Direct transcriptional activation of human caspase-1 by tumor suppressor p53.

The tumor suppressor protein p53 is a sequence-specific DNA-binding protein, and its biological responses are very often mediated by transcriptional activation of various target genes. Here we show that caspase-1 (interleukin-1beta converting enzyme), which plays a role in the production of proinflammatory cytokines and in apoptosis, is a transcriptional target of p53. Caspase-1 mRNA levels increased upon overexpression of p53 by transfection in MCF-7 cells. Human caspase-1 promoter showed a sequence homologous to the consensus p53-binding site. This sequence bound to p53 in gel shift assays. A caspase-1 promoter-reporter construct was activated 6-8-fold by cotransfection with normal p53 but not by mutant p53 (His(273)) in HeLa, as well as MCF-7, cells. Mutation of the p53-binding site in caspase-1 promoter abolished transactivation by p53. Treatment of p53-positive MCF-7 cells with the DNA-damaging drug, doxorubicin, which increases p53 levels, enhanced caspase-1 promoter activity 4-5-fold, but similar treatment of MCF-7-mp53 (a clone of MCF-7 cells expressing mutant p53) and p53-negative HeLa cells with doxorubicin did not increase caspase-1 promoter activity. Doxorubicin treatment increased caspase-1 mRNA levels in MCF-7 cells but not in MCF-7-mp53 or HeLa cells. These results show that endogenous p53 can regulate caspase-1 gene expression.

The tumor suppressor protein p53 plays an important role in mediating response to stress such as that induced by DNA damage and hypoxia resulting in either growth arrest or apoptosis (1)(2)(3). It is a sequence-specific DNA-binding protein, and its biological effects are generally mediated by transcriptional activation of various target genes (1)(2)(3). The p53 gene is mutated in over 50% of human tumors and in some inflammatory disorders like rheumatoid arthritis (2)(3)(4). These p53 mutations are clustered in the sequence-specific DNA-binding domain of the molecule leading to inactivation of its sequence-specific transactivation function (2).
Caspase-1, also known as interleukin-1␤ converting enzyme, is a member of the cysteine protease family, which cleaves cellular substrates after aspartic acid (5)(6)(7). The primary function of caspase-1 is the proteolytic processing of the precursors of proinflammatory cytokines such as interleukin-1␤ into active cytokines (5)(6)(7). In addition caspase-1 is also involved in some forms of apoptosis (5)(6)(7). Caspase-1 knockout mice are developmentally normal but are defective in the production of mature cytokines interleukin-1␤ and interleukin-18. These mice are resistant to septic shock and show a partial defect in apoptosis (8,9).
Several p53-responsive genes have been identified by using different approaches and various cell types (10,11). These p53-responsive genes include various functional categories such as those involved in apoptosis, cell cycle, signal transduction, angiogenesis, etc. (11). Induction of various genes by p53 is dependent on the type of inducer used, and even with the same inducer it may be cell type-dependent (11). However none of the members of the caspase family have been identified as a transcriptional target of p53. Here we report that human caspase-1 is a transcriptional target of exogenous, as well as endogenous, p53. In addition we have identified a site in the caspase-1 promoter that is required for transcriptional activation by p53.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfections-The cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum at 37°C in a CO 2 incubator. The transfections were done using LipofectAMINE PLUS TM reagent (Life Technologies, Inc.) according to manufacturer's instructions. All the plasmids for transfections were prepared by using Qiagen columns.
CAT Assay-Cells grown in 35-mm dishes were transfected with 250 ng of pCAT-ICE, 150 ng of pCMV⅐SPORT-␤GAL (Promega), and 500 ng of wild-type p53, mutant p53 (His 273 ), or control plasmids. Lysates were prepared 30 h post-transfection from HeLa cells and 48 h post-transfection from MCF-7 cells using reporter lysis buffer from Promega according to the manufacturer's instructions. For CAT assay 40 l of lysate was mixed with 2 l of 14 C-labeled chloramphenicol (25 Ci ml Ϫ1 ; 54 mCi mmol Ϫ1 ) and 10 l of acetyl coenzyme A (3.5 mg ml Ϫ1 ) in a total volume of 60 l and incubated at 37°C for 3 h. Relative CAT activities were calculated after normalizing with ␤-galactosidase enzyme activities.
Electrophoretic Mobility Shift Assay-Double-stranded oligonucleotide corresponding to the putative p53-binding site in caspase-1 pro-* This work was supported by a research grant from the Department of Biotechnology, Government of India (to G. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Reporter Plasmids-The promoter region of human caspase-1 gene from nucleotide position Ϫ182 to ϩ42 relative to the transcription start site was cloned in pCAT-Basic vector (Promega) and designated as pCAT-ICE-wt (15). Mutated promoter-reporter plasmid named as pCA-T-ICE-mt was constructed using primers mut-1, 5Ј-GGGAAAAGAAA-TAAAGAAATTCATATGAATTCACAGTGAGTATTTCC-3Ј, and mut-2, 5Ј-GGAAATACTCACTGTGAATTCATATGAATTTCTTTATTTCTTTT-CCC-3Ј, by PCR-based site-directed mutagenesis using overlap extension PCR (16). The nucleotide sequence of the mutant, as well as wild-type promoter, in these constructs was confirmed by automated sequencing.

