The Candidate Tumor Suppressor Gene ZAC Is Involved in ...

The Candidate Tumor Suppressor Gene ZAC Is Involved in Keratinocyte Differentiation and Its Expression Is Lost in Basal Cell Carcinomas Eugenia Basyuk,1,2 Vincent Coulon,3 Anne Le Digarcher,1 Marjorie Coisy-Quivy,2 Jean-Pierre Moles,3 Alberto Gandarillas,2,3 and Laurent Journot1 1

Institut de Geńomique Fonctionnelle, CNRS-UMR5203, INSERM-U661, Universite´ Montpellier 1, Universite´ Montpellier 2; 2Institut de Geńe´tique Molećulaire, CNRS-UMR5535, Universite´ Montpellier 2; and 3Laboratoire de Dermatologie Molećulaire, IURC, EA3754, Montpellier, France

Abstract ZAC is a zinc finger transcription factor that induces apoptosis and cell cycle arrest in various cell lines. The corresponding gene is maternally imprinted and localized on chromosome 6q24-q25, a region harboring an unidentified tumor suppressor gene for a variety of solid neoplasms. ZAC expression is lost or down-regulated in some breast, ovary, and pituitary tumors and in an in vitro model of ovary epithelial cell transformation. In the present study, we examined ZAC expression in normal skin and found a high expression level in basal keratinocytes and a lower, more heterogeneous, expression in the first suprabasal differentiating layers of epidermis. In vitro, ZAC was up-regulated following induction of keratinocyte differentiation. Conversely, ZAC expression triggered keratinocyte differentiation as indicated by induction of involucrin expression. Interestingly, we found a dramatic loss of ZAC expression in basal cell carcinoma, a neoplasm characterized by a relatively undifferentiated morphology. In contrast, ZAC expression was maintained in squamous cell carcinomas that retain the squamous differentiated phenotype. Altogether, these data suggest a role for ZAC at an early stage of keratinocyte differentiation and further support its role in carcinogenesis. (Mol Cancer Res 2005;3(9):483 – 92)

Introduction Zac1 is a mouse zinc finger transcription factor, which induces apoptosis and cell cycle arrest in vitro, and abrogates tumor formation in nude mice (1). Zac1 binds DNA and

Received 2/22/05; revised 8/5/05; accepted 8/11/05. Grant support: Centre National de la Recherche Scientifique, La Ligue Nationale contre le Cancer, and L’Association pour la Recherche contre le Cancer (A. Gandarillas and L. Journot), La Ligue Nationale contre le Cancer fellowship (V. Coulon), and European Molecular Biology Organization fellowship (A. Gandarillas). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Requests for reprints: Laurent Journot, Institut de Geńomique Fonctionnelle, 141, rue de la cardonille, F-34094 Montpellier Cedex 5, France. Phone: 33-467142-932; Fax: 33-467-542-432. E-mail: [email protected] Copyright D 2005 American Association for Cancer Research. doi:10.1158/1541-7786.MCR-05-0019

Mol Cancer Res 2005;3(9). September 2005

displays transactivation or transrepression activities depending on the precise arrangement of its binding sites (2), and accumulating evidence suggests its involvement in the control of cell proliferation and differentiation. Thus, inhibition of Zac1 expression by antisense oligonucleotides enhances pituitary cell proliferation (3), and expression of Lot1, the rat orthologue of Zac1, is lost during the spontaneous transformation of rat ovary surface epithelial cells in vitro (4). Interestingly, Zac1/Lot1 is negatively regulated by epidermal growth factor, suggesting that it might be involved in a feedback loop controlling cell proliferation on mitogen activation (5). In addition to its role as a transcription factor, Zac1 serves as a transcriptional coactivator or corepressor of nuclear receptors and binds to the nuclear receptor coactivators CBP/p300 and p160 (6). More recently, it was suggested that Zac1 is a transcriptional coactivator for p53 (7) and that it regulates the proapoptotic gene Apaf-1 (8). The human orthologue ZAC/LOT1/PLAGL1 is widely expressed in normal tissues and the corresponding protein binds DNA and displays transactivation activity (9). As its mouse counterpart, human ZAC displays an antiproliferative activity by inducing apoptosis and cell cycle arrest in transfected cells. In addition, alternative splicing of human ZAC transcript generates two mRNA species encoding proteins with different number of zinc finger domains (5 versus 7), which differentially regulate apoptosis and cell cycle arrest (10). Noteworthy, ZAC maps to chromosome 6q24-q25 (9, 11), a region frequently deleted in many solid tumors and thought to harbor at least one tumor suppressor gene. Loss of heterozygosity at 6q24-q25 occurs in breast and ovary cancer, melanomas, astrocytomas, and renal cell carcinomas. The functional properties of ZAC and the chromosomal location of the corresponding gene make it a plausible candidate for the tumor suppressor gene at 6q24-q25. In this context, loss or down-regulation of ZAC expression is observed in a large proportion of mammary (12) and ovary (11, 13) tumor-derived cell lines and tumors, in nonfunctioning pituitary adenomas (14), and in head and neck squamous cell carcinoma (SCC; ref. 15). Expression profiling of human breast tumors (16), lung adenocarcinomas (17), and breast tumor – derived cell lines (16, 18) with microarrays confirmed downregulation of ZAC expression. We found that ZAC expression is reinduced in breast tumor – derived cell lines by treatment with the demethylating agent azacytidine, suggesting that the

