An Integrated Microsatellite Length Analysis Using an Automated Fluorescent DNA Sequencer

[CANCER RESEARCH 56, 2688-2691. June 15, I996| Advances in Brief

Yasushi Toll,' Eiji Oki, Shinya Oda, Masaaki Tomoda, Shinichi Tomisaki, Yuji Ichiyoshi, Shinji Ohno, and Keizo Sugimachi Cancer Center, Kyushu University Hospital [Y. T., E. O., S. Od.], and Department of Surgery II, Faculty of Medicine, Kyushu University Â¡M.T.. S. T.. Y. !.. S. Oh.. K. SJ 3-1-1 Maidashi, Higashi-ku, Fukuoka 812. Japan

Abstract suming, and the usage of the radioisotope considerably limits the routine application of MI in the clinical field. Furthermore, there seem Analyzing microsatellite instability (MI) in malignant tumors is thought to be problems in the objective and reliable assessment of MI because to be useful for screening cancer patients to identify those patients with a lane-to-lane variation may sometimes interfere with the precise as higher risk of developing second malignant tumors. In this paper, we report a new, accurate, and efficient method of detecting MI using an sessment of the difference in mobility of the PCR products, which automated fluorescent DNA sequencer and a computer that automatically makes it difficult to objectively assess or quantitate the strength of the calculates the size, height, and area of each fluorescent product, making it bands. In this report, we describe a new, accurate, and efficient possible to assess MI more accurately and more rapidly. The primers for method of detecting MI in human malignancies, using PCR with two amplification of each microsatellite locus are labeled by two different primers labeled by different fluorescent dyes and an automated fluo fluorescent dyes, rox (red) and fam (blue). The rox-labeled primer was rescent DNA sequencer. This method may give a more rapid and used for the tumor, whereas the fam-labeled primer was used for the accurate sizing of the PCR products of microsatellites, making the corresponding normal tissue. Two amplified products from both the routine assessment of MI possible in many clinical fields. tumor and the normal tissue were co-loaded into a single lane of the sequencing gel and were analyzed. MI could be detected based on the Materials and Methods presence of different waving patterns. Furthermore, several loci could also be analyzed simultaneously for MI in a single lane. Using this method, we Surgical Specimens. All the samples were surgically obtained. Gastric examined the frequency of MI in gastric cancer. The results showed that cancers and the corresponding normal mucosa were surgically resected from 5 of 22 (22.7%) gastric cancers were Mi-positive, which corresponds to the 21 patients at Kyushu University Hospital during the period from 1994 to findings of previous reports that used the radioisotopic method. The 1995. Immediately after surgical resection, the carcinoma and the paired improved method may open up the possibility of performing routine normal gastric epithelium were snap-frozen in liquid nitrogen. The carcinoma examination of MI in many cancer patients and offers hope for the tissue specimens were obtained from a portion of the resected tissue while potential clinical application of MI analysis as a follow-up evaluation of omitting any necrotic or ulcerated parts that were clearly separated from the cancer patients. normal adjacent tissue. Preparation of Genomic DNA. Genomic DNA was extracted from the Introduction surgically resected gastric carcinoma and the corresponding normal gastric epithelium, using the standard proteinase K and phenol/chloroform methods. Microsatellites are short, repeated DNA sequences found inter The concentration of the extracted DNA was then measured using a model spersed throughout the genome ( 1), and they have been used for the DU7500 spectrophotometer (Beckman Instruments. Fullerton, CA). analysis of genetic linkage and alÃeleloss because they show length Primers for PCR of Microsatellites. In this study, we used the primer polymorphisms (2). MI2 was recently reported to be associated with pairs for four microsatellite markers (including D2S123, D5S107, D11S904, and D13S175). All of these markers were dinucleotide repeats. The nucleotide both hereditary nonpolyposis colon cancers (3) as well as a wide sequences of these primers have been described previously (18). One of the variety of other human malignancies not associated with hereditary paired primers was labeled with two different fluorescent dyes (i.e., rox or nonpolyposis colon cancer, including: gastric cancers (4-9); sporadic fam), which made it possible to detect only one strand of amplified DNA. The colorectal cancers (10, 11); pancreas cancers (4); lung cancers (12); fluorescent-labeled primers were purchased from the Takara Shuzo Co., Ltd. endometrial cancers (13); and so on. Furthermore, an abnormality in (Tokyo, Japan). the DNA mismatch repair system increases the risk for the develop Fluorescent PCR, PAGE, and Computer Analyses of MI. The cycling ment of multiple primary malignant tumors (14), and MI is caused by conditions of fluorescent PCR are the same for all of the loci examined in this study and consisted of 35 cycles at 95Â°Cfor 30 s for denaturing, at 55Â°Cfor defects in such DNA mismatch repair genes as HMSH2, hMLHl, 30 s for annealing, and at 72Â°Cfor 30 s tor extension. PCR was performed in hPMSl, and hPMS2 (15-17). Therefore, analysis of MI in malignant a 50-/J.1 solution |10 mM Tris-HCl (pH 8.3). 50 mM KC1. 1.5 min MgCl2, 200 tumors is thought to be useful for rapid screening to detect any /XMdeoxynucleotide triphosphate. 0.5 units of Taq polymerase (Takara Shuzo abnormalities or mutations found in these mismatch repair genes, Co., Ltd.), 25-50 ng of genomic DNA. and 10 pmol of each primer], using a identifying a population of patients who are at higher risk of devel GeneAmp PCR System 9800 thermal cycler (Perkin-Elmer Cetus, Norwalk, oping second malignant tumors (14). CT). The fluorescent PCR products were separated using 6% urea-polyacryl A PCR using a radioisotope followed by urea-polyacrylamide de amide denaturing gels in a model 373A automated fluorescent DNA sequencer naturing gel electrophoresis and autoradiography has been used to (Applied Biosystems, Foster City, CA). The GeneScan Analysis software detect MI in malignant tumors. However, this method is time-con- (Applied Biosystems) was used for the assessment of MI. Statistical Analyses. The statistical analyses were performed using the )f test and Fisher's exact test. P < 0.05 was judged to be statistically significant. Received 3/11/96; accepled 4/29/96. The costs of publication of this article were defrayed in part by the payment of page Results and Discussion charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' To whom requests for reprints should be addressed. We used two different fluorescent dyes for the amplification of one 2 The abbreviation used is: MI. microsatellite instability. microsatellite locus, and then we electrophoresed these two PCR 2688

