Elsevier

Gene

Volume 265, Issues 1–2, 7 March 2001, Pages 133-145
Gene

Isolation and characterization of mouse MUC18 cDNA gene, and correlation of MUC18 expression in mouse melanoma cell lines with metastatic ability

https://doi.org/10.1016/S0378-1119(01)00349-3Get rights and content

Abstract

The cell surface adhesion molecule human MUC18 (huMUC18 or Mel-CAM) has been postulated to play a key pathogenic role in metastatic melanoma progression. To establish an immunocompetent syngeneic mouse model that would greatly facilitate our understanding of the role of MUC18 in the metastatic behavior of melanoma, we cloned and characterized the mouse MUC18 (muMUC18) cDNA gene. The gene was amplified by RT-PCR and RACE of the poly(A)+RNA isolated from the mouse melanoma cell line B16F10/Queens. The cloned muMUC18 cDNA gene contained 28 nucleotides of 5′-UTR, 908 nucleotides of 3′-UTR, and an open reading frame (ORF) of 1947 nucleotides encoding a protein of 648 amino acids, which is two amino acids longer than huMUC18. The size of the muMUC18 mRNA is about 3 kb with a shorter 3′-UTR than the huMUC18 mRNA (about 3.3 kb). Besides, the sequence in the 3′ UTR of the two mRNAs is diverse with only 31% identity. The 5′-UTR and coding sequences of the muMUC18 cDNA are 72.4 and 80.6% identical to those of huMUC18, respectively. The deduced amino acid sequence of the muMUC18 cDNA is 76.2% identical to that of huMUC18. The amino acid sequences deduced from MUC18 cDNA sequences from six other mouse melanoma cell lines are identical except one to three residues, suggesting that the muMUC18 cDNA sequence determined in this report is correct. The muMUC18 protein is predicted to be slightly more acidic than the human protein. The levels of muMUC18 mRNA and protein in nine mouse melanoma cell lines were directly proportional to their ability to establish metastatic colonies in lungs of syngeneic mice. Most biological functions of the muMUC18 may be similar to the huMUC18.

Introduction

Cell adhesion molecules (CAMs) include members of several gene families, such as integrins, lectin-containing CAMs (selectins), calcium-dependent binding CAMs (cadherins and L-CAM), and calcium-independent binding N-CAM superfamily (ICAM, VCAM, MUC18) (Edelman and Crossin, 1991). CAMs play important roles in organ formation during embryogenesis, maintenance of tissue architecture, inflammatory and immune responses, and wound healing (Edelman and Crossin, 1991). In addition, aberrant expression of different CAMs has been suggested to play a central role in tumor development and metastasis of neoplasms (Behrens, 1993) including melanoma (Tang and Honn, 1994-1995). For example, the integrins αV, α4, and β3, ICAM, MUC18, and HLA-DR are over-expressed in metastatic melanomas, whereas VCAM and E-cadherin are under-expressed (Herlyn, 1993). Over-expression of αVβ3 by K1735C23 melanoma cells has been shown to increase motility and invasiveness of these cells in vitro and directly correlated with tumor formation and metastasis in mice (Li et al., 1998). In contrast, re-expression of E-cadherin by genetic manipulation reverses the invasiveness of tumor cells (Vleminckx et al., 1991).

Human MUC18 (huMUC18, Mel-CAM), with an apparent molecular weight of 113,000, was originally identified as a glycoprotein antigen expressed on the surface of human malignant melanoma cells (Lehmann et al., 1987). HuMUC18 is not expressed in most normal tissues (for a review by Shih, 1999) including melanocytes, but may be detected in hair follicular cells (Lehmann et al., 1987), smooth muscle cells (Lehmann et al., 1987), vascular endothelial cells (Bardin et al., 1996), normal breast epithelium (Shih, 1999), cerebellum (Sers et al., 1993), and activated T cells (Pickl et al., 1997). HuMUC18 is weakly and less frequently expressed in benign nevi and some primary melanomas. In contrast, huMUC18 is strongly and more frequently expressed in vertical growth melanomas (Lehmann et al., 1987). The malignant potential of cutaneous melanoma is directly correlated with the vertical thickness of the lesion (Breslow, 1970). Thus, the over-expression of huMUC18 has been implicated in the development of malignant melanoma (Johnson, 1994-1995). Consistent with this hypothesis, the stable expression of transfected huMUC18 cDNA gene in a low-tumorigenic and non-metastatic SB-2 human cutaneous melanoma cell line has significantly increased its tumorigenecity and metastatic capability in nude mice (Xie et al., 1997). The stable expression of transfected huMUC18 cDNA gene in other low-tumorigenic and non-metastatic human cutaneous melanoma cell lines (Xp-44 and SK-2) have only significantly increased the metastatic capability, but not tumorigenecity, of the cell line XP-44 in SCID mouse models (Schlagbauer-Wadl et al., 1999). However, the knowledge learned from these xenografts in immune-deficient mouse models is limited in its usage for extrapolating to understand human melanomas.

