Infection of the gastrointestinal tract by the human polyomavirus, JCV, which has been frequently detected in raw urban sewage, can occur via intake of contaminated water and food. In light of earlier reports on the tumorigenecity of JCV, we investigated the presence of the JCV genome and the expression of viral proteins in a collection of 27 well-characterized epithelial malignant tumors of the large intestine. Results from gene amplification revealed the presence of the viral early genome in 22 of 27 samples. Expression of the viral oncogenic early protein, T-antigen, and the late auxiliary protein, Agnoprotein, was observed in >50% of the samples. The absence of the viral capsid protein in the tumor cells excludes productive replication of the virus in neoplastic cells. Laser capture microdissection confirmed the presence of the JCV genome and expression of T-antigen in precancerous villous adenomas and regions of invasive adenocarcinoma. The ability of JCV T-antigen to interact with β-catenin and the nuclear detection of β-catenin in T-antigen-positive cells suggests dysregulation of the Wnt pathway in the tumor cells. The coproduction of T-antigen and β-catenin in colon cancer cells enhanced transcription of the c-myc promoter, the downstream target of β-catenin. These observations provide evidence for a possible association of JCV with colon cancer and suggest a novel regulatory role for T-antigen in the deregulation of the Wnt signaling pathway through β-catenin in tumors of the gastrointestinal tract.

Results from serological studies have indicated an asymptomatic infection of ∼90% of the adult population by age 15 with the human polyomavirus, JCV (1). It is believed that childhood exposure to JCV, most likely through the upper respiratory tract, causes persistent infection of kidney epithelial cells where the virus remains in a latent state during the life of the infected individual. In patients with impaired cell-mediated immunity, JCV enters the brain and its productive replication in glial cells causes the fatal demyelinating disease of the central nervous system, Progressive Multifocal Leukoencephalopathy (for review see Ref. 2). Moreover, recent studies have established an association of JCV with a broad range of human brain tumors, most notably medulloblastomas (3, 4, 5). JCV can transform cells that are manifested by distinct morphological changes, rapid division, prolonged life spans, and the ability to form dense foci in culture (6, 7, 8). The oncogenic potential of this virus has also been established in several experimental animals (for review see Ref. 9). Although the precise mechanism responsible for JCV-induced cellular transformation and tumor formation is not fully understood, it is believed that the viral early protein, T-antigen, through interaction with several cell regulatory proteins, including tumor suppressors and cell cycle regulators, promotes uncontrolled progression of cells through the cell cycle (10, 11). Furthermore, recent studies have revealed the potential of JCV T-antigen to modulate several signaling pathways, including insulin-like growth factor I and Wnt in cells derived from JCV-induced mouse medulloblastoma (12, 13).

JCV is detected in the urine of 20–80% of adults (14). Results from genotyping of JCV excreted by a diverse group of humans has revealed that infection with JCV is influenced by the geographical origin, ethnic group, and age (14, 15). Earlier surveys of raw sewage from urban areas have shown the detection of JC viral particles in sewage samples from widely divergent areas (16, 17, 18), suggesting a potential reentry of JCV and/or viral DNA into the human population through the intake of virus-contaminated water and food. In support of this notion, earlier studies have revealed the presence of JCV DNA sequences in the upper and lower human gastrointestinal tract (19, 20). Considering the tumorigenicity of JCV and its potential intake through the intestine, we performed a comprehensive study on a collection of well-characterized malignant epithelial tumors of the colon for the presence of JCV DNA sequence and expression of the viral early and late proteins. Specimens were obtained from Thomas Jefferson University Hospital (Philadelphia, PA) and the American British Cowdray Medical Center I. A. P. (México D.F., México).

A total of 27 cases of adenocarcinoma in patients between 41 and 91 years of age was selected for the initial evaluation of tumor location and expression of various marker proteins. Seven tumors were obtained from the cecum, 2 from the ascendant colon, 3 from the transverse region, 4 from the descendent colon, and 11 from the sigmoid (Table 1). All tumors were positive for EMA3 and all produced cytokeratin. None of the samples expressed desmin, whereas most showed various levels of CEA (Table 1). Histological assessment of the tumors as compared with a normal area of the colonic mucosa, which is characterized by interdigitated columnar and goblet cells, revealed papillary structures lined with atypical columnar cells characteristic of villous adenoma in seven of the samples, as well as isles of in situ and invasive neoplastic epithelial cells. The invasive glandular structures contained pleomorphic cells with atypical nuclei and frequent mitotic figures. Necrotic areas within the tumors were frequently observed. Immunohistochemistry against EMA showed positive reactivity in columnar cells of normal tissue as well as in preneoplastic and neoplastic cells. Furthermore, tumor cells showed strong immunoreactivity with antibodies recognizing cytokeratin and CEA (data not shown).

For the detection of the JCV gene sequences, total DNA was extracted from paraffin-embedded tissues and evaluated by PCR techniques using three pairs of primers derived from the viral early genome, T-antigen, the viral late auxiliary gene encoding Agnoprotein, and the viral late capsid gene VP1, as detailed in Fig. 1,A. The amplified DNAs were analyzed by Southern blot hybridization using oligonucleotide DNA probes specific for the amplified JCV sequences. Results from several experiments revealed that 22 of the samples (81.5%) contained the early region of the JCV genome, whereas 16 samples (59.2%) contained DNA sequences corresponding to Agnogene. The region of the JCV genome spanning VP1 was detected in four samples (14.8%). Fig. 1 illustrates a representative Southern blot, and Table 2 summarizes the results from PCR amplification/Southern blotting. Detection of JCV DNA sequences provided a rationale to investigate the expression of the viral T-antigen as well as Agnoprotein and VP1 in these samples. Results from immunohistochemistry revealed the expression of T-antigen in 17 (62.9%) samples (Table 2). Although no immunoreactivity with anti-VP1 antibody was observed in either tumor or nontumor regions of the samples, 12 specimens (44.4%) showed positive staining with anti-Agnoprotein antibody (Table 2). Fig. 2 illustrates results from immunohistochemical labeling in which T-antigen was detected in an area of a tumoral emboli located in the submucosal lymphatic vessels and absent in the suprayacent normal colonic mucosa (Fig. 2,A). In the preneoplastic villous adenoma, the columnar cells show nuclear immunoreactivity for T-antigen (Fig. 2,B). In the invasive tumor, some of the epithelial glandular neoplastic cells exhibited nuclear accumulation of JCV T-antigen (Fig. 2,C). Cytoplasmic perinuclear accumulation of Agnoprotein was evident in a polyp as well as in neoplastic cells (Fig. 2, D and E, respectively). No evidence of VP1 expression could be observed in various regions of the samples (Fig. 2, F and G).

