Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Translational Cancer Medicine 2008: Cancer Clinical Trials and Personalized Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Cell Growth & Differentiation
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ntougkos, E.
Right arrow Articles by Sellar, G. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ntougkos, E.
Right arrow Articles by Sellar, G. C.
Clinical Cancer Research Vol. 11, 5764-5768, August 15, 2005
© 2005 American Association for Cancer Research

Human Cancer Biology

The IgLON Family in Epithelial Ovarian Cancer: Expression Profiles and Clinicopathologic Correlates

Evangelos Ntougkos1, Robert Rush1, Diane Scott1, Tobias Frankenberg1, Hani Gabra2, John F. Smyth1 and Grant C. Sellar1

Authors' Affiliations: 1 Cancer Research UK Centre, University of Edinburgh, Edinburgh and 2 Department of Cancer Medicine, Imperial College, London, United Kingdom

Requests for reprints: Grant C. Sellar, Cancer Research UK Centre, The University of Edinburgh, Crewe Road South, Edinburgh EH4 2XR, United Kingdom. Phone: 44-131-777-3541; Fax: 44-131-777-3520; E-mail: grant.sellar{at}cancer.org.uk.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: The IgLON family of cell adhesion molecules, comprising OPCML, HNT, LSAMP, and NEGR1, has recently been linked to cancer, through two of its members being proposed as tumor suppressors. We examined the expression profile of the family in human sporadic epithelial ovarian cancer and the normal ovary.

Experimental Design: We determined the expression level of each IgLON in a panel comprising 57 tumor and 11 normal ovarian samples by quantitative real-time reverse transcription-PCR. The results were statistically tested for associations with clinicopathologic variables.

Results: OPCML, LSAMP and NEGR1 exhibited reduced expression in the tumor samples relative to the normal samples, whereas HNT expression was elevated. Statistically significant changes were specific to histologic type. The expression levels of individual IgLONs were correlated, the most significant finding being a positive correlation between LSAMP and NEGR1. LSAMP expression was also negatively correlated with overall survival and was found to be a negative predictor of outcome.

Conclusions: The expression of the IgLON family is altered in sporadic epithelial ovarian tumors in comparison to the normal ovary. In our small but representative cohort of patients, we have found significant correlations and associations in expression and clinicopathology that suggest a wider role of the family in ovarian cancer.

Ovarian cancer is the leading cause of gynecologic death in both the U.S. and the U.K., in fact, it ranks fourth as a cause of death from any cancer in women (Cancer Research UK statistics; ref. 1).3 The incidence of the disease increases with age, peaking in around the seventh to eighth decade (2). Approximately 90% of ovarian cancers are epithelial ovarian cancers (EOC), with 5% to 10% being familial in origin (3, 4). Although, if diagnosed early, it has a very good prognosis, with a 92% 5-year survival rate, most cases (75%) are diagnosed at a late stage of disseminated disease when the prognosis is much less favorable, with only a 25% 5-year survival rate (5). Treatment of ovarian cancer is usually based on surgical cytoreduction and platinum-based chemotherapy (6). Although primarily a chemoresponsive disease, chemoresistance develops in most cases, resulting in the high aforementioned mortality rate.

The IgLONs are an immunoglobulin subfamily of glycosylphosphatidylinositol-anchored cell adhesion molecules comprising opioid binding protein/cell adhesion molecule–like (OPCML/OBCAM; ref. 7), limbic system–associated membrane protein (LSAMP/LAMP; ref. 8), neurotrimin (HNT; ref. 9), and neuronal growth regulator 1 (NEGR1/Kilon; ref. 10). Our knowledge of the functions of these proteins is limited and mainly derives from studies of chick and rat brain, the tissue where they are primarily expressed. Their patterns of expression in the brain are overlapping but also distinct, with OPCML being mainly expressed in the gray matter (11), HNT in the sensorimotor cortex and the cerebellum (9), LSAMP in cortical and subcortical neurons of the limbic system (8), and NEGR1 in the cerebrum and brain stem (10). In these studies, it is suggested that the IgLONs could have an important role in cell adhesion and cell-cell recognition, mediated by both homo- and heterophilic interactions in cis and trans that are likely to be stabilized by clustering of the proteins on the cell surface (12). It has been recently suggested that IgLONs function mainly as heterodimers termed Diglons (13).

