SEX DETERMINATION ASSAY FOR DEGRADED OR LOW QUALITY DNA BASED ON PYROSEQUENCING OF HIGH COPY NUMBER LOCI By Amanda Buszek A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Forensic Science 2012 ABSTRACT SEX DETERMINATION ASSAY FOR DEGRADED OR LOW QUALITY DNA BASED ON PYROSEQUENCING OF HIGH COPY NUMBER LOCI By Amanda Buszek Sex determination is of utmost concern in the forensic community for developing the biological profile of an individual. Current molecular techniques are successful with high molecular weight DNA, but are inadequate for low quality or quantity DNA samples. In this study, a more sensitive pyrosequencing assay was developed for sexing challenging samples, such as those originating from aged bone or hair shafts. DYZ1, a high copy repetitive element on the Y chromosome, was multiplexed with Ya5 Alu repetitive element as a human and female control. Amplification with biotinylated primers allowed for solid-phase PCR product preparation and pyrosequencing of all DNAs. This assay was developed with high molecular weight male and female control DNAs and resulted in distinct pyrograms for both sexes. This sexing assay was then tested on several suboptimal sources of DNA: fingerprints deposited on glass, shed cells on worn T-shirts, hair shaft extracts, and ancient skeletal remains. Pyrosequencing sexing results were compared to previous sexing results from standard amelogenin, real-time PCR amelogenin, and/or real-time Alu/DYZ1 testing approaches. Of the 216 extracts tested, 59.5% generated a correct sex result using the Alu/DYZ1 assay. If samples that did not yield sequencing results were excluded from the data set, 80.5% of all samples were correctly sexed. Ultimately, the pyrosequencing Alu/DYZ1 sexing assay was more sensitive and accurate than the amelogenin sexing methods. ACKNOWLEDGEMENTS I would first like to thank my advisor, Dr. David Foran, for all of his support throughout my research and graduate career at Michigan State University. I am grateful for the opportunities and experiences I have gained while under his guidance that have prepared me for a successful professional career. I would also like to acknowledge Dr. Jeff Landgraf and Dr. Charles Corley for sitting on my thesis committee and providing advice for my research. I am also extremely thankful for Brianne Kiley and Carrie Jackson whose previous research and assay development was the basis for this study. Thanks also go to Suzanne Shunn, Elizabeth Graffy, and Sarah Thomasma for their work obtaining and extracting the samples used in this study. I am especially thankful for the loving support, patience, and motivation from my friends, family, and fellow graduate students in the Michigan State University Forensic Sciences Program. iii TABLE OF CONTENTS LIST OF TABLES......................................................................................................................... vi
 LIST OF FIGURES ...................................................................................................................... vii
 INTRODUCTION .......................................................................................................................... 1
 Sex Determination ...................................................................................................................... 1
 Anthropological Sex Determination ....................................................................................... 1
 Molecular Sex Determination ................................................................................................. 2
 Limitations of Traditional Molecular Sexing ......................................................................... 3
 Repetitive Elements Assayed in Forensics ................................................................................. 5
 Sex Chromosome Repetitive Elements................................................................................... 6
 Sensitivity Concerns of Multicopy Loci Assays..................................................................... 7
 Improvements in Molecular Sexing............................................................................................ 8
 Current DNA Sequencing Technology....................................................................................... 9
 Sanger Sequencing.................................................................................................................. 9
 Pyrosequencing ..................................................................................................................... 10
 Pyrosequencing Application and Improvements .................................................................. 13
 Goals and Objectives of This Study.......................................................................................... 14
 MATERIALS AND METHODS.................................................................................................. 15
 Assay Development on Control DNA ...................................................................................... 15
 Alu Primer Optimization....................................................................................................... 15
 DYZ1 Primer Optimization .................................................................................................. 18
 Multiplexed PCR Alu/DYZ1 Assay ..................................................................................... 20
 Primer Contamination Control.............................................................................................. 20
 Pyrosequencing the Alu/DYZ1 Multiplex ............................................................................ 21
 Application of Assay to Degraded and Low Quality DNA ...................................................... 24
 RESULTS ..................................................................................................................................... 26
 Assay Development .................................................................................................................. 26
 Alu Primer Optimization and Amplification ........................................................................ 26
 DYZ1 Primer Optimization and Amplification .................................................................... 27
 Alu/DYZ1 Multiplex PCR Assay ......................................................................................... 28
 Pyrosequencing the Alu/DYZ1 Multiplex ................................................................................ 29
 Application of the Assay to Degraded and Low Quality DNAs............................................... 32
 Sexing Touch DNAs with Alu/DYZ1 Assay........................................................................ 34
 Sexing Contact DNAs with Alu/DYZ1 Assay...................................................................... 36
 Sexing Hair Shafts with Alu/DYZ1 Assay ........................................................................... 38
 Sexing Skeletal Remains with Alu/DYZ1 Assay ................................................................. 41
 DISCUSSION............................................................................................................................... 44
 CONCLUSION............................................................................................................................. 54
 iv WORKS CITED ........................................................................................................................... 56
 v LIST OF TABLES Table 1. Alu Primer Sequences.................................................................................................... 17
 Table 2. Alu Primer Pairs and Amplicon Lengths....................................................................... 18
 Table 3. DYZ1 Primer Sequences ............................................................................................... 19
 Table 4. DYZ1 Primer Pairs and Amplicon Lengths................................................................... 19
 Table 5. Pyrosequencing Dispensation Orders ............................................................................ 22
 Table 6. Dispensation Order A with Projected Sequences .......................................................... 23
 Table 7. Dispensation Order F with Projected Sequences........................................................... 24
 Table 8. Alu Amplification of Control DNA............................................................................... 27
 Table 9. DYZ1 Amplification of Control DNA........................................................................... 28
 Table 10. Sexing Results of Touch DNAs................................................................................... 35
 Table 11. Sexing Results of Contact DNAs................................................................................. 37
 Table 12. Sexing Results of Hair Shaft DNAs ............................................................................ 39
 Table 13. Sexing Results for Ancient Skeletal Remains ............................................................. 42
 vi LIST OF FIGURES Figure 1. Location of Repetitive Elements on the Y Chromosome............................................... 7
 Figure 2. The General Chemical Cascade of Pyrosequencing..................................................... 11
 Figure 3. Example Pyrogram ....................................................................................................... 12
 Figure 4. Map of DYZ1 and Alu with Tested Primers ................................................................ 16
 Figure 5. Alu/DYZ Assay Agarose Gel....................................................................................... 28
 Figure 6. Male DNA Pyrogram Obtained with Dispensation A.................................................. 29
 Figure 7. Male DNA Pyrogram Obtained with Dispensation F................................................... 30
 Figure 8. Female DNA Pyrogram Obtained with Dispensation F ............................................... 30
 Figure 9. Negative Control Pyrogram.......................................................................................... 31
 Figure 10. Portion of Female Pyrogram Showing Incorporation Error ....................................... 32
 Figure 11. Representative Pyrogram Lacking Baseline Drift...................................................... 33
 Figure 12. Representative Pyrogram with Baseline Drift............................................................ 33