Next, the LCM system was used to extract DNA from specific areas of paraffin-embedded sections, which expressed positive immunoreactivity with anti-T-antigen antibody. Three regions of the specimens were selected representing normal mucosa, villous adenoma, and the invasive adenocarcinoma (Fig. 3,A). Results from gene amplification revealed detection of sequences corresponding to the JCV early genome in DNA obtained from villous adenomas and invasive adenocarcinoma (Fig. 3 B). A very weak signal corresponding to JCV sequences was observed upon amplification of DNA from a normal appearing region in juxtaposition with a precancerous areas, suggesting the presence of few JCV-positive cells in this region.

Although the importance of the presence of JCV DNA in the genesis of human epithelial malignant tumors of the colon remains unclear, the detection of viral proteins, including T-antigen, in the tumor cells indicate the potential involvement of JCV in pathways leading to the development and/or progression of cancer. Examination of p53, a cell regulatory protein that loses its activity and becomes more stable upon mutation or association with viral oncoproteins, including JCV T-antigen, in this collection of colon cancers revealed the detection of p53 in 19 samples (70.3%) of which 13 were also positive for T-antigen (Table 2). Fig. 4 represents immunohistochemistry of a T-antigen-positive sample in which nuclear p53 expression was observed in neoplastic glandular cells inside lymph vessels underlying a normal epithelium area (Fig. 4,A). The columnar cells lining the villous polyps and the neoplastic cells in the glandular structures were similar to the pattern of T-antigen expression and showed immunoreactivity to p53 (Fig. 4, B and C, respectively). Results from double immunofluorescence labeling of sections with anti-p53 and anti-T-antigen showed abundant columnar cells in a villous adenoma with nuclear localization of both p53 and T-antigen (Fig. 4, D–F). Colocalization of T-antigen and p53 suggests a possible interaction of these two proteins that can lead to uncontrolled proliferation of the tumor cells.

In addition to p53, T-antigen may affect other regulatory events that are implicated in cancer development. Earlier studies have demonstrated that β-catenin, a key component of Wnt signaling pathway, is affected in colon cancer. Mutations in β-catenin that prolong its stability in the cytoplasm permits the association of β-catenin with TCF/LEF transcription factors (for review see Ref. 21) and facilitates the nuclear import of β-catenin where it activates transcription of a series of cellular genes that are involved in rapid cell proliferation (22). It has been shown that mutations in the phosphorylation residues of β-catenin within exon 3 stabilize β-catenin by preventing its proteasome degradation (for review see Ref. 23). Evaluation of the β-catenin gene in T-antigen-positive colon specimens showed no mutations in the region of the protein-spanning exon 3 (unpublished data). However, results from immunohistochemistry revealed nuclear detection of β-catenin in five cases that showed strong positive nuclear immune reaction for T-antigen (Table 2, also see Fig. 5,A). Similarly, cytoplasmic and nuclear labeling of TCF-4 was evident in the neoplastic cells (Table 2 and Fig. 5,B), whereas TCF-1 and LEF-1 were mostly detected in the cytoplasm (LEF-1 staining is depicted in Fig. 5,C). These observations suggest that the Wnt signaling pathway may be affected by the expression of the JCV genome in these tumor cells. Interestingly, results from double labeling of the tumor cells with anti-T-antigen and anti-β-catenin antibodies showed colocalization of T-antigen and β-catenin in the nuclei of neoplastic columnar cells (Fig. 5, compare D and E to F).

To investigate the possible interaction of JCV T-antigen with β-catenin, protein extract from a colon cancer cell line, HCT116, transfected with a plasmid-expressing T-antigen was immunoprecipitated with anti-T-antigen antibody, and the immunocomplex was analyzed by Western blot with anti-β-catenin antibody. As shown in Fig. 6,A, a band corresponding to β-catenin was present in the immunocomplex pulled down by anti-T-antigen antibody but not with control preimmune sera. Accordingly, the level of c-myc gene expression, the downstream target for β-catenin, was enhanced in the cells (Fig. 6,B). No changes in the level of housekeeping protein, Grb-2, was detected. To assess the functional importance of T-antigen and β-catenin’s presence in the cells, transfection of HTC116 cells was carried out in the presence of plasmids containing the c-myc promoter driving the reporter luciferase gene. According to the results illustrated in Fig. 6,C, the expression of β-catenin had no stimulatory effect upon the c-myc promoter. However, in the presence of T-antigen, β-catenin significantly enhanced transcription of the c-myc promoter. T-antigen alone caused a modest increase in the activity of c-myc promoter, suggesting its cooperation with endogenous β-catenin in these cells. To investigate the subcellular localization of β-catenin in the absence and presence of JCV T-antigen, HCT116 cells were transfected with a plasmid-expressing T-antigen. As shown in Fig. 6 D, among the group of cells, β-catenin was colocalized in the nucleus of the cells that also expresses T-antigen. In the other cells where T-antigen is not present, β-catenin was detected in the cytoplasm. Taken together, these observations suggest that the association of T-antigen with β-catenin may translocate this protein to the nucleus.

Clinical Samples.

A total of 27 paraffin-embedded tumors of the colon was collected from the pathology archives of the following institutions: 17 samples were obtained from the American British Cowdray Hospital in Mexico City, Mexico, and 10 samples were obtained from Thomas Jefferson University in Philadelphia, Pennsylvania. The tumors were histologically graded and immunohistochemically characterized according to the WHO Classification of Tumors of the Gastrointestinal Tract (24).

DNA Extraction and Analysis.

A dedicated microtome was used to section the paraffin blocks. Furthermore, the block and blade holder were periodically autoclaved, and a new, disposable blade was used for each specimen. The sections were handled with a disposable, one-time use applicator to prevent contamination. DNA was extracted from ∼10 sections of 10 μm in thickness from each of the tissue samples by using the QIAamp Tissue Kit, according to the manufacturer’s instructions (Qiagen, Valencia, CA).

PCR amplification was performed by using three individual sets of primers: PEP1 and PEP2 (nucleotides 4255–4272 and 4408–4427, respectively), which amplify sequences in the NH2-terminal region of JCV T-antigen, VP2 and VP3 (nucleotides 1828–1848 and 2019–2039, respectively), which amplify a portion of the VP1 capsid gene sequence, and AGNO1 and AGNO2 (nucleotides 280–298 and 438–458, respectively), which amplify a region within coding region of JCV Agnoprotein. Amplification was carried out on 500 ng of template DNA with AmpliTaq DNA Polymerase (Perkin-Elmer) in a total volume of 50 μl. PCR was performed in the presence of 2.5 mm MgCl2 and 0.5 mm of each primer (Oligos, Etc., Guilford, CT). A Perkin-Elmer Gene Amp 9700 PCR System dedicated and used in this study was run using 9600 ramping conditions with denaturation at 95°C for 10 min, followed by 45 cycles of denaturation at 95°C for 15 s, annealing for 30 s, and extension at 72°C for 30 s. The annealing steps were performed at temperatures of 55°C for the PEP primers, 57°C for the Agno primers, and 54°C for the VP primers. As a termination step, the extension time of the last cycle was increased to 7 min. Samples amplified in the absence of template DNA served as a negative control, whereas inclusion of serial dilutions of the plasmid, pBJC, containing the JCV genome as template served as a positive control for each procedure.

