Robert J. Steighner¹, Lois Tully², Justin Karajala¹, and Edwin Huffine^1
1Armed Forces DNA Identification Laboratory (AFDIL), Office of the Armed Forces Medical Examiner, The Armed Forces Institute of Pathology, Rockville, MD
²University of Maryland at Baltimore (UMAB), School of Medicine, Division of Human Genetics, Baltimore, MD

× Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø

Mitochondrial DNA (mtDNA) typing is increasingly used in the forensic identification of skeletal remains. The discriminatory power of the technique arises from the polymorphic characteristics of the hypervariable regions 1 and 2 located within the mitochondrial D-loop. Typing of samples is made by PCR amplifying and sequencing both regions of the test sample and a matrilineal reference sample. Barring mutation, maternally related sequences will be identical due to the haploid inheritance characteristics of mitochondria. Regardless of whether the sequences match or not, the procedure is arduous, requiring a considerable commitment in time and money before any results are obtained. A one year evaluation project was initiated using denaturing gradient gel electrophoreses (DGGE) as an additional method to support the mtDNA sequence data generated in our laboratory. Previously sequenced samples from non-probative cases were amplified with primers designed for DGGE. The samples were mixed with an identical, but unrelated sequence obtained from our collection of mtDNA sequence types, heat denatured, reannealed, and analyzed along with the individual samples by DGGE. By exploiting the differential melting characteristics of polymorphic sequences, DGGE is capable of separating sequence pairs differing by a single base pair. If a single matching band was observed in both the mixed sample lane and independent sample lanes, the two input sequences are identical and the sequence call confirmed. If more then one sequence is present, multiple bands are seen in the mixed sample lane and the uniqueness of the sequences established. Multiple bands appearing in the independent sample lanes are indicative of either sample heteroplasmy or contamination.

Due to the high sensitivity of the PCR reaction, contamination must be monitored closely. Although reagent blanks and negatives may aid in identifying contamination arising systematically, they are not effective at identifying samples which are contaminated prior to arriving at the laboratory or which may occur sporadically during sample processing. Sequence data obtained from mixed samples is difficult to interpret because more than one peak is present at the different polymorphic positions between the two samples. Samples suspected of being contaminated were analyzed by DGGE. In cases where a mtDNA type could be hypothesized by discounting the mixed base pair positions, the sample was mixed with a bona fide sequence of that type from our collection. Support for the sequence call is obtained if the mixed sample lane and independent sample lanes match. If the hypothesized call is incorrect, additional bands, not present in the independent sample lanes will appear in the mixed sample lane. Although effective at providing additional support for the original sequence call, the technique works best in cases of low level contamination where a predominant sequence call is possible. In cases where the individual samples in a mixture are present in equivalent amounts, problems discriminating between the contaminate and genuine mtDNA type arise. In those cases, the individual bands must be excised from the gel and sequenced before a decision can be made as to which is the authentic sequence.

Heteroplasmy, or the presence of two distinct sequence types usually differing by a single base pair, appears to be much more common than had previously been thought. The identification of heteroplasmic positions in a sample is important due to the potential increase in the discriminatory potential those positions represent and to eliminate any possibility of obtaining an error in the final mtDNA type made. Heteroplasmy is usually seen as a single mixed base pair in an otherwise pristine sequence electrophoretogram, however the call if often missed due to differences in the concentration of the predominant sequence and heteroplasmic variant. DGGE allows heteroplasmic samples to be identified at levels of sensitivity unobtainable through conventional sequencing. Over ten examples of heteroplasmy in the mtDNA HV1 region have now been identified in our laboratory by DGGE, the individual bands representing the different sequence types were excised from the gels, and the position of heteroplasmy determined unambiguously by sequencing the isolated bands.

The opinions and assertions expressed herein are solely those of the author and are not to be construed as official or as the views of the United States Department of Defense or the United States Department of the Army.