The p63 transcription factor has a pivotal role in epithelial morphogenesis. Multiple transcripts of the TP63 gene are generated because of alternative promoter usage and splicing. ΔNp63α is the predominant isoform of p63 observed during epithelial morphogenesis and in human cancers. Loss of ΔNp63α expression has been shown to promote invasiveness in a subset of human cancer cell lines. Here, we studied whether the regulation of VDR by ΔNp63α controls the invasiveness of an epidermoid cancer cell line. We demonstrate that VDR expression is induced by all p63 isoforms, including ΔNp63α. Endogenous ΔNp63α protein was observed to bind to the VDR promoter, and silencing of endogenous ΔNp63α resulted in diminished VDR expression. Although silencing of p63 inhibits VDR expression leading to an increase in cell migration, overexpression of p63 or VDR results in reduced cell migration as a result of increased VDR expression. Therefore, it is conceivable that p63 inhibits cell invasion by regulating VDR expression. Finally, we observed that expression of p63 and VDR overlaps in the wild-type mouse skin, but a reduced or complete absence of VDR expression was observed in skin from p63-null mice and in p63-null mouse embryonic fibroblasts. In conclusion, we demonstrate a direct transcriptional regulation of VDR by ΔNp63α. Our results highlight a crucial role for VDR in p63-mediated biological functions.
The p63 transcription factor belongs to the p53 family of proteins and shows high structural similarity to the TP53 and TP73 genes (Osada et al., 1998; Shimada et al., 1999; Yang et al., 1999). p63 contains three functional domains, an N-terminal transactivation (TA), a central DNA-binding domain and an oligomerization domain. A high degree of conservation is observed within the DNA-binding domains of p63 and p53, therefore p63 transactivates several p53-responsive genes (Shimada et al., 1999). The TP63 gene generates six different isoforms: three TA isoforms (TAp63γ, TAp63β and TAp63α) and three ΔN isoforms (ΔNp63α, ΔNp63β and ΔNp63α) because of the differential promoter usage and alternative splicing. Although p63 is not mutated in human cancers, and unlike p53 is not considered to be a tumor suppressor, TP63 mutations have been observed in several human developmental syndromes (Barrow et al., 2002; van Bokhoven et al., 2001; van Bokhoven and McKeon, 2002).
p63 has a major role in development, particularly in epithelial morphogenesis during embryonic development (Mills et al., 1999; Yang et al., 1999). p63, in particular ΔNp63α, is highly expressed in basal layers of stratified epithelium and is essential for maintenance of epithelial tissue integrity. p63-knockout mice exhibit severe developmental defects, including defects in limbs, hair follicles, teeth and epidermal development (Mills et al., 1999; Yang et al., 1999). The developmental abnormalities observed in p63–/– mice were mainly attributed to defects in stratified epithelial maintenance and differentiation. Additionally, although TAp63 initiates the epithelial stratification program, ΔNp63 is indispensable for the maintenance of epithelial stem cell proliferation and differentiation, thus making ΔNp63α the most physiologically relevant isoform to study in most epithelial systems (Koster et al., 2004).
The loss of p63 expression is correlated with cancer progression in various cancers, including prostate and bladder cancers (Park et al., 2004; Parsons et al., 2001; Urist et al., 2002). An amplified level of the ΔNp63α isoform, which promotes proliferation by inducing the pro-survival proteins, has been observed in squamous cell carcinoma (Hu et al., 2002; Senoo et al., 2002; Sniezek et al., 2004; Wu et al., 2005). Moreover, it has been shown that TAp63 isoforms promote growth arrest and apoptosis in several cancer cell lines by inducing anti-proliferative genes (Kommagani et al., 2006; Shimada et al., 1999; Spiesbach et al., 2005), and by inhibiting the pro-proliferative genes (Senoo et al., 2002; Wu et al., 2005). Interestingly, downregulation of ΔNp63α expression leads to an increase in cell motility and invasiveness of squamous cell carcinoma cells (Higashikawa et al., 2007). Altogether, the above studies suggest isoform specific functions of p63 and therefore delineating the role of specific p63 isoforms is necessary to understand the overall biology of p63.
