In Drosophila melanogaster, the wings apart-like (wapl) gene encodes a protein that regulates heterochromatin structure. Here, we characterize a novel human homologue of wapl (termed human WAPL; hWAPL). The hWAPL mRNA was predominantly expressed in uterine cervical cancer, with weak expression in all other normal and tumor tissues examined. hWAPL expression in benign epithelia was confined to the basal cell layers, whereas in dysplasias it increasingly appeared in more superficial cell layers and showed a significant correlation with severity of dysplasia. Diffuse hWAPL expression was found in all invasive squamous cell carcinomas examined. In addition, NIH3T3 cells overexpressing hWAPL developed into tumors on injection into nude mice. Furthermore, repression of hWAPL expression by RNA interference induced cell death in SiHa cells. These results demonstrate that hWAPL is associated with cell growth, and the hWAPL expression may play a significant role in cervical carcinogenesis and tumor progression.
The wings apart-like (wapl) gene of Drosophila melanogaster encodes a protein that regulates heterochromatin structure (1) . Mutations of wapl prevent the normal close apposition of sister chromatids in heterochromatin regions but do not appear to affect either heterochromatin condensation or chromosomal segregation (1) . This evidence suggests that wapl is required to hold sister chromatids together in mitotic heterochromatin. wapl has also been implicated in both heterochromatin pairing during female meiosis and the modulation of position effect variegation (1) . In addition, a P element screen of Drosophila identified wapl as a modifier of chromosome inheritance (2) .
Among all varieties of cancer, uterine cervical cancer is unique because of its association with high-risk human papillomavirus (HPV) infection, with strains like HPV-16 and HPV-18. High-risk HPVs encode two oncoproteins, E6 and E7, which subvert crucial cellular regulatory mechanisms that reactivate and maintain DNA synthesis in the host cell. E6 accelerates proteosomal degradation of the p53 tumor suppressor, and E7 inactivates the retinoblastoma protein, interfering with the action of both p16INK4a (3) and the cyclin-dependent kinase inhibitor p21Cip1 (4 , 5) . Both the E6 and E7 high-risk HPV oncoproteins independently induce genomic instability in normal human cells (6 , 7) . Only a small portion of precursor lesions infected with HPV, however, develops into invasive carcinomas (8) . Therefore, additional genetic and microenvironmental factors subsequent to HPV infection are thought to play an important role in the initiation and progression of cervical neoplasia (8, 9, 10) .
In this study, we describe the isolation and characterization of a novel human wapl-related gene termed human WAPL (hWAPL). We have also demonstrated that hWAPL has the characteristics of an oncogene and is associated with uterine cervical cancer.
MATERIALS AND METHODS
cDNA Cloning and Construction of the hWAPL Expression Vector.
To isolate the complete hWAPL cDNA sequence, we used a human testis Marathon-Ready cDNA kit (Clontech, Palo Alto, CA).
To create an expression vector encoding hWAPL, a HindIII-EcoRI cDNA fragment containing the complete coding region of hWAPL was amplified by PCR using the primers 5′-TTAAGCTTTGAAACTGGTGTCAAAATGACATCCAGATT-3′ and 5′-TTGAATTCAAGCAATGTTCCAAATATTCAATCACTCTAGAG-3′ and inserted into the hemagglutinin (HA)-tagged mammalian expression vector, pHM6 (HA-hWAPL; Roche Diagnostics, Mannheim, Germany).
Northern Blot and Quantitative Real-Time PCR Analysis.
RNA isolation (11) and Northern blot analysis (11 , 12) were performed as described. The 674-bp DpnII fragment of hWAPL cDNA was used as a probe and labeled with 32P using the Rediprime II random prime labeling system (Amersham Biosciences, Piscataway, NJ). A human β-actin cDNA control probe (Clontech) was used as a control.
First-strand cDNA synthesis was performed as described (13) . Real-time PCR analysis was performed using the Smart Cycler System (Cepheid, Sunnyvale, CA) with SYBR Green I (Cambrex, Washington, DC). Real-time PCR used the hWAPL-specific primers 5′-GAATTCATAGGCACAGCGCTGAACTGTGTG-3′ and 5′-TTGAATTCCTAGCAATGTTCCAAATATTCA-3′ and β-actin-specific primers 5′-GGGAAATCGTGCGTGACATTAAG-3′ and 5′-TGTGTTGGCGTACAGGTCTTTG-3′. Reaction mixtures were denatured at 95°C for 30 s and then were subjected to 40 PCR cycles at 95°C for 3 s, 68°C for 30 s, and 87°C for 6 s. hWAPL mRNA levels were normalized to β-actin signals.