483

484 Basyuk et al.

methylation status of ZAC gene critically modulates its expression (12). This hypothesis was further supported by the demonstration of ZAC imprinting in human (19, 20) and mice (21) and by the finding that methylation of ZAC promoter is responsible for its observed monoallelic expression (22, 23). Although the above-mentioned data support a role for ZAC as a tumor suppressor gene, its precise mechanism of action and target genes are mostly unknown at present. Because ZAC is potentially involved in the homeostasis of mammary and ovary epithelial cells, we assessed its expression in different epithelial tissues, including epidermis. Human epidermis is a stratified, constantly renewed epithelium in which proliferation and differentiation are compartmentalized and tightly controlled. It consists of basal, suprabasal (lower spinous, upper spinous, and granular), and cornified layers. The basal layer comprises two populations of keratinocytes: stem cells and transit amplifying cells that initiated their differentiation program after a rapid phase of clonal expansion (24, 25). Differentiating keratinocytes detach from the basal membrane, increase in size, and transform into a cornified envelope that sheds from the surface of the skin (26). Therefore, the epidermis is an excellent model to study the regulation of proliferation and differentiation in vivo and its alteration during carcinogenesis. In the present study, we analyzed ZAC expression during differentiation of human primary keratinocytes in vitro and in normal and pathologic skin samples, including psoriasis

and two common types of nonmelanoma skin cancer [i.e., basal cell carcinomas (BCC) and SCCs]. Our data suggest a role for ZAC in keratinocyte differentiation and BCC formation.

Results ZAC Expression in Normal Skin We first examined ZAC expression in five normal human skin samples using nonradioactive in situ hybridization with digoxigenin-labeled riboprobes. As a positive control for in situ hybridization, we used a probe for the differentiation marker cytokeratin K1, which prominently labeled the suprabasal spinous layers of epidermis and, to a lesser extent, the granular layer but not the basal layer (Fig. 1A). As a negative control, identical samples were hybridized with a probe for lacZ that produced no signal (Fig. 1B). In situ hybridization with the ZAC probe showed a strong and homogenous expression in the basal layer of epidermis (Fig. 1C and D). We detected as well a faint and more heterogeneous expression in the spinous layers. Interestingly, in one normal skin sample, we detected ZAC expression in all suprabasal keratinocytes, including the lower and upper spinous and granular layers (data not shown), indicative of some individual variations. We also observed ZAC expression in the dermal fibroblasts (Fig. 1D) and in the epithelial cells of the sudoriferous glands (Fig. 1E). In contrast, no expression was observed in the endothelium of the blood vessels (Fig. 1F).

FIGURE 1. ZAC expression in normal skin. Sections from normal skin samples were processed for in situ hybridization with probes for cytokeratin 1 (A), h-galactosidase (B), or ZAC (C-F). A-D. Epidermis. E. Sudoriferous gland. F. Blood vessel. A, B, and D-F. The field is 1.5 1.2 mm. C. The field is 3 2.4 mm. Mol Cancer Res 2005;3(9). September 2005

ZAC Is Involved in Keratinocyte Differentiation

ZAC Expression Is Induced in Early Differentiating Keratinocytes In vitro To gain insights into the role of ZAC in epithelial cell biology, we analyzed ZAC expression during keratinocyte differentiation in vitro. Primary culture of human keratinocytes is widely used to study factors that regulate stem cell proliferation, the initiation of terminal differentiation and tissue assembly, and the process of neoplastic transformation. Primary keratinocytes from human foreskin were cultured under lowcalcium and low-serum conditions in a specialized keratinocyte growth medium that keeps these cells as a monolayer, in an undifferentiated stage, for extended periods of time (27, 28). Terminal differentiation under these conditions can be initiated in keratinocytes, and a proportion of cells within the monolayer express the differentiation marker involucrin (29) and eventually shed into the culture medium (30). When the calcium concentration is raised, cells start to form desmosomes and stratify with selective migration of involucrin-positive cells out of the basal layer (31). We detected low level of ZAC expression in the keratinocyte cultures grown under lowcalcium conditions by in situ hybridization (Fig. 2B). Because addition of calcium to the culture medium increases the proportion of differentiating keratinocytes in the stratified layer, this system allows the evaluation of ZAC expression during differentiation. We did real-time reverse transcriptionPCR (RT-PCR) with total RNAs derived from control keratinocytes cultured in low-calcium medium and from keratinocytes at different time points after raising calcium concentration (Fig. 2A). We observed an induction of ZAC expression in keratinocytes stimulated with calcium as early as 8 hours following calcium addition. ZAC expression levels remained high until 24 hours and then gradually declined at 48 and 72 hours. This finding was further corroborated by in situ hybridization data (Fig. 2B and C). We noted a heterogeneous distribution of ZAC signal in cells grown in the presence of 1.5 mmol/L calcium where cells in cluster expressed high levels of ZAC (Fig. 2C), whereas isolated cells displayed low ZAC expression, reminiscent of the situation in control cells (Fig. 2B). To further assess the regulation of ZAC expression during keratinocyte differentiation, we used another differentiation system (i.e., anchorage-independent growth in methylcellulose). In contrast to the calcium-induced stratification system, keratinocytes grown under these conditions exhibit a certain extent of synchronization. Keratinocytes are irreversibly committed by 5 to 6 hours and undergo terminal differentiation by 24 hours when most of them express the differentiation marker involucrin (32). Total RNAs from adherent keratinocytes or from keratinocytes placed in suspension for different period of time were analyzed by Northern blot hybridization with ZAC cDNA as a probe (Fig. 3A). To control for RNA quality and equal loading, we rehybridized the Northern blots with a probe against 18S rRNA. We repeated the same experiment with an independent preparation of primary keratinocytes and analyzed ZAC expression levels using realtime RT-PCR (Fig. 3B). In both experiments, loss of anchorage resulted in a transient increase in ZAC expression after 3 to 6 hours in suspension. At that time, keratinocytes loose their ability to proliferate and become irreversibly Mol Cancer Res 2005;3(9). September 2005

