Phosphorylated TP63 Induces Transcription of RPN13, Leading to NOS2 Protein Degradation*

Head and neck squamous cell carcinoma cells exposed to cisplatin display ATM-dependent phosphorylation of the most predominant TP63 isoform (ΔNp63α), leading to its activation as a transcription factor. Here, we found that the phospho-ΔNp63α protein binds to the genomic promoter of RPN13 through the TP63-responsive element. We further found that the phospho-ΔNp63α protein associates with other transcription factors (DDIT3 (also known as CHOP), NF-Y, and NF-κB), activating RPN13 gene transcription. Furthermore, cisplatin-induced and phospho-ΔNp63α-dependent RPN13 gene transcription leads to NOS2 degradation. Finally, we show that RPN13 knockdown by siRNA essentially rescues NOS2 from cisplatin-dependent inactivation. These data provide a novel mechanism for the phospho-ΔNp63α-dependent regulation of NOS2 function in cells upon cisplatin treatment, contributing to the cell death pathway of tumor cells.

Misfolding and aggregation of proteins may play an important part in the pathogenesis of cancer because cells utilize a physiologic aggresome pathway targeting various proteins into proteasome-dependent degradation in response to stress (1)(2)(3)(4). There is emerging evidence that inhibiting the aggresome pathway leads to accumulation of misfolded proteins and to cell death (apoptosis or autophagy) in tumor cells (1)(2)(3)(4). One of these misfolded proteins, inducible NOS2, was shown to be targeted into an aggresome/proteasome pathway, which leads to termination of NO production by NOS2 (5)(6)(7)(8). NOS2 plays a critical role in massive NO production in a variety of cell types under physiologic and pathophysiologic conditions (9 -13). Understanding the molecular and cellular processes responsible for controlling NO production by NOS2 is critical for devising therapeutic strategies for many pathologic conditions, including cancer (14 -20).
We previously found that the NAP110 (NOS2-associated protein of 110 kDa; also known as ADRM1 (adhesion regulatory molecule 1) or RPN13 (regulatory particle non-ATPase subunit 13)) forms protein-protein complexes with NOS2. This interaction modulates NOS2 activity by blocking its ho-modimerization (21,22) and likely targeting NOS2 into a membrane fraction of cells, as reported by others (23). RPN13 is predominantly expressed as a non-glycosylated 42-kDa protein and is the ortholog of the yeast proteasome subunit Rpn13 (24 -27). Furthermore, RPN13 functions as a novel 19 S proteasome cap-associated protein, acting as a receptor for ubiquitin, and recruits the deubiquitinating enzyme UCH37 to the 26 S proteasome (24 -27).
Although RPN13 gene expression can be potentially induced by IFN␥ in gastric cancer cells, its transcriptional regulatory machinery is largely unknown (28). We thus undertook the study of transcription factors (TFs) 2 implicated in RPN13 gene regulation in head and neck squamous cell carcinoma (HNSCC) cells exposed to cisplatin, the most used agent in chemotherapy for human cancers (29 -32). Among a few of the TFs controlling RPN13 gene transcription, we found TP63 (tumor protein 63).
The TP53 homolog TP63 is a novel TF implicated in the regulation of genes involved in DNA damage response and chemotherapeutic stress in tumor cells (33). Because of the two independent promoters, the TP63 gene encodes two types of protein isoforms, one with a long transactivation domain and one with a short transactivation domain (34). The latter is designated ⌬Np63. Because of several alternative splicing events, TP63 produces three isoforms with various lengths of the C terminus (␣, ␤, and ␥). ⌬Np63␣ is the longest TP63 protein among the ⌬Np63 isoforms and is the most predominant isoform expressed in HNSCC cells (35)(36)(37). We previously showed the importance of the ATM (ataxia telangiectasia mutated)-dependent phosphorylation of TP63 for its transcriptional activity in HNSCC cells upon cisplatin exposure (36,37). Here, we define a novel molecular mechanism underlying the effect of the cisplatin-induced and phospho-⌬Np63␣-dependent up-regulation of RPN13 gene expression on NOS2 proteasome-dependent degradation.
