Oncogene (2001) 20, 5215 ± 5218 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc
Alterations of repeated sequences in 5' upstream and coding regions in colorectal tumors from patients with hereditary nonpolyposis colorectal cancer and Turcot syndrome Michiko Miyaki*,1,5, Takeru Iijima1,5, Kiyotaka Shiba3, Toshihiko Aki4, Yumi Kita1, Masamichi Yasuno2, Takeo Mori2, Toshio Kuroki5 and Takeo Iwama6 1
Hereditary Tumor Research Project, Tokyo Metropolitan Komagome Hospital, Tokyo 113-8677, Japan; 2Department of Surgery, Tokyo Metropolitan Komagome Hospital, Tokyo 113-8677, Japan; 3Cancer Institute, Tokyo 170-8455, Japan; 4Yamaguchi University, Yamaguchi 755-0805, Japan; 5Institute of Molecular Oncology, Showa University, Tokyo 142-8666, Japan; 6Kyoundo Hospital Sasaki Institute, Tokyo 101-0062, Japan
One of the characteristics of tumors from patients with germline mutations of DNA mismatch repair genes is instability at microsatellite regions (MSI). We analysed alterations at repeated sequences of coding regions, as well as those of 5' upstream regions, in 29 MSI-High colorectal tumors from patients with hereditary nonpolyposis colorectal cancer (HNPCC) and Turcot syndrome. We found that repeated sequences in 5' upstream regions were altered in these tumors, at considerable frequencies. The (A)10 repeat in the promoter region (position 7178*7169) of the GAPDH gene was altered in 17% of the tumors. The (A)10(TA)9 in the 5' upstream region (position 7318*7291) of the mitochondrial isoleucyl tRNA synthetase gene (IleRS-A), coded in nuclear DNA, was altered in 59% of the tumors, whereas (A)9 in the 5' upstream region (position 7859*7851) of cytoplasmic isoleucyl tRNA synthetase gene (IleRS-B) was not altered. Alteration at repeated sequences in the coding regions were 72% at TGFbRII(A)10, 24% at IGFIIR(G)8, 45% at BAX(G)8, 55% at E2F4(CAG)13, 66% at caspase-5 (A)10, 31% at MBD4(A)10, 55% at hMSH3(A)8 and 34% at hMSH6(C)8. The number of altered genes increased with the advancement of carcinoma according to Dukes categories: mean numbers of altered genes within these 10 genes were 2.6 for Dukes A, 4.7 for Dukes B and 7.8 for Dukes C. The mean number for adenomas was 2.0. These results suggest that the MSI phenotype also causes alteration of 5' upstream regions which may aect apoptosis and some mitochondrial functions in HNPCC and Turcot tumors, and that accumulation of altered genes with repeated sequences is associated with the progression of HNPCC and Turcot colorectal tumors. Oncogene (2001) 20, 5215 ± 5218.
*Correspondence: M Miyaki, Hereditary Tumor Research Project, Tokyo Metropolitan Komagome Hospital, 3-18-22 Honkomagome, Bunkyo-ku Tokyo 113-8677, Japan; E-mail:
[email protected] Received 5 February 2001; revised 12 April 2001; accepted 30 April 2001
Keywords: mutation; repeated sequence; 5' upstream region; HNPCC; Turcot; Dukes stage Hereditary nonpolyposis colorectal cancer (HNPCC) syndrome (Lynch and Smyrk, 1996) is characterized by germline mutation in one allele of a DNA mismatch repair gene (Fishel et al., 1993; Leach et al., 1993). We have previously reported, in HNPCC tumors, somatic mutations also occurring in the other allele (Konishi et al., 1996). This causes loss of mismatch repair function, resulting in replication error at microsatellite loci (Aaltonen et al., 1993; Ionov et al., 1993) as well as repeated sequences of coding regions, such as (A)10 of TGFbRII (Markovitz et al., 1995) and (G)8 of BAX (Rampino et al., 1997) genes. HNPCC tumors are assumed to also have alterations at numerous other repeated sequences, instead of less frequent mutation of APC, p53, and K-ras genes (Konishi et al., 1996), and almost no loss of heterozygosity at various chromosomal regions (Thibodeau et al., 1993; Konishi et al., 1996). Alterations of several other coding regions with repeated sequences have been analysed, but, not all target genes of replication error have been revealed. Furthermore, the nature of alteration at 5' upstream or promoter region, which may regulate the transcription of genes, is still unclear. To examine whether these regions are altered in HNPCC tumors, we analysed repeated sequences in the 5' upstream region of three genes. Since we have observed that colorectal tumors from a patient with Turcot syndrome (characterized by brain and colorectal cancer at a young age) showed severe replication error (Miyaki et al., 1997a), we analysed changes of these regions in Turcot tumors. We also made simultaneous analyses of alteration of repeated sequences in eight coding regions, in both HNPCC and Turcot cases, to assess their contribution to carcinogenesis, and the extent of these alterations was compared with the Dukes stage of tumors. Twenty-four primary colorectal tumors, including 22 carcinomas, one adenoma, and one hyperplastic polyp, were obtained from 14 Japanese HNPCC patients, after obtaining informed consent. These patients
Alteration of repeated sequence in HNPCC and Turcot tumor M Miyaki et al
5216
included 10 cases with germline mutations of the hMSH2, hMLH1, and hMSH6 genes (Miyaki et al., 1995, 1997b), and four cases without identi®ed germline mutation but ful®lling the criteria of HNPCC. Five primary colorectal tumors, including three carcinomas and two adenomas, were obtained from a Turcot syndrome patient with germline mutation of hPMS2 (Miyaki et al., 1997a). All HNPCC and Turcot tumors exhibited high instability at microsatellite loci (MSI-H), as previously reported (Konishi et al., 1996; Miyaki et al., 1997a). The promoter region of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene has (A)10 repeat at the position of 7178 * 7169 (Ercolani et al., 1988). Mutation of 1-bp deletion was detected in ®ve of 24 HNPCC tumors, as shown in Figure 1 and Table 1. GAPDH has been known to have various functions, such as glycolytic enzyme and uracil glycosylase, and this protein has been reported to also have activity as a mediator of cell death (Sawa et al., 1997). A change of the number of A repeats at the promoter region has been demonstrated to lower the transcription level of this gene (Aki et al., 1997). Accordingly, it may be possible that HNPCC tumors with alteration of this (A)10 result in resistance to cell death. The mitochondrial isoleucyl t-RNA synthetase gene (IleRS-A) is coded in nuclear DNA, and it has (A)10(TA)9 repeat at the 5' upstream region (position
Figure 1 Examples of alterations at repeated sequences at 5' upstream regions in tumors from patients with HNPCC and Turcot syndrome. (a) GAPDH Promoter(A)10; (b) 5' Upstream(A)10(TA)9 of IleRS-A Oncogene
Table 1 Alterations of repeated sequences at 5' upstream and coding regions in colorectal tumors from patients with HNPCC and Turcot syndrome
Repeated sequence 5' Upstream region GAPDH(A)10 IleRS-A(A)10(TA)9 IleRS-B(A)9 Coding region TGFbRII(A)10 IFGIIR(G)8 BAX(G)8 E2F-4(CAG)13 Caspase-5(A)10 MBD4(A)10 hMSH3(A)8 hMSH6(C)8
No of tumors with alteration/ No of tumors analysed HNPCC Turcot
Total (%)
5/24 15/24 0/24
0/5 2/5 0/5
5/29 (17) 17/29 (59) 0/29 (0)
17/24 6/24 13/24 14/24 16/24 9/24 13/24 8/24
4/5 1/5 0/5 2/5 3/5 0/5 3/5 2/5
21/29 7/29 13/29 16/29 19/29 9/29 16/29 10/29
(72) (24) (45) (55) (66) (31) (55) (34)
DNA samples were ampli®ed by PCR under the same conditions as previously reported (Konishi et al., 1996) and electrophoresed through denaturing or non-denaturing acrylamide gel. DNA fragments in abnormal bands were sequenced by dideoxy termination method. Primers used for ampli®cation of 5' upstream (A)10 of GAPDH gene were: sense 5'-CACACGCTCGGTGCGTGCCCA-3', and antisense 5'-GTGCGCCCGTAAAACCGCTAGTA-3'; for 5' upstream (A)10(TA)9 of IleRS-A gene were: sense 5'-CCCAGATTGAGATGTGCAGT-3', and antisense 5'-GCTCATAGTCCAATCGCTGT-3'; for 5' upstream (A)9 of IleRS-B gene were: sense 5'CATGACCGTGCCACTGCACT-3', and antisense 5'-CCAAAGCCGCTTCTCCTAT-3'. Primers used for ampli®cation of TGFbRII(A)10, BAX(G)8, IGFIIR(G)8, E2F-4(CAG)13, caspase5(A)10, MBD4(A)10, hMSH3(A)8 and hMSH6(C)8 regions were the same as those previously reported (Konishi et al., 1996; Rampino et al., 1997; Souza et al., 1996; Yoshitaka et al., 1996; Schwartz et al., 1999; Riccio et al., 1999; Malkhosyan et al., 1996)
7318 * 7291) (Shiba et al., 1994). This repeat was found to be altered in 15 of 24 HNPCC, and in two of ®ve Turcot, tumors. Alteration occurred at (A)10 as well as (TA)9, several tumors showing alteration to (A)9(TA)8. Some tumors with alteration showed absence of wild-type allele; examples being shown in Figure 1. However, (A)9 repeat at the 5' upstream region (position 7859 * 7851) of the cytoplasmic isoleucyl t-RNA synthetase gene (IleRS-B), which is also coded in nuclear DNA (Shiba et al., 1994), was not altered in HNPCC or Turcot tumors. The reason for less mutability at (A)9 of IleRS-B is unclear, despite that (A)8 in coding regions of the hMSH3 gene was altered at considerable frequencies in these tumors. One possible explanation is that alteration of this (A)9 region of IleRS-B has no in¯uence on tumorigenesis in HNPCC and Turcot patients. High mutability at (A)10(TA)9 of IleRS-A suggests that alteration of 5' upstream of this gene has a certain contribution to tumorigenesis. However, since function of the 5' upstream region of IleRS-A in relation to gene expression has not been con®rmed yet, it will be necessary to examine whether the alteration of expression level of mitochondrial-type aminoacyl tRNA is caused by mutation of the 5' upstream. To compare the extent of alterations in 5' upstream regions with those in coding regions, we also analysed
Alteration of repeated sequence in HNPCC and Turcot tumor M Miyaki et al
alterations of repeated sequences of coding regions in HNPCC and Turcot tumors (Table 1). Deletions of 1 ± 2 bp and insertions of 1 bp at TGFbRII(A)10 (Markoviz et al., 1995) were detected in 17 of 24 HNPCC, and in four of ®ve Turcot, tumors, and the majority of cases with this mutation had no wild-type allele. Since TGFbRII mediates growth inhibition of epithelial cells by TGF-b, loss of receptor II function by gene mutation appears to result in a growth advantage in a signi®cant percentage of colorectal tumors from HNPCC and Turcot patients. Alteration at (G)8 of the IGFIIR gene (Souza et al., 1996), which has a function to activate TGF-b, was found in 24% of our tumor samples. Frameshift mutations of BAX(G)8 (Rampino et al., 1997) were detected in 13 of 24 of our HNPCC carcinomas, but not at all in the Turcot carcinomas, two of Turcot carcinomas having p53 mutations. Alternative exclusion was observed between BAX and p53 mutation in both HNPCC and Turcot carcinomas, which was consistent with the ®nding that p53 is a direct transcriptional activator of BAX (Miyashita and Reed, 1995). Such inactivating mutation of the proapoptotic protein BAX appears to have the same role in HNPCC carcinogenesis as that of inactivation of p53 in non-HNPCC carcinogenesis (conversion from adenoma to carcinoma). A growth control gene, E2F4, has a (CAG)13 repeat in the coding region, and this repeat has been found to be altered in sporadic gastrointestinal tumors with microsatellite instability (Yoshitaka et al., 1996). In HNPCC tumors, 14 of 24 exhibited 3-, 6-, and 9-bp deletions, and some tumors with mutation had no wild-type allele. Two of three Turcot carcinomas also showed deletion mutations. Such a high frequency of alteration of E2F-4 suggests that this gene is a target of genetic instability in HNPCC and Turcot tumors, although in¯uence of this alteration on the function of E2F-4 has yet to be Table 2
clari®ed. Recently, (A)10 repeat in the caspase-5 gene has been observed in sporadic endometrial, colon, and stomach tumors with microsatellite instability (Schwartz et al., 1999). In the present study, alterations of this region were detected in 16 of 24 HNPCC, and three of ®ve Turcot, colorectal tumors. Almost all alterations were 1-bp deletion, and four tumors had 2bp deletion and 1-bp insertions. Caspase-5 has been demonstrated to cleave pro-caspase-3, resulting in activation of caspase-3, which is involved in apoptotic cell death (Kamada et al., 1997). This suggests that alteration of caspase-5 may give a growth advantage to cancer cells through inhibition of apoptosis. The (A)10 tract of the DNA repair gene MBD4 has been reported to be altered in tumors with MSI (Riccio et al., 1999). In our cases, this region was altered in nine of 24 HNPCC tumors, suggesting that genomic instability may be increased in these HNPCC tumors. In addition to inactivation of causative mismatch repair genes, 13 of 24 HNPCC tumors and three of ®ve Turcot tumors had 1 ± 2 bp deletion and 1 bp insertion at (A)8 of the hMSH3 gene (Malkhosyan et al., 1996). Similar frameshift mutations were detected at (C)8 of the hMSH6 gene (Malkhosyan et al., 1996) in eight of 24 of our HNPCC, and two of our ®ve Turcot, tumors. Alteration in these secondary genes, which are components of the mismatch repair complex, may cause more severe replication error in HNPCC and Turcot tumors, although these alterations contribute indirectly in tumorigenesis. Table 2 shows alterations of genes with repeated sequences with respect to tumor progression through Dukes A to C. Although all tumors showed MSI-High, the extent of altered genes was notably higher in Dukes C stage compared with Dukes A and B. Mean numbers of altered genes within 10 genes were estimated from this table as 2.6 for Dukes A, 4.7 for Dukes B and 7.8
5217
Examples of alterations at repeated sequences in 5' upstream and coding regions in colorectal tumors from HNPCC and Turcot patients
Tumor HNP10 Ad2 Turcot Ad1 Turcot Ad3 HNP10 Ca1 HNP6 Ca HNP16 Ca2 HNP16 Ca4 HNP5 Ca1 HNP5 Ca3 Turcot Ca1 HNP5 Ca2 HNP13 Ca HNP14 Ca1 HNP1 Ca HNP3 Ca HNP7 Ca HNP8 Ca
Age at diagnosis 47 18 18 47 27 52 52 43 43 18 43 52 46 44 48 35 43
Site
Dukes stage
MSI
Rectum Adenoma 4/5 S-colon Adenoma 4/5 S-colon Adenoma 4/5 D-colon A 5/5 A-colon A 3/5 D-colon A 3/5 Rectum A 5/5 T-colon A 5/5 T-colon A 5/5 S-colon A 5/5 A-colon B 5/5 T-colon B 4/5 S-colon B 5/5 A-colon C 5/5 D-colon C 5/5 A-colon C 5/5 A-colon C 4/5
GAPDH IleRSA TGF-bRII IGFIIR BAX E2F-4 Caspase-5 MBD4 hMSH3 hMSH6 (A)10 (A)10(TA)9 (A)10 (G)8 (G)8 (CAG)13 (A)10 (A)10 (A)8 (C)8 7 7 7 7 7 7 7 7 7 7 7 7 7 + + + 7
7 7 7 7 7 7 7 + + 7 7 + + + + 7 +
+ + + 7 7 + 7 + 7 + 7 + 7 + + + +
7 7 7 7 7 7 7 7 7 7 7 + 7 7 7 + +
7 7 7 7 7 7 + 7 7 7 + + + + + 7 +
7 7 7 + + 7 7 + 7 7 + 7 + + + + +
7 7 7 7 + 7 7 + + + 7 + + + + + +
7 7 7 7 7 7 7 7 + 7 7 7 + 7 + + 7
+ + 7 7 7 7 7 + + + + 7 7 + + + 7
7 + 7 7 7 7 7 7 7 + + 7 7 7 + + +
This table includes carcinomas in which Dukes stages could be determined, and adenomas. Tumor column indicates patient number followed by the tumor number. A: ascending; D: descending; S: sigmoid; T: transverse. +: repeated sequence was altered; 7: not altered; MSI indicates number of altered microsatellite loci/number of loci analysed (as previously reported, Konishi et al., 1996) Oncogene
Alteration of repeated sequence in HNPCC and Turcot tumor M Miyaki et al
5218
for Dukes C. In the case of adenoma the mean number was 2.0. These data suggest that it is possible to observe gradual development of HNPCC carcinomas by the number of altered genes with repeated sequence. None of 30 samples of MSI-negative sporadic colorectal carcinomas showed alteration of repeated sequences within these 10 genes. Genetic changes involved in carcinogenesis of HNPCC and Turcot syndrome appear to be dierent from those in the usual adenoma-carcinoma sequence (Konishi et al., 1996). The present results suggest that genetic changes occurring in repeated sequences at 5' upstream regions of certain genes, besides repeats in coding regions, are targets of replication error. More-
over, gradual accumulation of alteration of genes with repeated sequences may be involved in progression of HNPCC carcinomas. Repeated sequences at both 5' upstream and coding regions in more genes should be analysed, to further clarify the mechanism of development and/or progression of colorectal cancer in HNPCC and Turcot syndrome. Acknowledgments This work was supported in part by the Project `HighTechnology Research Center' from Ministry of Education, Science, Sport and Culture of Japan, and Smoking Research Foundation.
References Aaltonen LA, PetromaÈki P, Leach FS, Sistonen P, PylkkaÈnen L, Mecklin J-P, JaÈrvinen H, Powell SM, Jen J, Hamilton SR, Petersen GM, Kinzler KW, Vogelstein B and de la Chapelle A. (1993). Science, 260, 812 ± 816. Aki T, Yanagisawa S and Akanuma H. (1997). J. Biochem., 122, 271 ± 278. Ercolani L, Florence B, Denaro M and Alexander M. (1988). J. Biol. Chem., 263, 15335 ± 15341. Fishel R, Lescoe MK, Rao MRS, Copeland NG, Jenkins NA, Garber J, Kane M and Kolodner G. (1993). Cell, 75, 1027 ± 1038. Ionov Y, Peinado MA, Malkosyan S, Shibata D and Perucho M. (1993). Nature, 363, 558 ± 561. Kamada S, Funahashi Y and Tsujimoto Y. (1997). Cell Death Dier., 4, 473 ± 478. Konishi M, Kikuchi-Yanoshita R, Tanaka K, Muraoka M, Onda A, Okumura Y, Kishi N, Iwama T, Mori T, Koike M, Ushio K, Chiba M, Nomizu S, Konishi F, Utsunomiya J and Miyaki M. (1996). Gastroenterology, 111, 307 ± 317. Leach FS, Nicolaidis NC, Papadopoulos N, Liu B, Jen J, Parsons R, Peltomaki P, Sistonen P, Aaltonen LA, Nystrom-Lahti M, Guan XY, Zang Z, Meltzer PS, Yu JW, Kao FT, Chen DJ, Cerosalitti KM, Fournier REK, Todd S, Lewis T, Leach RJ, Naylor SL, Weissenbach J, Mecklin JP, Jarvinen H, Petersen GM, Hamilton SR, Green J, Jass J, Watson P, Lynch HT, Trent JM, de la Chapelle A, Kinzler KW and Vogelstein B. (1993). Cell, 75, 1215 ± 1225. Lynch HT and Smyrk T. (1996). Cancer, 78, 1149 ± 1167. Malkhosyan S, Rampino N, Yamamoto H and Perucho M. (1996). Nature, 382, 499 ± 500. Markovitz S, Wang J, Myero L, Parsons R, Sun L-Z, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B, Brattain M and Willson JKV. (1995). Science, 268, 1336 ± 1338. Miyaki M, Konishi M, Muraoka M, Kikuchi-Yanoshita R, Tanaka K, Iwama T, Mori T, Koike M, Ushio K, Chiba M, Nomizu S and Utsunomiya J. (1995). J. Mol. Med., 73, 515 ± 520.
Oncogene
Miyaki M, Nishio J, Konishi M, Kikuchi-Yanoshita R, Tanaka K, Muraoka M, Nagato M, Chong J-M, Koike M, Terada T, Kawahara Y, Fukutome A, Tomiyama J, Chuganji Y, Momoi M and Utsunomiya J. (1997a). Oncogene, 15, 2877 ± 2881. Miyaki M, Konishi M, Tanaka K, Kikuchi-Yanoshita R, Muraoka M, Yasuno M, Igari T, Koike M, Chiba M and Mori T. (1997b). Nat. Genet., 17, 271 ± 272. Miyashita T and Reed JC. (1995). Cell, 80, 293 ± 299. Rampino N, Yamamoto H, Ionov Y, Li Y, Wawai H, Reed JC and Perucho M. (1997). Science, 275, 967 ± 969. Riccio A, Aaltonen LA, Godwin AK, Loukola A, Peresepe A, Salovaara R, Masciullo V, Genuardi M, ParavatouPetsotas M, Bassi DE, Ruggeri BA, Klein-Szanto AJP, Testa JR, Neri G and Ballacosa A. (1999). Nature Genetics, 23, 266 ± 268. Sawa A, Khan AA, Hester LD and Synder SH. (1997). Proc. Natl. Acad. Sci. USA, 94, 11669 ± 11674. Schwartz Jr S, Yamamoto H, Navarro M, Maestro M, Reventos J and Perucho M. (1999). Cancer Res., 59, 2995 ± 3002. Shiba K, Suzuki N, Shigesada K, Namba Y, Shimmel P and Noda T. (1994). Proc. Natl. Acad. Sci. USA, 91, 7435 ± 7439. Souza RF, Apple R, Yin J, Wang S, Smolinski KN, Abraham JM, Zou T-T, Shi Y-Q, Lei J, Cottrell J, Cymes K, Biden K, Simms L, Leggett B, Lynch PM, Frazier M, Powell SM, Harpaz N, Sugimura H, Young J and Meltzer SJ. (1996). Nat. Genet., 14, 255 ± 257. Thibodeau SN, Bren G and Schaid D. (1993). Science, 260, 816 ± 819. Yoshitaka T, Matsubara N, Ikeda M, Tanino M, Hanafusa H, Tanaka N and Shimizu K. (1996). Biochem. Biophys. Res. Commun., 227, 553 ± 557.