FIGURE 2. ZAC expression in primary keratinocytes is induced by calcium. Normal keratinocytes were grown in vitro in a specialized medium with low-calcium concentration. A. ZAC expression level was quantified by real-time RT-PCR at different time points after calcium concentration was raised to 2 mmol/L. Normalization of ZAC expression level was done using geometric averaging of B2M, ARP, and hypoxanthine phosphoribosyltransferase expression levels. Points, mean of two independent experiments done in triplicate; bars, SE. *, P < 0.01, Student’s t test. B and C. In situ hybridization on keratinocytes grown in low calcium (B) and 24 hours after raising calcium concentration (C).

committed to differentiation. After 12 hours in suspension, DNA synthesis in the keratinocytes is inhibited. At this time point, ZAC expression decreased significantly and remained low after 24 hours in suspension when the cells terminally differentiate (Fig. 3A and B). Interestingly, trypsinization of the cells rapidly up-regulated ZAC expression as evidenced by Northern blot (Fig. 3C) or real-time RT-PCR (data not shown) analyses. ZAC-Induced Keratinocyte Differentiation In vitro Because ZAC was expressed during early stages of keratinocytes differentiation in normal skin and in primary

485

486 Basyuk et al.

keratinocytes, we tested whether it was able to induce keratinocytes differentiation in vitro. We transfected HaCaT, a keratinocyte-derived cell line, with plasmids encoding ZAC splice variants, ZAC and ZACD2, which are both expressed in normal epidermis (data not shown). ZACD2 is a variant of ZAC, which lacks the two NH2-terminal zinc fingers, and differs from ZAC in its ability to control cell cycle exit and apoptosis (10). A green fluorescent protein – encoding vector was transfected as a control in our experiments. Thirty-six hours after transfection, we did immunostainings for ZAC and the differentiation marker involucrin. The number of cells positive for both involucrin and ZAC was determined and compared with the number of green fluorescent protein – and involucrin-positive cells in the control. ZAC or ZACD2 expression led to a significant increase of involucrin level in transfected cells (Fig. 4). These results suggest that ZAC expression is sufficient to promote keratinocyte differentiation in vitro.

FIGURE 3. ZAC expression is regulated by anchorage-independent growth. A and B. Normal keratinocytes were grown in a specialized medium containing methylcellulose for the indicated period of time. A. Total RNAs were prepared and analyzed by Northern blotting with a ZAC probe and a 18S rRNA probe. B. ZAC expression level was quantified as in Fig. 2A but for the control genes B2M, ARP, glyceraldehyde-3phosphate dehydrogenase, and PP1A. Points, mean of a representative experiment done in triplicate; bars, SE. *, P < 0.01, Student’s t test. C. Northern blot analysis of ZAC and 18S rRNA expression in keratinocytes before and after trypsin treatment.

ZAC Expression in Pathologic Skin Samples To find out whether the up-regulation of ZAC expression observed during differentiation in vitro was associated with the decrease of the proliferative potential and the initiation of terminal differentiation in vivo, we analyzed ZAC expression in hyperproliferative disorders of the skin (i.e., psoriasis and skin tumors). Psoriasis is a pathology of the skin characterized by an increased proliferation rate and abnormal differentiation that leads to thickening of the proliferative and differentiated layers and to an increase of the epidermal turnover (33). We did in situ hybridization on sections of psoriatic human skin biopsies. In all five samples examined, we observed rather extended ZAC expression with respect to normal skin. ZAC mRNA was expressed in the thickened basal and suprabasal layers of the psoriatic epidermis (Fig. 5A and C). The expression was strong and homogeneous in the basal layers and much weaker and heterogeneous in the suprabasal layers. Control hybridization with the h-galactosidase probe produced no signal (data not shown) and a probe for cytokeratin K1 – labeled suprabasal layers (Fig. 5B and D). Thus, we observed increased expression of ZAC in psoriatic epidermis, consistent with the delay in the initiation of terminal differentiation that is typical for this type of skin pathology. We further examined ZAC expression in two types of skin cancer, BCCs and SCCs. Morphologically, SCC is characterized by an irregular and highly disorganized keratinocyte proliferation, forming diffuse structures (34). SCC is locally more invasive than BCC. It consists of squamous cells, which are usually more differentiated than BCC-forming cells, and it has a high ability to develop metastasis. We found a strong ZAC expression in all eight SCCs examined by in situ hybridization (Table 1; Fig. 6A and B). On one section, the epidermis overlaying the tumor was present and we noted that ZAC was expressed in the basal and suprabasal layers. We also observed ZAC expression in the lymphocytes infiltrating the tumors (data not shown). Altogether, our data suggest that ZAC is not involved in SSC formation. BCC is the most common type of human cancer (35). Morphologically, this tumor is characterized by tumor nodules Mol Cancer Res 2005;3(9). September 2005


