that regulate G1 progression through regulation of their stability.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
While the control of p38 MAP kinase activity has significant implications for regulating apoptosis, its role in other aspects of cell proliferation are not well understood. Work by Kyriakis and others has demonstrated a role for the p38 MAP kinase pathway in the inhibition of G1 progression (14, 19). The elaboration of a p38 MAP kinase-regulated network for cell cycle control could have significant implications for an underappreciated role for the p38 MAP kinase pathway in G1 progression. Until recently, the relevant p38 MAP kinase substrates for cell cycle regulation have not been identified.

In this paper, we have identified the HBP1 transcriptional repressor as a p38 MAP kinase substrate. As described in the Results, the identification of the HBP1 transcriptional repressor and of its regulatory mechanism by the p38 MAP kinase is consistent with possible roles in G1 regulation. Our previous work has shown that HBP1 represses the expression of critical growth regulatory genes that can inhibit G1 progression in cells and animals (25, 27, 28). In the present paper, both cell and animal models were used to show that phosphorylation of HBP1 by p38 MAP kinase provides for increased HBP1 protein stability. In fact, the apparent phosphorylation of a single serine (at aa 401) appeared to protect the HBP1 proteins from ubiquitin-mediated degradation. The comparison of specific activity in transcriptional repression (Fig. 6) between WT HBP1 and the (S-A)401HBP1 mutant highlights the role of p38 Map kinase in controlling HBP1 protein levels but not intrinsic transcriptional activity. The observation that both the WT HBP1 and the (S-A)401HBP1 mutant proteins had equivalent specific activity (the ratio of the level of activity to the amount of protein) indicates that the mutation of serine 401 did not grossly affect overall protein structure, but the consequences were limited to protein stability. This result also indicated that the major factor in the overall decrease in HBP1 DNA binding activity with SB203580 (Fig. 2 and 3) is a decrease in HBP1 protein levels and not of intrinsic activity (Fig. 6). While the study in this paper clearly does not represent an exhaustive mapping of all potential HBP1 phosphorylation sites, the data do provide good support for serine 401 as a functionally relevant regulatory site for HBP1 protein stability in the context of p38 MAP kinase pathway functions.

A consequence of controlling HBP1 protein stability and levels through the p38 MAP kinase is the important regulation of G1 progression. The data shown in Fig. 8 illustrate that p38 MAP kinase activity can affect the cell cycle through HBP1. As summarized by the data shown in Fig. 9, HBP1 is a logical and informative substrate for p38 MAP kinase in cell cycle regulatory mechanisms. Our findings and recently published papers suggest that the p38 MAP kinase pathway may be a new molecular switch in a G1 regulatory network. The model in Fig. 9 proposes that p38 MAP kinase may coordinate aspects of G1 progression by regulating the stability of certain G1 proteins. While we were completing this work, links between p38 MAP kinase and the G1 proteins cyclin D1, p21CIP, and p53 were reported (3, 10, 11). In the example of p21CIP, the phosphorylation by p38 MAP kinase also triggers protein stabilization (10). The p21CIP protein is well established as a G1 inhibitor through the blockading of CDK2 activity. Alternatively, the cyclin D1 protein, which can drive G1 progression, is destabilized through p38 MAP kinase phosphorylation (3). Strikingly, the independent definition of p38 MAP kinase substrates (e.g., HBP1, p21, and cyclin D1) with implications for cell cycle regulation highlight a previously unappreciated, but important role for the p38 MAP kinase pathway in cell cycle control. These results predict that fluxes in p38 MAP kinase activity may have significant impact on the cellular and tissue-proliferative state.

Numerous reviews have highlighted the complexity of G1 regulation, including several components (p16, p53, and RB) that are tumor suppressor genes. Other studies have highlighted that biochemical regulation is equally important to genetic regulation, particularly in the case of p53 and RB (for a review, see reference 5). In this way, p38 MAP kinase regulation of key substrates like cyclin D1, p21CIP, and HBP1 may have added importance. The finding that HBP1 and p38 MAP kinase may enforce a cell cycle arrest suggests that both may be part of tissue mechanisms that prevent inappropriate proliferation. In the context of the cell cycle, regulating p38 MAP kinase activity could have an impact on G1 transitions by globally regulating effector proteins such as p21CIP and HBP1. Thus, HBP1 and p38 MAP kinase may be part of a potential tumor suppression mechanism, since the reproliferation of otherwise differentiated cells often represent premalignant changes. In fact, the HBP1 protein may be involved in tumor suppression. We have shown that HBP1 can suppress the Wnt oncogenic pathway in terms of both signaling and growth (25). In addition, HBP1 lies within a region (7q31.1) that is mutated in numerous cancers, leading to our speculation that HBP1 may be a new tumor suppressor gene (25). While the role of p38 MAP kinase in tumor suppression has not been widely investigated, a recent report suggests that inhibition of p38 MAP kinase activity can promote transformation in response to RAF, a downstream effector of the canonical RAS pathway (22). This study could also suggest that induction of p38 MAP kinase activity may be a barrier to transformation. Our results with p38 MAP kinase regulation and cell cycle control complement the studies. Thus, the potential role of p38 MAP kinase activity in tumor suppression may be both interesting and complex.

