Familial Cancer 2: 9–13, 2003. 2003 Kluwer Academic Publishers. Printed in the Netherlands.
The inframe MSH2 codon 596 deletion is linked with HNPCC and associated with lack of MSH2 protein in tumours Astrid T. Stormorken1, Wolfram Müller2, Annika Lindblom3, Ketil Heimdal1, Steinar Aase4, Inger Marie Bowitz Lothe5, Tove Norèn5, Juul T. Wijnen6, Gabriela Möslein7 and Pål Møller1 1 Section of Genetic Counselling, Department of Cancer Genetics, The Norwegian Radium Hospital, Oslo, Norway; 2 Department of Pathology, Heinrich Heine University, Düsseldorf, Germany; 3 Department of Clinical Genetics, Karolinska Hospital, Stockholm, Sweden; 4 Department of Pathology, Telemark Central Hospital, Skien, Norway; 5 Department of Pathology, Ullevål Hospital, Oslo, Norway; 6 MGC-Department of Human and Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands; 7 Department of Surgery, Heinrich Heine University, Düsseldorf, Germany Received 14 June 2002; accepted in revised form 28 October 2002
Key words: colon cancer, immunohistochemistry, inframe mutation, HNPCC, MSH2
Abstract Hereditary nonpolyposis colorectal cancer (HNPCC) may be caused by germline truncating mutations in DNA mismatch repair (MMR) genes. Whether or not missense or inframe mutations are disease-associated has become a practical clinical problem, because predictive genetic testing is employed to select high-risk persons for clinical examinations. Clinical examinations may reveal polyps to be removed and prevent cancer. One large kindred applying for health care had a N596del mutation in the MSH2 gene. The aim of this study was to determine whether or not the inframe mutation in this family was associated with disease, and to examine the tumours for presence of the MSH2 protein by immunohistochemistry. We demonstrated that the mutation was linked to disease with lod score 5.7 in the family, and all but one examined tumours lacked the MSH2 protein.
Introduction HNPCC may be caused by germline truncating mutations in DNA mismatch repair (MMR) genes [1–6]. The mismatch repair genes encode protein products recognizing and correcting errors that arise when DNA is replicated [3, 7]. It has been assumed that in addition to truncating mutations in the MMR genes, missense mutations and inframe deletions may also cause disease [6, 8, 9]. This has become a practical clinical problem, because demonstration of disease-predisposing mutations is instrumental to select high-risk persons for clinical examinations. Early detection and removal of adenomas or carcinomas may prevent colorectal cancer deaths [10]. We had the opportunity to examine a large HNPCC kindred with a N596del mutation in the MSH2 gene. The same MSH2 inframe deletion has also been reported by others and has been suggested to be causative of disease [11]. The aim of this study was to determine whether or not the inframe deletion in this family was associated
with disease, and to examine the tumours for presence of MSH2 protein.
Materials and methods The pedigree is presented in Figure 1. Four index persons (VI.3, VI.4, VI.16 and VII.11) were referred to the Norwegian Radium Hospital for genetic counselling. Family members contacted us after having been informed by the index persons or other family members, or they were referred by a physician. They were offered genetic counselling and those with a first-degree relative with an HNPCC-related cancer were offered follow-up according to international guidelines [12]. The pedigree was traced backwards and laterally. Site and classification of all cancers and polyps and age at diagnosis were obtained and verified in the medical files. The medical files were obtained with written permission from all living persons, or with permission from decendants. One branch (index person VI.3)
Correspondence to: Pål Møller, MD, PhD, Section of Genetic Counselling, Department of Cancer Genetics, The Norwegian Radium Hospital, N-0130 Oslo, Norway. Tel: +47-22935675; Fax: +47-22935219; E-mail:
[email protected]
Figure 1. Pedigree of the family examined. All spouses and non-affected persons are omitted from the pedigree. IV.3 and IV.4 have common ancestors, not depicted in this pedigree.
A = adenoma. B = breast cancer. Bt = brain tumour. C = colorectal cancer. Cx = cancer of uterine cervix. E = endometrial cancer. G = gastric cancer. H = hyperplastic polyp. M = multiple myeloma. O = ovarian cancer. P = carcinoma of the prostate. S = sarcoma. U = urinary tract cancer. 99 = age unknown. + = mutation positive.