Southern blot was performed by resolving 10 μl of each PCR reaction on 2% agarose gel electrophoresis. The gels were then treated for 15 min each with 0.2 m HCl for depurination, 1.5 m NaCl/0.5 m NaOH for denaturation, and 1.5 m NaCl/0.5 m Tris-HCl (pH 7.4) for neutralization, followed by transfer of the amplified fragments from the gel to nylon membranes (Hybond-N; Amersham). The membranes were then prehybridized for 1 h in Ultrahyb solution (Amersham), followed by hybridization in the same solution containing 5 × 106 cpm/ml [γ32P]ATP end-labeled oligonucleotide specific for JCV. Probes used for Southern blotting included JCV probe (nucleotides 4303–4327) to detect fragments amplified with PEP primers, VP probe (nucleotides 1872–1891) for those amplified with VP primers, and Agno probe (nucleotides 425–445) for sequences amplified with the Agno 1 and 2 primers as shown in Fig. 2. Membranes were hybridized overnight, washed, and autoradiographed as described previously (5).

Histological and Immunohistochemical Analysis.

Paraffin-embedded tissue previously fixed in 10% buffered formalin was sectioned at 4-μm thickness and mounted onto charged slides. Sections were placed in an oven at 65°C to melt the paraffin and then deparaffinized in three changes of xylene for 30 min each. Sections were then rehydrated through a graded series of alcohols up to water, and nonenzymatic antigen retrieval was performed in 0.01 m sodium citrate (pH 6.0) at 97°C in a vacuum oven for 35 min. After a cooling period of 25 min, sections were rinsed with PBS, and endogenous peroxidase was quenched by incubating the slides in methanol/3% H2O2 for 30 min at room temperature.

Sections were blocked in 2% horse serum and incubated overnight with primary antibodies at room temperature in a humidified chamber. Antibodies to characterize the epithelial nature of the tumors included a mouse monoclonal anti-EMA, (clone E29, 1:50 dilution; Dako), a rabbit polyclonal antibody against cytokeratin (wide spectrum, 1:2000 dilution; Dako), a mouse monoclonal anti-desmin to exclude any sarcomas (clone D33, 1:100 dilution; Dako), and a mouse monoclonal antihuman CEA (clone II-7, 1:100 dilution; Dako). Antibodies used to detect viral proteins included a mouse monoclonal antibody against SV40 large T-antigen that cross-reacts with JCV T-antigen (clone pAb416, 1:100 dilution; Oncogene Science), a rabbit polyclonal antibody against Agnoprotein (4), and a mouse monoclonal antibody against the JCV capsid protein VP-1 (1:1000 dilution; a kind gift from Dr. Walter Atwood, Brown University, Providence, RI). The antioncogenic gene product p53 was detected by using a mouse monoclonal antibody that recognizes both mutant and wild-type p53 (clone DO-7, 1:100 dilution; Dako). Proteins of the Wnt pathway were analyzed by using the following antibodies: a mouse monoclonal antihuman adenomatous polyposis coli (clone F-3, 1:500 dilution; Dako); a mouse monoclonal anti-β-catenin antibody (clone E-5, 1:100 dilution; Santa Cruz Biotechnology); a goat polyclonal antibody against TCF-1 (1:200 dilution; Santa Cruz Biotechnology); a goat polyclonal antibody against TCF-4 (1:250 dilution; Santa Cruz Biotechnology); and a goat polyclonal antibody for LEF-1 (1:200 dilution; Santa Cruz Biotechnology).

After rinsing the sections in PBS, the slides were incubated for 1 h at room temperature with biotinylated antimouse or antirabbit secondary antibodies and then were rinsed in PBS. The tissue was subsequently incubated with avidin-biotin-peroxidase complexes for 1 h at room temperature according to the manufacturer’s instructions (Vector Laboratories), and finally, the sections were developed with a diaminobenzidine substrate (Sigma, St. Louis, MO), counterstained with hematoxylin, and coverslipped with Permount (Fisher, Pittsburgh, PA). To assess the fraction of immunolabeled cells in specimens from each patient case, the labeling index defined as the percentage of positive cells of the total number of tumor cells counted was determined.

Double-labeling Immunofluorescence.

For immunofluorescence and double labeling of paraffin embedded sections, deparaffinization, antigen retrieval, endogenous peroxidase quenching and blocking were performed as described above. For immunofluorescent double labeling of HCT116 cells in culture, the cells were washed with PBS, fixed in cold acetone, and blocked in PBS containing 5% horse serum and 0.1% BSA for 2 h. Sections were then incubated with mouse anti-T-antigen antibody (Oncogene Science, clone pAb416, 1:100 dilution) for 16 h followed by washing in PBS and incubation in antimouse rhodamine antibody (1:200 dilution; Vector Laboratories). Next, sections were incubated with mouse monoclonal antibody, which recognizes either β-catenin (clone E-5, 1:100 dilution; Santa Cruz Biotechnology) or p53 (Dako, clone DO-7, 1:100 dilution) for 16 h followed by washing in PBS and incubation in antimouse fluorescein antibody (1:200 dilution; Vector Laboratories). Finally, sections were washed in PBS and mounted in aqueous mounting media (Vector Laboratories).

LCM.

Representative sections of colon cancer tissue were selected and formalin-fixed, paraffin-embedded sections were cut at 5 μm and mounted on glass slides. Immunohistochemistry against the JCV T-antigen protein was performed to identify and selectively dissect T-antigen-positive cell populations. Sections were subsequently dehydrated in graded ethanol solutions (95% ethanol, 2 × 5 min, 100% 3 × 5 min) and cleared in xylene (3 × 5 min). After air-drying for 30 min, laser capture was performed under direct microscopic visualization of the T-antigen-positive immunolabeled areas by laser activation of thermoplastic film mounted on optically transparent LCM caps (Arcturus Engineering, Mountain View, CA). The PixCell II LCM System (Arcturus Engineering) was set to the following parameters: 15-μm laser spot size; 40-mW power; and 3.0-ms duration. Cancer cells were captured by focal melting of the membrane through a carbon dioxide laser pulse activation. Normal epithelial crypts, villous adenomas, and invasive adenocarcinoma components were individually dissected for all cases.

Transient Transfection 15, Protein Extraction, and Analysis.

