Differential Regulation of Vitamin D Responsive Elements inNormal and Transformed Keratinocytes

Zhongjian Xie and Daniel D. BikleEndocrine Unit, VA Medical Center, University of California, San Francisco, California, U.S.A.

Squamous cell carcinomas (SCC) derived from humanepidermis fail to differentiate normally under the influ-ence of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] despitethe presence of the vitamin D receptor. Previous studiesfrom our laboratory showed that phospholipase C-γ1(PLC-γ1) was upregulated transcriptionally by1,25(OH)2D3 in normal human keratinocytes, and a vit-amin D responsive element (VDRE) in its promoterregion has been identified. To examine the inducibilityof human PLC-γ1 transcription by 1,25(OH)2D3 and/orretinoic acid in SCC cell lines, we transiently transfectedSCC4 and SCC12B2 cells with human PLC-γ1 promoter-luciferase constructs containing the VDRE and tested theresponse of these constructs to 1,25(OH)2D3 and/or all-trans retinoic acid. The induction of the human PLC-γ1VDRE by 1,25(OH)2D3 was synergistic with all-transretinoic acid in normal human keratinocytes, but noneof the constructs was induced by 1,25(OH)2D3 and/or

Phospholipase C (PLC) is a family of isoenzymes thathydrolyzes phosphatidylinositol-4,5-bisphosphate into twosecond messengers, inositol-1,4,5-triphosphate and diacyl-glycerol (Rhee et al, 1989; Majerus et al, 1990). Diacylgly-cerol is the physiologic activator of protein kinase C

(PKC), and inositol triphosphate induces the release of calcium frominternal storage (Berridge and Irvine, 1989). Calcium mobilization andPKC activation are essential for many cellular processes, such as cellproliferation and differentiation (Rana and Hokin, 1990). Severaldistinct PLC isoenzymes have been purified from several mammaliantissues, and several forms have been molecularly cloned. Comparisonof the deduced amino acid sequences has indicated that PLC aredivided into three types (PLC-β, PLC-γ, and PLC-δ), and each typecontains several subtypes (Rhee et al, 1989, 1992). PLC-γ1 has attractedthe most attention because PLC-γ1 is activated by the growth factorswith intrinsic tyrosine kinase activity, such as platelet derived growthfactor (Wahl et al, 1989), epidermal growth factor (Margolis et al, 1989;Meisenhelder et al, 1989), fibroblast growth factor (Burgess et al, 1990),and nerve growth factor (Kim et al, 1991). PLC-γ1 is overexpressed

Manuscript received April 3, 1997; revised December 14, 1997; accepted forpublication January 14, 1998.

Reprint requests to: Dr. Daniel D. Bikle, Endocrine Unit, VA MedicalCenter, 4150 Clement Street (111N), San Francisco, CA 94121.

Abbreviations: DR6, direct repeat 6; KGM, keratinocyte growth medium;NHK, normal human keratinocyte; 1,25(OH)2D3, 1,25-dihydroxyvitamin D;RA, retinoic acid; RAR, retinoic acid receptor; VDR, vitamin D receptor;VDRE, vitamin D responsive element.

0022-202X/98/$10.50 · Copyright © 1998 by The Society for Investigative Dermatology, Inc.

730

all-trans retinoic acid in SCC4 and SCC12B2 cells. Incontrast, the construct containing the VDRE of thehuman 24-hydroxylase gene was induced several fold by1,25(OH)2D3 in normal human keratinocytes and by both1,25(OH)2D3 and all-trans retinoic acid in SCC4 andSCC12B2 cells. DNA mobility shift assays showed thatboth the vitamin D receptor and the retinoic acid receptorin SCC4 and SCC12B2 cells bound the human PLC-γ1VDRE similarly to that seen in normal keratinocytes.The data indicate that the VDRE in the human PLC-γ1gene is not functional in SCC4 and SCC12B2 cells, unlikenormal human keratinocytes, even though vitamin Dreceptors bind normally to it. Failure of transcriptionalcontrol of the PLC-γ1 gene by 1,25(OH)2D3 suggests thelack of a cofactor(s) linking the VDRE to the transcrip-tional machinery. Key words: all-trans retinoic acid/1,25-dihydroxyvitamin D/phospholipase C-γ1/24-hydroxylase. JInvest Dermatol 110:730–733, 1998

