HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 24, 18424-18431, June 16, 2000

This Article Abstract Full Text (PDF) An addition or correction has been published All Versions of this Article: 275/24/18424    most recent M001140200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chang, D. K. Articles by Boland, C. R. Search for Related Content PubMed PubMed Citation Articles by Chang, D. K. Articles by Boland, C. R. Social Bookmarking                      What's this?

Steady-state Regulation of the Human DNA Mismatch Repair System^*

Dong Kyung Chang, Luigi Ricciardiello, Ajay Goel, Christina L. Chang, and C. Richard Boland^§

From the Department of Medicine and Cancer Center, University of California, San Diego, California 92093 and  Veterans Affairs Medical Center, La Jolla, California 92093-0688

Received for publication, February 11, 2000, and in revised form, March 14, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Steady-state levels of human DNA mismatch repair (MMR) transcripts and proteins were measured in MMR-proficient and -deficient cell lines by the newly developed competitive quantitative reverse transcription- polymerase chain reaction and Western analysis normalized with purified proteins. In MMR-proficient cells, hMSH2 is the most abundant MMR protein and is expressed 3 to 5 times more than hMLH1. The hMLH1 protein was expressed 1.5 to 2.5 times more than hPMS2. Steady-state levels of mRNA expression correlated well with protein expression. hMSH2-mutated LoVo cells did not express detectable hMSH3 or hMSH6 proteins. Similarly, hMLH1-mutated HCT116 cells did not express detectable hMLH1 or hPMS2 protein, whereas in hMLH1-restored HCT116+ch3 cells, hPMS2 protein was re-expressed. In hMSH6-mutated HCT15 cells, both hMSH3 protein and mRNA were increased. In SV40-transformed lung fibroblasts, all MMR mRNAs and proteins examined were expressed at levels 1.5-5-fold higher than in their nontransformed counterpart. The steady-state levels of MMR proteins indicate that substantially more hMutS proteins, which are involved in DNA mismatch recognition, are present in comparison with the hMutL proteins. Stability of hMSH3 and hMSH6 proteins appears to depend upon the presence of the hMSH2 protein, and, similarly, the stability of the hPMS2 protein depends upon hMLH1. When the hMSH6 is mutationally inactivated, hMSH3 increases by both transcriptional up-regulation and enhanced protein stability. A balanced up-regulation of all of the components was seen after viral transformation in a fibroblast model. Quantitative changes of the MMR components are a potential mechanism to modify the DNA MMR capabilities of a cell.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The human mismatch repair (MMR)¹ system ensures replication fidelity by correcting postreplication errors that have escaped the DNA proofreading function of DNA polymerase. Defects in the MMR system result in the development of a genetically unstable mutator phenotype and render the cell more susceptible to neoplastic transformation (reviewed in Refs. 1-3). The MMR system functions through the interactions between several proteins, such as hMSH2, hMSH3, hMSH6, hMLH1, hPMS2, and recently discovered hMLH3 (1-4). hMutS- is a heterodimer of hMSH2 and hMSH6 that binds to mismatched nucleotides or single insertion/deletion loops (5-7). In addition, hMSH2 can dimerize with hMSH3 creating the hMutS- complex, which binds larger DNA insertion/deletion mispaired loops (5, 7, 8). Subsequently, activated hMutS- or hMutS- interacts with hMutL- (a heterodimer of hMLH1 with hPMS2) to direct DNA repair (9, 10).

Recent studies suggest that maintaining an appropriate balance of the components of the human MMR system is critical for its proper repair function. In hMSH6-defective HCT15 cells, hMSH3 protein is highly expressed, whereas in hMSH2-defective LoVo cells, hMSH3 and hMSH6 proteins are not detected (11). The suggested cause for the absence of the hMSH3 and hMSH6 proteins is that they are relatively unstable without forming a heterodimer complex with hMSH2 (11, 12). When hMSH3 was overexpressed by methotrexate-induced simultaneous amplification of the dihydrofolate reductase and hMSH3 genes in CHOR and HL60R cells, the relative amount of hMSH6 protein decreased (12, 13). In all likelihood, this decrease was caused by excess hMSH3 protein competing with hMSH6 for dimerization with the common partner protein, hMSH2 (12, 13).

The human MMR system may be regulated in several different biological situations. Immunohistochemical studies show that the hMSH2 protein is readily detected in the proliferating portion of esophageal and intestinal epithelium (14-16) and increases at least 12-fold in proliferating cells (16). Since the major role of the MMR system occurs during the immediate postreplicative stage, it seems plausible that the MMR system may be up-regulated in highly replicating cells.

Although a stable relationship among the components of the MMR system is important in maintaining their function as a complex, the exact quantitative profile of each MMR component is not yet known. Limited information as to the genetic regulation of the MMR system is available. Recently, some quantitative approaches have been attempted to demonstrate that functional defects of the MMR system may result in microsatellite instability and carcinogenesis (17, 18).