OPCML has been shown to exhibit functional characteristics of a tumor suppressor gene in an ovarian cancer cell line in vitro and also in vivo when xenografted into nude mice (14). Abrogated expression was found in 83% of sporadic EOC samples examined. The mechanism underlying OPCML silencing was shown to be CpG island hypermethylation. This first report of an IgLON involvement in cancer was followed by a further one, where the LSAMP promoter was shown to be methylated in 26% of a panel of sporadic clear cell renal cell carcinomas (15). Moreover, expression of LSAMP in clear cell renal cell carcinoma lines inhibited cell proliferation.

We have investigated the expression of the four members of the IgLON family in a panel of 11 normal human ovaries and 57 tumor samples from patients with EOC. Our primary objective has been to observe each gene separately, as well as to attempt to draw conclusions from the interrelationships of data that emerge. In this extensive study, we report significant associations between the expression levels of the IgLONs and various clinicopathologic variables.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Patient samples. Primary human ovarian tumor material and nonmalignant tissues were obtained from patients having undergone gynecologic surgery in the Lothian University Hospitals National Health Service Trust. Institutional ethical approval was granted for this work by the Lothian University National Health Service Trust, Medicine/Clinical Oncology Research Ethics Subcommittee. Tissue samples were excised and stored in liquid nitrogen. Nonmalignant tissue samples were derived from patients that underwent bilateral oophorectomies for suspected malignancy, but were found to have benign histologies; samples were collected from apparently normal contralateral ovaries.

In this study, a panel comprising 57 sporadic ovarian tumors and 11 normal ovaries was used. The features of the panel are shown in Table 1. Overall, our unselected cohort is very representative of the presentation of 1,400 women from the Scottish community over 20 years to a single center in a completely unselected fashion from primary care (16). Our present cohort is exactly as expected for stage distribution in a Scottish population; although it is skewed slightly towards a more adverse grade distribution. The overrepresentation of grade 3 tumors may account for a slightly shorter median survival of 2.25 years compared with 2.50 years.


View this table:
[in this window]
[in a new window]
  		Table 1. The panel of normal ovaries and EOC samples

 
RNA preparation. Total RNA was prepared from the ovarian tissues and DNase I-treated using the Absolutely RNA reverse transcription-PCR Miniprep Kit (Stratagene, La Jolla, CA), as per the manufacturer's instructions.

Quantitative real-time reverse transcription-PCR. Quantitative reverse transcription-PCR was done using Rotorgene 2000 and 3000 real-time thermal cyclers (Corbett Research, Australia). Fifteen-microliter reactions were set up using 40 ng of DNase I-treated total RNA, 0.3 µmol/L each of forward and reverse primer and the Quantitect SYBR Green one-step reverse transcription-PCR kit (Qiagen, Crawley, United Kingdom), according to the manufacturer's guidelines. Primers were designed using Primer 3 v0.2. Primer sequences are available upon request. All reactions were carried out in triplicate for the standard curve samples and in quadruplicate for the experimental samples and the negative control (no template). Fluorescence was detected using the FAM channel (source 470 mm, detector 510 nm). Analysis and quantification was done using Rotorgene v4.6 and v5 software. The relative quantification method was employed by extrapolation from a standard curve and calculation of expression levels as ratios to ß-actin.

Cycling conditions were set as follows: reverse transcription at 50°C for 30 minutes, followed by a 15-minute polymerase activation at 95°C, and 40 cycles of denaturation at 94°C for 15 seconds, annealing at 57°C for 30 seconds and extension at 72°C for 30 seconds; after a final extension at 72°C for 1 minute, product melting was set across a 72°C to 99°C temperature ramp, with 5-second steps of 1°C. Data were corrected for inter-run variation using a "normalizer" sample that was included in all runs.

