This Article Abstract Full Text (PDF) Other Versions of this Article: MCB.00657-06v1 27/7/2713    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Song, L. Articles by Huang, C. Search for Related Content PubMed PubMed Citation Articles by Song, L. Articles by Huang, C.

p85 Acts as a Novel Signal Transducer for Mediation of Cellular Apoptotic Response to UV Radiation

Nelson Institute of Environmental Medicine, New York University School of Medicine, 57 Old Forge Rd., Tuxedo, New York 10987,¹ Pennington Biomedical Research Center, Louisiana State University, 6400 Perkins Rd., Baton Rouge, Louisiana 70808,² The Hormel Institute, University of Minnesota, 801 16th Ave. NE, Austin, Minnesota 55912,³ Department of Microbiology and Immunology, University of Western Ontario, Sibens-Drake Research Institute, 1400 Western Rd., London, Ontario N6G 2V4, Canada⁴

Received 15 April 2006/ Returned for modification 12 June 2006/ Accepted 29 December 2006

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion References

 
Apoptosis is an important cellular response to UV radiation (UVR), but the corresponding mechanisms remain largely unknown. Here we report that the p85 regulatory subunit of phosphatidylinositol 3-kinase (PI-3K) exerted a proapoptotic role in response to UVR through the induction of tumor necrosis factor alpha (TNF-) gene expression. This special effect of p85 was unrelated to the PI-3K-dependent signaling pathway. Further evidence demonstrated that the inducible transcription factor NFAT3 was the major downstream target of p85 for the mediation of UVR-induced apoptosis and TNF- gene transcription. p85 regulated UVR-induced NFAT3 activation by modulation of its nuclear translocation and DNA binding and the relevant transcriptional activities. Gel shift assays and site-directed mutagenesis allowed the identification of two regions in the TNF- gene promoter that served as the NFAT3 recognition sequences. Chromatin immunoprecipitation assays further confirmed that the recruitment of NFAT3 to the endogenous TNF- promoter was regulated by p85 upon UVR exposure. Finally, the knockdown of the NFAT3 level by its specific small interfering RNA decreased UVR-induced TNF- gene transcription and cell apoptosis. The knockdown of endogenous p85 blocked NFAT activity and TNF- gene transcription, as well as cell apoptosis. Thus, we demonstrated p85-associated but PI-3K-independent cell death in response to UVR and identified a novel p85/NFAT3/TNF- signaling pathway for the mediation of cellular apoptotic responses under certain stress conditions such as UVR.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

 
Exposure to UV radiation (UVR), a ubiquitous environmental stress factor, is implicated in many medical problems, such as sunburn and skin cancer. Multiple types of damage to the cellular components (DNA, RNA, and proteins) can be observed upon UVR exposure (1, 24, 60). Apoptosis is an important response against oncogenesis by eliminating genetically damaged cells. Thus, the deregulation of the apoptotic response to UV radiation contributes greatly to the process of UVR-induced carcinogenesis (44, 60).

In addition to the direct cellular damage, UVR induces the production and release of some inflammatory cytokines, such as tumor necrosis factor alpha (TNF-) (33, 42, 44), interleukin-1 (IL-1) (5), and IL-6 (27). TNF- induction has been demonstrated to be an acute-phase response to UVR and contributes to UVR-induced cell apoptosis (29, 42, 44). TNF- was originally established as an important cytokine for the mediation of the immune response, and the corresponding mechanism for the modulation of TNF- gene expression in the immune system has been well characterized (10, 16, 48, 49). Many of the early studies have demonstrated that the nuclear factor of activated T cells (NFAT) is the key factor responsible for TNF- induction in T and B lymphocytes exposed to various stimuli (10, 13, 48, 49).

NFAT is a transcription factor family that includes five members: NFAT1 (NFATp or NFATc2), NFAT2 (NFATc or NFATc1), NFAT3 (NFATc4), NFAT4 (NFATc3 or NFATx), and NFAT5 (Ton EBP). Different NFAT members have distinct tissue expression profiles. NFAT1 and NFAT2 are expressed predominantly in the lymphocytes and have been proven to play a pivotal role in human TNF- gene transcription in response to various types of stimulation (13, 48, 49). NFAT4 is also detected in cells derived from the immune system, and the overexpression of NFAT4 in transgenic mice leads to enhanced TNF- expression in T cells (7). NFAT5 is distinct from the other four NFAT family members and is involved in the induction of TNF- gene transcription during hypertonic stress alone (12). Unlike other NFAT family members, NFAT3 is preferentially expressed in the nonlymphoid tissues. Whether this factor plays a regulatory role in TNF- expression in other somatic cells outside the immune system remains to be elucidated.

Class IA phosphatidylinositol 3-kinase (PI-3K) is a central component for transducing signals essential for multiple cellular processes including cell proliferation, differentiation, motility, and survival. PI-3K is a heterodimer consisting of a 110-kDa catalytic subunit (p110) and a regulatory subunit. To date, five regulatory subunits of PI-3K in mammalian cells have been identified, including two 85-kDa proteins (p85 and p85ß), two 55-kDa proteins (p55 and p55), and one 50-kDa protein (p50) (14, 15). Of these isoforms, p85 is predominantly and ubiquitously expressed in most tissues and is thought to be the major element of response to most stimuli (13). One established role for the regulatory subunit of PI-3K is to recruit the p110 catalytic subunit to the membrane and then catalyze the conversion of phosphatidylinositol 4,5,-bisphosphate into the second massager molecule phosphatidylinositol 3,4,5-triphosphoate, which activates a wide range of the downstream targets, including the serine/threonine kinase Akt (14, 15). Although the PI-3K/Akt signaling pathway mediates cell survival under multiple stress conditions (14, 32), its role in the determination of cell fates in response to UVR appears to be cell type dependent (9, 40, 41, 54, 58).