inf iltrating dermis, with characteristic peripheral palisading. BCC nodules are composed of small nondifferentiated cells with a basal cell-like phenotype (36). Familial BCCs, called Gorlin syndrome or nevoid BCC syndrome, have loss-offunction mutations of PATCHED (PTCH), which encodes a SONIC HEDGEHOG (SHH) receptor (37, 38). We examined 13 sporadic and 5 familial BCCs by in situ hybridization and found loss of ZAC expression in all sporadic BCC (Table 1). ZAC expression was absent from all tumor nodules but weakly present in the basal and suprabasal layers of the epidermis overlaying the tumor on the same tissue section (Fig. 6C-E). In several cases, ZAC expression was not detected even in the epidermis overlaying the tumor. In the biopsies of BCC taken from patients with Gorlin syndrome, ZAC expression was completely lost in three of five cases. In the other two cases, it was strongly diminished when compared with normal skin taken from the same patients where expression was noted in the basal and suprabasal layers of epidermis (Fig. 6F). In these cases, a very weak ZAC expression was observed in the middle of the tumoral nodules but not on their periphery where most proliferating cells appear to be located (39). ZAC Expression Is Not Regulated by the SHH Pathway Constitutive activation of the SHH pathway is sufficient to induce BCC in transgenic mice (40) and human skin (41). The observed loss or decrease of ZAC expression in all BCC examined suggested a possible down-regulation of ZAC by the activation of the SHH pathway. To test this hypothesis, we analyzed ZAC expression in primary keratinocytes grown in the presence of recombinant SHH. It was shown previously that SHH opposes calcium-induced cell cycle arrest in differentiating keratinocytes (42). Primary keratinocytes were grown to 60% to 80% confluence in low-calcium conditions before the calcium concentration was raised to 1.5 mmol/L and SHH was simultaneously added to the culture medium. Total RNAs from cells cultivated with or without SHH for 4 and 24 hours were analyzed by RT-PCR. To test for RNA integrity, each RNA sample was analyzed for actin expression. No effect on ZAC expression was observed after addition of SHH to the culture medium (Fig. 7). To verify that recombinant SHH was active in this experiment, the up-regulation of BMP-2B, a member of the transforming growth factor-h family and a known target of SHH in mammals (43), was verified and confirmed by RT-PCR (Fig. 7).

Discussion

FIGURE 4. ZAC promotes involucrin expression in HaCaT cells. HaCaT cells were transfected with green fluorescent protein (GFP ), ZAC, or ZACD2 and stained with 4V ,6-diamidino-2-phenylindole (blue ), anti-ZAC (red), and anti-involucrin (green) antisera. Top, high-magnification pictures of ZAC- and ZACD2-positive cells; bottom, involucrinpositive cells and transfected cells (i.e., green fluorescent protein – or ZAC-positive cells) were scored and the ratios of involucrin-positive cells to transfected cells were calculated for each condition. Columns, mean of three independent experiments; bars, SD. *, P < 0.01, Student’s t test.


We undertook the present study to test a possible involvement of ZAC in epidermal differentiation and to evaluate its implication in the development of skin cancer. We detected ZAC expression in the basal and, to a lower level, suprabasal layers of normal epidermis. Using two models of keratinocyte differentiation in vitro, we observed an early upregulation of ZAC expression following induction of differentiation. Interestingly, in both models, ZAC induction is transient and followed by a down-regulation. These observations are in good agreement with the distribution of ZAC mRNA in normal skin. Keratinocytes initiate differentiation within the basal layer where ZAC is abundantly expressed. The differentiation process

487

488 Basyuk et al.

FIGURE 5. ZAC expression in psoriasis samples. Two psoriasis samples (A-D) were analyzed by in situ hybridization for ZAC (A and C) and cytokeratin 1 (B and D) expression. A-C. The field is 1.2 1.2 mm. D. The field is 2.4 2.4 mm.

proceeds during the migration of these cells toward the outmost layers of the skin where ZAC expression gradually disappears. In addition, we showed that ZAC expression in an immortalized keratinocyte cell line promotes keratinocyte differentiation. Collectively, this set of data suggests that ZAC expression contributes to keratinocyte differentiation. We found a complete loss or a significant decrease in ZAC expression in all sporadic and familial BCCs examined. This observation is in agreement with the proposed tumor suppressor function for ZAC and with the observed loss or down-regulation of ZAC expression in breast, ovary, and pituitary tumors (12 – 14). BCC is the most common type of neoplasm and its incidence is dramatically increasing. BCC metastasis occurs only locally and sporadically in relation to sun exposure. Its incidence is increased significantly in some rare genetic disorders, such as Gorlin syndrome. Familial BCCs and about one-third of sporadic BCCs have loss-offunction mutations of PATCHED (37, 38), which encodes a receptor for the morphogen SHH. Activation of PATCHED negatively regulates the activity of SMOOTHENED, a seventransmembrane domain receptor controlling the activation of the GLI family of transcription factors (44, 45). Moreover, human sporadic BCCs consistently express GLI-1 (46). In agreement with these data, overexpression of Shh (40) or GLI-1 (47) is sufficient to induce BCC-like tumors in transgenic mice. Interestingly, ZAC and GLI-1 display opposite patterns of expression. Indeed, GLI-1 expression is not detected in basal keratinocytes of the normal skin, which display high expression of ZAC in our in situ experiments. Conversely, GLI-1 is