in primary human breast carcinoma (Arteaga et al, 1991), humancolorectal cancer (Park et al, 1994), familial adenomatous polyposis(Noh et al, 1994), and hyperproliferative epidermal diseases (Nanneyet al, 1992). The amount of PLC-γ1 protein is higher in neoplastickeratinocyte cell lines than in normal keratinocytes (NHK) (Punnonenet al, 1994). Differentiating keratinocytes express more PLC-γ1 proteinthan undifferentiated keratinocytes (Punnonen et al, 1993). Theseobservations suggest that PLC-γ1 might be involved in the regulationof cell proliferation and differentiation.

NHK differentiation is induced by 1,25-dihydroxyvitamin D3[1,25(OH)2D3] (Hosomi et al, 1983; Smith et al, 1986; Bikle and Pillai,1993; Bikle et al, 1993). The mechanism by which 1,25(OH)2D3induces keratinocyte differentiation is thought to include changes inintracellular calcium, PLC, and PKC activation. PLC-γ1 in NHK isupregulated by 1,25(OH)2D (Pillai et al, 1995). This regulation takesplace at the transcriptional level, and we have recently identified avitamin D responsive element (VDRE) at –786 bp to –803 bp in thehuman PLC-γ1 promoter region (Xie and Bikle, 1997). In contrast,several of the squamous cell carcinoma (SCC) lines fail to differentiatenormally under the influence of 1,25(OH)2D3, despite normal levelsof the vitamin D receptor (VDR) (Ratnam et al, 1996); however, 24-hydroxylase, which is one of the key enzymes in the metabolism ofvitamin D, is stimulated in its activity by 1,25(OH)2D3 in both NHKand SCC (Bikle et al, 1991). To examine the inducibility of humanPLC-γ1 transcription by 1,25(OH)2D3 in SCC lines, we transientlytransfected the SCC4 cell line derived from a carcinoma of humantongue and the SCC12B2 cell line derived from a carcinoma of humanfacial epidermis with human PLC-γ1 promoter-luciferase constructs orhuman 24-hydroxylase promoter-luciferase constructs. We found that

VOL. 110, NO. 5 MAY 1998 VITAMIN D-RESPONSIVE ELEMENTS, KERATINOCYTES 731

the VDRE in the human PLC-γ1 gene was not responsive to1,25(OH)2D3, all-trans retinoic acid (all-trans RA), or their combina-tion in SCC4 and SCC12B2 cells, unlike the VDRE in the human24-hydroxylase gene. Correspondingly, the PLC-γ1 mRNA was upreg-ulated by 1,25(OH)2D3 synergistic with all-trans RA in NHK, butnot in SCC4 and SCC12B2 cells; however, the DNA mobility shiftassay showed that both the VDR and the retinoic acid receptor (RAR)bound to the human PLC-γ1 VDRE in nuclear extracts from NHK,SCC4 and SCC12B2 cells indicating that the defect in transcriptionalregulation in SCC4 and SCC12B2 cells lies elsewhere.

MATERIALS AND METHODS

Cell culture NHK were isolated from neonatal human foreskins and grownin serum free keratinocyte growth medium (KGM, Clonetics, San Diego, CA;Pillai et al, 1988). Briefly, keratinocytes were isolated from newborn humanforeskins by trypsinization (0.25% tyrosine, 4°C, overnight), and primarycultures were established in KGM containing 0.07 mM calcium. Second passagekeratinocytes were plated with KGM plus 0.03 mM calcium for the experiments.SCC4 and SCC12B2 cells were plated under the same conditions as were NHK.