We have developed a new method, competitive quantitative reverse transcription-polymerase chain reaction (CQ-RT-PCR) for measuring the exact amount of each MMR mRNA. We then compared the amount of each mRNA to the expression of the corresponding protein. We successfully determined the exact amount of each MMR component in the balanced steady state for both mRNAs and proteins in MMR-proficient cell lines. Moreover, the quantitative changes in abnormal MMR-defective cell lines were studied at both the mRNA and protein level. We also found that all human MMR mRNA and proteins measured were up-regulated in highly proliferating SV40-transformed cell lines in comparison with their normal counterpart.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Models-- The MMR-proficient colon cancer cell lines SW480 and CaCo2, the erythroleukemia cell line HEL, the cervical cancer cell line HeLa, the human fibroblast cell lines MRC5 and WI38, and the simian virus (SV) 40-transformed WI38/VA13 cells (19) were cultured in Isocove's modified Dulbecco's medium with 10% fetal bovine serum (Life Technologies, Inc., Grand Island, NY) at 37 °C in 5% CO₂. The MMR-deficient colon cancer cell lines HCT116, LoVo, DLD1, HCT15, and SW48 were also grown in these conditions. The WI38 (CCL-75) and WI38/VA13 (CCL-75.1) cell lines were purchased from ATCC. HCT116+ch3, which is MMR proficient by the stable transfection of chromosome 3 bearing a wild-type copy of the hMLH1 gene, was prepared as originally described (20) and grown in 400 µg/ml G418. All cells were harvested at log phase and stored at 80 °C.

Multiplex RT-PCR-- We designed oligonucleotides for simultaneously measuring six human MMR and -actin transcripts. The MacVector^R software (Oxford Molecular Group, Oxford, United Kingdom) was used for designing the candidate primer sequences with the optimal annealing temperatures ranging between 51 and 56 °C. To avoid primer-primer dimerizations, we excluded complementary sequences between the three nucleotides at both sides of each primer. By including at least one intronic sequence, we chose oligonucleotides to amplify fragments whose size could be readily distinguished from those obtained from genomic DNA. Computerized PCR simulation with Amplify 1.2^TM software (University of Wisconsin, Madison, WI) ruled out possible nonspecific PCR products. The primer sequences are shown in Table I.

Total cellular RNA was extracted with TRIzol (Life Technologies, Inc.) following the manufacturer's instructions. cDNA was synthesized at 37 °C for 60 min from 1 µg of total cellular RNA by reverse transcription in 20-µl reactions containing 25 µg/ml random hexamer (Roche Molecular Biochemicals, Indianapolis, IN); 10,000 units/ml of Moloney murine leukemia virus reverse transcriptase (Life Technologies, Inc.); 50 mM Tris-HCl, pH 8.3, 75 mM KCl, 3 mM MgCl₂, 10 mM dithiothreitol, 0.4 mM dNTPs, 2,000 units/ml RNase Inhibitor (Stratagene, La Jolla, CA), and 8.5 µl of diethyl pyrocarbonate-treated water. Reactions were stopped by heat inactivation for 10 min at 90 °C and then quickly chilled on ice.

A PTC200 DNA Engine (MJ Research, Watertown, MA) was used to amplify 2 µl of cDNA products in 50-µl reactions containing 10 mM Tris-HCl, pH 8.3, 10 mM KCl, 0.4 mM each dNTP, and 100 units/ml AmpliTaq DNA Polymerase-Stoffel fragment (Perkin-Elmer, Wellesley, MA), in the presence of 100 pM -actin primers, 340 pM hMSH6 primers, 120 pM hMSH2 primers, 230 pM hPMS2 primers, 90 pM hMLH1 primers, 110 pM hMSH3 primers, and 100 pM hPMS1 primers. Hot-start PCR was performed by adding the DNA polymerase at 80 °C after initial denaturation. The PCR amplification cycles consisted of denaturation at 95 °C for 3 min; 30 cycles of denaturation at 95 °C for 30 s, annealing at 54 °C for 30 s, and extension at 72 °C for 45 s; and a final elongation at 72 °C for 10 min. PCR products were separated on a 1% agarose gel, stained with 0.5 µg/ml ethidium bromide, and visualized by ultraviolet (UV) light.

Constructing pMMR Multi-template Plasmid and Generating MMR RNA Standards-- To facilitate a more stringent and accurate quantification of human MMR mRNAs, we developed a CQ-RT-PCR technique by applying an approach originally described (21). First, the plasmid pMMR was developed by PCR-oligonucleotide overlap extension. The pMMR plasmid contains a 491-bp DNA multi-template composed of the six human MMR and the -actin primer pair sequences, and a portion of the hMSH3 gene sequence obtained from the MMR-proficient SW480 cell line was included as a spacer between the pair sequences (Fig. 1). Because the six MMR primer sets used for multiplex RT-PCR were already screened to rule out nonspecific interactions, we used the same primer sequences to construct the pMMR plasmid. This deoxynucleotide cassette was inserted into the TA-cloning vector pCR2.1 (Invitrogen, Carlsbad, CA). The RNA standard was generated by in vitro transcription using the pMMR plasmid as a template. After linearization with HindIII, the pMMR plasmid was transcribed in vitro using T7 RNA polymerase following the conditions recommended by the manufacturer (Promega, Madison, WI). The total length of the RNA product of the pMMR plasmid was 619 bp.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Structure of the pMMR multi-template plasmid for generating the MMR RNA standard. Six primer set sequences for MMR multiplex PCR and -actin primers were used for constructing pMMR by PCR-oligonucleotide overlap extension. The arrangement of priming sites was designed to yield PCR products that are different in size from those of the target RNA. The MMR RNA standard was generated by in vitro transcription using this plasmid.