Immunohistochemistry and Immunoblot Analysis.
To generate mouse monoclonal antibodies against hWAPL, we immunized mice against a 6 × histidine-tagged hWAPL COOH terminus (amino acids 814-1037) fusion protein. Spleen cells of an immunized mouse were fused with P3UI mouse myeloma cells as described previously (14) . Of the 128 hybrids generated, one clone (clone R929) showed exclusive reactivity with hWAPL by ELISA. We used the supernatant of this clone as anti-hWAPL antibody.
Immunohistochemical assays were performed on formalin-fixed, paraffin-embedded sections using Ventana HX System Benchmark (Ventana Medical Systems Inc., Tucson, AZ). Immunohistochemical stains for hWAPL were interpreted semiquantitatively by assessing the intensity and extent of staining on the entire tissue sections present on the slides as described (9) .
Immunoblot analyses were performed as described previously (15) . The anti-HA (Roche Diagnostics; 3F10) and monoclonal anti-α-tubulin clone B-5-1-2 (Sigma Chemical Co., St. Louis, MO; T-5168) antibodies were purchased.
Animals and Treatment.
BALB/cAJc1-nu female mice (4 weeks old) were purchased from Charles River Japan, Inc. (Kanagawa, Japan).
The tumorigeneicity of the stable NIH3T3 transformants overexpressing hWAPL in vivo was examined as described previously (16) .
Cell Culture and small interfering RNA (siRNA) Transfection.
SiHa and NIH3T3 cells were grown in DMEM (Sigma) containing 10% fetal bovine serum at 37°C in a 5% CO2 environment. For the transfection of siRNA, we generated siRNAs using a Silencer siRNA Construction Kit (Ambion, Austin, TX). siRNA transfection was performed in DMEM without serum using Oligofectamine Reagent (Invitrogen Japan, Tokyo, Japan) and Opti-MEM I (Invitrogen Japan).
For cell quantitation, we harvested the cells from the wells of a 12-well plate and resuspended them in 100 μl of PBS. Trypan blue solution (100 μl, 0.4%; Sigma) was added to each sample, and viable cell numbers were quantitated using an erythrometer. The results shown are representative of three independent cell count analyses.
Molecular Cloning of hWAPL.
To isolate wapl-related genes from human cells, we searched DNA databases and identified a cDNA fragment, KIAA0261 (17) , and three expressed sequence tag clones, BE410177, BF79516, and BE257022, containing the KIAA0261 sequence. We also performed 5′ rapid amplification of cDNA ends. From these DNA sequences, we cloned and confirmed the full-length coding region sequence of the cDNA containing KIAA0261. We named this gene hWAPL (GenBank accession no. AB065003) to reflect its homology to wapl. The hWAPL gene product shows high sequence similarity in the WAPL-conserved region (amino acids 627-1169, 34% identical and 56% similar) and low similarity throughout the other regions to the wapl gene product. Several additional stretches of amino acids are also present in wapl protein (Fig. 1A) ⇓ .
High-Level Expression of hWAPL in Human Cervical Cancer.
As wapl is involved in sister chromatid cohesion, hWAPL may modify chromosomal inheritance. Deregulation of the expression of genes involved in chromosomal inheritance directly induces a variety of disorders associated with aneuploidy, including birth defects and cancer. Northern blot analysis detected hWAPL mRNA expression in several invasive cervical cancer samples, examined in tandem with additional human cancers and normal tissues (Fig. 1B) ⇓ . We confirmed the hWAPL expression in cervical cancers by quantitative real-time PCR analysis of tumor and normal tissue samples. The levels of hWAPL mRNA expression in cervical cancers were significantly higher than the levels observed in either normal cervical controls or endometrial, ovarian, breast, lung, stomach, renal, and colon cancers (Fig. 1C) ⇓ .