 vii INTRODUCTION Determining the sex of human tissue is routine in forensic casework and medical genetic laboratories that perform prenatal diagnoses and screen for organ transplants. In forensics, sexing human remains is essential in order to develop a physical profile of a suspect or victim. Correct sex determination can limit a pool of suspects, or aid in identifying missing persons or victims of mass disaster. Visual (anthropological) sexing is effective on full adult skeletons, but cannot be applied to sub-adults or remains lacking specific bones. Molecular approaches to sexing remains can alleviate problems with skeletal age or completeness. Typical DNA analysis of submitted biological evidence in crime laboratories includes amplification of autosomal short tandem repeats (STRs) and X and Y chromosomal loci for sex determination. Although current anthropological and molecular sexing techniques are extremely effective for uncompromised remains, they often fail when applied to challenging (low quality or quantity) samples. Sex Determination Anthropological Sex Determination In cases with intact adult skeletal remains, anthropological techniques are utilized to develop a biological profile, which includes estimates of sex, stature, and age. Limitations arise, however, when sub-adult skeletons are examined or when the analyzed remains do not include the bones essential for anthropological sexing: generally the skull and/or pelvis. The rate of positive identification is low in these instances, with the complete biological profile of an individual remaining uncertain. Even if the skull and most of the skeleton is recovered, the 1 biological profiles of an estimated 44% of individuals still cannot be properly determined (Komar and Potter, 2007). Molecular Sex Determination Alternatives to anthropological sexing are molecular techniques based on sequence comparisons of the X and Y chromosomes. Some authors have focused solely on the presence or absence of the Y chromosome for determining male or female respectively (Honda et al., 1990; Hummel and Herrmann, 1994). If amplification using the polymerase chain reaction (PCR) is not successful, however, which is common in degraded or low quantity samples, Y chromosome analyses will result in these samples being identified as female, when in fact they are male. As advised by the International Society for Forensic Haemogenetics (1992), sex testing must not rely on the absence of amplification; sizing of a separate, control region is required to avoid males testing as females. This may include targeting specific loci of both sex chromosomes, or targeting one sex locus and an autosomal locus. A common approach for molecular sexing is based on the single copy locus amelogenin, which encodes a protein for tooth enamel development (Nakahori et al., 1991). Homologous copies of the gene are located on both sex chromosomes, with the X chromosome having a 6 bp deletion (Sullivan et al., 1993) within the first intron. The gene is located on the short arm of the X chromosome at p22.1–22.3 and in the pericentric region on the short arm of the Y chromosome at p11.2 (Nakahori et al, 1991; Salido et al., 1992). Based on the Sullivan et al. (1993) assay, sizing of this region is possible by amplification across the homologous X and Y regions by PCR with one pair of primers. Both male and female DNA will result in the smaller (X) product, while male DNA will have an additional larger (Y) product. These amplicons may 2 be visualized by gel or capillary electrophoresis; the latter is standard practice in forensic laboratories, which use STR multiplex kits that incorporate primers for the amelogenin locus, such as the AmpFlSTR® Identifiler™ PCR Amplification Kit (Applied Biosystems, Foster City, CA). Limitations of Traditional Molecular Sexing The amelogenin assay relies on the presence of a sufficient quantity of DNA, which for the Identifiler Kit is 0.5–1.25 ng, to produce a signal adequate for detection (Applied Biosystems, 2007). Unfortunately, many forensic samples, such as aged bone, teeth, or hair shafts, yield little or no nuclear DNA, rendering traditional molecular sexing methods unsuccessful (Andréasson and Allen, 2003). Similarly, the above-mentioned methods assay single copy loci and are susceptible to ‘stochastic sampling effects’: in low quantity DNAs, unequal sampling may result in a heterozygous locus showing just one allele, termed allelic dropout (Budowle and van Daal, 2009), and a male sample would incorrectly sex as female. General sensitivity improvements for PCR-based assays include reducing the reaction volume (Leclair et al., 2003), changing concentrations of key components such as the polymerase, or for degraded DNA, moving primer positions closer to the locus of interest (MiniFiler™; Applied Biosystems). While the former can concentrate PCR products used for downstream capillary electrophoresis, it also increases inhibitor effects and manual pipetting error is more likely to occur when a reduced PCR reaction size is utilized (Leclair et al., 2003). Another method to increase PCR sensitivity is nested PCR, which consists of two stages of amplification. The first amplifies across a larger target region; the second requires addition of one or two internal (nested) primers for additional rounds of amplification. While the higher 3 PCR cycle number increases the amount of product, especially for degraded DNA, there is additional risk of spurious amplification, along with contamination during primer addition (Strom and Rechitsky, 1998), and this method is generally not considered acceptable in forensic laboratories. Applying the amelogenin assay to degraded DNA, Mannucci et al. (1994) introduced another potential sensitivity improvement. They worked with skeletal remains that had been anthropologically sexed and were able to consistently confirm sex from as little as 20 pg DNA (about three diploid cell copies) by increasing amplification cycles to 39. Increasing PCR cycles above STR kit manufacturers' recommendations of 28–30 (Applied Biosystems) has been used for DNA analysis under 100 pg in order to increase sensitivity; 31 cycles yields a theoretical 16fold improvement in sensitivity (Caragine et al., 2009) and 34 cycles a theoretical 64-fold improvement (Gill et al., 2000; Whitaker et al., 2001). As noted above, results of these tests must be carefully analyzed, however, because more cycles also increases allelic drop-in and risk of contamination (Gill et al., 2000). Mannucci et al. (1994) also recommended performing tests in replicate whenever cycling parameters are increased, helping to identify stochastic sampling effects or instances of contamination. Even when applied to high molecular weight DNA, the amelogenin assay has limitations pertaining to genetic differences among individuals. Maciejewska and Pawłowski (2009) found cases in which the X homologue of the amelogenin locus did not amplify, confirmed by replicate testing. Subsequent sequencing showed a C to G transversion at the 3'-most base of the reverse primer-annealing location, which resulted in improper annealing, and thus no amplification. Santos et al. (1998) detailed a similarly troubling genotype: a deletion within the Y chromosome. When sized by amelogenin testing, this deletion gave false female results because it caused a 4 band that was similar in size to the female band. Deletions at this locus have been found in a variety of populations at various frequencies, e.g., 0.02% in Austria and 1.85% in India (Steinlechner et al., 2002; Thangaraj et al., 2002). Although these anomalies are rare, Santos et al. (1998) concluded that Y-specific loci, without an X homologue, should be used to ensure accurate sex results in forensic testing. Repetitive Elements Assayed in Forensics One approach to overcome sampling difficulties of single-copy loci is to assay elements that are present in many copies and located throughout the genome, occurring as repeats of core sequences ranging from a single base pair up to thousands of base pairs. Microsatellites, termed STRs in the forensic field, are characterized as repeats smaller than six nucleotides, while larger repeats are designated minisatellites (Ellegren, 2004). Once viewed as "junk" or "trash" DNA, repetitive elements are important genetic mapping tools for evolutionary studies, and play key roles in gene transcription, splicing, and expression (Jurka et al., 1992). STRs are popular DNA markers for human identification because they are easy to assay and have high variability among individuals. Mini-STRs, multiplexes with reduced amplicon size, allow for more sensitivity and have improved low copy and degraded DNA analyses in forensic casework (Chung et al., 2004). Genome-wide repeats are larger than microsatellites and may consist of core repeats of over one thousand base pairs. These repeats are found in many copies throughout the genome, occurring as isolated units or clustered closely together. Present on all 21 chromosomes, the Alu sequence is the most abundant repetitive element in the human genome. This sequence has upwards of 1 million copies of a 280 bp consensus sequence and accounts for more than 10% of the genome (Mighell et al., 1997). Alu sequences, constituting four principal subfamilies, were 5 aptly named after discovering that the restriction enzyme AluI cuts within the consensus Alu sequence (Mighell et al., 1997). Nicklas and Buel (2006) found that the subfamily Ya5 was detectable down to 0.5 pg DNA, which was 50 times more sensitive than single-copy genes (23 pg). Sex Chromosome Repetitive Elements Repetitive elements of the X or Y chromosome are as polymorphic as autosomal elements and are ideal targets for sexing assays. On the Y chromosome, repetitive elements span both the long (Yq) and short (Yp) arms and comprise 60% of the chromosome (Roewer et al., 1992). STRs of the Y chromosome are a valuable tool for male lineage determination in forensics, while larger repeat elements have been analyzed in genetic and fertility screenings (Rahman et al., 2004). One family of these elements is DYZ, subfamilies of which are numbered in order of frequency (Fig. 1). DYZ1, the most abundant subfamily, is present within the heterochromatin and is made up of a 3.4 kb repeat with up to 4000 copies (Rahman et al., 2004). DYZ2 is made up of a 2.5 kb repeat with about 2000 copies (Schmid et al., 1990), containing mainly AT-rich sequences as well as one Alu sequence per repeat (Tyler-Smith et al., 1988), and extends through the entire heterochromatic region and much of the distal end of Yq. The DYZ1 and DYZ2 subfamilies make up 70% of the chromosome's repetitive elements. DYZ3 is a 170 bp repeat with about 35 copies and the only subfamily present within the centromeric region (Tyler-Smith et al., 1988). DYZ4 and DYZ5 (Yp11.2) are present on Yp and are the smallest subfamilies of DYZ (Müller et al., 1986). DYZ5 is a 137 bp repeat with 15–50 copies (Müller et al., 1986). 