In addition to forming a complex with the p110 catalytic subunit, p85 exists in a monomeric form due to the greater abundance of p85 than of p110 in many cell types (51). It has been proven that monomeric p85 is able to act as a mediator for transducing the insulin-like growth factor 1-dependent cellular response (33). Moreover, p85 is also involved in the apoptotic response under oxidative stress in a PI-3K-independent manner (57). Therefore, p85 is a multifunctional protein and its role under stress conditions may go beyond its effect through the well-known PI-3K pathway.

Here we addressed a novel function of p85 in UVR response by the genetic abolishment or targeted disruption of p85 expression in mouse embryonic fibroblasts (MEFs). Our results reveal a p85-dependent but PI-3K-independent apoptotic response upon UVR and further identify a novel p85/NFAT3/TNF- pathway for the specific mediation of type B UV (UVB) radiation-induced cell death. This is the first time, to the best of our knowledge, that the role of p85 in the activation of NFAT3 in response to UVR has been established, and this role may have medical implications in the prevention of UVR-related diseases.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Plasmids, antibodies, and reagents. The TNF--luciferase reporter plasmid, in which the transcription of the luciferase reporter gene is driven by the upstream 5'-flanking region of the TNF- gene promoter (nucleotides –1260 to +60), has been described previously (56, 59). The NF-B-, AP-1-, and NFAT-luciferase reporter plasmids were also described previously (23, 55). The expression plasmid containing full-length mouse p85 cDNA (pSR-p85) and the empty vector pSR were kindly provided by W. Ogawa (Kobe University School of Medicine, Japan) and described previously (19). The plasmids containing cDNA for the dominant negative mutant of PI-3K (p85), a deletion mutant of p85 lacking the binding site with the p110 catalytic subunit and thus blocking PI-3K activity, were described in previous studies (19, 30). The adenoviral plasmid expressing the dominant negative mutant of Akt (AD-DN-Akt/K179M), which has a kinase-dead mutation with a single-amino-acid replacement of lysine 179 (K179M), was constructed and kindly provided by Bing-Hua Jiang (West Virginia University, Morgantown, WV). The adenovirus was produced and amplified by transfecting HEK 293 cells with the plasmids. The antibodies used in the Western blot and super gel shift assays were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and Cell Signal Technology (Beverly, MA), respectively. Cyclosporine A (CsA) and wortmannin were purchased from Sigma (St. Louis, MO).

Cell culture. p85^+/+ and p85^–/– MEFs isolated from wild-type and p85^–/– mice were kindly provided by Y. Terauchi and T. Kadowaki (University of Tokyo, Tokyo, Japan) and were described in a previous study (47). The TNF- receptor I (TNF-RI)^–/–, TNF-RII^–/–, TNF-RI/II^–/– MEFs and the corresponding wild-type MEFs were the primary cultures derived from the TNF-RI, TNF-RII, and TNF-RI/II gene knockout mice and the wild-type mice, which were kindly provided by Sung O. Kim (University of Western Ontario, Canada). The MEFs were maintained in Dulbecco's modified Eagle medium (DMEM; Calbiochem, San Diego, CA) supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin, and 2 mM L-glutamine (Life Technologies, Inc., Rockville, MD).

siRNA. The small interfering RNAs (siRNAs) targeting the mouse p85, mouse TNF-, mouse NFAT3, and human COX-2 genes were designed with the aid of the siRNA Target Finder (http://www.ambion.com/techlib/misc/siRNA_finder.html; Ambion Inc., Austin, TX), and the corresponding expression plasmids were constructed using the GeneSuppressor system (Imgenex Co., San Diego, CA). The target sequence of mouse TNF- siRNA was 5'-AATTCGAGTGACAAGCCTGTA-3', the target sequence of human COX-2 siRNA was 5'-AGACAGATCATAAGCGAGGA-3', and the target sequence of mouse p85 siRNA was 5'-TCGAAGGACTGGAATGTTCGACTCT-3'. The target sequence of mouse NFAT3 siRNA was described previously (31). The siRNA oligonucleotides specifically targeting the mouse TNF-RI and TNF-RII genes were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Cell transfection. All of the stable and transient transfections were performed with Lipofectamine reagents (Life Technologies, Inc., Rockville, MD) according to the manufacturer's instructions. To stably transfect the p85^+/+ and p85^–/– MEFs with the luciferase reporter constructs, the luciferase reporter plasmids (expressing TNF-, NF-B, AP-1, or NFAT) were introduced into the cells together with lacZ-hygro⁺, a vector conferring hygromycin resistance. The resulting drug-resistant stable transfectants were identified by using hygromycin B selection. For transient transfection, the cells were transfected with the luciferase reporter constructs and then subjected to the next-step exposure 30 h after transfection. To disrupt the expression of p85, TNF-, or NFAT3, p85^+/+ MEFs were transfected with the corresponding siRNA recombinant construct and the stable transfectants were identified by using G418 selection. To knock down the expression of TNF-RI and/or TNF-RII, the p85^+/+ MEFs were transfected with the commercially available TNF-RI or TNF-RII siRNA oligonucleotides either separately or in combination. The cells were then harvested 48 h later and subjected to the next-step stimulations. For the restoration of p85 expression in the p85^–/– MEFs, the cells were cotransfected with the plasmid pSR-p85 and lacZ-hygro⁺, and the stable transfectants were obtained by using hygromycin B selection. The wild-type MEFs with p85 overexpression were established by cotransfection of the MEFs with the p85 construct and the lacZ-hygro⁺ vector. To interfere with the activity of Akt, the wild-type MEFs were infected with AD-DN-Akt/K179M and the cells were then exposed to UVB radiation 36 h after infection with the virus.

RT-PCR. Total RNA was extracted with TRIzol reagent (Gibco), and cDNA was synthesized with the ThermoScript reverse transcription-PCR (RT-PCR) system (Invitrogen, Carlsbad, CA). For the detection of TNF- induction, a pair of oligonucleotides (5'-CCAGACCCCACACTCAGAT-3' and 5'-AACACCCATTCCCTCACAG-3') were synthesized as described in a previous report (18) and used as the specific primers to amplify mouse TNF- cDNA. The mouse ß-actin cDNA was amplified by using the primers 5'-GACGATGATATTGCCGCACT-3' and 5'-GATACC ACGCT TGCTCTGAG-3'.