Northern hybridization Total RNA was isolated from NHK, SCC4, andSCC12B2 cells treated with 1,25(OH)2D3 at a final concentration of 10–9 M,all-trans RA at a final concentration of 10–6 M, or vehicle (ethanol) for 24 h,using STAT-60 kitTM (Tel-Test ‘‘B,’’ Friendswood, TX), according to theprocedures recommended by the manufacturer. The isolated RNA (60 µgloaded in each lane) was electrophoresed through a 0.8% agarose-formaldehydegel, transferred to a nylon membrane (Hybond-N1; Amersham, ArlingtonHeights, IL) using a PosiBlot 30–30 Pressure Blotter (Stratagene, La Jolla, CA),and immobilized by baking the membrane at 80°C for 2 h. The human PLC-γ1 cDNA probe (gift from Dr. John Imboden, University of California, SanFrancisco, CA) was labeled with 32P-dCTP (Amersham) by Random Prime-IT, II labeling kit (Stratagene), and purified by NucTrap Probe PurificationColumns (Stratagene). The membrane was prehybridized and hybridized in5 3 sodium citrate/chloride buffer, 5 3 Denhardt’s solution, 0.5% sodiumdodecyl sulfate, and 20 µg salmon sperm DNA per ml with the 32P labeledhuman PLC-γ1 cDNA. Following hybridization at 65°C overnight, themembrane was washed in solutions with decreasing ionic strength and increasingtemperature, to a final stringency of 0.1 3 sodium citrate/chloride buffer, 0.1%sodium dodecyl sulfate, at 65°C. The [32P]cDNA-mRNA hybrids werevisualized by exposing to X-ray film and quantitated by NIH Image. Thehuman PLC-γ1 mRNA was normalized to the levels of 18S ribosomal RNAin the same RNA blots as determined by rehybridization of the filter with a32P-end labeled cDNA for 18S RNA.

DNA transfection and luciferase assay The NHK, SCC4, and SCC12B2cells were transfected with PLC-γ1-luciferase chimeric plasmids (Xie and Bikle,1997) using a polybrene method 24 h after plating cells in 60 mm culturedishes (Jiang et al, 1991). Cells were cotransfected with 0.2 µg of pRSVβ-gal(Carroll and Taichman, 1992), a β-galactosidase expression vector that containsa β-galactosidase gene that is driven by a Rous Sarcoma Virus promoter andenhancer, which was used as an internal control to normalize for transfectionefficiency. 1,25(OH)2D3 and/or all-trans RA was added to the cells 24 h aftertransfection. The final concentration of 1,25(OH)2D3 was 10–9 M. The finalconcentration of all-trans RA was 10–6 M. The same amount of ethanol(vehicle) was added to the control plates. The final concentration of the ethanolwas 1% in the culture medium. Cells were harvested 24 h after addition of1,25(OH)2D3. The cells were lysed and the cell extracts were assayed forluciferase activities using a Luciferase Assay System (Promega, Madison, WI). Theβ-galactosidase activities were assayed using a Galacto-LightTM kit (TROPIX,Bedford, MA). A pGL-3-control vector (Promega) containing SV40 promoterand SV40 enhancer, which we have shown to be unresponsive to 1,25(OH)2D3,was included in each transfection experiment as a control. Every experimentwas done in triplicate and was repeated at least three times.

DNA mobility shift assay The nuclear extracts were made from NHK,SCC4, and SCC12B2 cells according to the method described by Abmayr andWorkman (1994). Synthetic oligonucleotides used for the DNA mobility shiftassay were end-labeled by T4 polynucleotide kinase. The DNA-protein reactionswere performed in a total of 17 µl: nuclear extracts (12 µg protein) wereincubated with 2 µg of poly (dI:dC) (Pharmacia, Piscataway, NJ) and 10,000cpm of 32P labeled probe in 10 ml binding buffer (20 mM HEPES, pH 7.9,20% glycerol, 50 mM KCl, 0.5 mM dithiothreitol) at 30°C for 25 min.Unlabeled competitors were added before incubation. In the super gel-shiftreaction, a polyclonal anti-VDR antibody or polyclonal RAR-γ1 antibody(Affinity Bioreagents, Golden, CO) was added to the DNA-protein reactionand incubated for an additional 25 min. Protein-DNA complexes were

Figure 1. The human PLC-γ1 mRNA level in NHK, SCC4, andSCC12B2 cells in the presence or absence of 1,25(OH)2D3 and/or all-trans RA. NHK, SCC4, and SCC12B2 cells were grown to preconfluence inKGM containing 0.03 mM calcium, at which point 1,25(OH)2D3 (VD), all-trans RA, or the same amount of vehicle (ethanol) was added. The cultureswere maintained for 24 h, and the cells harvested for RNA. The RNA wasisolated, electrophoresed, and blotted on to a membrane. The membrane washybridized with the probes for human PLC-γ1 cDNA and 18S RNA (a, c).The resultant autoradiographs were scanned, and the ratios of PLC-γ1/18Splotted (b, d).

electrophoresed in a 6% nondenaturing polyacrylamide gel in 1 3 gel shiftrunning buffer (50 mM Tris, 380 mM glycerin, 2 mM ethylenediaminetetraacetic acid, pH 8.5).