All constructs were sequenced from the 5Ј-and 3Ј-ends. 5 ϫ 10 4 cells were plated per well in a 24-well plate. pGL3 luciferase reporter constructs (100 ng; Promega) and Renilla luciferase plasmid pRL-SV40 (1 ng; Promega) were introduced into cells using FuGENE 6 (Roche Applied Science) as described previously (34). 48 h after transfection, luciferase assays were performed using the Dual-Luciferase reporter assay kit (Promega). Firefly luciferase activity was normalized to Renilla luciferase activity for each transfected well (34). For each experimental trial, wells were transfected in triplicate, and each well was assayed in triplicate. Luciferase activity was normalized to the activity produced from the empty vector. Cell lysates were cleared by centrifugation, 10 l was added to 50 l of firefly luciferase substrate, and light units were measured in a luminometer. Renilla luciferase activities were measured in the same tube after the addition of 50 l of Stop & Glo reagent. Values for the firefly luciferase activity were divided by the Renilla luciferase activity to normalize for differences caused by unequal transfection efficiency (34).
RT-PCR-a first strand cDNA synthesis kit (Invitrogen) was used for reverse transcription. RT-PCR was performed with Taq DNA polymerase (Invitrogen) for 24 -30 cycles at 94°C for 1 min, 55°C for 1 min, and 72°C for 30 s as described previously (34,37). As a control, we amplified the GAPDH mRNA with primers 5Ј-CTACATGGTTTACATGT-3Ј (sense, ϩ121) and 5Ј-TGCCCTCAACGACCACT-3Ј (antisense, ϩ920), yielding the 800-bp PCR product. For RPN13 amplification, we used primers 5Ј-TCATCTTCCCTGAC-GACTG-3Ј (sense, ϩ301) and 5Ј-GCCTGCTGGGAAA-CATGA-3Ј (antisense, ϩ570), yielding the 270-bp PCR product. A quantitative PCR (qPCR) assay was performed using the StepOnePlus real-time PCR system kit (Applied Biosystems) with SYBR Green Universal PCR Master Mix. The same primers described above were used. Values for RPN13 were normalized to values for GAPDH, and values obtained from the control untreated samples were designated as 1. Experiments were performed in triplicate.
As a negative control, the binding of TF was tested with the nonspecific region primers representing the RPN13 ORF (ϩ301 to ϩ570). To quantify the binding of TFs to the RPN13 gene promoter sequences, we used qPCR with the abovementioned primers for various TFs. ChIP-PCR values were normalized to GAPDH values. For each TF, values obtained from the input samples were designated as 1. Experiments were performed in triplicate.
RNA Isolation, cDNA Synthesis, and Hybridization-Total RNA was isolated from 5 ϫ 10 7 cells using guanidine isothiocyanate lysis buffer (TRIzol, Invitrogen). Poly(A) ϩ RNA was subsequently isolated using biotinylated oligo(dT) primers and streptavidin coupled to magnetic beads (PolyATract, Promega). The RiboClone cDNA synthesis system (Promega) was then used to convert 2 g of poly(A) ϩ RNA into cDNA. For hybridization, 10 g of total RNA was fractionated on a 1.3% formaldehyde-agarose gel and transferred to a nitrocellulose filter. The 1.3-kbp RPN13 cDNA probe was labeled with [␣-32 P]dCTP by random priming (Invitrogen) and used to hybridize the filter. The filter was then autoradiographed for 16 h on Kodak XAR film at Ϫ80°C.
