1 Istituto Chimica Biomolecolare-Sezione di Sassari, C.N.R., Tramariglio, Alghero; 2 Istituto di Anatomia Patologica, Università di Sassari, Sassari; 3 Istituto Nazionale Tumori G. Pascale, Napoli; 4 Chirurgia Generale II, Università di Cagliari, Cagliari; 5 Chirurgia Generale e Vascolare e Oncologia Clinica, Università di Bari, Bari; 6 Anatomia Patologica, Ospedali Riuniti, Foggia; 7 Medicina Nucleare, Università di Sassari, Viale San Pietro, Sassari, Italy
Received 12 February 2003; revised 7 April 2003; accepted 15 April 2003
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Microsatellite instability (MSI) is due to defective DNA mismatch repair (MMR) and has been detected at various rates in colorectal carcinoma (CRC). The role of MSI in colorectal tumorigenesis was assessed further in this study by both microsatellite analysis of two CRC subsets [unselected patients (n = 215) and patients <50 years of age (n = 95)], and mutation screening of the two major MMR genes MLH1 and MSH2 among familial CRC cases.
Patients and methods:
PCR-based microsatellite analysis was performed on paraffin-embedded tissues. In CRC families, MLH1/MSH2 mutation analysis and MLH1/MSH2 immunostaining were performed on germline DNA and MSI+ tumour tissues, respectively.
Results:
The MSI+ phenotype was detected in 75 (24%) patients, with higher incidence in early-onset or proximally located tumours. Among 220 patients investigated for family cancer history, MSI frequency was markedly higher in familial [18/27 (67%)] than in sporadic [32/193 (17%)] cases. Three MLH1 and six MSH2 germline mutations were identified in 14 out of 36 (39%) CRC families. Prevalence of MLH1/MSH2 mutations in CRC families was significantly increased by the presence of: (i) fulfilled Amsterdam criteria; (ii) four or more CRCs; or (iii) one or more endometrial cancer. While MSH2 was found mostly mutated, almost all [8/9 (89%)] familial MSI+ cases with loss of the MLH1 protein were negative for MLH1 germline mutations.
Conclusions:
Both genetic (for MSH2) and gene-silencing (for MLH1) alterations seem to be involved in CRC pathogenesis.
Key words: colorectal cancer, gene inactivation, microsatellite instability, MLH1/MSH2 germline mutations
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Tumours with a non-functional DNA mismatch repair (MMR) display a genomic instability, as inferred by detection of ubiquitous somatic variation in the length of microsatellite sequences [2, 5]. Microsatellite instability (MSI) is characterized by small insertions or deletions within short tandem repeats in tumour DNA when compared with the corresponding normal DNA. MSI has been demonstrated in patients with hereditary non-polyposis colorectal carcinoma (HNPCC), an inherited cancer syndrome that also predisposes to endometrial cancer, which represents the most common extracolonic malignancy in HNPCC families [6]. Frequency of MSI has been reported ranging from 8% to 18% in unselected CRCs, and from 70% to 90% in tumours from HNPCC patients [710]. Genetic instability seems to decrease moving from carcinomas originating in the proximal colon to those located in distal colon and rectum, where MSI frequency has been found at a very low rate (in some reports it was as little as
2%) [11, 12].
MSI has been associated with the presence of mutations in the two principal MMR genes, MLH1 and MSH2 (although MLH1 is altered most frequently) [13]. Patients with CRC tumours displaying MSI have a greater chance of presenting with a family history positive for colorectal cancer and/or synchronous CRCs [9, 10], strongly suggesting a hereditary cancer predisposition (MSI is considered a molecular feature particularly frequent in familial CRC) [12, 13].
The prognostic value of MSI in colorectal cancer is highly debated. Several studies have suggested a favourable outcome for patients with the MSI+ phenotype in colorectal and gastric cancer [10, 14, 15], whereas some recent results on MSI+ CRC patients have demonstrated a better survival only for cases with Dukes B and C disease stages, receiving adjuvant chemotherapy (survivals among untreated patients were found to be similar in the two groups with and without the MSI+ phenotype) [11, 16].
To investigate further the prevalence of MSI in CRC, as well as the existence of any clinicopathological difference in CRC patients with and without MSI that might be representative of distinctive tumour behaviour, a PCR-based microsatellite analysis was performed on archival tissues from patients with CRC at various stages of disease and originating from different regions of southern Italy. The incidence of CRC in this geographical area has been reported as being similar to that of other Western countries, with 70 new cases per year per 100 000 inhabitants [17]. Statistical correlation between the presence of MSI and histopathological or clinical parameters was thus inferred. To evaluate the prevalence and spectrum of MLH1 and MSH2 germline mutations in this area, CRC families with at least three affected members underwent mutation analysis of both of these two MMR genes.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
All patients for which family history was investigated were informed about the aims of the study; those on whom blood sampling was performed gave written informed consent.