RESULTS

Human PLC-γ1 mRNA was induced by 1,25(OH)2D3 in NHK,but not in SCC4 and SCC12B2 cells Previous studies in ourlaboratory showed that the effect of 1,25(OH)2D3 on the human PLC-γ1 mRNA expression was biphasic. The peak stimulation was at10–10–10–9 M (Pillai et al, 1995). This study showed that the mRNAfor PLC-γ1 in NHK was induced over 2-fold by 1,25(OH)2D3 (10–9

M) or all-trans RA (10–6 M), and 6-fold by their combination. Incontrast, the mRNA level for PLC-γ1 in SCC4 and SCC12B2 cellsdid not change after the treatment of 1,25(OH)2D3 (10–9 M) or all-trans RA (10–6 M), or their combination. The basal mRNA levels inSCC4 and SCC12B2 cells were much higher than in NHK in theabsence of 1,25(OH)2D3 (Fig 1). In other experiments, the mRNAlevel in either NHK or SCC4 cells was not increased by all-trans RAat a concentration of 10–8 M, whereas, as for the higher dose of all-trans RA described above, the mRNA level for PLC-γ1 was increasedup to 6-fold by the combination of 1,25(OH)2D3 (10–9 M) and all-trans RA (10–8 M) in NHK; no stimulation was seen in SCC4 cells(data not shown).

The VDRE in the human PLC-γ1 promoter is induced by1,25(OH)2D3 and all-trans RA in NHK but not in SCC4 andSCC12B2 cells Nine constructs containing nine deletional fragmentsspanning 1135 bp to –877 bp in the 59 flanking region of the humanPLC-γ1 gene (Xie and Bikle, 1997) were transfected into NHK orSCC4 and SCC12B2 cells in the presence or absence of 1,25(OH)2D3(Fig 2a, b) The profile of the basal promoter activity obtained fromthe nine constructs transfected into SCC4 cells is different from thattransfected into NHK. Deletions to –748 bp stimulated basal activityin SCC4 cells but had no effect in NHK. When the 59 deletionsreached –613 bp, the promoter activity was greatly reduced in SCC4cells, but not in NHK. The constructs containing the fragment 1135to –877 or 1135 to –828 were responsive to 1,25(OH)2D3 whentransfected into NHK, but no response was seen when transfected intoSCC4 cells. The construct containing the human PLC-γ1 VDRE,–786 bp to –803 bp in the human PLC-γ1 promoter (Xie and Bikle,1997) linked to an SV40 promoter, showed a 2-fold increase by1,25(OH)2D3 at a concentration of 10–9 M in promoter activity, whichwas potentiated by all-trans RA at a concentration of 10–6 M whentransfected into NHK (Fig 2c). All-trans RA at a concentration of

732 XIE AND BIKLE THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 2. The human PLC-γ1 promoter activities in NHK, SCC4, andSCC12B2 cells in the presence or absence of 1,25(OH)2D3 and/or all-trans RA. Nine 59 deletional fragments spanning from 1135 bp to –877 bpin the human PLC-γ1 gene were ligated to the luciferase gene in a pGL-3-basic vector. The fragment –768 to –803 was ligated to the SV40 promoterand luciferase gene in a pGL-3-promoter vector. The construct containing theSV40 promoter and SV40 enhancer was used as a control that is known not tobe induced by 1,25(OH)2D3. The constructs were transfected into NHK (a)or SCC4 (b) as described in Materials and Methods. The luciferase activities weremeasured following 24 h exposure to 1,25(OH)2D3 (VD) at a final concentrationof 10–9 M or vehicle, divided by β-galactosidase activity, and expressed as thepercentage of activity of the construct containing SV40 promoter and SV40enhancer in the absence of 1,25(OH)2D3. The activity obtained with the pGL-3-basic vector showed the vector background. A similar experiment wasperformed in NHK (c), SCC4 (d), and SCC12B2 (e) cells transfected withconstructs containing the human PLC-γ1 gene fragment –768 to –803 in thepresence or absence of 1,25(OH)2D3 and/or all-trans RA. The results arenormalized to β-galactosidase activity. A construct containing the vitamin Dresponsive region at –143 to –293 in the human 24-hydroxylase gene (24-hydroxylase) was used as a positive control.