Isolation of a Genomic RPN13 Clone and Partial Characterization-A human placenta genomic library in the Lambda FIX II vector (946207, Stratagene) was screened with a 1.3-kbp cDNA probe encoding full-length human RPN13 (28) according to the manufacturer's recommendations. A total of 1 ϫ 10 6 plaques were screened, yielding three positive clones. To screen clones by PCR, we then used the following primer pair from the 5Ј-part of the RPN13 ORF: 5Ј-ACCGT-GACTCCGGATAAGCGGA-3Ј (sense, 109 -130) and 5Ј-CTGCATCCAGAAGAAAAGCC-3Ј (antisense, 330 to 311). One clone was found positive. Lambda FIX phage DNA was isolated from this clone and digested with EcoRI, XbaI, and PstI. One EcoRI fragment (3.8 kbp), four XbaI fragments (3.0, 2.3, 1.7, and 1.4 kbp), and two PstI fragments (5.5 and 4.2 kbp) were subcloned into the pBluescript SKII ϩ vector (Stratagene). To map the transcription start site (TSS), the 5Ј-end of the RPN13 mRNA was cloned after 5Ј-rapid amplification of cDNA ends (5Ј-RACE; Invitrogen). Poly(A) ϩ RNA for 5Ј-RACE was reversed-transcribed using the RPN13-specific primer 5Ј-GCCATTATCTCCGGCGTCAG-3Ј (ϩ894 to ϩ875) and Superscript II reverse transcriptase (Invitrogen) for 30 min at 52°C. Gene-specific cDNA was PCR-amplified using the RPN13-specific primer 5Ј-GCCGCTGGCCCCC-AGTGCCCCAGG-3Ј (ϩ423 to ϩ400) and the universal 5Ј-RACE anchor primer 5Ј-GGCCACGCGTCGACTAG-TACGGGGGGGGGG-3Ј. The resulting PCR product was cloned into the pCR3.1 vector and sequenced. The primer extension assay was carried out by reverse transcription of 10 g of total RNA using the 32 P-labeled primer 5Ј-CTTCAGG-GACATCTTTCCCGCCCG-3Ј (positions ϩ101 to ϩ79 of the RPN13 ORF). Annealing was carried out for 10 min at 70°C. Reverse transcription was carried out at 42°C for 50 min. The reaction product was treated with DNase-free RNase at 37°C for 15 min. Extension products were resolved on an 8% sequencing gel. As negative controls, we used tRNA and a sample without RNA.
NOS Activity Assay-Cells were sonicated in 50 mM Tris-HCl (pH 7.4,) 1 mM EDTA, 0.1 mM tetrahydrobiopterin, 2 mM dithiothreitol, 10% (v/v) glycerol, 25 g/ml aprotinin, 25 g/ml leupeptin, 100 M PMSF, 10 M FMN, and 10 M FAD. A reaction mixture was made of cell lysate containing 50 g of protein, 50 l of [U-14 C]arginine (50,000 cpm), 5 l of 5 mM FAD, 5 l of 100 M tetrahydrobiopterin, and 1 l of 30 M calmodulin, and the volume was brought up to 250 l with 50 mM Tris-HCl (pH 7.4) and 1 mM EDTA. The reaction was initiated by the addition of 50 l of 10 mM NADPH, incubated for 60 min at 37°C, and applied to a Dowex 50W-H ϩ column, and the flow-through was counted in a liquid scintillation counter (21,22).

RESULTS
Cisplatin Induces the Phospho-⌬Np63␣-dependent Expression of RPN13 in HNSCC Cells-Using the mRNA expression array and ChIP-on-chip array, we previously showed that exposure of HNSCC cells to 10 g/ml cisplatin for 16 h leads to up-regulation of many mRNA transcripts associated with the DNA damage, cell cycle, and apoptosis pathways (37). Among these mRNA targets, we found RPN13 (also known ADRM1). We previously showed that cisplatin treatment induces the ATM-dependent phosphorylation of ⌬Np63␣, subsequently modulating various genes implicated in cell survival (33,(35)(36)(37). Using the isogenic HNSCC cell lines expressing wild-type ⌬Np63␣ or ⌬Np63␣-S385G (⌬Np63␣ with an altered ability to be phosphorylated by ATM kinase), we previously showed the failure of ⌬Np63␣-S385G to regulate gene transcription (36,37).