HCT116 cells were transfected with 5 μg of pCMV-T-antigen by the calcium phosphate precipitation method. Total protein extracts were prepared after 48 h according to procedures described previously (12). Protein extract (250 μg) was incubated with anti-T-antigen antibody or preimmune sera, and the immunocomplexes were separated by SDS-PAGE, transferred to nitrocellulose, and analyzed by Western blot using anti-β-catenin antibody. For direct Western blot, 50 μg of protein extract were analyzed by SDS-PAGE and transferred to nitrocellulose filter incubated with anti-c-myc or anti-Grb-2 antibodies. For luciferase assay, protein extracts were prepared from HCT116 cells transfected with 3 μg pDel-1 (kindly provided by Dr. B. Vogelstein, Johns Hopkins Hospital, Baltimore, MD) reporter plasmid alone or together with pC52MT-WTβ-cat (kindly provided by Dr. F. Fagotto, Max Placnk Institute, Tubingen, Germany) and pCMV-T-antigen by standard methods (25) and assayed for luciferase activity as described previously (12).

Once believed to be a highly neurotropic virus whose genes are preferentially expressed in astrocytes and oligodendrocytes of human brain, recent studies have provided evidence for the presence of the JCV genome in a broad range of human cell types and tissues. For example, the JCV genome has been detected in tonsillar stromal cells (26), B lymphoid cells (27), kidney epithelial cells (28), and upper and lower parts of the gastrointestinal tract, including the mucosa of the colon (19, 20, 29). Moreover, a segment of JCV DNA has been detected in a broad range of tumors of glial and nonglial origin, including gliomas, ependymomas, and medulloblastomas (3). Thus far, expression of the viral proteins, including T-antigen and Agnoprotein, has been shown in only tumors originating from the central nervous system in the absence of productive viral infection.

Results from the studies presented here reveal the expression of JCV proteins in a series of well-defined nonneural origin clinical specimens of the lower gastrointestinal tract and provide some clues as to the molecular events that are affected by JCV T-antigen in tumor cells. According to the proposed genetic model for colorectal tumorigenesis (30), accumulation of a series of genetic variations resulting in the activation of an oncoprotein and inactivation of tumor suppressors proteins orchestrate the development of tumors at various stages in the colon. Whereas some of these alterations may be inherited such as the mutation in chromosome 5, which is seen in familial adenomatous polyposis (31), others such as viral infection with the ability to alter many regulatory events can be environmentally acquired. The T-antigen of polyomaviruses, including JCV, has an established oncogenic capability likely because of its ability to interact with and inactivate several tumor suppressor proteins such as p53 and pRb. These events can lead to suppression of p21WAF-1, the downstream regulator of p53 and the liberation of the E2F family of transcription factors from pRb. Consequently, p21WAF-1 and E2F stimulate several proteins, the function of which is essential for cell cycle progression and rapid cell proliferation. Detection of p53 and its colocalization with T-antigen in the tumor cells suggests that p53 is functionally inactive. This notion is supported by results from immunohistochemistry where p21WAF-1 was not detected in the tumors cells (data not shown).

β-catenin is an integral component of the Wnt signaling pathway whose nuclear import contributes to colorectal carcinogenesis (21). Our results demonstrate that in T-antigen-positive tumor cells, β-catenin is found in the nuclei where T-antigen is present. Furthermore, we found that T-antigen and wild-type β-catenin can form a complex leading to speculate that the association of these two proteins results in the nuclear entry of β-catenin in neoplastic cells. Although the functional consequence of this interaction in cancer development remains to be investigated, results from cotransfection studies show cooperativity between β-catenin and T-antigen in inducing transcription from the c-myc promoter, a known downstream target gene of β-catenin. This observation is in agreement with earlier reports showing the involvement of the Wnt signaling pathway and up-regulation of c-myc in colon cancer (for review see Refs. 23, 32). Taken together, the results presented in this manuscript extend earlier reports on the presence of JCV DNA sequences in the human upper and lower gastrointestinal tract and colorectal cancers (19, 20) and, for the first time, demonstrate the expression of the viral oncoprotein, T-antigen, and late Agnoprotein in tumor cells containing viral DNA sequences. The presence of the JCV genome and, more importantly, expression of its proteins in tumor cells suggests a role, perhaps as a cofactor, in the development of tumors of the gastrointestinal tract. These observations should invite additional investigation of the association of this oncogenic virus with other human tumors, the development of animal models for human cancers using JCV as a tool, and the investigation of pathways that may be differentially deregulated by JCV in various tumors.

Fig. 1.

Structural organization of the JCV genome and detection of JCV DNA in colon cancer. The numbers within the inner circle show the map position with 0.0 being the EcoRI site (33). The thick solid arrow on the left depicts the viral early protein T-antigen, whereas the shaded arrows on the right point to the positions of the viral late proteins, Agnoprotein, and the capsid protein, VP1. The positions of the PCR primers are shown by thin arrows outside of the circle. The size of the amplified DNAs and the location of the DNA probes used for Southern blot hybridization specific for the amplified T-antigen, Agnoprotein, and VP1 sequences are shown. Numbers over the hybridization blots indicate the case number on Table 2. −, negative control; +, positive control; M, molecular weight standards. Size of the different JCV PCR product is indicated.

Fig. 1.

Structural organization of the JCV genome and detection of JCV DNA in colon cancer. The numbers within the inner circle show the map position with 0.0 being the EcoRI site (33). The thick solid arrow on the left depicts the viral early protein T-antigen, whereas the shaded arrows on the right point to the positions of the viral late proteins, Agnoprotein, and the capsid protein, VP1. The positions of the PCR primers are shown by thin arrows outside of the circle. The size of the amplified DNAs and the location of the DNA probes used for Southern blot hybridization specific for the amplified T-antigen, Agnoprotein, and VP1 sequences are shown. Numbers over the hybridization blots indicate the case number on Table 2. −, negative control; +, positive control; M, molecular weight standards. Size of the different JCV PCR product is indicated.

Close modal
Fig. 2.

Expression of JCV proteins in human colon cancer. A, dissection of biopsy with an area of normal mucosa on top and tumoral emboli on the bottom was immunostained with anti-T-antigen antibody. Immunoreactivity was observed in tumor cells but not in normal areas. B, immunostaining of a villous adenoma with T-antigen shows nuclear labeling of columnar cells. C, invasive epithelial glandular neoplastic cells displayed immunoreactivity with anti-T-antigen antibody. Cytoplasmic perinuclear detection of Agnoprotein in a polyp and in neoplastic cells is shown (D and E, respectively). Negative immunoreactivity of both normal and neoplastic areas were observed with anti-VP1 antibody (F and G, respectively). Magnification: ×200 (A, F, and G); ×400 (B–E).

Fig. 2.