Statistical analysis. All statistical analyses were conducted in SPSS v12 (SPSS Inc., Chicago, IL). Nonparametric comparisons between unmatched samples were made using the Mann-Whitney (U) test for two samples or the Kruskal-Wallis ({chi}2) test for more than two samples. Nonparametric correlations were tested using the Spearman rank (rs) test. Regression models were made using linear regression analysis. Survival curves were produced using the Kaplan-Meier method and tested with the log-rank test. Survival models were estimated using Cox proportional hazards regression analysis.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
IgLON expression levels differ between normal ovaries and EOC. In order to evaluate the RNA expression levels of the four IgLON family members in the panel that comprised 11 normal human ovaries and 57 ovarian tumor samples, a real-time quantitative reverse transcription-PCR approach was followed. The results are shown in Fig. 1. For OPCML, LSAMP, and NEGR1, the majority of tumors exhibit reduced expression as compared with the normal ovary samples. In contrast, tumor samples have increased HNT expression in relation to nonmalignant ones. The distribution of values in all four genes is not normal, hence, nonparametric approaches were chosen for statistical analysis.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 1. IgLON expression in EOC and normal ovaries. The relative expression levels of the four IgLON genes, OPCML (A), HNT (B), LSAMP (C), and NEGR1 (D) were investigated in a panel of EOC samples and normal human ovaries using a quantitative reverse transcription-PCR approach. Black, normal samples; gray, tumor samples.

 
Associations of IgLON expression with EOC histologic types. The expression data prompted us to take a closer look at each gene separately in the different histologic types and make comparisons to expression in the normal ovaries. In all four IgLONs, there are statistically significant differences between normal ovaries and tumor subgroups (Fig. 2). More specifically, OPCML expression is reduced in clear cell carcinomas (U = 19, P = 0.039); LSAMP expression is reduced in endometrioid tumors (U = 28, P = 0.012); and NEGR1 expression is reduced in serous (U = 59, P = 0.014), clear cell (U = 9, P = 0.009), and endometrioid (U = 7, P = 0.001) tumors. In contrast, HNT expression is elevated in serous carcinomas as compared with normal ovaries (U = 113, P = 0.008).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2. IgLON expression according to histologic type. The EOC sample panel was categorized by histologic type, i.e., into serous, clear cell, mucinous, and endometrioid tumors. The median relative expression levels of OPCML (A), HNT (B), LSAMP (C), and NEGR1 (D) are shown for these subgroups as compared with normal ovary samples. *, P < 0.05; **, P < 0.01.

 
Correlations in expression between IgLONs. As all the four genes that constitute the focus of this study are structurally and functionally related, a question that arises is whether their individual expression levels are interlinked. The following statistically significant correlations of expression between IgLONs were found: LSAMP and NEGR1 (rs = 0.553, P < 0.001); OPCML and NEGR1 (rs = 0.389, P = 0.001); OPCML and LSAMP (rs = 0.356, P = 0.003). In all these cases, the correlations are positive.

Associations between IgLON expression and clinicopathologic variables. In an attempt to establish whether expression of IgLON family members could be a function of age as a continuous variable, the expression of each gene was tested against age. However, evidence to support associations was not found, either by correlation testing or by regression analysis.

Neither was stage found to differentiate IgLON expression; nonetheless, there is a statistically significant difference in LSAMP expression among tumors of different grades ({chi}2 = 6.73, P = 0.035). In fact, as can be seen in Fig. 3, this difference is even more significant when comparing grade 3 tumors with grades 1 and 2 tumors: undifferentiated carcinomas have lower levels of LSAMP than moderately or well-differentiated ones (U = 148, P = 0.017). In comparison to the large study of Clark et al. (16), grade 1 tumors are underrepresented in our relatively small sample cohort. The low number of grade 1 cases in our cohort has prevented us from comparing expression in these samples against grades 2 or 3, which is a more customary comparison.



View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3. LSAMP expression according to differentiation status. The EOC sample panel was divided into tumors of poor differentiation, i.e., of grade 3 histology, and tumors that are moderately or well-differentiated, i.e., of grades 1 or 2 histology. Columns, median relative expression of LSAMP in each of these tumor subgroups.

 
IgLON expression and survival. The impact of IgLON expression levels on the survival of the patients was tested at two levels. Firstly, correlation between expression of each gene and overall survival was investigated. This highlighted LSAMP as having a negative correlation (rs = –0.296, P = 0.025).

Patients were then divided into high and low expressers for each gene, using the median values of the normal samples as cutoff points. Survival curves were plotted and high expressers were compared with low expressers (data not shown). However, we were not able to show statistical significance for any of the four genes on this division.

To address whether the expression level of an IgLON gene can predict clinical outcome, we tested the expression variables, as well as the clinicopathologic ones, for prognostic value by Cox regression analysis. As can be seen in Table 2, in univariate analyses, log-transformed LSAMP expression, as well as stage, were found to be negative predictors of overall survival. Both variables were still found to be significant predictors of overall survival when tested together by multivariate analysis.