We first found that cisplatin treatment substantially increased the RPN13 mRNA levels in wild-type ⌬Np63␣ cells but not in ⌬Np63␣-S385G cells as shown by RT-PCR assay (Fig. 1A). By qPCR, we further found that cisplatin treatment increased the RPN13 mRNA levels in wild-type ⌬Np63␣ cells by 12.4 Ϯ 0.9-fold (Fig. 1B). Upon inspection of the 1.5-kbp RPN13 promoter sequence (UCSC Genome Bioinformatics) using TFSEARCH software, we observed that the specific TF cognate sequences are present in the 1.5-kbp human RPN13 promoter (Fig. 2).
Cloning and Initial Analysis of the Human RPN13 Promoter-To define the molecular mechanism of cisplatin-induced RPN13 gene expression in HNSCC cells and a potential involvement of phospho-⌬Np63␣, we cloned and analyzed the human RPN13 gene promoter. We screened a human placenta genomic library in the Lambda FIX II vector with a 1.3-kbp cDNA probe encoding full-length human RPN13 (28). A total of 1 ϫ 10 6 plaques were screened, yielding three positive clones. Restriction mapping of the Lambda FIX II phage DNA positive clone showed one EcoRI fragment (3.8 kbp), four XbaI fragments (3.0, 2.3, 1.7, and 1.4 kbp), and two PstI fragments (5.5 and 4.2 kbp), as shown in Fig. 4A. Partial sequencing of the XbaI clones revealed that the 3.0-and 2.3-kbp fragments contained sequences of the 5Ј-part of the RPN13 coding region (Fig. 4A). Nucleotide sequence analysis of a 3.8-kbp EcoRI subclone and a 5.5kbp PstI subclone demonstrated that these clones contained 620 bp and 1.7 kbp, respectively, of the 5Ј-flanking region of the RPN13 gene. Analysis of the 1.7-kbp fragment revealed a 230-bp-long intron found in the coding region of the RPN13 gene (located between ϩ333 and ϩ334 bp of the ORF).
To map the TSS, the 5Ј-end of the RPN13 mRNA was cloned after 5Ј-RACE. Poly(A) ϩ RNA for 5Ј-RACE was isolated from HNSCC cells exposed to 10 g/ml cisplatin for 6 h based on time course analysis of mRNA expression in HNSCC cells (Fig. 4B). RNA was reversed-transcribed using the RPN13-specific primer (ϩ894 to ϩ875 bp of the RPN13 ORF) and then amplified with the RPN13-specific primer (ϩ423 to ϩ399 bp) and the universal 5Ј-RACE anchor primer, yielding a single band of ϳ459 bp (Fig. 5A). The resulting PCR product was cloned and sequenced, and it showed identity to the RPN13 cDNA sequence (Fig. 5B), suggesting that the TSS is a G residue (Ϫ70), as confirmed by the primer extension assay (Fig. 5B).