6 Figure 1. Location of Repetitive Elements on the Y Chromosome Figure from Kiley, 2009. Approximate locations of SRY, AMEL-Y (amelogenin), DYZ4, and DYZ5 on the short arm (Yp) and DYZ1 and 2 on the long arm (Yq) of the Y chromosome. DYZ3 is located in the centromeric region and DYZ1 is within the heterochromatin region of Yq. Commercial kits exist for assaying Y chromosome and autosomal STRs, and multiplexes have been developed for X STRs (Diegoli and Coble, 2011). Like the Y chromosome, the X chromosome has specific repetitive elements: the DXZ family is the most common, subfamilies of which are named in order of frequency. DXZ4 is a widely studied repeat element located on Xq (Luptáková et al., 2011). It is a highly polymorphic marker composed of 12–100 three-kb repeats arranged in tandem (Giacalone, 1992). Sensitivity Concerns of Multicopy Loci Assays Due to the high abundance of repetitive elements, especially Alu, contamination by minute amounts of human DNA during primer production, DNA isolation, or PCR preparation, may be detrimental to assays. Meier et al. (1993) found bacterial contamination of primers and other reagents in their screening for various eubacteria. Using long-wave ultraviolet (UV) irradiation in combination with chemical treatment, they were able to eliminate contamination of PCR reagents. In a similar manner, Kiley (2009) found that during production, human contamination of primers may occur, which can be overcome by UV irradiation of PCR master 7 mixes (all reagents except polymerase and NTPs) for 30 seconds to 1 minute (0.25 and 0.5 2 J/cm , respectively) but treatments over 1.5 minutes rendered PCR unsuccessful. Improvements in Molecular Sexing To overcome problems with the amelogenin assay's effectiveness on degraded or low copy number DNA and avoid stochastic effects, small and/or multicopy loci can be targeted, which increase the number of targets that are potentially available for amplification. Due to their large size and highly repetitive nature, the DYZ and DXZ families are suitable targets for sensitive screening techniques, particularly sex determination. Palmirotta et al. (1997) used nested PCR to assay DXZ4 and the testis-determined factor gene (sex determining region Y, SRY) of the Y chromosome on ancient bones (600–1300 AD). Presence of a DXZ4 amplicon served as a positive control and was clearly distinguished by size from the SRY product (91 and 204 bp, respectively). All molecular results in this assay were consistent with anthropological sexing of the bones. Luptáková et al. (2011) also targeted the DXZ4 sequences and a region of the SRY gene to determine the sex of remains from the Middle Ages. Their DXZ4 and SRY amplicon sizes, however, were reduced to 85 and 102 bp, respectively, for increased amplification success. Although the 204 bp amplicon from Palmirotta et al. (1997) is short enough for routine typing on modern DNA, Luptáková et al. (2011) showed that the smaller amplicon was more successful on degraded samples. In both of these studies, however, amplification of a high copy X chromosome locus was compared to that of a single copy Y chromosome locus. Nicklas and Buel (2006) developed a highly sensitive sexing assay to quantify total and male DNA that 8 targeted two multicopy loci: Ya5, an Alu subfamily, and DYZ5. The authors were able to detect and correctly sex samples with as little as 0.5 pg of DNA. A multiplex assay targeting DYZ1 and Ya5 to sex challenging forensic samples was developed by Jackson (2006) and improved by Kiley (2009). Their work used real-time PCR with TaqMan® technology to improve specificity. This assay proved far more sensitive than the standard amelogenin sexing method, capable of detecting as little as 0.0623 pg (~1/100 of a cell) of control DNA, and even lower quantities of artificially degraded DNA. In addition to targeting multicopy loci, more sensitive techniques can also be used to yield sexing results for degraded or low copy DNA, including nested PCR (previously discussed) and pyrosequencing (detailed below). Current DNA Sequencing Technology Sanger Sequencing Sanger sequencing is an electrophoretic DNA sequencing approach based on incorporation of chain-terminating nucleotides (Sanger et al., 1977). Although this technique is used throughout scientific disciplines and is the main mode of sequencing in the field of forensics, it has limitations. Sequence reads generally begin 50–300 nucleotides from the primer and can span from 15 to many hundreds of nucleotides (Sanger et al., 1977). Degraded or low quality DNA, however, may not have intact targets that even span a few hundred base pairs. Therefore, traditional Sanger sequencing is often not successful in forensic samples, so alternative methods have been developed (Hummel, 2003). 9 Pyrosequencing Pyrosequencing is a sensitive and robust approach to determine the nucleotide sequence of DNA. It is ideal for short- and medium-length DNA sequencing of up to about 150 bp and was the first commercially available alternative to Sanger methodologies for de novo sequencing (Melamede, 1985; Gharizadeh et al., 2003; Nyrén, 2007). This technique has been utilized for a variety of testing, such as single nucleotide polymorphism (SNP) and STR analysis, quantification and discrimination of mixtures, species identification, whole-genome sequencing, and sex determination (Fakhrai-Rad, 2002; Harrison et al., 2006; Divne et al., 2010; Andréasson et al., 2006; Batlitzki-Korte, 2005; Margulies et al., 2005; Tschentscher et al., 2008). It is a sequencing-by-synthesis approach for single- or double-stranded DNA that detects nucleotide incorporation immediately adjacent to the primer in real time. Although other sequencing-bysynthesis methods exist, some of which monitor polymerase activity, pyrosequencing is the most sensitive method because it relies on the detection of pyrophosphate (PPi) (Nyrén, 2007). During a pyrosequencing reaction, deoxynucleotides (dNTPs) are systematically dispensed and PPi is released as the DNA polymerase incorporates the complementary nucleotide (Fig. 2). Accumulation of PPi is measured after a chemical cascade in which sulfurylase converts PPi to ATP and firefly luciferase converts ATP to light, which is detected by a photon detector (Ronaghi et al., 1996). Light detection is translated by software into a pyrogram (Fig. 3), in which peak height corresponds to relative quantity of light, and thus incorporated deoxynucleotides. Because dATP is also a good substrate for firefly luciferase, the modified base deoxyadenosine α-thiotriphosphate (dATPαS) is used because it does not interact with firefly luciferase, yet is still a good substrate for polymerase (Ronaghi et al., 2001). 10 Figure 2. The General Chemical Cascade of Pyrosequencing The chemical cascade of pyrosequencing begins with nucleotide incorporation by a DNA polymerase, in which pyrophosphate (PPi) is released. PPi ultimately is converted into ATP, the substrate for lucerifase to emit light. Based on Fakhrai-Rad et al., 2002. 11 = Figure 3. Example Pyrogram X-axis corresponds to dispensation order of nucleotides and y-axis to the relative abundance. Enzyme ("E") and substrate ("S") peaks correspond to internal controls. The resultant sequence of this pyrogram is TAAG. Pyrosequencing is accurate, flexible, and conducive to automation, but sample preparation and purity are crucial and account for the majority of testing time (Nordström et al., 2001). Amplified products must be purified prior to pyrosequencing using a solid-phase or enzymatic approach. Solid-phase preparation utilizes PCR product resulting from one biotinylated primer that can be purified by selection with streptavidin-coated beads. Biotinylated PCR product is incubated with and binds to streptavidin, and is then washed with ethanol, sodium hydroxide to denature the double-stranded DNA, and wash buffer to clean and neutralize the PCR product of residual sodium hydroxide (King and Scott-Horton, 2007). These wash steps also ensure removal of all pyrophosphate from PCR. The clean, single-stranded product is then incubated with sequencing primer to allow for annealing before pyrosequencing is carried out. 12 Alternatively, enzymatic procedures degrade PCR primers, PPi, and unincorporated nucleotides for preparation of a double-stranded product. Enzyme reactions for pyrosequencing preparation require as little as 4 minutes and several enzyme combinations are effective: shrimp or calf alkaline phosphatase and exonuclease I, apyrase and inorganic pyrophosphatase, or apyrase with ATP sulfurylase and exonuclease I (Nordström et al., 2000). Apyrase speeds up the pyrosequencing process by degrading unincorporated nucleotides, which, in other sequencing techniques, are removed by washings before any new nucleotides are added to the reaction. The order in which the four nucleotides are dispensed during a pyrosequencing reaction is programmable, and depends on the sequence being analyzed. For SNP analysis, dispensations include every predicted base until the point of a polymorphism; then both nucleotides are dispensed, followed by the remaining bases of the known sequence (Fakhrai-Rad, 2002). Resulting pyrograms then reveal which SNP was present. This is referred to as directed dispensation order. Conversely, cyclic dispensation is used when a sequence is not known and dispensation orders cannot be easily predicted. This approach consists of repeats of the four nucleotides (e.g. [CATG]x) being dispensed in order to determine the unknown sequence. Pyrosequencing Application and Improvements Since its development, advances in pyrosequencing have resulted in more cost-effective reactions: $0.20/DNA fragment (Pourmand et al., 2002) and likely even lower costs currently. The use of a vacuum for solid-phase preparation is the main price-reducing improvement. It relies on vacuum filtration with 10-µm-diameter pores to capture and hold the biotin-labeled PCR product bound to streptavidin-coated beads (34 µm) while transferring through a series of washing solutions. Complete aspiration of the washing solutions through the vacuum allows for 13 maximum contact with the biotin-labeled PCR product (Dunker et al., 2003). Filters can be reused after bead release, performing well after 100 repeated assay preparations (Dunker et al., 2003; Qiagen, 2008). Preparation takes approximately 30 seconds and platforms with 24–96 vacuum probes allow for increased processing rates and reduced reagent volumes and cost. Pyrosequencing multiplexes are used to simultaneously assay more than one target using multiple sequencing primers, and have been optimized to detect sequence variants and screen for SNPs (Pourmand et al., 2002). These advances are applicable to forensics, as they have eliminated the need for sometimes controversial methods, including nested PCR or stringent PCR conditions for testing low quantities of DNA (Gharizadeh et al., 2003). While pyrosequencing many targets at once, directed dispensations of nucleotides produces unique pyrograms for the multiplexed assay (Divne et al., 2010). Advantages of this approach include shortened processing times allowing for higher throughput of samples, and decreased PCR template preparation, and genotyping costs (Pourmand et al., 2002). Goals and Objectives of This Study Although sexing assays using Alu and DYZ1 sequences have been developed previously, none used pyrosequencing technology to clearly distinguish male and female samples. The goal of this study was to create a highly sensitive sex-determination assay for pyrosequencing of the Alu and DYZ1 sequences. Several primer pairs were tested for PCR amplification of both regions and optimized for multiplexing, with one biotinylated primer for each target sequence. Once the assay was optimized on high quality, control male and female DNAs, it was applied to various low quantity and quality DNA samples, similar to those found in a forensic laboratory, including touch and contact DNA samples, hair shafts, and ancient skeletal remains. 