Competitive Quantitative RT-PCR-- Known amounts of RNA standard molecules were serially diluted 2-2.5-fold, mixed with 1 µg of total cellular RNA per reaction tube, and reverse transcribed. Subsequently, 2 µl of cDNA products per MMR species was amplified by PCR in a 50-µl reaction volume containing 20 mM Tris-HCl, pH 8.4, 50 mM KCl, 0.2 mM dNTPs, 1.2 mM (for hMSH2, hMSH3, hMSH6, hMLH1, and hPMS1) or 1.5 mM (for hPMS2) MgCl₂, 50 units/ml Taq DNA polymerase (Life Technologies, Inc.), and one of the MMR primer sets. For hMSH2, hMSH3, hMLH1, and hPMS1, 100 pM primers were used; for hMSH6 and hPMS2, 200 pM primers were used.

The amplification consisted of denaturation at 95 °C for 2 min, 30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s (hMSH2, hMSH6, hMLH1, and hPMS1) or 54 °C (hMSH3 and hPMS2), and extension at 72 °C for 45 s; and a final elongation at 72 °C for 10 min. PCR products were electrophoresed through a 1% agarose gel stained with 0.5 µg/ml ethidium bromide and visualized by UV light.

The band intensities were compared with an Alphaimager 2000 image densitometer (Alpha Innotech Co., San Leandro, CA). To determine the amounts of sample mRNA, we calculated the isomolar point between bands obtained with the RNA standard and the sample mRNA (Fig. 2). When the ratios of the band intensities between the PCR products obtained from the RNA standard and target mRNA are plotted against the number of RNA standard molecules, the number of RNA standard molecules equivalent to the ratio 1 is the pre-corrected copy number of sample target mRNA.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Determination of the isomolar point. Increasing amounts of RNA standard molecules were reverse transcribed with a constant amount of SW480 total cellular RNA (1 µg) simultaneously in the same reaction. Following reverse transcription, a cDNA reaction mixture equivalent to 100 ng of total cellular RNA was amplified using hMSH2-specific primers. PCR products were electrophoresed in a 1% agarose gel and bands were visualized by ethidium bromide staining and analyzed with an image densitometer. The ratios of band intensities of the PCR products from RNA standard and target RNA were plotted against the amounts of RNA standard. When the ratio of band intensities reaches 1, the amount of target RNA molecules is equivalent to the amount of the RNA standard. In this example, 100 ng of SW480 total cellular RNA contains hMSH2 mRNA equivalent to 17.7 amol of RNA standard. With remaining cDNA reaction mixture and specific primers, all six MMR mRNAs can be measured.

Validation of CQ-RT-PCR Using Full-length MMR cDNA-- To establish the accuracy of the absolute quantification yielded by the CQ-RT-PCR, we compared the measured amounts of the cDNAs generated from the artificially constructed RNA standard against each of the full-length MMR cDNAs. We then calculated the appropriate correction factors by creating a standard curve comparing known amounts of the full-length coding sequence of MMR cDNAs with the multi-template DNA for the RNA standard. The primers used to generate the full coding sequences of MMR system are the following: for hMLH1, 5'-GACGTTTCCTTGGCTCTTCTGGCG-3' (forward), 5'-GGAATACAGAGAAAGAAGAACACATCCC-3' (backward); for hMSH2, 5'-GTGAGGAGGTTTCGACATGGC-3' (forward), 5'-TCCATTACTGGGATTTTTCACGTAG-3' (backward); for hPMS2, 5'-ATCGGGTGTTGCATCCATGGAGC-3' (forward), 5'-CCATACAGTGACTACGGTCAGTTCTG-3' (backward).

PCR reactions were performed using SW480 cDNAs as templates in 1× ExpandLong Template PCR buffer 3 (detergents and 2.25 mM MgCl₂ included; Roche Molecular Biochemicals), 0.5 mM each dNTP, 50 units/ml Thermostable Taq, and Pwo DNA polymerase mixture (Roche), and 300 pM primers. An initial denaturation at 94 °C for 2 min was followed by 10 cycles of amplification consisting of denaturation at 94 °C for 10 s, annealing at 55 °C for 30 s, and extension at 68 °C for 2 min, and 20 additional cycles with incremental increases of the extension time of 20 s per cycle, followed by a final extension at 68 °C for 7 min. PCR products were electrophoresed through a 1% agarose gel, and the amplified fragments were sliced from the gel with a sterile razor blade. Gel slices were purified with UltraClean GelSpin DNA Purification Kit (McFrugal's Lab Depot, San Diego, CA) following the manufacturer's instructions. The lengths for each full MMR PCR product were 2840 bp for hMSH2, 2341 bp for hMLH1, and 2624 bp for hPMS2.

hMSH3 and hMSH6 full-length cDNAs were prepared by restricting the hMSH3-pGEM7z(+) plasmid with XhoI and HindIII and the hMSH6-pBSKS plasmid with XhoI and BamHI. These full-length cDNAs were used for generating the correction factors.