View this table:
[in this window]
[in a new window]
  		Table 2. Survival models

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
When the link between a member of the IgLON family, OPCML, and EOC was put forward (14), questions were raised regarding the role of this gene in ovarian physiology and ovarian cancer pathology. The realization that the family as a whole might be associated with ovarian or other types of cancer also emerged. This probability was subsequently strengthened by the finding of another IgLON, LSAMP, being associated with clear cell renal cell carcinomas as a putative tumor-suppressor gene (15). Could there be a case for the IgLON family being of importance in cancer?

To answer this question, one needs a powerful global approach that examines the family as a whole. This study has been based on a well-characterized panel of ovarian tumors and normal ovaries, which has come under scrutiny for the expression of each IgLON separately, but also for expression patterns and associations deduced in a statistically sound manner.

Here we present evidence that, at least with respect to EOC, it is not just OPCML, but the other members as well, which may be of relevance. We have shown that the expression of all the IgLONs differs significantly between normal ovaries and EOC at the RNA level. The expression levels of OPCML, LSAMP, and NEGR1 are all reduced in EOC, whereas that of HNT is increased. Confirmation of these trends at the protein level will merit investigating whether, apart from OPCML, there could be a case for LSAMP and NEGR1 acting as tumor suppressors in sporadic EOC. It has been shown that the cause of the reduction in expression in the case of OPCML in EOC is mainly epigenetic (14), and there is evidence that the same applies for LSAMP in clear cell renal cell carcinomas (15). Thus, it seems likely that the reduction in LSAMP and NEGR1 expression in EOC may be attributed to epigenetic mechanisms. In addition, we have found a significant positive correlation, the strongest in this study, between the expression of LSAMP and that of NEGR1. Consequently, it could be hypothesized that both genes need not be genetically or epigenetically silenced in EOC; the reduction in expression of the one could result in a reduction in expression of the other, potentially through a mechanism of coordinated transcriptional regulation. Although epigenetic modification is an age-related variable which affects gene expression, we have not found an association between altered IgLON expression and age in this study.

We have shown that in EOC, the importance of each IgLON might be specific to histologic type. OPCML expression is significantly reduced in clear cell carcinomas, LSAMP in endometrioid, and NEGR1, exhibiting the most striking reduction in expression, in serous, clear cell, and endometrioid tumors. In contrast, HNT is significantly elevated in serous carcinomas. With regard to mucinous carcinomas, it should be noted that the lack of statistical significance could be accounted for by the under-representation of this type of tumor in the panel.

Malignant transformation in ovarian cancer is characterized by a differentiation process, in contrast with most other tumor types, where the converse occurs (17). Normal ovarian epithelium is not fully differentiated, retaining certain mesenchymal characteristics; upon transformation, differentiation drives the epithelium toward certain histologic morphologies, depending on the type of the tumor. Because IgLONs have partially distinct patterns of histologic subtype expression, they could be part of a different pathway that guides this differentiation process.

Interestingly, LSAMP expression is significantly higher in moderately and well-differentiated tumors. This indicates that the level of LSAMP expression is likely to be a biological function of differentiation. It could be speculated that the significance of reduction of LSAMP expression in the process of malignant transformation of the ovarian surface epithelium could lie in the cells acquiring a less differentiated phenotype, which is considered to be of higher malignant potential.

Furthermore, LSAMP was found to be a negative prognostic factor in both univariate and multivariate models of EOC in addition to being negatively correlated with overall survival. In both cases, the impact of LSAMP on survival is not high. This link may seem paradoxical, given that its overall tumor levels are significantly lower than in normal ovaries. However, the role of residual LSAMP in the tumors might be antithetical to its original functions in the normal ovary. In terms of the multifactorial output described as survival, ovarian tumors could be disadvantaged by not expressing LSAMP and losing its novel functions. The true reason behind this paradoxical finding remains as yet undetermined.

The analysis of the present study, albeit deriving from a relatively small cohort and requiring confirmation in a larger study, has pointed to a very interesting and novel association of the IgLON family with ovarian cancer. In order to integrate OPCML and its relatives into models of ovarian carcinogenesis, we need to establish their functions in the normal ovary, their localization within the ovarian environment and the pathway that links them to ovarian cancer. In addition, the functional importance of homo- and heterodimers needs to be addressed. Finally, profiling the expression of the IgLONs in other types of neoplasia might reveal that this hitherto mysterious family has a wider importance in cancer.