14 MATERIALS AND METHODS Pre-amplification procedures were performed on fresh bench paper while wearing lab coat, protective sleeves, facemask, and two pairs of gloves. Outer gloves were replaced between 2 sample preparations. Reagents were UV-irradiated for 5 min per side (~2.5 J/cm ), if applicable, in a Spectrolinker XL-1500 UV Crosslinker (Spectronics Corporation; Westbury, NY). Assay Development on Control DNA Alu Primer Optimization Alu PCR and sequencing primers (Fig. 4, Table 1) were designed by Nicklas and Buel (2007). Additional primers were designed using Beacon Designer 7 (BD7) Software (PREMIER BioSoft International; Palo Alto, CA). Melting temperatures were calculated using BioMath Calculator (Promega) or provided by BD7. Primers were obtained from Integrated DNA Technologies (IDT®; Coralville, IA) or Sigma-Aldrich® (St. Louis, MO) and purified by standard desalting during production. Biotinylated (5') sequencing primers were purified by high-performance liquid chromatography during production. Primers were diluted in 18.2 MΩ filter-sterilized water (pore size 0.22 µm), and were filtered through Microcon Ultracel® YM100 centrifugal filters (Millipore Corporation; Billerica, MA) for 12 min at 500 x g Alu forward and reverse primers were tested for optimal concentrations at 0.2 µM, 0.5 µM, 0.9 µM, 2 µM, 3 µM, 5 µM, and 20 µM, in 10 µL-reactions consisting of 0.75 U AmpliTaq Gold® DNA polymerase (Applied Biosystems; Foster City, CA), 2.5 mM MgCl2, GeneAmp® 10X PCR Buffer II, 0.2 mM each dNTP, 1 ng male (AmpFlSTR® Control DNA 007) or female (Female Control DNA 9947a) control DNA, and sterile water to volume. 15 RA RB A B RV RD FII RBK RC RX F2 FPBK/ RPBK FIV RY F4 F3 FI FIII RZ RW F1 F5 FV FBK Figure 4. Map of DYZ1 and Alu with Tested Primers (a) Map of DYZ1 with forward (solid arrow) and reverse (dashed arrow) primers. (b) Map of Alu with forward (solid arrow) and reverse (dashed arrow) primers. Hash marks separate 10 bp regions. For interpretation of the references to color in this an all other figures, the reader is referred to the electronic version of this thesis. 16 Table 1. Alu Primer Sequences Alu forward (F) and reverse (R) primer sequences with melting temperatures (Tm), in ºC. Tm marked with * indicates primers designed by Nicklas and Buel (2006) with melting temperature calculated using BioMath Calculator; all others were designed using Beacon Designer 7 Software, which generated the listed melting temperature. Primer Name FI FII FIII FIV FV FBK FPBK RV RW RX RY RZ RBK RPBK Length (bp) 18 19 24 18 20 24 15 19 16 20 18 17 20 15 Primer Sequence (5'–3') AACGGTGAAACCCCGTCT GCAGGAGAATGGCGTGAAC TAAAAATACAAAAATTAGCCGGGC AATTAGCCGGGCGTAGTG AGATCGAGACCATCCCGGCT GAGATCGAGACCATCCCGGCTAAA GGGCGTAGTGGCGGG CCTCAGCCTCCCAAGTAGC GCGCGATCTCGGCTCA GCCCTGCTAATTTTTTGTAT TTGAGACGGAGTCTCGCT AGGCTGGAGTGCAGTGG CTCAGCCTCCCAAGTAGCTG CCCGCCACTACGCCC Tm (ºC) 56.8 56.6 52.5 55.0 60.4 59* 53* 57.6 58.4 48.8 55.5 57.8 56* 53* Thirteen combinations of PCR master mixes were compared for amplification of male and female control DNA, including appropriate negative controls. Amplicons ranged in size from 47 to 298 bp (Table 2). Five microliters of PCR products were separated by 4% agarose gel electrophoresis using xylene cyanol loading dye and stained with ethidium bromide; if properly sized amplicons were produced, PCR was performed with biotinylated primers. 17 Table 2. Alu Primer Pairs and Amplicon Lengths Alu primer pairs used for PCR amplification, with resultant amplicon lengths (bp). Forward and reverse abbreviations refer to those in Table 1. Pair Number 1 2 3 4 5 6 7 8 9 10 11 12 13 Forward FI FI FII FII FIII FIII FIV FIV FV FV FBK FPBK FBK Reverse RZ RW RX RY RY RZ RV RX RW RV RBK RBK RPBK Amplicon Length (bp) 160 144 140 98 165 138 239 195 169 298 115 47 83 DYZ1 Primer Optimization Forward and reverse primers and applicable sequencing primers (Fig. 4, Table 3) were designed with BD7 or by Kiley (2009) and obtained in the same manner as Alu primers. Dilution, amplification, separation, and storage proceeded according to the parameters detailed above. 18 Table 3. DYZ1 Primer Sequences DYZ1 forward (F) and reverse (R) primer sequences with melting temperatures (Tm) in °C. Primer Name F1 F2 F3 F4 F5 RA RB RC RD Length (bp) 22 23 23 23 24 22 21 25 24 Sequence (5'-3') Tm (ºC) TGTCCATTACACTACATTCCCT 53.3 TCCATTCTATTCTCTTCTACTGC 51.7 ATTCCATTCCAATCCATTCCTTT 52.8 ATTCCAATCCATTCCTTTCCTTT 52.7 GGCCTGTCCATTACACTACATTCC 57.4 GAATTGAATGGAATGGGAACGA 53.4 GAATGGAATGGGAACGAATGG 54.0 GCAGTAGAAGAGAATAGAATGGAAT 52.4 GGAGTGAAATTGTATGCAGTAGAA 53.0 Amplification of ten reactions with different combinations of forward and reverse primers (Table 4) was performed with male control DNA, including appropriate negative controls. Amplicons ranged in size from 57 to 138 bp. Gel electrophoresis was used to separate products as noted above, and if properly sized amplicons were produced, PCR was performed with biotinylated primers. Table 4. DYZ1 Primer Pairs and Amplicon Lengths DYZ1 primer pairs used for PCR amplification, with resultant amplicon lengths (bp). Primer abbreviations refer to those in Table 3. Pair Number 1 2 3 4 5 6 7 8 9 10 Forward F1 F1 F2 F2 F3 F3 F4 F4 F5 F5 Reverse RC RD RA RB RC RB RA RD RB RC 19 Amplicon Length (bp) 100 115 62 57 60 94 94 70 138 104 Multiplexed PCR Alu/DYZ1 Assay Primers for amplification of the Alu and DYZ1 loci were combined in 20 µL reactions, including 0.75 U AmpliTaq Gold® DNA polymerase, 2 µM Alu-F, DYZ1-F, Alu-R, and DYZ1R primers, 1 ng male or female control DNA, GeneAmp® 10X PCR Buffer II, 0.2 mM each dNTP, and water to volume. A temperature gradient assay was run on a Mastercycler® gradient thermal cycler (Eppendorf Scientific Inc.; Westbury, NY) to determine optimal annealing temperature between 48 and 64°C at the following cycling parameters: initial hold at 94°C for 10 min, 35 cycles of 94°C denaturation for 30 s, annealing for 1 min, 72°C extension for 20 s, and a final extension at 72°C for 5 min. Optimized conditions were used for subsequent experiments. Separation and storage of products was performed according to the parameters detailed above. Primer Contamination Control Based on previous methods utilized in the Michigan State University Forensic Biology Laboratory (Kiley, 2009), DYZ1/Alu multiplex PCR reaction master mixes were subjected to 2 2 2 2 UV irradiation for 0 min (0 J/cm ), 15 s (0.125 J/cm ), 30 s (0.25 J/cm ), 1 min (0.5 J/cm ), 2 2 2 2 2 min (1 J/cm ), 3 min (1.5 J/cm ), 4 min (2 J/cm ), and 5 min (2.5 J/cm ). Master mixes included all PCR reagents except polymerase, dNTPs, and DNA, which were added after irradiation. UVirradiated master mixes were also tested for optimal final volumes of 10, 15, and 25 µL. Irradiation times that produced no amplification in negative controls and proper positive results were used for the remainder of the study. 20 Pyrosequencing the Alu/DYZ1 Multiplex Pyrosequencing was conducted on a fully automated microtiter plate-based PyroMark Q24 Pyrosequencer (Qiagen; Germantown, MD) with manual sample preparation on a PyroMark® Q24 vacuum workstation in SQA mode according to the manufacturer's instructions, using Streptavidin™ Sepharose High Performance Beads (GE Healthcare; South San Francisco, CA). Sequencing reactions were run using method parameter code 001 Rev. A, correlating to the cartridge used. Sequencing reaction volumes were tested for optimal amount of PCR product between 5 and 20 µL. Primers were used at 0.4 µM each, as suggested by the manufacturer, for multiplex sequencing. Data were analyzed using PyroMark® Q24 Software 2.0.6. Dispensation orders were generated specific to the target sequence(s) and sequencing primers (Table 5), using the following guidelines, based on manufacturer's recommendations: 1. Core order is sequence of target immediately adjacent to sequencing primer 2. Repeat dispensation of nucleotides for polynucleotide stretches 3. Dispense erroneous bases at beginning and scatter throughout 4. End dispensation with simple repeated pattern (e.g. [CAGT]2) 21 Table 5. Pyrosequencing Dispensation Orders Dispensation order of nucleotides for pyrosequencing, specific to sequencing primer and target sequence. All dispensations were followed by (AAGGTTCC)2-3 Sequencing Primer F3 F4 RBK or RPBK FBK FBK + F3 Dispensation Order GCCTTTTCAGCTTAGCATTTCCAT TTCGCTTAGCTATTCCTATTCTAT TGAAGTAAAGCTGGATGCATAACGCCTCGATCGT AGCGTGATCGTCAGCTACGTGCAGTCACTCGACC a. ATGCGATTCTACCGACTTGAC b. AACGTCGACTGCATCATCTC c. TATCGTTCTACCGACTTGAC d. TACGTCGAGTGCTCAT e. TACGTCGACTGCTCATC f. TACGTCGACTGTCATC Based on dispensation orders for the multiplex assay, sequences for a male and female result can be projected and are detailed in Tables 6 (order a) and 7 (order f). 22 Table 6. Dispensation Order A with Projected Sequences Alu and DYZ1 sequences based on dispensation order (a) using sequencing primers F3 and FBK. Numerical value or dash (-) indicates number of bases. Dispensation order A T G C G A 2T C T A 2C G A C 2T G A C 2A 2G 2T 2C 2A 2G 2T 2C 2A 2G 2T 2C Alu 4 1 2 1 1 3 4 1 1 1 1 1 1 1 23 DYZ1 2 3 1 1 1 2 1 1 1 2 2 1 2 1 1 - Table 7. Dispensation Order F with Projected Sequences Alu and DYZ1 sequences based on dispensation order (f) using sequencing primers F3 and FBK. Numerical value or dash (-) indicates number of bases. Dispensation Order T A C G T C G A C T G T C A T C 2A 2G 2T 2C 2A 2G 2T 2C Alu 4 1 2 1 1 3 4 1 1 1 1 1 1 1 1 1 1 DYZ1 2 3 1 1 1 2 1 2 2 1 2 1 1 2 2 1 2 1 Application of Assay to Degraded and Low Quality DNA The optimized assay was applied to 216 DNA samples maintained in the Forensic Biology Laboratory at Michigan State University: 55 touch DNAs (IRB #07-577), 34 contact DNAs (IRB #10-410), 72 hair shaft extracts (IRB#03-306), and 55 ancient bone samples (Shunn, 2005). Two microliters of DNA extract were added to 20 µL PCR reactions. If less extract was available, 2 µL of water was added to the tubes, vortexed, and used for PCR. Touch and contact 24 DNA quantities had been obtained previously using a Quantifiler® Human DNA Quantification Kit on an iQ5 Optical System (BioRad; Hercules, CA). Touch DNAs were collected from laid fingerprints on glass and STR profiles were obtained for samples that quantified less than 50 pg/µL using AmpFlSTR® Minifiler™ PCR Amplification Kit. Touch DNAs were chosen for Alu/DYZ1 sexing if they quantified at ≤ 50 pg/µL (Thomasma, in press). Contact DNAs were obtained from T-shirts worn by 20 individuals and STR profiles were obtained using an AmpFlSTR® Identifiler™ PCR Amplification Kit. Contact DNAs were chosen for Alu/DYZ1 sexing if they had exhibited inconclusive amelogenin results and/or quantified less than 50 pg/µL, and showed no signs of contamination or mixtures (Thomasma, unpublished). DNAs from hair shafts had been extracted previously by alkaline and organic methods (Graffy and Foran, 2005) and were analyzed for sex using real-time assays of amelogenin and Alu/DYZ1 (Kiley, 2009). Sex was known for all touch DNAs, contact DNAs, and hair shafts. The bone extracts were obtained from 14 adult and three subadult skeletons interred between 1743 and 1781 in Fort Michilimackinac, Michigan (Shunn, 2005). Adult remains had been anthropologically sexed, but no genetic sexing had been performed. Resultant pyrograms were compared to control male and female pyrograms and identified as male, female, or absent (obtained no sequence). 