An Update on the Military's Program of Skeletal Remains Identification Using Mitochondrial DNA Sequence Analysis: Identification of Soldiers Killed During th eVietnam War, Korean War, and World War II

The opinions and assertions expressed herein are solely those of the authors 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.

As a consultative service for the Army Central Identification Laboratory in Hawaii (CILHI), the Armed Forces DNA Identification Laboratory (AFDIL) has been performing mitochondrial DNA (mtDNA) testing for the past four years on skeletal remains from World War II to the Vietnam War (1,2). Two years ago, the US government requested that AFDIL expand its capabilities. Renovation of a laboratory facility in Rockville, Maryland was conducted. AFDIL occupied the facility and hired 23 new personnel on April 1, 1994.

The new facility consists of approximately 5,000 square feet of service laboratory space which is broken into five main areas. In accordance with TWGDAM guidelines (3), the extraction of DNA from material such as ancient skeletal remains (greater than 30 years of age) is separated from other extraction activities. Extraction of DNA from evidentiary material thought to contain higher levels of DNA is conducted in a second laboratory. Extraction of DNA from blood reference sources is performed in a third laboratory. As many phases of the extraction process as possible are conducted under a hood (laminar flow, dead space, chemical). Amplification reactions are prepared under hoods as well. PCR amplification and product analysis are separated from all extraction activities. Finally, reagent preparation and DNA primer synthesis are conducted in a separate laboratory to ensure contamination-free reagents. Separation of the testing activities, as outlined above, has been very successful.

Prior to commencing with production operations, the 21 new technical staff members were given six months of intensive training. The final stage of training consisted of supervised, duplicate testing on active cases. In one instance, two trainees conducted testing on a blind proficiency case submitted by CILHI. The two trainees completed testing, obtained the same results, and were able to correctly type the remains based on a follow-up report from CILHI. This was evidence that the training process was successful, and further illustrated the reliability of the testing methods. Full casework production commenced on October 1, 1994.

As of October 1, 1995, AFDIL has processed 75 skeletal remains cases representing approximately 117 unaccounted for military service-members. The mtDNA sequence of the remains has been successfully generated in greater than 90% of the cases. Eighteen (18) identifications and two (2) exclusions have resulted. In addition, twenty-seven (27) comparisons are currently being conducted, while cases representing seventy (70) individuals are awaiting maternal family references. Therefore, in the near future, AFDIL expects to help identify greater than one hundred (100) individuals.

In addition of CILHI cases, AFDIL has performed mtDNA testing on criminal and historical cases. A notable case, which was completed recently, is the identification of Tsar Nicholas II (in preparation). Although the British Forensic Science Service (FSS) provided evidence supporting the identification of the Tsar two years ago, the Tsar had an unusual sequence type which was slightly different from the maternal family references (4). At position 16169 of the mtDNA genome (5), the Tsar had two sequence types (i.e. C and T), whereas the maternal references had a single type (i.e. T). Although this was scientifically explainable, doubt was cast on the identification causing the Russian government not to sanction the identification. In June of 1995, the Russian forensic scientist Dr. Pavel Ivanov brought the remains of Georgij Romanov (the brother of Nicholas II) to AFDIL. Georgij had died in 1899 and was buried in St. Petersburg. His remains were exhumed in 1994. Portions of the remains from both Nicholas II and Georgij were analyzed. The Tsar showed identical heteroplasmic pattern seen by the British FSS (i.e. C and T at 16169). Georgij also showed heteroplasmy at 16169. Although the heteroplasmy was the same, the ratio of the two nucleotides was inverted. These were very compelling results which strongly confirmed the identification of the Tsar. Including the existence of heteroplasmy, the chance of finding such a mtDNA sequence type in the general population has been calculated conservatively as greater than one in a million. The historical implications of this case are obvious (the Tsar will be buried in St. Petersburg on February 25th of 1996), and the ability to obtain results on remains greater than 95 years of age further supports the robustness of mtDNA testing.

Finally, although the current protocols are robust and proven reliable, the methodologies of mtDNA testing performed by AFDIL have evolved significantly (1,2; see MtDNA Testing Methods), and are constantly being improved. One unique aspect of mtDNA testing is that the position of the amplification primers and the sequencing primers can move without changing the essence of the results. Consequently, the search for better primers is an ongoing process. The extraction and purification techniques employed by AFDIL have been very successful (1,2,6). Nevertheless, alternative methods are being evaluated. In particular, a silica-based method has worked very well and is being further validated for implementation into casework (7). This method has proven to reduce the level of contamination in the extraction reagents by reducing the number of sample transfers and by allowing all reagents to be UV irradiated before use. In essence, methods development is an ongoing quality improvement process.