The cDNA of huMUC18 has been isolated and sequence-characterized; it encodes 646 amino acids (GenBank access #M28882 and Wu et al., 1999a). Similar to ICAM, the amino acid sequence of huMUC18 protein contains five immunoglobulin-like domains, and thus huMUC18 is categorized as a member of the immunoglobulin superfamily (Lehmann et al., 1989). HuMUC18 appears to play a role in cell-cell heterotypic adhesion and communication, but its ligand has not been identified. HuMUC18 may form a link in cell-cell signaling for growth and differentiation. However, the corresponding mouse MUC18 cDNA has not been cloned, nor its biological properties characterized.

A syngeneic mouse model with a complete immune system would greatly facilitate our ability to study the mechanism by which MUC18 facilitates metastatic behavior of melanoma cells. It also offers many advantages over a xenograft mouse system for metastatic studies: (a) It has a complete wild type immune system, thus the model is similar to the real situation in human melanoma cases. (b) The system is useful for testing biological responses of various therapeutic means before they are to be applied to human cases. (c) The significance of a gene in promoting or potentiating melanoma metastasis can be evaluated in knockout mice and in knockin mice.

We have used the methods of RACE and RT-PCR to amplify and isolate the mouse MUC18 (muMUC18) cDNA gene from seven mouse melanoma cell lines. Identical muMUC18 cDNA sequence was found except with minor variations. The muMUC18 cDNA sequence was compared to that of huMUC18 cDNA and found to have a high identity. But the size of the muMUC18 mRNA is shorter than that of huMUC18 mRNA with a more diverse 3′-UTR. We found that the levels of the expression of muMUC18 mRNA and protein in nine mouse melanoma cell lines were directly correlated with their metastatic capabilities. From the evidence, we conclude that most biological functions of the muMUC18 may be similar to the huMUC18. This work has been presented (Wu et al., 1999b).

Section snippets

Cell lines and cell culturing conditions

The human melanoma cell line SK-Mel-28 from the ATCC and mouse melanoma cell lines B16F1, B16F10, K1735-3, K1735-9, K1735-10, K1735-4, and K1735-m2 from Dr Isaiah J. Fidler (M.D. Anderson medical Center/U of Texas, Houston, TX) were maintained in Eagle's MEM supplemented with sodium pyruvate and 10% fetal bovine serum (GIBCO/BRL, Life Technology). Mouse melanoma cell lines B16F10/Queens and B16F10D5.1G4 (Knisley and Niederkorn, 1990) from Dr Jerry Y. Niederkorn (U of Texas Southwestern Medical

Expression of huMUC18-like mRNA in two mouse melanoma cell lines

A previous study using a human MUC18 cDNA probe to detect the expression of huMUC18-like mRNA in mouse melanoma cells suggested that mouse B16F1 melanoma cells express very low levels (about 1%) of huMUC18-like mRNA (Lehmann et al., 1989). Additionally, the size of huMUC18-like mRNA was shown to be similar to that of huMUC18 mRNA (Lehmann et al., 1989). If this result is correct, to facilitate RT-PCR amplification of muMUC18 cDNA it would be highly desirable to search for other mouse melanoma

Acknowledgements

We thank Dr Richard W. Compans for critical reading the manuscript. We thank Ms. Li Xin for providing primary culture of human melanocytes. This work was supported by the Winship Cancer center melanoma seed grant, Emory Skin Diseases Research Center feasibility pilot study fund, Emory University Research committee, and in part by a grant from NCI (R21C69764) and a grant from the Department of Defense (PC992041) (G.-J. Wu). Hsiuchin Yang was supported by a postdoctoral fellowship grant from the

References (30)

  • A. Breslow

    Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma

    Ann. Surg.

    (1970)
  • P. Chomczyuski et al.

    Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction

    Anal. Biochem.

    (1987)
  • G.M. Edelman et al.

    Cell adhesion molecules: implications for a molecular histology

    Ann. Rev. Biochem.

    (1991)
  • M. Herlyn
  • J.P. Johnson

    Identification of molecules associated with the development of metastasis in human malignant melanoma

    Invasion Metastasis

    (1994-1995)
  • Cited by (0)

    View full text