25 RESULTS Assay Development Alu Primer Optimization and Amplification Initial PCR amplification produced amplicons in all negative controls, regardless of primer pair tested. In subsequent rounds of amplification with filtered master mixes, eight combinations (primer pairs 1, 2, 6, 8, 10–13) produced amplicons (Table 8) in positive male and female controls. Correctly sized product from female control DNA was produced using pair 3, but no amplification of the male control DNA occurred. Although a product was produced using pair 10, it was much smaller than the predicted size. Primer pairs 3 and 10, therefore, were removed from further testing. Of the remaining pairs, 11–13 were used for subsequent testing because they had the smallest amplicons (Table 2). Negative PCR controls amplified using these pairs, where "negative" bands were as intense as positive control bands, but UV irradiation of 2 master mixes for 3 minutes (1.5 J/cm ) eliminated contamination. Results were consistent among primers from both distributors (pairs 1–10) and between replicate testing of pairs 11–13, which were only obtained from Sigma-Aldrich®. Optimal primer concentrations were 2 µM for FBK, RBK, FPBK, and RPBK. 26 Table 8. Alu Amplification of Control DNA Alu target amplification (+/-) using control DNAs and primer pairs listed in Table 2. Amplification of incorrectly sized product indicated by *. Alu Primer Pair 1 2 3 4 5 6 7 8 9 10 11 12 13 Male + + + + +* + + + Female + + +* + + + + +* + + + Negative - DYZ1 Primer Optimization and Amplification Initial amplification of male control DNA produced correctly sized amplicons using all master mix combinations, except those with primer pairs 1 and 4 (Table 9). Subsequent PCR with filtered master mixes and UV irradiation for 3 minutes did not affect amplification. Negative controls and female DNA did not amplify. Optimal primer concentrations were 2 µM for F1, F3, and RC. 27 Table 9. DYZ1 Amplification of Control DNA DYZ1 target amplification (+/-) using control DNAs and primers pairs listed in Table 4. DYZ1 Primer Pair 1 2 3 4 5 6 7 8 9 10 Male + + + + + + + + Female - Negative - Alu/DYZ1 Multiplex PCR Assay Amplification of control DNAs using DYZ1 primer pair 5 and Alu pair 11 produced amplicons that could be visually differentiated on a gel, 60 bp and 115 bp, respectively (Fig. 5). Optimized PCR annealing temperature for these primers ranged from 54.4°C to 58.8°C. N F M  Alu (115 bp)  DYZ1 (60 bp) Figure 5. Alu/DYZ Assay Agarose Gel Products of negative (N), female (F), and male (M) amplification of DYZ1 and Alu targets using master mix with primer pairs 5 and 11, respectively. Size of amplicon is included in parentheses. Band in negative is primer dimer. 28 Pyrosequencing the Alu/DYZ1 Multiplex Pyrosequencing was successful with sequencing primers F3 and F4 for the DYZ1 target using the dispensation orders listed in Table 5. Pyrosequencing the Alu target was only successful using primer FBK with the listed dispensation orders. Sequencing with primers RPBK or RBK resulted in inaccurate incorporation across polynucleotide stretches, although the order was correct. Assaying both DYZ1 and Alu targets simultaneously with sequencing primers F4 and FBK and dispensation order (a) resulted in clean DYZ1 sequence from male controls; however no Alu sequence was detected. Sequencing with primers F3 and FBK resulted in pyrograms with a clean combination of DYZ1 and Alu sequences for male control DNAs, and a pure Alu sequence for female control DNAs. A representative pyrogram of male DNA is shown in Figure 6. Figure 6. Male DNA Pyrogram Obtained with Dispensation A Representative pyrogram of male DNA using dispensation order (a) with sequencing primers F3 and FBK. 29 Although dispensation order (a) resulted in clean sequences, extraneous and duplicate bases made it necessary to test more concise orders (Table 5). Order (f) had the highest signal intensity and least extraneous bases while still having clean sequences for male and female control DNAs. Representative pyrograms of positive and negative controls are given in Figures 7 (male), 8 (female) and 9 (negative). Note that the signal intensity for the male and female DNA is approximately 100, while the negative sample is less than 4. Figure 7. Male DNA Pyrogram Obtained with Dispensation F Representative pyrogram of male DNA using dispensation order (f) with sequencing primers F3 and FBK. Figure 8. Female DNA Pyrogram Obtained with Dispensation F Representative pyrogram of female DNA using dispensation order (f) with sequencing primer F3 and FBK. 30 Figure 9. Negative Control Pyrogram Representative pyrogram of a negative control using dispensation order (f) and sequencing primers F3 and FBK. Twenty microliters of control DNA PCR product frequently resulted in nucleotide incorporation between dispensations, shown in Figure 12. Five microliters was the minimum volume of control DNA placed into the immobilization master mix that yielded clean pyrograms with no spurious incorporation. One anomaly existed that prevented perfect sequencing results, but it did not interfere with sequence analysis. Regardless of the dispensation order used, results did not reflect the beginning Alu sequence of AAAACGT; only one dATPαS was incorporated in all positive control results. Figures 7 and 8 show this anomaly, but an enlarged pyrogram is given in Figure 10 for better resolution. Instead of correctly sequencing the Alu target, these results show "ACGGT". This result occurred in all control and sample pyrograms, so measures to improve pyrosequencing were not evaluated. 31 Figure 10. Portion of Female Pyrogram Showing Incorporation Error The results of these sequencing results are "ACGGT." Note the second peak (A) is not four times the height of the third peak (C), as should occur if the correct sequence of Alu was obtained ("AAAACGGT"). Application of the Assay to Degraded and Low Quality DNAs An artifact frequently found while analyzing degrade and low quality DNAs was baseline drift. One explanation provided by Pyromark user manual was large fluctuations in ambient temperature and/or unstable camera temperatures. Allowing the machine to acclimate to ambient temperatures for one hour (recommended by manufacturer) did not resolve the artifact, but two hours notably decreased baseline drift. Over time, warm weather counteracted the two hour acclimation period and baseline drift was more prevalent. Note the difference in drift from a pyrogram produced on April 4 (Fig. 11) and on July 16 (Fig. 12). 32 Figure 11. Representative Pyrogram Lacking Baseline Drift Pyrogram produced on April 4, 2012 showing no baseline drift. Figure 12. Representative Pyrogram with Baseline Drift Pyrogram produced on July 16, 2012 showing moderate baseline drift. The box denotes spurious peak incorporation between two thymine dispensations. Optimal volume of PCR product from touch, contact, hair, and skeletal DNAs placed into the immobilization master mix was 10 µL. Addition of less than 10 µL repeatedly produced pyrograms with peaks heights too low to analyze (not shown). 33 Sexing Touch DNAs with Alu/DYZ1 Assay Sex determination of the 55 touch samples and the known sex for each donor are displayed in Table 10. Forty-five (82%) touch DNAs, 20 male and 25 female, matched the known sex. Eight samples lacked sufficient extract volume for the PCR reaction and correct sexing results were obtained from 6 of them, while the remaining two (ST-28, and 54) failed to amplify. Three samples (ST-1 and 7, which quantified previously at 0.00 pg/µL, and ST-50) failed to amplify, despite having sufficient extract remaining. Incorrect male results were obtained from five touch DNAs (ST-19, 30, 42, 46, 49), while no samples incorrectly sexed as female. 34 Table 10. Sexing Results of Touch DNAs Touch DNAs previously quantified (in pg/µL) with known sex and sexing results from Alu/DYZ1 assay, are noted as male (M), female (F), or no sequence was obtained (-). (*) represents DNAs that had less than 2 µL extract available for PCR. DNAs that did not amplify are denoted by (‡). Results inconsistent with known sex are bolded. ID ST-1‡ ST-2 ST-3 ST-4 ST-5 ST-6 ST-7‡ ST-8 ST-9 ST-10 ST-11 ST-12 ST-13 ST-14 ST-15 ST-16 ST-17 ST-18 ST-19 ST-20 ST-21 ST-22 ST-23* ST-24 ST-25 ST-26 ST-27 ST-28*‡ Quantity (pg/µL) 0 0 0 6.77 8.07 8.39 9.22 10.8 11 11.4 12.5 12.6 14 15.3 15.8 16.1 18.7 21.6 22.1 23 24.1 24.6 24.6 24.6 24.9 25 26.3 26.4 Known M M M M F F M F F F F M F F F F M F F M F F F F F M M F Alu/DYZ1 M M M F F F F F F M F F F F M F M M F F F F F M M - ID ST-29 ST-30 ST-31 ST-32 ST-33 ST-34 ST-35 ST-36* ST-37* ST-38* ST-39 ST-40 ST-41* ST-42 ST-43 ST-44 ST-45 ST-46 ST-47* ST-48 ST-49 ST-50‡ ST-51 ST-52 ST-53 ST-54*‡ ST-55 35 Quantity (pg/µL) 26.4 26.5 27.6 28.7 30.5 31.1 31.2 35 37.5 37.7 39 39.6 40.7 41.1 41.5 42.3 42.5 43.1 43.9 44.4 45.5 46 47.2 47.3 49.7 50.2 54.3 Known M F M M F F F M F M M F M F M F F F M M F F M F M F F Alu/DYZ1 M M M M F F F M F M M F M M M F F M M M M M F M F Sexing Contact DNAs with Alu/DYZ1 Assay Sex determination for the 33 contact DNAs (ST 56–89), with amelogenin typing results and the known sex for each donor, are displayed in Table 11. Twenty-six (82%) contact DNAs, 16 male and 10 female, matched the known sex when analyzed using the Alu/DYZ1 assay. Eighteen (69%) of the correctly sexed samples lacked sexing results from screening of the amelogenin locus and five had incorrect female amelogenin results. Having sufficient volume for PCR, sexing results were not obtained from one sample (ST-68) because it failed to amplify. Incorrect male results were generated from five contact DNAs (ST-62–65 and 67) and two (ST-73 and 78) incorrectly sexed as female. When previously sexed with amelogenin assay, amplification did not occur for several samples, while five (ST-79, 80, 84, 86, and 87) yielded incorrect female results. The incorrect sexing results with amelogenin were not repeatable in the pyrosequencing sexing assay. 36 Table 11. Sexing Results of Contact DNAs Contact DNAs previously quantified (in pg/µL) with known sex and amelogenin (Amel) (Thomasma, unpublished) and Alu/DYZ1 assay sexing results listed as male (M), female (F), or no sequence obtained (-). Samples that did not amplify are denoted by (‡). No X allele was detected in the sample marked with (*). Sexing results inconsistent with known are bolded. ID ST-60 ST-67 ST-68‡ ST-73 ST-74 ST-81 ST-69 ST-70 ST-71 ST-76 ST-88 ST-59 ST-85 ST-78 ST-56 ST-58 ST-79 ST-84 ST-82 ST-77 ST-61 ST-63 ST-87 ST-86 ST-65 ST-80 ST-75 ST-72 ST-83 ST-89 ST-57 ST-64 ST-62 ST-66 Quantity (pg/µL) 0 0 0 0 0 2.76 4.28 4.35 5.41 5.76 5.83 6.03 6.92 7.17 7.5 7.66 8.92 9.03 10.7 14.3 16.5 17.9 18.9 20.3 20.4 25.3 25.7 28.2 29.8 33.2 41.2 51 101 101 Known M F F M M F F F F M F M M M F F M M F M M F M M F M M M F M F F F M 37 Amel F M* F F F F F F M F F F - Alu/DYZ1 M M F M F F F F M F M M F F F M M F M M M M M M M M M F M F M M M Sexing Hair Shafts with Alu/DYZ1 Assay Sex determination for the 72 hair shaft DNAs and the known sex for each donor are displayed in Table 12, as are sexing results from real-time assays for amelogenin and Alu/DYZ1. All but two hairs (10 and 16) were correctly sexed using the real-time Alu/DYZ1 assay, while several exhibited inconclusive sexing results with the real-time amelogenin assay and a few typed incorrectly. When analyzed using the pyrosequencing Alu/DYZ1 assay, 36 (50%) hair shaft DNAs extracted from the 30 hairs, 15 male and 21 female, matched the known sex, three of which (1a, 15a, and 19c) had insufficient volumes for PCR. Two hair shaft extracts (18a and 19b) that failed to amplify or produce sequences did not have sufficient volume for PCR. Twenty-nine (40%) hair shaft DNAs completely lacked pyrosequencing sex results. Two of these extracts (18a and 19b) had insufficient volume for PCR. Eight hairs (5, 6, 7, 21, 22, 23, 28, and 30) lacked sequence data from any of the extracts. Six extracts from six different hair donors were incorrectly sexed: three falsely sexed as male and three as female. Although incorrect results were obtained, each hair had at least one extract that was correctly sexed. 