Cell apoptosis assay. Cells seeded in 6-well plates (2 x 10⁵/well) were starved by replacing the medium with 0.1% FBS-DMEM 24 h prior to UV radiation. Culture plates were then covered with a thin layer of fresh medium (0.1% FBS-DMEM) and exposed to UVR. Control cultures were identically processed but not irradiated. The UVB light source was purchased from UVP Inc. (Upland, CA) and emitted >95% 302-nm-wavelength UVB light. Exposure to UVB radiation for 1, 2, or 4 min corresponds to a dose of 1, 2, or 4 kJ/m² of UVB radiation. After the indicated time periods, the UVR-treated and control cells were harvested and fixed in 75% ethanol and then stained with propidium iodide solution (50 µg/ml) in the presence of RNase A. Propidium iodide is excluded by the viable cells, but it can penetrate the membranes of the apoptotic or dead cells. The apoptotic profile was determined by flow cytometry utilizing a Beckman-Coulter EpicsXL flow cytometer on the FL3 channel, and gating was set to exclude debris and cellular aggregates. Twenty thousand events were counted for each analysis, and two to four independent experiments for each group were conducted. The subdiploid DNA peak, immediately adjacent to the G₀/G₁ peak, represented apoptotic cells and was quantified by histogram analyses.

Western blot assay. Cellular protein extracts were prepared with a cell lysis buffer (10 mM Tris-HCl [pH 7.4], 1% sodium dodecyl sulfate, 1 mM Na₃VO₄) and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The membranes were probed with the primary antibodies indicated in the figures and the alkaline phosphatase-conjugated second antibody. Signals were detected by using the ECF Western blotting system as described in previous reports (24, 30).

Luciferase reporter assay. MEFs transiently or stably transfected with the luciferase reporter constructs were seeded into 96-well plates (8 x 10³/well) and subjected to the various treatments when cultures reached 80 to 90% confluence. Cellular lysates were prepared at the time points indicated in the figures, and the luciferase activities were determined by using a luminometer (Wallac 1420 Victor 2 multilabel counter system) as described in previous studies (24, 30). The results are expressed as relative activities normalized to the luciferase activity in the control cells without treatment.

Electrophoretic mobility shift assays (EMSA). Nuclear proteins were prepared with the CelLytic-NuCLEAR extraction kit (Sigma, St. Louis, MO) according to the manufacturer's protocols. Five micrograms of nuclear protein was subjected to the gel shift assay by incubation with 1 µg of poly(dI-dC) DNA carrier in DNA binding buffer (10 mM Tris [pH 8.0], 150 mM KCl, 2 mM EDTA, 10 mM MgCl₂, 10 mM dithiothreitol, 0.1% bovine serum albumin, 20% glycerol) in a final volume of 10 µl on ice for 10 min. Then a volume corresponding to 10⁵ cpm (10⁸ cpm/µg) of the ³²P-labeled double-stranded oligonucleotide (2 µl) was added, and the reaction mixture was incubated at room temperature for 30 min. For competition experiments, 20-fold molar excesses of the unlabeled oligonucleotides or the irrelevant cold probes were added before the addition of the probe. For the super gel shift assay, nuclear extracts were incubated with 2 µg of either the preimmune serum or the specific anti-NFAT3 antibody for 30 min at 4°C before the addition of the probe. DNA-protein complexes were resolved by electrophoresis on 5% nondenaturing glycerol-polyacrylamide gels.

The following synthetic oligonucleotides (5' to 3') were used as probes or competitors in EMSA experiments (core sequences are underlined): GCCCAAAGAGGAAAATTTGTTTCATACAG (NFAT site on the murine IL-2 promoter), GGCCCAAAGACCTTTATTTGTTTCATACAG (mutated NFAT site on the murine IL-2 promoter), CCCCAACTTTCCAAACCCT (nucleotides –185 to –167 containing the putative NFAT site on the mouse TNF- promoter; designated probe TNF--NFAT1), GGAAGTTTTCCGAGGGTTGAATGA (nucleotides –79 to –56 containing the putative NFAT site on the mouse TNF- promoter; designated probe TNF--NFAT2), GAACATGTCTTGACATGTTC (p53 site on the mouse p21 promoter), GAGCCGCAAGTGACTCAGCGCGGGCG (AP-1 consensus sequence), and GAGTTGAGGGGACTTTCCCAGGC (NF-B consensus sequence).

Site-directed mutagenesis. Site-directed mutagenesis was performed with the QuikChange II XL site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's protocol. The following synthetic oligonucleotide primers (lowercase letters indicate the mutated positions) were used for the construction of the mutants: 5'-CTGGGTGTCCCCCAACgTTaaAAACC CTCTGCCCC-3' (designated –1260 –178) and 5'-CTCCACCAAGGAAGTTagaCGAGGGTTGAATGAGAGC-3' (designated –1260 –72). The nucleotide sequences of the mutants were confirmed by automatic DNA sequencing.

ChIP assay. The chromatin immunoprecipitation (ChIP) assay was performed with the EZ ChIP kit (Upstate) according to the manufacturer's protocol. Briefly, cells were untreated or treated with UVB radiation (1 kJ/m²) for 12 h and then genomic DNA and the proteins were cross-linked with 1% formaldehyde. The cross-linked cells were pelleted, resuspended in lysis buffer, and sonicated to generate 200- to 500-bp chromatin DNA fragments. After centrifugation, the supernatants were diluted 10-fold and then incubated with anti-NFAT3 antibody or the control rabbit immunoglobulin G at 4°C overnight. The immune complex was captured by protein G agarose saturated with salmon sperm DNA and then eluted with the elution buffer. DNA-protein cross-linking was reversed by heating at 65°C for 4 h. DNA was purified and subjected to PCR analysis. To specifically amplify the region containing the putative NFAT-responsive elements on the mouse TNF- promoter, PCR was performed with the following pair of primers: 5'-CGATGGAGAAGAAACCGAGACAGA-3' (forward) and 5'-AGGGAGCTTC TGCTGGCTGGCTGT-3' (reverse). The primers that targeted the region 1 kb upstream of the NFAT binding sites on the TNF- promoter were also used in the PCR analysis to support the specificity of the ChIP assay: 5'-CCTGCACCATCTGTGAAACCCAAT-3' (forward) and 5'-AATCCACCATGTC TCTGGGAGCTG-3' (reverse).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Genetic evidence that p85 plays a key role specifically in UVR-induced apoptosis. p85 has been shown to perform multiple functions dependent on or independent of PI-3K activity (13, 33, 57). To investigate the role of p85 in responses to UVR, primary MEFs isolated from the p85 gene knockout mice and the corresponding wild-type mice with an otherwise identical genetic background were used in the present study.

