| Objective: Tumor tissue,and particularly formalin fixed paraffin embedded(FFPE)tissue,is a kind of challenging samples often encountered in forensic casework.Neoplastic tissue manifests a wide variety of genetic alterations and DNA extracted from FFPE tissue is prone to degradation.Due to the high mutation rate of STR in tumor tissue and with long amplicon size(100-400bp),the traditional capillary electrophoresis(CE)-based fragment analysis methods to detect length variation in STR markers are no longer suitable for the detection of tumor tissues.Compared with STR,SNP is characterized by low mutation rate and smaller amplicon size.Therefore,the aim of this study is to explore the SNP mutation rate of tumor tissues,and the advantages of SNP compared with STR in detection of degradated formalin fixed tumor tissue.This study also performs MPS on tumor tissue with the aim of providing a reference for the selection of genetic markers and technology in the case involving tumor tissue.Methods:1 Sample collection: 124 formalin-fixed tumor tissues within 24 hours were obtained(including 85 digestive system cancer cases,16 urinary system cancer cases,13 respiratory system cancer cases,10 head and neck cancer cases and their corresponding normal tissues).Degraded samples were from 24 formalin-fixed tissues stored at-20℃ for five years,including 12 paired tumor/normal tissues from the patients of digestive system cancer.2 DNA extraction: DNA was extracted from the 124 paired tumor/normal tissues using TIANamp FFPE DNA Kit according to the manufacturers’ instructions,and then the DNA was quantified with a NanoDrop ND-1000 spectrophotometer.DNA was extracted from 12 degraded paired tumor/normal tissues with GeneReadTM DNA FFPE Kit,and the concentration and degradation index of DNA was detected with Quantifiler? Trio DNA Kit in 7500 Real Time PCR System.3 SNP typing: A 55-SNP multiplex SNaPshot assay was applied to amplify 55 autosomal SNPs and amelogenin sex marker.Separations were performed by capillary electrophoresis,using an Applied Biosystems 3130 Genetic Analyzer.Data were collected using Data Collection software v3.0,and allele assignments were performed.Then the SNP genotype of tumor tissues was compared with the normal tissues of the same patient to note the mutation type and allele.4 Verification of SNP mutation: Unreliable SNP genotypes were verified by repeated detection,singleplex SNaPshot system,or Sanger sequencing.5 STR genotyping: Twenty STR loci of degraded specimens were genotyped with PowerPlex? 21 multiplex genotyping system.The STR alleles were separated on an ABI 3130 genetic analyzer.6 Amplification of single STR locus: The specimens with suspected mutation alleles were verified using single STR locus amplification system,and the alleles were separated on polyacrylamide gel electrophoresis.7 Miseq FGx sequencing: Twenty four degraded samples were sequenced using Foren SeqTM DNA Signature Prep Kit(primer mixture A)according to the manufacturer’s protocol that allows for targeted amplification and sequencing of 27 autosomal STRs,24 Y-STRs,7 X-STRs and 94 identity informative SNPs by Miseq FGx sequencing platform.The analysis and interpretation of the data was performed with ForenSeqTM Universal Analysis Software V1.2.16347(Illumina,CA).Standard analysis settings were used for UAS genotype assignment and manual alterations on the genotype calls were recorded by the software.8 Statistical analysis: Statistical analysis was performed using Excel and SPSS v16.0 software.Chi-square test was used to compare the difference of SNP mutation rate between digestive system tumors and other tumor tissues.The correlation between the amplicon size of the STR locus and the allele detection rate,and the degradation index of the samples and the detection rate of STR alleles were analyzed by linear correlation test.Mean and standard deviation were used to describe the average depth of coverage,the average allele coverage ratio,the constituent ratio of STR and SNP.Rank test and t test were used to compare the difference of the average depth of coverage,the average allele coverage ratio,the constituent ratio of STR and SNP,between the degraded tumors and degraded normal samples.Correlation of the average depth of coverage and the amplicon size of STR and SNP,and the average depth of coverage and degraded index was performed by linear correlation analysis.Results:1 The SNP mutation rate of tumor tissuesForty-nine suspected SNPs’ mutations were found in 124 paired tumor/normal tissues by 55-SNP multiplex SNaPshot assay.Then these 49 SNPs were verified by singleplex SNaPshot system,and the final results showed that the 4 mutational SNPs were found in 3 specimens,including one digestive system tumor tissue(rs9905977:C/T→T,rs722290:C/G→G),and two urinary system tumor tissues(rs7041158:C/T→C),(rs10092491:C/T→C).The 4 SNPs were verified by Sanger sequencing,except for the rs7041158,the other 3 SNP genotypes were consistent with those of singleplex SNaPshot system based on their sequences.There was no difference in SNP mutation rate between digestive system tumor and other tumor tissues(P>0.05).The SNP mutation rate was significantly lower than that of STR previously detected in our laboratory both in individual and locus levels(P<0.05).2 Comparison of detection rates of SNP and STR in degraded tissue specimensThe results showed that the degradation indexes of 23 samples ranging from 1.01~8.57,and only 1 sample less than 1.The SNP genotypes of the 24 degraded specimens were completely consistent with the non-degraded DNA from the same individuals and the successful genotyping rate was 100%.However,33 allele dropouts were observed