38 Table 12. Sexing Results of Hair Shaft DNAs Hair shaft DNAs previously quantified (in pg/µL) with known sex and sexing results from realtime amelogenin (Amel) and Alu/DYZ1 real-time (RT) assays (Kiley, 2009) and Alu/DYZ1 pyrosequencing assay (Alu/DYZ1) listed as male (M), female (F), inconclusive (I), or no sequence was obtained (-). Hairs were extracted by standard organic (O) and alkaline (A) digestion. (*) represent samples that had less than 2 µL extract for PCR. Samples that did not amplify are denoted by (‡). Hairs with the same sample ID number are from the same donor. Sexing results inconsistent with known are bolded. ID 1a* 1b 1c 2a 2b 2c 2d 3a 3b 4a 4b 5a‡ 5b‡ 6a‡ 6b‡ 7a‡ 7b‡ 8a‡ 8b‡ 9a 9b 10a 10b 11a‡ 11b‡ 12a 12b 13a 13b 14a 14b RT Extraction Known Amel Alu/DYZ1 A F F F O A A F F F A O O A F I F O A M M M O A F I F O A F F F O A F F F O A F M F O A M F M O A F F M O A F F F O A M I M O A F I F O A M I M O 39 Alu/DYZ1 F F F F F F F M M M M M F F M M F F M M Table 12 (cont'd) 15a* A 15b A 15c O 16a A 16b A 16c O 17a A 17b A 17c O 18a*‡ A 18b A 18c‡ O 19a‡ A 19b*‡ A 19c* A 19d‡ O 20a A 20b O 21a‡ A 21b‡ O 22a‡ A 22b‡ O 23a‡ A 23b‡ O 24a A 24b O 25a A 25b O 26a A 26b O 27a‡ A 27b A 27c‡ O 27d O 28a‡ A 28b O 29a‡ A 29b O 30a‡ A 30b O M I M F I M M I M F F F F I F F F F F F F F M F F F F M M M F F F M F M F F F F F F F I F F F F 40 M M M F M F M M F F F F F M F M F M F F F F - Sexing Skeletal Remains with Alu/DYZ1 Assay Sex determination for the 55 bone extracts from 17 burials and the estimated anthropological sex for each (Shunn, 2005) are given in Table 13. Sexing results were obtained from 33 bone extracts (60%). Twenty-one (38%) bone extracts, 11 male and 10 female, matched the known sex when analyzed using the Alu/DYZ1 assay. Nine adults were sexed consistently with the anthropological sex, including "likely" sexes. Twenty extracts (36%) lacked sequencing results, 10 of which did not amplify, resulting in four complete burials (9, 11, 13, and 17) with no sexing results. Six bone extracts, only one entire burial (6), sexed inconsistently with the anthropological sex. The remaining bones, except 9, 11, 13, and 17, were sexed, but results could not be confirmed due to unknown sex of adult or subadult remains. 41 Table 13. Sexing Results for Ancient Skeletal Remains Bone DNAs from skeletal remains interred in Fort Michilimackinac, Michigan between 1743 and 1781 with anthropological sex (known) and sexing result from Alu/DYZ1 assay listed as male (M), female (F), unknown (U), or no sequence was obtained (-). Skeletal remains with the same sample ID number are from the same burial. Age is give by adult (A) or subadult (S). (*) represent samples that had less than 2 µL extract for PCR. Samples that did not amplify are denoted by (‡). Samples with the same ID number are from the same burial. Sexing results inconsistent with known are bolded. ID 1a* 1b‡ 1c 1d‡ 2a* 2b 2c 2d 3a 3b* 3c 3d 3e 4a 4b 4c 4d 5a* 5b* 5c 6a 6b‡ 7a* 7b 7c 7d 8a 8b*‡ 8c*‡ 8d 8e‡ Age A Known M A F A Likely F A Likely M A M A M A Likely F S Unknown Alu/DYZ1 F M F F M M F F M M M M M M F F F F F F - ID 9*‡ 10a* 10b 11a 12a* 12b 12c 12d 12e 13a‡ 14a* 14b‡ 14c 14d 15a 15b* 15c 15d‡ 15e 16a* 16b 16c 16d 17a* Age A A Known M M S A Unknown Likely F A A M Unknown A Likely M A Unknown S Unknown 42 Alu/DYZ1 F M F F F F M M M M M M M M - In total, 59.5% of all tested extracts were correctly sexed using the Alu/DYZ1 assay. If samples that yielded no sequence were excluded from the data set, 80.5% of all samples were correctly sexed. The touch DNAs had the most correctly sexed samples (82% of all samples or 90% of samples that gave sequence) and the ancient skeletal remains had the least number of correctly sexed samples (38% or 62%). In general, contact DNA extracts were the most robust as no samples had completely evaporated. Eight touch DNAs (15%), 5 hair shafts (7%), and 15 bone extracts (27%) partially or completely evaporated. 43 DISCUSSION Determining the sex of human body fluids, tissues, or remains is essential for developing a physical profile of a suspect or victim. Visual sexing is effective for high quality skeletal remains, however analysis of incomplete and subadult skeletons is unreliable, and not feasible at the tissue or body fluid level. Sequence comparisons of the X and Y chromosomes are made for molecular sex determination when visual estimates are inadequate, and assays targeting the single copy locus amelogenin have become the gold standard in forensics. Sexing low quality or quantity DNA samples, such as those originating from aged bone or hair shafts, may be unsuccessful using the standard amelogenin assay. In this study, a more sensitive pyrosequencing sexing assay of multicopy loci Alu and DYZ1 was developed to overcome limitations of the standard single copy locus amelogenin analysis. The development of novel techniques often requires substantial troubleshooting, which was certainly the case in the current study. Troubleshooting primarily centered on PCR and pyrosequencing sample preparation. The extreme sensitivity of the assay meant that any minute amount of human DNA was detrimental to PCR of the multicopy loci, made evident by the contamination of all commercially produced Alu primers tested. Precautions to ensure no contamination with human DNA must be strictly followed for these types of assays. Unfortunately, quality control in oligo-generating companies apparently fails to meet these standards. Contact with IDT proved fruitless, as the only explanation given for contamination was a "synthesis error" (Ryan Wilson, IDT, personal communication). The company synthesized new primer batches under reportedly better conditions, but no change in quality resulted. No attempt was made to contact Sigma regarding the quality of their primers, but it 44 would be prudent to do so in the future so they can implement necessary precautions to avoid contamination. Removal of human DNA from primers was attempted using two methods: primer filtration and UV-irradiation of master mixes. Filtering of primers prior to addition to PCR master mixes eliminated most amplification in negative controls, as the filters were designed to retain DNA larger than 125 bp (double-stranded) or 300 bp (single-stranded) (Millipore, 2005). However, the Alu and DYZ1 amplicons were 115 and 60 bp respectively, thus DNA fragments of this size could flow through with the primers, explaining the amplification in negative controls that occurred even after filtering. Based on previous findings by Kiley (2009) and protocols used by Armed Forces DNA Identification Laboratory, UV irradiation can successfully be used to 2 eliminate primer contamination. UV irradiation of master mixes for three minutes (1.5 J/cm ) following primer filtration negated negative control amplification. Periods longer than three minutes apparently damaged the primers and prevented amplification. Note that the three minute irradiation period may not be applicable to other assays, and will be influenced by the intensity 2 of the UV source used. For example, 30 seconds of irradiation (0.25 J/cm ) was the standard for the real-time Alu/DYZ1 assay generated by Kiley (2009), which employed the same Alu primers as in this study. The same UV source was also used in both studies, thus it is possible UV emission decreased over time. On the other hand, large discrepancies in oligo synthesis and contamination control within and among companies could account for the decreased irradiation effectiveness since 2009. Given unknown quality control procedures of other reagent vendors, all supplies were extensively irradiated to destroy any DNA. Following the implementation of these precautions, no DNA amplification occurred in negative controls while Alu and DYZ1 amplified properly in all positive controls. 45 Once optimal PCR conditions were established, pyrosequencing parameters were refined, including PCR product cleanup and dNTP dispensation. Cyclic dispensation orders that were not specific to the target sequence or sequencing primers were first tested. These gave erroneous results (not shown) with incorporation of every nucleotide dispensed. Replacing cyclic with directed dispensation, according to the guidelines outlined in Materials and Methods, gave clean sequences for positive control DNA. Dispensing the same nucleotide multiple times at stretches of identical bases was recommended by the company to eliminate minus shift, a phenomenon where a small portion of the template fails to incorporate a nucleotide and cannot continue with the rest of the sequence until the missing base is dispensed again. Additional bases, however, were not needed for the Alu/DYZ1 target sequence, as seen in Figure 6. For instance, the seventh and eighth bases were two thymines (T), but no incorporation occurred with the second T dispensation, thus minus shift was not occurring. The most beneficial effect from changing dispensation order occurred in the pyrogram pattern that reflected a combined Alu/DYZ1 sequence (Table 7). This made sex determination straightforward: results consisted of Alu sequence for both sexes, with "blank" regions in female pyrograms where DYZ1 sequence would occur (Figs. 7 and 8). After many trials, the optimal volume of biotinylated PCR product in pyrosequencing preparation was determined. The initial cleanup was based on protocols provided by Pyromark installation personnel, which included placing 20 µL of PCR product into the immobilization master mix, while subsequent tests followed recommendations in the Pyromark user manual (5– 20 µL PCR product). A volume of 5 µL for positive control and 10 µL of sample PCR product produced the most reliable and successful pyrograms. Addition of too much template (20 µL) resulted in pyrograms with wide peaks and peaks present (nucleotide incorporation) between 46 dispensations. According to manufacturer troubleshooting guides, however, similar results could have been caused by cross-talk, when light from one well appears in neighboring wells, or incorrect annealing buffer pH. Pyrosequencing of touch and contact DNA using less than 10 µL of template DNA resulted in low or missing peaks in the pyrogram, similar to a negative result. Ten microliters or more eliminated wide and absent peaks without the need to adjust pH or photon detector alignment. Appropriate use of the vacuum preparation station is crucial for generating a singlestranded substrate required for pyrosequencing, and was optimized with some effort due to conflicting advice between Pyromark personnel and user manuals. Equipment installers recommended filling all wash containers to the same level before beginning vacuum preparation (approximately 50 mL). In contrast, the user manual lists volumes to which washing vessels should be filled: all at 50 mL, save denaturation solution (40 mL). After multiple unsuccessful pyrosequencing runs, all solutions