Validation of PowerPlexÔ STR Multiplex and Amelogenin Sex Identification Typing Kits Using the FMBIO II® Fluorescent Scanner: Forensic Casework And High-Throughput Convicted Offender Databanking
George R. Riley, Ph.D., Vanora M. Kean, Ph.D., Anne G. Pace, BS, Howard C. Coleman, BS, BA, and Teresa H. Aulinskas, Ph.D. GeneLex Corporation, 2203 Airport Way South, Seattle, WA 98134, (800) 523-6487
× Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø
INTRODUCTION
The laboratories at GeneLex Corporation have validated the GenePrint
Ô PowerPlexÔ identity and Amelogenin gender typing kits (Promega Corporation, Madison, WI) and have sucessfully applied this technology to forensic and paternity casework and convicted offender DNA profile databanking (Fig 1).This identity system allows for the simultaneous single tube amplification and fluorescent labeling of eight polymorphic short tandem repeat (STR) loci, the CTTv (CSF1PO, TPOX, TH01, vWA), and GammaSTR
Ô (D16S539, D7S820, D13S317, D5S818) quadriplexes and the gender specific Amelogenin locus. Single lane analysis of all nine loci is accomplished using polyacrylamide gel electrophoresis and the Hitachi FMBIO II® Laser Activated Fluorescent Scanner. The FMBIO II® performs automated three color detection of the CTTv loci, the GammaSTRÔ loci and the internal sizing standards. The internal sizing standard automatically adjusts for lane-to-lane differences that may occur during electrophoresis. The overall system produces a matching probability of greater than 1 X 108 on less than 1000 cells. Testing can be completed in two or three days and is cost effective relative to the amount of information that is obtained.The TWGDAM recommended validation program included evaluation of procedures for consistency, reproducibility, and sensitivity; the analysis of standard and unusual specimens, mixed specimens, and non-probative evidence; and included proficiency and specimen exchange programs. The manufacturer suggests using 1-2 ng of template DNA, however DNA quantities from approximately 0.25 to 50 ng per reaction produced interpretable results. Amelogenin was detectable to 0.1 ng. Two DNAs, mixed in varying ratios, can be reliably typed up to a 1:2 to 1:4 ratio, depending on the locus, using 1 ng template DNA. PowerPlex
Ô and Amelogenin alleles detected on the FMBIO II® were consistent in repeat typing and produced the same profile for DNA extracted from various tissues. PowerPlexÔ fluorescent CTT profiles agreed with silver stained CTT profiles obtained by more than 13 other laboratories in repeat typing of proficiency samples. Analysis of specimens from previously adjudicated cases corroborated results obtained with RFLP, DQA1, Polymarker, D1S80 and silver stained CTT.The practical application of the PowerPlex
Ô system in routine casework and high-throughput convicted offender profiling has given us experience in troubleshooting operational problems and interpretation of unusual DNA profiles. Operational problems that we have encountered include stutter and other artifactual bands; and uneven allele intensities and lane migration. Genetic phenomena that we have observed include novel microvariants, off-ladder alleles and three band patterns. Our experience with the PowerPlexÔ Amelogenin combination indicates that it is a powerful technology that may significantly improve the utility of DNA testing for the criminal justice system.METHODS
Purified human DNA was obtained from a variety of known exemplars and questioned evidence by one of three methods: Chelex extraction, inorganic extraction (Dykes salting-out method), or by organic extraction (phenol-chloroform). Extraction method had no effect on typing results. Quantitation of human DNA was by slot blot using radiolabeled probe to the locus D17Z1. Amplification using the GenePrint
Ô PowerPlexÔ identity and Amelogenin sex identification kits (Promega Corporation, Madison, WI) was as recommended in the manufacturers protocol (Promega) for the Perkin Elmer 480 thermalcycler. Generally, 0.5 5 ng DNA was amplified with a standard Taq DNA polymerase in 25 ul reaction volumes. Amplified samples were separated by electrophoresis together with internal sizing standards on 43 cm long, 4.5% acrylamide:bisacrylamide (19:1) (Amresco), 1 x TBE-7 M urea gels, poured in low fluorescence glass plates (Hitachi), using a BRL SA-32 electrophoresis apparatus with extender (Life Technologies). Electrophoresis was at 50 W for ~ 2.25 hr (pre-run for ~ 0.75 hr at 50W). Gels run later in the study had an external anodized aluminum heat sink (GeneLex Corp.) added to the front of the electrophoresis apparatus to improve the edge-to edge linearity of sample migration and were run at 65 W for ~ 1.75 hr (pre-run for 0.5 hr). Bands were detected by scanning with an FMBIO II® (Hitachi) and were analyzed with Hitachi Analysis 6.0 software. Alleles were identified using Hitachi STARcall® software running in Microsoft Excel.RESULTS AND CONCLUSIONS
Multiplex PCR amplification of STR loci of human DNA derived from a variety of samples, using the combined GenePrint
Ô PowerPlexÔ identity and Amelogenin gender typing kits, gave reliable typing results. The TWGDAM recommended validation program included evaluation of procedures for consistency, reproducibility, and sensitivity; the analysis of standard and unusual specimens, mixed specimens, and non-probative evidence; and included proficiency and specimen exchange programs.Evaluation of Procedures
Included in the evaluation of procedures was a comparison of 4.0% versus 4.5% acrylamide, comparison of amplification setup with and without ice, the effect of diffusion on band resolution, and a band precision study. Separations with both 4.0% and 4.5% polyacrylamide gave good resolution. Results indicated that amplification setup should be done on ice. Setup at room temperature gave variable results, with poor amplification of some loci. Post-electrophoresis diffusion was insignificant within a period of several hours. Gels rescanned 18 hour post-electrophoresis showed some loss of band resolution, particularly among the lower molecular weight alleles, however the bands remained interpretable. Band size precision was measured both lane-to-lane within gels and from gel-to-gel. Band size precision varied with allele size, from a standard deviation of 0.1 bp to <0.5 bp.