Recently, we have demonstrated the transcriptional activation of vitamin D receptor (VDR) by the TAp63γ isoform (Kommagani et al., 2006). Vitamin D receptor is a ligand-dependent transcription factor, and is a natural receptor for the secosteroid hormone vitamin D. Vitamin D and its analogues promote anti-proliferative activities in several cancer cell lines and are used for cancer chemotherapy (Banerjee and Chatterjee, 2003). High expression of VDR is associated with favorable prognosis in colon cancers and loss of VDR expression is associated with an increase in metastatic proteins such as SNAIL (Palmer et al., 2004; Pena et al., 2005). In this report, we studied regulation of VDR by ΔNp63α, the most physiologically relevant p63 isoform. We clearly demonstrated that all isoforms of p63, including ΔNp63α, induce VDR expression. Downregulation of ΔNp63α resulted in a concomitant decrease in VDR expression. Our results indicate that ΔNp63α-mediated regulation of VDR is associated with inhibition of invasiveness and migration. Finally, p63 expression is essential for VDR expression during embryonic development, further suggesting the p63-mediated regulation might have a role in several cellular processes.
Transcriptional activation of VDR by all the p63 isoforms
We have previously reported the transcriptional activation of VDR by the most potent transactivator of all the p63 isoforms, TAp63γ, however, the regulation of VDR by other isoforms of p63 has not yet been studied. This is important because the most abundant isoform of p63 expressed in basal epithelial cells of the skin is ΔNp63α, and VDR has also been shown to be expressed in the basal and upper layers of differentiating skin. To determine the effect of different p63 isoforms on endogenous VDR expression, we overexpressed all six different isoforms of p63 (TAp63γ, TAp63β, TAp63α, ΔNp63γ, ΔNp63β and ΔNp63α) in two p63–/– cell lines (H1299 and HeLa), and monitored the expression levels of VDR at transcript and protein levels. As a positive control, we monitored the transcript and protein levels of p21, which has been shown to be positively regulated by all isoforms of p63 except ΔNp63α (Barbieri et al., 2005; Petitjean et al., 2008). A significant increase in VDR transcript levels was observed in cells transfected with all isoforms of p63 when compared with levels in cells transfected with control vector (Fig. 1A,B). As expected, we also observed a significant increase in p21 transcript levels with TAp63γ, TAp63β and TAp63α, and ΔNp63β isoforms but not ΔNp63γ and ΔNp63α. Given the fact that the TA isoforms contain the full-length transactivation domain, the effects of the ΔNp63 isoforms, except ΔNp63β, were expectedly modest when compared with TA isoforms. Consistent with the transcript levels TAp63γ, TAp63β, TAp63α and ΔNp63β isoforms induced the VDR protein to much higher levels than those of ΔNp63γ and ΔNp63α isoforms. Similarly, all isoforms of p63 except ΔNp63α, induced the p21 protein levels, albeit by a modest amount with the ΔNp63γ isoform (Fig. 1). These results clearly show that all p63 isoforms, including ΔNp63α, can transcriptionally activate VDR.
ΔNp63α does not inhibit TAp63-mediated induction of VDR.
Several studies have reported that the ΔNp63α isoform exerts a dominant-negative effect towards p53 and TAp63 isoforms. Since we observed the induction of VDR by both ΔNp63α and TAp63 isoforms, we next examined the effect of ΔNp63α on TAp63-mediated activation of VDR. H1229 cells were transfected with TAp63γ isoform alone or with increasing concentrations of ΔNp63α isoform and mRNA transcript levels of VDR, p21 and IGFBP3 (Insulin growth factor binding protein-3) were monitored. Both p21 and IGFBP-3 were shown to be repressed by ΔNp63α and therefore were used as controls for the ΔNp63α-mediated dominant-negative effect on TAp63 (Barbieri et al., 2005; Westfall et al., 2003). As reported earlier, TAp63γ, but not ΔNp63α, upregulated transcription of both p21 and IGFBP3. As shown above in Fig. 1, both exogenous TAp63γ and ΔNp63α upregulated the transcript levels of VDR (Fig. 2A). Interestingly, although ΔNp63α clearly downregulated TAp63-mediated induction of p21 and IGFBP-3, ΔNp63α did not affect TAp63-mediated induction of VDR (Fig. 2A). Consistent with the results on mRNA transcript levels, co-transfection of ΔNp63α did not lead to a significant downregulation of TAp63-mediated induction of VDR protein levels (Fig. 2B). Therefore, these results demonstrate that ΔNp63α does not act in a dominant-negative manner towards TAp63-mediated induction of VDR.