Expression of JCV proteins in human colon cancer. A, dissection of biopsy with an area of normal mucosa on top and tumoral emboli on the bottom was immunostained with anti-T-antigen antibody. Immunoreactivity was observed in tumor cells but not in normal areas. B, immunostaining of a villous adenoma with T-antigen shows nuclear labeling of columnar cells. C, invasive epithelial glandular neoplastic cells displayed immunoreactivity with anti-T-antigen antibody. Cytoplasmic perinuclear detection of Agnoprotein in a polyp and in neoplastic cells is shown (D and E, respectively). Negative immunoreactivity of both normal and neoplastic areas were observed with anti-VP1 antibody (F and G, respectively). Magnification: ×200 (A, F, and G); ×400 (B–E).

Close modal
Fig. 3.

LCM and gene amplification. A, selection of the cells for laser capture was guided by immunohistochemical detection of JCV T-antigen in villous adenoma and invasive adenocarcinoma. Top: three different areas with normal histological features with no visible labeling with anti-T-antigen antibody as well as precancerous (villous adenoma) and invasive adenocarcinoma with strong immunoreactivity with anti-T-antigen antibody are shown before laser capture. Middle: after the thermoplastic film is removed, the tissue left behind after laser capture show punched holes in the remaining section. Bottom: the precise removal of the target cells is confirmed by microscopic visualization before processing for DNA extraction. B, gene amplification followed by Southern blot hybridization using a pair of primers that recognize a sequence of T-antigen followed by Southern blot hybridization using probes that specifically detect the amplified JCV T-antigen gene. The position of the 173-bp-amplified JCV DNA fragment is shown by an arrow. Magnification: ×100 (all panels).

Fig. 3.

LCM and gene amplification. A, selection of the cells for laser capture was guided by immunohistochemical detection of JCV T-antigen in villous adenoma and invasive adenocarcinoma. Top: three different areas with normal histological features with no visible labeling with anti-T-antigen antibody as well as precancerous (villous adenoma) and invasive adenocarcinoma with strong immunoreactivity with anti-T-antigen antibody are shown before laser capture. Middle: after the thermoplastic film is removed, the tissue left behind after laser capture show punched holes in the remaining section. Bottom: the precise removal of the target cells is confirmed by microscopic visualization before processing for DNA extraction. B, gene amplification followed by Southern blot hybridization using a pair of primers that recognize a sequence of T-antigen followed by Southern blot hybridization using probes that specifically detect the amplified JCV T-antigen gene. The position of the 173-bp-amplified JCV DNA fragment is shown by an arrow. Magnification: ×100 (all panels).

Close modal
Fig. 4.

Detection of p53 and its colocalization with T-antigen in colon cancer. A, immunohistochemistry against p53 shows negative reactivity in normal epithelium (top) and robust labeling in neoplastic cells inside lymph vessels (bottom). Strong immunoreactivity with anti-p53 antibody was observed in villous polyp (B) and neoplastic cells in glandular structures (C). Double labeling of columnar cells with anti-p53 antibody and anti-T-antigen shows nuclear colocalization of these proteins in neoplastic cells. The presence of p53, T-antigen, and colocalization of both proteins in columnar cells of a villous adenoma are depicted in D–F, respectively. Magnification: ×200 (A); ×400 (B and C); ×1000 (D–F).

Fig. 4.

Detection of p53 and its colocalization with T-antigen in colon cancer. A, immunohistochemistry against p53 shows negative reactivity in normal epithelium (top) and robust labeling in neoplastic cells inside lymph vessels (bottom). Strong immunoreactivity with anti-p53 antibody was observed in villous polyp (B) and neoplastic cells in glandular structures (C). Double labeling of columnar cells with anti-p53 antibody and anti-T-antigen shows nuclear colocalization of these proteins in neoplastic cells. The presence of p53, T-antigen, and colocalization of both proteins in columnar cells of a villous adenoma are depicted in D–F, respectively. Magnification: ×200 (A); ×400 (B and C); ×1000 (D–F).

Close modal
Fig. 5.

Expression of Wnt pathway proteins and colocalization of β-catenin and T-antigen. A, immunohistochemical labeling of neoplastic cells with anti-β-catenin shows both, nuclear and cytoplasmic staining. B, immunostaining of the tumor cells with anti-TCF-4 antibody demonstrates strong nuclear staining. C, treatment of tumor cells with anti-LEF-1 antibody shows predominant cytoplasmic with scattered nuclear staining (arrows). D, columnar cells show nuclear labeling of tumor cells with anti-T-antigen antibody (shown by an arrow; E) The same section shows nuclear (arrow) and cytoplasmic (arrowhead) accumulation of β-catenin in the tumor areas. F, superimposition of the double-labeled tumor cells with anti-T-antigen and anti-β-catenin show frequent nuclear colocalization of the two proteins (arrow). Magnification: ×400 (A–C); ×1000 (D–F).

Fig. 5.

Expression of Wnt pathway proteins and colocalization of β-catenin and T-antigen. A, immunohistochemical labeling of neoplastic cells with anti-β-catenin shows both, nuclear and cytoplasmic staining. B, immunostaining of the tumor cells with anti-TCF-4 antibody demonstrates strong nuclear staining. C, treatment of tumor cells with anti-LEF-1 antibody shows predominant cytoplasmic with scattered nuclear staining (arrows). D, columnar cells show nuclear labeling of tumor cells with anti-T-antigen antibody (shown by an arrow; E) The same section shows nuclear (arrow) and cytoplasmic (arrowhead) accumulation of β-catenin in the tumor areas. F, superimposition of the double-labeled tumor cells with anti-T-antigen and anti-β-catenin show frequent nuclear colocalization of the two proteins (arrow). Magnification: ×400 (A–C); ×1000 (D–F).

Close modal
Fig. 6.

Interaction of JCV T-antigen with β-catenin in colon cancer cells. A, the human colon cancer cell line, HCT116, was transfected with the T-antigen expression plasmid pCMV-T-antigen. After 48 h, total protein extracts were prepared and incubated with either anti-T-antigen antibody or control preimmune serum for 16 h. The immunocomplexes were analyzed by Western blot using anti-β-catenin antibody. Lane 1 represents direct Western blot analysis of protein extract from HCT116 cells. The band representing β-catenin is depicted by an arrow. B, Western blot analysis of protein extract from untransfected and pCMV-T-antigen-transfected cells using anti-c-myc (top) and Grb-2 bottom antibodies. C, HCT116 cells were transfected with 3 μg of a plasmid containing the c-myc promoter upstream of a luciferase reporter gene either alone or together with pCMV-β-catenin and pCMV-T-antigen plasmids. The level of luciferase activity was tested after 36 h. The experiment was repeated three times, and the results with standard errors are illustrated. D, HCT116 cells were transfected with 2 μg of pCMV-T-antigen plasmid. Cells were fixed after 48 h, and expression and subcellular localization of T-antigen and β-catenin were examined by immunocytochemistry using specific antibodies (as described in “Materials and Methods”). The arrowhead shows cells that express T-antigen, whereas the arrows depict representative cells that do not receive express T-antigen.