In summary, the work in this paper highlights the functions of the p38 MAP kinase pathway in cell cycle control by demonstrating the regulation of HBP1 protein stability by phosphorylation. As described, HBP1 suppresses G1 progression and has implications as a tumor suppressor. In keeping with recent reports, p38 MAP kinase may be a kinase trigger for a network that regulates G1 progression through protein stability. This fundamental observation adds a new dimension to the role of p38 MAP kinase in controlling cell proliferation but adds complexity to existing knowledge of the p38 MAP kinase pathway. In addition to documented roles in apoptosis, the work in this paper and others suggest that signals that inhibit p38 MAP kinase activity could promote cell cycle progression and tumorigenesis in certain contexts. By contrast, the signals that activate p38 MAP kinase may be tumor suppressive in two ways: by triggering apoptosis and by suppressing cell proliferation. The latter would occur by inducing the levels of proteins like HBP1 and p21CIP, while concomitantly inhibiting proteins like cyclin D1. The work in this paper opens new avenues for investigations into the cell cycle regulatory networks that may be triggered by the p38 MAP kinase pathway and may have implications for tumor suppression mechanisms.


   ACKNOWLEDGMENTS
 
We acknowledge the technical assistance of Creighton Tuzon in the early stages of this work and of Steve Berasi in providing the 4XJ vector before publication. We also thank Martin Obin and Vimla Band for discussions on ubiquitin-mediated protein degradation.

This work was supported by grants to A.S.Y. (National Institutes of Health [NIH] grant GM44634, NIH grant CA94187, and U.S. Army grant BC990538) and to K.E.P (NIH ES11518). R.J.D. is an Investigator of the Howard Hughes Medical Institute. The support of the GRASP Digestive Disease Center at New England Medical Center (P30 DK34928) and the use of their core facilities were gratefully appreciated.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Biochemistry, Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111. Phone: (617) 636-6850. Fax: (617) 636-2409. E-mail: amy.yee{at}tufts.edu. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Ambrosino, C., and A. R. Nebreda. 2001. Cell cycle regulation by p38 MAP kinases. Biol. Cell 93:47-51.[CrossRef][Medline]

2. American Institute of Nutrition. 1977. Report of the American Institute of Nutrition ad hoc Committee on Standards for Nutritional Studies. J. Nutr. 107:1340-1348.

3. Casanovas, O., F. Miro, J. M. Estanol, E. Itarte, N. Agell, and O. Bachs. 2000. Osmotic stress regulates the stability of cyclin D1 in a p38SAPK2-dependent manner. J. Biol. Chem. 275:35091-35097.[Abstract/Free Full Text]

4. Chang, L., and M. Karin. 2001. Mammalian MAP kinase signalling cascades. Nature 410:37-40.[CrossRef][Medline]

5. Classon, M., and E. Harlow. 2002. The retinoblastoma tumour suppressor in development and cancer. Nat. Rev. Cancer 2:910-917.[CrossRef][Medline]

6. Ellinger-Ziegelbauer, H., K. Kelly, and U. Siebenlist. 1999. Cell cycle arrest and reversion of Ras-induced transformation by a conditionally activated form of mitogen-activated protein kinase kinase kinase 3. Mol. Cell. Biol. 19:3857-3868.[Abstract/Free Full Text]

7. Hess, P., G. Pihan, C. L. Sawyers, R. A. Flavell, and R. J. Davis. 2002. Survival signaling mediated by c-Jun NH2-terminal kinase in transformed B lymphoblasts. Nat. Genet. 32:201-205.[CrossRef][Medline]

8. Juo, P., C. J. Kuo, S. E. Reynolds, R. F. Konz, J. Raingeaud, R. J. Davis, H. P. Biemann, and J. Blenis. 1997. Fas activation of the p38 mitogen-activated protein kinase signalling pathway requires ICE/CED-3 family proteases. Mol. Cell. Biol. 17:24-37. (Erratum, 17:1757, 1997.)[Abstract]

9. Khwaja, A., and J. Downward. 1997. Lack of correlation between activation of Jun-NH2-terminal kinase and induction of apoptosis after detachment of epithelial cells. J. Cell Biol. 139:1017-1023.[Abstract/Free Full Text]

10. Kim, G.-Y., S. E. Mercer, D. Z. Ewton, Z. Yan, K. Jin, and E. Friedman. The stress-activated kinases p38alpha and JNK1 stabilize p21CIP by phosphorylation. J. Biol. Chem., in press.

