Forensic STR analysis reveals DNA contamination previously undetected during clinical analysis of chronically inflamed tissues

1. Introduction 

Short tandem repeat (STR) analysis is used in forensic identification efforts, and to detect alterations in diseased tissues. Cancerous tissues may accumulate allelic changes in the form of microsatellite instability (MSI) and loss of heterozygosity (LOH) [1]. Analysis of this instability is the basis of molecular genetic detection of disease progression prior to visible dysplasia [2], and provides information about the nature of existing tumors, familial risk, chemo-sensitivity, and prognosis [1], [2], [3]. Alternatively, forensic analysis for human identification is complicated by the presence of genetic instability such as loss of an allele (LOH), or expansion/contraction of an allele (MSI). These changes can alter a profile from that present in healthy tissue, or mimic the presence of multiple DNA contributors. Previous studies have addressed the effects of carcinoma on STR loci commonly used for forensic individualization [4], [5]. We originally sought to investigate whether patients surgically treated for ulcerative colitis (UC), a chronic inflammatory disease resulting in an increased risk of colon cancer [6], exhibited instability in clinically relevant STRs pre- and post-surgery, and whether this instability was similarly found in loci used for forensic individualization.

PCR-based analysis of STRs is highly sensitive, allowing the analysis of low amounts of damaged and degraded DNA, such as that found in formalin-fixed paraffin-embedded tissues (FFPET). As always with very sensitive techniques, the potential impact of contamination must be assessed. Multiplex forensic STR kits have been used to determine the origin of tissues in contaminated or potentially switched specimens. Contamination, acquired through batch processing or shared equipment, is usually indicated by pieces of foreign or unexpected tissue present in a sample, triggering further review of the case [7]. Here we describe DNA contamination that could have gone undetected during routine pathology and genetic screening, due to the lack of visible contamination, and absence of indications of a DNA mixture. This contamination could have led to improper assessments of MSI and LOH.

2. Materials and methods 

111 blocks of FFPET originally obtained from biopsies taken at the time of ileo-colectomy [14 colon, 21 terminal ileum and 32 unaffected lymph node (reference samples)], and at multiple time points between 1 and 9 years post-surgery (44 ileal pouch reviews) from 13 patients with UC. Tissues were micro-dissected from routinely prepared slides, and DNA was extracted by an organic solvent method.

For analysis of clinically relevant STRs – 9p22-p21 (BDMF and IFNA), 17p13.1 (p53), 3p21 (close to the VHL gene; 3p), and 5q22–q23 (APC) – extracts were amplified using HotStar Mastermix (Qiagen, Valencia, CA, USA). Fluorescently labeled primer sequences were as follows. 3p: Fwd [5′-FAM-ACCACATGAGCCAATTCTGT-3′], Rev [5′-ACCCAATTATGGTGTTGTTACC-3′]; APC: Fwd [5′-NED-ACTCACTCTAGTGATAAATCGGG-3′], Rev [5′-AGCAGATAAGACAGTATTACTAGTT-3′]; p53: Fwd [5′-FAM-GAATCCGGGAGGAGGTTG], Rev [5′-AACAGCTCCTTTAATGGCAG]; IFNA: Fwd [5′-VIC-TGCGCGTTAAGTTAATTGGTT-3′], Rev [5′-GTAAGGTGGAAACCCCCACT-3′]; BDNF: Fwd [5′-NED-AGCTAAGTGAAACCTCATCTCTGTCT-3′], Rev [5′-ACCCTAGCACTGATGGTATAGTCT-3′].

For each sample 4–100pg of DNA was amplified in a final volume of 20μl by the following program: 95°C—15min; (94°C—30s; 54°C—30s; 72°C—1min) for 35 cycles; 72°C—10min. The Identifiler® Multiplex System (Applied Biosystems, Foster City, CA, USA) was used according to the manufacturer's recommendations. Amplified products were analyzed using capillary electrophoresis. MSI and LOH (partial and complete) were assessed as indicated in Cawkwell et al. [8].

3. Results and discussion 

4. Conclusions 

In contrast to forensic STR loci, which are chosen for high rates of heterozygosity, high variability in the population, and a lack of correlation with a physical trait or pathology, STR regions used in clinical analysis are linked to genes that are highly conserved, and their variability remains unknown. In this report we have shown that the accuracy of clinical molecular testing for genetic instability could be compromised by undetected DNA contamination, leading to erroneous determinations of genetic instability. This could result in improper diagnosis or therapy to a patient. Contamination affected reference as well as diseased tissues, and cannot be distinguished from genetic instability. Therefore, we urge the forensic individualization of all archival samples undergoing genetic testing to ensure they are single source, and no tissue mix-ups have occurred.

Role of funding 

This study was funded by the New York City Office of Chief Medical Examiner (New York, NY, USA), and the University of Pennsylvania School of Medicine (Philadelphia, PA, USA), and was not extramurally funded. Employees of both sources participated in study design, execution, and the decision to submit this work.

Conflicts of Interest 

None.