Fig. 6.

Interaction of JCV T-antigen with β-catenin in colon cancer cells. A, the human colon cancer cell line, HCT116, was transfected with the T-antigen expression plasmid pCMV-T-antigen. After 48 h, total protein extracts were prepared and incubated with either anti-T-antigen antibody or control preimmune serum for 16 h. The immunocomplexes were analyzed by Western blot using anti-β-catenin antibody. Lane 1 represents direct Western blot analysis of protein extract from HCT116 cells. The band representing β-catenin is depicted by an arrow. B, Western blot analysis of protein extract from untransfected and pCMV-T-antigen-transfected cells using anti-c-myc (top) and Grb-2 bottom antibodies. C, HCT116 cells were transfected with 3 μg of a plasmid containing the c-myc promoter upstream of a luciferase reporter gene either alone or together with pCMV-β-catenin and pCMV-T-antigen plasmids. The level of luciferase activity was tested after 36 h. The experiment was repeated three times, and the results with standard errors are illustrated. D, HCT116 cells were transfected with 2 μg of pCMV-T-antigen plasmid. Cells were fixed after 48 h, and expression and subcellular localization of T-antigen and β-catenin were examined by immunocytochemistry using specific antibodies (as described in “Materials and Methods”). The arrowhead shows cells that express T-antigen, whereas the arrows depict representative cells that do not receive express T-antigen.

Close modal

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.

1

This work was made possible by grants awarded by NIH (to K. K.).

3

The abbreviations used are: EMA, epithelial membrane antigen; CEA, carcinoembryonic antigen; LCM, laser capture microdissection; TCF, T cell factor, LEF, leukemia enhancing factor.

Table 1

Clinical and immunohistochemical evaluation of human colon cancera

No.OriginDiagnosisAge (yr)/GenderLocationStageImmunohistochemistry
EMAKeratinDesminCEA
Mexico Adenocarcinoma 61/Fb Cecum Dukes C1 ++c ++ − 
Mexico Adenocarcinoma 55/F Sigmoid colon Dukes B1 ++ − ++ 
Mexico Adenocarcinoma 74/F Descendent colon Dukes B1 − 
Mexico Adenocarcinoma 54/F Descendent colon Dukes C1 ++ +++ − − 
Mexico Mixed Adeno + squamous 62/F Sigmoid colon Dukes B3 +++ +++ − +f 
Mexico Adenocarcinoma + villous polyp 80/F Ascendent colon Dukes B2 ++ ++ − 
Mexico Adenocarcinoma + villous polyp 74/M Sigmoid colon Dukes C1 ++ − ++ 
Mexico Adenocarcinoma + villous polyp 69/F Ascendent colon Dukes B2 − 
Mexico Adenocarcinoma + villous polyp 52/F Sigmoid colon Dukes C1 ++ ++ − ++ 
10 Mexico Adenocarcinoma 51/M Sigmoid colon Dukes B2 − ++ 
11 Mexico Adenocarcinoma + villous polyp 75/F Cecum Dukes B2 ++ +++ − 
12 Mexico Adenocarcinoma 81/M Cecum Dukes B2 +++ ++ − +++ 
13 Mexico Adenocarcinoma + villous polyp 55/F Sigmoid colon Dukes C2 ++ − +++ 
14 Mexico Adenocarcinoma (poor dif) 41/M Sigmoid colon Dukes B2 +++ − +++ 
15 Mexico Adenocarcinoma (poor dif) 85/M Descendent colon Dukes B2 +++ +++ − ++ 
16 Mexico Adenocarcinoma 71/M Descendent colon Dukes B2 − +++ 
17 Mexico Adenocarcinoma 62/F Sigmoid colon Dukes B3 +++ +++ − − 
18 USA Adenocarcinoma + villous polyp 74/M Transv colon Dukes B1 ++ − ++ 
19 USA Adenocarcinoma 69/F Cecum Dukes B1 ++ +++ − − 
20 USA Adenocarcinoma 82/M Sigmoid colon Dukes B1 ++ − +++ 
21 USA Adenocarcinoma (poor dif) 65/F Hepatic flexure Dukes B2 +++ − +++ 
22 USA Adenocarcinoma 81/M Cecum Dukes C2 +++ − +++ 
23 USA Adenocarcinoma 54/M Sigmoid colon Dukes C2 ++ − +++ 
24 USA Adenocarcinoma 52/F Rectosigmoid Dukes B2 − − ++ 
25 USA Mucinous adenocarcinoma 71/F Cecum Dukes B2 ++ ++ − 
26 USA Adenocarcinoma 65/F Cecum Dukes B1 ++ +++ − ++ 
27 USA Adenocarcinoma (poor dif) 91/F Transv colon Dukes B2 +++ +++ − +++ 
No.OriginDiagnosisAge (yr)/GenderLocationStageImmunohistochemistry
EMAKeratinDesminCEA
Mexico Adenocarcinoma 61/Fb Cecum Dukes C1 ++c ++ − 
Mexico Adenocarcinoma 55/F Sigmoid colon Dukes B1 ++ − ++ 
Mexico Adenocarcinoma 74/F Descendent colon Dukes B1 − 
Mexico Adenocarcinoma 54/F Descendent colon Dukes C1 ++ +++ − − 
Mexico Mixed Adeno + squamous 62/F Sigmoid colon Dukes B3 +++ +++ − +f 
Mexico Adenocarcinoma + villous polyp 80/F Ascendent colon Dukes B2 ++ ++ − 
Mexico Adenocarcinoma + villous polyp 74/M Sigmoid colon Dukes C1 ++ − ++ 
Mexico Adenocarcinoma + villous polyp 69/F Ascendent colon Dukes B2 − 
Mexico Adenocarcinoma + villous polyp 52/F Sigmoid colon Dukes C1 ++ ++ − ++ 
10 Mexico Adenocarcinoma 51/M Sigmoid colon Dukes B2 − ++ 
11 Mexico Adenocarcinoma + villous polyp 75/F Cecum Dukes B2 ++ +++ − 
12 Mexico Adenocarcinoma 81/M Cecum Dukes B2 +++ ++ − +++ 
13 Mexico Adenocarcinoma + villous polyp 55/F Sigmoid colon Dukes C2 ++ − +++ 
14 Mexico Adenocarcinoma (poor dif) 41/M Sigmoid colon Dukes B2 +++ − +++ 
15 Mexico Adenocarcinoma (poor dif) 85/M Descendent colon Dukes B2 +++ +++ − ++ 
16 Mexico Adenocarcinoma 71/M Descendent colon Dukes B2 − +++ 
17 Mexico Adenocarcinoma 62/F Sigmoid colon Dukes B3 +++ +++ − − 
18 USA Adenocarcinoma + villous polyp 74/M Transv colon Dukes B1 ++ − ++ 
19 USA Adenocarcinoma 69/F Cecum Dukes B1 ++ +++ − − 
20 USA Adenocarcinoma 82/M Sigmoid colon Dukes B1 ++ − +++ 
21 USA Adenocarcinoma (poor dif) 65/F Hepatic flexure Dukes B2 +++ − +++ 
22 USA Adenocarcinoma 81/M Cecum Dukes C2 +++ − +++ 
23 USA Adenocarcinoma 54/M Sigmoid colon Dukes C2 ++ − +++ 
24 USA Adenocarcinoma 52/F Rectosigmoid Dukes B2 − − ++ 
25 USA Mucinous adenocarcinoma 71/F Cecum Dukes B2 ++ ++ − 
26 USA Adenocarcinoma 65/F Cecum Dukes B1 ++ +++ − ++ 
27 USA Adenocarcinoma (poor dif) 91/F Transv colon Dukes B2 +++ +++ − +++ 
a