Due to the alternative splicing mechanism, the p85 gene (Pik3r1) encodes three protein isoforms, including the full-length p85 and the two truncated splicing variants, p50 and p55, which share identical C-terminal sequences with p85 but have the N-terminal section of the p85 polypeptide replaced by unique sequences of 6 and 34 amino acids, respectively (25). The p85^–/– MEFs are derived from the mice with a single deletion of exon 1A (containing the initiation codon for p85) of the Pik3r1 gene, which results in the selective abolishment of p85 expression without disrupting the p50 and p55 splicing variants corresponding to the intact exons 1B and 1C (containing the initiation codons for p50 and p55, respectively) (47). We also confirmed the specific absence of p85 expression in p85^–/– MEFs, in which the levels of p50 and p55 were normal, by Western blot assay with the antibody raised against the common C terminus of p85, p50, and p55 (Fig. 1A). Another isoform of the 85-kDa regulatory subunit of PI-3K, p85ß, was encoded by a distinct sequence of cDNA, and its expression level was not altered by the selective disruption of the p85 gene (47).

View larger version (41K): [in this window] [in a new window]   FIG. 1. p85 exerted proapoptotic activity specifically in response to UVR. (A) Identification of p85, p50, and p55 expression in p85+/+ and p85–/– MEFs. (B) p85+/+ and p85–/– MEFs were exposed to different doses of UVB radiation, and cell apoptosis was determined by flow cytometry analysis 24 h later. AP, apoptosis; +, present; –, absent. (C) Quantification data from three independent experiments described in the legend to panel B. (D) p85+/+ and p85–/– MEFs were treated with arsenite (10 µM), and the level of cell apoptosis was determined 12 and 24 h later. Results from a representative experiment and the percentage of the apoptotic cells in each group are presented. (E) p85+/+ and p85–/– MEFs were exposed to different doses of UVB radiation, and the cleavage of caspase 3 and PARP was detected 12 h later. (F) Forced expression of p85 partially restored the wild-type response in p85–/– MEFs (P85–/–/P85). (G) p85+/+ and p85–/– MEFs, p85–/– MEFs transfected with a vector, and p85–/– MEFs expressing p85 were exposed to UVB radiation (2 kJ/m2), and cell apoptosis was detected 24 h later. A representative result and the percentage of the apoptotic cells in each group are presented. (H) p85–/– MEFs transfected with a vector and p85–/– MEFs expressing p85 were exposed to different doses of UVB radiation, and the cleavage of caspase 3 and PARP was detected 12 h later.

To rule out the possibility that the resistance to UVR-induced apoptosis in p85^–/– MEFs was due to gene changes other than that resulting in the p85 deficiency during the establishment of the p85^–/– MEFs, p85^–/– MEFs were transfected with a plasmid containing the full-length p85 cDNA for stable expression (Fig. 1F). Compared to the control vector transfectant, p85^–/– MEFs with the forced expression of p85 showed a partially restored apoptotic response under UVR (Fig. 1G). Consistently, UVR-induced caspase 3 and PARP cleavage was also present in the p85^–/– cells in which p85 had been reintroduced (Fig. 1H). These data clearly demonstrate that p85 acts as a mediator specifically for UVR-induced cell apoptosis in MEFs.

The impaired apoptotic response in p85-null MEFs upon UVR was mediated by a PI-3K-independent pathway. The PI-3K-dependent pathway has been proven to be a critical cell survival cascade under multiple stress conditions (13, 32), but its role in the response to UVR remains controversial (9, 40, 41, 54, 58). Optimal signaling through PI-3K depends on a critical molecular balance of the regulatory and catalytic subunits (50, 51). Therefore, p85 deficiency might lead to the deregulation of the PI-3K-dependent pathway and affect the cell apoptotic response. To assess this possibility, we detected the activation status of Akt and p70^S6k, two important downstream targets of PI-3K that our group has proven to be involved in the response to UVR in another mouse cell line (22, 24, 54). As shown in Fig. 2A, no significant difference in UVR-induced Akt or p70^S6k phosphorylation between the p85-null and the wild-type MEFs was observed. Glycogen synthase kinase 3ß (GSK3ß), a downstream target of Akt in many cellular responses (36), could also be induced by UVB radiation in both p85^+/+ and p85^–/– cells at comparable levels (Fig. 2A). These data suggest that p85 deficiency does not affect the activation status of the PI-3K-dependent pathway in the UVR response.

View larger version (55K): [in this window] [in a new window]   FIG. 2. The resistance of p85–/– MEFs to UV-induced apoptosis was unrelated to PI-3K-dependent pathway activation. (A) p85+/+ and p85–/– MEFs were exposed to UVB radiation (1 kJ/m2), and the phosphorylation statuses of Akt (P-Akt), GSK3ß (P-GSK3ß), and p70S6k (P-p70S6k were detected by a Western blot assay at the indicated time periods after UVR exposure. (B) p85+/+ MEFs were infected with the adenovirus expressing DN-Akt or that expressing GFP (AD-GFP). The overexpression of DN-Akt or GFP in p85+/+ MEFs was detected 36 h postinfection. (C) Cells described in the legend to panel B were subjected to UVB radiation exposure 36 h after infection with the virus, and the activation of GSK3ß and the inducible cleavage of caspase 3 were detected 12 h later. (D) Cells described in the legend to panel B were subjected to UVB radiation exposure, and the incidence of cell death was determined by flow cytometry analysis 30 h after radiation. AP, apoptosis; +, present; –, absent. (E) p85+/+ MEFs stably transfected with the p85 construct or the control vector (vc) were exposed to UVB radiation. The activation of Akt and GSK3ß and the inducible cleavage of caspase 3 were detected 12 h later. (F) Cells treated as described in the legend to panel E were subjected to flow cytometry analysis 30 h after UVR exposure to determine the percentage of apoptotic cells. (G) p85+/+ MEFs were pretreated with wortmannin (200 nM) or its carrier dimethyl sulfoxide (DMSO) for 1 h, followed by exposure to UV radiation. The activation of Akt and the inducible cleavage of caspase 3 and PARP were detected 12 h later. (H) Cells treated as described in the legend to panel G were subjected to flow cytometry analysis 30 h after UVR exposure.