were carefully measured to the appropriate volumes listed in the user manual. Successful sequencing runs resulted, likely due to balanced pH in the reaction. According to Amy Smith (Qiagen technical support assistant, personal communication), as long as the volume of washing buffer is greater than NaOH, residual NaOH can be rinsed from the probes to avoid adulterating PCR product or sequencing reagents, especially the sequencing primers. Residual NaOH in the reaction alters the pH enough to cause wide peaks or, in the most extreme cases, complete pyrosequencing failure. Additional discrepancies between Pyromark personnel and user manuals in proper vacuum use pertained to time periods that samples were exposed to each washing solution. In an ideal situation with new probes and a perfectly functioning vacuum, five seconds is appropriate, as recommended in the user manual. However, as probes are used and become less efficient or if 47 solution volumes are higher than 40 or 50 mL, longer aspiration time may be necessary, as instructed by Qiagen technical support personnel. Based on the quality of pyrograms obtained from each approach, visual monitoring until the vacuum aspirates the entire volume of liquid without drying the probe is more essential for proper clean-up than a timed five second aspiration. Both vacuum preparation adjustments discussed above were undertaken at the same time as modified dispensation orders, so ultimate pyrosequencing success could not be attributed to a single modification. The only artifact common to all control pyrograms was an apparent lack of dATPαS incorporation at the beginning of the Alu target, where four adenine residues should exist. Instead, only one or two dATPαS(s) were incorporated. Assuming correct primer and target sequence, minus shift should then have occurred at the beginning of every pyrogram. Instead, sequencing proceeded as predicted and sex determination was possible. Moreover, incorrect dATPαS incorporation only occurred at the beginning of the sequence and the downstream multi-A region was correctly pyrosequenced, verifying that the modified ATP did not interfere with incorporation. This could be explained by faulty synthesis of the forward Alu sequencing primer (Table 1), with three adenines at the 3' end, which would overlap a few bases with the target sequence to result in the observed pyrograms (Fig. 10). Alternatively, if Alu primers hybridized to another region promoting amplification of the same sized target, resulting sequences would have been much different than those found. Not all DNA extracts analyzed with the Alu/DYZ1 assay were in ideal condition, as many had partially or completely evaporated and most had not been previously quantified, or lacked reliable quantification results. Evaporation likely resulted from long-term storage of the extracts, ranging from a few months, none of which had evaporated, to many years. Hair shafts 48 and skeletal remains had been extracted, analyzed, and stored over 7 years earlier (Shunn, 2005; Graffy and Foran, 2005). Collected more recently, contact (6 months earlier) and touch DNAs (2 years earlier) (Thomasma, unpublished; Thomasma and Foran, in press), showed no signs of evaporation. However, even with little or no remaining extract, rinsing tubes with water seemed to elute enough residual DNA to generate successful amplification and sequencing. Additionally, neither hair shaft nor ancient skeletal DNAs had been quantified, as they would likely have little or no detectable nuclear DNA, and quantification would result in numerous samples at 0 pg/µL. All the touch and contact DNA quantified below levels desired for MiniFiler and Identifiler Kits, which recommend at least 0.5 ng. Given this, the high number of correct sexing results is a noteworthy improvement considering the extremely low quantities of nuclear DNA available. Correct sexing results were obtained from DNAs that quantified at 0.00 pg/µL, which refutes the idea that low quantity DNAs are necessarily unsuitable for genetic analyses. Consequently, DNA quantities determined prior to this study were often not predictive of sexing success. For example, six DNAs quantified at 0 pg/µL and three had correct sexing results using the Alu/DYZ1 pyrosequencing assay (Tables 10 and 11); of the remaining samples, two (ST-1 and 68) yielded no results, and the last (ST-67) sexed incorrectly. Due to stochastic sampling, it is possible that inaccurate quantification results were obtained using Quantifiler, which compares unknowns to a set of standard DNAs, the lowest of which is 23 pg/µL. More sensitive quantification methods that assay multicopy loci are now commercially available, such as the Investigator Quantiplex Kit that is reportedly sensitive down to 1 pg/µL DNA or less (Qiagen, 2011), and may have more precisely quantified the touch and contact DNAs and resulted in fewer incorrect 0 pg/µL results. 49 STRs were previously amplified from contact DNAs using an Identifiler Kit, twelve of which had sexing results based on amelogenin (Thomasma, unpublished; Table 11). While seven correctly sexed, the other five (ST-79, 80, 84, 86, and 87), all of which had low amounts of DNA (<26 pg/ µL) incorrectly typed as female. However, differences in sexing results in contact samples containing higher quantities of DNA quantities also produced erroneous sexing result using amelogenin. For example, ST-62 and 66 each quantified at 101 pg/µL, about 50 pg less than the amount of DNA recommended for STR typing with Identifiler, but only the latter correctly sexed, while the former generated no sexing results. The limitation of amelogenin sexing, however, did not interfere with Alu/DYZ1 pyrosequencing assay success. Notably, the samples that incorrectly sexed based on amelogenin, all as female, each correctly sexed as male using the Alu/DYZ1 pyrosequencing assay. Conversely, of the seven contact DNAs that were correctly sexed with amelogenin, the Alu/DYZ1 pyrosequencing assay gave four false male results (ST-62–64, and 67), which may reflect the presence of minute amounts of male DNA in the female samples. The initial collection of hair shafts tested in this research was for a comparison study between alkaline and organic DNA extraction methods of hairs (Graffy and Foran, 2005). The extracts were used in a more recent sexing study based on real-time PCR of the amelogenin locus and an Alu/DYZ1 multiplex (Kiley, 2009). While Kiley designated hair shaft sex based on the best result from alkaline or organic extracts, every extract from each hair was analyzed separately with the pyrosequencing assay, summing to 72 extracts from 30 hair shaft donors. Eighty-four percent of the extracts that yielded pyrosequencing results were correctly sexed, equating to at least one extract for every hair. There was improvement in sexing with the pyrosequencing assay over both real-time analysis of amelogenin and Alu/DYZ1, as several 50 were only correctly sexed with the new assay (Table 12). On the other hand, analyzing the same 30 hair shaft extracts with the pyrosequencing Alu/DYZ1 assay resulted in less overall amplification success than either method studied by Kiley (2009). It is unclear why samples that had previously sexed correctly did not even amplify in the current study, but degradation of DNA while stored may account for the decreased success. There was no indication that one hair DNA extraction technique stored or sequenced better than the other, except that alkaline extracts were the only ones that had evaporated (Table 12), likely due to their storage in Microcon collection tubes. It is somewhat surprising that the pyrosequencing assay was less effective than the real-time assay, especially when considering that the DYZ1 amplicon was smaller than the one targeted in Kiley's real-time assay, although the severity of degradation is unknown in these extracts. It would be beneficial to retest inconsistent pyrosequencing samples, including those that did not amplify, with the real-time Alu/DYZ1 assay to see if the earlier results could be repeated. Sexing discrepancies among real-time amelogenin, real-time Alu/DYZ1, and pyrosequencing Alu/DYZ1 techniques may have had different causes. For example, in her comparison of alkaline and organic extraction techniques, Graffy (2005) amplified and sequenced a short region of mitochondrial DNA, but had difficulty analyzing hair 10 as it consistently contained mixed sequence, suggesting contamination. Further, although Kiley (2009) correctly determined hair 10 to be female, her real-time Alu/DYZ1 sexing tests produced signals for both Alu and DYZ1, with Ct values 15 cycles apart, when a typical male samples would have Alu and DYZ1 Ct values within a cycle or two of each other. This indicates the presence of very low quantities of male DNA relative to Alu, thus it is possible the pyrosequencing assay accurately sexed samples based on the DNA present, including low levels 51 of contamination in hair 10 and other extracts (for example, hairs 3 and 24), despite being inconsistent with the known source. Sexing of hair shaft and bone DNAs seemed to produce higher instances of baseline drift (Figs. 11 and 12). Comparing pyrograms collected over the course of several months, drift was more prevalent when temperatures were highest, despite machine acclimation to ambient temperatures. Additionally, hair and bone extracts likely had much less DNA than touch or contact DNAs, resulting in less PCR product. The main cause for increased baseline drift in these samples is unclear, as both low DNA quantities and severe temperature fluctuations may have generated the artifact. Retesting hair and bone extracts in more mild temperature conditions could exclude it as the cause of drift. Pyrosequencing sex results of bone extracts were compared to anthropological sex estimates of adult burials because sex was not known for these ancient remains (Shunn, 2005). More than one bone was collected from each burial and Shunn (2005) extracted DNA from several portions of each bone. In that study, however, ancestry determination was based on the results of one extract from each bone. All of the stored DNAs, regardless of burial, were analyzed with the pyrosequencing Alu/DYZ1 assay, resulting in 55 tested extracts, which generated some conflicting data within a burial. First, many extracts failed to amplify and/or sequence (Table 13). Second, male and female results were generated from the same burial. For example, the anthropologically estimated sex of burial 1 was male, but using the Alu/DYZ1 pyrosequencing assay led to one male and one female result. If one extract from each burial was randomly selected and used to test molecular sexing consistency with the anthropological sex, accuracy of this assay would change and could show improvement. In total, nine burials were sexed consistently with visual estimates and four lacked pyrosequencing results, an improvement 52 over the ten burials that could not be visually estimated or were considered "likely" male or female (Table 13). The ability to sex subadults also corroborates the greater effectiveness of the pyrosequencing assay over visual sex estimates. 