Cisplatin-induced Phospho-⌬Np63␣ Interacts with Other Transcription Factors That Regulate the RPN13 Promoter-
To examine the molecular mechanism underlying RPN13 FIGURE 3. Immunoblot analysis of the protein levels of RPN13 and TFs (tested with the indicated antibodies) potentially regulating RPN13 expression. Total, nuclear, and cytoplasmic lysates were used as indicated. The loading control used for total and cytoplasmic lysates was ␤-actin, and that for nuclear lysates was TOP2A. CIS, cisplatin. gene transcriptional regulation, we tested whether the phospho-⌬Np63␣ protein along with other TFs (defined in the human RPN13 promoter sequence) (Fig. 2) were endogenously bound to the RPN13 promoter. We used the ChIP assay with antibodies to ⌬Np63␣, phospho-⌬Np63␣, DDIT3, and NF-B (p65 subunit) on wild-type ⌬Np63␣ or ⌬Np63␣-S385G cells exposed to control medium or 10 g/ml cisplatin for 16 h. We found that cisplatin treatment of wild-type ⌬Np63␣ cells induced phospho-⌬Np63␣ protein binding to the RPN13 promoter sequences (Fig. 7A). The binding of the phospho-⌬Np63␣ protein was dramatically increased in wildtype ⌬Np63␣ cells compared with the non-phosphorylated ⌬Np63␣ protein in ⌬Np63␣-S385G cells (Fig. 7A, upper (input) and lower (ChIP) panels). Similarly, cisplatin dramatically induced the binding of DDIT3 to the RPN13 promoter (Fig.  7A). However, NF-B (p65 subunit) binding to the RPN13 promoter minimally increased in wild-type ⌬Np63␣ cells upon cisplatin exposure (Fig. 7A). In contrast, the binding of ⌬Np63␣, phospho-⌬Np63␣, and DDIT3 to the RPN13 promoter was totally abolished in ⌬Np63␣-S385G cells, whereas NF-B (p65 subunit) binding essentially remained (Fig. 7B, upper (input) and lower (ChIP) panels). The TFs tested showed no binding to the nonspecific regions of the RPN13 promoter (data not shown). By qPCR, we assessed the quantitative differences in TF binding to the RPN13 promoter in wild-type ⌬Np63␣ and ⌬Np63␣-S385G cells under control and cisplatin-treated conditions (Fig. 7C). We observed that cisplatin treatment substantially increased the binding of ⌬Np63␣, phospho-⌬Np63␣, and DDIT3 by 10.5-14.1-fold in wild-type ⌬Np63␣ cells while having a minimal effect in ⌬Np63␣-S385G cells (Fig. 7C).
To further examine the mutual protein-protein interactions of the various TFs bound to the RPN13 promoter, we coprecipitated protein-protein complexes with antibody to ⌬Np63␣ in wild-type ⌬Np63␣ and ⌬Np63␣-S385G cells treated with control medium or 10 g/ml cisplatin for 16 h. We found that cisplatin treatment induced complex formation between phospho-⌬Np63␣, DDIT3, and NF-YA, whereas complexes between ⌬Np63␣ and NF-B (p65 subunit) or STAT3 essentially remained unchanged in cells upon cisplatin exposure (Fig. 8, left panels). However, in ⌬Np63␣-S385G cells, neither control medium nor cisplatin led to complex formation between ⌬Np63␣ and other regulators of RPN13 transcription (Fig. 8, right panels).
We found that cisplatin treatment along with IFN␥ exposure dramatically induced RPN13 expression in wild-type ⌬Np63␣ cells (Fig. 9A, left panels) in contrast to ⌬Np63␣-S385G cells (right panels). However, NOS2 expression substantially decreased in wild-type ⌬Np63␣ cells (Fig. 9A, left  panels), whereas no changes were observed in ⌬Np63␣-S385G cells (right panels). We further found that RPN13-NOS2-UCH37 protein complex formation substantially increased in wild-type ⌬Np63␣ cells upon cisplatin exposure (Fig. 9A, left panels) but minimally increased in ⌬Np63␣-S385G cells (right panels). Using the proteasome inhibitor lactacystin, we observed no changes in NOS2 protein levels in both cell lines exposed to control medium or cisplatin treatment (Fig. 9A, lowest panel). Using the radioactive assay for converting arginine to citrulline (21,22), we examined NO production in HNSCC cells exposed to control medium or 10 g/ml cisplatin for 16 h grown without or with IFN␥. To ensure that we exclusively monitored the calcium-independent activity of NOS2, we tested the total NOS activity in the presence of EDTA as described previously (21,22,41). We observed that, in wild-type ⌬Np63␣ cells, IFN␥ dramatically increased NOS2 activity in the absence of cisplatin (Fig. 9B,  samples 1 and 2), whereas cisplatin treatment substantially modulated IFN␥-induced NOS2 activity (samples 3 and 4). We further found that, in ⌬Np63␣-S385G cells, both control medium and cisplatin had no effect on the IFN␥-dependent activation of NOS2 activity (Fig. 9B, samples 5-8).