Diagnosis of the tumors is based on the WHO Classification of Tumours of the Digestive System, as described in “Materials and Methods.”

b

F, female; M, male. Age of the patient at the time of surgical resection is shown.

c

−, indicates negative immunoreactivity; +, 1–30% cell positivity; ++, 31–60% cell positivity; +++, >61% cell positivity; f = indicates focal positivity.

Table 2

DNA and protein analysis of human colon cancer

No.PCR/SouthernImmunohistochemistrya
T-antigenVP-1AgnoViral proteinsCellular proteins
T-antigenVP-1Agnop53APCβ-CateninTCF-4TCF-1LEF-1
− − − + cy − + cy 
− ++ − − +++ − − − − − 
− − − − − − − − − − − 
− − − ++ − − + cy − ++ cy 
− − − − − − + cy + cy + cy 
− − +f ++ ++ cy + nu + cy + nu − + cy 
− − − − ++ nu + cy + cy ++ cy 
− − − − + cy + cy ++ cy 
− − − − − − − − − − + cy 
10 − − − +++ + cy + nu − + cy + cy 
11 − +++ − +++ ++ − − − + cy 
12 − +++ − ++ +++ + cy + nu ++ cy + cy 
13 − − − − − − ++ − − − − − 
14 − − − − − − − − − − − 
15 − − ++ − +f − − + cy + nu − − 
16 − − − − − − + cy + nu − + cy 
17 − − − − ++ − − − − 
18 − +++ − + cy + nu + cy + nu − − 
19 − − − − +++ + cy + cy + nu + cy 
20 − ++ − +++ ++ + cy + nu + cy + nu − − 
21 − − − − − − ++ − + cy + cy + nu + cy − 
22 − − − − − + cy + nu − − 
23 − − − − +f − − − + cy + cy 
24 − − ++ ++ − − + cy + nu − − 
25 − − − + cy + cy + nu − + cy 
26 − − − − ++ − − + cy + nu − − 
27 − − − − − − − − − + cy − 
No.PCR/SouthernImmunohistochemistrya
T-antigenVP-1AgnoViral proteinsCellular proteins
T-antigenVP-1Agnop53APCβ-CateninTCF-4TCF-1LEF-1
− − − + cy − + cy 
− ++ − − +++ − − − − − 
− − − − − − − − − − − 
− − − ++ − − + cy − ++ cy 
− − − − − − + cy + cy + cy 
− − +f ++ ++ cy + nu + cy + nu − + cy 
− − − − ++ nu + cy + cy ++ cy 
− − − − + cy + cy ++ cy 
− − − − − − − − − − + cy 
10 − − − +++ + cy + nu − + cy + cy 
11 − +++ − +++ ++ − − − + cy 
12 − +++ − ++ +++ + cy + nu ++ cy + cy 
13 − − − − − − ++ − − − − − 
14 − − − − − − − − − − − 
15 − − ++ − +f − − + cy + nu − − 
16 − − − − − − + cy + nu − + cy 
17 − − − − ++ − − − − 
18 − +++ − + cy + nu + cy + nu − − 
19 − − − − +++ + cy + cy + nu + cy 
20 − ++ − +++ ++ + cy + nu + cy + nu − − 
21 − − − − − − ++ − + cy + cy + nu + cy − 
22 − − − − − + cy + nu − − 
23 − − − − +f − − − + cy + cy 
24 − − ++ ++ − − + cy + nu − − 
25 − − − + cy + cy + nu − + cy 
26 − − − − ++ − − + cy + nu − − 
27 − − − − − − − − − + cy − 
a

Immunohistochemistry: −, indicates negative immunoreactivity; +, 1–30% cell positivity; ++, 31–60% cell positivity; +++, >61% cell positivity; f, indicates focal positivity; cy, cytoplasmic immunoreactivity; nu, nuclear immunoreactivity; APC, adenomatous polyposis coli.

We thank past and present members of the Center for Neurovirology and Cancer Biology for their insightful discussion and sharing of ideas and reagents. We also thank Cynthia Schriver for editorial assistance and preparation of the manuscript.

