Evaluation of reliability of STR typing in human colon carcinomas tissues used for identification purpose

1. Introduction 

Short tandem repeats (STRs) are highly polymorphic microsatellites. This is due to a large number of repeat-structured alleles per locus, which are inherited in a Mendelian fashion. Modern gene technology based on DNA amplification using PCR enables simultaneous analysis of many STR loci in the genome. Analysis of STR offers an efficient and reliable way of discriminating between sample source and individuals. So STR typing technology has become an important and widely used tool in forensic casework.

Clinical tissue samples are not usually employed in forensic casework, but can be used for identification of unknown bodies and paternity testing even if the putative father is dead. Many archival pathology specimens can be tested without problems, since STRs are stable in most tissues, even in post-mortem tissues [1]. The use of tumor tissue for forensic applications, however, has certain difficulties. Tumor DNA is known to harbor genetic alterations not only in defined genes, but also in repetitive DNA sequences. Loss of heterozygosity (LOH) and microsatellite instability (MSI) are important characteristics of many tumors [2]. Although these characteristics are usually defined by the analysis of di-nucleotide and mono-nucleotide markers, DNA instability may well extend to other short tandem repeat families, for example tetra-nucleotide repeats which are commonly used in forensic casework. This leads to the assumption that the reliability of STR typing with commercially available multiplex kits common in forensic analysis may be questionable. Changes in these STRs are still under discussion. Hoff-Olsen et al. described STR instability in 19% of colorectal carcinomas [3]. Rubocki et al. demonstrated LOH in D13S317 in a bladder cancer sample [4]. In contrast, Orlandi et al. could not find any alterations in breast carcinomas [5]. To contribute new data to this discussion we analyzed 77 human colon carcinomas tissue for LOH and MSI in tetra-nucleotide (AmpFlSTR Identifiler Kit) repeats.

2. Materials and methods 

2.2. Analysis of STR loci 

The analysis of STR loci was done with the commercially available AmpFlSTR Identifiler kit according to the manufacturers instructions. This kit contains all reagents to coamplify the 15 STRs D3S1358, vWA, FGA, D8S1179, D21S11, D18S51, D5S818, D13S317, D16S539, TH01, TPOX, CSF1PO, D7S820, D2S1338, D19S433 and the amelogenin gene. PCR amplification reactions were carried out using a GeneAmp PCR system 9700 (Applied Biosystems). The amplified products were separated by capillary electrophoresis on AB PRISM 3130 Genetic Analyzers (Applied Biosystems, Foster City, CA, USA) according to the manufacture's recommended protocols. The sample run data were analyzed together with an allelic ladder and positive and negative controls using GeneMapper® ID Software Version 3.2. Allele designation was established following the recommendations of the DNA commission of the ISFG.

In the case of different allelic patterns between normal and tumor tissue in any of the STR loci, PCR and electrophoresis of this normal/tumor pair were repeated twice. Only variations occurring identically in all three assays are mentioned here. All results from normal and tumor tissue were analyzed in a double-blind manner and then compared. Concerning the partial loss of one allele, pLOH was scored if one allele was >50% decreased in tumor DNA when compared with the same allele in normal control DNA. The following equation was used to calculate pLOH and LOH:

Allelic loss is indicated by an LOH value of less than 0.5 or higher than 2.0.

3. Results and discussion 

In our study we detected four kinds of changes between normal and tumor tissue including partial Loss of heterozygosity (pLOH), complete Loss of heterozygosity (LOH), occurrence of an additional allele (Add) and occurrence of a new allele (New) instead of that found in normal tissue.

Only 36% of samples presented the same genotypes in normal and tumor tissue specimens, the rest of samples had at least one altered locus. The overall variation detectable rate was 11.28%, of these, 79.1% was pLOH, 7.9% was LOH, 7.9% was Add and 5% was New. Of the above four changes, the incidence rate of pLOH, LOH, Add and New was respectively 8.93%, 0.89%, 0.89% and 0.57%. The most frequently altered loci were D18S51, D5S818, FGA and D19S433 whose variation rate were all greater than 15%. As CSF1PO and D5S818 are both located on the 5 chromosome, from the view of chromosome variation rate. The chromosome 5 was most frequently affected (31.17%), and the chromosome 18 take second place (28.57%). Only pLOH was found at five loci including vWA, TPOX, TH01, D13S317 and amelogenin gene. The differences in the extent of alterations at the 15 STR loci and the amelogenin gene were shown in Table 1.

Table 1. Distribution of four kinds of changes over the 15 STR loci and the amelogenin gene in all investigated tumors.

	Locus	PLOH	LOH	Add	New
	AMEL	6 (7.8%)	0	0	0
	CSF1PO	7 (9.1%)	2 (2.6%)	0	1 (1.3%)
	D13S317	10 (12.9%)	0	0	0
	D16S539	3 (3.9%)	0	0	1 (1.3%)
	D18S51	19 (24.7%)	2 (2.6%)	0	1 (1.3%)
	D19S433	10 (12.9%)	3 (3.9%)	0	0
	D21S11	4 (5.2%)	1 (1.3%)	0	0
	D2S1338	4 (5.2%)	0	1 (1.3%)	0
	D3S1358	5 (6.5%)	0	1 (1.3%)	0
	D5S818	10 (12.9)	1 (1.3%)	2 (2.6%)	1 (1.3%)
	D7S820	3 (3.9%)	1 (1.3%)	0 (	2 (2.6%)
	D8S1179	6 (7.8%)	0	2 (2.6%)	0
	FGA	8 (10.4%)	1 (1.3%)	5 (6.49)	1 (1.3%)
	TH01	3 (3.9%)	0	0	0
	TPOX	3 (3.9%)	0	0	0
	vWA	9 (11.7%)	0	0	0

In cases of frequent four kinds of changes, a wide panel of markers should be used to obtain reliable results from the STR analysis. Moreover, a microscopically verified histopathological diagnosis should be obtained, and if possible, a marginal cut from non-malignant tissue should be used for forensic or paternity analysis.

Conflict of interest 

None.

Role of funding 

This study was supported by the Grants from Ministry of Science and Technology, PR China (GY0604).