Western Blot Analysis of MMR Proteins-- After extraction as described previously (16), protein concentrations were measured with the BCA protein assay kit (Pierce, Rockford, IL) following the manufacturer's instructions. Protein aliquots were then mixed with an identical amount of Laemmli gel loading buffer and placed in a boiling water bath for 5 min. Proteins were separated by 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto an Hybond ECL nitrocellulose membrane (Amersham Pharmacia Biotech). Membranes were blocked for 1 h with 5% nonfat milk in phosphate-buffered saline-Tween 20 (PBS-T) and incubated at 4 °C overnight with a primary antibody at a dilution of 1:1,000 for mouse monoclonal anti-hMSH2 antibody from CalBiochem (La Jolla, CA), goat monoclonal anti-hMSH6 antibody from Santa Cruz Biotechnology (Santa Cruz, CA), and rabbit polyclonal anti-hMSH3 antibody. For mouse monoclonal anti-hMLH1 antibody from PharMingen (San Diego, CA), a dilution of 1:500 was used; for mouse monoclonal anti-hPMS2 antibody from PharMingen a dilution of 1:250 was used. Blots were washed with a large volume of PBS-T and incubated with 1:7,500 dilution of a horseradish peroxidase-conjugated secondary anti-mouse IgG antibody (Amersham Pharmacia Biotech) for hMSH2, hMLH1, and hPMS2; a 1:5,000 dilution of anti-goat IgG antibody (Santa Cruz Biotechnology) for hMSH6; and a 1:5,000 dilution of anti-rabbit IgG antibody (Santa Cruz Biotechnology) for hMSH3. To evaluate the absolute amount of each MMR protein in several cell lines, we compared the band densities with the standard curve obtained by each serially diluted purified hMutS-, hMutS-, and hMutL- heterodimer proteins provided by Dr. Giancarlo Marra and Dr. Joseph Jiricny. The amounts of the purified proteins were measured using the BCA protein assay kit (Pierce), and normalized by Coomassie Blue staining.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Semi-quantification of Human MMR mRNA by Multiplex RT-PCR-- We first used conventional RT-PCR for each DNA MMR component with each primer set and confirmed that the primers did not produce nonspecific bands (data not shown). Subsequently, we standardized the multiplex RT-PCR for simultaneous evaluation of the six human MMR genes (Fig. 3). A semiquantitative analysis was achieved by comparing the relative intensity of each band with the intensity obtained from the -actin mRNA. All cancer cell lines, irrespective of the tissue of origin, showed a higher level of all MMR transcripts than the normal human fibroblast lines.

View larger version (48K): [in this window] [in a new window]   Fig. 3.   Multiplex RT-PCR for six human MMR mRNA. Each band density was standardized to -actin. The amounts of mRNAs in each cell line were compared in reference to SW480.

Validity of CQ-RT-PCR-- For generating the RNA standard, we constructed the pMMR plasmid based on the six MMR multiplex PCR primer sequences. We confirmed the validity of CQ-RT-PCR by comparing the measured values obtained from 50, 100, 150, and 200 ng of total RNA collected from the SW480 cells (Fig. 4). An exact linear relationship was observed between the amounts of total RNA and the measured MMR mRNA. A reliable range for the intra-assay coefficient of variation was generated for measuring each MMR mRNA using the formula: CV = s/ × 100% (s, standard deviation; , mean of each value). This provided a CV of 9.3% for hMSH2, 24.3% for hMSH3, 33.4% for hMSH6, 7.8% for hMLH1, 17.4% for hPMS2, and 34.8% for hPMS1.

View larger version (43K): [in this window] [in a new window]   Fig. 4.   A linear relationship was demonstrated between total RNA and the hMSH2 mRNA measured in SW480 with a reliable range of coefficient of variation (CV). The intra-assay CV for the measurement of each MMR mRNA was 9.3% for hMSH2, 24.3% for hMSH3, 33.4% for hMSH6, 7.8% for hMLH1, 17.4% for hPMS2, and 34.8% for hPMS1. The arrow indicates the isomolar point.

For the cDNAs generated from each target mRNA and RNA standard, the exact quantification of the mRNA copy number by CQ-RT-PCR might vary due to differential efficiency of PCR amplifications. We evaluated this differential by using the full-length cDNAs (Fig. 5). We found that the value obtained by CQ-RT-PCR was overestimated although the relationship between the measured and real values was consistently linear. To determine the exact copy number of mRNA, we adopted correction factors from the slope of the linear regression equation: hMSH2, 0.26; hMSH3, 0.38; hMSH6, 0.26; hMLH1, 0.32; and hPMS2, 0.52. The correction factor for hPMS1 was not obtained because its full-length cDNA was not available. Finally, the absolute copy number could be calculated by multiplying the measured value obtained by the CQ-RT-PCR with the correction factor for each gene.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Correction factors. To establish the accuracy of the absolute quantification by competitive quantitative RT-PCR, we compared the equivalent measured amount using the cDNA from the RNA standard against the full-length MMR c-DNA. Based on the amount of each full-length MMR c-DNA, correction factors were derived for the value measured by competitive quantitative RT-PCR. The slope of the linear regression equation was adopted as a correction factor. The obtained correction factors are as follows: hMSH2, 0.26*; hMSH3, 0.38; hMSH6, 0.26; hMLH1, 0.32; and hPMS2, 0.52.

Quantification of Human MMR mRNA by CQ-RT-PCR-- Comparing the amounts of mRNA among different genes using conventional quantification methods, such as Northern blot analysis or quantitative RT-PCR, is virtually impossible. However, the RNA standard generated with the pMMR plasmid allowed us to achieve this comparison since we could reach the absolute, rather than the relative, quantification (Table II).

We determined that the mRNA amounts of all the components of the MMR system were 3 to 4 times higher in cancer cells than in normal fibroblasts, which correlates well to our initial observations when we used multiplex RT-PCR alone. Moreover, we observed that the hMSH2 mRNA was the most abundant of all human MMR mRNAs, whereas the amount for the hMSH6 mRNA was 3-5 times higher compared with hMSH3. The mRNAs of hMutS, especially of hMutS- (composed of hMSH2 and hMSH6), were 6-14 times higher than those of hMutL-. Among the components of hMutL-, hMLH1 mRNA was slightly more abundant than hPMS2 mRNA, although this value was not definite.

Steady-state Levels of Human MMR Proteins in MMR-proficient Cell Lines-- In an effort to correlate the concentration of mRNA with protein, we evaluated the protein levels among the different MMR components. Using the purified proteins of the MMR system, we estimated the amounts of each MMR protein in cells by comparing these intensities to the serially diluted purified proteins (Fig. 6, Table III). We normalized the absolute amounts of purified protein by Coomassie Blue staining and BCA protein assay. The purified proteins were prepared in their heterodimer states as hMutS-, hMutS-, and hMutL-, which ensures maximal protein stability. As previously reported, the ratios between hMSH2 and hMSH6 (in hMutS-), hMSH2 and hMSH3 (in hMutS-), and hMLH1 and hPMS2 (in hMutL-) were stoichiometrically 1:1 when they were heterodimerized (7, 11, 22).

View larger version (46K): [in this window] [in a new window]   Fig. 6.   Western analysis for each human MMR protein. The left seven lanes show the bands for each serially diluted purified MMR protein. MMR proteins in each cell line were compared by reference to the same amounts of each purified protein. hMSH2 protein was most abundant and approximated the amounts of hMSH6 plus hMSH3. hMutS- proteins were in excess of hMutL- proteins. hMLH1 protein was more abundant than hPMS2. All MMR proteins from cancer cell lines were consistently more abundant than those for the normal fibroblast cell lines (MRC5 and WI38).

The steady-state levels of MMR proteins in MMR-proficient cells revealed that hMSH2 is the most abundantly expressed MMR protein. hMSH6 was expressed 4-12 times as abundantly as hMSH3, and the sum of hMSH3 and hMSH6 proteins was approximately equal to the amount of hMSH2 protein. hMSH2 (an obligatory component of the DNA mismatch recognition complex) was expressed 3-5 times more abundantly than hMLH1 (an obligatory component of the system that responds to or signals mismatch recognition). hMLH1 was expressed 1.5-2.5 times as abundantly as hPMS2. Probes were not available to measure the expression of hPMS1. Steady-state levels of mRNA expression correlated well with protein expression in all instances.

The Interactive Regulation of MMR mRNA and Protein Levels in MMR-deficient Cell Lines-- To evaluate how the MMR system is regulated in an MMR-deficient setting, we examined several colon cancer cell lines with known MMR mutations and compared them with MMR proficient cells (Fig. 7, Table II). As expected, for LoVo cells, which express a truncated hMSH2 protein due to a large deletion encompassing exons 3 through 8 (23), we did not amplify any mRNA sequence since our products spanned from exons 2 through 4 (from nucleotide 355 to 798 of the complete coding sequences). In SW48 cells, mRNA expression for hMLH1 was not detected either, due to methylation of the promoter (24). In HCT116 cells that carry a mutation of hMLH1 at codon 252 of exon 9 resulting in a truncated protein (23), a faint PCR product was visualized, which may reflect the instability of the mRNA due to protein truncation. However, in HCT116+ch3 cells, which express hMLH1 protein by complementation of a wild-type gene on chromosome 3 (20), hMLH1 mRNA was visualized well. Interestingly, in HCT116 cells, hPMS2 protein was not detected although the mRNA was intact. When the hMLH1 protein was restored, as in HCT116+ch3 cells, hPMS2 protein also reappeared with no significant change in the hPMS2 mRNA level. In SW48, the hPMS2 protein expression was weak although the mRNA was fully expressed. DLD1 and HCT15 cells, which both carry a frameshift mutation in the hMSH6 gene (23), showed decreased levels of hMSH6 mRNA and no detectable hMSH6 protein. As was seen for the hMLH1 gene in HCT116 cells, the decrease of hMSH6 mRNA might reflect the instability of the mRNA due to protein truncation. As expected, hMSH3 protein was increased in all of these cell lines due to the higher probability of binding with hMSH2, which increases protein stability. hMSH3 mRNA was also 2-3-fold increased in HCT15 compared with the MMR-proficient cells (SW480, CaCo2, HEL, and HeLa). This increase suggests that hMSH3 is transcriptionally up-regulated to compensate for the hMSH6 deficiency. In LoVo cells, hMSH3 mRNA was also increased 3-fold, however, neither hMSH3 nor hMSH6 proteins were detectable, probably due to destabilization after failing to form heterodimers with the hMSH2 protein.

View larger version (92K): [in this window] [in a new window]   Fig. 7.   A, multiplex RT-PCR for SW480 and the MMR-defective cell lines; B, Western analysis for the same cell lines. hMLH1-mutated HCT116 shows a very faint mRNA with no protein at the proper size, and undetectable hPMS2 protein. In the hMLH1-corrected HCT116+ch3, hMLH1 mRNA and protein were restored, and hPMS2 protein was also restored with no significant changes in mRNA levels. SW48, in which the hMLH1 gene is silenced by promoter hypermethylation, show no mRNA or protein, and undetectable hPMS2 protein, either. LoVo cells show no hMSH2 mRNA and protein. hMSH3 and hMSH6 protein were also nearly undetectable. hMSH6-mutated DLD1 and HCT15 cells show a faint hMSH6 mRNA with no protein, and interestingly, increased hMSH3 mRNA and protein.

Transcriptional Up-regulation of Human MMR System in SV40transformed WI38 Fibroblasts-- As shown in Figs. 3 and 6, cancer cell lines derived from several different tissues (colon, uterine cervix, and blood) have higher levels of both MMR mRNAs and proteins in comparison with non-neoplastic lung fibroblasts. We then tested whether transformation of these fibroblasts leads to up-regulation of the MMR system, a more direct comparison of the impact of neoplastic change on these genes. Although the SV40-transformed variant 13 (WI38/VA13) of the WI38 human fibroblast is not a malignant cell line, SV40 transfection results in cell immortalization, a higher rate of cell proliferation, and chromosomal instability (19). All of the MMR mRNAs and proteins examined were expressed 1.5-5-fold higher in WI-38/VA13 cells than in the normal counterpart (Fig. 8, Table II).

View larger version (74K): [in this window] [in a new window]   Fig. 8.   A, multiplex RT-PCR; and B, Western blot of MMR system of WI38 and SV40-transformed WI38/VA13 fibroblasts. In the transformed fibroblast line, All of the MMR components are expressed at an increased level. C, Coomassie Blue staining confirmed same loading of total proteins.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The DNA MMR system has been extensively studied in prokaryotes and yeast, but the human system is more complex and less well understood. The stoichiometry and quantitative interaction among the components of human DNA MMR are assumed to be similar to that in lower organisms (1). However, these issues have not been directly studied for all of the components of the system. Quantification of mRNA levels is necessary to evaluate the regulation of MMR gene expression. Northern blot hybridization, a conventional method for quantifying mRNA levels, is of limited value when sample sizes are small and the message is minimally expressed. We used multiplex RT-PCR to measure the relative amounts of mRNA for multiple individual genes between different samples. However, this technique is not sufficiently accurate to measure absolute transcript numbers. We developed a novel technique, CQ-RT-PCR using an internal RNA standard to measure the steady-state relationships among the components of the MMR system within individual cell lines. Furthermore, we used purified proteins as standards to measure absolute amounts of MMR proteins in individual cell lines.

In six different DNA MMR-proficient human cell lines, we performed RT-PCR to determine relative mRNA levels for each cell. We were able to measure absolute amounts of DNA MMR mRNA by CQ-RT-PCR for five of these genes; we also determined the steady-state expression of the same five proteins. In DNA MMR-proficient cell lines, the relationship between mRNA and protein for each MMR gene was linear, suggesting that the level of each protein was primarily determined by transcription, with similar kinetic characteristics for each of the components of this system.

In DNA MMR-proficient cell lines, the hMutS components were expressed substantially more abundantly than the hMutL components. The hMSH2 protein is an essential component of the human MMR system and is expressed 3-5 times more abundantly than hMLH1. hMSH2 must heterodimerize with either hMSH3 or hMSH6 for mismatched DNA to be recognized (5), and presumably a complex between mismatched DNA and hMutS proteins transduces a signal through the hMutL system to initiate DNA repair by ADP-ATP exchange (10, 25). The precise role of the hMutL complexes is not yet completely understood. In E. coli, MutL is thought to accelerate an ATP-dependent translocation of the MutS-MutL complex to a hemimethylated GATC Dam site bound by MutH to achieve strand-specific repair (3, 9). Recent data have suggested that the hMSH2-hMSH6 heteroduplex may function as a molecular switch. Fishel (10) proposed that hMutL complexes may act as an adenine nucleotide regulator for hMutS complexes in a signal transduction series. A partial crystal structure of the N-terminal fragment from Escherichia coli suggests that MutL is homologous to the ATPase-containing DNA gyrase, which binds to DNA directly as does MutS (26, 27). Coimmunoprecipitation experiments indicate that the hMLH1-hPMS2 heterodimer interacts with hMSH2, requiring mispaired DNA and ATP (28). In every model of DNA MMR, the MutS components serve to recognize mispaired DNA, and MutL components function downstream to complete repair. The stoichiometric interaction of MutS and MutL is less clear. Since MutL proteins are unnecessary as long as mispaired nucleotides are not recognized, the initial complex may serve to initiate a repair cascade through a smaller number of MutL proteins.

Understanding the steady-state relationships among the components of the DNA MMR system, in addition to informing their functional roles in repair, may provide insights into hereditary nonpolyposis colorectal cancer (HNPCC) as well. Among the most important causes of HNPCC are germline mutations in hMSH2 (29, 30), an obligatory partner with either hMSH6 or hMSH3 in the mispair recognition complex (5). In fact, hMSH2 and hMLH1 germline mutations account for most of the highly penetrant forms of HNPCC (29, 30). hMSH6 is quantitatively the most important protein partner with hMSH2 and is expressed 4-12 times more abundantly than hMSH3. Germline mutations in hMSH6 account for relatively fewer cases of familial colorectal cancer than can be attributed to hMSH2 and hMLH1(29, 30), and the phenotype is relatively attenuated in the case of hMSH6 germline mutations (31). To date, germline mutations in hMSH3 have never been linked to familial colorectal cancer (29, 30).

Recently, the hMLH1-hPMS1 heterodimer has been purified and termed hMutL- (32), and a new human DNA MMR component hMLH3, which also forms a complex with hMLH1, has been reported (4). In our study, hMLH1 was consistently detected at higher levels than hPMS2, particularly at the protein level. Unfortunately, probes were not available to accurately quantitate hPMS1 and hMLH3 at the message or protein level. We speculate that the combined amounts of hPMS1, hPMS2, and hMLH3 proteins might be approximately equal to hMLH1.

We investigated in detail the interactive regulation of each component of the MMR system in MMR-deficient cell lines missing different components of the system. In LoVo cells, which are defective in hMSH2 (23), hMSH3 and hMSH6 proteins were also nearly undetectable (our data, and Ref. 11). Of interest, compensatory transcriptional up-regulation of the hMSH3 gene can be seen in the absence of hMSH6, as is the case in HCT15 cells. Similarly, absence of the hMLH1 affects expression of the hPMS2 protein, as seen in HCT116 cells. Interactions among the components of the MMR system can become particularly complicated. Mutations of hMSH2 or hMLH1, which functionally inactivate the system, can also facilitate somatic mutations of hMSH3 and hMSH6, both of which bear repetitive mononucleotide tracts in their coding sequences (33).

An inactivating mutation in hMSH6 in combination with intact hMSH2 is associated with an up-regulation of hMSH3, which may compensate, to some degree, for the defective hMutS- function. There is some redundancy in function between hMutS- and hMutS- as both can repair small insertion/deletion mispaired loops (34). Moreover, the lack of hMSH6 protein permits a compensatory increase in hMutS- by increasing the stability of hMSH3 following its dimerization with hMSH2 (11). Additionally, we confirmed the transcriptional up-regulation of hMSH3 in the presence of inactivating mutation in hMSH6. This observation helps explain the relatively weaker mutator effect of a mutated hMSH6 gene, and the observation that germline mutations in hMSH6 are associated with an attenuated HNPCC phenotype (31).

The colon cancer cell line HCT116 has undergone mutational inactivation of the hMLH1 gene and is defective in DNA MMR activity (23). The SW48 cell line expresses no hMLH1 due to hypermethylation of its promoter (24). The level of the hMutS- mRNAs and proteins are variable in these cell lines, indicating that the hMutS and hMutL systems do not quantitatively regulate each other. The stability of hPMS2 appears to depend upon the presence of an intact hMLH1 (our data and Ref. 35). As seen in the HCT116+ch3 cell line, the reconstitution of hMLH1 also restores expression of the hPMS2 protein.

We have previously reported that the human MMR system is regulated throughout the cell cycle in MMR-proficient cell lines (16). We noted that the MMR-proficient cancer cell lines showed higher levels of MMR mRNAs and proteins than we saw in the two normal lung fibroblast lines. The normal lung fibroblast cell line WI38 has been stably transformed and immortalized with the SV40 virus (19). The entire DNA MMR system underwent transcriptional up-regulation after SV40 transformation. The promoters of the MMR genes hMSH2, hMLH1, hPMS2, and hPMS1 have characteristic features of housekeeping genes, including abundant CG islands and no TATA box (36-39). However, these sequences show potential binding sites for transcriptional activators including AP-1 (in the case of the hMSH2 and hPMS1 promoters) (36, 38) and T-antigen (in the hMSH2 promoter) (38). The presence of these binding sites suggests that the MMR genes might be up-regulated in different physiological or biological settings. As cell cycledependent regulation of the MMR system has been previously reported (16), up-regulation of the MMR system in the synthetic and postsynthetic phases might reflect an increased need for active MMR function in actively proliferating cells. Increased recruitment of cells into the cell cycle from G₀ phase or an increased proportion of S and G₂/M phase cells by active proliferation may be responsible for up-regulation of the system.

In summary, we have studied the steady-state expression of human DNA MMR transcripts and proteins, where a hierarchy of expression reflects the relative roles of these proteins in DNA repair. hMSH2 is the most abundantly expressed of these proteins, and the presence of hMSH2 protein serves to stabilize hMSH6 and hMSH3. The defect of hMSH6 interactively up-regulates hMSH3 transcription. The hMSH6 protein is substantially more highly expressed than hMSH3, an observation worthy of additional functional exploration. hMLH1 also behaves like a chaperone of stability for the other hMutL components. The components of the hMutS system are expressed severalfold more than components of the hMutL system, which suggests the presence of a signaling cascade in MMR function. Finally, SV40 transformation of a fibroblast cell line led to an up-regulation of the entire MMR system. The DNA MMR system is regulated primarily at a transcriptional level; however, mutational inactivation of the obligatory components of the system, hMSH2 and hMLH1, lead to post-translational down-regulation of its heterodimerizing partners. These findings have implications for the physiological regulation of the DNA MMR system and for the interpretation of germline mutations in hereditary colorectal cancer.

	ACKNOWLEDGEMENTS

We thank Dr. Giancarlo Marra and Dr. Joseph Jiricny for providing plasmids hMSH3-pGEM7z(+) and hMSH6-pBSKS, purified proteins hMutS-, hMutS-, and hMutL-, and anti-hMSH3 antibodies.

	FOOTNOTES

^* This work was supported by the research service of the Department of Veterans Affairs and National Institutes of Health Grant RO1-CA72851 (to C. R. B.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed. Tel.: 858-822-0300; Fax: 858-822-0301; E-mail: crboland@ucsd.edu.

Published, JBC Papers in Press, March 29, 2000, DOI 10.1074/jbc.M001140200

	ABBREVIATIONS

The abbreviations used are: MMR, mismatch repair; CQ-RT-PCR, competitive quantitative reverse transcription-polymerase chain reaction; SV40, simian virus 40; HNPCC, hereditary nonpolyposis colorectal cancer.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Z. Hong, J. Jiang, K. Hashiguchi, M. Hoshi, L. Lan, and A. Yasui Recruitment of mismatch repair proteins to the site of DNA damage in human cells J. Cell Sci., October 1, 2008; 121(19): 3146 - 3154. [Abstract] [Full Text] [PDF]

				I. Marinovic-Terzic, A. Yoshioka-Yamashita, H. Shimodaira, E. Avdievich, I. C. Hunton, R. D. Kolodner, W. Edelmann, and J. Y. J. Wang Apoptotic function of human PMS2 compromised by the nonsynonymous single-nucleotide polymorphic variant R20Q PNAS, September 16, 2008; 105(37): 13993 - 13998. [Abstract] [Full Text] [PDF]

				I.-Y. Chang, M. Jin, S. P. Yoon, C.-K. Youn, Y. Yoon, S.-P. Moon, J.-W. Hyun, J. Y. Jun, and H. J. You Senescence-Dependent MutS{alpha} Dysfunction Attenuates Mismatch Repair Mol. Cancer Res., June 1, 2008; 6(6): 978 - 989. [Abstract] [Full Text] [PDF]

				V. O'Brien and R. Brown Signalling cell cycle arrest and cell death through the MMR System Carcinogenesis, April 1, 2006; 27(4): 682 - 692. [Abstract] [Full Text] [PDF]

				A. T. Stormorken, I. M. Bowitz-Lothe, T. Noren, E. Kure, S. Aase, J. Wijnen, J. Apold, K. Heimdal, and P. Moller Immunohistochemistry Identifies Carriers of Mismatch Repair Gene Defects Causing Hereditary Nonpolyposis Colorectal Cancer J. Clin. Oncol., July 20, 2005; 23(21): 4705 - 4712. [Abstract] [Full Text] [PDF]

				H. Nakagawa, J. C. Lockman, W. L. Frankel, H. Hampel, K. Steenblock, L. J. Burgart, S. N. Thibodeau, and A. de la Chapelle Mismatch Repair Gene PMS2: Disease-Causing Germline Mutations Are Frequent in Patients Whose Tumors Stain Negative for PMS2 Protein, but Paralogous Genes Obscure Mutation Detection and Interpretation Cancer Res., July 15, 2004; 64(14): 4721 - 4727. [Abstract] [Full Text] [PDF]

				J. Plaschke, S. Kruger, B. Jeske, F. Theissig, F. R. Kreuz, S. Pistorius, H. D. Saeger, I. Iaccarino, G. Marra, and H. K. Schackert Loss of MSH3 Protein Expression Is Frequent in MLH1-Deficient Colorectal Cancer and Is Associated with Disease Progression1 Cancer Res., February 1, 2004; 64(3): 864 - 870. [Abstract] [Full Text] [PDF]

				G. Plotz, J. Raedle, A. Brieger, J. Trojan, and S. Zeuzem N-terminus of hMLH1 confers interaction of hMutL{alpha} and hMutL{beta} with hMutS{alpha} Nucleic Acids Res., June 15, 2003; 31(12): 3217 - 3226. [Abstract] [Full Text] [PDF]

				V. T. Mihaylova, R. S. Bindra, J. Yuan, D. Campisi, L. Narayanan, R. Jensen, F. Giordano, R. S. Johnson, S. Rockwell, and P. M. Glazer Decreased Expression of the DNA Mismatch Repair Gene Mlh1 under Hypoxic Stress in Mammalian Cells Mol. Cell. Biol., May 1, 2003; 23(9): 3265 - 3273. [Abstract] [Full Text] [PDF]

				X. Wu, J. L. Platt, and M. Cascalho Dimerization of MLH1 and PMS2 Limits Nuclear Localization of MutL{alpha} Mol. Cell. Biol., May 1, 2003; 23(9): 3320 - 3328. [Abstract] [Full Text] [PDF]