10–8 M had no effect on the promoter activity (data not shown);however, no increase was found when transfected into SCC4 (Fig 2d)and SCC12B2 cells (Fig 2e). SR11246, a retinoid 3 receptor (RXR)selective analog, had no effect on the human PLC-γ1 VDRE in eitherNHK or SCC4 cells (data not shown). The vitamin D responsiveregion at –143 to –293 in the human 24-hydroxylase gene (Chen andDeLuca, 1995), subcloned into the pGL-3-promoter vector, was usedas a positive control in the transfection experiments. In contrast to thePLC-γ1 VDRE, the 24-hydroxylase VDRE construct showed a 2–3-fold induction by 1,25(OH)2D3 in NHK, SCC4, and SCC12B2 cells,and a 2-fold induction by all-trans RA in SCC4 and SCC12B2 cells,but not in NHK (Fig 2c, d, e).

Both the VDR and the RAR in SCC4 and SCC12B2 cells bindto the VDRE in the human PLC-γ1 gene To examine if theVDR in SCC4 and SCC12B2 cells is able to bind the VDRE in thehuman PLC-γ1 gene, the nuclear extracts isolated from NHK, SCC4,or SCC12B2 cells were incubated with a 38 bp oligonucleotidecontaining the direct repeat-6 (DR6) type VDRE (AGGT-CAgaccacTGGACA) of the human PLC-γ1 gene (Xie and Bikle,1997) in a DNA mobility shift assay. The results showed that a singlemain complex formed after the incubation of the oligonucleotide withthe nuclear extracts isolated from NHK, SCC4, or SCC12B2 cells.

Figure 3. The binding of VDR in NHK, SCC4, and SCC12B2 cells withfragments of the 59 flanking region of the human PLC-γ1 gene. Bindingof the human PLC-γ1 gene fragment –786 to –803 containing a DR6 typeVDRE with the VDR or RAR in NHK, SCC4, and SCC12B2 cells.The arrows indicate VDR-RAR-DNA and antibodies-VDR-RAR-DNAcomplexes. VDR-Ab, anti-VDR antibody; RAR-γ1-Ab, anti-retinoic acidreceptor antibody; NHK, normal human keratinocytes; SCC4, SCC4 cells;SCC12B2, SCC12B2 cells.

The binding complex was shifted to a higher molecular form uponthe addition of an antibody specific to VDR or an antibody specificto RARγ1 (Fig 3). Antibodies to RXRα had no effect (Fig 3). Theresults indicate that the VDR and RAR-γ1 in SCC4 and SCC12B2cells bind VDRE in the human PLC-γ1 gene, similarly to that in NHK.

DISCUSSION

The PLC-γ1 protein and mRNA expression are upregulated by1,25(OH)2D3 in human keratinocytes (Pillai et al, 1995). This regulationtakes place at the transcriptional level with the 1,25(OH)2D3 receptorbinding to a DR6 type VDRE in the promoter (Xie and Bikle, 1997).SCC are not induced to differentiate by 1,25(OH)2D3 in termsof transglutaminase or involucrin production or cornified envelopeformation, despite normal levels of the VDR. These results indicatethat the mediators of 1,25(OH)2D3 action on gene expression otherthan the VDR may be missing or defective in SCC4 cells (Ratnamet al, 1996). In this study, we extended these observations by examiningthe human PLC-γ1 mRNA level and promoter activities in thepresence or absence of 1,25(OH)2D3 in NHK, SCC4, and SCC12B2cells. Our results indicate that PLC-γ1 mRNA expression and theVDRE in the human PLC-γ1 promoter are not regulated by1,25(OH)2D3 in SCC4 and SCC12B2 cells, unlike that in NHK;however, this is not due to a universal failure of 1,25(OH)2D3 to acton SCC4 and SCC12B2 cells because the 24-hydroxylase VDRE wasnormally regulated by 1,25(OH)2D3 in these cells. These data suggestthat the inability of the human PLC-γ1 promoter in SCC4 andSCC12B2 cells to respond to 1,25(OH)2D3 could be due to a selectivedefect in the ability of the VDR to activate the VDRE in the humanPLC-γ1 promoter.