expressed in all sporadic and familial BCCs, where ZAC is lost. We thus assessed whether the SHH-PTCH-SMOH-GLI pathway modulates ZAC expression and found no regulation of ZAC by recombinant SHH. Hence, ZAC loss of expression is apparently not directly controlled by the aberrant activation of the SHH pathway in BCC. Besides PTCH, very few other tumor suppressor genes have been involved in BCC formation and/or development. One interesting candidate is TP53 for which a specific, UV-induced mutation spectrum has been reported in sporadic BCC (48, 49). The recent demonstration that mouse Zac1 behaves as a p53 coactivator (7) suggests that the loss of ZAC expression in sporadic BCC may correlate with a decrease in the p53-induced response to DNA damage. Noteworthy, Zac1 cotransfection was reported to potentiate p53-mediated stimulation of p21WAF1/Cip1 expression (7), which is involved in keratinocyte differentiation (50). Our data regarding ZAC expression in vitro and in vivo suggest a role for ZAC in the early stages of keratinocyte differentiation. Due to its functional properties (1, 9, 10), one would predict that ZAC would be involved in the cell cycle arrest, which usually accompanies the induction of terminal differentiation in most tissues. Indeed, in normal epidermis, the progressive loss of ZAC expression in the suprabasal layers could be correlated with an irreversible proliferation shutdown. However, several lines of evidence indicate that ZAC is most likely not directly involved in negative regulation of cell cycle progression in the epidermis. First, the maintaining of ZAC expression in the hyperproliferative psoriasis and SCC indicates Mol Cancer Res 2005;3(9). September 2005


no link between ZAC expression and a proliferation block. Second, epidermal differentiation is not invariably accompanied by a complete withdrawal from the cell cycle, because keratinocytes retain the ability to synthesize DNA (51, 52) and continue to grow in size during differentiation (26). Finally, SCCs and psoriasis retain the capacity to differentiate and express ZAC, whereas BCC cells display a nondifferentiated phenotype and do not express ZAC. All these data suggest that ZAC plays a role in early events of keratinocyte differentiation, at a step distinct from the control of the cell cycle progression. Interestingly, keratinocytes of suprabasal layers of normal and psoriatic epidermis and SCC display increased cell size, whereas BCCs are formed by small-sized cells. ZAC may thus have a role in the cellular growth that takes place at the initiation of differentiation that is altered in SCCs and lost in BCCs. In addition, it was shown recently that cytokeratin 14 is a direct target of mouse Zac1 and that several other cytokeratins are positively or negatively regulated by Zac1 (2). Consistent with the putative role of its human counterpart in keratinocytes differentiation, Zac1 was observed to repress cytokeratin 19, which has been proposed as an interfollicular stem cell marker (53), and to activate cytokeratin 14, a component of the basal layer where ZAC is also present (24).

Table 1. ZAC expression in normal and pathologic skin samples was assessed by in situ hybridization and visual scoring n

Diagnosis

308 281 280 221 220 392 314 394 466 366-2 436 218bs 448 354 355 357 315 318 366-1 430 432 345 397 287 442 431 467 442n 455 428 356 408 427

BCC BCC BCC BCC BCC BCC BCC BCC BCC BCC BCC BCC BCC BCC (NBCCS) BCC (NBCCS) BCC (NBCCS) BCC (NBCCS) BCC (NBCCS) SCC SCC SCC SCC SCC SCC SCC SCC Normal skin Normal skin Normal skin Normal skin Normal skin (NBCCS) Normal skin (scalp) Normal skin (scalp)

ZAC expression

+/ +/ + +++ ++ + + + ++ + ++ + ++ ++ +++ + +

NOTE: , no ZAC expression; +/ , very weak ZAC expression; +, ++, and +++, increasing levels of ZAC expression. Abbreviation: NBCCS, a sample from a patient with nevoid BCC (Gorlin) syndrome.


Finally, our data also unveil a potential role for the loss of the candidate tumor suppressor gene ZAC in the development of BCC. Additional experiments aiming at identifying ZAC target genes will provide insights into the process of keratinocyte differentiation that is lost in BCC.