p85-null MEFs were competent for the activation of JNKs and p38K in response to UVR. The c-Jun N-terminal kinases (JNKs) and p38 kinases (p38K) are two important protein kinases that mediate cell apoptosis induced by various stresses, including UVR (11). To test whether p85 transmits the apoptotic signal through these two pathways, UVR-induced phosphorylation of JNKs and p38K in the wild-type and p85^–/– MEFs was analyzed. As shown in Fig. 3, JNK and p38K phosphorylation under UVR exhibited the same dynamics in both p85^+/+ and p85^–/– MEFs. Thus, we exclude the involvement of JNKs and p38K-dependent pathways in p85-mediated cell apoptosis in response to UVR.

View larger version (88K): [in this window] [in a new window]   FIG. 3. Both p85+/+ and p85–/– MEFs are competent for UVB radiation-induced phosphorylation of JNKs (P-JNKs) and p38K (P-p38K). (A) Induction of JNK and p38K phosphorylation at different time periods after a single dose of UVB radiation (1 kJ/m2). (B) Induction of JNK and p38K phosphorylation 1.5 h after treatment with different doses of UVB radiation.

p85 mediated UVR-induced apoptosis by the induction of TNF- expression.

View larger version (44K): [in this window] [in a new window]   FIG. 4. p85-mediated UVB radiation-induced apoptosis by the induction of TNF- gene transcription. (A) p85+/+ and p85–/– MEFs stably transfected with TNF--luciferase reporter constructs were exposed to different doses of UVB radiation, arsenite, nickel chloride, or H2O2. The luciferase activities were determined 12 h (for UVB radiation and H2O2) or 24 h (for arsenite and nickel chloride) later. The results were expressed as the relative TNF- promoter activity normalized to the luciferase activity of the cells without any treatment. (B) p85+/+ and p85–/– MEFs were exposed to different doses of UVB radiation (upper panel) or H2O2 (lower panel), and the transcription of TNF- mRNA was detected by a RT-PCR assay 12 h later. The relative levels of expression of TNF- mRNA normalized to the expression in the ß-actin control are presented. (C and D) The reintroduction of exogenous p85 partially reactivated TNF- gene transcription in p85–/– MEFs [P85–/–(P85)] 24 h after UV radiation. TNF-Luc, TNF--luciferase reporter construct. (E) p85+/+ MEFs stably transfected with TNF--specific siRNA or the control siRNA were exposed to different doses of UVB radiation, and the inducible TNF- gene transcription was detected by a RT-PCR assay 12 h after UV radiation. (F) Stable transfection with TNF--specific siRNA decreased the expression of the TNF- precursor (30 kDa) in the wild-type cells. (G) Cells described in the legend to panel E were exposed to a single dose of UVB radiation (2 kJ/m2), and the level of cell apoptosis was determined 24 h later. AP, apoptosis; +, present; –, absent. (H) The cleavage of caspase 3 and PARP was inhibited in a stable transfectant with TNF- siRNA 12 h after UV radiation.

An earlier study revealed that p85 has an effect on the mediation of the apoptotic response to H₂O₂-induced oxidative stress (57). UV radiation also yields reactive oxidative species in vivo (28). Therefore, we wondered whether p85-mediated TNF- up-regulation was a common response to the oxidative stress or whether it was specific to certain UVR damaging signals. To clarify this issue, wild-type and p85-null MEFs stably transfected with TNF--luciferase reporter plasmids were treated with different oxidative reagents, including H₂O₂, arsenite, and NiCl₂ (4, 52). We found that all of these tested oxidative reagents failed to obviously induce the transcriptional activation of the TNF- promoter (Fig. 4A and B). Therefore, we conclude that the effect of p85 on TNF- transcriptional activation is specifically induced by UVR damage and is unrelated to the oxidative stress.

To further define whether the p85-mediated TNF- induction contributed to UVR-induced cell apoptosis, the wild-type MEFs were stably transfected with a plasmid expressing a specific siRNA targeting the mouse TNF- gene and a control irrelevant siRNA (human COX-2 siRNA). The efficacy of the stably expressed TNF- siRNA in the suppression of UVR-induced TNF- mRNA transcription was verified by a RT-PCR assay (Fig. 4E). Western blot analysis further demonstrated that the stable expression of TNF- siRNA caused a significant decline in the level of expression of the cellular TNF- precursor in the whole-cell extracts (Fig. 4F). However, we failed to detect the release of soluble TNF- into the cell culture supernatant upon UV radiation. Actually, the level of induction of TNF- proteins by UVB radiation was reported to be extremely low and usually beyond the limit of detection by regular immunoassays (46).We further provided evidence that inhibiting TNF- induction reduced the incidence of UVR-induced apoptosis (Fig. 4G) and partially inhibited caspase 3 and PARP cleavage in response to UVR (Fig. 4H) while there were no obvious changes in the apoptosis-resistant status of the p85-null cells with or without TNF- siRNA transfection (data not shown). These data indicate that the inducible p85-mediated TNF- gene transcription represents an apoptotic pathway in response to UVR.