   Footnotes
 
Grant support: Cancer Research UK.

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.

3 http://cancerresearchuk.org/aboutcancer/statistics/mortality. Back

Received 11/22/04; revised 5/16/05; accepted 5/26/05.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975–2001, with a special feature regarding survival. Cancer 2004;101:3–27.[CrossRef][Medline]
  2. Runnebaum IB, Stickeler E. Epidemiological and molecular aspects of ovarian cancer risk. J Cancer Res Clin Oncol 2001;127:73–9.[CrossRef][Medline]
  3. Stratton JF, Gayther SA, Russell P, et al. Contribution of BRCA1 mutations to ovarian cancer. N Engl J Med 1997;336:1125–30.[Abstract/Free Full Text]
  4. Wenham RM, Lancaster JM, Berchuck A. Molecular aspects of ovarian cancer. Best Pract Res Clin Obstet Gynaecol 2002;16:483–97.[CrossRef][Medline]
  5. Holschneider CH, Berek JS. Ovarian cancer: epidemiology, biology, and prognostic factors. Semin Surg Oncol 2000;19:3–10.[CrossRef][Medline]
  6. Ozols RF. Update on the management of ovarian cancer. Cancer J 2002;8 Suppl 1:S22–30.
  7. Schofield PR, McFarland KC, Hayflick JS, et al. Molecular characterization of a new immunoglobulin superfamily protein with potential roles in opioid binding and cell contact. EMBO J 1989;8:489–95.[Medline]
  8. Levitt P. A monoclonal antibody to limbic system neurons. Science 1984;223:299–301.[Abstract/Free Full Text]
  9. Struyk AF, Canoll PD, Wolfgang MJ, Rosen CL, D'Eustachio P, Salzer JL. Cloning of neurotrimin defines a new subfamily of differentially expressed neural cell adhesion molecules. J Neurosci 1995;15:2141–56.[Abstract]
  10. Funatsu N, Miyata S, Kumanogoh H, et al. Characterization of a novel rat brain glycosylphosphatidylinositol-anchored protein (Kilon), a member of the IgLON cell adhesion molecule family. J Biol Chem 1999;274:8224–30.[Abstract/Free Full Text]
  11. Hachisuka A, Nakajima O, Yamazaki T, Sawada J. Localization of opioid-binding cell adhesion molecule (OBCAM) in adult rat brain. Brain Res 1999;842:482–6.[CrossRef][Medline]
  12. McNamee CJ, Reed JE, Howard MR, Lodge AP, Moss DJ. Promotion of neuronal cell adhesion by members of the IgLON family occurs in the absence of either support or modification of neurite outgrowth. J Neurochem 2002;80:941–8.[CrossRef][Medline]
  13. Reed J, McNamee C, Rackstraw S, Jenkins J, Moss D. Diglons are heterodimeric proteins composed of IgLON subunits, and Diglon-CO inhibits neurite outgrowth from cerebellar granule cells. J Cell Sci 2004;117:3961–73.[Abstract/Free Full Text]
  14. Sellar GC, Watt KP, Rabiasz GJ, et al. OPCML at 11q25 is epigenetically inactivated and has tumor-suppressor function in epithelial ovarian cancer. Nat Genet 2003;34:337–43.[CrossRef][Medline]
  15. Chen J, Lui WO, Vos MD, et al. The t(1;3) breakpoint-spanning genes LSAMP and NORE1 are involved in clear cell renal cell carcinomas. Cancer Cell 2003;4:405–13.[CrossRef][Medline]
  16. Clark TG, Stewart ME, Altman DG, Gabra H, Smyth JF. A prognostic model for ovarian cancer. Br J Cancer 2001;85:944–52.[CrossRef][Medline]
  17. Auersperg N, Wong AS, Choi KC, Kang SK, Leung PC. Ovarian surface epithelium: biology, endocrinology, and pathology. Endocr Rev 2001;22:255–88.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ntougkos, E.
Right arrow Articles by Sellar, G. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ntougkos, E.
Right arrow Articles by Sellar, G. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Cell Growth & Differentiation