VDR is a direct target of ΔNp63a
ΔNp63α lacks the full-length N-terminal transactivation domain, but it contains a short unique transactivation domain that is involved in activating multiple target genes by directly binding to their promoter regions (Helton et al., 2006). Since we showed that VDR expression is upregulated by ΔNp63α, we next examined whether VDR is a direct target of ΔNp63α. First, we tested whether ΔNp63α can transactivate a luciferase reporter under the control of the full-length VDR promoter (VDR-Luc). Co-transfection of VDR-Luc, along with increasing concentrations of ΔNp63α, led to a dose-dependent increase in VDR-Luc activity (Fig. 3A). To confirm that ΔNp63α can bind to the VDR promoter under physiological conditions, we used A431 cells, an epidermoid cancer cell line that has endogenous p63. By using pan p63, ΔN-isoform-specific and α-isoform-specific antibodies, we were able to confirm that A431 cells predominantly express the ΔNp63α isoform (Fig. 3B) (Romano et al., 2006). To confirm that VDR is a direct target of ΔNp63α, we next performed chromatin immunoprecipitation (ChIP) assay on intact A431 cells to determine the occupancy of endogenous ΔNp63α protein on the VDR promoter. As shown in Fig. 3C, endogenous ΔNp63α protein bound to the VDR promoter region using the 4A4 antibody that recognizes all p63 isoforms, and H-129 antibody that is directed against the α-isoforms of p63. As reported earlier, ΔNp63α is shown to transrepress p21, and hence, as expected, we also observed the binding of ΔNp63α on the p21 promoter region. The β-actin promoter was used as a negative control (Fig. 3C). Altogether, these results demonstrate that ΔNp63α induces VDR through direct transcriptional activation.
Silencing of ΔNp63α leads to reduced VDR expression
To determine whether endogenous VDR expression levels are under the control of ΔNp63α, A431 cells were transfected with three different siRNAs against p63 to silence endogenous ΔNp63α and rule out any off-target effects. The first two siRNAs, sip63_1 and sip63_2, are against the 3′UTR whereas sip63_3 is against the cDNA. Although each of the siRNAs used target all isoforms of p63, A431 cells predominantly express ΔNp63α (Fig. 3B) and thus ΔNp63α is the isoform that is being silenced. As shown in Fig. 4A, compared with control siRNA-transfected cells, cells transfected with any of the three different p63 siRNAs showed a significant reduction in ΔNp63α mRNA transcript levels, indicating effective silencing of p63 in these cells. Silencing of endogenous ΔNp63α led to a concomitant downregulation of VDR transcript levels, indicating the requirement of endogenous ΔNp63α for the basal expression of VDR. Consistent with the transcript levels, silencing of ΔNp63α with three different siRNAs led to a significant reduction in both ΔNp63α as well as VDR protein levels (Fig. 4B). These results clearly demonstrate that VDR is a direct target of ΔNp63α and that ΔNp63α is essential for basal expression of VDR in A431 cells.
Regulation of VDR by ΔNp63α is associated with cell migration and invasion
Recently, downregulation of p63 expression has been shown to result in an enhanced invasiveness of A431 cells (Higashikawa et al., 2007). Since we found that silencing of p63 in A431 cells led to a downregulation in VDR expression (Fig. 4), we next tested whether ΔNp63α-mediated upregulation of VDR has a role in inhibition of invasiveness in A431 cells. To address this, the effects of either p63 or VDR silencing on cell migration and cell invasion were assessed by wound-healing and Matrigel-based assays, respectively. As shown in Fig. 5A, silencing of either p63 or VDR in A431 cells leads to an increase in wound closure, compared with control siRNA-transfected cells, indicating enhanced cell migration. Using Matrigel-membrane-based invasion assays, another measure of invasiveness and cell migration, we observed that knockdown of either p63 or VDR resulted in significant increase in number of invading cells (Fig. 5B). Since loss of VDR observed upon silencing of p63 leads to enhanced invasion of A431 cells, this suggests that p63-mediated regulation of VDR has a role in inhibiting the migration and invasiveness of A431 cells.