It is generally believed that TNF- exerts its biological effects by binding with its cognate membrane TNF receptor (17). However, due to the extremely low level of the released TNF- induced by UVB radiation (as described here and in reference 46), we wondered if this trace amount of TNF- triggers cellular apoptosis through the classical TNF receptor-dependent manner in response to UVR. To address this issue, primary cultures of TNF-RI^–/–, TNF-RII^–/–, and TNF-RI/II^–/– MEFs (Fig. 5A) were exposed to UVR, and the incidence of cell apoptosis was detected by flow cytometry analysis. Unexpectedly, the knockdown of either membrane TNF-RI or TNF-RII or both of the two receptors rendered the cells more sensitive than the wild-type cells, not resistant to, UVR-induced apoptosis under various conditions (Fig. 5B). Consistently, the incidence of cell death in response to UV radiation also increased when the wild-type cells were transfected with TNF-RI- or TNF-RII-specific siRNA either separately or in combination compared with that among the control siRNA-transfected cells (Fig. 5C to H). Altogether, these results suggest that TNF- induced by UVR appears to mediate the prodeath effect independently of the membrane TNF receptor.

View larger version (48K): [in this window] [in a new window]   FIG. 5. Knockdown of TNF- receptor expression increased UVR-induced cell apoptosis. (A) Identification of TNF-RI–/–, TNF-RII–/–, and TNF-RI/II –/– MEFs. (B) TNF-RI–/–, TNF-RII–/–, and TNF-RI/II–/– MEFs and the corresponding wild-type MEFs were exposed to different doses of UVB radiation, and then the level of cell apoptosis was determined 24 h later. AP, apoptosis; +, present; –, absent. (C, E, and G) Wild-type cells were transfected with TNF-RI- or TNF-RII-specific siRNAs and the corresponding control siRNAs either separately (C and E) or in combination (G). The levels of expression of TNF-RI and/or TNF-RII were detected 48 h after transfection (D, F, and H). Cells described in the legend to panels C, E, and G were subjected to UVB radiation exposure 48 h after transfection, and the level of cell apoptosis was determined 24 h after UV radiation.

NFAT, but not NF-B, AP-1, or Egr-1, was involved in p85-mediated TNF- gene transcription in response to UVR.

View larger version (50K): [in this window] [in a new window]   FIG. 6. NFAT was involved in p85-mediated TNF- induction in response to UVR. (A) p85+/+ and p85–/– MEFs stably transfected with NF-B-luciferase reporter constructs were exposed to UV radiation, and the inducible luciferase activities were determined 12 h later. (B) p85+/+ and p85–/– MEFs were exposed to UVR (1 kJ/m2), and the cytosolic and nuclear extracts were prepared 12 h after radiation. The expression level of Egr-1 was monitored by a Western blot assay. (C) AP-1 transcriptional activity was determined to be similar to that of NF-B as described in the legend to panel A. (D) A Western blot assay showed comparable levels of phosphorylation of Jun (P-c-Jun), Fos, and ATF (P-ATF2) family members in response to UVR in p85+/+ and p85–/– MEFs. (E) p85+/+ and p85–/– MEFs and the p85–/– MEFs expressing p85 [P85–/–(P85)] were transiently transfected with the NFAT-luciferase reporter constructs (NFAT-Luc) and exposed to UVB radiation, and the NFAT transcriptional activity was determined 24 h later. (F) p85+/+ MEFs stably transfected with TNF--luciferase reporter constructs were pretreated with different doses of CsA for 2 h and then exposed to UV radiation (1 kJ/m2). The luciferase activities were determined 12 h later. (G) Cells described in the legend to panel F were pretreated with a single dose of CsA (5 µM) for 2 h and then exposed to different doses of UV radiation. The luciferase activities were determined 24 h later.

The process of UVR-induced NFAT3 nuclear translocation and the subsequent binding of NFAT3 to the TNF- promoter were regulated by p85 in MEFs. To address which NFAT family member might be involved in p85-regulated TNF- gene transcription in response to UVR, whole-cell extracts of p85^+/+ MEFs, p85^–/– MEFs, and the null mutant cells in which p85 expression was restored were subjected to the Western blot assay to detect the expression profiles of the distinct NFAT family members. As shown in Fig. 7A, NFAT3 is the predominant form of this family of proteins expressed in the wild-type MEFs. The deficiency of p85 led to a reduced NFAT3 basal level, and the return of p85 into the null mutant cells restored NFAT3 expression. Although the expression of NFAT4 also could be detected in both the wild-type and the null mutant cells, its role in the p85-regulated response appeared to be minimal because its level of expression was much lower than that of NFAT3. There was no detectable expression of NFAT1, 2, or 5 in the MEFs. These data suggest that NFAT3 may be the major NFAT family member that is involved in the p85-regulated response to UVR.