Tissue samples and DNA extraction
Tissue samples from CRC patients were collected at the Departments and Institutes of Pathology participating in the study. Patients originated from both Sardinia and other geographical areas of southern Italy (in both cases, the incidence of CRC is similar to that found in other Western countries [17]). Family history of cancer was evaluated by questionnaire interviews with CRC patients after initial surgical treatment. Patients were classified as sporadic when no significant evidence of CRC tumours in first- and second-degree relatives was registered. Cases were classified as familial when at least three family members were affected by CRC. Presence of HNPCC syndrome was ascertained on the basis of the Amsterdam criteria, as follows: (i) three relatives with CRC, one of whom is a first-degree relative of the other two; (ii) CRC present in at least two generations; (iii) CRC diagnosed in at least one relative aged <50 years; and (iv) absence of familial adenomatous polyposis [9].
Genomic DNA was isolated from tumour samples and corresponding normal tissues, as described previously [18]. Tumour samples were estimated to contain at least 80% neoplastic cells by light microscopy.
Microsatellite analysis
Primers used to amplify simple sequence repeats at the five marker loci were as reported previously by a panel of experts [19]. The presence of MSI (referred to as MSI+) was defined by detection of at least two unstable (due to deletions or insertions) microsatellite markers in tumour DNA when compared with normal DNA. For each marker, PCR was carried out as described previously [18], using fluorescence-labelled primers. PCR products were analysed on a 377 Sequencer (Applied Biosystems, Foster City, CA).
Mutation analysis of the MLH1/MSH2 genes
Mutation analysis was performed on DNA from peripheral blood samples of familial CRC cases. Primers sequences for MLH1/MSH2 genes were as reported previously [20]. PCR products corresponding to all exons and intronexon boundaries of these two MMR genes were analysed by denaturing high-performance liquid chromatography (DHPLC), as described previously [20]. All PCR products with abnormal DHPLC profiles were sequenced on the ABI3100 Automated Sequencer (Applied Biosystems) [20].
To assess whether MLH1/MSH2 germline variants detected by sequencing were mutations or polymorphisms, 53 unrelated normal individuals (corresponding to 106 chromosomes), originating from the same geographical area and with no family history for cancer, were used as controls.
Expression analysis of the MLH1/MSH2 genes
Expression of the MLH1/MSH2 proteins was detected by immunohistochemistry (IHC) on 2-µm sections of formalin-fixed, paraffin-embedded tissues, following previously described protocols [21]. Staining was evaluated semiquantitatively, using normal epithelial cells or the centres of lymphoid follicles as internal controls. IHC scoring was performed by at least two investigators.
Statistical analysis
Differences in dichotomous variables were evaluated using Pearsons 2 test or by Fishers exact test. The exact coefficient for sample proportion analysis was performed to determine all significant parameters (below the P = 0.05 level). All analyses were performed using the statistical package SPSS/7.5 for Windows.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
Molecular analysis of MLH1/MSH2 genes in familial CRC cases
DNA from 44 CRC patients as probands among 36 families with at least three affected members (including 16 families from the aforementioned population screening and 20 newly selected families originating from the same geographical area) was analysed for germline mutations of the MLH1 and MSH2 genes. All coding regions and splice boundaries of these two genes were screened by DHPLC analysis; PCR products with abnormal denaturing profiles compared with normal controls were sequenced using an automated approach.
As shown in Table 3, three MLH1 and six MSH2 germline mutations were detected in our CRC families (one family proband presented with two MSH2 germline mutations: Val342Ile and IVS6+1GT). With the exception of the MSH2-Gly322Asp, previously described as a low penetrance variant [22], none of the remaining mutations has been reported previously in the database (missense variants present unknown functional significance due to the lack of any information about the effects on protein function). Altogether, 14 (39%) families were found to be positive for germline mutations in these two MMR genes [four (11%) in MLH1 and 10 (28%) in MSH2]. In CRC families, positivity to the Amsterdam criteria, or the presence of at least four CRC cases or association with endometrial cancer (at least one family member with it) were all factors significantly increasing the prevalence of MLH1/MSH2 germline mutations (Table 4).
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Using a reference subset of five polymorphic markers (two mononucleotide repeats and three dinucleotide repeats), DNA from paraffin-embedded tissues of 310 CRC patients was investigated for genome-wide MSI. As recommended by a panel of experts [19], alterations in two or more locus markers were needed to identify tumours with high MSI (also referred to as MSI+ phenotype). Altogether, the presence of MSI was detected in 75 (24%) CRC cases.