1
Weber T., Major E. O. Progressive multifocal leukoencephalopathy: molecular biology, pathogenesis and clinical impact.
Intervirology
,
40
:
98
-111,  
1997
.
2
Berger J. R., Concha M. Progressive multifocal leukoencephalopathy: the evolution of a disease once considered rare.
J. Neurovirol.
,
1
:
15
-18,  
1995
.
3
Del Valle L., Gordon J., Assimakopolou M., Enam S., Geddes J. F., Varakis J., Katsetos C. D., Croul S. E., Khalili K. Detection of JC virus DNA sequences and expression of the viral regulatory protein, T-antigen, in tumors of the central nervous system.
Cancer Res.
,
61
:
4287
-4293,  
2001
.
4
Del Valle L., Gordon J., Enam S., Delbue S., Croul S., Abraham S., Radhakrishnan S., Assimakoupoulou M., Katsetos C. D., Khalili K. Expression of human neurotropic polyomavirus JCV late gene product AGNO protein in human medulloblastoma.
J. Natl. Cancer Inst. (Bethesda)
,
94
:
267
-273,  
2002
.
5
Krynska B., Del Valle L., Croul S. E., Gordon J., Katsetos C., Carbone M., Giordano A., Khalili K. Detection of human neurotropic JC virus DNA sequence and expression of the viral oncogenic protein in pediatric medulloblastomas.
Proc. Natl. Acad. Sci. USA
,
96
:
11519
-11524,  
1999
.
6
Frisque R. J., Rifkin D. B., Walker D. L. Transformation of primary hamster brain cells with JC virus and its DNA.
J. Virol.
,
35
:
265
-269,  
1980
.
7
Howley P. M., Rentier-Delrue F., Heilman C. A., Law M. F., Chowdhury K., Israel M. A., Takemoto K. K. Cloned human polyomavirus JC DNA can transform human amnion cells.
J. Virol.
,
36
:
878
-882,  
1980
.
8
Vacante D. A., Traub R., Major E. O. Extension of JCV virus host range to monkey cells by insertion of a simian virus 40 enhancer into the JC virus regulatory region.
Virology
,
170
:
353
-361,  
1989
.
9
Del Valle L., Gordon J., Ferrante P., Khalili K. JV virus in experimental and clinical brain tumorigenesis Khalili K. Stoner G. eds. .
Human Polyomaviruses: Molecular and Clinical Perspectives
,
409
-430, Wiley-Liss New York  
2001
.
10
Dyson N., Bernards R., Friend S. H., Gooding L. R., Hassell J. A., Major E. O., Pipas J. M., van Dyke T., Harlow E. Large T antigens of many polyomaviruses are able to form complexes with the retinoblastoma protein.
J. Virol.
,
64
:
1353
-1356,  
1990
.
11
Krynska B., Otte J., Franks R., Khalili K., Croul S. Human ubiquitous JCV·CY T-antigen gene induces brain tumors in experimental animals.
Oncogene
,
18
:
39
-46,  
1998
.
12
Gan D. D., Reiss K., Carrill T., Del Valle L., Croul S., Giordano A., Fishman P., Khalili K. Involvement of Wnt signaling pathway in murine medulloblastoma induced by human neurotropic JC virus.
Oncogene
,
20
:
4864
-4870,  
2001
.
13
Lassak A., Del Valle L., Peruzzi F., Wang J. Y., Croul S., Khalili K., Reiss K. Insulin receptor substrate 1 translocation to the nucleus by the human JC virus T-antigen.
J. Biol. Chem.
,
277
:
17231
-17238,  
2002
.
14
Agostini H. T., Yanagihara R., Davis V., Ryschkewitsch C. F., Stoner G. L. Asian genotypes of JC virus in Native Americans and in a Pacific Island population: markers of viral evolution and human migration.
Proc. Natl. Acad. Sci. USA
,
94
:
14542
-14546,  
1997
.
15
Kitamura T., Kunitake T., Guo J., Timinaga T., Kawabe K., Yogo Y. Transmission of the human polyomavirus JC virus occurs both within the family and outside the family.
J. Clin. Microbiol.
,
32
:
2359
-2363,  
1994
.
16
Bofill-Mas S., Pina S., Girones R. Documenting the epidemiologic patterns of polyomaviruses in human populations by studying their presence in urban sewage.
Appl. Environ. Microbiol.
,
66
:
238
-245,  
2000
.
17
Bofill-Mas S., Formiga-Cruz M., Clemente Casares P., Calafell F., Girones R. Potential transmission of human polyomaviruses through the gastrointestinal tract after exposure to virions or viral DNA.
J. Virol.
,
75
:
10290
-10299,  
2001
.
18
Bofill-Mas S., Girones R. Excretion and transmission of JCV in human population.
J. Neurovirol.
,
7
:
345
-349,  
2001
.
19
Ricciardiello L., Laghi L., Ramamirtham P., Chang C. L., Chang D. K., Randolph A. E., Boland C. R. JC virus DNA sequences are frequently present in the human upper and lower gastrointestinal tract.
Gastroenterology
,
119
:
1228
-1235,  
2000
.
20
Ricciardiello L., Chang D. K., Laghi L., Goel A., Chang C. L., Boland R. Mad-1 is the exclusive JC virus strain present in the human colon, and its transcriptional control region has a deleted 98-base-pair sequence in colon cancer tissues.
J. Virol.
,
75
:
1996
-2001,  
2001
.
21
Alexander N., Wong C. S., Pignatelli M. β-Catenin: a linchpin in colorectal carcinogenesis?.
Am. J. Pathol.
,
160
:
389
-401,  
2002
.
22
Peifer M. β-Catenin as oncogene: the smoking gun.
Science (Wash. DC)
,
275
:
1752
-1753,  
1997
.
23
Morin P. J. β-Catenin signaling and cancer.
Bioessays
,
21
:
1021
-1030,  
1999
.
24
Hamilton, S. R., and Aaltonen, L. A. (eds.). World Health Organization classification of tumours: pathology and genetics of tumours of the digestive system. Lyon, France: International Agency for Research on Cancer, 2001.
25
Ausubel F., Brent R., Kingston R., Moore D., Seidman J., Smith J., Struhl K. .
Current protocols in molecular biology
, John Wiley and Sons, Inc. New York  
1996
.
26
Monaco M. C., Atwood W. J., Gravell M., Tornatore C. S., Major E. O. JC virus infection of hematopoietic progenitor cells, primary B lymphocytes, and tonsillar stromal cells: implications for viral latency.
J. Virol.
,
70
:
7004
-7012,  
1996
.
27
Tornatore C., Berger J. R., Houff S. A., Curfman B., Meyers K., Winfield D., Major E. O. Detection of JC virus DNA in peripheral lymphocytes from patients with and without progressive multifocal leukoencephalopathy.
Ann. Neurol.
,
31
:
454
-462,  
1992
.
28
Dorries K., ter Meulen V. Progressive multifocal leukoencephalopathy: detection of papovavirus JC in kidney tissue.
J. Med. Virol.
,
11
:
307
-317,  
1983
.
29
Laghi L., Randolph A. E., Chauhan D. P., Marra G., Major E. O., Neel J. V., Boland C. R. JC virus DNA is present in the mucosa of the human colon and in colorectal cancers.
Proc. Natl. Acad. Sci. USA
,
96
:
7484
-7489,  
1999
.
30
Fearon E. R., Vogelstein B. A genetic model for colorectal tumorigenesis.
Cell
,
61
:
759
-767,  
1990
.
31
Solomon E., Voss R., Hall V., Bodmer W. F., Jass J. R., Jeffreys A. J., Lucibello F. C., Patel I., Rider S. H. Chromosome 5 allele loss in human colo-rectal carcinomas.
Nature (Lond.)
,
328
:
616
-619,  
1987
.
32
Dobbie Z., Muller P. Y., Heinimann K., Albrecht C., D’Orazio D., Bendik I., Muller H., Bauerfeind P. Expression of COX-2 and Wnt pathway gene in adenomas of familial adenomatous polyposis patients treated with meloxicam.
Anticancer Res.
,
22
:
2215
-2220,  
2002
.
33
Frisque R. J., Bream G. L., Cannella M. T. Human polyomavirus JC virus genome.
J. Virol.
,
51
:
458
-469,  
1984
.