View larger version (69K): [in this window] [in a new window]   FIG. 7. NFAT3 nuclear translocation and the DNA binding ability of NFAT3 with the mouse TNF- gene promoter were regulated by p85 under UVR. (A) The expression profiles of the five NFAT family members in p85+/+ and p85–/– MEFs and the null mutant cells in which p85 was reintroduced were determined by a Western blot assay. Commercially available HeLa whole-cell extracts (for NFAT5) or Ramos nuclear extracts (for NFAT1 through 4) served as the positive controls for each antibody. (B) Cytosolic and nuclear extracts (indicated as C and N, respectively) from UVR-treated and untreated p85+/+ and p85–/– MEFs and the p85–/– MEFs expressing p85 were prepared and subjected to a Western blot assay to detect the nuclear translocation of NFAT3 under UV radiation. (C) p85+/+ MEFs were pretreated with different doses of CsA for 2 h, followed by exposure to UV radiation. The cytosolic or nuclear distribution of NFAT3 was determined 12 h after radiation. (D) Sequences containing two putative NFAT binding sites on the TNF- gene promoter were used as probes in EMSA experiments. The consensus NFAT binding sequence is underlined and italicized. (E and F) p85+/+ and p85–/– MEFs were exposed to UVR (1 kJ/m2). The nuclear extracts prepared 12 h after radiation were subjected to the gel shift assay with TNF--NFAT1 (E) or TNF--NFAT2 (F) as a probe. A super gel shift assay was performed in the presence of the antibody (Ab) specific for NFAT3 or the preimmune serum. IgG, immunoglobulin G; +, present; –, absent. (G) A 20-fold molar excess of the unlabeled TNF--NFAT2, IL-2/NFAT, mutant IL-2/NFAT, NF-B, p53, or AP-1 cold probe was added to the binding reaction mixtures of p85+/+ MEFs to determine the specific binding. (H) p85+/+ MEFs were pretreated with different doses of CsA for 2 h, followed by exposure to UV radiation. Nuclear extracts were prepared 12 h after radiation and subjected to a gel shift assay with the TNF--NFAT1 oligonucleotide as a probe. (I) p85+/+ and p85–/– MEFs and MEFs with p85 reintroduced [P85–/–(P85)] were exposed to UVB radiation, and the ability of NFAT to bind to the TNF- gene promoter was determined 12 h later by using the TNF--NFAT2 oligonucleotide as a probe. (J) Soluble chromatin prepared from UVB radiation-treated and -untreated p85+/+ and p85–/– MEFs and the p85–/– MEFs expressing p85 was subjected to a ChIP assay using normal serum or anti-NFAT3 antibody. Immunoprecipitated chromatin DNA was PCR amplified with primers that specifically annealed to the region flanking NFAT-responsive elements (indicated as NFAT-RE) or the sequence 1 kb upstream of the NFAT binding sites (indicated as the US sequence) on the TNF- gene promoter. (K) p85+/+ MEFs were stably transfected with the parental TNF--luciferase reporter construct (–1260) or the individual site-directed mutants (–1260 –178 and –1260 –72). The stable transfectants were treated with different doses of UVB radiation, and the luciferase activities were determined 24 h after radiation.

To further assess the link between NFAT3 activation and p85-mediated TNF- induction under UVR, EMSA experiments were performed to address whether the two putative NFAT binding sites, which contained the GGAAA NFAT core sequence and were located at positions –178 and –72 relative to the transcription start site of the TNF- gene (Fig. 7D), could act as the functional NFAT-responsive elements in the response to UVR. For this purpose, two oligonucleotides including these sequences were synthesized and used as probes (named TNF--NFAT1 and TNF--NFAT2) in a gel shift assay. As expected, both of the probes efficiently bound to the nuclear proteins from the UVB radiation-treated p85^+/+ MEFs (lanes 2 in Fig. 7E and F) but the formation of the retarded DNA-protein complexes was significantly inhibited in p85^–/– MEFs (lanes 4 in Fig. 7E and F). The specific binding of the nuclear proteins with these two probes was confirmed by competition experiments. As shown in Fig. 7G, the major retarded band detected by the TNF--NFAT2 probe in wild-type cells was efficiently outcompeted by a 20-fold molar excess of the unlabeled TNF--NFAT2 cold probe and the oligonucleotide containing the NFAT binding sequence on the murine IL-2 promoter but not by the same amount of the mutant IL-2/NFAT, NF-B, AP-1, and p53 cold probes. Similar results were also obtained in the TNF--NFAT1 cold probe competition assay (data not shown). When the nuclear proteins of UVR-treated wild-type cells were subjected to the super gel shift assay, we found that the addition of the specific anti-NFAT3 antibody to the reaction mixtures caused the supershift of the inducible complexes generated with both probes (lanes 6 in Fig. 7E and F) while the isotype control antibody did not alter the retarded complexes in the samples (lanes 5 in Fig. 7E and F), demonstrating the presence of NFAT3 in both DNA-protein complexes. In addition, CsA pretreatment efficiently inhibited the formation of the specific complexes in a dose-dependent manner (Fig. 7H and data not shown). The reintroduction of p85 significantly restored the binding of the nuclear extracts with the two probes in p85-null cells (Fig. 7I and data not shown). These data indicate that the interaction of the CsA-sensitive NFAT3 with the TNF- gene promoter was regulated by p85 under UVR.

To further confirm the results from EMSA experiments, we next performed the ChIP assay by using the specific antibody against NFAT3 or normal serum. The detection of the TNF- promoter among the antibody-captured genomic DNA fragments was performed by PCR amplification with primers designed to specifically recognize the region containing NFAT-responsive elements. As shown in Fig. 7J, efficient promoter amplification from the input samples from p85^+/+ and p85^–/– MEFs and the null mutant cells expressing p85 could be observed, with or without UV radiation. The NFAT3 antibody strongly coimmunoprecipitated with the target TNF- promoter region DNA from UVB radiation-treated p85^+/+ cells but not with that from the UVB radiation-untreated samples, indicating the inducible recruitment of NFAT3 to the endogenous TNF- promoter in the wild-type MEFs under the simulated conditions. Furthermore, no signal was observed in the control immunoglobulin G-immunoprecipitated samples, supporting the specific binding of NFAT3 with the TNF- promoter. In contrast, samples from UVB radiation-treated p85^–/– cells immunoprecipitated with anti-NFAT3 antibody did not show detectable amplification of TNF- promoter DNA while the immunoprecipitation of the chromatin from UVB radiation-simulated null mutant cells expressing p85 with anti-NFAT3 antibody led to a marked restoration of the enrichment with the target promoter DNA. Further specificity of this ChIP assay was demonstrated by the inability to detect the occupancy of NFAT3 on a region 1 kb upstream of the NFAT-responsive elements on the TNF- promoter. Thus, these results strongly demonstrate the p85-dependent recruitment of NFAT3 onto the endogenous TNF- promoter after UVB radiation in vivo.

Given the efficient association of the functional NFAT3 with the TNF- gene promoter region, we next determined the contribution of these NFAT3 binding sites to the overall transcriptional induction of TNF- by UVB radiation. To this end, site-directed mutagenesis was performed using the parental TNF--luciferase reporter construct (nucleotides –1260 to +60) as a template. As shown in Fig. 7K, mutation of either of the two NFAT binding sites caused a 75 to 90% reduction in the induction of TNF- gene transcription by UVB radiation, strongly confirming the critical role of NFAT3 in TNF- induction under UVR.