Consistency and Reproducibility
Results indicated that multiplex amplification using the combined GenePrint
Ô PowerPlexÔ identity and Amelogenin gender typing kits was both consistent and reproducible. Standard specimens included fresh blood, bloodstains on 903 paper or cotton cloth, and buccal swabs obtained from donors of known type. Unusual specimens included DNA isolated from different tissues of the same individual taken at autopsy (liver, brain) and DNA isolated from cigarette butts. All typing results were consistent and samples from the same individual gave the same typing results, as expected. Generally, there was no preferential amplification within a locus and alleles were fairly balanced. Typically, when data was obtained at one STR locus, the other loci were also interpretable. However, in some samples, particularly in highly degraded samples, the smaller alleles were interpretable when the larger ones were not.Standard specimens were exchanged with the Promega Corporation and the Indiana State Police laboratories. Results between all laboratories were consistent and reproducible. Examination of proficiency testing results published by CTS (Collaborative Testing Services) indicated that PowerPlex
Ô fluorescent CTT profiles agreed with silver stained CTT profiles obtained by our laboratory and by more than 13 other laboratories in repeat typing of proficiency test samples.Sensitivity
The sensitivity of the PowerPlex
Ô system was studied by amplification of 50, 25, 10, 5, 2.5, 1, 0.5, 0.25, 0.1, 0.5, and 0.025 ng of a genomic DNA sample heterozygous for 7 of 8 loci. The manufacturer suggests using 1-2 ng of template DNA, however DNA quantities from 50 to approximately 0.25 ng per reaction (depending on the locus) produced interpretable results. Significant stutter, particularly in the vWA locus, was observed at concentrations of greater than 2.5 ng input DNA. Amelogenin was detectable at DNA concentrations as low as 0.1 ng of input DNA.DNA Mixtures
Two purified DNA samples, mixed after extraction, were reliably typed up to a 1:2 to 1:4 ratio, depending on the locus, using 1 ng template DNA (this translates to 0.2 to 0.3 ng of the minor DNA component). Increasing the level of input template DNA would increase the ratio at which the minor component can reliably be typed. Analysis of a variety of mixed DNA samples revealed the PowerPlex
Ô STR multiplex to be a powerful tool for interpretation of mixed samples. Figure 2 shows how four of six possible contributors can be excluded from an equal mixture of two individuals (lane 9) using the PowerPlexÔ multiplex system (National Institute of Standards and Technology STR study). Alleles that exclude individuals are circled at each locus (CTTv loci illustrated).Non-probative Casework
Testing specimens from previously adjudicated casework corroborated previous results obtained by this laboratory using RFLP, DQA1, Polymarker, CTT, and D1S80. Studies included non-probative evidentiary stain materials (blood and semen containing samples) for both inclusionary and exclusionary cases. Repeat testing of paternity cases also corroborated previous results obtained by RFLP and CTT (silver stain) testing.
DNA extracted from various substrates gave reproducible results. Inhibitors of amplification, for example those from blue denim, may co-extract with human DNA. Results indicated that the addition of BSA (160 ug/ml final concentration) or additional Taq polymerase (5 U/25 ul) to reactions containing inhibitors from blue denim relieved the inhibition.
Unusual Genetic Variants and Artifactual Bands
Microvariants, off-ladder alleles and three-band patterns have been observed (Figure 3). We have observed two three band patterns in typing approximately 2000 samples, one at the TPOX locus and one at the D7S820 locus. Repeat extraction and amplification produced the same results. Both three band patterns await further confirmation by single-locus amplification.
Typing results indicated that stutter was proportionally higher in the vWA locus. In addition to the stutter bands (n-4) that are seen associated with authentic alleles in vWA and D5S818, we have noted that when samples are overloaded, there may be additional artifactual bands. These migrate at approximately (n-2/3), (n-5/6) and (n-6/7) in vWA and (n-2/3) in D5S818. These bands may be more intense than the n 4 stutter bands, but are easily identified by their anomalous size and reduced intensity compared to the authentic alleles which they accompany.
Operational Considerations
The combined PowerPlex
Ô multiplex/FMBIO® scanner system is remarkably robust and allows easy recognition of operational problems. These are minor and are usually associated with gel manufacture or gel loading/running and include gel bubbles, perturbation of the gel during polymerization and inconsistent gel thickness. The use of CXR fluorescent ladder as an internal marker for allele sizing makes the system robust enough to correctly identify alleles in this situation.PowerPlex
Ô Casework StatusThe PowerPlex
Ô STR multiplex is currently in use in our laboratory for convicted-offender databanking, forensic casework, and paternity testing. The combination of the PowerPlexÔ multiplex and the FMBIO II® scanner has provided for the development of an efficient, high-throughput profiling system at a cost of less than $50 per sample for offender profiling. As of September 1997, more than 2000 samples have been tested and the current rate of processing is approximately 800 samples per month.The PowerPlex
Ô Amelogenin multiplex combination is performing well in forensic casework, both with single source samples and with mixed samples of two or more individuals (Figure 5). It has also performed well in both criminal and civil paternity testing, where it has been used for cases involving amniocyte samples and samples with very limited or degraded DNA.Our experience with the combination of the PowerPlex
Ô STR multiplex and the FMBIO II® scanner indicates that it is a powerful technology that may significantly improve the utility of DNA testing for the criminal justice system.Go to proceedings home page