Next, to demonstrate that p63-mediated inhibition of cell invasiveness occurs via VDR, we established A431-ΔNp63α and A431-VDR stable pool cells that express exogenous ΔNp63α and VDR, respectively. As a control, we established A431-eGFP stable cells that express enhanced green fluorescence protein. As shown in Fig. 6A, A431-ΔNp63α cells showed increased levels of ΔNp63α, which correlated with increased VDR expression when compared with A431-eGFP cells. A431-VDR cells also showed increased VDR expression compared with A431-eGFP cells. Both A431-ΔNp63α and A431-VDR stable cells, when subjected to Matrigel-invasion assays, showed decreased cell invasion compared with A431-eGFP control cells, demonstrating that both ΔNp63α and VDR can inhibit cell invasion and supporting our hypothesis that ΔNp63α might be inhibiting cell invasion by inducing VDR expression (Fig. 6A). To confirm this, we investigated whether overexpression of either ΔNp63α or VDR could rescue the increased cell invasion observed upon p63 silencing. All three stable cells were transfected with control siRNA or p63 siRNA and Matrigel-based invasion assays were performed. As shown in Fig. 6B, an increase in the number of invading cells was observed in A431-eGFP cells transfected with p63 siRNA cells compared with control siRNA-transfected cells (Fig. 6B, compare lanes 1 and 2). A431-ΔNp63α cells transfected with control siRNA inhibited cell invasion and did not result in an increase in number of invading cells when compared with A431-eGFP cells transfected with control siRNA (Fig. 6B, compare lanes 3 and 1). A431-ΔNp63α cells transfected with p63 siRNA showed a slight increase in the number of invading cells (Fig. 6B, compare lanes 3 and 4), which could be attributed to silencing of the endogenous p63. However, the overexpressed ΔNp63α in A431-ΔNp63α cells was still sufficient to rescue the effects observed upon p63 silencing shown in Fig. 5B. Finally, A431-VDR cells transfected with p63 siRNA also did not show increased cell invasion (compare lanes 2 and 6) confirming that overexpression of VDR can also rescue the effects of silencing ΔNp63α. These results suggest that increased VDR levels observed upon re-expression of ΔNp63α or VDR, is sufficient to counteract the increase in invasiveness observed with loss of endogenous p63 and VDR. Thus, these results provide evidence that p63-mediated regulation of VDR has a role in inhibition of the migration and invasiveness of A431 cells.
p63 is essential for VDR expression during embryonic development
The functional role of p63 and VDR in epidermal development is well established. p63-knockout mice exhibit defects in hair development, and VDR-null mice have also been shown to exhibit severe alopecia (Xie et al., 2002). Given that p63 expression is essential for VDR expression in A431 cells, we next determined whether p63 is essential for VDR expression during embryonic development. First, we assessed the mRNA transcript levels of p63 and VDR in mouse embryonic fibroblasts (MEFs) from both wild-type and p63-null mice. A significant reduction in VDR transcript levels were observed in p63 null MEFs when compared with wild-type MEFs (Fig. 7A). Next, histochemical staining on skin sections obtained from wild-type and p63-null mice were analyzed for VDR and ΔNp63 expression. As shown in Fig. 7B, the expression of p63 and VDR overlapped in the basal keratinocytes of the interfollicular epidermis and the outer root sheath of the hair follicles of skin in newborn (NB), postnatal day 4 (P4) and day 16 (P16) wild-type mice. Conversely, staining of ΔNp63 and VDR in p63-knock out mice showed a lack of both p63 and VDR expression in the single layered skin. Altogether, these results suggest that ΔNp63-mediated regulation of VDR also has an important role in development.