Knockdown of NFAT3 decreased TNF- gene transcription and cell apoptosis in response to UVR. To further confirm the contribution of NFAT3 activation to p85-mediated TNF- induction and cell apoptosis, wild-type MEFs were cotransfected with the specific NFAT3 siRNA and the TNF--luciferase reporter constructs. The knockdown of NFAT3 expression in the NFAT3 siRNA stable transfectants compared with the expression in control siRNA-transfected cells was verified by a Western blot assay (Fig. 8A). When the cells were exposed to UV radiation, TNF- promoter-driven luciferase activities were almost abolished by the disruption of NFAT3 expression (Fig. 8B). RT-PCR analysis further confirmed the inhibitory effect of NFAT3 siRNA on UV-induced TNF- gene transcription (Fig. 8C). Accordingly, cell apoptosis in response to UV radiation was also significantly reduced in NFAT3 siRNA-transfected cells, as evidenced by both flow cytometry and the inducible cleavage of caspase 3 (Fig. 8D and E). These data further demonstrate that NFAT3 is the critical mediator of UVB radiation-induced TNF- gene transcription and cell apoptosis.

View larger version (42K): [in this window] [in a new window]   FIG. 8. Knockdown of NFAT3 expression reduced UVB radiation-induced TNF- gene transcription and cell apoptosis. (A) p85+/+ MEFs were cotransfected with NFAT3 siRNA or the control siRNA and the TNF--luciferase reporter constructs, and the corresponding stable transfectants were identified. The inhibition of endogenous NFAT3 expression by NFAT3 siRNA transfection was confirmed by Western blot assay. (B and C) Cells described in the legend to panel A were exposed to UVB radiation, and the inducible transcription of the TNF- gene was assessed by using a luciferase reporter assay (B) and RT-PCR analysis (C). (D) Cells described in the legend to panel A were exposed to UVB radiation, and cell apoptosis was detected 30 h after radiation. AP, apoptosis; +, present; –, absent. (E) Cells described in the legend to panel A were exposed to UVB radiation, and the inducible cleavage of caspase 3 was detected 24 h after UVR exposure.

p85 siRNA blocked UVR-induced apoptosis by the inhibition of NFAT3 activity and TNF- gene transcription.

View larger version (36K): [in this window] [in a new window]   FIG. 9. p85 siRNA blocked UVR-induced apoptosis, the DNA binding activity of NFAT, and TNF- gene transcription. (A) The level of p85 expression was specifically knocked down by transfection with p85 siRNA, while the levels of p50 and p55 were not altered. (B) p85+/+ MEFs stably transfected with p85 siRNA and the control siRNA were exposed to UVB radiation, and then cell apoptosis was detected 30 h after radiation. AP, apoptosis; +, present; –, absent. (C) Cells described in the legend to panel B were treated with UVB radiation, and the nuclear extracts were prepared 12 h after UVR exposure and then subjected to the gel shift assay using the TNF--NFAT2 oligonucleotide as a probe. (D) Cells described in the legend to panel B were transiently transfected with the TNF--luciferase reporter construct and then exposed to UV radiation 30 h after transfection. Inducible TNF- gene transcription was detected 12 h after UVR exposure. (E) The transcription of TNF- mRNA was detected in the cells described in the legend to panel B by a RT-PCR assay 12 h after UV radiation.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

It has been demonstrated that p85 is a multifunctional protein and serves as a critical mediator in various physiological processes, dependent on or independent of PI-3K activity (33, 43, 45, 51, 57). One previous study has revealed the involvement of p85 in the p53-mediated apoptotic response to oxidative stress, which is unrelated to the activation of the PI-3K signal transduction pathway, suggesting the potential role of p85 in transmitting cell damage signaling (57). Our data in the present study provide further strong evidence for a p85-dependent but PI-3K activity-independent cellular death response which is mediated by the induction of TNF- expression. These results suggest that p85 transmits apoptotic signaling by employing different pathways under different types of stimulation. In addition, we noticed that the genetic abolishment or targeted disruption of p85 expression did not alter the levels of other PI-3K regulatory subunits (p50, p55, and p85ß) (Fig. 1A and 9A). Therefore, p85 has the unique proapoptotic activity, probably due to its distinct protein structure, which cannot be found in the other isoforms.

Although we identified the novel p85/NFAT3/TNF- pathway for the mediation of the UVR-induced apoptotic effect, it is still unknown how this mediation is linked to the UVR damage signaling. A previous study showed that the level of p85 expression was altered in response to oxidative stress (57). However, unlike the investigators in that study, we did not observe the detectable up- or down-regulation of p85 protein levels in response to UVR (Fig. 4H and data not shown). We therefore propose that the model for the p85-mediated cell death response under oxidative stress may not be applied to the role of p85 in the response to UV radiation, although the oxidative stress is also a common biological effect of UVR damage. Further supporting this notion, our results indicate that p85 did not mediate an apoptotic effect but rather appeared to confer a protective effect on cells exposed to other well-known oxidative reagents such as arsenite and nickel compounds. Thus, the specific proapoptotic role of p85 reported here may be elicited by UV radiation via an event unrelated to those of oxidative stress.

How p85 is functionally linked to NFAT3 in response to UVR remains to be elucidated. We observed an obviously decreased basal level of NFAT3 due to the deficiency of the p85 gene, and the reintroduction of p85 into the null mutant cells restored NFAT3 expression (Fig. 7A). However, NFAT3 mRNA levels in p85^+/+ and p85^–/– MEFs were comparable (data not shown). Therefore, p85 appears to exert some regulatory function for the stability of NFAT3 in the resting fibroblasts through an unknown mechanism. This result was further supported by the decreased NFAT3 level in the p85 siRNA-transfected MEFs (data not shown). Upon UV radiation, nuclear import and export of NFAT3 was observed in p85^+/+ and p85^–/– MEFs, respectively (Fig. 7B), indicating that p85 regulates the activity of NFAT3 by modulation of its cytoplasm-nucleus shuttling in response to UVR. As we know, nuclear import and export are the most important steps for NFAT activation and are tightly regulated by the phosph