Papillomavirus-mediated neoplastic progression is associated with reciprocal changes in JAGGED1 and manic fringe expression linked to notch activation - PubMed (original) (raw)

Papillomavirus-mediated neoplastic progression is associated with reciprocal changes in JAGGED1 and manic fringe expression linked to notch activation

Karthikeyan Veeraraghavalu et al. J Virol. 2004 Aug.

Abstract

Infection by high-risk human papillomaviruses (HPV) and persistent expression of viral oncogenes E6 and E7 are causally linked to the development of cervical cancer. These oncogenes are necessary but insufficient for complete transformation of human epithelial cells in vivo. Intracellular Notch1 protein is detected in invasive cervical carcinomas (ICC), and truncated Notch1 alleles complement the function of E6/E7 in the transformation of human epithelial cells. Here we investigate potential mechanisms of Notch activation in a human cervical neoplasia. We have analyzed human cervical lesions and serial passages of an HPV type 16-positive human cervical low-grade lesion-derived cell line, W12, that shows abnormalities resembling those seen in cervical neoplastic progression in vivo. Late-passage, but not early-passage, W12 and progression of the majority of human high-grade cervical lesions to ICC showed upregulation of Notch ligand and Jagged1 and downregulation of Manic Fringe, a negative regulator of Jagged1-Notch1 signaling. Concomitantly, an increase in Notch/CSL (CBF1, Suppressor of Hairless, Lag1)-driven reporter activity and a decrease in Manic Fringe upstream regulatory region (MFng-URR)-driven reporter activity was observed in late-passage versus early passage W12. Analysis of the MFng-URR revealed that Notch signaling represses this gene through Hairy Enhancer of Split 1, a transcriptional target of the Notch pathway. Expression of Manic Fringe by a recombinant adenovirus, dominant-negative Jagged1, or small interfering RNA against Jagged1 inhibits the tumorigenicity of CaSki, an ICC-derived cell line that was previously shown to be susceptible to growth inhibition induced by antisense Notch1. We suggest that activation of Notch in cervical neoplasia is Jagged1 dependent and that its susceptibility to the influence of Manic Fringe is of therapeutic value.

PubMed Disclaimer

Figures

FIG. 1.

FIG. 1.

Evaluation of Jagged1 and Manic Fringe expression in CaSki cells. (A) Total cell extracts from CaSki and NCK were analyzed by immunoblotting with anti-Jagged1 (TS1.15SH; DSHB) and anti-Notch1 (bTAN-20; DSHB) antibodies. Lane M, molecular mass marker with bands of indicated sizes. Jagged1 (∼175-kDa band) and the cytoplasmic portion of Notch1 (∼110-kDa band) are indicated by arrowheads. As a protein-loading control, the same blot was reprobed with anti-β-actin antibody. (B) Twenty-five micrograms of total RNA isolated from CaSki was subjected to Northern blot analysis with a 32P-labeled Manic Fringe cDNA probe. Total RNA isolated from A431 cells stably expressing Manic Fringe (A431-mFNG) was loaded as a positive control. The 1.9-kb Manic Fringe transcript is indicated by an arrowhead. The lower panel shows 28S and 18S rRNA as loading controls.

FIG. 2.

FIG. 2.

Early versus late-passage W12 cells show alterations in expression of Jagged1 and Manic Fringe. (A) Total protein lysate from the cells mentioned were immunoblotted with the various antibodies indicated (panels 1 to 6). Lanes loaded with lysates from W12 late passages 49 and 56 are indicated in boldface type. An NIH 3T3 lysate was used as a negative control for Jagged1 expression. A CaSki lysate was loaded as a positive control for Notch1 and Jagged1 expression. The immunoblot of β-actin shows the protein loading control. (B) Total cell extracts from the cell lines mentioned were immunoblotted with anti-Manic Fringe antibody. As a protein-loading control, the same blot was reprobed with anti-β-actin antibody.

FIG. 3.

FIG. 3.

Expression pattern of Jagged1 and Manic Fringe on organotypic raft cultures. Panels A, B, and C show immunohistochemical staining of Notch1 in organotypic rafts generated by W12p9, W12p59, and CaSki cells, respectively. Panels D, E, and F show expression of Jagged1 and panels G, H, and I show expression of Manic Fringe in rafts generated by W12p9, W12p59, and CaSki cells, respectively.

FIG. 4.

FIG. 4.

