Summary. After distal gastrectomy in a patient withearly gastric cancer, 27 regional lymph nodes around thestomach were evaluated for the existence of metastasis.There was a 0IIa+IIc type tumor 2.0x1.5cm in size in thegastric angle of the lesser curvature according to theJapanese Classification of Gastric Carcinoma (JCGC).Histologically, the lesion extended no deeper than themuscularis mucosae. The cancer stage was so early thatno metastasis was expected to occur but a lymph nodewith metastasis was found in one lymph node along thecommon anterior hepatic artery (station No.8a). Thishistological type was a little different from that of aprimary tumor. The doctor began to suspect that thelymph node with metastasis might have been fromanother patient by mistake. Therefore, DNA typingusing the AmpFlSTR® Identifiler® kit was performed informaldehyde-fixed paraffin-embedded (FFPE) tissues:2 parts of gastric mucosa without cancer, one part ofgastric mucosa with cancer, 4 lymph nodes withoutmetastasis, and the lymph node station No.8a withmetastasis. STR typing was successful in 6~14 STR lociand amelogenin gene, and the detected STR type was thesame in all samples. Compared with the STR type usingDNA from the patient’s blood, the lymph node stationNo.8a was from the same patient. The lymph node withmetastasis turned out to be not from another patient.Therefore, we suggest that DNA typing using theAmpFlSTR® Identifiler® Kit for FFPE samples is usefulin such clinical cases.

Key words: DNA typing, AmpFlSTR® Identifiler® Kit,Gastric cancer, Lymph node metastasis, Formaldehyde-fixed paraffin-embedded (FFPE) tissues

Introduction

Recent advances in cancer screening, early detectionand therapy are remarkable. Early-stage cancer andlymph node metastases that had been overlookedformerly are nowadays found. However, if early-stagecancer with metastasis is found, a possible mistake thatsomeone may take a sample with metastasis for anotherpatient’s should be avoided. In that case proof that thesample is not another patient’s is necessary.

In general, tissues from an autopsy and tumor tissuefrom the surgery room are stored in 10~20% formalinfor a period of 3 days to 1 week, and then they areembedded in paraffin. Formalin is a solution offormaldehyde, and 10% formalin usually contains3.5~3.8% formaldehyde (HCHO). Early formalinfixation preserves the cytomorphology in tissues verywell, and the analysis of cell morphology usingimmunological techniques provides very importantinformation. However, it is known that the preservationof cytomorphology by protein-protein and protein-nucleic acid crosslinks causes DNA fragmentation,degradation and crosslinks among nucleic acids(Impraim et al., 1987; Tokuda et al., 1990; Noguchi etal., 1997). Furthermore, the oxidation of formaldehydeby dissolved oxygen produces a strong acid, formic acid,so the fixation solution becomes very acidic, and thenthe DNA of formalin-fixed tissues is thought to befragmented.

Generally, a polymerase chain reaction (PCR) that

Useful DNA typing using AmpFlSTR® Identifiler® Kit forformaldehyde-fixed paraffin-embedded (FFPE) tissues inearly gastric cancer patient with lymph node metastasisHisako Motani-Saitoh1, Hiroyuki Inoue1, Tohru Tanizawa2, Yoshihiro Nabeya3, Daisuke Yajima1, Mutsumi Hayakawa1, Yayoi Sato1, Yukio Nakatani2, Hisahiro Matsubara3 and Hirotaro Iwase1

1Department of Legal Medicine, Graduate School of Medicine, Chiba University, Inohana, Chuo-ku, Chiba, Japan, 2Department of

Diagnostic Pathology, Graduate School of Medicine, Chiba University, Inohana, Chuo-ku, Chiba, Japan and 3Department of Frontier

Surgery, Graduate School of Medicine, Chiba University, Inohana, Chuo-ku, Chiba, Japan

Histol Histopathol (2009) 24: 1139-1145

Offprint requests to: Hisako Motani, Department of Legal Medicine,Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. e-mail: hms1466@faculty.chiba-u.jp

http://www.hh.um.es

Histology andHistopathology

Cellular and Molecular Biology

amplifies the target region from fragmented anddegraded DNA has been used in forensic DNA typing.Recently, personal identification using STR genotyping,including CODIS 13 core loci, has been used in variousforensic laboratories in the world (Budowle et al., 1998,2001; Butler, 2005) and STR genotying has mainly beencarried out with the commercially available AmpFlSTR®

Identifiler® PCR Amplification Kit (AppliedBiosystems, Foster City, CA) (Applied Biosystems,2001; Holt et al., 2002; Butler, 2005). Using this kit, theSTR type in 15 STR loci, including the CODIS 13 lociand amelogenin gene, can be detected (Collins et al.,2004; Reid et al., 2004; Yoshida et al., 2005, 2006).