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1996 American Association for Cancer Research. MICROSATKLLITE LENGTH ANALYSIS BY FLUORESCENT PCR products in one lane. A rox-labeled primer was used to amplify the cently by tamra dye (yellow) was co-loaded for each lane (data not microsatellites in genomic DNA extracted from the tumor samples shown). This enabled us to measure the size of the PCR product at a (which generated red signals), whereas a fam-labeled primer was used 1-bp accuracy. In Fig. 2, six different PCR products of three inde to amplify the microsatellites in genomic DNA from the correspond pendent microsatellite loci (D5SI07, DÃŒIS904,and D13SI75) were ing normal tissue (which generated blue signals). Because the strength analyzed simultaneously in one lane. This case was classified as of fluorescence differs between rox and fam, we estimated the most Mi-positive (as shown in Fig. 2) because a novel alÃelethat was 2 bp appropriate loading volume for each PCR product and the proportion longer than the largest alÃelein the normal counterpart appeared at the of the mixture of PCR products derived from tumor DNA and normal D5SÃŒ07locus in the cancer DNA. These findings demonstrate that MI DNA. As a result, approximately 1 volume of the PCR products in several loci can be assessed in one lane (Fig. 2/4), providing a very labeled by fam to 5-10 volumes of those labeled by rox resulted in the efficient method to screen many samples for MI at the same time. best condition for assessment of MI (Fig. 1). We thus decided to load Furthermore, because the strength of the signals are presented as peak 1 fi\ of the 50-/nl rox-labeled PCR product for the tumor and 0.1 jal heights by the computer (Fig. 2ÃŸ),this assay is thought to be more of the 50-ju.l fam-labeled PCR product for the normal counterpart. quantitative than the radioisotope-based assay used previously. The Furthermore, we also co-loaded the amplified PCR products of several slightly larger peak sizes of the PCR products using the rox-labeled microsatellite loci to one lane. primers (compared to those of the PCR products using the fam-labeled In this study, we used the genomic DNAs of human gastric cancer primers) are considered to be caused by the differences in polarity and specimens and the corresponding normal gastric mucosa from the molecular weight between the two fluorescent dyes used for primer same patient. Fig. 2 demonstrates the representative results showing labeling. MI. To measure size precisely, the DNA size-marker labeled fluores- Therefore, we decided to examine the incidence of MI in gastric cancers in Japan, using our system. Microsatellite loci examined were D2S123, 055/07, D1IS904, and D13S175. MI was observed in 5 of 2000-1 22 (22.7%) gastric cancer cases in D2S123 (2/22), D5SÃŒ07(3/22), or D2S123 DI1S904 (1/22). One case had MI in both D5S107 and D1IS904, whereas no MI was found in DI3S175. The raw data are presented in Figs. 2 and 3. The observed incidence of MI was similar to that 1500 reported in previous studies (6-8), although it was somewhat lower 0) than some of the reported findings (4, 5, 9). The clinicopathological > features of Mi-positive and -negative cases are shown in Table 1. 0) Lymphatic involvement was observed in all Mi-positive cases, whereas it was observed in only 7 of 17 Mi-negative cases (P < 0.05). (0 Furthermore, all Mi-positive cases had lymph node mÃ©tastases,were 0) a in the more advanced stages, and were not histologically differenti 500 ated, although these findings were not statistically significant. The Mi-positive tumors were localized on the more distal portions of the stomach as compared to the Mi-negative tumors (P < 0.05). To make these results more convincing, however, more cases need to be ana 0.0 0.2 0.4 0.6 0.8 lyzed. Previous reports have demonstrated confusing findings regard PCR product ing the clinicopathological comparison between Mi-positive and -negative groups. Han et al. (5) reported that poorly differentiated adenocarcinomas of the stomach showed more MI than well-differ 8000 D5S107 entiated ones, whereas Rhyu et al. (4) showed no significant correla tion between MI and the degree of differentiation. Nakashima et al. (8) reported a significant correlation between MI and negative lymph node metastasis and negative lymphatic involvement, in strong con trast to our results. These inconsistencies might be caused by the lack of an objective and accurate method for detection of MI in addition to the differences in both the geography and the various races examined. We believe that our method will be useful as a more objective and easier method for the detection of MI because the automated DNA sequencer and the computer software (GeneScan) are widely available in many laboratories throughout the world. The abnormality in the DNA mismatch repair system causing Ml increases the risk for development of multiple primary malignant tumors (14). Therefore, cancer patients with Ml must be carefully followed-up for the possible early detection of second primary cancer. PCR product For this purpose, the screening method to detect MI should be rapid, easy, accurate, and objective. In this paper, we have described a new Fig. I. Determination of the optimal mixture ratio of fam-based and rox-hased PCR method for the detection of MI, using fluorescent PCR. By using this products of the microsatcllite loci. Various amounts of the PCR products of microsatellite new method, MI can be assessed and quantitated very effectively by loci were electrophoresed with an automated DNA sequencer: the peak heights were measured with GeneScan software, and the values from two experiments were plotted. computer, making the assessment more objective and accurate in The results of two microsatellite loci ID2S123 and D5SIÃ›7) are shown as representative comparison to the radioisotopic method used mainly in previous findings. This figure shows that the comparable fluorescent strength of the PCR products for the computer analysis is achieved when one volume of the fam-lubeled PCR products studies. During the preparation for this paper, Cawkwell et al. (IO) is co-loaded to a lane with about 5-10 volumes of the rox-labeled PCR products. reported a similar method, the basis of which is essentially the same 2689

D13S175 D5S107 D11S904

Â¿600 Â¿800 2000 Â¿800, ,300,0

Fig. 2. Integrated microsatellite length analysis. An example of gastric carcinoma with MI is shown. In A. three independent microsatellite loci (D5S107. D1IS904. and D13SI75) were simultaneously analyzed in a single lane. The red signal was derived from the genomic DNA of the gastric carcinoma amplified with rox-labeled primers, whereas the blue I Lane 6: 9363N I Uni i- 9353T signal originated from the genomic DNA of corresponding normal mucosa amplified with lam-labeled primers. In B. the signals expanded to the x-axis of D5S107, and the results of the GeneScan analysis are also shown. A B novel alÃele(arrow) was observed in the signal from the carcinoma (red) in D5S107 the D5SI07 loci, as compared to the paired signals of the normal mucosa 2265 (blue). PeakNo1 Size148.31 Height177 Area1410 (blue) 2 (blue) 150.41 520 4485 3 (blue) 152.50 211 2031 4(blue)1 154.49148.99 436135 40911102

(red) 2 (red) 151.06 415 4017 3 (red) 153.14 122 1185 4 (red) 155.12 280 2579 5 (red)Peak157.16Peak 248Peak 2551

Casel D5S107 D11S197 Table 1 Clinicopafhological features of gastric carcinomas with regard to Ml Â¿120 Â¿170 Â¿220 Â¿270 Â¿320 j. 675 JZ7ÃŽS Â¿775 Â¿82S Â¿8 MIVariablesSexMaleFemaleLocation

17)11657237107111611HI713417(1170710PNS"P<0.05NSNSNS/><0.05NSNSNS"=5)5(10l4Â»(1514050532504105Negative(n=

tumorUpperMiddleLowerWholeHistologyDifferentiatedUndifferentiatedSerosa!of

invasionNegativePositiveHistologie Case 2 Case 3 D5S107 D2S123 metastasisNegativePositiveLymphaticlymph node Â¿805 ,2955 Â¿90S Â¿955 Â¿005 involvementNegativePositiveVascular

involvementNegativePositivePeritoneal

disseminationNegativePositiveLiver

metastasisNegativePositiveHistological

stageI +IIIII + IVPositive(n Fig. 3. Presentation of the Mi-positive cases. The raw data from the three cases other NS. not statistically significant. than the one shown in Fig. 2 are presented. Case I showed MI at two loci. Arrow, novel alÃele;red, carcinoma; blue, normal mucosa.

2690

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1996 American Association for Cancer Research. MICROSATELLITE LENGTH ANALYSIS BY FLUORESCENT PCR as ours. However, our method is more efficient and demonstrative 5. Han. H. J.. Yanagisawa. A.. Kalo, Y.. Park, J. G., and Nakamura, Y. Genetic instability in pancreatic cancer and poorly differentiated type of gastric cancer. because the PCR products from both the tumor and the normal tissue Cancer Res., 53: 5087-5089. 1993. can be analyzed in a single lane with two different colors, and the 6. Mironov. N. M.. Aguelon. M. A.. Potapova. G. I., Omori. Y.. Gorbunov. O. V., problems of lane-to-lane variation can be avoided. Furthermore, we Klimenkov, A. A., and Yamasaki. H. Alterations of (CA)â€žDNA repeats and tumor have succeeded in analyzing three microsatellite loci in a single lane suppressor genes in human gastric cancer. Cancer Res.. 54: 41-44, 1994. 7. Tamura. G.. Sakata. K.. Maesawa. C.. Suzuki, Y.. Terashima. M.. Satoh, K., (Fig. 2). Because the sequencing gel for the automated DNA se Sekiyama, S.. Suzuki. A.. Eda. Y., and Satodate. R. Microsatellite alterations in quencer we used had 24 lanes/gel, a maximum of 12 patients can be adenoma and differentiated adenocarcinoma of the stomach. Cancer Res., 55: 1933- assessed in one assay when 6 microsatellite loci per patient are 1936, 1995. 8. Nakashima. H.. Inoue. H.. Mori, M., Ueo, H.. Ikeda, M.. and Akiyoshi. T. Microsat examined. If the products (with an appropriate mixture ratio of rox- ellite instability in Japanese gastric cancer. Cancer (Phila.). 75: 1503-1507, 1995. and fam-based PCR) are coprecipitated with alcohol and then con 9. Chong. J. M., Fukayama. M.. Hayashi. Y., Takizawa. T.. Koike. M.. Konishi. M.. Kikuchi-Yanoshita. R.. and Miyaki. M. Microsatellite instability in the progression of centrated, it would be possible to examine even more microsatellite gastric carcinoma. Cancer Res., 54: 4595-4597. 1994. loci in the same lane. On the other hand, the use of another dye with 10. Cawkwell, L.. Li, D., Lewis, F. A., Martin, I., Dixon, M. F., and Quirke, P. more fluorescent strength than rox (i.e., hex, which produces a green Microsatellile instability in colorectal cancer: improved assessment using fluorescent PCR. Gastroenterology. 109: 465-471, 1995. signal) may reduce the loading volume/locus and make it possible to 11. Aaltonen, L. A., Peltomaki. P.. Mecklin, J. P., Jarvinen, H.. Jass, J. R.. Green, J. S., apply more samples to one well of the gel. Lynch, H. T.. Watson. P.. Tallqvist, G.. Juhola, M., Sistonen, P.. Hamilton, S. R., In conclusion, we have described a new and improved method for Kinzler. K. W.. Vogelstein. B., and de la Chapelle, A. Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Can the detection of MI that is more accurate, objective, and rapid than the cer Res., 54: 1645-1648, 1994. method used previously. We believe that this new method may open 12. Shridhar. V.. Siegfried. J.. Hunt, J., Alonso, M. d. M., and Smith, D. I. Genetic up the possibility of performing routine MI examinations on many instability of microsatellite sequences in many non-small cell lung carcinomas. Cancer Res.. 54: 2084-2087. 1994. cancer patients and offers hope for the potential clinical application of 13. Risinger. J. I.. Berchuck. A.. Kohler, M. F., Watson. P., Lynch. H. T.. and Boyd. MI as a follow-up evaluation of cancer patients, which could lead J. Genetic instability of microsatellites in endometrial carcinoma. Cancer Res., 53: eventually to more accurate assessment and the increased significance 5100-5103, 1993. 14. Horii, A., Han, H. J.. Shimada. M.. Yanagisawa. A.. Kalo, Y., Ohta, H., Yasui, of MI in the clinical field. W.. Tahara, E., and Nakamura. Y. Frequent replication errors at microsatellite loci in tumors of patients with multiple primary cancers. Cancer Res., 54: 3373-3375. Acknowledgments 1994. 15. Fishel, R., Lescoe, M. K., Rao, M. R.. Copeland, N. G.. Jenkins, N. A., Garber, J., We thank Dr. Brian T. Quinn tor comments and Ms. J. Tsuchihashi for help Kane. M.. and Kolodner. R. The human mulalor gene homologue MSH2 and its association with hereditary nonpolyposis colon cancer. Cell. 75: 1027-1038, 1993. in the statistical analysis. 16. Bronner, C. E., Baker, S. M.. Morrison. P. T.. Warren. G.. Smith. L. G., Lescoe. M. K.. Kane. M., Earabino, C.. Lipford. J.. Lindblom. A.. Tannergard. P.. Bollag. References R. J.. Godwin. A. R.. Ward. D. C.. Nordenskjold, M.. Fishel. R., Kolodner, R., and 1. Jeffreys, A. J., Wilson. V.. Neumann. R.. and Keyte, J. Amplification of human Liskay. R. M. Mutation in the DNA mismatch repair gÃ¨nehomologue hMLHl is minisatellites by PCR: toward DNA fingerprinting of single cells. Nucleic Acids Res.. associated with hereditary nonpolyposis colon cancer. Nature (Lond.). 368: 258-261. 16: 10953-10971. 1988. 1994. 2. Weber, J. L.. and May. P. E. Abundant class of human DNA polymorphisms which 17. Nicolaides, N. C., Papadopoulos, N., Liu, B., Wei, Y. F.. Carter, K. C., Ruben, S. M., can be typed using PCR. Am. J. Hum. Genet., 44: 388-396. 1989. Rosen. C. A., Haseltine. W. A., Fleischmann, R. D., Fraser, C. M., Adams, M. D., 3. Aaltonen. L. A.. Peltomaki. P., Leach, F. S., Sistonen. P.. Pylkkanen. L.. Mecklin. Venter. J. C.. Dunlop. M. G., de la Chapelle, A., Vogelstein, B., and Kinzler, K. W. J. P., Jarvinen. H., Powell, S. M.. Jen. J.. Hamilton. S. R.. Peterson. G. M.. Kinzler. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature K. W., Vogelstein. B., and de la Chapelle. A. Clues to the pathogenesis of familial (Lond.), 371: 75-80. 1994. colorectal cancer. Science (Washington DC). 260: 812-816. 1993. 18. Weissenbach. J.. Gyapay. G.. Dib, C., Vignai, A.. Morissette. J.. and Millasseau. P. 4. Rhyu, M. G., Park, W. S.. and Meltzer. S. J. Microsatellite instability occurs A second-generation linkage map of the human genome. Nature (Lond.). 359: frequently in human gastric carcinoma. Oncogene, Â°:29-32, 1994. 794-801, 1992.

2691

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1996 American Association for Cancer Research. An Integrated Microsatellite Length Analysis Using an Automated Fluorescent DNA Sequencer

Yasushi Toh, Eiji Oki, Shinya Oda, et al.

Cancer Res 1996;56:2688-2691.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/12/2688

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/56/12/2688. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.