To investigate the connection between hWAPL expression and oncogenesis in cervical malignancies, we examined the expression of hWAPL by immunohistochemistry in a series of clinical samples of the various grades of cervical dysplasia [cervical intraepithelial neoplasia (CIN) I–III] and invasive squamous cell carcinoma. We found nuclear immunostaining for hWAPL in all samples (Fig. 2A) ⇓ . hWAPL expression in benign squamous epithelia was confined to the basal and parabasal cell layers. In contrast, hWAPL expression in squamous dysplasia and invasive carcinoma increasingly appeared in the more superficial cell layers and was significantly increased compared with the adjacent benign epithelia (P = 0.0002 for CIN I, P = 0.0003 for CIN II, P = 0.0001 for CIN III, and P = 0.0001 for invasive squamous cell carcinoma; Wilcoxon’s signed rank test). CIN I and II cases showed hWAPL expression in the basal 50 and 70% of the epithelial thickness, respectively, whereas CIN III and invasive squamous cell carcinoma showed hWAPL expression in the full thickness of the dysplastic epithelia (Fig. 2A) ⇓ . Furthermore, the mean hWAPL staining score increased remarkably with increasing grade of dysplasia (Fig. 2B) ⇓ . These data strongly suggest that the unscheduled high-level expression of hWAPL may play a significant role in cervical carcinogenesis and tumor progression.
hWAPL Has Oncogenic Characteristics.
Because we observed high-level expression of hWAPL in tumors, we sought to determine whether hWAPL overexpression promotes tumor development. We transfected NIH3T3 cells with an HA-tagged hWAPL expression vector (HA-hWAPL 3T3) or HA expression vector (HA-3T3). Then, we compared the ability of HA-hWAPL 3T3 with HA-3T3 cells to grow as tumors in nude mice. We injected 106 cells into three s.c. sites of each nude mouse. HA-hWAPL 3T3 cells produced tumors in all nude mice within 10 days after injection of cells (100%, n = 18; Fig. 3A ⇓ ). HA-3T3 failed to produce tumors in any mice (0%, n = 18). We confirmed high hWAPL expression levels in the resultant tumors by Western blot analysis (Fig. 3B) ⇓ . These results suggest that hWAPL has the characteristics of an oncogene.
Repression of hWAPL Expression Induces Cell Death.
We examined hWAPL function by suppressing hWAPL expression. Initial attempts to generate a WAPL-deficient mouse demonstrated that the loss of WAPL was embryonic lethal (data not shown). Therefore, we designed two 21-nucleotide, double-stranded siRNAs, siRNA(I) and siRNA(II), to repress hWAPL expression (Refs. 18 and 19 ; Figs. 1A ⇓ and 4A ⇓ ). We examined various human cancer-derived cell lines and found that cervical cancer-derived cell lines containing both HPV-positive and -negative cells exhibited higher levels of hWAPL expression compared with the other cell lines (data not shown). Then, we examined the effects of suppressing hWAPL in a cervical cancer-derived cell line, SiHa. siRNA transfection at a concentration of either 1 nm siRNA(I) or siRNA(II) reduced hWAPL mRNA levels (Fig. 4B) ⇓ . siRNA(I) was more effective at reducing hWAPL mRNA than siRNA(II). Thus, we used siRNA(I) in the subsequent experiments. hWAPL protein levels were also significantly reduced after siRNA(I) transfection (Fig. 4C) ⇓ . Interestingly, siRNA(I) repressed the growth of the cells and subsequently induced cell death (Fig. 4, D and E) ⇓ . siRNA(II) repressed cell growth in a similar manner as siRNA(I) (Fig. 4D) ⇓ , suggesting that the effects of these siRNAs on proliferation and viability are likely caused by the repression of hWAPL expression. Similar results were obtained in another cervical cancer-derived cell line, CaSki, with 10 nm siRNA(I) (data not shown). On the contrary, we did not observe any effects of siRNA(I) on cells expressing relatively low levels of hWAPL, such as Saos-2 and HCT116 (data not shown).
To investigate the fate of cells transfected with siRNA(I), we analyzed siRNA-transfected cells by flow cytometry (Fig. 5) ⇓ . In siRNA(I)-transfected cells, the population of cells exhibiting S phase DNA content increased (Fig. 5 ⇓ ; 48 and 72 h). In addition, there was an increase in the number of apoptotic cells exhibiting subG1 DNA content (Fig. 5 ⇓ ; 72 h). Many cells showing S phase DNA content may also be apoptotic cells at G2-M phase. Taken together, these results suggest that a malfunction in the hWAPL pathway activates an S phase checkpoint or another apoptotic pathway and consequently leads to cell death.