An interaction of VDR with RAR has been reported in the DR6type VDRE in the human osteocalcin gene. This VDRE is not onlyresponsive to 1,25(OH)2D3, but also to all-trans RA (Schule et al,1990). In this report, the DR6 type VDRE located between –786 bpand –803 bp upstream from the transcriptional start site in the humanPLC-γ1 gene is functional in NHK, but not in SCC4 and SCC12B2cells. The promoter activities as well as the PLC-γ1 mRNA level inNHK, but not in SCC4 and SCC123B2 cells, were increased by either1,25(OH)2D3 or all-trans RA, and synergistically induced by thecombination of both. In contrast, the construct containing the DR3type VDRE in the human 24-hydroxylase gene was found to beresponsive to 1,25(OH)2D3 in NHK, SCC4, and SCC12B2 cells,but responsive to all-trans RA only in SCC4 and SCC12B2 cells.Furthermore, when the interaction of 1,25(OH)2D3 and RA on theexpression of the differentiating markers involucrin and transglutaminasewas evaluated, RA blocked the increase in involucrin and transglutamin-ase mRNA induced by 1,25(OH)2D3 (Gibson et al, 1997). At thispoint the mechanism by which 1,25(OH)2D3 and all-trans RA regulate

VOL. 110, NO. 5 MAY 1998 VITAMIN D-RESPONSIVE ELEMENTS, KERATINOCYTES 733

the expression of involucrin and transglutaminase is not clear. Thedifferential effect of RA on the stimulation by 1,25(OH)2D3 of PLC-γ1 versus involucrin and transglutaminase suggests different mechanisms.This is an area of particular interest in our laboratory. The mechanismsunderlying the differential response of a gene to an activator like1,25(OH)2D3 and RA within different cells, or even within the samecells at different stages of differentiation, remain elusive. Selective lossof such mechanisms during transformation to the malignant state is ofparticular interest.

The activation of vitamin D responsive genes by 1,25(OH)2D3 isthrough the VDR binding to its VDRE. VDR homodimers andVDR-RAR heterodimers have been reported to bind to DR6 typeVDRE (Carlberg, 1993; Carlberg et al, 1993; Schrader et al, 1993,1994). Our data indicate that all NHK, SCC4, and SCC12B2 cellscontain VDR and RAR-γ1 capable of binding to the DR6 typeVDRE in the human PLC-γ1 gene; however, nuclear receptors suchas VDR interact with additional coactivating or silencing factors inaddition to RXR or RAR to mediate activation and repression ofgene expression (Cavailles et al, 1994; Halachmi et al, 1994; Baniahmadet al, 1995; Horlein et al, 1995; Chen and Evans, 1995; Kurokawaet al, 1995; Kamei et al, 1996; Chakravarti et al, 1996). Identifyingsuch cofactors is expected to provide important clues for the selectivefailure of SCC4, SCC12B2, and other transformed squamous celllines to respond to the anti-proliferating, prodifferentiating actions of1,25(OH)2D3.

These studies were supported by grants from the National Institute of Health(PO1AR39448, R01AR38386) and the American Institute for Cancer Research.