In this study, STR typing was performed using DNAsamples from formaldehyde-fixed paraffin-embedded(FFPE) tissues containing a lymph node with metastasisin an early gastric cancer patient (Romero et al., 1997;Zehner and Lasczkowski, 2000). We report thesuccessful personal identification by appropriatetemplate DNA in PCR amplification with theAmpFlSTR® Identifiler® Kit.

Case

The patient with early gastric cancer was female and62 years old. She underwent a distal gastrectomy(Billroth-I) with D2 lymph node dissection according tothe Japanese Classification of Gastric Carcinoma(JCGC) (Japanese Gastric Cancer Association, 1998;Koyanagi et al., 2004; Townsend et al., 2006).Macroscopic findings revealed no lymph nodemetastasis. The length of the resected lesser curvatureand the duodenum was 11 cm and 0.5 cm, respectively(Fig. 1). There was a 0IIa+IIc type tumor 2.0x1.5cm insize in the gastric angle of the lesser curvature whichcould be seen with the naked eye. Histologically, therewere well differentiated tubular adenocarcinomas, andmoderately differentiated tubular adenocarcinomas withfused glands (Kumar et al., 2003). The latter tumorelement formed nests and proliferated deeply as downgrowth, but the lesion extended no deeper than themuscularis mucosae (Fig. 2a,b). There were no venousinvasions or lymphatic vessel invasions.

However, a lymph node with metastasis was foundin the lymph node station No.8a out of 27 lymph nodesdissected. However, the histology of the lymph nodewith metastasis was a little different from that of theprimary tumor (Fig. 3a,b). In the primary tumor, thetumor element with high-grade atypical cells showedCK7(+), CK20(-), MUC5(-), MUC6(-), CD10(+),CDX2(+), and p53(-), and the tumor element with low-grade atypical cells showed CK7(-), CK20(+), MUC5(-),MUC6(-), CD10(-), CDX2(+), and p53(-). However, thelymph node with metastasis showed CK7(+), CK20(-),MUC5(-), MUC6(-), CD10(-), CDX2(-), and p53(-).While both tumor elements of a primary tumor showedKi-67(+) in almost all tumor cells, the percentage of Ki-67(+) cells was less than 50% in the lymph node withmetastasis. In addition, while the gastric cancer of the

primary tumor showed some degree of the intestinal typecell differentiation, the lymph node with metastasis didnot show any apparent intestinal type character.

It is possible that this lymph node was metastasizedfrom the cancer of another organ. However, thesesamples might be obtained from another patient.Therefore, we needed to examine whether these sampleswere obtained from the same patient or not. After givinginformed consent from the patient, personalidentification using the AmpFlSTR® Identifiler® Kit wasperformed for FFPE tissues: 2 parts of normal gastricmucosa, one part of malignant gastric mucosa, 4 normallymph nodes, and a lymph node with metastasis.

Materials and methods

Samples

Eight parts of FFPE tissues were used among thesamples removed from the patient with early gastriccancer for histologic diagnosis: 2 parts of gastric mucosawithout cancer, one part of tumor gastric mucosa withcancer, 4 lymph nodes without cancer, and the lymphnode station No.8a with metastasis. These samples werefixed with 20% formalin neutral buffer solution (Wako,Osaka, Japan) for two weeks. The standardconcentration of formalin is usually 10% and the usualperiod is 3 to 5 days. However, this fixed solution wasnot a simple formalin solution but a buffered formalinsolution. Afterwards, these samples were embedded inparaffin and stored at room temperature. Furthermore, aperipheral blood sample was obtained from the samepatient.

DNA extraction

After the surplus paraffin was removed from the

1140

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues

Fig. 1. Resected stomach after distal gastrectomy.