Materials and Methods Cell Culture HaCaT keratinocyte cells were maintained in DMEM plus 10% fetal bovine serum and transfected using Lipofectamine 2000 (IVG, Cergy, France). Plasmids encoding ZAC and ZACD2 were described previously (10). Monolayer cultures of primary human keratinocytes were prepared from neonatal foreskins as described (54). Cells were maintained in serumfree keratinocyte growth medium K-SFM (IVG). Stratification was induced by addition of 2 mmol/L calcium or 10% fetal bovine serum to the medium. Cells were harvested for RNA isolation at different time points after induction. To study the effect of SHH on ZAC expression, the keratinocytes were grown to 60% to 80% confluence before the concentration of calcium in the medium was raised to 1.5 to 2 mmol/L. Recombinant SHH (Research Diagnostics, Inc., Flanders, NJ) was simultaneously added at a final concentration of 20 Ag/mL. Cells were harvested for RNA isolation at different time intervals after SHH addition. To induce terminal differentiation, the keratinocytes were trypsinized and maintained in suspension in 1.75% methylcellulose as described previously (32). RNA Isolation and Northern Blot Analysis Total RNA was isolated from keratinocyte cultures using Trizol (LTI, Cergy, France). Northern blots of total RNAs (20 Ag) were hybridized with a human ZAC probe (nucleotides 6842,800 of the human cDNA; Genbank accession no. AJ006354) and a probe for 18S rRNA. Hybridization was in Church buffer [0.5 mol/L phosphate buffer (pH 7), 7% SDS, 2% bovine serum albumin] and final washing step was at 65jC in 1 SSC and 0.2% SDS. Real-time and Semiquantitative PCR Analysis cDNAs were generated from 1 Ag total RNAs treated with DNase I by using random hexamers and Moloney murine leukemia virus reverse transcriptase (LTI). Real-time PCR was done using Applied Biosystems (Courtaboeuf, France) SYBR Green PCR mix according to the manufacturer’s instructions. Housekeeping genes used to normalize ZAC expression levels were selected using the geNorm process (55). The sequences of the primers for real-time PCR are ZAC 5V-CCACTCACAGGAGCTGATGAAA and 5V-GCAGCCTTCAGTTGGAATGAA, B2M 5V-TGACTTTGTCACAGCCCAAGATA and 5VCGGCATCTTCAAACCTCCA, ARP 5V-CTCCAAGCAGATGCAGCAGA and 5V-CCCATCAGCACCACAGCC, hypoxanthine phosphoribosyltransferase 5V-CGGCTCCGTTATGGCG and 5V-GGTCATAACCTGGTTCATCATCAC, and PP1A 5VGTCGACGGCGAGCCC, 5V-TCTTTGGGACCTTGTCTGCAA, and glyceraldehyde-3-phosphate dehydrogenase 5V-TGTTCGACAGTCAGCCGC and 5V-GGTGTCTGAGCGATGTGGC. The sequences of the primers for semiquantitative PCR are

489

490 Basyuk et al.

FIGURE 6. ZAC expression in skin tumors. Two SCCs (A and B), two sporadic BCCs (C and D), a BCC from a patient with Gorlin syndrome (E), and normal skin from the same patient (F) were analyzed for ZAC expression by in situ hybridization. Sections were counterstained with hematoxylin. Arrows, tumor nodules in BCC samples; arrowheads, tumor on SCC sections.

ZAC 5V-AACCGGAAAGACCACCTGAAAAACCAC and 5V-GTCGCACATCCTTCCGGGTGTAGAA (amplicon 303 bp), actin 5V-ACTGGCATCGTGATGGACTCC and 5V-GTTGGCGTACAGGTCTTTGCG (amplicon 444 bp), and BMP-2B 5V-CACCATGATTCCTGGTAACC and 5V-TCTCCAGATGTTCTTCGTGG (amplicon 390 bp). Semiquantitative PCR was done as follows: 2 minutes at 94jC followed by 20 to 30 cycles: 30 seconds at 94jC, 30 seconds at 55jC for BMP-2B primers, 57jC for ZAC primers, and 60jC for actin primers, and 30 seconds at 72jC. Lower numbers of cycles were used to verify linearity of the amplification process. Half of the PCR reaction was run on a 6% polyacrylamide gel and analyzed by Southern blotting.

nucleotide – long RNA probe, corresponding to nucleotides 1,889 to 2,182 of the human cDNA (Genbank accession no. AJ006354). The specificity of this probe was shown previously (56). For detection of cytokeratin 1, we used a

In situ Hybridization In situ hybridization on sections of normal skin, psoriasis, and skin tumors obtained after biopsy was done as described previously (56). For detection of ZAC, we used a 294-

FIGURE 7. ZAC expression is not repressed by recombinant SHH. Normal keratinocytes were grown in low-calcium medium. The calcium concentration was raised to 1.5 mmol/L for the indicated period of time in the presence (+) or absence ( ) of recombinant SHH. Total RNAs were isolated and analyzed by RT-PCR for ZAC, actin, and BMP-2B expression.



1-kb probe corresponding to the coding region of cytokeratin 1 cDNA, which was hydrolyzed before hybridization. Following ISH, sections of BCC and SCC were stained with hematoxylin. Immunocytochemistry Cells grown on coverslips were fixed with 4% formaldehyde in PBS for 30 minutes, washed with PBS thrice, and permeabilized with 0.1% Triton in PBS for 10 minutes. Double labeling was done with anti-ZAC rabbit polyclonal antibodies (diluted 1:1,000) and anti-involucrin mouse monoclonal antibodies SY5 (Interchim, Montluçon, France), diluted 1:100 in PBS containing 1% bovine serum albumin. Secondary antibodies anti-rabbit FITC (Sigma, San Quentin Fallavier, France) and anti-mouse CY3 (Sigma) were diluted 1:300 and 1:3,000, respectively, in PBS with 1% bovine serum albumin. No crossreactivity was observed in control experiments. The coverslips were mounted in 4V,6-diamidino-2-phenylindole – containing medium (56).

Acknowledgments We thank Drs. Edouard Bertrand, Jean-Marie Blanchard, and Annie Varrault for critical reading of the article and helpful comments.