with STR genotyping in 8 samples,of which the degradation index was higher than 2.6,and the fragment size of the 75.8% allele was longer than 300 bp.There was a negative correlation between the amplicon size of STR and the allele detection rate(r=-0.629,P<0.05),and a negative correlation was also observed between the degradation index of samples and the allele detection rate except for two samples with mild degradation(r=-0.659,P<0.05).This study validated that the long-term formalin-fixed tissues were susceptible to degradation,and the SNP was more suitable for detecting these tissues than STR typing system.3 Miseq FGx sequencing analysisThe average depth of coverage across the STR(Amel excepted)of 12 degraded tumor tissues was 2384 reads per locus(×)(±1884×).The average allele coverage ratio of STR locus was 0.61(±0.13).The proportion of alleles,stutter and noise of STR loci was 0.92(0.81~1.00),0.06(0.00~0.13)and 0.02(0.00~0.07)respectively.The average depth of coverage across the STR(Amel excepted)of 12 degraded normal samples was 2158×(±1710×).The average allele coverage ratio of STR locus was 0.73(±0.11).The proportion of alleles,stutter and noise of STR loci was 0.92(0.81~1.00),0.06(0.00~0.13),and 0.02(0.00~0.07)respectively.There was no difference in the average depth of coverage of STR and the proportion of the alleles,stutter and noise of STR loci between degraded tumor samples and degraded normal samples(P>0.05).The average allele coverage ratio of degraded normal samples STR locus was higher than that of degraded tumor samples(P<0.05).There was a negative correlation between the average depth of coverage and fragment size of STR in degraded tumor and normal samples.With the increase of fragment size of STR,the average depth of coverage of STR locus decreases gradually.There was no correlation between the average depth of coverage of STR and degradation index of samples in degraded tumor samples(P>0.05).The average depth of coverage of STR was negatively correlated with degradation index of normal samples(P<0.05).The average depth of coverage across the SNP of 12 degraded tumor samples was 741×(±661×).The average allele coverage ratio of SNP loci was 0.61(±0.12).The proportion of alleles and of SNP loci was 0.9998(0.9925~1.0000)and 0.0002(0.0000~0.0075)respectively.The average depth of coverage across the SNP of 12 degraded normal samples was 633×(±552×).The average allele coverage ratio of SNP loci was 0.77(±0.12).The proportion of alleles and noise of SNP loci was 0.9995(0.9892~1.0000)and 0.0005(0.0000~0.0108)respectively.There was no difference in the average depth of coverage of SNP between degraded tumor samples and degraded normal samples(P>0.05).But there was difference in average allele coverage ratio of SNP loci,the proportion of the alleles and noise of SNP loci,between degraded tumor samples and degraded normal samples(P<0.05).There was a negative correlation between the average depth of coverage and fragment size of SNP loci in degraded tumor and normal samples.With the increase of fragment size of SNP,the average depth of coverage of SNP loci decreases gradually.There was no correlation between the depth of coverage of SNP and degradation index of samples in degraded tumor and normal samples(P>0.05).To evaluate the concordance of STR genotype between Miseq FGx sequencing platform and traditional CE technique,24 degraded samples were typed with ForenSeqTM DNA Signature Prep Kit(primer mixture A)and PowerPlex? 21 Kit.While 21 STR with allele sequence variants and 3 STR with different alleles between the two methods for all typable alleles.For degraded samples,STR allele dropout was detected in both Miseq FGx sequencing platform and CE.There was no difference in allele detection rate of STR loci between Miseq platform and CE(P>0.05).A total of 30 allele dropouts were observed in Penta E and D12S391 with Miseq FGx sequencing platform.The allele dropout was mainly observed in PentaE which accounted for 93.33% of the non-detectable allele and the fragment size was 362bp~467bp.A total of 33 allele dropouts were observed in 11 STR locus with CE,and the fragment size of 75.8% allele was more than 300 bp.To assess the concordance of identity informative SNPs,24 degraded samples were typed with ForenSeqTM DNA Signature Prep Kit(primer mixture A)on Miseq FGx and a 55-SNP multiplex SNaPshot assay.Concordance was evaluated on the 43 overlapping ii SNPs in these two panels.Results were concordant between the two panels for all typable alleles,except at the loci rs9905977,rs722290,rs338882 and rs7041158.Allele dropout was observed in one sample for rs9905977 and rs722290 when typed with multiplex SNaPshot assay.Allele and locus dropout was observed in three samples for rs338882 with Mi Seq FGx.Allele dropout was observed in one sample for rs7041158 with MiSeq FGx.There was no difference in allele detection rate of SNP loci between Miseq FGx and SNaPshot(P>0.05).Conclusion:1 The mutation rate of SNP in tumor tissues was significantly lower than that of STR,so SNP is more suitable for individual identification and kinship analysis involving tumor tissues.2 Because of long-term formalin fixed tissues susceptible to degradation,SNP has more advantages than STR in the degraded tumor tissues.3 Miseq FGx sequencing platform showes a higher sensitivity to tumor tissues and degraded specimens compared to multiplex SNaPshot assay,and the preliminary evaluation illustrated the ForenSeqTM DNA Signature Prep Kit is capable of producing reliable and accurate sequencing data for forensic marker.Miseq FGx sequencing platform can be used as an effective tool for the detection of tumor tissues. |
PDF Full Text Request