RESULTS AND DISCUSSION
The level of caspase-1 mRNA was determined by RT-PCR analysis in response to transient overexpression of human wild-type p53 in MCF-7 cells. Caspase-1 mRNA level increased severalfold by overexpression of p53 as compared with the control-transfected cells or untransfected cells (Fig. 1A). This increase in the caspase-1 mRNA level was not the result of induction of apoptosis by p53, because treatment of MCF-7 cells with some apoptosis-inducing agents, staurosporine, and cycloheximide did not increase the caspase-1 mRNA level (Fig.  1B). Treatment with staurosporine in fact decreased the level of caspase-1 mRNA. There are five isoforms of caspase-1 mRNA (17). Using another set of primers, we found that the ␣ form, which is proapoptotic, was induced by p53 (Fig. 1, C and D), and ␤, ␥, and ␦ forms were not induced. By using appropriate primers we found that the ⑀-isoform was also not induced (data not shown).
Examination of the nucleotide sequence of human caspase-1 promoter (18) showed a sequence homologous to the consensus p53-binding site at nucleotide position Ϫ85 to Ϫ66 relative to the transcriptional start site (Fig. 2A). A caspase-1 promoterreporter construct (pCAT-ICE-wt) containing this region (nucleotide position Ϫ182 to ϩ42) was activated over 6 -8-fold by cotransfection with normal p53 in MCF-7, as well as HeLa, cells (Fig. 2, B and C). In these experiments the ratio of p53 to reporter plasmid was 1:1 with HeLa cells (Fig. 2C) and 2:1 with MCF-7 cells (Fig. 2B). At a higher ratio (2:1) of p53 to reporter plasmid in HeLa cells there was an over 12-fold increase in activation of transcription from this promoter (Fig. 2D). Mutant p53 (His 273 ) did not activate this transcription in p53negative HeLa cells, but in MCF-7 cells, which are p53-positive, it gave a small (less than 2-fold) increase in activity (Fig.  2, B and C). The control plasmid (pCAT-Basic) gave much lower activity and did not show any activation by p53 (data not shown). Mutation of the putative p53-binding site in caspase-1 promoter completely abolished transactivation by p53 (Fig.  2D). These observations suggest that there is only one functional p53-responsive site in this region (Ϫ182 to ϩ42) of caspase-1 promoter.
To determine whether p53 binds to the putative p53-binding site in human caspase-1 promoter, we carried out electro- Transcriptional Activation of Caspase-1 by p53 10586 phoretic mobility shift assays using a synthetic oligonucleotide corresponding to this site (Fig. 3A). Binding to this oligonucleotide was seen with nuclear extract prepared from MCF-7 cells treated with doxorubicin, which is known to increase the p53 protein level (Fig. 3B, lane 2). This binding was competed out with a 50-fold excess of unlabeled self-oligonucleotide and also with a consensus p53-binding oligonucleotide but not with a mutated oligonucleotide in which the p53-binding core sequence CATG was mutated to AATT (Fig. 3, A and B, lanes [3][4][5]. A polyclonal antibody to p53 immunodepleted the shifted band (Fig. 3B, lane 6). These results suggest that the binding to this oligonucleotide corresponding to the putative p53-binding site in caspase-1 promoter is specific and dependent on p53.
To address the role of endogenous p53 in regulating endogenous caspase-1 gene expression, MCF-7 cells were treated with doxorubicin, which increases the level of p53 protein. Treatment of MCF-7 cells with doxorubicin enhanced the caspase-1 mRNA level 4 -5-fold (Fig. 4). Similar treatment of MCF-7-mp53, a clone of MCF-7 cells expressing mutant p53 (His 273 ) or p53-negative HeLa cells, did not increase the caspase-1 mRNA level (Fig. 4). The basal level of caspase-1 mRNA was higher in MCF-7-mp53 that decreased upon treatment with doxorubicin.The MCF-7-mp53 cell line was obtained by transfection of MCF-7 cells with the His 273 mutant of p53 followed by selection with G418. This mutant of p53 is known to function as a dominant inhibitor of wild-type p53 function (19). Treatment of A549 cells (which have normal p53) with doxorubicin also resulted in an increase in the caspase-1 mRNA level (Fig. 4). These results showed that endogenous p53 can regulate expression of the endogenous caspase-1 gene.
To determine the role of endogenous p53 in regulating caspase-1 promoter, MCF-7, MCF-7-mp53, and HeLa cells were transfected with caspase-1 promoter-reporter plasmid, and after 24 h they were treated with doxorubicin for 40 or 48 h. Doxorubicin treatment resulted in a 4 -5-fold increase in caspase-1 promoter activity in MCF-7 cells but not in MCF-7-mp53 or HeLa cells (Fig. 5). These results showed that endogenous wild-type p53 can also activate transcription from the caspase-1 promoter, which is inhibited by the His 273 mutant of p53.
Ectopic expression of caspase-1 is known to induce apoptosis (17,20). The wild-type p53-induced apoptosis in MCF-7 cells was partially inhibited (50% inhibition) by YVAD-cmk (which preferentially inhibits caspase-1) but not by the caspase-3 family inhibitor DEVD-cmk (data not shown). Doxorubicin-induced apoptosis in MCF-7 cells was also partially inhibited (45% inhibition) by YVAD-cmk and not by DEVD-cmk (data not shown). These observations suggest that caspase-1 contributes in part to p53-mediated apoptosis. Apoptotic pathways are cell type-and stimulus-specific, and it is likely that caspase-1, along with other transcriptional targets, may play a role in p53-mediated apoptosis at least in some cells.
The primary role of caspase-1 is in the production of proinflammatory cytokines interleukin-1␤, interleukin-16, and interleukin-18 (5-7). Wild-type p53 is overexpressed in several inflammatory diseases (reviewed in Ref. 4), but its potential role in inflammation is not understood. Our results, showing that caspase-1 is transcriptionally activated by p53, suggest that p53 has a role in inflammation. Mutational inactivation of  Transcriptional Activation of Caspase-1 by p53 10587 p53 in human tumors would, therefore, lead to reduced inflammatory response, in addition to resistance to apoptosis.