REFERENCES

Abmayr SM, Workman JL: Current Protocol in Molecular Biology. John Wiley, New York, 1994Arteaga CL, Johnson MD, Todderud G, Coffey RJ, Carpenter G, Page DL: Elevated

content of the tyrosine kinase substrate phospholipase C-γ1 in primary human breastcarcinomas. Proc Natl Acad Sci USA 88:10435–10439, 1991

Baniahmad A, Leng X, Burris TP, Tsai SY, Tsai MJ, O’Malley BW: The tau4 activationdomain of the thyroid hormone receptor is required for release of a putativecorepressor(s) necessary for transcriptional silencing. Mol Cell Biol 15:76–86, 1995

Berridge MJ, Irvine RF: Inositol phosphate and cell signaling. Nature 341:197–205, 1989Bikle DD, Pillai S: Vitamin D, calcium and epidermal differentiation. Endocr Rev 14:3–

19, 1993Bikle DD, Pillai S, Gee E: Squamous carcinoma cell lines produce 1,25 dihydroxyvitamin

D, but fail to respond to its prodifferentiating effect. J Invest Dermatol 97:435–441, 1991Bikle DD, Gee E, Pillai S: Regulation of keratinocyte growth, differentiation, and vitamin

D metabolism by analogs of 1,25-dihydroxyvitamin D. J Invest Dermatol 101:713–718, 1993

Burgess WH, Dionne CA, Kaplow J, et al: Characterization and cDNA cloning ofphospholipase C-γ, a major substrate for heparin binding growth factor 1 (acidicfibroblast growth factor) activated tyrosine kinase. Mol Cell Biol 10:4770–4777, 1990

Carlberg C: RXR-independent action of the receptors for thyroid hormone, retinoid acidand vitamin D on inverted palindromes. Biochem Biophys Res Commun 195:1345–1353, 1993

Carlberg C, Bendik I, Wyss A, Meier E, Sturzenbecker LJ, Grippo JF, Hunziker W: Twonuclear signaling pathways for vitamin D. Nature 361:657–660, 1993

Carroll JM, Taichman LB: Characterization of the human involucrin promoter using atransient beta-galactosidase assay. J Cell Sci 103:925–930, 1992

Cavailles V, Dauvois S, Danielian PS, Parker MG: Interaction of proteins withtranscriptionally active estrogen receptors. Proc Natl Acad Sci USA 91:10009–10013, 1994

Chakravarti D, LaMorte VJ, Nelson MC, et al: Role of CBP/P300 in nuclear receptorsignaling. Nature 383:99–103, 1996

Chen K-S, DeLuca HF: Cloning of the human 1 α,25-dihydroxyvitamin D-3, 24-hydroxylase gene promoter and identification of two vitamin D-responsive elements.Biochem Biophys Acta 1263:1–9, 1995

Chen JD, Evans RM: A transcriptional co-repressor that interacts with nuclear hormonereceptors. Nature 377:454–457, 1995

Gibson DFC, Bikle DD, Harris J: All-trans retinoic acid blocks the antiproliferativeprodifferentiating actions of, 1, 25–dihydroxyvitamin D3 in normal humankeratinocytes. J Cellular Physiol 174:1–8, 1998

Halachmi S, Marden E, Martin G, MacKay I, Abbondanza C, Brown M: Estrogenreceptor-associated proteins: possible mediators of hormone-induced transcription.Science 264:1455–1458, 1994

Horlein AJ, Naar AM, Heinzel T, et al: Ligand-independent repression by the thyroidhormone receptor mediated by a nuclear receptor co-repressor. Nature 377:397–404, 1995

Hosomi J, Hosoi J, Abe E, Suda T, Kuroki T: Regulation of terminal differentiationof cultured mouse epidermal cells by 1α,25-dihydroxyvitamin D3. Endocrinology113:1950–1957, 1983

Jiang C-K, Connolly D, Blumenberg M: Comparison of methods for transfection ofhuman epidermal keratinocytes. J Invect Dermatol 97:969–973, 1991

Kamei Y, Xu L, Heinzel T, et al: A CBP integrator complex mediates transcriptionalactivation and AP-1 inhibition by nuclear receptors. Cell 85:403–414, 1996

Kim UH, Fink D, Kim HS, Park DJ, Contreras ML, Guroff G, Rhee SG: Nerve growthfactor stimulates phosphorylation of phospholipase C-γ in PC12 cells. J Biol Chem266:1359–1362, 1991

Kurokawa R, Soderstrom M, Horlein A, Halachmi S, Brown M, Rosenfeld MG, GlassCK: Polarity-specific activities of retinoic acid receptors determined by a co-repressor. Nature 377:451–454, 1995

Majerus PW, Ross TS, Cunningham TW, Caldwell KK, Jefferson AB, Bansal VS: Recentinsights in phosphatidylinositol signaling. Cell 63:459–465, 1990