We then examined whether siRNA silencing of RPN13 expression would affect the NOS2 level and activity in wild-type ⌬Np63␣ cells treated with control medium or 10 g/ml cisplatin for 16 h. We found that, in the presence of scrambled siRNA, IFN␥ and cisplatin increased RPN13 and decreased NOS2 expression, whereas RPN13 siRNA inhibited RPN13 expression and increased NOS2 expression (Fig. 10A), supporting the notion that RPN13 is intimately involved in NOS2 inactivation. Because we tested the total NOS activity under calcium-independent conditions and only negligible amounts of both NOS1 and NOS3 were found in HNSCC cells (supplemental Fig. 1, B-D), we concluded that only NOS2 activity  was monitored under these experimental conditions. Thus, our data showed that scrambled siRNA had no effect on RPN13-dependent inhibition of NOS (NOS2) activity (Fig.  10B, samples 1 and 2), whereas RPN13 siRNA reversed this inhibitory effect on NOS (NOS2) activity (samples 3 and 4).

DISCUSSION
We previously found that HNSCC cells exposed to cisplatin display an increase in the ATM-dependent phosphorylation of ⌬Np63␣, leading to activation of its transcriptional function (33)(34)(35)(36)(37). Using cDNA chip microarray expression and ChIP-on-chip array, we found gene targets induced/downregulated via a cisplatin/phospho-⌬Np63␣-dependent pathway (37). Among those genes, whose expression was up-regulated in HNSCC cells upon cisplatin exposure, we found a human homolog of yeast Rpn13, which serves as a ubiquitin receptor and associates with the deubiquitinating enzyme UCH37, recruiting the latter to the 26 S proteasome (24 -27). We previously showed that RPN13 physically interacts with NOS2, modulating its activity through blocking its homodimerization and potentially targeting nascent NOS2 polypeptides into a membrane-bound location (21)(22)(23). NOS2 inactivation is also likely to be achieved by targeting this enzyme into a proteasome-dependent degradation pathway, modulating NOS2 turnover and protecting cells against longterm and high-output NO production by this powerful enzyme (5)(6)(7)(8).
Accumulating evidence supports the notion that high-level NO might lead to a pro-apoptotic response, whereas low-level NO would likely play an anti-apoptotic role (10, 16 -18). NO affects cellular decisions of life and death either by turning on apoptotic pathways or by shutting them off (10, 16 -18). Proapoptotic pathways of NO appeared likely to be compatible with TP53 intrinsic mitochondria-dependent apoptosis (10, 16 -18, 42-50). Anti-apoptotic actions of NO range from an immediate interference with pro-apoptotic signaling cascades to long-lasting effects based on expression of cell-protective proteins, such as the ability of NO to block caspases by S-nitrosylation (10, 16 -18, 50). High concentrations of NO and FIGURE 9. RPN13 binds to NOS2 and targets it to proteasome-dependent degradation. A, wild-type ⌬Np63␣ and ⌬Np63␣-S385G cells were treated with control medium (Ϫ) or 10 g/ml cisplatin (CIS; ϩ) and 20 units/ml IFN␥ for 16 h. Immunoblotting was performed with the indicated antibodies. Immunoprecipitation (IP) of the RPN13-NOS2 protein complexes was performed with antibody to RPN13. Cells were treated with the 26 S proteasome inhibitor lactacystin (25 M) for 12 h. Separate experiments are shown as separate images. B, NOS enzymatic assay. Cells were treated with control medium (samples 1, 3, 5, and 7) or 20 units/ml IFN␥ (samples 2, 4, 6, and 8) and also incubated with 10 g/ml cisplatin (samples 3, 4, 7, and 8). Lysates of these cells were assayed individually for NOS2 activity. 50 g of cell lysate was assayed for the conversion of arginine to citrulline.  2 and 4), and all samples were also incubated with 20 units/ml IFN␥. Lysates of these cells were assayed individually for NOS2 activity. 50 g of cell lysate was assayed for the conversion of arginine to citrulline.