References 1. Spengler D, Villalba M, Hoffmann A, et al. Regulation of apoptosis and cell cycle arrest by Zac1, a novel zinc finger protein expressed in the pituitary gland and the brain. EMBO J 1997;16:2814 – 25. 2. Hoffmann A, Ciani E, Boeckardt J, Holsboer F, Journot L, Spengler D. Transcriptional activities of the zinc finger protein Zac are differentially controlled by DNA binding. Mol Cell Biol 2003;23:988 – 1003. 3. Pagotto U, Arzberger T, Ciani E, et al. Inhibition of Zac1, a new gene differentially expressed in the anterior pituitary, increases cell proliferation. Endocrinology 1999;140:987 – 96. 4. Abdollahi A, Godwin AK, Miller PD, et al. Identification of a gene containing zinc-finger motifs based on lost expression in malignantly transformed rat ovarian surface epithelial cells. Cancer Res 1997;57:2029 – 34. 5. Abdollahi A, Bao R, Hamilton TC. LOT1 is a growth suppressor gene downregulated by the epidermal growth factor receptor ligands and encodes a nuclear zinc-finger protein. Oncogene 1999;18:6477 – 87. 6. Huang SM, Stallcup MR. Mouse Zac1, a transcriptional coactivator and repressor for nuclear receptors. Mol Cell Biol 2000;20:1855 – 67. 7. Huang S-M, Scho¨nthal AH, Stallcup MR. Enhancement of p53-dependent gene activation by the transcriptional coactivator Zac1. Oncogene 2001;20:2134 – 43. 8. Rozenfeld-Granot G, Krishnamurthy J, Kannan K, et al. A positive feedback mechanism in the transcriptional activation of Apaf-1 by p53 and the coactivator Zac-1. Oncogene 2002;21:1469 – 76.

expression of ZAC/LOT1 in squamous cell carcinomas of head and neck. Head Neck 2004;26:338 – 44. 16. Perou CM, Jeffrey SS, van de Rijn M, et al. Distinctive gene expression patterns in human mammary epithelial cells and breast cancers. Proc Natl Acad Sci U S A 1999;96:9212 – 7. 17. Singhal S, Amin KM, Kruklitis R, et al. Alterations in cell cycle genes in early stage lung adenocarcinoma identified by expression profiling. Cancer Biol Ther 2003;2:291 – 8. 18. Ross DT, Scherf U, Eisen MB, et al. Systematic variation in gene expression patterns in human cancer cell lines. Nat Genet 2000;24:227 – 35. 19. Gardner RJ, Mackay DJG, Mungall AJ, et al. An imprinted locus associated with transient neonatal diabetes mellitus. Hum Mol Genet 2000;9:589 – 96. 20. Kamiya M, Judson H, Okazaki Y, et al. The cell cycle control gene ZAC/ PLAGL1 is imprinted—a strong candidate gene for transient neonatal diabetes. Hum Mol Genet 2000;9:453 – 60. 21. Piras G, El Kharroubi A, Kozlov S, et al. Zac1 (Lot1), a potential tumor suppressor gene, and the gene for e-sarcoglycan are maternally imprinted genes: identification by a substractive screen of novel uniparental fibroblast lines. Mol Cell Biol 2000;20:3308 – 15. 22. Abdollahi A, Pisarcik D, Roberts D, Weinstein J, Cairns P, Hamilton TC. LOT1 (PLAGL1/ZAC1), the candidate tumor suppressor gene at chromosome 6q24-25, is epigenetically regulated in cancer. J Biol Chem 2003;278:6041 – 9. 23. Varrault A, Bilanges B, Mackay DJG, et al. Characterization of the methylation-sensitive promoter of the imprinted ZAC gene supports its role in transient neonatal diabetes mellitus. J Biol Chem 2001;276:18653 – 6. 24. Fuchs E, Byrne C. The epidermis: rising to the surface. Curr Opin Genet Dev 1994;4:725 – 36. 25. Watt FM, Jones PH. Expression and function of the keratinocyte integrins. Dev Suppl 1993;185 – 92. 26. Watt FM, Green H. Involucrin synthesis is correlated with cell size in human epidermal cultures. J Cell Biol 1981;90:738 – 42. 27. Hennings H, Michael D, Cheng C, Steinert P, Holbrook K, Yuspa SH. Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell 1980;19:245 – 54. 28. Wille JJJ, Pittelkow MR, Shipley GD, Scott RE. Integrated control of growth and differentiation of normal human prokeratinocytes cultured in serumfree medium: clonal analyses, growth kinetics, and cell cycle studies. J Cell Physiol 1984;121:31 – 44. 29. Murphy GF, Flynn TC, Rice RH, Pinkus GS. Involucrin expression in normal and neoplastic human skin: a marker for keratinocyte differentiation. J Invest Dermatol 1984;82:453 – 7. 30. Watt FM, Green H. Stratification and terminal differentiation of cultured epidermal cells. Nature 1982;295:434 – 6. 31. Boisseau AM, Donatien P, Surleve-Bazeille JE, et al. Production of epidermal sheets in a serum free culture system: a further appraisal of the role of extracellular calcium. J Dermatol Sci 1992;3:111 – 20. 32. Adams JC, Watt FM. Fibronectin inhibits the terminal differentiation of human keratinocytes. Nature 1989;340:307 – 9. 33. McKay IA, Leigh IM. Altered keratinocyte growth and differentiation in psoriasis. Clin Dermatol 1995;13:105 – 14. 34. Kim J, Shin DM. Biomarkers of squamous cell carcinoma of the head and neck. Histol Histopathol 1997;12:205 – 18.