Histomorphological characterization of organotypic raft cultures. Photographs show hematoxylin and eosin (H & E) histology of raft tissue generated by early passage W12p9 (A), late-passage W12p59 (B), and CaSki cells (C). Panels D, E, and F show immunohistochemical staining for CK-10 (CK10) in rafts generated by W12p9, W12p59, and CaSki cells, respectively. Panels G, H, and I show immunohistochemical staining for CK-19 (CK19) in rafts generated by W12p9, W12p59, and CaSki cells, respectively. Dysplastic pockets of cells in rafts generated by W12p59 are indicated by arrows.

FIG. 5.

FIG. 5.

Endogenous CSL-dependent Notch signaling in W12 cells. (A) W12p10 and W12p56 cells were transfected with either a HES1-Luc reporter (1 μg) containing two intact CSL binding sites or a mutant HES1-Luc reporter (1 μg) that lacked both of the CSL binding sites. The graph shows the increase in reporter activity of HES1-Luc over mutant HES1-Luc in W12p10 and W12p56 cells. (B) W12p56 cells were transfected with HES1-Luc reporter (1 μg) or mutant HES-Luc reporter (1 μg). The graph shows the increase in reporter activity of HES1-Luc over mutant HES1-Luc in W12p56 cells. In addition to the reporter constructs, cells were either cotransfected with 3 μg of mentioned plasmids (mock vector [pcDNA3-Neo], pcDNA3-Sol hJag1, pLK01-puro mock vector, or pLK01-Si hJag1) or treated with 20 μM GI or DMSO (vehicle control) or infected with recombinant adenovirus expressing GFP alone (Ad-GFP) or Manic Fringe (Ad-mFNG). The results shown represent the means ± standard errors of the means of the results from three independent transfection experiments. Transfection efficiency was normalized by using the dual luciferase assay system.

FIG. 6.

FIG. 6.

Expression of Notch pathway genes in CIN III to SCC transition. (Left panel) Representative photomicrographs show expression of mentioned Notch pathway genes in CIN III and SCC cases as determined by mRNA in situ hybridization. (A and B) Jagged1 (Jag1); (E and F) Notch1; (I and J) HES1; (M and N) Manic Fringe (MFng). FITC-labeled RNA probes and an alkaline phosphatase-conjugated anti-FITC antibody-based detection system were employed. Antisense staining is in purple (indicated by arrows), and the sections were counterstained with fast green. (Middle panel) Representative photomicrographs show immunohistochemical detection of mentioned Notch pathway genes in CIN III and SCC cases. (C and D) Jagged1 (Jag1); (G and H) full-length Notch1; (K and L) HES1; (O and P) Manic Fringe (MFng). Arrows indicate areas of positive DAB staining. The counterstain is either hematoxylin or fast green. Photomicrographs were taken under ×40 magnification. (Q) Graph shows the percentage of CIN III and SCC cases scored positive for the Notch1, Jagged1, Manic Fringe, and HES1 transcripts based on the results obtained from RNA in situ hybridization. The number of positive cases over the total number of cases analyzed is mentioned below the graph. The criteria employed for scoring samples positive or negative are discussed in Materials and Methods.

FIG. 7.

FIG. 7.

HES1 negatively regulates Manic Fringe promoter activity. (A) Features of the 3.5-kb 5′ MFng-URR gene deduced from the human genomic DNA clone RP5-889J22 on chromosome 22q13.1. Sequence examination revealed a LhX2 binding element (nucleotide position −650), multiple HES1 interacting N-box elements (CACNAG) around nucleotide position −250, and multiple GREs at nucleotide position −30 upstream of the predicted transcriptional start site (nucleotide position +1). This 3.5-kb 5′ MFng-URR gene was cloned upstream of the CAT reporter gene (pMfP-CAT). (B) The promoterless CAT reporter (pBasic-CAT [1 μg]) and MFng-URR-driven CAT reporter (pMfP-CAT [1 μg]) constructs were transfected into W12p10, W12p56, HeLa, CaSki, and SiHa cell lines. The graph represents the increase in reporter activity of pMfP-CAT over pBasic-CAT. (C) Series of 5′ deletion MFng-URR-driven CAT reporter constructs lacking various transcriptional elements. pMfΩ1P-CAT retained the URR up to nucleotide position −650, spanning the LhX2, HES1, and GREs. pMfΩ2P-CAT retained the URR up to nucleotide position −250, spanning only the HES1 N-box and GREs. pMfΩ3P-CAT retained the URR up to nucleotide position −50, spanning only the GRE. (D) The graph represents the normalized CAT reporter activity of various Manic Fringe promoter deletion constructs in W12p10 cells cotransfected with 3 μg of pcDNA3-HES1 (HES1), pcDNA3-AcN1 (AcN1), pGLhX2, or mock vector (Neo). (E) CaSki cells were transfected with the pBasic-CAT (1 μg) or pMfP-CAT (1 μg) reporter construct. The graph shows the increase in reporter activity of pMfP-CAT over pBasic-CAT in CaSki cells. In addition to the reporter constructs, cells were either cotransfected with 3 μg of mentioned plasmids (pLK01-puro mock vector, pLK01-puro Si hJag1, pcDNA3-Sol hJag1, or pcDNA3-Hes1) or treated with 20 μM GI. The last bar represents the increase in reporter activity of pMfΩ3P-CAT over pBasic-CAT in CaSki cells. The results shown in panels B, D, and E represent the means ± standard errors of the means of the results from three independent transfection experiments. Transfection efficiency was normalized by using the dual luciferase assay system.