10 A. T. Stormorken et al.
HNPCC caused by one codon deleted in MSH2 fulfilled the Amsterdam I criteria [13], the three others fulfilled the Amsterdam II criteria [14]. The test results from two index persons (VI.3 and VI.16) have previously been reported [6, 8] and included an inframe MSH2 codon 596 deletion. The same two index persons were tested extensively for other mutations with negative results [9]. The mutation was found in what at that time seemed to be two separate families. The family had emigrated from Russia to Finland and further to Norway. Mutation testing was done with informed consent and all information was kept in the medical files. The mutation was tested for in all available blood samples from affected persons (V.14, VI.3, VI.11, VI.16, VII.1, VII.3, VII.4,VII.8, VII.9, VII.10 and VII.11) by sequencing. Formalin-fixed, paraffin-embedded tissue sections from HNPCC-related adenocarcinomas or tubular adenomas were obtained from 10 persons (V.6, VI.3, VI.11, VI.16, VII.4, VII.6, VII.8, VII.9, VII.10 and VII.11) (Figure 1). Immunohistochemistry with monoclonal antibodies against MLH1, MSH2 and MSH6 on tumour samples was performed using standard procedures [15, 16]. Lod score was computed by the MLINK programme under the assumption of no recombination between the mutation (as a genetic marker) and the disease. The computation was done seperately for the two families and the resulting lod-scores then added. For generation I–IV (in whom disease status was lacking), disease status was coded as unknown.
Results Genealogically, the branches of the family were related as shown in Figure 1. As seen, the right branch is apparently not part of the family. However, all persons in the oldest generations lived in a small, isolated community in the forest on the Norwegian–Swedish border. We have conflicting information on exact relations in the oldest generations. The possibility that the mutation had arisen more than once in this small population was not considered. We chose to present the structure giving the lowest calculated lod score. Tumour spectrum was typical of HNPCC as shown in Table 1. The fifth, sixth and seventh generations all had cases of HNPCC-related tumours. In the presented family 10 persons had 2 or more primary neoplasms. The index person VI.3 had 3 metachronous coloncarcinomas, when he was 38 years, 43 years and 52 years old, respectively. He is now 68 years old and has several tubular adenomas and an adenocarcinoma of the prostate. One 44-year-old person (VII.9) was diagnosed with 3 synchronous adenocarcinomas: in the ovary, in the uterus and in the cervix. The pedigree shows two cases of skipped generations. We have information that V.7 was still alive at the age of 88 and that VI.9 died of multiple myeloma at the age of 70. The number of
11 Table 1. Tumours and median age at diagnosis in a family with an MSH2 inframe deletion. Tumour
Number of tumours
Median age (range)
Colon/rectum, total Males Females
24
50 (33–86) 50 (33–80) 50 (42–86)
Endometrium
10
Other, HNPCC related Ovarium Renal pelvis Uterine cervix Brain Stomach
09
15 09
48 (43–73) 04 02 01 01 01
Total 43 (not including tubular adenomas) Tubular adenomas Males Females
48 (30–87) 46.5 (44–53) 76.5 (72–81) 44 31 87 50 (30–87)
11 04 07
54 (31–84) 55 (54–67) 47 (31–84)
tumours and median age at diagnosis are shown in Table 1. No member of the eighth generation is affected, apart from one person with a hyperplastic polyp (VIII.1). Twelve of the seventeen non-affected firstdegree relatives of affected persons in generation VII are under 25 years of age and too young for surveillance. The median age of the remaining 5 is 30 years (range 27–35). The mutation was tested for and found in all available blood samples from affected persons (V.14, VI.3, VI.11, VI.16, VII.1, VII.3, VII.4, VII.8, VII.9, VII.10 and VII.11). The results from immunohistochemistry are shown in Table 2. Eight tumours lacked MSH2 protein and six of these had concurrent loss of staining for MSH6 protein. In two tumours, the immunohistochemical findings were not consistent with the patients’ mutational status. One tubular adenoma from VII.4 did not show loss of staining, and one colon carcinoma from patient VI.11 showed focal staining for all three antibodies. In the index person VI.3, we were able to do immunohistochemistry on both a tumour from the colon and tissue from a bone metastasis from an adenocarcinoma of the prostate. Both tumours showed loss of MSH2 protein and concurrent loss of MSH6 protein. The mutation was linked to the disease with a lod score of 5.7.
Discussion The present study links the disease in the kindred to the MSH2 mutation both by genetic linkage and by lack of MSH2 protein in the tumours. Only affected persons were tested for the mutation, but two cases only of skipped generations indicate that the mutation has a high penetrance. Presence of mutation was associated with
12
A. T. Stormorken et al.
Table 2. Tumours, immunohistochemical and mutation analysis from 13 affected relatives in an MSH2 family. Patient
V.6 V.14 VI.3 VI.11 VI.16 VII.1 VII.3 VII.4 VII.6 VII.8 VII.9 VII.10 VII.11 a b c
Tumour
Colon carcinoma Adenocarcinoma ovary Colon carcinoma Adenocarcinoma of prostate Colon carcinoma Colorectal cancer Tubular adenoma Hyperplastic polyp Tubular adenoma Colorectal cancer Endometroid adenocarcinoma ovary Adenocarcinoma ovary Endometrial carcinoma Endometrial carcinoma
Age at diagnosis
67 48 43 68 51 35 47 26 31 44 45 44 52 45
Expression of MMR proteinsa
MSH2 mutation
MLH1
MSH2
MSH6
+ NT + + +b + NT NT + + + + + +c
0 NT 0 0 +b 0 NT NT + 0 0 0 0 0c
0 NT 0 0 +b NT NT NT + 0 0 0 0c +c
NT Yes Yes Yes Yes Yes Yes Yes NT Yes Yes Yes Yes
As determined by immunohistochemistry: NT = not tested; 0 = abnormal lack of staining; + = staining present. In this patient the immunohistochemistry showed variable staining for all three antibodies, see text. Poor quality of tissue/no normal tissue as positive control.
the usual phenotypic features of the HNPCC syndrome as shown in Table 1 [17]. The demonstrated mutation has been reported several times by others [11] and has been suggested to be causative. The index persons were tested extensively. No other variation was found by analyzing all the MSH2 exons with denaturing gradient gel electrophoresis, solid phase sequencing and Southern blot analysis [8, 9]. The possibility of the inframe deletion acting as a modifying or low penetrant factor interacting with another gene was excluded by lod score, the extensive search for additional mutations, and by the demonstrated lack of MSH2 protein. Nevertheless, the mutation is linked to the disease in the family and may be used for predictive genetic testing. Tumours lacking normal MSH2 protein also lacked the MSH6 protein. This has also been reported [16, 18, 19] and is most likely the result of abrogation of the MutSα complex formed by the MSH2 and MSH6 proteins and acting as a mismatch recognition factor [19]. The complex is required to repair base-pair and single-base insertion/deletion mismatches. It has been shown that MSH2 has two interaction regions with MSH6 [20] and that the MSH6 protein is unstable in the absence of MSH2 [21, 22]. Thus, absence of the MSH6 protein may be expected if no normal MSH2 protein is present. It is not clarified why an inframe deletion predicted to cause a protein short of one amino acid is associated with absence of the whole protein. One might have expected to find normal or increased levels of mutated protein as observed in tumours with distorted p53 protein [23]. Missense mutations causing lack of RNA transcript have been reported [24]. It is possible that the protein was present, but not detected by the immunostaining method applied. But, if so, we would not have expected MSH6 to be lacking as well. In two tumours, immunohistochemical findings were
not consistent with the patients’ mutational status. The adenoma from VII.4 undoubtedly showed presence of all three proteins. Adenomas are not that common at the age of 31 years, but still it might represent a sporadic adenoma in a mutation carrier. It was less than three millimetres and located in the distal colon. Alternatively, the explanation might be that MMR malfunction and loss of protein expression does not initiate adenoma development, but becomes present at an early stage of tumour genesis [25]. The patient’s adenoma might not have reached this stage and this could explain the presence of protein. A colorectal tumour from VI.11 showed focal staining for all three antibodies. This might be explained by technical problems: the tumour block was 20 years old. Microsatellite analysis has not been done, but could have been a help in clarifying this. If the tumour from an index patient was MSI positive and the tumours from VII.4 and VI.11 were MSI stable, the latter might represent sporadic tumours in mutation carriers. Increased sensitivity by the Amsterdam II criteria was demonstrated in this family. All branches fulfilled the Amsterdam II criteria, while only one branch fulfilled the Amsterdam I criteria. In conclusion, the family demonstrated that the N596del mutation in the MSH2 gene is highly penetrant, and confers the usual expressions of disease. Presence or absence of the mutation is now being used for predictive genetic testing in the kindred. Note added in proof After this manuscript had been accepted for publication, another family fulfilling the Amsterdam I criteria was demonstrated to have the same mutation. They immediately confirmed that they originated from the same area, but could not confirm any person in Figure 1 to be relatives.
HNPCC caused by one codon deleted in MSH2 Acknowledgements We are indebted to Eldbjørg Hanslien, Section of Genetic Counselling, The Norwegian Radium Hospital, for technical assistance and to the Departments of Pathology at various Norwegian hospitals for providing the stored paraffin blocks.
References 01. Strand M, Prolla T, Liskay M et al. Destabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature 1993; 365: 274–6. 02. Fishel R, Lescoe MK, Rao MRS et al. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 1993; 75: 1027–38. 03. Leach FS, Nicolaides NC, Papadopoulos N et al. Mutations of a MutS homolog in hereditary non-polyposis colorectal cancer. Cell 1993; 75: 1215–35. 04. Bronner CE, Baker SM, Morrison PT et al. Mutations in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 1994; 368: 258–61. 05. Papadopoulos N, Nicolaides NC, Wei Y et al. Mutation of a mutL homolog in hereditary colon cancer. Science 1994; 263: 1625–9. 06. Wijnen J, de Leeuw W, Vasen H et al. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Genet 1999; 23: 142–4. 07. Dunlop MG, Farrington SM, Carothers AD et al. Cancer risk associated with germline DNA mismatch repair gene mutations. Hum Mol Genet 1997; 6: 105–110. 08. Wijnen JT, Meera Khan P, Vasen H et al. HNPCC families not complying with the Amsterdam criteria show extremely low frequency of mismatch repair gene mutations. Am J Hum Genet 1997; 61; 329–35. 09. Wijnen JT, van der Klift H, Vasen H et al. MSH2 genomic deletions are a frequent cause of HNPCC. Nat Genet 1998; 20: 326–8. 10. Järvinen H, Aarnio M, Mustonen H et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 2000; 118: 829–34.
13 11. http://www.nfdht.nl/database2/msh2-12.htm 12. http://www.nfdht.nl/guidelines.htm 13. Vasen HFA, Mecklin J-P, Meera Khan P et al. The International Collaborative Group on hereditary non-polyposis colorectal cancer (ICG-HNPCC). Dis Colon Rectum 1991; 34: 424–5. 14. Vasen FA, Watson P, Mecklin J-P et al. The International Collaborative Group on hereditary non-polyposis colorectal cancer (ICG-HNPCC). Gastroenterology 1999; 116: 453–1456. 15. Thibodeau SN, French AJ, Rosche PC et al. Altered expression of hMSH2 and hMLH1 in tumours with microsatellite instability and genetic alterations in mismatch repair genes. Cancer Res 1996; 56: 4836–40. 16. de Leeuw WJF, Dierssen JW, Vasen HFA et al. Prediction of a mismatch repair gene defect by microsatelite instability and immunohistochemical analysis in endometrial tumours from HNPCC patients. J Pathol 2000; 192: 328–35. 17. Vasen HF, Stormorken A, Menko FH et al. MSH2 mutation carriers are at higher risk of cancer than MLH1 mutation carriers. A study of 138 HNPCC families. J Clin Oncol 2001; 19: 4074–80. 18. Wagner A, Hendriks Y, Meijers-Heijboer EJ et al. Atypical HNPCC owing to MSH6 germline mutations: analysis of a large Dutch pedigree. J Med Genet 2001; 38: 318–22. 19. Schweizer P, Moiso AL, Kuismanen SA et al. Lack of MSH2 and MSH6 characterizes endometrial but not colon carcinomas in hereditary nonpolyposis colorectal cancer. Cancer Res 2001; 61: 2813–15. 20. Guerette S, Wilson T, Gradia S et al. Interactions of human hMSH2 with hMSH3 and hMSH2 with hMSH6: examination of mutations found in hereditary nonpolyposis colorectal cancer. Mol Cell Biol 1998; 18: 6616–23. 21. Chang DK, Ricciardiello L, Goel A et al. Steady-state regulation of the human DNA mismatch repair system. J Biol Chem 2000; 275: 18424–31. 22. Marra G, Iaccarino I, Lettierei T et al. Mismatch repair deficiency associated with overexpression of the MSH3 gene. Biochemistry 1998; 95: 8568–73. 23. Johnson P, Gray D, Mowat M et al. Expression of wild-type p53 is not compatible with continued growth of p53-negative tumor cells. Mol Cell Biol 1991; 11: 1–11. 24. Lynch E, Ostermeyer E, Lee M et al. Inherited mutations in PTEN that are associated with breast cancer, Cowden disease, and juvenile polyposis. Am J Hum Genet 1997; 61: 1254–60. 25. Rijcken FEM, Hollema H, Kleibeuker JH. Proximal adenomas in hereditary non-polyposis colorectal cancer are prone to rapid malignant transformation. Gut 2002; 50: 382–6.