Margolis B, Rhee SG, Felder S, et al: EGF induces tyrosine phosphorylation of phospholipaseC-II. a potential mechanism for EGF-receptor signaling. Cell 57:1101–1107, 1989

Meisenhelder J, Suh PG, Rhee SG, Hunter T: Phospholipase C-γ is a substrate for thePDGF and EGF receptor protein tyrosine kinases in vivo and in vitro. Cell 57:1109–1122, 1989

Nanney LB, Gates RE, Todderud G, King LE Jr, Carpenter G: Altered distribution ofphospholipase C-γ1 in benign hyperproliferative epidermal diseases. Cell Growth Diff3:233–239, 1992

Noh D-Y, Lee YH, Kim SS, Kim YI, Ryu S-H, Suh P-G, Park JG: Elevated content ofphospholipase C-γ1 in colorectal cancer tissues. Cancer 73:36–41, 1994

Park J-G, Lee YH, Kim SS, Park KJ, Noh D-Y, Ryn SH, Suh P-G: Overexpressionof phospholipase C-γ1 in familial adenomatous polyposis. Cancer Res 54:2240–2244, 1994

Pillai S, Bikle DD, Hincenbergs M, Elias PM: Biochemical and morphologicalcharacterization of growth and differentiation of normal neonatal human keratinocytesin a serum-free medium. J Cell Physiol 134:229–237, 1988

Pillai S, Bikle DD, Su M-J, Ratnam A, Abe J: 1,25-Dihydroxyvitamin D3 upregulatesthe phosphatidylinositol signaling pathway in human keratinocytes by increasingphospholipase C level. J Clin Invest 96:602–609, 1995

Punnonen K, Denning M, Lee E, Li L, Rhee SG, Yuspa SH: Keratinocyte differentiationis associated with changes in the expression and regulation of phospholipase Cisoenzymes. J Invest Dematol 101:719–726, 1993

Punnonen K, Denning MF, Rhee SG, Yuspa SH: Differences in the regulation ofphosphatidylinositol-specific phospholipase C in normal and neoplastic keratinocytes.Mol Carcinogenesis 10:216–225, 1994

Rana RS, Hokin LE: Role of phosphoinositides in transmembrane signaling. Physiol Rev70:115–164, 1990

Ratnam AV, Bikle DD, Su MJ, Pillai S: Squamous carcinoma cell lines fail to respond to1, 25-dihydroxyvitamin D despite normal level of the vitamin D receptor. J InvestDermatol 106:522–525, 1996

Rhee SG, Choi KD: Regulation of inositol phospholipid-specific phospholipase Cisoenzymes. J Biol Chem 267:12393–12396, 1992

Rhee SG, Suh P-G, Ryu SH, Lee SY: Studies of inositol phospholipid-specific phospholipaseC. Science 244:546–550, 1989

Schrader M, Bendik I, Becker-Andre M, Carlberg C: Interaction between retinoic acidand vitamin D signaling pathways. J Biol Chem 268:17830–17836, 1993

Schrader M, Muller KM, Carlberg C: Specificity, flexibility of vitamin D signaling.Modulation of the activation of natural vitamin D response elements by thyroidhormone. J Biol Chem 269:5501–5504, 1994

Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM: Jun-Fos andreceptors for vitamins A and D recognize a common response element in the humanosteocalcin gene. Cell 61:497–504, 1990

Smith EL, Walworth NC, Holick MF: Effect of 1α, 25-dihydroxyvitamin D3 onthe morphologic and biochemical differentiation of cultured human epidermalkeratinocytes grown in serum-free conditions. J Invest Dermatol 86:709–714, 1986

Wahl MI, Olashaw NE, Nishibe S, Rhee SG, Pledger WJ, Carpenter G: Platelet-derivedgrowth factor induces rapid and sustained tyrosine phosphorylation of phospholipaseC-γ1 in quiescent BALB/c 3T3 cells. Mol Cell Biol 9:2934–2943, 1989

Xie Z, Bikle DD: Cloning of the human phospholipase C-γ1 promoter and identificationof a DR6-type vitamin D responsive element. J Biol Chem 272:6573–6577, 1997