The role of p63 in epithelial tissue development, in particular in stratified epithelial morphogenesis is well established (Candi et al., 2007). The proliferative potential of epithelial stem cells in stratified epithelia is dependent on p63 (Senoo et al., 2007). The predominant isoforms of p63 observed during development are TAp63α and ΔNp63α; however, the functional relevance of the other p63 isoforms still needs to be addressed (Koster et al., 2004). The vitamin D receptor (VDR), which is a member of the nuclear receptor family, has also been implicated in epidermal differentiation. The lack of VDR expression in VDR-null mice leads to development of alopecia because of defects in hair follicle regeneration (Xie et al., 2002), and reduced expression of epidermal differentiation markers including involucrin and loricrin (Bikle et al., 2002; Hawker et al., 2007). Our results demonstrated that all isoforms of p63 can induce VDR expression (Fig. 1), suggesting that severe epidermal defects observed in p63-knockout mice could result from lack of VDR expression. Initiation and execution of epithelial morphogenesis involves sequential activation of TAp63 isoforms, followed by ΔNp63 isoforms. Thus, it would be intriguing to see whether VDR is a vital component of both TAp63- and ΔNp63-associated functions during development.
ΔNp63α is the most predominant isoform of p63 expressed in the basal layers of stratified epithelia and is overexpressed in several cancers. ΔNp63α antagonizes the functions of p53 and TAp63 by forming heterotetramers or by competitive binding to specific DNA sequences (Yang et al., 1998). In support of this notion, several studies have shown downregulation of TAp63-mediated transactivation by ΔNp63α (Barbieri et al., 2005; Petitjean et al., 2008). Our results clearly showed that, although we observed a significant downregulation of TAp63-mediated induction of p21 and IGFBP-3 by ΔNp63α, TAp63γ-mediated VDR induction was not affected by the ΔNp63α isoform (Fig. 2), thus arguing against the assumed dominant-negative function for ΔNp63α towards all targets regulated by TAp63. Although ΔNp63 lacks the full-length N-terminal TA domain, the ΔNp63α isoform has been shown to activate target genes transcriptionally (Helton et al., 2006; Wu et al., 2005). Our results clearly show the induction of VDR by ΔNp63α isoform at both protein and mRNA levels (Fig. 1). This was further supported by a significant dose-dependent increase in VDR-Luc activity with increasing concentrations of ΔNp63α, suggesting a direct transcriptional activation of VDR by ΔNp63α (Fig. 3A). We confirmed that VDR is a direct transcriptional target of ΔNp63α by performing ChIP assay, which showed binding of endogenous ΔNp63α to the VDR promoter in A431 cells (Fig. 3C) and that silencing of endogenous ΔNp63α led to a significant decrease in basal VDR expression (Fig. 4).
The ability of cancer cells to invade is an essential component of metastasis and depends upon several factors. Recently, downregulation of p63 has been shown to increase the cell migration and invasiveness of cancer cell lines such as A431 (Barbieri et al., 2006; Higashikawa et al., 2007). The active form of vitamin D has also been shown to inhibit the invasiveness of prostate cancer cells through VDR (Tokar and Webber, 2005). Our results clearly demonstrate that downregulation of both ΔNp63α and VDR results in increased cell migration and cell invasion (Fig. 5A,B). Thus, our findings shed light on a novel role for ΔNp63α in ligand-independent function for VDR, particularly in inhibition of cell migration and invasion. Although ΔNp63α inhibits cell invasion via VDR, it is possible that other less-prominent isoforms of p63 are also able to mediate this inhibition because all p63 isoforms were able to induce VDR expression (Fig. 1). The abundance of ΔNp63α in the basal skin cells, however, allows for better characterization of the possible mechanisms of cell invasion. The zinc-finger transcription factor SNAIL has been shown to promote invasion and metastasis by repressing multiple proteins including E-cadherin and ΔNp63α (Barrallo-Gimeno and Nieto, 2005; Higashikawa et al., 2007). Interestingly, direct transcriptional activation of E-cadherin by vitamin D has also been shown to promote growth arrest in colon cancer cells (Palmer et al., 2001). Finally, VDR is repressed by SNAIL and a negative correlation between SNAIL and VDR is reported in a colon cancer cell line (Palmer et al., 2004; Pena et al., 2005). Thus, it is likely that at least in A431 cells, SNAIL represses the ΔNp63α-VDR-E-cadherin axis to promote invasiveness of cancer cells.
Several reports implicate a functional role for ligand-independent VDR in epidermal development (Ellison et al., 2007). The vitamin D receptor has been shown to be essential for normal keratinocyte stem cell function and severe epidermal defects were observed in Vdr-deficient mice (Cianferotti et al., 2007; Xie et al., 2002). However, the factors involved in molecular regulation of VDR during epithelial morphogenesis are not fully studied. We observed a significant downregulation of VDR expression levels in p63-deficient MEFs and both p63 and VDR were colocalized in basal layers of epidermis in newborn wild-type mice (Fig. 7). In contrast to wild-type mice, p63-deficient mice lack well-formed skin and only a single layer of cells can be observed. Histochemical analysis on single-layered skin from p63-deficient mice showed the complete abolishment in VDR expression (Fig. 7B). Additionally, we have recently shown a differential effect of naturally occurring p63 mutants on VDR expression, suggesting that deregulation of VDR by p63 mutants contributes to the pathophysiology of diseases associated with p63 mutations (Khokhar et al., 2008). Altogether, these results suggest that expression of VDR during embryonic development is dependent on p63 expression. However, additional studies are required to understand the precise role of VDR in p63-mediated functions, in particular in epithelial morphogenesis. In conclusion, our study identifies an association between two transcription factors, ΔNp63α and VDR, which might have a critical role in cancer cell invasion and epidermal differentiation.
Materials and Methods
Cell lines and plasmids
Human non-small lung carcinoma cell line; H1299, human cervical carcinoma cell line; HeLa and human epidermoid carcinoma cell line A431 were obtained from ATCC and were maintained in Dulbecco's modified Eagle's medium (DMEM) with 8% fetal bovine calf serum (FBS) and 250 U penicillin and 250 μg streptomycin (PS). VDR full-length promoter luciferase construct (VDR-Luc) was constructed as described previously (Kommagani et al., 2006). Primary MEFs were obtained from wild-type or p63–/– mice, kindly provided by Elsa Flores (University of Texas M.D. Anderson Cancer Center, Houston, TX). A431-eGFP and A431-ΔNp63α pooled stable cells were generated as follows. The ΔNp63α cDNA containing lentiviral expression plasmid was constructed by amplifying the ΔNp63α cDNA sequence from ΔNp63α plasmid and then cloning it into modified pLentiV6 vector. The primers used for amplifying the ΔNp63α cDNA were as follows: sense 5′-GGAGGATCCATGTTGTACCTGGAAAACA-3′ and anti-sense 5′-AAACTCGAGTCACTCCCCCTCCCTTT-3′. Primers used for amplifying the VDR cDNA were as follows: sense 5′-AAGGAAAAAAGCGGCCGCATGGAGGCAATGGCGGCCAGC-3′ and anti-sense 5′-CTAGTCTAGATCAGGAGATCTCATTGCCAAACA-3′. The modified pLentiV6 vector contains additional restriction sites and the pLentiV6-eGFP were obtained from Thomas Brown and Steven Berberich, respectively, both at Wright State University, Dayton, OH. The A431-eGFP, A431-ΔNp63α and A431-VDR stable cell lines were generated by infecting the parental A431 cells with lentivirus expressing eGFP, ΔNp63α and VDR, respectively. At 24 hours post-infection, transduced cells were selected in blasticidin (10 mg/ml) to obtain the final stable pool of A431 cells expressing eGFP, ΔNp63α or VDR. The stable expression of ΔNp63α and VDR was confirmed by performing immunoblot analysis with p63 and VDR specific antibodies. Expression plasmids encoding TAp63γ, TAp63α, TAp63β, ΔNp63γ, ΔNp63α and ΔNp63β have been described earlier (Caserta et al., 2006).
Reverse transcription PCR
Total RNA from cells was extracted using eZNA RNA isolation kit (Omega Bio-tek, Norcross, GA) as per manufacturer's protocol. A total of 1 μg RNA was used to synthesize cDNA by using TaqMan reverse transcription kit (Applied Biosystems, Foster City, CA). TaqMan based real-time PCR analysis was performed on ABI Prism7900HT sequence detection system using TaqMan 2× master mix and gene-specific assays on demand. Assays on demand (AOD) were used for VDR (Hs_0017213_m1), p21 (Hs_00355782_m1), p63 (Hs_00978340_m1), IGFBP-3 (Hs_00426287_m1), murine p63 (Mm_00495788_m1), murine VDR (Mm_00437297_m1) and murine Id2 (Mm_00711781_m1) (Applied Biosystems, Foster City, CA).
Whole-cell extracts were made using RIPA buffer (0.5% sodium deoxycholate, 1% NP-40, 0.1% SDS in phosphate-buffered saline, pH 7.4). Total protein amounts were measured using BCA protein detection method and equivalent amounts of protein extracts were resolved on SDS-PAGE gels and processed for immunoblotting analysis. Monoclonal anti-VDR, rabbit polyclonal anti-p21, monoclonal anti-p63 4A4 (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal ΔNp63 specific anti-p63 RR-14 (Satrajit Sinha, SUNY, Buffalo, NY), rabbit polyclonal p63α-specific anti-p63 H-129 (Santa Cruz Biotechnology) and monoclonal anti-β-actin (Sigma, St Louis, MO) antibodies were used to detect VDR, p21, p53, p63 and β-actin expression, respectively. Mouse monoclonal anti-p63 C-12 antibody was used for detecting exogenous ΔNp63α protein. A Westpico Chemiluminescent Substrate kit (Pierce, Rockford, IL) was used for detecting the chemiluminescence signal.
Chromatin immunoprecipitation of A431 cells
A431 cells were grown on three 15 cm tissue culture dishes. At around 70-80% confluency, cells were subjected to chromatin immunoprecipitation assay by using ChIP Kit (Upstate Cell Signaling Solutions, Waltham, MA). Briefly, formaldehyde was used to fix cells grown in culture plates and after fixation, cells were homogenized and nuclei were subjected to sonication to obtain fragmented chromatin. Fragmented chromatin was pre-cleared using protein G beads and subjected to immunoprecipitation using monoclonal anti-p63 4A4, rabbit polyclonal anti-p63 H-129 and normal mouse monoclonal IgG antibodies (Santa Cruz Biotechnology). Both immunoprecipitated and non-immunoprecipitated (Input) chromatin were reverse crosslinked and DNA was eluted. Eluted DNA was subjected to PCR amplification specific for VDR promoter, p21 promoter and β-actin gene using GoTaq Green PCR master mix (Promega, Madison, WI). Primers used for PCR amplification as follows: VDR (forward, 5′-CGGGGTACCCAGTAACAGGTTGCGACGGAG-3′ and reverse, 5′-CCCAAGCTTGATGATTATAGGTGCGGATACCCG-3′), p21 (forward, 5′-GGTACCGGCACTCTTGTTCCCCCAGGCTG-3′ and reverse, 5′-CTCGAGACCATCCCCTTCCTCACCTGAAAA-3′) and β-actin (forward, 5′-ATCGTGGGCCGCCCTAGGCA-3′ and reverse, 5′-TGGCCTTAGGGTTCAGAGGGG-3′). PCR conditions used were as follows, a total of 40 cycles were performed, each consisting of 30 seconds at 94°C, 30 seconds at 60°C and 45 seconds at 68°C.
Transfections and transactivation assays
For transient transfections, at around 70-80% confluency cells were transfected with appropriate plasmids using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) in serum-free DMEM medium. A total of 3 μg plasmid DNA was used for each condition and in a given experiment, the total amount of plasmids was adjusted to 3 μg using control vector. For transactivation assays, cells were plated on 12-well plates and transfected with 1000 ng VDR-Luc reporter and Renilla luciferase constructs along with desired plasmids. At 24 hours post-transfection, cells were harvested in passive lysis buffer and subjected to Dual luciferase assay as per manufacturer's protocol (Promega, Madison, WI). The relative luciferase activity was measured by calculating the ratio of Renilla luciferase activity to Firefly luciferase activity.