FIG. 8.

FIG. 8.

Jagged1-induced Notch activity is necessary for the maintenance of tumorigenicity in CaSki cells. (A) CaSki cells were transfected with either HES1-Luc reporter (1 μg) or mutant HES1-Luc reporter (1 μg). The graph shows the increase in reporter activity of HES-Luc over mutant HES1-Luc in CaSki cells. In addition to the reporter constructs, cells were either cotransfected with 3 μg of the mentioned plasmids (mock vector [pcDNA3-Neo], pcDNA3-Sol hJag1, pLK01-puro mock vector, or pLK01-puro Si hJag1) or treated with 20 μM GI or infected with recombinant adenovirus expressing GFP alone (Ad-GFP) or Manic Fringe (Ad-mFNG). The results shown on the graph represent the means ± standard errors of the means of the results from three independent transfection experiments. Transfection efficiency was normalized by using the dual luciferase assay system. Student's t test was used to obtain statistical significance compared between cells infected with Ad-GFP and Ad-mFNG. *, P < 0.05. (B) The graph shows the number of colonies on soft agar generated by CaSki cells treated with vehicle control (DMSO) or 20 μM GI. The results shown on the graph represent the means ± standard errors of the means of the results from three independent experiments, and data were generated by counting colonies in 10 random fieldsunder ×10 magnification. (C) The graph shows the number of colonies on soft agar generated by CaSki cells infected with Ad-GFP or Ad-mFNG. The results shown on the graph represent the means ± standard errors of the means of the results from three independent experiments, and data were generated by counting colonies in 30 random fields under ×10 magnification. (D) The graph shows tumor volumes in BALB/c nude mice generated by injection of either plain CaSki cells, CaSki cells transfected with mock vector (pcDNA3) or pcDNA3-Sol hJag1, CaSki cells infected with Ad-GFP or Ad-mFNG, and CaSki cells transfected with pLK01-puro or pLK01-si hJag1. Each bar represents the mean ± standard error of the mean of the tumor volume obtained at 3 weeks from five independent sets of experiments.

Similar articles

Cited by

References

    1. Aasen, T., M. B. Hodgins, M. Edward, and S. V. Graham. 2003. The relationship between connexins, gap junctions, tissue architecture and tumour invasion, as studied in a novel in vitro model of HPV-16-associated cervical cancer progression. Oncogene 22:6025-6036. - PubMed
    1. Alazawi, W., M. Pett, B. Arch, L. Scott, T. Freeman, M. A. Stanley, and N. Coleman. 2002. Changes in cervical keratinocyte gene expression associated with integration of human papillomavirus 16. Cancer Res. 6:6959-6965. - PubMed
    1. Allenspach, E. J., I. Maillard, J. C. Aster, and W. S. Pear. 2002. Notch signaling in cancer. Cancer Biol. Ther. 5:466-476. - PubMed
    1. Artavanis-Tsakonas, S., M. D. Rand, and R. J. Lake. 1999. Notch signalling: cell fate control and signal integration in development. Science 284:770-776. - PubMed
    1. Ascano, J. M., L. J. Beverly, and A. J. Capobianco. 2003. The C-terminal PDZ-ligand of JAGGED1 is essential for cellular transformation. J. Biol. Chem. 278:8771-8779. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources