Cancer Research Cancer Health Disparities Conference 2009 SU2C
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 Meeting Abstracts Online
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plaschke, J.
Right arrow Articles by Schackert, H. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Plaschke, J.
Right arrow Articles by Schackert, H. K.
[Cancer Research 64, 864-870, February 1, 2004]
© 2004 American Association for Cancer Research

Regular Articles

Loss of MSH3 Protein Expression Is Frequent in MLH1-Deficient Colorectal Cancer and Is Associated with Disease Progression1

Jens Plaschke1, Stefan Krüger1, Birgit Jeske1, Franz Theissig2, Friedmar R. Kreuz3, Steffen Pistorius4, Hans D. Saeger4, Ingram Iaccarino5, Giancarlo Marra5 and Hans K. Schackert1

Department of 1 Surgical Research, 2 Institute of Pathology, 3 Institute of Clinical Genetics, and 4 Department of Visceral, Thoracic and Vascular Surgery, Carl Gustav Carus Hospital, Dresden University of Technology, Dresden, Germany, and 5 Institute of Molecular Cancer Research, University of Zurich, Zurich, Switzerland


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Mononucleotide repeat sequences are particularly prone to frameshift mutations in tumors with biallelic inactivation of the mismatch repair (MMR) genes MLH1 or MSH2. In these tumors, several genes harboring mononucleotide repeats in their coding region have been proposed as targets involved in tumor progression, among which are also the MMR genes MSH3 and MSH6. We have analyzed the expression of the MSH3 and MSH6 proteins by immunohistochemistry in 31 colorectal carcinomas in which MLH1 was inactivated. Loss of MSH3 expression was identified in 15 tumors (48.5%), whereas all tumors expressed MSH6. Frameshift mutations at coding microsatellites were more frequent in MSH3 (16 of 31) than in MSH6 (3 of 31; Fisher’s exact test, P < 0.001). Frameshift mutations and allelic losses of MSH3 were more frequent in MSH3-negative tumors compared with those with normal expression (22 mutations in 30 alleles versus 8 mutations in 28 alleles; {chi}2, P = 0.001). Biallelic inactivation was evident or inferred for 60% of MSH3-negative tumors but none of the tumors with normal MSH3 expression. In contrast, we did not identify frameshift mutations in the (A)8 tract of MSH3 in a control group of 18 colorectal carcinomas in which the MMR deficiency was based on the inactivation of MSH2. As it has been suggested that mutations of MSH3 might play a role in tumor progression, we studied the association between MSH3 expression and disease stage assessed by lymph node and distant metastases status. Dukes stages C and D were more frequent in primary tumors with loss of MSH3 expression (9 of 13), compared with tumors with retained expression (1 of 14; Fisher’s exact test, P = 0.001), suggesting that MSH3 abrogation may be a predictor of metastatic disease or even favor tumor cell spread in MLH1-deficient colorectal cancers.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Deficiency of the mismatch repair (MMR) system results in the accumulation of point mutations and insertions/deletions of one or a few base pairs because mispairings and misalignments occurring during DNA replication remain undetected (1) . Most tumors from patients with hereditary nonpolyposis colorectal cancer, a highly penetrant, autosomal dominant cancer-susceptibility syndrome, show contraction/expansion of simple sequence motifs, termed microsatellites, in which insertion/deletion loops occur frequently during DNA synthesis. This phenotype is called microsatellite mutator phenotype or microsatellite instability (MSI), and it is present in both noncoding and coding sequences (2, 3, 4) . Hereditary nonpolyposis colorectal cancer is caused in the majority of cases by germ-line mutations in the MMR genes MLH1 or MSH2 (reviewed in Ref. 5 ). Moreover, 10–15% of sporadic colorectal, endometrial, and gastric cancers are MSI-H (2 , 3) , mostly based on epigenetic silencing of MLH1 by hypermethylation of the promoter region (6) . Clinical and pathological features, as well as genetic events involved in tumor progression, are different in MSI-H tumors compared with tumors without this phenotype (7 , 8) .

Several genes with proven or hypothesized function in tumor suppressor pathways such as TGFßRII, MSH3, MSH6, IGFIIR, BAX, TCF-4, or caspase-5 have been suggested to play a role in the progression of MMR-deficient tumors due to frequent frameshift mutations in homopolymeric runs of eight or more nucleotides in their coding regions (9, 10, 11, 12, 13, 14) . However, because of the high genetic instability of these tumors, it is difficult to establish which of these mutations contribute to carcinogenesis (target genes mutations) and which represent irrelevant mutations (bystander genes mutations). Five criteria were proposed in a consensus meeting to distinguish between targeted or bystander mutations in MSI-H tumors (15) . These include (a) a high mutational frequency; (b) biallelic inactivation; (c) a role of the target in a growth suppressor pathway; (d) alterations within the same pathway in MSS tumors; and (e) in vitro or in vivo functional analyses. However, the validity of some of these criteria have been recently disputed (16 , 17) . The allelic status of mutated target genes has not always been fully addressed in previous studies. Although monoallelic mutations might contribute to the transformation process, biallelic mutations have a higher effect in this process when the involved gene has an important role in cell proliferation or survival. These two somatic hits could be favored by a positive selection or an intrinsic propensity of the sequence (repeats and surrounding sequences) to mutate or both.

Frameshift mutations at coding (N)8 mononucleotide repeats were found in 22–53% and in 12–50% of MSI-H colorectal carcinomas in MSH3 and MSH6, respectively (Refs. 10 , 18 , 19 , reviewed in Ref. 20 ), raising the question of whether there is a selection in these tumors for cell clones harboring mutated MSH3 and MSH6 genes. Biallelic inactivation of MSH3 or MSH6 has been reported for a few tumors (21) , and loss of protein expression has been shown only in some colon cancer cell lines (Refs. 21 , 22 , reviewed in Ref. 23 ). Most MSI-H tumors are associated with a primary deficiency of MLH1 or MSH2 (5 , 6) . In these tumors, the heterodimeric partners of MLH1 (i.e., PMS2) and MSH2 (i.e., MSH3 and MSH6) are degraded (22 , 24 , 25) , leading to severe impairment of MMR. However, an increased mutation phenotype in cells deficient for both MLH1 and MSH6 has been reported compared with those with deficiency of MLH1 solely (26) . Thus, the mutation of a second MMR gene, the product of which is not a heterodimeric partner of the primarily mutated MMR protein, could be functionally relevant in which genetic instability is increased by affecting a more specific mismatch or insertion/deletion loop repair process within the MMR system or a different DNA repair process. Therefore, we hypothesized that functional loss of the MSH3 may contribute to tumorigenesis when MMR deficiency is based on loss of MLH1. Here, we have analyzed the protein expression of MSH3 and MSH6 in MLH1-deficient tumors by immunohistochemistry. We revealed that the protein expression of MSH3, but not MSH6, is frequently abrogated and associated with frameshift mutations at the coding (A)8 tract and/or allelic losses. It has been suggested that mutations of MSH3 might play a role in the progression of MMR-deficient tumors by increasing instability (10 , 27) . We therefore studied the association between protein expression and disease stage and found a higher frequency of metastatic disease among patients with MSH3-negative tumors.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Clinical Samples.
We analyzed 29 primary colorectal carcinomas and 2 metastases (according to intraoperative and histological findings) from which the primary colorectal carcinomas were not available. All 31 tumors originated from different patients, showed loss of MLH1 (and subsequently PMS2) protein expression, and showed normal, nuclear expression of MSH2, as identified by immunohistochemistry (25) . Twenty patients fulfilled the Bethesda guidelines for hereditary nonpolyposis colorectal cancer (28) and were recruited from diagnostics in patients suspected for hereditary nonpolyposis colorectal cancer. The remaining 11 tumors originated from patients that did not fulfill the Bethesda guidelines and were recruited from 146 unselected, consecutively collected colorectal cancers screened for MSI and immunohistochemical staining for MMR proteins (25) . In addition, 18 primary colorectal carcinomas with loss of MSH2 expression and normal nuclear MLH1 staining, originating from 16 Bethesda-positive patients (10 of them with identified MSH2 germ-line mutations), were included as control. All 49 tumors were MSI-H. Genomic DNA from peripheral blood or fresh frozen normal colonic mucosa and from microdissected fresh frozen or paraffin-embedded tumor tissue was extracted, applying the QIAamp blood and tissue kit (Qiagen, Hilden, Germany). Written informed consent was obtained from all patients investigated.

Immunohistochemistry.
Immunohistochemical staining was performed on 5-µm thick, formalin-fixed, paraffin-embedded tumor sections. Sections were deparaffinized and rehydrated through graded alcohols to water (containing 3% H2O2 for MSH3). Antigen retrieval was obtained by microwave treatment (25 min, 600 W) in 10 mM sodium citrate buffer (pH 6.0). After rinsing with 0.1 M PBS (pH 7.4), treatment with 2% normal rabbit (MSH6) or swine (MSH3) serum (Dako Corp., Via Real, CA) containing 0.1% BSA was performed for 30 min to block nonspecific proteins. Sections were incubated overnight at 4°C with mouse monoclonal antibody for MSH6 (clone 44, 250 µg/ml, 1:50; Transduction Laboratories, Lexington, United Kingdom) or with a rabbit polyclonal antibody for MSH3 (1:100) raised against an NH2-terminal polypeptide comprising amino acids 1–200 (29) . Sections were then rinsed with PBS and incubated with biotinylated antimouse (MSH6) or antirabbit (MSH3) IgG (1:100; Dako Corp.) for 1 h at room temperature, followed by washing with 1x PBS and 30 min treatment with a biotin-avidin complex (Vectastain Elite ABC, Vector Laboratories, Burlingame, CA). After brief washing in PBS and 0.1 M phosphate buffer, antibody staining was obtained by 15–30 min treatment with a freshly prepared developing solution (50 ml of phosphate buffer containing 25 mg of diaminobenzidine, 20 mg of ammonium chloride, 0.2% of ß-D-glucose, and 0.2 mg of glucose oxidase) at room temperature. Sections were then rinsed briefly with phosphate buffer and PBS, weakly counterstained with eosin (1% in 80% ethanol), dehydrated in graded alcohols, cleared in xylene, and coverslipped. Omitting the primary antibody in the process described above generated negative controls from all cases. The normal staining pattern for both antibodies was nuclear. Tumor cells were considered to have lost or reduced staining only in the presence of adjacent nonneoplastic cells with nuclear staining. Staining was evaluated independently by three examiners (J. P., F. T., and S. K.).

Coding Mononucleotide Repeats and Loss of Heterozygosity (LOH) Analyses.
For the detection of frameshift mutations, the (A)8 tract in exon 7 of MSH3 was amplified in a 109-bp fragment applying sense primer 5'-CCAGCTATCTTCTGTCATCTC-3' and antisense primer 5'-CATTTGTTCCTCACCTGCAA-3', and the (C)8 tract in exon 5 of MSH6 was amplified in a 91-bp fragment applying sense primer 5'-TATAGTCGAGGGGGTGATGG-3' and antisense primer 5'-GGCGTGATCCTTTAAGCTCT-3'. Both fragments were amplified simultaneously allowing detection of allelic losses by relative quantification between tumor and normal DNA, whereby reduction of 30–50% and of >50% of the signals from tumor-derived DNA were considered as loss of one allele and loss of both alleles, respectively. Analysis for allelic loss was performed only when the (N)8 tracts were not affected by frameshift mutations. Moreover, the (N)7 tracts in MSH3 exon 22 (primers: 5'-CTTAACCCTGTTTGTCACCCA-3' and 5'-TGGATCCAGTTTGCTTTCATC-3') and in MSH6 exon 4 (5'-AAGACACAAGGATCTAGGCGA-3' and 5'-AGCCACCAATGTCACTCTCA-3') and exon 5 (5'-GCCATCCTTGCATTACGAAG-3' and 5'-TAGGCTTTGCCATTTTCCTG-3') were analyzed. For the detection of LOH in MSH3 the polymorphic, imperfect 9-bp repeat in exon 1 (Ref. 30 ; consensus DNA sequence: SCYSCAGCG; occurring with 3–8 repeats) was amplified using sense primer 5'-GTTTTGAGCCGATTCTTCCAGTCT-3' and antisense primer 5'-TCCTCCTCCAGCCCTATCAT-3'. Cutoff levels for detection of LOH were applied as described by Cawkwell et al. (31) . PCRs contained 30–50 ng of DNA, 200 mM of each dinucleotide, 1.7 mM MgCl2, 200 nM of each primer, and 1 unit of TaqDNA polymerase in a total volume of 25 µl. Conditions were 30 s at 94°C, 30 s at 58°C, and 30 s at 72°C for 27 cycles with 5 min 94°C before and 4 min 72°C after cycling. Of each PCR, 0.5–2.0 µl were electrophoresed on automated laser fluorescence express sequencing devices according to standard protocols, and fragments were analyzed for size shifts and signal intensities with the program ALLELELINKS, version 1.00 (both Amersham Biotech, Freiburg, Germany).

Sequence Analysis.
For verification of frameshift mutations at the coding mononucleotides, PCR products were separated on agarose gels, excised, eluted, and cycle sequenced applying Cy5-labeled primers, the Thermo Sequenase Fluorescent Cycle Sequencing kit (Amersham Biotech) and automated laser fluorescence express devices according to standard protocols. In addition, all 24 exons of the MSH3 gene, including their flanking intronic regions, were sequenced from 3 MSH3-negative tumors with monoallelic mutations and from 3 tumors with normal MSH3 expression according to the availability of high molecular weight tumor DNA. The genomic sequence of MSH3 was inferred from GenBank accession no. NM_006961.1 containing the complete coding region of the gene. Primers are available on request.

Statistical Analysis.
Statistical associations were analyzed using Fisher’s exact test and {chi}2 test. Statistical significance was considered at a 5% confidence interval.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Protein Expression.
We first tested the polyclonal MSH3 antibody with immunohistochemical staining of 12 MSH2-negative colorectal tumors in which both MSH3 and MSH6 proteins were expected to be degraded (see "Introduction"). The staining turned out to be drastically reduced or absent in tumor cells of all of the specimens, whereas nuclei from normal crypt epithelial cells and infiltrating lymphocytes were normally stained, indicating the suitability of the applied antibody for the detection of MSH3 by immunohistochemistry (Fig. 1, A and B)Citation . We then analyzed expression of MSH3 and MSH6 in 31 tumors in which MLH1 (and consequently PMS2) was not expressed, whereas MSH2 was normally expressed. Compared with normal epithelial crypts or infiltrating lymphocytes, a severe reduction of MSH3-staining similar to that obtained in MSH2-deficient tumors was detected in 15 tumors, whereas MSH6 was normally expressed in all tumors (Fig. 1Citation C–J and Table 1Citation ). In addition, the 2 tumors KTnL20 (Fig. 1K)Citation and KTnL22 (Fig. 1, L and M)Citation showed a heterogeneous staining pattern for MSH3, harboring both tumor regions with normal and with drastically reduced expression. There was no difference for MSH3 expression between sporadic and proven/suspected hereditary form of MLH1-deficient colorectal cancer (Table 1)Citation .



View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Protein expression of MSH3 in colorectal tumor sections, analyzed by immunostaining. Bold arrows mark either MSH3-negative () or MSH3-positive () tumor cells, normal arrows (->) mark normal epithelial crypts or infiltrating lymphocytes. A and B, sections of a MSH2-negative tumor stained with MSH2 and MSH3, respectively. A severe reduction in staining intensity of MSH3 is due to protein degradation. C and D, sections of the MLH1-negative tumor KTnL26 stained with MLH1 and MSH3, respectively, showing a normal, nuclear expression of MSH3. E–H, sections of the MLH1-negative tumor KTnL25 stained with MLH1, MSH3, MSH2, and MSH6, respectively, showing loss of MSH3 expression along with normal expression for MSH2 and MSH6. I and J, tumors KTnL19 and KTnL6, respectively, stained with MSH3 showing loss of expression. Sections of the MLH1-negative tumors KTnL20 (K) and KTnL22 (L and M) showing both MSH3-negative and -positive regions. Magnification: A–J, x100; K, x400; L, x200; and M, x40.

 

View this table:
[in this window]
[in a new window]

  		Table 1 Protein expression and mutations of MSH3 and MSH6 in MLH1-deficient colorectal tumors

 
The residual staining rather than complete loss seen with the MSH3 antibody in MSH2-negative tumors might be explained by trace amounts of undegraded MSH3 protein left upon MSH2 deficiency. Traces of full-length protein were also visible in Western blots with extracts from MSH2-negative cancer cell lines, but with this technique, no full-length product was visible in a HCT116 cell line extract in which an homozygous or hemizygous 1-bp deletion at the coding (A)8 tract is present (data not shown). However, a very weak residual staining at immunohistochemistry was also observed in tumors with biallelic MSH3 mutations (Fig. 1, F and JCitation , and Table 1Citation ). The MSH3 antibody was raised against the first 200 amino acids, whereas the frameshift mutation at the (A)8 tract frequently observed in this (see below) and other studies affects amino acid 374. Therefore, the residual MSH3 expression in these tumors might be based on the generation of truncated proteins with partial stability. Finally, the residual staining could also be caused by recognition of similar domains in the proteome with low affinity or by low level of unspecific binding. Nonetheless, the reduction of staining intensity was severe enough to lead us to designate these tumors as MSH3 negative.

Frameshifts in Coding Mononucleotides and Allelic Losses.
Frameshift mutations at the (A)8 tract in exon 7 of MSH3 were found in 16 of the 31 MLH1-deficient tumors by fragment analysis and were verified by sequencing (Table 1Citation and Fig. 2, A and CCitation ). These mutations were more frequent among the tumors with lost or partially lost MSH3 expression compared with MSH3-positive tumors (Table 2)Citation . In 3 tumors (KTnL1, KT1nL2, and KTnL7), these mutations were homozygous or hemizygous (designated "ho" in Table 1Citation , column 5; Fig. 2ACitation ). All mutations were deletions removing one or two nucleotides. In contrast to the 16 MSH3 mutations, only 3 frameshift mutations, all heterozygous insertions (Table 1Citation , column 8, Fig. 2ACitation ), were found for the (C)8 tract in exon 5 of MSH6 (Fisher’s exact test, P < 0.001). Moreover, none of the control sample of 18 MSH2-negative tumors had frameshift mutations in the (A)8 tract of MSH3, whereas two insertions (one homozygous or hemizygous) and one deletion of single nucleotides at the (C)8 tract of MSH6 were found (Table 2)Citation . No frameshifts were detected in the N (7) tracts in exon 22 of MSH3 and in exons 4 and 5 of MSH6.



View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. A–C, frameshift mutations at (N)8 coding mononucleotide repeats of MSH3 exon 7 and MSH6 exon 5 in MLH1-negative colorectal tumors. Fragment analysis of the simultaneously amplified mononucleotide repeats showing heterozygous and homozygous or hemizygous frameshift mutations (A), and signal reduction indicative of loss of one allele (KTnL12) and indicative of loss of both alleles (KTnL17; B). C, sequence verification of mutations indicated by fragment analysis. D, loss of heterozygosity (LOH) at the imperfect 9-bp repeat in exon 1 of MSH3. KT, tumor DNA; KB, normal DNA; het, heterozygous; ho, homo- or hemizygous; mo-loss and bi-loss, relative signal reduction indicating monoallelic or biallelic loss, respectively.

 

View this table:
[in this window]
[in a new window]

  		Table 2 Frequencies of MSH3 and MSH6 frameshift mutations at coding mononucleotide repeats in MLH1- and in MSH2-negative tumors

 
Simultaneous amplification of the two (N)8 repeats in MSH3 and MSH6 allowed detection of allelic loss in tumors not affected by frameshift mutations in these repeats. Allelic loss at MSH3 was found in 5 tumors (Table 1Citation , column 5), whereby in 1 tumor (KTnL17), the fragment was deleted from both alleles as ascertained by a signal reduction of ~80% (loss of one allele would not exceed signal reduction of ~50%; Fig. 2BCitation ). Allelic loss at MSH6 was found in 3 tumors, all being monoallelic (Table 1Citation , column 8).

LOH analysis of tumors having the imperfect 9-bp repeat in exon 1 of MSH3 in a heterozygous state identified additional allelic losses in 6 of 9 (66.7%) MSH3-negative tumors, compared with 2 of 6 (33.3%) MSH3-positive tumors (Table 1Citation , column 6; Fig. 2DCitation ).

Taking in account frameshift mutations and allelic losses in MSH3 (frameshifts appearing homozygous, which cannot be distinguished from hemizygous situations, were counted twice, except when associated allelic loss could be determined, i.e., no more than 2 mutations were counted/tumor), 22 mutations (73.3%) for the 30 alleles of the 15 MSH3-negative tumors, compared with 8 mutations (28.6%) for the 28 alleles of the 14 MSH3-positive tumors were found ({chi}2 = 11.62, P = 0.001). Biallelic inactivation of MSH3 was evident or could be inferred for 9 of 15 (60.0%) MSH3-negative tumors but was not found for any of the MSH3-positive tumors (Table 3)Citation .


View this table:
[in this window]
[in a new window]

  		Table 3 Inactivation of MSH3 by frameshift mutations at coding mononucleotide repeats and allelic losses in MLH1-negative tumors

 
Sequencing of the Coding Region of MSH3.
No alterations or only monoallelic inactivation of MSH3 were found in 6 MSH3-negative cancers. Thus, the coding region of MSH3 was sequenced from DNA of 3 of these tumors (KTnL15, KTnL18, and KTnL19) in which high molecular weight tumor DNA was available. No additional mutations were found, but a signal imbalance in a common A to G polymorphism at nucleotide 3133 changing threonine to alanine (32) indicated allelic loss in KTnL18. This polymorphism is located at the 3' end of the gene (exon 23), >200 kb of genomic sequence away from the loci analyzed for allelic loss by the methods described above (exons 1 and 7). Sequencing of 3 tumors with normal MSH3 expression (KTnL5, KTnL9, and KTnL21) did not reveal additional mutations.

MSH3 Expression and Metastatic Spread in MLH1-Negative Colon Cancers.
Among 13 primary colorectal carcinomas showing loss of MSH3 expression, 9 were associated with lymph node and/or distant metastases (Dukes C and D). In contrast, among 14 patients with carcinomas showing normal MSH3 expression, only 1 patient had Dukes C stage of the disease, and none had a Dukes D stage cancer (Fisher’s exact test, P = 0.001). Both metastatic tumors analyzed in this study showed loss of MSH3 expression (Table 1)Citation . Disease of the 2 patients with tumors showing both MSH3-negative and -positive regions (KTnL20 and KTnL22) were Dukes B and Dukes A, respectively.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In MMR-deficient tumors, MSH3 and MSH6 are frequently affected by frameshift mutations at (N)8 tracts in their coding regions (10) . Evidence for loss of MSH3 protein expression has been shown for the MLH1-deficient cell line HCT116 by Western blotting (22) but has not been reported in primary tumors because of the lack of antibodies suitable for immunohistochemistry. Using a polyclonal antibody for the MSH3 protein, we have shown that in around half of the MLH1-deficient colorectal tumors, there is a loss of MSH3 expression, indicating that the MSH3 function is targeted frequently upon MLH1-based MMR deficiency. Accordingly, the frequency of frameshift mutations in the coding (A)8 repeats of MSH3 was significantly higher (51.6%) than in the (C)8 tract of MSH6 (9.7%), although randomly selected intronic (G)8 tracts have been shown to be 4.5 times more often mutated than (A)8 tracts in MMR-deficient tumors (33) . Moreover, out of a control panel of 18 MSH2-negative tumors, none had frameshift mutations at the MSH3 (A)8 tract.

Although about one-fourth of MMR-deficient carcinomas show LOH at loci often affected in colorectal tumors (34) , the majority of MMR-deficient carcinomas remain diploid or near diploid, which is different from the majority of sporadic, chromosomal unstable colorectal carcinomas (8 , 35 , 36) . Assuming diploidy for the MLH1-deficient carcinomas, both MSH3 alleles should be inactivated for loss of MSH3 expression to occur. The frameshift mutations at the (A)8 tract accounted for approximately half of the mutations expected. We identified allelic loss as a second type of genetic alterations involved. Biallelic inactivation of MSH3 was ascertained or inferred in 60% of MSH3-negative tumors and was not found in any of the tumors with normal MSH3 expression. Identified alterations account for approximately two-thirds of the mutations expected. Sequencing of the coding region of MSH3 in some tumors negative for MSH3 at immunohistochemistry but with only monoallelic or no alterations of MSH3 did not reveal additional mutations, except an allelic loss at a polymorphic site at the 3'-end of the gene in one tumor. This might indicate that the frequency of allelic loss affecting the gene has been underestimated in our data. In addition, genetic or epigenetic alterations of the promoter region of MSH3 cannot be excluded in our study. Because of the fairly accurate staining for MSH3 in immunohistochemistry and Western blots, we assume that the different approaches for mutation detection we used did not cover all of the genetic alterations affecting MSH3.

The (C)8 tract of MSH6 has also been reported as mutated in MMR-deficient tumors, albeit in a lower frequency compared with MSH3 (Ref. 10 , reviewed in Ref. 20 ). We found a low frequency (~10%) of MSH6 (C)8 tract frameshift mutations in MLH1-deficient tumors not different to the frequency found in MSH2-deficient tumors. Moreover, none of the MLH1-deficient tumors showed loss of protein expression and/or biallelic inactivation for MSH6 by the methods applied here. Baranovskaya et al. (26) showed that the inactivation of MSH6 in MLH1-deficient cells shifts the prevalence of mutations from frameshifts to base substitutions and increases the mutation rate, which might speed up tumor progression. Moreover, they reported several somatic missense mutations in MSH6 in MMR-deficient tumors (21) . In our series of MSI-H tumors we also identified a sporadic, MLH1-negative tumor with a somatic missense mutation of a highly conserved residue in MSH6 without loss of protein expression (data not shown). Therefore, MSH6 might also be targeted upon MLH1-based tumorigenesis.

As expected, frameshift mutations and allelic losses of MSH3 were significantly more frequent in tumors with loss of MSH3 expression. However, approximately half of MLH1-negative tumors with retained MSH3 expression harbored mono-allelic MSH3 alterations, raising the question whether these alterations will be followed by a second somatic hit during tumor progression. In our series, loss of MSH3 expression was associated with lymph node involvement and distant metastasis (Dukes C and D stages) in 9 of 13 tumors. In contrast, MSH3 was not expressed in 4 of 17 tumors from patients with Dukes A or B disease. Dukes stages C and D are associated with an increasingly worse prognosis, and metastasis is the major cause of death among colorectal cancer patients. These data support the notion that functional impairment of MSH3 is not required for the formation of MMR-deficient tumors but is selected for during tumor progression (27) and might be a predictor of metastatic disease. The co-occurrence of MSH3-negative and -positive regions seen in 2 tumors may pinpoint a transitional state in tumor progression. We analyzed 2 metastases that were both negative for MSH3 staining; however, this observation must be extended in a larger series of metastatic lesions of MLH1-negative primary carcinomas.