The frequency of this genetic alteration in the CRC population in southern Italy is slightly higher than that reported by others for unselected CRCs (usually percentages range from <10% to 18% [7, 11, 12]). However, our findings are strongly consistent with the previously reported data regarding a higher MSI frequency in CRC patients with earlier onset age (45 years) or proximal tumour location [2, 4]. Indeed, there is increasing evidence that proximal and distal colon tumours comprise distinct diseases, with marked differences in the expression of tumour-associated proteins as well as in the level of genetic alterations [23]. Conversely, no significant correlation between MSI and stage of disease was observed, suggesting that such genetic alteration may occur before tumour dissemination, and may also play a role in the first stages of CRC development.
Microsatellite instability has been documented at various rate in many tumour types, suggesting that the presence of MSI might be a marker for a tendency for replication errors in human cancers [24, 25]. In our series, the probability of detecting MSI seems to be increased by the presence of familial recurrence of CRC (17% among patients with less than three affected members versus 67% among patients with three or more CRC cases in a family). Considering the Amsterdam criteria, 11/15 (73%) HNPCC cases tested showed MSI (this is consistent with previously reported rates, varying from 70% to >90% [710]).
A large proportion of CRC families have been demonstrated to harbor both germline mutations of the MLH1/MSH2 genes (together accounting for 8085% of all mutations in HNPCC families [26]) and altered somatic immunoexpression of these two MMR genes (loss of the MLH1 protein has been described in 75% of the MSI+ tumours [8, 21]). In our series, percentages were markedly lower: 12/23 (52%) HNPCC families and 11/21 (52%) MSI+ tumours presented germline mutations and lack of expression of the MLH1/MSH2 genes, respectively. Even taking into consideration some limitations of the screening strategy (e.g. DHPLC analysis may not detect large deletions), the frequency of MLH1/MSH2 germline mutations in our CRC families was still lower than expected.
The detection of a much higher fraction of MSI+ familial CRCs presenting loss of MLH1 protein compared with those harboring a MLH1 mutation [nine (43%) versus one (5%) out of the 21 MSI+ tumours tested, respectively] strongly suggests that either genetic alterations not affecting the coding portion (i.e. within the regulatory regions) or epigenetic modifications may account for the MLH1 inactivation. In this regard, additional data from our group indicated that about one-fifth of MSI+ cases with tumour loss of the MLH1 protein and absence of MLH1 germline mutations presented MLH1 promoter hypermethylation at the somatic level using a methylation-specific PCR assay [27].
Our findings seem to support the criticisms of the general idea that inactivation of MLH1 in colorectal cancer is prevalently due to gene mutations in familial MSI+ cases and to epigenetic changes (mostly promoter hypermethylation) in sporadic MSI+ cases [25, 28]. Conversely, inactivation of the MSH2 gene seems to be strictly correlated with the presence of genetic mutations in its coding regions. Finally, the remaining familial cases presenting neither MLH1 nor MSH2 inactivations may have alterations in other known MMR genes (such as MSH6 or PMS2) or new genes, which await identification.
Considering the correlation with clinical parameters, preliminary data seem to indicate a favourable outcome in patients with MSI compared with those without MSI (not shown). Previous investigations have demonstrated a predictive value as prognostic factor for MSI in colorectal cancer, although in some cases a significant survival advantage has been observed only in MSI+ patients receiving chemotherapy [16, 29]. In this regard, an adequately powered, prospective trial should be performed to understand better the clinical behaviour of MSI+ tumours, as well as to define the real effect of MSI on prognosis.
![]() |
Acknowledgements |
---|
![]() |
Footnotes |
---|
Correspondence to: Dr G. Palmieri, Istituto Chimica Biomolecolare-Sezione di Sassari, C.N.R., Località Tramariglio-Alghero, 07040 Santa Maria La Palma, Sassari, Italy. Tel: +39-079-946706; Fax: +39-079-3961036; E-mail: gpalmieri{at}yahoo.com
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Ilyas M, Straub J, Tomlinson IP, Bodmer WF. Genetic pathways in colorectal and other cancers. Eur J Cancer 1999; 35: 19862002.[CrossRef][ISI][Medline]
3. Kinzler KW, Vogelstein B. Landscaping the cancer terrain. Science 1998; 280: 10361037.
4. Grady WM, Markowitz S. Genomic instability and colorectal cancer. Curr Opin Gastroenterol 2000; 16: 6267.[CrossRef][ISI]
5. Olschwang S, Hamelin R, Laurent-Puig P et al. Alternative genetic pathways in colorectal carcinogenesis. Proc Natl Acad Sci USA 1997; 94: 1212212127.
6. Salovaara R, Loukola A, Kristo P et al. Population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol 2000; 18: 21932200.
7. Percesepe A, Borghi F, Menigatti M et al. Molecular screening for hereditary nonpolyposis colorectal cancer: a prospective, population-based study. J Clin Oncol 2001; 19: 39443950.