Calcium-phosphate-based transfections for siRNA
Calcium-phosphate-mediated transfections were performed for siRNA transfections as described earlier (Kommagani et al., 2007). A431 cells were plated in six-well plates prior to the day of transfections and desired siRNAs were transfected using CaCl2 (125 mM final concentration) and 2× BBS (50 mM BES, 280 mM NaCl, 1.5 mM Na2HPO4). At 6 hours post-transfection, fresh medium was added, and after 24 hours, another round of transfection was performed as described above. All the siRNA used for the studies were purchased from Qiagen (Valencia, CA) and the target sequences used for siRNA were as follows: p63 siRNA_1 (CACCCTTATAGTCTAAGACTA), p63 siRNA_2 (AAGGGCCAGCGTGGCTCTAAA), p63 siRNA_3 (CCAGATGACATCCATCAAGAA) and VDR siRNA (CCGCGTCAGTGACGTGACCAA).
Wound-healing and Matrigel-based membrane invasion assays
For the wound-healing assay, A431 cells were transfected twice with desired siRNA and 24 hours post-transfection, a 200 μl pipette tip was used to create an artificial wound or a cleared area on monolayer cells. Scratched cells were washed and were re-fed with fresh medium. Immediately after the scratching (0 hours) and after incubation of cells at 37°C for 24 hours, phase-contrast images of the wound-healing process were taken with an inverted microscope. The distance of the wound area was measured on the images and set at 100% for 0 hours and the mean percentage of the total distance of the wound area was measured to calculate wound closure. The Matrigel-membrane-based invasion assays were performed using BD BioCoat Matigel Invasion chambers (BD Biosciences, Bedford, MA). At 24 hours post-transfection, A431 cells were counted and 1×105 cells were seeded onto invasion chambers. Fetal bovine serum was used as a chemoattractant and was added to the lower invasion chambers. After 24 hours, the non-invading cells were removed using cotton swabs and the invaded cells on the bottom of the lower membrane were stained with Diff-Quik stain kit. After chambers were air-dried, membranes were removed gently and were mounted on glass slides. Stained cells images from several fields were counted to calculate the fold change in the number of invading cells.
Immunofluorescence staining for p63 and VDR
Newborn animal whole embryo (for p63–/– animals) and wild-type skin tissue samples were fixed overnight in 10% NBF, dehydrated, paraffin-embedded and sectioned to 4 μm thickness. Slides were de-paraffinized and rehydrated through a graded alcohol series. Antigen retrieval was performed by boiling slides in a microwave for 20 minutes in antigen retrieval solution (10 mM sodium citrate, 0.05% Tween-20, pH 6.0). Slides were cooled for 20 minutes, rinsed briefly in PBS, circled with a PAP pen and blocked using reagents from the MOM Basic kit (Vector Labs). The primary antibodies used at the indicated dilutions in the MOM Basic kit were ΔNp63 (RR-14) [1:50; (Romano et al., 2006)] and VDR (1:50; Santa Cruz). Secondary antibodies used were anti-rabbit IgG Alexa Fluor 568 (1:750; Molecular Probes) and anti-mouse IgG Alexa Fluor 488 (1:250; Molecular Probes). Slides were mounted using Vectashield Mounting Medium with DAPI (Vector Labs) viewed with a Nikon FXA fluorescence microscope. Images were captured using a Nikon digital camera and analyzed using ImageJ, Adobe Photoshop, and Adobe Illustrator software. All experimental procedures involving mice were approved by the Institutional Animal Care and Use Committee of the State University of New York at Buffalo. The p63–/– mice were genotyped as previously described (Mills et al., 1999).
We thank members of the Kadakia laboratory for the technical assistance and helpful suggestions. This work was supported by a grant from the NCI/NIH (CA118315-2) to M.K. Deposited in PMC for release after 12 months.