The contribution of MSH3 deficiency to tumorigenesis may be supported by the fact that MSH6 germ-line mutations were found in kindreds with a later age of tumor onset when compared with families with MSH2 germ-line mutations (37) , and the apparent difference is presence or absence of the MSH3 protein, respectively. Similar to the contribution of MSH6 deficiency in MLH1-deficient tumor cells (26) , MSH3 deficiency also may increase the mutation rate and/or change the mutation spectrum in MLH1-deficient cells. The human MutSß complex (MSH2+MSH3) recognizes insertion/deletion loops of two and more nucleotides more efficiently than MutS{alpha} (MSH2+MSH6; Refs. 32 , 38 ). Besides its involvement in MMR, the yeast Msh2p-Msh3p complex is involved in the repair of DNA double-strand breaks by single-strand annealing (39 , 40) . This type of recombinational repair involves the Rad1p-Rad10p endonuclease but is independent of other MMR proteins such as Mlh1p and Msh6p (39) . Deletion of Msh3p in Saccharomyces cerevisiae increased the mitotic recombination rate of homeologous sequences (91% identity) ~9-fold (41) . Msh3p recognizes DNA loops and palindromic sequences in mitotic recombination in S. cerevisiae and had an antirecombinational role when heteroduplex sequences contained base-base mispairs (42) . The nonrejection of homeologous DNA increases the genetic exchange between divergent chromosomes or chromosome regions (43) . Therefore, the contribution of MSH3 inactivation to tumor progression might also be related to a function of this gene different from that in MMR.


   ACKNOWLEDGMENTS
 
We thank Monika Reichmann and Ansgard Rudek for excellent technical assistance.


   FOOTNOTES
 
Grant support: Deutsche Forschungsgemeinschaft Grant PL 311/1-1.

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.

Present address: I. Iaccarino, Instituto di Genetica e Biofisica "A. Buzzati Traverso," Consiglio Nazionale delle Ricerche, Napoli, Italy.

Requests for reprints: Jens Plaschke, Department of Surgical Research, Carl Gustav Carus Hospital, Dresden University of Technology, Fetscherstrasse 74, D-01307 Dresden, Germany. Phone: 49-351-458-3348; Fax: 49-351-458-5365; E-mail: plaschke{at}rcs.urz.tu-dresden.de

Received 9/ 5/03. Revised 10/31/03. Accepted 11/25/03.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kolodner R. D., Marsischky G. T. Eukaryotic DNA mismatch repair. Curr. Opin. Genet. Dev., 9: 89-96, 1999.[CrossRef][Medline]
  2. Ionov Y., Peinado M. A., Malkhosyan S., Shibata D., Perucho M. Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature (Lond.), 363: 558-561, 1993.[CrossRef][Medline]
  3. Thibodeau S. N., Bren G., Schaid D. Microsatellite instability in cancer of the proximal colon. Science (Wash. DC), 260: 816-819, 1993.[Abstract/Free Full Text]
  4. Aaltonen L. A., Peltomaki P., Leach F. S., Sistonen P., Pylkkanen L., Mecklin J. P., Järvinen H., Powell S. M., Jen J., Hamilton S. R., Peterson G. M., Kinzler K. W., Vogelstein B., de la Chapelle A. Clues to the pathogenesis of familial colorectal cancer. Science (Wash. DC), 260: 812-816, 1993.[Abstract/Free Full Text]
  5. Peltomaki P. Role of DNA mismatch repair defects in the pathogenesis of human cancer. J. Clin. Oncol., 21: 1174-1179, 2003.[Abstract/Free Full Text]
  6. Kane M. F., Loda M., Gaida G. M., Lipman J., Mishra R., Goldman H., Jessup J. M., Kolodner R. Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res., 57: 808-811, 1997.[Abstract/Free Full Text]
  7. Lothe R. A., Peltomaki P., Meling G. I., Aaltonen L. A., Nystrom-Lahti M., Pylkkanen L., Heimdal K., Andersen T. I., Moller P., Rognum T. O. Genomic instability in colorectal cancer: relationship to clinicopathological variables and family history. Cancer Res., 53: 5849-5852, 1993.[Abstract/Free Full Text]
  8. Olschwang S., Hamelin R., Laurent-Puig P., Thuille B., De Rycke Y., Li Y. J., Muzeau F., Girodet J., Salmon R. J., Thomas G. Alternative genetic pathways in colorectal carcinogenesis. Proc. Natl. Acad. Sci. USA, 94: 12122-12127, 1997.[Abstract/Free Full Text]
  9. Markowitz S., Wang J., Myeroff L., Parsons R., Sun L., Lutterbaugh J., Fan R. S., Zborowska E., Kinzler K. W., Vogelstein B., Brattain M., Willson J. K. V. Inactivation of the type II TGF-ß receptor in colon cancer cells with microsatellite instability. Science (Wash. DC), 268: 1336-1338, 1995.[Abstract/Free Full Text]
  10. Malkhosyan S., Rampino N., Yamamoto H., Perucho M. Frameshift mutator mutations. Nature (Lond.), 382: 499-500, 1996.[CrossRef][Medline]
  11. Souza R. F., Appel R., Yin J., Wang S., Smolinski K. N., Abraham J. M., Zou T. T., Shi Y. Q., Lei J., Cottrell J., Cymes K., Biden K., Simms L., Leggett B., Lynch P. M., Frazier M., Powell S. M., Harpaz N., Sugimura H., Young J., Meltzer S. J. Microsatellite instability in the insulin-like growth factor II receptor gene in gastrointestinal tumours. Nat. Genet., 14: 255-257, 1996.[CrossRef][Medline]
  12. Rampino N., Yamamoto H., Ionov Y., Li Y., Sawai H., Reed J. C., Perucho M. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science (Wash. DC), 275: 967-969, 1997.[Abstract/Free Full Text]
  13. Duval A., Gayet J., Zhou X. P., Iacopetta B., Thomas G., Hamelin R. Frequent frameshift mutations of the TCF-4 gene in colorectal cancers with microsatellite instability. Cancer Res., 59: 4213-4215, 1999.[Abstract/Free Full Text]
  14. Schwartz S., Jr., Yamamoto H., Navarro M., Maestro M., Reventos J., Perucho M. Frameshift mutations at mononucleotide repeats in caspase-5 and other target genes in endometrial and gastrointestinal cancer of the microsatellite mutator phenotype. Cancer Res., 59: 2995-3002, 1999.[Abstract/Free Full Text]
  15. Boland C. R., Thibodeau S. N., Hamilton S. R., Sidransky D., Eshleman J. R., Burt R. W., Meltzer S. J., Rodriguez-Bigas M. A., Fodde R., Ranzani G. N., Srivastava S. 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., 58: 5248-5257, 1998.[Abstract/Free Full Text]
  16. Perucho M. Correspondence re C. R. Boland 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., 59: 249-256, 1999.[Free Full Text]
  17. Perucho M. Tumors with microsatellite instability: many mutations, targets and paradoxes. Oncogene, 22: 2223-2225, 2003.[CrossRef][Medline]
  18. Yin J., Kong D., Wang S., Zou T. T., Souza R. F., Smolinski K. N., Lynch P. M., Hamilton S. R., Sugimura H., Powell S. M., Young J., Abraham J. M., Meltzer S. J. Mutation of hMSH3 and hMSH6 mismatch repair genes in genetically unstable human colorectal and gastric carcinomas. Hum. Mutat., 10: 474-478, 1997.[CrossRef][Medline]
  19. Yamamoto H., Sawai H., Weber T. K., Rodriguez-Bigas M. A., Perucho M. Somatic frameshift mutations in DNA mismatch repair and proapoptosis genes in hereditary nonpolyposis colorectal cancer. Cancer Res., 58: 997-1003, 1998.[Abstract/Free Full Text]
  20. Duval A., Hamelin R. Mutations at coding repeat sequences in mismatch repair-deficient human cancers: toward a new concept of target genes for instability. Cancer Res., 62: 2447-2454, 2002.[Abstract/Free Full Text]
  21. Ohmiya N., Matsumoto S., Yamamoto H., Baranovskaya S., Malkhosyan S. R., Perucho M. Germline and somatic mutations in hMSH6 and hMSH3 in gastrointestinal cancers of the microsatellite mutator phenotype. Gene (Amst.), 272: 301-313, 2001.[CrossRef][Medline]
  22. Chang D. K., Ricciardiello L., Goel A., Chang C. L., Boland C. R. Steady-state regulation of the human DNA mismatch repair system. J. Biol. Chem., 275: 18424-18431, 2000.[Abstract/Free Full Text]
  23. Jiricny J., Marra G. DNA repair defects in colon cancer. Curr. Opin. Genet. Dev., 13: 61-69, 2003.[CrossRef][Medline]
  24. Drummond J. T., Genschel J., Wolf E., Modrich P. DHFR/MSH3 amplification in methotrexate-resistant cells alters the hMutS{alpha}/hMutSß ratio and reduces the efficiency of base-base mismatch repair. Proc. Natl. Acad. Sci. USA, 94: 10144-10149, 1997.[Abstract/Free Full Text]
  25. Plaschke J., Krüger S., Pistorius S., Theissig F., Saeger H. D., Schackert H. K. Involvement of hMSH6 in the development of hereditary and sporadic colorectal cancer revealed by immunostaining is based on germline mutations, but rarely on somatic inactivation. Int. J. Cancer, 97: 643-648, 2002.[CrossRef][Medline]
  26. Baranovskaya S., Soto J. L., Perucho M., Malkhosyan S. R. Functional significance of concomitant inactivation of hMLH1 and hMSH6 in tumor cells of the microsatellite mutator phenotype. Proc. Natl. Acad. Sci. USA, 98: 15107-15112, 2001.[Abstract/Free Full Text]
  27. Duval A., Rolland S., Compoint A., Tubacher E., Iacopetta B., Thomas G., Hamelin R. Evolution of instability at coding and non-coding repeat sequences in human MSI-H colorectal cancers. Hum. Mol. Genet., 10: 513-518, 2001.[Abstract/Free Full Text]
  28. Rodriguez-Bigas M. A., Boland C. R., Hamilton S. R., Henson D. E., Jass J. R., Khan P. M. A National Cancer Institute Workshop on Hereditary Nonpolyposis Colorectal Cancer Syndrome: meeting highlights and Bethesda guidelines. J. Natl. Cancer Inst. (Bethesda), 89: 1758-1762, 1997.[Free Full Text]
  29. Kleczkowska H. E., Marra G., Lettieri T., Jiricny J. hMSH3 and hMSH6 interact with PCNA and colocalize with it to replication foci. Genes Dev., 15: 724-736, 2001.[Abstract/Free Full Text]
  30. Nakajima E., Orimo H., Ikejima M., Shimada T. Nine-bp repeat polymorphism in exon 1 of the hMSH3 gene. Jpn. J. Hum. Genet., 40: 343-345, 1995.[CrossRef][Medline]
  31. Cawkwell L., Lewis F. A., Quirke P. Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction. Br. J. Cancer, 70: 813-818, 1994.[Medline]
  32. Acharya S., Wilson T., Gradia S., Kane M. F., Guerrette S., Marsischky G. T., Kolodner R., Fishel R. hMSH2 forms specific mispair-binding complexes with hMSH3 and hMSH6. Proc. Natl. Acad. Sci. USA, 93: 13629-13634, 1996.[Abstract/Free Full Text]
  33. Zhang L., Yu J., Willson J. K., Markowitz S. D., Kinzler K. W., Vogelstein B. Mononucleotide repeat sequence variability in mismatch repair-deficient cancers. Cancer Res., 61: 3801-3805, 2001.[Abstract/Free Full Text]
  34. Goel A., Arnold C. N., Niedzwiecki D., Chang D. K., Ricciardiello L., Carethers J. M., Dowell J. M., Wasserman L., Compton C., Mayer R. J., Bertagnolli M. M., Boland C. R. Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res., 63: 1608-1614, 2003.[Abstract/Free Full Text]
  35. Remvikos Y., Vogt N., Muleris M., Salmon R. J., Malfoy B., Dutrillaux B. DNA-repeat instability is associated with colorectal cancers presenting minimal chromosome rearrangements. Genes Chromosomes Cancer, 12: 272-276, 1995.[Medline]
  36. Lengauer C., Kinzeler K. W., Vogelstein B. Genetic instability in colorectal cancers. Nature (Lond.), 386: 623-627, 1997.[CrossRef][Medline]
  37. Wijnen J., de Leeuw W., Vasen H., van der Klift H., Moller P., Stormorken A., Meijers-Heijboer H., Lindhout D., Menko F., Vossen S., Möslein G., Tops C., Bröcker-Vriends A., Wu Y., Hofstra R., Sijmons R., Cornelisse C., Morreau H., Fodde R. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat. Genet., 23: 142-144, 1999.[CrossRef][Medline]
  38. Palombo F., Iaccarino I., Nakajima E., Ikejima M., Shimada T., Jiricny J. hMutSß, a heterodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA. Curr. Biol., 6: 1181-1184, 1996.[CrossRef][Medline]
  39. Sugawara N., Paques F., Colaiacovo M., Haber J. E. Role of Saccharomyces cerevisiae Msh2 and Msh3 repair proteins in double-strand break-induced recombination. Proc. Natl. Acad. Sci. USA, 94: 9214-9219, 1997.[Abstract/Free Full Text]
  40. Bertrand P., Tishkoff D. X., Filosi N., Dasgupta R., Kolodner R. D. Physical interaction between components of DNA mismatch repair and nucleotide excision repair. Proc. Natl. Acad. Sci. USA, 95: 14278-14283, 1998.[Abstract/Free Full Text]
  41. Datta A., Adjiri A., New L., Crouse G. F., Jinks Robertson S. Mitotic crossovers between diverged sequences are regulated by mismatch repair proteins in Saccaromyces cerevisiae. Mol. Cell Biol., 16: 1085-1093, 1996.[Abstract]
  42. Nicholson A., Hendrix M., Jinks-Robertson S., Crouse G. F. Regulation of mitotic homeologous recombination in yeast. Functions of mismatch repair and nucleotide excision repair genes. Genetics, 154: 133-146, 2000.[Abstract/Free Full Text]
  43. Radman M. Mismatch repair and the fidelity of genetic recombination. Genome, 31: 68-73, 1989.[Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
A. C. Haugen, A. Goel, K. Yamada, G. Marra, T.-P. Nguyen, T. Nagasaka, S. Kanazawa, J. Koike, Y. Kikuchi, X. Zhong, et al.
Genetic Instability Caused by Loss of MutS Homologue 3 in Human Colorectal Cancer
Cancer Res., October 15, 2008; 68(20): 8465 - 8472.
[Abstract] [Full Text] [PDF]

Home page
 BioinformaticsHome page
C. H. Busold, S. Winter, N. Hauser, A. Bauer, J. Dippon, J. D. Hoheisel, and K. Fellenberg
Integration of GO annotations in Correspondence Analysis: facilitating the interpretation of microarray data
Bioinformatics, May 15, 2005; 21(10): 2424 - 2429.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Pathol.Home page
B Perez-Ordonez, N N Huynh, K W Berean, and R C K Jordan
Expression of mismatch repair proteins, {beta} catenin, and E cadherin in intestinal-type sinonasal adenocarcinoma
J. Clin. Pathol., October 1, 2004; 57(10): 1080 - 1083.
[Abstract] [Full Text] [PDF]

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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plaschke, J.
Right arrow Articles by Schackert, H. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Plaschke, J.
Right arrow Articles by Schackert, H. K.

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 Meeting Abstracts Online