8. Kuismanen SA, Holmberg MT, Salovaara R et al. Genetic and epigenetic modification of MLH1 accounts for a major share of microsatellite-unstable colorectal cancers. Am J Pathol 2000; 156: 17731779.
9. Peltomaki P, de la Chapelle A. Mutations predisposing to hereditary nonpoliposis colorectal cancer. Adv Cancer Res 1997; 71: 93119.[ISI][Medline]
10. Gonzalez-Garcia I, Moreno V, Navarro M et al. Standardized approach for microsatellite instability detection in colorectal carcinomas. J Natl Cancer Inst 2000; 92: 544549.
11. Elsaleh H, Shannon B, Iacopetta B. Microsatellite instability as a molecular marker for very good survival in colorectal cancer patients receiving adjuvant chemotherapy. Gastroenterology 2001; 120: 13091310.[ISI][Medline]
12. Nilbert M, Planck M, Fernebro E et al. Microsatellite instability is rare in rectal carcinomas and signifies hereditary cancer. Eur J Cancer 1999; 35: 942945.[CrossRef][ISI][Medline]
13. Debniak T, Kurzawski G, Gorski B et al. Value of pedigree/clinical data, immunohistochemistry and microsatellite instability analyses in reducing the cost of determining hMLH1 and hMSH2 gene mutations in patients with colorectal cancer. Eur J Cancer 2000; 36: 4954.[CrossRef][ISI][Medline]
14. Leung SY, Yuen ST, Chung LP et al. hMLH1 promoter methylation and lack of hMLH1 expression in sporadic gastric carcinomas with high-frequency microsatellite instability. Cancer Res 1999; 59: 159164.
15. Ward R, Meagher A, Tomlinson I et al. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut 2001; 48: 821829.
16. Hemminki A, Mecklin JP, Jarvinen H et al. Microsatellite instability is a favorable prognostic indicator in patients with colorectal cancer receiving chemotherapy. Gastroenterology 2000; 119: 921928.[ISI][Medline]
17. Budroni M, Tanda F. Tumor Registry of Sassari: I tumori in Sardegna negli anni novanta. Sassari, Italy: Tipografia Moderna, Inc. 1998.
18. Palmieri G, Manca A, Cossu A et al. Microsatellite analysis at 10q25-q26 in Sardinian patients with sporadic endometrial carcinoma: identification of specific patterns of genetic alteration. Cancer 2000; 89: 17731782.[CrossRef][ISI][Medline]
19. Boland CR, Thibodeau SN, Hamilton SR et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998; 58: 52485257.[Abstract]
20. Baldinu P, Cossu A, Manca A et al. Microsatellite instability and mutation analysis of candidate genes in unselected sardinian patients with endometrial carcinoma. Cancer 2002; 94: 31573168.[CrossRef][ISI][Medline]
21. Chaves P, Cruz C, Lage P et al. Immunohistochemical detection of mismatch repair gene proteins as a useful tool for the identification of colorectal carcinoma with the mutator phenotype. J Pathol 2000; 191: 355360.[CrossRef][ISI][Medline]
22. Liu B, Nicolaides NC, Markowitz S et al. Mismatch repair gene defects in sporadic colorectal cancers with microsatellite instability. Nat Genet 1995; 9: 4855.[ISI][Medline]
23. McKay JA, Rooney PH, Ross VG et al. Marked differences in tumor-associated protein expression and genetic stability between proximal and distal colon tumours. Eur J Cancer 2001; 37 (Suppl 6): S272.
24. Lengauer C, Kinzier KW, Volgestein B. Genetic instabilities in human cancers. Nature 1998; 396: 643649.[CrossRef][ISI][Medline]
25. Sturzeneker R, Bevilacqua RA, Haddad LA et al. Microsatellite instability in tumors as a model to study the process of microsatellite mutations. Hum Mol Genet 2000; 9: 347352.
26. Heinimann K, Scott RJ, Buerstedde JM et al. Influence of selection criteria on mutation detection in patients with hereditary nonpolyposis colorectal cancer. Cancer 1999; 85: 25122518.[CrossRef][ISI][Medline]
27. Strazzullo M, Cossu A, Baldinu P et al. High-resolution methylation analysis of hMLH1 promoter in sporadic endometrial and colorectal carcinomas. Cancer 2003; In press.
28. Kuismanen SA, Holmberg MT, Salovaara R et al. Epigenetic phenotypes distinguish microsatellite-stable and -unstable colorectal cancers. Proc Natl Acad Sci USA 1999; 96: 1266112666.
29. Elsaleh H, Powell B, McCaul K et al. P53 alteration and microsatellite instability have predictive value for survival benefit from chemotherapy in stage III colorectal carcinoma. Clin Cancer Res 2001; 7: 13431349.