In this study, we report the isolation and characterization of a novel human gene termed hWAPL. We were unable to identify additional genes similar to wapl within the human genome sequence database. Thus, although the high-sequence conservation between hWAPL and wapl is limited to a third of the protein sequence encoded by wapl (Fig. 1A) ⇓ , we consider hWAPL to be the human homologue of wapl. We did not find any protein sequence motifs in hWAPL, except for the WAPL-conserved region (Fig. 1A) ⇓ . We therefore expect that hWAPL has similar functions to the wapl protein. Two hybridization signals for hWAPL were visible by Northern blot analysis (Fig. 1B) ⇓ . Western blot analysis, however, detected only a single band for hWAPL (Fig. 2C) ⇓ . In addition, we did not obtain additional nucleotide sequences similar to the open reading frame of hWAPL by PCR analysis with various PCR primers (data not shown). Thus, we consider that the two hybridization signals may reflect the difference of the length of the untranslated regions of the hWAPL mRNA.
High-level expression of hWAPL was observed in cervical cancers (Fig. 1, B and C) ⇓ . Furthermore, hWAPL-overexpressing 3T3 cells developed into tumors on injection into nude mice (Fig. 3) ⇓ . These results suggest that hWAPL has oncogenic characteristics. Cervical cancer is a serious health problem, with ∼500,000 women developing the disease each year worldwide. In many developing countries, it is the most common cause of cancer death and years of life lost because of cancer (20) . Although the fundamental role of high-risk HPV infection in the pathogenesis of cervical carcinoma is well established, other factors are thought to play a role in cervical carcinogenesis (8 , 21) . Because all of uterine cervical samples examined were HPV positive (data not shown), it is still to be confirmed whether hWAPL expression is inducible by HPV infection. However, HPV-positive normal cervical tissue samples exhibited low hWAPL expression (Fig. 1, B and C ⇓ and data not shown), and an HPV-negative, uterine cervical cancer-derived cell line, C33A, showed high hWAPL expression (data not shown). Thus, hWAPL expression is likely to be more closely related with cervical carcinogenesis than HPV infection. Recently, Acs et al. (9) found significant correlation among expression of Epo receptor, p16INK4a, and bcl-2 in benign and dysplastic squamous epithelia. In our results, hWAPL showed similar expression pattern to Epo receptor and p16INK4a in benign and dysplastic cervical squamous epithelia and invasive squamous cell carcinomas (Fig. 2, A and B) ⇓ . Although we did not find any evidence for hWAPL being involved in hypoxia-inducible Epo signaling, hWAPL may cooperate with the Epo signaling in the progression of cervical neoplasia. These observations indicate that hWAPL overexpression can be used as a useful diagnostic tool in the detection of cervical dysplasia like p16INK4a (22) and Epo receptor (9) . In addition, our results provide the necessity to investigate the potential of hWAPL as a cancer therapeutic target.
Loss of WAPL was embryonic lethal in mouse (data not shown), and repression of hWAPL expression in SiHa cells led to cell death (Fig. 4) ⇓ . Flow cytometry analysis demonstrated that malfunction of hWAPL may cause apoptosis and/or arrest of cells at S phase (Fig. 5) ⇓ . In addition, Drosophila wapl is associated with regulation of chromatin organization (1) . Thus, we expect that hWAPL is also associated with regulation of chromatin structure, and deregulation of hWAPL expression may induce chromosomal instability. Although additional investigations are necessary to elucidate the actual function of hWAPL in normal and malignant cells, our results have demonstrated that the novel oncogene, hWAPL, is one of the essential genes for development and cell growth and may play a significant role for cervical carcionogenesis and tumor progression.
We thank K. Yoshida, R. Iwata, R. Tsujimoto, K. Kitamura, M. Sugiura, and M. Takaoka for their technical assistance.
Grant support: Grant-in-Aid for Scientific Research on Priority Area (C) and Grant-in-Aid for Encouragement of Young Scientists from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and a grant from Core Research for Evolutional Science and Technology, Japan Science and Technology Corp.
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.
Note: T. Ohbayashi is currently at the Horizontal Medical Research Organization, Kyoto University Faculty of Medicine, Kyoto, Japan.
Requests for reprints: Masahiko Kuroda, Department of Pathology, Tokyo Medical University, 6-1-1, Shinjuku, Shinjuku-ku, Tokyo, 160-8402, Japan. Fax: 81-3-3352-6335; E-mail:
- Received December 8, 2003.
- Revision received March 15, 2004.
- Accepted March 16, 2004.
- ©2004 American Association for Cancer Research.