53 CONCLUSION Based on the results presented here, it is clear that the Alu/DYZ1 pyrosequencing assay is effective for determining sex of degraded and low quantities of DNA, as it correctly sexed over 80% of all tested DNAs that yielded sequencing results. The Alu/DYZ1 assay was more sensitive than traditional and real-time based PCR assays of amelogenin, and showed similar effectiveness as a real-time Alu/DYZ1 assay. Further testing comparing the two Alu/DYZ1 assays could resolve conflicting sexing results, and improve the sensitivity of the pyrosequencing assay. This assay is so sensitive that minute amounts of male DNA will be detected if present in female samples, hence taking the aforementioned precautions to avoid all sources of contamination is imperative to generating accurate sexing results. When anthropological means cannot be used and standard molecular sexing approaches are ineffective, this assay can reliably provide sexing results that are necessary to develop a physical profile of ancient remains or suspects and victims of a crime. 54 WORKS CITED 55 WORKS CITED Andréasson H, Allen M. Rapid quantification and sex determination of forensic evidence materials. Journal of Forensic Sciences 2003; 48(6):1280-1287. Andréasson H, Nilsson M, Budowle B, Frisk S, Allen M. Quantification of mtDNA mixtures in forensic evidence material using pyrosequencing. International Journal of Legal Medicine 2006; 120:383-390. Applied Biosystems. 2007. AmpFlSTR® Identifiler ® PCR Amplification Kit. User Guide. Foster City, CA. Batlitzki-Korte, B, Anslinger K, Bartsch C, Rolf B. Species identification by means of pyrosequencing the mitochondrial 12S rRNA gene. International Journal of Legal Medicine 2005; 119:291-294. Budowle B, van Daal A. Extracting evidence from forensic DNA analyses: future molecular biology directions. BioTechniques 2009; 46:339-350. Caragine T, Mikulasovich R, Tamariz J, Bajda E, Sebestyen J, Baum H, Prinz M. Validation of testing and interpretation protocols for low template DNA samples using AmpFlSTR® Identifiler®. Croatian Medical Journal 2009; 50:250-267. Chung DT, Drábek J, Opel KL, Butler JM, McCord BR. A study on the effects of degradation and template concentration on the amplification efficiency of the STR miniplex primer sets. Journal of Forensic Sciences 2004; 49(4):733-740. Diegoli TM, Coble MD. Development and characterization of two mini-X chromosomal short tandem repeat multiplexes. Forensic Science International: Genetics 2011; 5:415-421. Divne A-M, Edlund H, Allen M. Forensic analysis of autosomal STR markers using Pyrosequencing. Forensic Science International: Genetics 2010; 4:122-129. DNA recommendations–1992 report concerning recommendations of the DNA Commission of the International Society for Forensic Haemogenetics relating to the use of PCR-based polymorphism [technical note]. International Journal of Legal Medicine 1992; 105:63-64. Dunker J, Larsson U, Petersson D, Forsell J, Schiller A-L, Alderborn A, Berg LM. Parallel DNA template preparation using a vacuum filtration sample transfer device. BioTechniques 2003; 34:862-868. Ellegren H. Microsatellites: simple sequences with complex evolution. Nature Reviews Genetics 2004; 5:435-445. 56 Fakhrai-Rad H, Pourmand N, Ronaghi M. Pyrosequencing: an accurate detection platform for single nucleotide polymorphisms. Human Mutation 2002; 19:479-485. Gharizadeh B, Ghaderi M, Donnelly D, Amini B, Wallin K-L, Nyrén P. Multiple-primer DNA sequencing method, Electrophoresis 2003; 24:1145-1151. Giacalone J, Friedes J, Francke U. A novel GC-rich human macrosatellite VNTR in Xq24 is differentially methylated on active and inactive X chromosomes. Nature Genetics 1992; 1:137-143. Gill P, Whitaker J, Flaxman C, Brown N, Buckleton J. An investigation of the rigor of interpretation rules for STRs derived from less than 100 pg of DNA. Forensic Science International 2000; 112:17-40. Graffy EA, Foran DR. A simplified method for mitochondrial DNA extraction from head hair shafts. Journal of Forensic Sciences 2005; 50(5):1-4. Harrison C, Musgrave-Brown E, Bender K, Carracedo A, Morling N, Schneider P, Syndercombe-Court D. A sensitive issue: pyrosequencing as a valuable forensic SNP typing platform. International Congress Series 2006; 1288:52-54. Honda K, Harihara S, Nakamura T, Hirai M, Misawa S. Sex identification of analysis of DNA extracted from hard tissues. Japanese Journal of Legal Medicine 1990; 44(4):293-301. Hummel S. Ancient DNA typing: methods, strategies, and applications. New York: SpringerVerlag, 2003;81-109. Hummel S, Herrmann B. Y-chromosomal DNA from ancient bones. In: Herrmann B, Hummell S editors. Ancient DNA. New York: Springer-Verlag, 1994;205-210. Jackson CB. 2006. A more sensitive sex determination assay. Thesis for the Degree of M.S. School of Criminal Justice. East Lansing (MI): Michigan State University. Jurka J, Walichiewicz J, Milosavljevic A. Prototypic sequences for human repetitive DNA. Journal of Molecular Evolution 1992; 35:286-291. Kiley BM. 2009. A highly sensitive sex determination assay for low quality DNA. Thesis for the Degree of M.S. School of Criminal Justice. East Lansing (MI): Michigan State University. King CR, Scott-Horton T. Pyrosequencing: a simple method for accurate genotyping. In: Marsh S, editor. Pyrosequencing protocols. Totowa, NJ: Humana, 2007;39-55. Komar DA, Potter WE. Percentage of body recovered and its effect on identification rates and cause and manner of death determination. Journal of Forensic Sciences 2007; 52(3):528531. 57 Leclair B, Sgueglia JB, Wojtowicz PC, Juston AC, Frégeau CJ, Fourney RM. STR DNA typing: increased sensitivity and efficient sample consumption using reduced PCR reaction volumes. Journal of Forensic Sciences 2003; 48(5):1001-1013. Luptáková L, Bábelovà A, Omelka R, Kolena B, Vondràkovà M, Bauerovà M. Sex determination of early medieval individuals through nested PCR using a new primer set in the SRY gene. Forensic Science International 2011; 207:1-5. Maciejewska A, Pawłowski R. A rare mutation in the primer binding region of the amelogenin X homologue gene. Forensic Science International: Genetics 2009; 3:265-267. Mannucci A, Sullivan KM, Ivanov PL, Gill P. Forensic application of a rapid and quantitative DNA sex test by amplification of the X-Y homologous gene amelogenin. International Journal of Legal Medicine 1994; 106:190-193. Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bemben LA, et al. Genome sequencing in open microfabricated high density picoliter reactors. Nature 2005; 437(7057):376-380. Meier A, Persing DH, Finken M, Böttger EC. Elimination of contaminating DNA within polymerase chain reaction reagents: implications for a general approach to detection of uncultured pathogens. Journal of Clinical Microbiology 1993; 31(3):646-652. Melamede RJ. 1985. Automatable process for sequencing nucleotide. US Patent 4863849. Mighell AJ, Markham AF, Robinson PA. Alu sequences. Federation of European Biochemical Societies 1997; 417:1-5. Millipore. 2005. Microcon® Centrifugal Filter Devices. User Guide. Billerica, MA. Müller U, Lalande M, Donlon T, Latt SA. Moderately repeated DNA sequences specific for the short arm of the human Y chromosome are present in XX males and reduced in copy number in an XY female. Nucleic Acids Research 1986; 14(3):1325-1340. Nakahori Y, Takenaka O, Nakagome Y. A human X-Y homologous region encodes "amelogenin". Genomics 1991; 9:264-269. Nicklas JA, Buel E. Simultaneous determination of total human and male DNA using a duplex real-time PCR assay. Journal of Forensic Sciences 2006; 51(5):1005-1015. Nordström T, Nourizad K, Ronaghi M, Nyrén P. Method enabling pyrosequencing on doublestranded DNA. Analytical Biochemistry 2000; 282: 186-193. Nordström T, Gharizadeh B, Pourmand N, Nyrén P, Ronaghi M. Method enabling fast partial sequencing of cDNA clones. Analytical Biochemistry 2001; 292:266-271. 58 Nyrén P. The history of pyrosequencing. In: Marsh S, editor. Pyrosequencing protocols. Totowa, NJ: Humana, 2007;1-13. Palmirotta R, Verginelli F, Di Tota G, Battista P, Cama A, Caramiello S, Capasso L, MarianiCostantini R. Use of a multiplex polymerase chain reaction assay in the sex typing of DNA extracted from archaeological bone. International Journal of Osteoarchaeology 1997; 7:605-609. Pourmand N, Elahi E, Davis RW, Ronaghi M. Multiplex pyrosequencing. Nucleic Acids Research 2002; 30(7):1-5. Qiagen. 2008. Pyromark Q24 user manual v 1.01. Valencia, CA. Qiagen. 2011. Investigator ® Quantiplex Handbook. Valencia, CA. Rahman MM, Bashamboo A, Prasad A, Pathak D, Ali S. Organisational variation of DYZ1 repeat sequences on the human Y chromosome and its diagnostic potentials. DNA and Cell Biology 2004; 23(9):561-571. Roewer L, Arnemann J, Spurr NK, Grzeschik KH, Epplen JT. Simple repeat sequences on the human Y chromosome are equally polymorphic as their autosomal counterparts. Human Genetics 1992; 89:389-394. Ronaghi M, Karamohamed S, Pettersson B, Uhlén M, Nyrén P. Real-time DNA sequencing using detection of pyrophosphate release. Analytical Biochemistry 1996; 242:84-89. Ronaghi M. Pyrosequencing sheds light on DNA sequencing. Genome Research 2001; 11:3-11. Salido EC, Yen PH, Kaprovnikar K, Yu L, Shapior LJ. The human enamel protein gene amelogenin is expressed from both the X and Y chromosomes. American Journal of Human Genetics 1992; 50:303-316. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Biochemistry 1977; 74(12):5463-5467. Santos RF, Pandya A, Tyler-Smith C. Relability of DNA-based sex tests. Nature Genetics 1998; 18:103. Schmid M, Guttenbach M, Nanda I, Studer R, Epplen JT. Organization of DYZ2 repetitive DNA on the human Y chromosome. Genomics 1990; 6:212-218. Shunn SL. 2005. DNA based ancestry analysis of human skeletal remains from Fort Michilimackinac (1743–1781). Thesis for the Degree of M.S. School of Criminal Justice. East Lansing (MI): Michigan State University. 59 Steinlechner M, Berge B, Niederstätter H, Parson W. Rare failures in the amelogenin sex test. International Journal of Legal Medicine 2002; 116:117-120. Strom CM, Rechitsky S. Use of nested PCR to identify charred human remains and minute amounts of blood. Journal of Forensic Sciences 1998; 43:696-700. Sullivan KM, Mannucci A, Kimpton CP, Gill P. A rapid and quantitative DNA sex test: fluorescence-based PCR analysis of X-Y homologous gene amelogenin. BioTechniques 1993; 15(4):637-641. Thangaraj K, Reddy AG, Singh L. Is the amelogenin gene reliable for gender identification in forensic casework and prenatal diagnosis? International Journal of Legal Medicine 2002; 116:121-123. Thomasma SM. Unpublished. A comparison of trace DNA collection methods for worn and handled clothing. Thesis for the Degree of M.S. School of Criminal Justice. East Lansing (MI): Michigan State University. Thomasma SM, Foran DR. The influence of swabbing solutions on DNA recovery from touch samples. Journal of Forensic sciences. In press. Tschentscher F, Frey UH, Bajanowski T. Amelogenin sex determination by pyrosequencing of short PCR products. International Journal of Legal Medicine 2008; 122:333-335. Tyler-Smith C, Taylor L, Müller U. Structure of a hypervariable tandemly repeated DNA sequence on the short arm of the human Y chromosome. Journal of Molecular Biology 1988; 203:837-848. Whitaker JP, Cotton EA, Gill P. A comparison of the characteristics of profiles produced with AmpFlSTR SGM Plus multiplex system for both standard and low copy number (LCN) STR DNA analysis. Forensic Science International 2001; 123(2-3):215-223. 60