NO signaling is crucial for effecting long-lasting changes in cells, including gene expression, cell cycle arrest, cell death, and cell differentiation (9, 16 -18, 51, 52). Using cDNA microarray technology to study the kinetics of gene activation by NO, Enikolopov and co-workers (9,51,52) determined that NO induces three distinct waves of gene activity. The first wave is induced within 30 min of exposure to NO and represents the primary gene targets of NO (9,51,52). It is followed by subsequent waves of gene activity that may reflect further cascades of NO-induced gene expression (9,51,52). Numerous reports showed the functional interrelationship of NOS2 and TP53 (11-14, 41, 45-50). TP53 activation was documented to play a central role in the inflammatory stress response associated with cancer, changing the cellular microenvironment, reactive oxygen species production, and sets of genes ultimately induced by both hypoxia and NO in human cancer (11, 12, 16 -18). At the same time, TP53 was shown to cooperate with NO in the negative regulation of tumorigenesis (41). NO is a potent activator of TP53 through rapid down-regulation of MDM2, promotion of TP53 nuclear retention, inhibition of MDM2-mediated TP53 nuclear export, and modulation of TP53 phosphorylation and stability (45)(46)(47)(48)(49)(50).
Because TP63 is a member of the TP53 family of TFs, we hypothesized that ⌬Np63␣, the TP63 isoform that is expressed predominantly in HNSCC cells, might play a role in NOS2 regulation as part of the cellular response to cisplatin (33,(35)(36)(37). In this study, we employed two cellular models available to us: HNSCC cells expressing wild-type ⌬Np63␣ and the ⌬Np63␣-S385G mutant (with an altered ability to be phosphorylated by ATM). We found that wild-type ⌬Np63␣ cells exposed to cisplatin displayed an increase in RPN13 expression at the RNA and protein levels. However, ⌬Np63␣-S385G cells failed to support this RPN13 up-regulation. We further found that cisplatin treatment induced both phospho-⌬Np63␣ and DDIT3 protein levels and subsequently RPN13 gene transcription. However, the protein levels of activated NF-B (p50 and p65 subunits) imported to the nucleus remained unchanged in HNSCC cells upon cisplatin exposure. ⌬Np63␣-S385G cells failed to show ⌬Np63␣ phosphorylation and detectable levels of DDIT3, which is consistent with our previous observations (37). Using a series of 5Ј-deletions of the RPN13 promoter, we then found that phospho-⌬Np63␣ bound to the TP63-responsive element in combination with DDIT3, NF-YA, and NF-B dramatically activated RPN13 gene transcription. We also found that cisplatin-induced RPN13 up-regulation led to a decrease in NOS2 protein levels and inactivation of NOS2 activity through the RPN13/proteasome degradation pathway. Finally, we found that RPN13 siRNA essentially rescued NOS2 from cisplatin-dependent inactivation in HNSCC cells.
Our results provide a novel link between the ⌬Np63␣-dependent transcriptional regulation of RPN13 and the NOS2 proteasome degradation in HNSCC cells upon cisplatin exposure, which might be part of the potential response of tumor cells to cisplatin-induced cell death through apoptotic or autophagic pathways (32). Other reports support this notion, further emphasizing the role for NO/NOS2 regulation in the response of cancer cells to cisplatin exposure (13, 18, 20, 39 -44, 53). For example, blocking NOS activity with N -amino-Larginine was shown to modulate cisplatin-resistant ovarian cancer cells (OV-2008-C13) to become cisplatin-sensitive (13). Additionally, NOS2 expression in melanoma cells was shown to correlate with poor patient survival after platinum chemotherapy (54,55). Depletion of endogenously produced NO was shown to potentially enhance cisplatin-induced death of melanoma cells (54,55). Other approaches, including altering protein-protein interactions and proteasome inhibition, would also be considered as future venues to study molecular and cellular mechanisms potentially underlying tumor resistance to chemotherapy (56 -58).