DNA from skeletal remains is extracted using a modification of the method described by Fisher et al. (4). In brief, the outer surface of approximately 5 grams of compact bone is cleaned and sanded with a Dremmel tool (Sears, Roebuck and Co.) to remove surface contaminants introduced during exposure to the environment and casual handling during recovery of the remains. Approximately 2 grams of the clean bone is crushed into small pieces with a stainless steel chisel and a mortar, and the small pieces ground into a powder using a Micro-Mill Grinder (Fisher Scientific, Pittsburgh, PA) or a Warring Blender ( J & H Berge, Inc., South Plainfield, NJ). The powder is suspended in 3 ml of extraction buffer (10mM Tris-HCl, pH 8.0, 100 mM NaCl, 2% sodium dodecyl sulfate, and 10 mM EDTA) in the presence of 0.5 mg/ml Proteinase K and incubated at 56° C for 12-18 hours. Insoluble material is pelleted by centrifugation and discarded. The supernatant is purified in three steps:

Four overlapping regions of mitochondrial DNA are amplified using the polymerase chain reaction (PCR) (Table 1). When performing casework, whole blood extracts are routinely amplified using the forward primer of primer set I and the reverse primer of primer set II to amplify the entire HV1 region. In turn, HV2 is amplified using the forward primer of primer set III and the reverse primer from primer set IV.

The amplification reaction mixture for all four primer sets contain 10 pmol each primer, 200 µM each dNTP, 1X Perkin-Elmer polymerase buffer (10mM Tris-HCl, pH 8.3, 50mM KCl, 1.5 mM MgCl2), 1 unit of Perkin-Elmer AmpliTaq DNA polymerase, and approximately 10-200 pg of template in a 50 µl reaction. The Perkin-Elmer 9600 Thermal Cycler parameters for primer set 1 are as follows: 94°C for 30 seconds, followed by 38 cycles for skeletal remains or 32 cycles for blood reference samples at 94°C for 20 seconds, 62°C for 20 seconds, and 72°C for 30 seconds, followed by a 4°C hold. The amplification parameters for primer sets 2,3 and 4 are as follows; 94°C for 30 seconds, followed by 38 cycles for skeletal remains or 32 cycles for blood references at 94°C for 20 seconds, 56°C for 10 seconds, and 72°C for 30 seconds, followed by a 4°C hold.

The PCR product is run on a 2% agarose yield gel to evaluate the quality and quantity of the product. The amplification products are then purified by micro-dialysis on an Amicon Centricon 100 column. The product is washed three times with 2 mL of sterile deionized water and the final wash is concentrated to a volume of approximately 40 µl.

Approximately 20-200 ng of PCR product is used for DNA sequence analysis. The protocol included in the Perkin-Elmer Taq DyeTerminator Cycle Sequencing Kit is used to sequence the PCR template, with the exception that 2 minutes of extension time are used. The primers used for sequencing are the same as those for PCR amplification (Table 1). Following cycle sequencing, the reaction products are purified using Boehringer Mannheim Quick Spin G-50 columns and the protocol provided by Perkin-Elmer. The purified reaction products are ultimately separated on an Applied Biosystems (ABD) 373A DNA Sequencer, and the results analyzed using the software provided. Sequence comparisons are made using the ABD SeqNav (Sequence Navigator) software.

Primer Set I
Forward Primers
	F15990 5'-TTA ACT CCA CCA TTA GCA CC-3'
Reverse Primers
	R16236 5'-TGG CTT TGG AGT TGC AGT TG-3'
	R16239 5'-TGG CTT TGG AGT TGC AGT TG-3'*
Primer Set II
Forward Primers
	F16159 5'-TAC TTG ACC ACC TGT AGT AC-3'	(Budowle et al. 1990)
	F16163 5'-TGA CCA CCT GTA GTA CAT AA-3'*
Reverse Primers
	R16391 5'-GAG GAT GGT GGT CAA GGG AC-3'*
	R16395 5'-CAC GGA GGA TGG TGG TCA AG-3'
	R16401 5'-TGA TTT CAC GGA GGA TGG TG-3'	(Vigilant et al. 1989)
Primer Set III
Forward Primers
	F34 5'-CAC CCT ATT AAC CAC TCA CG-3'*
	F48 5'-CTC ACG GGA GCT CTC CAT GC-3'
Reverse Primers
	R251 5'-TGG AAA GTG GCT GTG CAG AC-3'
	R255 5'-TGT GTG GAA AGT GGC TGT GC-3'
	R266 5'-GTT ATG ATG TCT GTG TGG AA-3'*
Primer Set IV
Forward Primers
	F164 5'-CTC ATC CTA TTA TTT ATC GC-3'
	F174 5'-TAT TTA TCG CAC CTA CGT TC-3'*
Reverse Primers
	R362 5'-GCT GGT GTT AGG GTT CTT TG-3'
	R370 5'-CTG GTT AGG CTG GTG TTA GG-3'*
	R465 5'-TGA GAT TAG TAG TAT GGG AG-3'

* Primers which are considered by AFDIL to be the best at the time this manuscript was written.

An Update on the Military's Program of Skeletal Remains Identification Using Mitochondrial DNA Sequence Analysis: Identification of Soldiers Killed During the Vietnam War, Korean War and World War II