11. Kishi, H., K. Nakagawa, M. Matsumoto, M. Suga, M. Ando, Y. Taya, and M. Yamaizumi. 2001. Osmotic shock induces G1 arrest through p53 phosphorylation at Ser33 by activated p38MAPK without phosphorylation at Ser15 and Ser20. J. Biol. Chem. 276:39115-39122.[Abstract/Free Full Text]

12. Kyriakis, J. M., and J. Avruch. 2001. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol. Rev. 81:807-869.[Abstract/Free Full Text]

13. Lavender, P., L. Vandel, A. J. Bannister, and T. Kouzarides. 1997. The HMG-box transcription factor HBP1 is targeted by the pocket proteins and E1A. Oncogene 14:2721-2728.[CrossRef][Medline]

14. Lavoie, J. N., G. L'Allemain, A. Brunet, R. Muller, and J. Poussegur. 1996. Cyclin D1 expression is regulated positively by the p42/p44 MAPK and negatively by the p38/HOG MAPK pathway. J. Biol. Chem. 271:20608-20618.[Abstract/Free Full Text]

15. Lemercier, C., K. Duncliffe, I. Boibessot, H. Zhang, A. Verdel, D. Angelov, and S. Khochbin. 2000. Involvement of retinoblastoma protein and HBP1 in histone H10 gene expression. Mol. Cell. Biol. 20:6627-6637.[Abstract/Free Full Text]

16. Lin, K. M., W. G. Zhao, J. Bhatnagar, W. D. Zhao, J. P. Lu, S. Simko, A. Schueneman, and G. E. Austin. 2001. Cloning and expression of human HBP1, a high mobility group protein that enhances myeloperoxidase (MPO) promoter activity. Leukemia 15:601-612.[CrossRef][Medline]

17. Liu, Z. G., H. Hsu, D. V. Goeddel, and M. Karin. 1996. Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NK-kappaB activation prevents cell death. Cell 87:565-576.[CrossRef][Medline]

18. Mendelson, K. G., L. R. Contois, S. G. Tevosian, R. J. Davis, and K. E. Paulson. 1996. Independent regulation of JNK/p38 MAP kinases bymetabolic oxydative stress in the liver. Proc. Natl. Acad. Sci. USA 93:12908-12913.[Abstract/Free Full Text]

19. Molnar, A., A. M. Theodoras, L. I. Zon, and J. M. Kyriakis. 1997. Cdc42Hs, but not Rac1, inhibits seru-stimulated cell cycle progression at G1/S through a mechanism requiring p38/RK. J. Biol. Chem. 272:13229-13235.[Abstract/Free Full Text]

20. Park, D. S., L. Stefanis, C. Y. I. Yan, S. E. Farinelli, and L. A. Greene. 1996. Ordering the cell death pathway: differential effects of BCL2, an interleukin-1-concerting enzyme family protease inhibitor, and other survival agents on JNK activation in serum/nerve growth factor-deprived PC12 cells. J. Biol. Chem. 271:21898-21905.[Abstract/Free Full Text]

21. Pearson, G., F. Robinson, T. Beers Gibson, B. E. Xu, M. Karandikar, K. Berman, and M. H. Cobb. 2001. Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr. Rev. 22:153-183.[Abstract/Free Full Text]

22. Pruitt, K., W. M. Pruitt, G. K. Bilter, J. K. Westwick, and C. J. Der. 2002. Raf-independent deregulation of p38 and JNK mitogen-activated protein kinases are critical for Ras transformation. J. Biol. Chem. 277:31808-31817.[Abstract/Free Full Text]

23. Raingeaud, J., S. Gupta, J. S. Rogers, M. Dickens, J. Han, R. J. Ulevitch, and R. J. Davis. 1995. Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J. Biol. Chem. 270:7420-7426.[Abstract/Free Full Text]