9. Varrault A, Ciani E, Apiou F, et al. hZAC encodes a zinc finger protein with antiproliferative properties and maps to a chromosomal region frequently lost in cancer. Proc Natl Acad Sci U S A 1998;95:8835 – 40.

35. Katz MH. Nonmelanoma skin cancer. Md Med J 1997;46:239 – 42.

10. Bilanges B, Varrault A, Mazumdar A, et al. Alternative splicing of the imprinted candidate tumor suppressor gene ZAC regulates its antiproliferative and DNA binding activities. Oncogene 2001;20:1246 – 53.

37. Gailani MR, Stahle-Backdahl M, Leffell DJ, et al. The role of the human homologue of Drosophila patched in sporadic basal cell carcinomas. Nat Genet 1996;14:78 – 81.

11. Abdollahi A, Roberts D, Godwin AK, et al. Identification of a zinc-finger gene at 6q25: a chromosomal region implicated in development of many solid tumors. Oncogene 1997;14:1973 – 9.

38. Hahn H, Wicking C, Zaphiropoulous PG, et al. Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 1996;85:841 – 51.

12. Bilanges B, Varrault A, Basyuk E, et al. Loss of expression of the candidate tumor suppressor gene ZAC in breast cancer cell lines and primary tumors. Oncogene 1999;18:3979 – 88.

39. Grimwood RE, Ferris CF, Mercill DB, Huff JC. Proliferating cells of human basal cell carcinoma are located on the periphery of tumor nodules. J Invest Dermatol 1986;86:191 – 4.

13. Cvetkovic D, Pisarcik D, Lee C, Hamilton TC, Abdollahi A. Altered expression and loss of heterozygosity of the LOT1 gene in ovarian cancer. Gynecol Oncol 2004;95:449 – 55.

40. Oro AE, Higgins KM, Hu Z, Bonifas JM, Epstein EH, Scott MP. Basal cell carcinomas in mice overexpressing Sonic Hedgehog. Science 1997;276:817 – 21.

14. Pagotto U, Thomas A, Theodoropoulou M, et al. The expression of the antiproliferative gene ZAC is lost or highly reduced in non functioning pituitary adenomas. Cancer Res 2000;60:6794 – 8. 15. Koy S, Hauses M, Appelt H, Friedrich K, Schackert HK, Eckelt U. Loss of


36. Maloney ME. Histology of basal cell carcinoma. Clin Dermatol 1995;13: 545 – 9.

41. Fan H, Oro AE, Scott MP, Khavari PA. Induction of basal cell carcinoma features in transgenic human skin expressing Sonic Hedgehog. Nat Med 1997;3: 788 – 92. 42. Fan H, Khavari PA. Sonic Hedgehog opposes epithelial cell cycle arrest. J Cell Biol 1999;147:71 – 6.

491

492 Basyuk et al.

43. Bitgood MJ, McMahon AP. Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 1995; 172:126 – 38. 44. Hahn H, Wojnowski L, Miller G, Zimmer A. The patched signaling pathway in tumorigenesis and development: lessons from animal models. J Mol Med 1999; 77:459 – 68. 45. Ruiz i Altaba A. Gli proteins and Hedgehog signaling: development and cancer. Trends Genet 1999;15:418 – 25. 46. Dahmane N, Lee J, Robins P, Heller P, Ruiz i Altaba A. Activation of the transcription factor Gli1 and the Sonic Hedgehog signalling pathway in skin tumours. Nature 1997;389:876 – 81. 47. Nilsson M, Unden AB, Krause D, et al. Induction of basal cell carcinomas and trichoeptheliomas in mice overexpressing GLI-1. Proc Natl Acad Sci U S A 2000;97:3438 – 43. 48. Moles J-P, Moyret C, Guillot B, et al. p53 gene mutations in human epithelial skin cancers. Oncogene 1993;8:583 – 8. 49. Ouhtit A, Nakazawa H, Armstrong BK, et al. UV-radiation-specific p53 mutation frequency in normal skin as a predictor of risk of basal cell carcinoma. J Natl Cancer Inst 1998;90:523 – 31.

50. Dotto GP. p21WAF1/Cip1: more than a break to the cell cycle? Biochim Biophys Acta 2000;1471:M43 – 56. 51. Gandarillas A, Davies D, Blanchard JM. Normal and c-Mycpromoted human keratinocyte differentiation both occur via a novel cell cycle involving cellular growth and endoreplication. Oncogene 2000;19: 3278 – 89. 52. Zanet J, Pibre S, Jacquet C, Ramirez A, de Alboran IM, Gandarillas A. Endogenous Myc controls mammalian epidermal cell size, hyperproliferation, endoreplication and stem cell amplification. J Cell Sci 2005; 118:1693 – 704. 53. Larouche D, Hayward C, Cuffley KLG. Keratin 19 as a stem cell marker in vivo and in vitro . Methods Mol Biol 2005;289:103 – 10. 54. Rheinwald JG. Serial cultivation of normal human epidermal keratinocytes. Methods Cell Biol 1980;21A:229 – 54. 55. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of realtime quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002;3:RESEARCH0034. 56. Basyuk E, Bertrand E, Journot L. Alkaline fixation drastically improves signal of in situ hybridization. Nucleic Acids Res 2000;28:e46.