FFPE tissues blocks, about 30~40 mg from the blockswere cut using sterile disposal forceps and a scalpel, andplaced in a 1.5-mL microcentrifuge tube. To remove theparaffin from the FFPE samples, 1 mL of 100% xylenewas added to the tube, and mixed by vortexing for 15sec. To melt the paraffin, the tissues were incubated at1,200 rpm for 10 min at 50°C, centrifuged at 14,000rpmfor 10 min, and then the xylene was discarded. Thisxylene treatment was performed twice. Then, 1 mL of99.5% ethanol was added to the pellet. The pellet wasmixed by vortexing for 15 sec, and agitated at 1,200 rpmfor 10 min at 20°C. It was centrifuged at 14,000 rpm for10 min, and then the ethanol was discarded. Theseethanol washings were performed twice. The pellet wasleft for about two hours so that the ethanol wouldevaporate. Then, two kinds DNA extraction methodswere carried out about one sample. One is theRecoverAll™ Total Nucleic Acid Isolation Kit for FFPE(Applied Biosystems) with slight modifications. Thepellet was put into a solution 400 µL of digestion buffer,20 µL of Proteinase K (QIAGEN, Hilden, Germany),and 20 µL of 1M dithiothreitol (DTT), and incubated forseveral hours at 1,200 rpm at 70°C. To achieve completedigestion, 20 µL of Proteinase K and 20 µL of 1M DTTwere added to the pellet. The solution was incubated for1~2 days at 1,200 rpm at 56°C. Then, DNA wasextracted according to the manufacturer’s instruction.The other was the phenol/chloroform method (Sambrooket al., 1989). The pellet was put into a solutioncontaining 300 µL of Buffer ATL (QIAGEN), 20 µL ofProteinase K (QIAGEN), and 10 µL of 1M DTT. Thesolution was incubated for 24 hr at 1,200 rpm at 56°C.Also, an extra 20 µL of Proteinase K and 20 µL of 1MDTT were added, and then the tubes were incubated for1~2 days at 1,200 rpm at 56°C. Then, they were heated

for 10 min at 99°C. DNA was extracted 3 times withphenol, followed by phenol-chloroform at a 1:1 ratioonce, and precipitated by the addition of a 1/20 volumeof 3M sodium acetate and a 2.5 volume of absoluteethanol. The pellet was washed with 70% ethanol andair-dried, and then purified by Microcon® YM-100(Millipore, Bedford, MA).

Blood DNA was extracted from 100 µL whole bloodusing QIAamp® DNA Mini Kit (QIAGEN).

The absorbance of the DNA at 260 nm and 280 nmwas measured. Calculation of the DNA concentration isbased on the absorbance at 260 nm (a solution with anOD260 of 1 contains approximately 50 µg of DNA permilliliter). Furthermore, DNA purity is judged on thebasis of the ratio of A260 to A280. A low ratio indicatescontamination by protein.

Analysis of STR loci

To amplify D8S1179, D21S11, D7S820, CSF1PO,D3S1358, TH01, D13S317, D16S539, D2S1338,D19S433, vWA, TPOX, D18S51, D5S818, FGA and theamelogenin loci, multiplex PCR using 1-800 ng of targetDNA was performed with the AmpFlSTR® Identifiler®

PCR Amplification Kit, according to the manufacturer’srecommendations (Applied Biosystems, 2001). PCRamplification was carried out using a GeneAmp® PCRSystem 9700 Thermal Cycler (Applied Biosystems). Theamplified products were electrophoresed on an ABIPRISM® 310 Genetic Analyzer (Applied Biosystems).Analysis of the amplified products and allele designationwas performed automatically using GeneMapper™ IDsoftware (version 3.2, Applied Biosystems). Todesignate the detected alleles, the highest value was6000 relative fluorescent units (RFU) and the lowest

1141

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues

Fig. 2. Gastric cancer. A, x 20; B, x100

value (threshold) was 150 RFU or 50 RFU.

Results

The DNA yields were 984 ng in the normal lymphnodes, 343 ng in normal gastric mucosa, 155 ng in thelymph node with metastasis, and 793 ng in the gastriccancer on average per 1mg of the FFPE tissues using theRecoverAll™ Total Nucleic Acid Isolation Kit. Usingthe phenol/chloroform method, the DNA yields were 5.7

µg in the normal lymph node, 3.5 µg in the normalgastric mucosa, 622 ng in the lymph node withmetastasis, and 3.2 µg in the gastric cancer. Judgingfrom the DNA quality and quantity, thephenol/chloroform method was better than the DNAextraction kit. The PCR amplification was performedusing 1ng of DNA as a template according to themanufacturer’s protocol (Applied Biosytems, 2001), butno alleles were detected. Then, PCR was carried outwith various concentrations of template DNA, so that the

1142

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues

Fig. 3. The lymph node station No.8a with metastasis. A, x 20; B, x 100

Fig. 4. A. Electropherogram of STR results obtained from gastric cancer using AmpFlSTR® Identifiler® PCR Amplification Kit. B. Electropherogram ofSTR results obtained from the lymph node station No.8a with metastasis.

alleles came to be detected. Concerning theconcentrations of template DNA, the typing succeededusing 400 ng or 50 ng in the normal lymph nodes, 400ng or 100 ng in the normal gastric mucosa, 300 ng in thegastric cancer (Fig. 4a), and 200 ng in the lymph nodewith metastasis (Fig. 4b). Furthermore, the number ofSTR loci with successful typing increased when thelowest peak amount was set from 150 RFU to 50 RFU(Fig. 5). STR typing using DNA from the gastric cancerwas the most typical among all tissues. STR typing of 14loci was successful with 150 RFU, but the typing of theD2S1338 locus was unsuccessful. With 50 RFU, STRtyping of all loci was successful. In the lymph nodestation No.8a with metastasis, STR typing wassuccessful in 6 loci with 150 RFU, and 9 loci with 50RFU. The number of STR loci with successful typingwas the lowest among all tissues.

STR typing by the detection of PCR products ofshort size was successful at 5 loci: D8S1179, D3S1358,D19S433, vWA, and D5S818 in all 8 tissues. However,STR typing did not succeed in several STR loci by thedetection of PCR products of long size. The detectionrate was very low at the CSF1PO and D2S1338 loci, andeven lower at D7S820. In regard to the amelogenin gene,the typing was successful in all samples.

The result of STR and amelogenin typing obtainedfrom the 8 FFPE tissues coincided with that from thepatient’s blood DNA. Consequently, the lymph nodestation No.8a with metastasis turned out to be from thesame patient.

Discussion

Formalin fixtation leads to 2 phenomena--crosslinking of protein-protein and protein-nucleic acid,and DNA fragmentation due to strand breakage. DNAfragmentation is caused because of the increased acidityof the solution for formalin fixation. Therefore, it is said

that PCR amplification is difficult if the size of the PCRproduct is long (Greer et al., 1991a,b; Romero et al.,1997). In this study, STR typing was successful in manySTR loci by the detection of PCR product of short size,but it was unsuccessful in several loci when detection ofPCR product of long size was attempted. Our experiencewas that the size of the detectable PCR product in STRloci was short when the period of formalin fixation waslong. When the formalin fixation was more than oneweek, the STR loci of more than 200 bp becameundetectable. There are some reports in regard to suchphenomena (Greer et al., 1991a,b; Yamada et al., 1994).A simple formalin dilution solution has an acidity ofabout pH 4, but buffered formalin is adjusted to pH7.4~7.6 by a phosphate buffer. Neutral pH can preventDNA fragmentation by formalin fixation. In this case,the tissue samples were stored in 20% buffered formalinfor 2 weeks due to the “golden week”, a long holiday inJapan. It was higher than the standard concentration(10%) and longer than the usual period (3 to 5 days).However, it is thought that STR typing was successfulbecause the fixed solution was not a simple formalinsolution but a buffered formalin solution. Therefore, thetissue samples should be fixed in buffered formalinsolution when considering DNA typing after histologicaldiagnosis.

In this case, DNA was extracted from FFPE tissuesusing a DNA extraction kit and the phenol/chloroformmethod, and the absorbance of the DNA at 260 nm and280 nm was measured. The ratio of A260 to A280 shouldbe greater than 1.75. A lower ratio is an indication thatsignificant amounts of protein remain in the preparation.The ratio by the phenol/chloroform method was higherthan that by the DNA extraction kit. Thephenol/chloroform method showed DNA with goodquantity and quality, and PCR with good amplification,as compared with the DNA extraction kit. Moreover, thephenol/chloroform method used here involved a heatstep after enzymatic digestion. Some studies showed thatPCR-amplifiable DNA was yielded by using alkali plusheat pretreatment of the FFPE samples (Shi et al., 2002,2004; Gilbert et al., 2007a). Crosslinking in this way canbe heat-reversed. Therefore, it is thought that heattreatment is likely to uncross some of the DNA, makingit easier to amplify. Accordingly, it is thought that thephenol/chloroform method with heat treatment ispreferable to DNA extraction.

DNA amplification with the AmpFlSTR®

Identifiler® Kit requires 10 µ L of DNA at arecommended concentration of 0.05-0.125 ng/µ L.However, STR typing was not successful with thisconcentration at all, so we performed PCR amplificationusing various DNA concentrations, and STR typingturned out to be succeessful in many STR loci. As aresult, too much or too little concentration of DNA wasnot appropriate for STR typing, and the appropriateconcentration for PCR amplification differed among thetissues and the extracted DNA. The DNA concentrationwas measured by the absorbance of the DNA at 260 nm.

1143

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues

Fig. 5. Number of amelogenin and STR loci detected using AmpFlSTR®

Identifiler® PCR Amplification Kit.

However, this DNA contained PCR-amplifiable andunamplifiable DNA. Moreover, the grade of digestion ofprotein-protein and protein-nucleic acid crosslinksinfluences the result of STR typing. Therefore, it ispreferable that PCR amplification is carried out atvarious concentrations in the case of formalin-fixedsamples.

Clinically, STRs can be used to study geneticalterations in tumors. A genetic deletion common tomany types of cancer is referred to as the loss ofheterozygosity (LOH). Numerous examples of LOH incancer have been described (Silva et al, 1997, 1998; Paiet al., 2002; Poetsch et al., 2004; Vauhkonen et al., 2004)and some have been mapped in areas located in closeproximity to markers employed in human identitytesting. Despite this fact, LOH has rarely been observedfor STR loci commonly employed in forensic testing.Although we observed an allelic imbalance at theD8S1179 locus, this did not cause any difficulty in STRtyping.

Recently, multiplex PCR and minisequencing ofSNPs have been studied for human identification(Sanchez et al., 2003, 2006). Gilbert et al. demonstratedthat multiplex PCR with minisequencing (MPMS) willprove useful some day when FFPE materials areanalyzed (Gilbert et al., 2007b). In forensic science, theAmpFlSTR® MiniFiler™ PCR Amplification Kit(Applied Biosystems) has become commerciallyavailable for forensic samples (Butler et al., 2003;Chung et al., 2004; Coble and Butler, 2005; Hill et al.,2007). This amplification kit provides information on the8 largest and most difficult to amplify loci in theAmpFlSTR® Identifiler® Kit and SGM Plus®. Withamplicon sizes of less than 270 bp, MiniFiler™ isoptimized for use with the most difficult types ofsamples, including those which are degraded and containinhibitors. This kit may be useful in such formalin-fixedsamples to recover the larger loci, which cannot beamplified or which drop out.

Over 2 years of follow-up after the DNA typing, thispatient has survived without malignancy. Clinical tissuesamples are not usually employed in forensic casework.However, personal identification using DNA typing inforensic science proved very useful for clinical medicaltreatment. We think that such forensic DNA typing maybe developed in the future. Furthermore, there is anotherissue in forensic casework for clinical cases. Usually,DNA typing costs 100,000 yen ($ 800) in Japan, so weare confronted with the problem of expenses. In Japan,when it is suspected that a patient's sample may havebeen mistaken for another patient’s, the budget for sucha problem is not appropriated. In future, it should bedecided who bears the expense for personalidentification in such clinical cases.

References

Applied Biosystems. (2001). AmpFlSTR® Identifiler™ PCR AmplificationKit user’s manual. Foster City, CA: Applied Biosystems.

Budowle B., Moretti T.R., Niezgoda S.J. and Brown B.L. (1998). CODISand PCR-based short tandem repeat loci: law enforcement tools. In:Proceedings of the Second European Symposium on HumanIdentification. 1998 June 10-12; Innsbruck, Austria. Madison, WI:Promega Corporation, 73-88.

Budowle B., Shea B., Niezgoda S. and Chakraborty R. (2001). CODISSTR loci data from 41 sample populations. J. Forensic Sci. 46, 453-489.

Butler J.M. (2005). Forensic DNA typing: biology, technology, andgenetics of STR markers. 2nd ed. Elsevier. New York.

Butler J.M., Shen Y. and McCord B.R. (2003). The development ofreduced size STR amplicons as tools for analysis of degraded DNA.J. Forensic Sci. 48, 1054-1064.

Chung D.T., Drábek J., Opel K.L., Butler J.M. and McCord B.R. (2004).A study on the effects of degradation and template concentration onthe amplification efficiency of the STR miniplex primer sets. J.Forensic Sci. 49, 733-740.

Coble M.D. and Butler J.M. (2005). Characterization of new miniSTRloci to aid analysis of degraded DNA. J. Forensic Sci. 50, 43-53.

Collins P.J., Hennessy L.K., Leibelt C.S., Roby R.K., Reeder D.J. andFoxwall P.A. (2004). Developmental validation of a single-tubeamplification of the 13 CODIS STR loci, D2S1338, D19S433, andAmelogenin: The AmpFlSTR® Identifiler™ PCR Amplification Kit. J.Forensic Sci. 49, 1265-77.

Gilbert M.T.P., Haselkorn T., Bunce M., Sanchez J.J., Lucas S.B.,Jewell L.D., Marck E.V. and Worobey M. (2007a). The isolation ofnucleic acids from fixed, paraffin-embedded tissues-which methodsare useful when? PLoS ONE 2, e537.

Gilbert M.T.P., Sanchez J.J., Haselkorn T., Jewell L.D., Lucas S.B.,Marck E.V., Børsting C., Morling N. and Worobey M. (2007b).Multiplex PCR with minisequencing as an effective high-throughputSNP typing method for formalin-fixed tissue. Electrophoresis 28,2361-2367.

Greer C.E., Lund J.K. and Manos M.M. (1991a). PCR amplification fromparaffin-embedded tissues: Recommendations on fixatives for long-term storage and prospective studies. PCR Meth Appl. 1, 46-50.

Greer C.E., Peterson S.L., Kiviat N.B. and Manos M.M. (1991b). PCRamplification from paraffin-embedded tissues: Effects of fixative andfixation time. Am. J. Clin. Pathol. 95, 117-124.

Hill C.R,, Kline M.C., Mulero J.J., Lagacé R.E., Chang C-W., HennessyL.K. and Butler J.M. (2007). Concordance study between theAmpFlSTR® MiniFiler™ PCR Amplification Kit and conventionalSTR typing kits. J. Forensic Sci. 52, 870-873.

Holt C.L., Buoncristiani M., Wallin J.M., Nguyen T., Lazaruk K.D. andWalsh P.S. (2002). TWGDAM validation of AmpFlSTR® Identifiler™PCR amplification kits for forensic DNA casework. J. Forensic Sci.47, 66-96.

Impraim C.C., Saiki R.K., Erlich H.A. and Teplitz R.L. (1987). Analysis ofDNA extracted from formalin-fixed paraffin-embedded tissues byenzymatic amplification and hybridization with sequence-specificoligonucleotides. Biochem. Biophys. Res. Commun. 142, 710-716.

Japanese Gastric Cancer Association. (1998). Japanese Classificationof Gastric Carcinoma. 2nd Engl. ed. Gastric cancer. 1, 10-24.

Koyanagi H., Matsuno S., Kitajima M. and Kato H. (2004). Standardtextbook. 10th ed. Igaku-shoin. Japan. (In Japanese).

Kumar V., Cotran R.Z. and Robbins S.L. (2003). Robbins BasicPathology. 7th ed. Elsevier Saunders. USA.

Noguchi M., Furuya S., Takeuchi T. and Hirohashi S. (1997). Modifiedformalin and methanol fixation methods for molecular biological and

1144

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues

morphological analyses. Pathol. Int. 47, 685-691.Pai C.Y., Hsieh L.L., Tsai C.W., Chiou F.S., Yang C.H. and Hsu B.D.

(2002). Allelic alterations at the STR markers in the buccal tissuecells of oral cancer patients and the oral epithelial cells of healthybetel quid-chewers: an evaluation of forensic applicability. ForensicSci. Int. 129, 158-167.

Poetsch M., Petersmann A., Woenckhaus C., Protzel C., Dittberner T.,Lignitz E. and Kleist B. (2004). Evaluation of allelic alterations inshort tandem repeats in different kinds of solid tumors--possiblepitfalls in forensic casework. Forensic Sci. Int. 145, 1-6.

Reid T.M., Wolf C.A., Kraemer C.M., Lee S.C., Baird M.L. and Lee R.F.(2004). Specificity of sibship determination using the ABI identifilermultiplex system. J. Forensic Sci. 49, 1262-1264.

Romero R.L., Juston A.C., Ballantyne J. and Henry B.E. (1997). Theapplicability of formalin-fixed and formalin fixed paraffin embeddedtissues in forensic DNA analysis. J. Forensic Sci. 42, 708-714.

Sambrook J., Fritsch E.F. and Maniatis T. (1989). Molecular cloning: ALaboratory Manual. 2nd ed. Cold Spring Harbor Laboratory Press.New York.

Sanchez J.J., Børsting C., Hallenberg C., Buchard A. and Hernandez A.and Morling N. (2003). Multiplex PCR and minisequencing of SNPs -a model with 35 Y chromosome SNPs. Forensic Sci. Int. 137, 74-84.

Sanchez J.J., Phillips C., Børsting C., Balogh K., Bogus M., FondevilaM., Harrison C.D., Musgrave-Brown E., Salas A., Syndercombe-Court D., Schneider P.M., Carracedo A. and Morling N. (2006) Amultiplex assay with 52 single nucleotide polymorphisms for humanidentification. Electophoresis 27. 1713-1724.

Silva F., Amorim A. and Gusmao L. (1998). Somatic instability in gastrictumors at STRs used in forensic genetics. In: Progress in forensicgenetics. Olaisen B., Brinkmann B. and Lincoln P.J. (Eds.). Vol. 7.Elsevier. Amsterdam. pp 424-426.

Silva F., Gusmao L., Alves C., Seruca R, David L. and Amorim A.(1997). Tetra- and penta-nucleotide short tandem repeat instabilityin gastric cancer. Electrophoresis 18, 1633-1636.

Shi S.R., Cote R.J., Wu L., Liu C., Datar R., Shi Y., Liu D., Lim H. and

Taylor C.R. (2002). DNA extraction from archival formalin-fixed,paraffin-embedded tissue sections based on the antigen retrivalprinciple: Heating under the influence of pH. J. Histochem.Cytochem. 50, 1005-1011.

Shi S.R., Datar R., Liu C., Wu L., Zhang Z., Cote R.J. and Taylor C.R.(2004). DNA extraction from archival formalin-fixed, paraffin-embedded tissues: heat-induced retrival in alkaline solution.Histochem. Cell Biol. 122, 211-218.

Tokuda Y., Nakamura T., Satonaka K., Maeda S., Doi K., Baba S. andSugiyama T. (1990). Fundamental study on the mechanism of DNAdegradation in tissues fixed in formaldehyde. J. Clin. Pathol. 43,748-751.

Townsend C.M., Beauchamp R.D., Evers B.M. and Mattox K.L. (2006).Sabiston Textbook of Surgery, the biological basis of modernsurgical practice. 17th ed. Elsevier.

Vauhkonen H., Hedman M., Vauhkonen M., Kataja M., Sipponen P. andSajantila A. (2004). Evaluation of gastrointestinal cancer tissues asa source of genetic information for forensic investigations by usingSTRs. Forensic Sci. Int. 139, 159-167.

Yamada M,, Yamamoto Y., Tanegashima A., Kane M., Ikehara Y.,Fukunaga T. and Nishi K. (1994). Determination of ABO genotypeswith DNA extracted from formalin-fixed paraffin-embedded tissues.Int. J. Legal Med. 106, 285-287.

Yoshida K., Takahashi K., and Kasai K. (2005). Allele frequencies of 15loci using AmpFlSTR Identifiler Kit in Japanese population. JForensic Sci. 50, 718-719.

Yoshida K., Takahashi K. and Kasai K. (2006). Validation ofAmpFlSTR® Identifiler® PCR Amplification Kit in application toforensic evidential samples. Rep. Natl. Res. Inst. Police Sci. 57, 49-56. (In Japanese).

Zehner R. and Lasczkowski G. (2000). Paternity-testing on paraffin-embedded abortion tissue: preparation of fetal cells may beindispensable. J. Forensic Sci. 45, 1332-1334.

Accepted March 18, 2009

1145

Forensic DNA typing for formaldehyde-fixed paraffin-embedded (FFPE) tissues