24. Raingeaud, J., A. J. Whitmarsh, T. Barrett, B. Derijard, and R. J. Davis. 1996. MKK3- and MKK6-regulated gene expression is mediated by the p38 mitogen-activated protein kinase signal transduction pathway. Mol. Cell Biol. 16:1247-1255.[Abstract]

25. Sampson, E. M., Z. K. Haque, M. C. Ku, S. G. Tevosian, C. Albanese, R. G. Pestell, K. E. Paulson, and A. S. Yee. 2001. Negative regulation of the Wnt-beta-catenin pathway by the transcriptional repressor HBP1. EMBO J. 20:4500-4511.[CrossRef][Medline]

26. Shih, H. H., S. G. Tevosian, and A. S. Yee. 1998. Regulation of differentiation by HBP1, a target of the retinoblastoma protein. Mol. Cell Biol. 18:4732-4743.[Abstract/Free Full Text]

27. Shih, H. H., M. Xiu, S. P. Berasi, E. M. Sampson, A. Leiter, K. E. Paulson, and A. S. Yee. 2001. HMG box transcriptional repressor HBP1 maintains a proliferation barrier in differentiated liver tissue. Mol. Cell Biol. 21:5723-5732.[Abstract/Free Full Text]

28. Tevosian, S. G., H. H. Shih, K. G. Mendelson, K. A. Sheppard, K. E. Paulson, and A. S. Yee. 1997. HBP1: a HMG box transcriptional repressor that is targeted by the retinoblastoma family. Genes Dev. 11:383-396.[Abstract/Free Full Text]

29. Tournier, C., P. Hess, D. D. Yang, J. Xu, T. K. Turner, A. Nimnual, D. Bar-Sagi, S. N. Jones, R. A. Flavell, and R. J. Davis. 2000. Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway. Science 288:870-874.[Abstract/Free Full Text]

30. Wang, X., C. H. McGowan, M. Zhao, L. He, J. S. Downey, C. Fearns, Y. Wang, S. Huang, and J. Han. 2000. Involvement of the MKK6-p38{gamma} cascade in {gamma}-radiation-induced cell cycle arrest. Mol. Cell. Biol. 20:4543-4552.[Abstract/Free Full Text]

31. Wang, X. Z., and D. Ron. 1996. Stress-induced phosphorylation and activation of the transcription factor CHOP (GADD153) by p38 MAP Kinase. Science 272:1347-1349.[Abstract]

32. Whitmarsh, A. J., C. Y. Kuan, N. J. Kennedy, N. Kelkar, T. F. Haydar, J. P. Mordes, M. Appel, A. A. Rossini, S. N. Jones, R. A. Flavell, P. Rakic, and R. J. Davis. 2001. Requirement of the JIP1 scaffold protein for stress-induced JNK activation. Genes Dev. 15:2421-2432.[Abstract/Free Full Text]

33. Wilkinson, M. G., and J. B. Millar. 2000. Control of the eukaryotic cell cycle by MAP kinase signaling pathways. FASEB J. 14:2147-2157.[Abstract/Free Full Text]

34. Xu, M., K. A. Sheppard, C. Y. Peng, A. S. Yee, and H. Piwnica-Worms. 1994. Cyclin A/CDK2 binds directly to E2F-1 and inhibits the DNA-binding activity of E2F-1/DP-1 by phosphorylation. Mol. Cell Biol. 14:8420-8431.[Abstract/Free Full Text]

35. Yang, D. D., C. Y. Kuan, A. J. Whitmarsh, M. Rincon, T. S. Zheng, R. J. Davis, P. Rakic, and R. A. Flavell. 1997. Absence of excitotoxicity-induced apoptosis in the hippocampus of mice lacking the Jnk3 gene. Nature 389:865-870.[CrossRef][Medline]

36. Yew, P. R. 2001. Ubiquitin-mediated proteolysis of vertebrate G1- and S-phase regulators. J. Cell Physiol. 187:1-10.[CrossRef][Medline]

37. Zhuma, T., R. Tyrrell, B. Sekkali, G. Skavdis, A. Saveliev, M. Tolaini, K. Roderick, T. Norton, S. Smerdon, S. Sedgwick, R. Festenstein, and D. Kioussis. 1999. Human HMG box transcription factor HBP1: a role in hCD2 LCR function. EMBO J. 18:6396-6406.[CrossRef][Medline]

Molecular and Cellular Biology, December 2003, p. 8890-8901, Vol. 23, No. 23
0270-7306/03/$08.00+0     DOI: 10.1128/MCB.23.23.8890-8901.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles: