Legal Medicine 18 (2016) 31–36
Contents lists available at ScienceDirect
Legal Medicine journal homepage: www.elsevier.com/locate/legalmed
Short Communication
An alternate method for extracting DNA from environmentally challenged teeth for improved DNA analysis Sheree Hughes-Stamm a,⇑, Frauke Warnke b, Angela van Daal a a b
Faculty of Health Sciences and Medicine, Bond University, Gold Coast, QLD 4223, Australia School of Dentistry and Oral Health, Griffith University, Parklands Dr, Southport, QLD 4215, Australia
a r t i c l e
i n f o
Article history: Received 29 November 2013 Received in revised form 2 November 2015 Accepted 25 November 2015 Available online 26 November 2015 Keywords: Forensic DNA typing Teeth Dentine DNA extraction
a b s t r a c t A grinding-free method to extract DNA from teeth via a direct minimal-invasive retrograde approach to the pulp cavity and dentine was compared to a standard grinding/pulverisation method. This alternate method uses endodontic dental files to access the root canals and pulp cavity for tissue and dentine harvest via the apical end of the roots and avoids mechanical damage to the crown and root morphology. In contrast, other methods require pulverisation of the whole root or tooth, transection or destruction of the occlusal surface to gain access to the DNA in the root canals and pulp chamber. This study compared two methods for preparing dentine powder from the roots of environmentally challenged teeth for forensic DNA analysis. We found that although the filing method was more laborious, and produced less dentine powder, the amount of amplifiable DNA per milligram of powder was substantially higher with the filing method compared to grinding the entire root. In addition, the number of short tandem repeat (STR) alleles detected and the peak height ratios of the STR profiles were notably higher. Although several other methods of extracting DNA-rich tissue from the pulp chamber of teeth have previously been reported, the method presented in this study is minimally invasive, thereby allowing the preservation of tooth and crown morphology. Ó 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In cases of fatalities involving fire or advanced decomposition, the human tissues that remain for DNA identification are usually bone and teeth. Teeth are the hardest tissue in the human body [1] and are resistant to adverse conditions such as humidity, heat, high temperature and microbial action [2,3]. Teeth have traditionally been used as a source of DNA when all other tissues are lost or have failed to yield sufficient DNA for identification. Such cases include ancient [4] and war [5] remains, severely burnt remains and Tsunami victims [6]. Dental pulp is encased in extremely tough dental enamel and is therefore a protected source of both mitochondrial and nuclear DNA for genotyping. Dental enamel is acellular and is therefore not a source of DNA, but sufficient quantities of DNA are found in the pulp cavity, crown dentine and root to support DNA extraction and downstream analysis. Although many laboratories pulverise the entire tooth, ⇑ Corresponding author at: Department of Forensic Science, College of Criminal Justice, Sam Houston State University, Huntsville, TX 77381, USA. E-mail addresses: [email protected] (S. Hughes-Stamm), f.warnke@griffith. edu.au (F. Warnke). http://dx.doi.org/10.1016/j.legalmed.2015.11.008 1344-6223/Ó 2015 Elsevier Ireland Ltd. All rights reserved.
the root dentine and root canal pulp chambers [7], and cells of the cementum [8] have been shown to produce the highest yields of DNA. The pulp and dentine form a structural and functional unit with mature odontoblasts lining the root canals with cellular projections extending into the dentine [9]. Dentine is a preferred target for DNA isolation as it is protected beneath a layer of cementum and enamel. Despite teeth providing such a valuable source of DNA from very ancient and/or degraded remains little is known about the relative amounts of DNA within different regions of teeth, or the potential impact of sampling methods on the resulting quality and quantity of DNA extracted from preserved teeth [8]. However, a major drawback is that the current methods to extract DNA from teeth require the partial or entire mechanical destruction of the tooth. Methods to extract material from teeth for DNA isolation include grinding the whole tooth [10], the root [11], sectioning of the tooth vertically [12,13] or horizontally at the enamel-cement (crown-root) junction [9], endodontic access to pulp cavity and dentine via the crown using a drill [14], and non-powdering methods by simply soaking the whole tooth or tooth sections in digestion buffers overnight [15,16]. The concept of using endodontic files for retrieving dentine for DNA analysis
32
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
from ancient teeth was first described by Cobb [17]. This technique was termed the Reverse-Root-Canal method as it accessed the dentine within the pulp chamber via the apical foramen of the roots. Maximal yield of dental DNA is achieved by grinding the entire tooth into a powder for extraction. However this gives rise to two significant disadvantages. Firstly, the tooth is completely destroyed eliminating any further radiographic, morphologic or restorative analysis that may be required for identification. This is an important consideration, as minimal sample damage is often a requirement for the analysis of ancient and museum remains. Secondly, the forceful mechanical destruction of the tooth may compromise the quality of the extracted DNA. A significant increase in PCR inhibition and DNA degradation has been observed compared with extracts from pure dental pulp [12]. The inclusion of sub-optimal dental tissue in the grinding and extraction processes may produce DNA samples that are more degraded, and/or include more PCR inhibitors. This problem may be overcome by sampling only those portions of the tooth that provide ample DNA of sufficient quality for successful genotyping. It was hypothesised that by approaching the root canal via the natural root-end opening using a manual endodontic file, drilling (and therefore heat) would be avoided, and a cleaner, richer and more pristine source of dentine for DNA extraction would be obtained. Our aim in this study was firstly to develop such a minimal invasive retrograde approach with miniature endodontic files to the pulp chambers and dentine, and secondly, to compare its efficacy for DNA extraction with a destructive standard tooth grinding method. 2. Materials and methods 2.1. Samples Teeth (n = 14) and reference buccal swabs were collected by dentists from seven patients undergoing routine dental extractions. Informed consent was obtained as per the Bond University Human Research Ethics Committee approval (RO-743). To minimise variation, two teeth of the same type (eg. two third molars, or two first molars) were recovered from each individual for comparison, and all teeth appeared healthy and devoid of caries and fillings. The teeth from the same individual were then both exposed to one of 7 environmental insults: buried for 12 or 24 months, surface exposure for 12 or 24 months, saltwater for 6 months, freshwater for 2 months, or fire in a cremation oven (6 min at 700 °C).
5 min, air-dried in a hood, and UV irradiated for 30 min prior to use. The file series increase in diameter in order to progressively widen the root canal and generate powdered dentine (Fig. 1C and D). Not all 14 files in the series were used for every tooth, but as required to widen each canal. The files were used only once, and then discarded. The powdered dentine (50–100 mg) was collected on sterile sheets of foil and transferred to 50 mL tubes for DNA extraction. An X-ray (Planmeca intra; 63 kV, 8 mA, 0.125 s) of each tooth was taken before and after each treatment in order to assess the condition of the root and pulp chamber. 2.2.2. Grinding method The roots were completely removed from the crown with a sterile chisel. The root tissue was ground to a fine powder in a 6770 SPEX freezer mill with cycle conditions consisting of a 10 min pre-cool, 2 1 min crush, 2 1 min cool. To monitor contamination, control samples were taken by swabbing the crushing cylinder prior to processing. 2.3. DNA purification Tooth powder (50–500 mg) was digested using a modified total demineralisation method [18] and extracted using the QIAGEN Blood Maxi kit [19,20]. Powder was incubated in a digestion buffer: 10 mL ATL Buffer (QIAGEN), 5 mL 0.5 M EDTA, 150 lL Proteinase K (20 mg/mL) and 200 lL 1 M DTT (half volumes for tooth powder) at 56 °C for 24 h in a shaking incubator. An additional 5 mL 0.5 M EDTA was added and returned to the shaking incubator at 56 °C for a further 24 h. 15 mL AL buffer (QIAGEN) and 150 lL Proteinase K (20 mg/mL) (half volumes for tooth powder) was added and incubated at 70 °C for one hour in a shaking incubator. Samples were centrifuged at 1000g for 5 min and the supernatant was transferred to new tubes and mixed with 15 mL 100% Ethanol and added to QIAGEN Blood maxi spin columns. The columns were centrifuged for 3 min at 2000g, washed with 10 mL AW1 buffer and centrifuged again. The filter was washed with 10 mL AW2 buffer and centrifuged again at 2000g for 3 min. Residual AW2 buffer was removed by further centrifugation at 2000g for 10 min. DNA was eluted by adding 3 mL AE buffer preheated to 72 °C and centrifuged at 2000g for 3 min. The elution process was repeated by running eluate back through the filter for maximal concentration. All samples were concentrated by centrifugation for 3 min at 4000g in Amicon Ultra-4 columns (Millipore) to a final volume of 200–400 lL. Samples were further concentrated using a 30 min centrifugation in a Speedvac (Thermo scientific) at the medium heat setting (100 lL final volume).
2.2. Sample preparation
2.4. DNA quantification and STR-typing
Teeth were soaked in 20% commercial bleach for 5 min then 100% ethanol for 5 min prior to drying at 30 °C overnight. Samples (n = 14) used to compare the efficiency of the two methods consisted of two molars from each individual (n = 7), each being treated in one of the two methods tested.
The quantity and quality of DNA recovered from teeth was assessed using the real-time quantitative (qPCR) quadruplex assay described [21]. All samples were analysed in triplicate for quantification and subsequent STR profiling. Samples were genotyped using the PowerPlexÒ ESI 16 (Promega) amplification kit as per manufacturer instructions. All reference samples were genotyped with 0.5 ng input template. The maximum sample volume (17.5 lL) was added to each reaction for all tooth samples except for sample 1 and sample 4 (filing extract only), where 0.5 ng of template was added. PCR was performed in 25 lL reaction volumes on a GeneAmp 9700 (Thermo Fisher Scientific) as per recommended cycling conditions. Capillary electrophoresis was performed on a 3130 Genetic Analyser (Thermo Fisher Scientific). Samples were prepared for fragment analysis as per PowerPlexÒ ESI 16 PCR amplification kit recommendations. Data analyses were performed using GeneMappper ID v 3.2.1 software (Thermo Fisher Scientific) with a 50 relative
2.2.1. Filing method Initially, the tips of each root (approximately 1–2 mm from apical foramen) were removed with a cutting disc on a Dremel StylusTM drill similar to a standard apidectomy in oral surgery. The apical opening allowed for direct access for endodontic files to the crucial root canal. Teeth were then sprayed with 20% bleach and 100% ethanol, allowed to air-dry or dried at 30 °C. Dentine and pulp were extracted via this root access for each tooth by a qualified dentist. A series of HedstrÖm endodontic files (sizes 08–80) (Fig. 1A and B) were used to scrape the dentine from inside each root and pulp cavity. Each file was soaked in 20% bleach for
33
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
Fig. 1. Instruments and process involved in the filing method to extract dentine for DNA purification. (A) A series of endodontic files were used to widen the root canals and generate powdered dentine. (B) A schematic diagram of the spiralled blade of the HedstrÖm files. (C) Files were repeatedly inserted into the root canal via the apex of the root. (D) An X-ray post-treatment with the file in situ showing the extent of access to the root canal and pulp cavity using this method.
fluorescence units (RFU) peak amplitude threshold for all dyes. The stutter threshold was 15%. The average number of alleles correct (expressed as a percentage) was determined by comparing the alleles present at each locus in the electropherograms from the experimental tooth samples to the true allele in the reference sample. The peak height ratio (PHR) was determined by dividing the height of the smaller peak by the height of the larger peak in each heterozygous pair. The average PHR was then calculated by summing the PHR of each heterozygous locus divided by the total number of heterozygous loci. A one-way ANOVA was performed on the DNA yield and STR data using SPSS v20.
3. Results and discussion Two methods for extracting dentine powder from teeth for DNA purification were investigated. A manual filing method of extracting dentine powder was evaluated as an alternative technique to a
commonly employed method of cutting the roots from the crown and grinding the entire root into a powder in a freezer mill or blender cup. Endodontic files were used to scrape the inside of each root canal to retrieve powdered dentine (targeting protected DNA nucleated cells) for DNA extraction without destroying tooth morphology. X-rays of each tooth before and after the filing process demonstrate the widening of each root canal as a result of the dentine harvest (Suppl. Fig. 1). 3.1. DNA yield As expected, more tooth powder was obtained using the grinding method compared to the filing method (Table 1). However, when the amount of amplifiable DNA from both methods was quantified, the trend was opposite (Table 1). Four of the seven samples (2, 3, 4 and 7) yielded more amplifiable DNA using the filing method compared to the grinding method (average 6-fold (±3-fold) increase). Two samples (1 and 6) yielded less amplifiable
Table 1 Comparative amounts of dentine powder, the total amount of amplifiable DNA retrieved, and the DNA yields using the two methods tested in addition to the amount of DNA into each PCR reaction. Sample 1 = buried for 12 months, 2 = buried for 24 months, 3 = surface exposure for 12 months, 4 = surface exposure for 24 months, 5 = saltwater for 6 months, 6 = freshwater for 2 months, fire exposure (700 °C for 6 min). Sample
1 2 3 4 5 6 7
Amount of dentine powder (mg)
Total amount of amplifiable DNA (ng)
DNA yield (pg DNA/mg dentine powder)
DNA template in PCR (ng)
Filing
Grinding
Filing
Grinding
Filing
Grinding
Filing
Grinding
100 90 70 140 60 50 50
280 320 300 250 280 350 500
17.93 2.61 0.42 5.49 0.1 0.48 0.18
34.94 0.26 0.13 0.89 0.09 0.64 0
179.3 29 6 39.21 1.67 9.6 3.6
124.79 0.81 0.43 3.56 0.32 1.83 0
0.50 0.46 0.07 0.50 0.02 0.08 0.03
0.50 0.05 0.02 0.16 0.02 0.11 0.00
34
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
DNA in total using the filing method than the grinding method and one sample (5) produced comparable amounts. No DNA was detected in sample 7 when using the grinding method. Biological variation between the teeth and/or the small sample size may both contribute to the difference in DNA yield. An interesting comparison is the DNA yield per milligram of dentine powder extracted. This is an indication of efficiency in the recovery of amplifiable DNA with the two different powdering methods. The filing method yielded more amplifiable DNA per
milligram of powder than grinding the entire root in all samples regardless of the environmental exposure. However the increase in DNA yield was variable over the seven samples (ranging from 1.5 to 35.5-fold) (Table 1). Due to the small sample size and wide dispersion of values, statistical significance was not reached (P = 0.054). However, these data may suggest that the filing method targets the DNA-rich region of the root canal and pulp cavity, resulting in a higher proportion of nucleated cells per milligram of powder. Despite the fact that grinding the entire tooth root
Fig. 2. The success of STR profiles from DNA recovered via the two methods tested in this study. Comparative DNA quality from teeth via the two methods tested as measured by (A) average number of alleles recovered and (B) average peak height ratio (PHR) of each STR profile. Sample 1 = buried for 12 months, 2 = buried for 24 months, 3 = surface exposure for 12 months, 4 = surface exposure for 24 months, 5 = saltwater for 6 months, 6 = freshwater for 2 months, fire exposure (700 °C for 6 min). STR analysis was performed in triplicate. Data is presented as average ± Std Dev.
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
generates more powder, and in general more DNA, the filing method is more efficient at retrieving amplifiable DNA. The suboptimal powder generated by crushing the entire root may contain PCR inhibitors that reduce the efficiency of the PCR reaction during quantification. However, no significant delay in cycle threshold (Ct) of the internal PCR control was observed with the mechanically ground samples versus the filed samples (data not shown). This result was contrary to a previous finding that reported significant PCR inhibition after crushing the tooth using mechanical force [12]. Samples from donor 1 yielded substantially more DNA than all other teeth in this study regardless of the method used to extract the dentine (Table 1). These data suggest three possibilities; (1) contamination with exogenous DNA, (2) biological variation, meaning that this individual had a higher amount of DNA-rich tissue in their teeth than the other individuals investigated, and (3) the treatment environment. This tooth was buried for a period of 12 months and therefore may have been protected from the harsh environmental effects such as sunlight, humidity and microbial action that may degrade the DNA, although, teeth are thought to be resistant to these adverse conditions [2,3,22]. Due to the low sample numbers in this preliminary study, no explanation for the DNA yields can be drawn. However, contamination was eliminated as a source of ‘extra’ DNA as no additional alleles were detected during downstream STR typing of this sample.
3.2. STR analysis The quality of DNA recovered from teeth using both methods was assessed via STR analysis. The average number of correct alleles detected (Fig. 2A), and the average peak height ratio (PHR) of each sample profile (Fig. 2B) were the two key indicators measured. The filing method yielded more consistent and overall better results than those seen with the grinding method. Greater variation in both allele recovery and PHR was observed with the samples that were mechanically ground. Overall, the number of alleles amplified using DNA extracted from the filed teeth was significantly (P = 0.03) higher than DNA from teeth that were processed using the grinding method. In five of the seven teeth (1, 3, 4, 5 and 7) the filing method generated more complete STR profiles than the grinding method (Fig. 2A). It should also be noted that in three of those five samples (samples 3, 4 and 7) a complete (or near complete) profile was obtained with the filing method when very poor results were obtained by grinding the entire root (620% of alleles amplified). The STR profiles from the same three teeth also displayed considerable allelic imbalance using the grinding method (0–40% average PHR) (Fig. 2B). The ground samples showed wide variation in the average PHR (Fig. 2B). Three of the seven samples showed average PHR > 70% (±5.2%) and two with 40–50% (±2.9%). A profile PHR of 0% was calculated for one of the remaining two samples when only one allele was detected at heterozygote loci. An average PHR could not be calculated for the remaining sample as no alleles were detected. The filing method generally produced more balanced profiles than the grinding method, but statistical significance was not reached (P = 0.072). Six out of the seven filed samples produced profiles with an average PHR > 70% (±7%). One filed tooth (sample 2) had an average PHR of 25% (±19%). The wide variation seen in both the average PHR and number of alleles recovered from the filed sample of tooth 2 (Fig. 2) reflects the different profiling results from the triplicate STR reactions. Due to the total consumption of the DNA extract from this tooth sample, the PCR reaction could not be repeated. The more complete and balanced STR profiles obtained from the filed teeth may also be explained by the variation in the amount of DNA recovered using the two methods
35
leading to downstream differences in the amount of total DNA template in each PCR reaction (Table 1). The occurrence of drop-in alleles was monitored to assess the risk of contamination of both methods. More contamination was observed with the mechanically ground samples. In a total of 14 profiles (448 alleles) for each method (in duplicate), ten (2.2%) drop-in alleles were seen in ‘ground’ profiles as opposed to three (0.6%) in the ‘filing’ profiles. This result was surprising due to the substantially more direct handling of each tooth during the filing process. Although contamination may be introduced at every step of the process, one likely explanation is that the decontamination washing process of the whole teeth prior to grinding was not adequate to remove all exogenous DNA from the external surface of the roots, and the filing method therefore avoided removing powder from these surfaces. Of the three drop-in alleles seen in the filing method, two were consistent with handler profiles. However, this observation may not hold true if the decontamination process of whole teeth is improved and optimised to ensure all exogenous DNA is removed prior to processing. Results of this preliminary study constitute a proof of concept that the filing method described here is a viable method for extracting dentine powder from teeth exposed to various environmental insults, when tooth morphology must be retained. In this study, greater DNA yields were obtained using the filing method than grinding the entire tooth root, which in turn led to improved STR profiles. These results support a previous study which also showed poorer STR profiles after crushing the entire tooth or root compared with a form of endodontic access to the root canal (via the crown) for removing pulp and dentine [14]. The manual process of removing dentine with endodontic files was laborious and more time consuming than grinding the entire root (2–3 h vs 40 min). The significantly longer time required for extracting dentin powder from teeth may not be a major concern for museum and anthropological specimens, casework or research samples when the demand for DNA extractions from teeth is infrequent. However, when large numbers of tooth samples are required for processing and genotyping (such as in a mass disaster situation or missing persons cases), the extra time required per sample would not make this method viable. Nevertheless, the combined benefits of retaining tooth morphology and more complete and evenly balanced STR profiles may outweigh the disadvantages of the additional time and skill required to perform the filing process. Acknowledgements The authors would like to thank Dr. John Cosson, a maxillofacial surgeon for providing the teeth used in this study. In addition, Mr Tim Connolly, the Director of New Haven Funerals, kindly provided the use of a cremator oven and the plot of land required for the fieldwork. The authors would also like to thank Dr Patrick Warnke for his valuable assistance with the initial editing of this manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.legalmed.2015. 11.008. References [1] P.C. Malaver, J.J. Yunis, Different dental tissues as source of DNA for human identification in forensic cases, Croat. Med. J. 44 (3) (2003) 306–309. [2] A. Alvarez Garcia, I. Munoz, C. Pestoni, M.V. Lareu, M.S. Rodriguez-Calvo, A. Carracedo, Effect of environmental factors on PCR-DNA analysis from dental pulp, Int. J. Legal Med. 109 (3) (1996) 125–129.
36
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
[3] D. Marjanovic, A. Durmic-Pasic, N. Bakal, S. Haveric, B. Kalamujic, L. Kovacevic, et al., DNA identification of skeletal remains from the World War II mass graves uncovered in Slovenia, Croat. Med. J. 48 (4) (2007) 513–519. [4] C. Sosa, M. Baeta, C. Nunez, Y. Casalod, A. Luna, B. Martinez-Jarreta, Nuclear DNA typing from ancient teeth, Am. J. Forensic Med. Pathol. 33 (3) (2012) 211– 214. [5] I. Zupanic Pajnic, B. Gornjak Pogorelc, J. Balazic, T. Zupanc, B. Stefanic, Highly efficient nuclear DNA typing of the World War II skeletal remains using three new autosomal short tandem repeat amplification kits with the extended European Standard Set of loci, Croat. Med. J. 53 (1) (2012) 17–23. [6] Y.J. Deng, Y.Z. Li, X.G. Yu, L. Li, D.Y. Wu, J. Zhou, et al., Preliminary DNA identification for the tsunami victims in Thailand, Genomics Proteomics Bioinf. 3 (3) (2005) 143–157. [7] R. Gaytmenn, D. Sweet, Quantification of forensic DNA from various regions of human teeth, J. Forensic Sci. 48 (3) (2003) 622–625. [8] C.J. Adler, W. Haak, D. Donlon, A. Cooper, T.G. Consortium, Survival and recovery of DNA from ancient teeth and bones, J. Archaeol. Sci. 38 (2011) 956– 964. [9] A. Corte-Real, L. Andrade, M.J. Anjos, M. Carvalho, M.C. Vide, F. Corte-Real, et al., The DNA extraction from the pulp dentine complex of both with and without carious, Int. Congr. Ser. 1288 (2006) 710–712. [10] D. Sweet, D. Hildebrand, Recovery of DNA from human teeth by cryogenic grinding, J. Forensic Sci. 43 (6) (1998) 1199–1202. [11] N. Rohland, M. Hofreiter, Ancient DNA extraction from bones and teeth, Nat. Protoc. 2 (7) (2007) 1756–1762. [12] B.C. Smith, D.L. Fisher, V.W. Weedn, G.R. Warnock, M.M. Holland, A systematic approach to the sampling of dental DNA, J. Forensic Sci. 38 (5) (1993) 1194– 1209.
[13] R. Trivedi, P. Chattopadhyay, V.K. Kashyap, A new improved method for extraction of DNA from teeth for the analysis of hypervariable loci, Am. J. Forensic Med. Pathol. 23 (2) (2002) 191–196. [14] F. Tilotta, P. Brousseau, E. Lepareur, K. Yasukawa, P. de Mazancourt, A comparative study of two methods of dental pulp extraction for genetic fingerprinting, Forensic Sci. Int. 202 (1–3) (2010) e39–e43. [15] D.A. Bolnick, H.M. Bonine, J. Mata-Miguez, B.M. Kemp, M.H. Snow, S.A. LeBlanc, Nondestructive sampling of human skeletal remains yields ancient nuclear and mitochondrial DNA, Am. J. Phys. Anthropol. 147 (2) (2011) 293–300. [16] T. Kitayama, Y. Ogawa, K. Fujii, H. Nakahara, N. Mizuno, K. Sekiguchi, et al., Evaluation of a new experimental kit for the extraction of DNA from bones and teeth using a non-powder method, Legal Med. (Tokyo) 12 (2) (2010) 84–89. [17] J. Cobb, Ancient DNA recovered by a non-destructive method, Ancient Biomolecules 4 (4) (2002) 169–172. [18] M.J. Park, H.Y. Lee, U. Chung, S.C. Kang, K.J. Shin, Y-STR analysis of degraded DNA using reduced-size amplicons, Int. J. Legal Med. 121 (2) (2007) 152–157. [19] J. Davoren, D. Vanek, R. Konjhodzic, J. Crews, E. Huffine, T.J. Parsons, Highly effective DNA extraction method for nuclear short tandem repeat testing of skeletal remains from mass graves, Croat. Med. J. 48 (4) (2007) 478–485. [20] S.B. Seo, A. Zhang, H.Y. Kim, J.A. Yi, H.Y. Lee, D.H. Shin, et al., Technical note: efficiency of total demineralization and ion-exchange column for DNA extraction from bone, Am. J. Phys. Anthropol. 141 (1) (2009) 158–162. [21] S.R. Hughes-Stamm, K.J. Ashton, A. van Daal, Assessment of DNA degradation and the genotyping success of highly degraded samples, Int. J. Legal Med. 125 (3) (2011) 341–348. [22] C.Y. Shiroma, C.G. Fielding, J.A. Lewis Jr., M.R. Gleisner, K.N. Dunn, A minimally destructive technique for sampling dentin powder for mitochondrial DNA testing, J. Forensic Sci. 49 (4) (2004) 791–795.
Contents lists available at ScienceDirect
Legal Medicine journal homepage: www.elsevier.com/locate/legalmed
Short Communication
An alternate method for extracting DNA from environmentally challenged teeth for improved DNA analysis Sheree Hughes-Stamm a,⇑, Frauke Warnke b, Angela van Daal a a b
Faculty of Health Sciences and Medicine, Bond University, Gold Coast, QLD 4223, Australia School of Dentistry and Oral Health, Griffith University, Parklands Dr, Southport, QLD 4215, Australia
a r t i c l e
i n f o
Article history: Received 29 November 2013 Received in revised form 2 November 2015 Accepted 25 November 2015 Available online 26 November 2015 Keywords: Forensic DNA typing Teeth Dentine DNA extraction
a b s t r a c t A grinding-free method to extract DNA from teeth via a direct minimal-invasive retrograde approach to the pulp cavity and dentine was compared to a standard grinding/pulverisation method. This alternate method uses endodontic dental files to access the root canals and pulp cavity for tissue and dentine harvest via the apical end of the roots and avoids mechanical damage to the crown and root morphology. In contrast, other methods require pulverisation of the whole root or tooth, transection or destruction of the occlusal surface to gain access to the DNA in the root canals and pulp chamber. This study compared two methods for preparing dentine powder from the roots of environmentally challenged teeth for forensic DNA analysis. We found that although the filing method was more laborious, and produced less dentine powder, the amount of amplifiable DNA per milligram of powder was substantially higher with the filing method compared to grinding the entire root. In addition, the number of short tandem repeat (STR) alleles detected and the peak height ratios of the STR profiles were notably higher. Although several other methods of extracting DNA-rich tissue from the pulp chamber of teeth have previously been reported, the method presented in this study is minimally invasive, thereby allowing the preservation of tooth and crown morphology. Ó 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In cases of fatalities involving fire or advanced decomposition, the human tissues that remain for DNA identification are usually bone and teeth. Teeth are the hardest tissue in the human body [1] and are resistant to adverse conditions such as humidity, heat, high temperature and microbial action [2,3]. Teeth have traditionally been used as a source of DNA when all other tissues are lost or have failed to yield sufficient DNA for identification. Such cases include ancient [4] and war [5] remains, severely burnt remains and Tsunami victims [6]. Dental pulp is encased in extremely tough dental enamel and is therefore a protected source of both mitochondrial and nuclear DNA for genotyping. Dental enamel is acellular and is therefore not a source of DNA, but sufficient quantities of DNA are found in the pulp cavity, crown dentine and root to support DNA extraction and downstream analysis. Although many laboratories pulverise the entire tooth, ⇑ Corresponding author at: Department of Forensic Science, College of Criminal Justice, Sam Houston State University, Huntsville, TX 77381, USA. E-mail addresses: [email protected] (S. Hughes-Stamm), f.warnke@griffith. edu.au (F. Warnke). http://dx.doi.org/10.1016/j.legalmed.2015.11.008 1344-6223/Ó 2015 Elsevier Ireland Ltd. All rights reserved.
the root dentine and root canal pulp chambers [7], and cells of the cementum [8] have been shown to produce the highest yields of DNA. The pulp and dentine form a structural and functional unit with mature odontoblasts lining the root canals with cellular projections extending into the dentine [9]. Dentine is a preferred target for DNA isolation as it is protected beneath a layer of cementum and enamel. Despite teeth providing such a valuable source of DNA from very ancient and/or degraded remains little is known about the relative amounts of DNA within different regions of teeth, or the potential impact of sampling methods on the resulting quality and quantity of DNA extracted from preserved teeth [8]. However, a major drawback is that the current methods to extract DNA from teeth require the partial or entire mechanical destruction of the tooth. Methods to extract material from teeth for DNA isolation include grinding the whole tooth [10], the root [11], sectioning of the tooth vertically [12,13] or horizontally at the enamel-cement (crown-root) junction [9], endodontic access to pulp cavity and dentine via the crown using a drill [14], and non-powdering methods by simply soaking the whole tooth or tooth sections in digestion buffers overnight [15,16]. The concept of using endodontic files for retrieving dentine for DNA analysis
32
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
from ancient teeth was first described by Cobb [17]. This technique was termed the Reverse-Root-Canal method as it accessed the dentine within the pulp chamber via the apical foramen of the roots. Maximal yield of dental DNA is achieved by grinding the entire tooth into a powder for extraction. However this gives rise to two significant disadvantages. Firstly, the tooth is completely destroyed eliminating any further radiographic, morphologic or restorative analysis that may be required for identification. This is an important consideration, as minimal sample damage is often a requirement for the analysis of ancient and museum remains. Secondly, the forceful mechanical destruction of the tooth may compromise the quality of the extracted DNA. A significant increase in PCR inhibition and DNA degradation has been observed compared with extracts from pure dental pulp [12]. The inclusion of sub-optimal dental tissue in the grinding and extraction processes may produce DNA samples that are more degraded, and/or include more PCR inhibitors. This problem may be overcome by sampling only those portions of the tooth that provide ample DNA of sufficient quality for successful genotyping. It was hypothesised that by approaching the root canal via the natural root-end opening using a manual endodontic file, drilling (and therefore heat) would be avoided, and a cleaner, richer and more pristine source of dentine for DNA extraction would be obtained. Our aim in this study was firstly to develop such a minimal invasive retrograde approach with miniature endodontic files to the pulp chambers and dentine, and secondly, to compare its efficacy for DNA extraction with a destructive standard tooth grinding method. 2. Materials and methods 2.1. Samples Teeth (n = 14) and reference buccal swabs were collected by dentists from seven patients undergoing routine dental extractions. Informed consent was obtained as per the Bond University Human Research Ethics Committee approval (RO-743). To minimise variation, two teeth of the same type (eg. two third molars, or two first molars) were recovered from each individual for comparison, and all teeth appeared healthy and devoid of caries and fillings. The teeth from the same individual were then both exposed to one of 7 environmental insults: buried for 12 or 24 months, surface exposure for 12 or 24 months, saltwater for 6 months, freshwater for 2 months, or fire in a cremation oven (6 min at 700 °C).
5 min, air-dried in a hood, and UV irradiated for 30 min prior to use. The file series increase in diameter in order to progressively widen the root canal and generate powdered dentine (Fig. 1C and D). Not all 14 files in the series were used for every tooth, but as required to widen each canal. The files were used only once, and then discarded. The powdered dentine (50–100 mg) was collected on sterile sheets of foil and transferred to 50 mL tubes for DNA extraction. An X-ray (Planmeca intra; 63 kV, 8 mA, 0.125 s) of each tooth was taken before and after each treatment in order to assess the condition of the root and pulp chamber. 2.2.2. Grinding method The roots were completely removed from the crown with a sterile chisel. The root tissue was ground to a fine powder in a 6770 SPEX freezer mill with cycle conditions consisting of a 10 min pre-cool, 2 1 min crush, 2 1 min cool. To monitor contamination, control samples were taken by swabbing the crushing cylinder prior to processing. 2.3. DNA purification Tooth powder (50–500 mg) was digested using a modified total demineralisation method [18] and extracted using the QIAGEN Blood Maxi kit [19,20]. Powder was incubated in a digestion buffer: 10 mL ATL Buffer (QIAGEN), 5 mL 0.5 M EDTA, 150 lL Proteinase K (20 mg/mL) and 200 lL 1 M DTT (half volumes for tooth powder) at 56 °C for 24 h in a shaking incubator. An additional 5 mL 0.5 M EDTA was added and returned to the shaking incubator at 56 °C for a further 24 h. 15 mL AL buffer (QIAGEN) and 150 lL Proteinase K (20 mg/mL) (half volumes for tooth powder) was added and incubated at 70 °C for one hour in a shaking incubator. Samples were centrifuged at 1000g for 5 min and the supernatant was transferred to new tubes and mixed with 15 mL 100% Ethanol and added to QIAGEN Blood maxi spin columns. The columns were centrifuged for 3 min at 2000g, washed with 10 mL AW1 buffer and centrifuged again. The filter was washed with 10 mL AW2 buffer and centrifuged again at 2000g for 3 min. Residual AW2 buffer was removed by further centrifugation at 2000g for 10 min. DNA was eluted by adding 3 mL AE buffer preheated to 72 °C and centrifuged at 2000g for 3 min. The elution process was repeated by running eluate back through the filter for maximal concentration. All samples were concentrated by centrifugation for 3 min at 4000g in Amicon Ultra-4 columns (Millipore) to a final volume of 200–400 lL. Samples were further concentrated using a 30 min centrifugation in a Speedvac (Thermo scientific) at the medium heat setting (100 lL final volume).
2.2. Sample preparation
2.4. DNA quantification and STR-typing
Teeth were soaked in 20% commercial bleach for 5 min then 100% ethanol for 5 min prior to drying at 30 °C overnight. Samples (n = 14) used to compare the efficiency of the two methods consisted of two molars from each individual (n = 7), each being treated in one of the two methods tested.
The quantity and quality of DNA recovered from teeth was assessed using the real-time quantitative (qPCR) quadruplex assay described [21]. All samples were analysed in triplicate for quantification and subsequent STR profiling. Samples were genotyped using the PowerPlexÒ ESI 16 (Promega) amplification kit as per manufacturer instructions. All reference samples were genotyped with 0.5 ng input template. The maximum sample volume (17.5 lL) was added to each reaction for all tooth samples except for sample 1 and sample 4 (filing extract only), where 0.5 ng of template was added. PCR was performed in 25 lL reaction volumes on a GeneAmp 9700 (Thermo Fisher Scientific) as per recommended cycling conditions. Capillary electrophoresis was performed on a 3130 Genetic Analyser (Thermo Fisher Scientific). Samples were prepared for fragment analysis as per PowerPlexÒ ESI 16 PCR amplification kit recommendations. Data analyses were performed using GeneMappper ID v 3.2.1 software (Thermo Fisher Scientific) with a 50 relative
2.2.1. Filing method Initially, the tips of each root (approximately 1–2 mm from apical foramen) were removed with a cutting disc on a Dremel StylusTM drill similar to a standard apidectomy in oral surgery. The apical opening allowed for direct access for endodontic files to the crucial root canal. Teeth were then sprayed with 20% bleach and 100% ethanol, allowed to air-dry or dried at 30 °C. Dentine and pulp were extracted via this root access for each tooth by a qualified dentist. A series of HedstrÖm endodontic files (sizes 08–80) (Fig. 1A and B) were used to scrape the dentine from inside each root and pulp cavity. Each file was soaked in 20% bleach for
33
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
Fig. 1. Instruments and process involved in the filing method to extract dentine for DNA purification. (A) A series of endodontic files were used to widen the root canals and generate powdered dentine. (B) A schematic diagram of the spiralled blade of the HedstrÖm files. (C) Files were repeatedly inserted into the root canal via the apex of the root. (D) An X-ray post-treatment with the file in situ showing the extent of access to the root canal and pulp cavity using this method.
fluorescence units (RFU) peak amplitude threshold for all dyes. The stutter threshold was 15%. The average number of alleles correct (expressed as a percentage) was determined by comparing the alleles present at each locus in the electropherograms from the experimental tooth samples to the true allele in the reference sample. The peak height ratio (PHR) was determined by dividing the height of the smaller peak by the height of the larger peak in each heterozygous pair. The average PHR was then calculated by summing the PHR of each heterozygous locus divided by the total number of heterozygous loci. A one-way ANOVA was performed on the DNA yield and STR data using SPSS v20.
3. Results and discussion Two methods for extracting dentine powder from teeth for DNA purification were investigated. A manual filing method of extracting dentine powder was evaluated as an alternative technique to a
commonly employed method of cutting the roots from the crown and grinding the entire root into a powder in a freezer mill or blender cup. Endodontic files were used to scrape the inside of each root canal to retrieve powdered dentine (targeting protected DNA nucleated cells) for DNA extraction without destroying tooth morphology. X-rays of each tooth before and after the filing process demonstrate the widening of each root canal as a result of the dentine harvest (Suppl. Fig. 1). 3.1. DNA yield As expected, more tooth powder was obtained using the grinding method compared to the filing method (Table 1). However, when the amount of amplifiable DNA from both methods was quantified, the trend was opposite (Table 1). Four of the seven samples (2, 3, 4 and 7) yielded more amplifiable DNA using the filing method compared to the grinding method (average 6-fold (±3-fold) increase). Two samples (1 and 6) yielded less amplifiable
Table 1 Comparative amounts of dentine powder, the total amount of amplifiable DNA retrieved, and the DNA yields using the two methods tested in addition to the amount of DNA into each PCR reaction. Sample 1 = buried for 12 months, 2 = buried for 24 months, 3 = surface exposure for 12 months, 4 = surface exposure for 24 months, 5 = saltwater for 6 months, 6 = freshwater for 2 months, fire exposure (700 °C for 6 min). Sample
1 2 3 4 5 6 7
Amount of dentine powder (mg)
Total amount of amplifiable DNA (ng)
DNA yield (pg DNA/mg dentine powder)
DNA template in PCR (ng)
Filing
Grinding
Filing
Grinding
Filing
Grinding
Filing
Grinding
100 90 70 140 60 50 50
280 320 300 250 280 350 500
17.93 2.61 0.42 5.49 0.1 0.48 0.18
34.94 0.26 0.13 0.89 0.09 0.64 0
179.3 29 6 39.21 1.67 9.6 3.6
124.79 0.81 0.43 3.56 0.32 1.83 0
0.50 0.46 0.07 0.50 0.02 0.08 0.03
0.50 0.05 0.02 0.16 0.02 0.11 0.00
34
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
DNA in total using the filing method than the grinding method and one sample (5) produced comparable amounts. No DNA was detected in sample 7 when using the grinding method. Biological variation between the teeth and/or the small sample size may both contribute to the difference in DNA yield. An interesting comparison is the DNA yield per milligram of dentine powder extracted. This is an indication of efficiency in the recovery of amplifiable DNA with the two different powdering methods. The filing method yielded more amplifiable DNA per
milligram of powder than grinding the entire root in all samples regardless of the environmental exposure. However the increase in DNA yield was variable over the seven samples (ranging from 1.5 to 35.5-fold) (Table 1). Due to the small sample size and wide dispersion of values, statistical significance was not reached (P = 0.054). However, these data may suggest that the filing method targets the DNA-rich region of the root canal and pulp cavity, resulting in a higher proportion of nucleated cells per milligram of powder. Despite the fact that grinding the entire tooth root
Fig. 2. The success of STR profiles from DNA recovered via the two methods tested in this study. Comparative DNA quality from teeth via the two methods tested as measured by (A) average number of alleles recovered and (B) average peak height ratio (PHR) of each STR profile. Sample 1 = buried for 12 months, 2 = buried for 24 months, 3 = surface exposure for 12 months, 4 = surface exposure for 24 months, 5 = saltwater for 6 months, 6 = freshwater for 2 months, fire exposure (700 °C for 6 min). STR analysis was performed in triplicate. Data is presented as average ± Std Dev.
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
generates more powder, and in general more DNA, the filing method is more efficient at retrieving amplifiable DNA. The suboptimal powder generated by crushing the entire root may contain PCR inhibitors that reduce the efficiency of the PCR reaction during quantification. However, no significant delay in cycle threshold (Ct) of the internal PCR control was observed with the mechanically ground samples versus the filed samples (data not shown). This result was contrary to a previous finding that reported significant PCR inhibition after crushing the tooth using mechanical force [12]. Samples from donor 1 yielded substantially more DNA than all other teeth in this study regardless of the method used to extract the dentine (Table 1). These data suggest three possibilities; (1) contamination with exogenous DNA, (2) biological variation, meaning that this individual had a higher amount of DNA-rich tissue in their teeth than the other individuals investigated, and (3) the treatment environment. This tooth was buried for a period of 12 months and therefore may have been protected from the harsh environmental effects such as sunlight, humidity and microbial action that may degrade the DNA, although, teeth are thought to be resistant to these adverse conditions [2,3,22]. Due to the low sample numbers in this preliminary study, no explanation for the DNA yields can be drawn. However, contamination was eliminated as a source of ‘extra’ DNA as no additional alleles were detected during downstream STR typing of this sample.
3.2. STR analysis The quality of DNA recovered from teeth using both methods was assessed via STR analysis. The average number of correct alleles detected (Fig. 2A), and the average peak height ratio (PHR) of each sample profile (Fig. 2B) were the two key indicators measured. The filing method yielded more consistent and overall better results than those seen with the grinding method. Greater variation in both allele recovery and PHR was observed with the samples that were mechanically ground. Overall, the number of alleles amplified using DNA extracted from the filed teeth was significantly (P = 0.03) higher than DNA from teeth that were processed using the grinding method. In five of the seven teeth (1, 3, 4, 5 and 7) the filing method generated more complete STR profiles than the grinding method (Fig. 2A). It should also be noted that in three of those five samples (samples 3, 4 and 7) a complete (or near complete) profile was obtained with the filing method when very poor results were obtained by grinding the entire root (620% of alleles amplified). The STR profiles from the same three teeth also displayed considerable allelic imbalance using the grinding method (0–40% average PHR) (Fig. 2B). The ground samples showed wide variation in the average PHR (Fig. 2B). Three of the seven samples showed average PHR > 70% (±5.2%) and two with 40–50% (±2.9%). A profile PHR of 0% was calculated for one of the remaining two samples when only one allele was detected at heterozygote loci. An average PHR could not be calculated for the remaining sample as no alleles were detected. The filing method generally produced more balanced profiles than the grinding method, but statistical significance was not reached (P = 0.072). Six out of the seven filed samples produced profiles with an average PHR > 70% (±7%). One filed tooth (sample 2) had an average PHR of 25% (±19%). The wide variation seen in both the average PHR and number of alleles recovered from the filed sample of tooth 2 (Fig. 2) reflects the different profiling results from the triplicate STR reactions. Due to the total consumption of the DNA extract from this tooth sample, the PCR reaction could not be repeated. The more complete and balanced STR profiles obtained from the filed teeth may also be explained by the variation in the amount of DNA recovered using the two methods
35
leading to downstream differences in the amount of total DNA template in each PCR reaction (Table 1). The occurrence of drop-in alleles was monitored to assess the risk of contamination of both methods. More contamination was observed with the mechanically ground samples. In a total of 14 profiles (448 alleles) for each method (in duplicate), ten (2.2%) drop-in alleles were seen in ‘ground’ profiles as opposed to three (0.6%) in the ‘filing’ profiles. This result was surprising due to the substantially more direct handling of each tooth during the filing process. Although contamination may be introduced at every step of the process, one likely explanation is that the decontamination washing process of the whole teeth prior to grinding was not adequate to remove all exogenous DNA from the external surface of the roots, and the filing method therefore avoided removing powder from these surfaces. Of the three drop-in alleles seen in the filing method, two were consistent with handler profiles. However, this observation may not hold true if the decontamination process of whole teeth is improved and optimised to ensure all exogenous DNA is removed prior to processing. Results of this preliminary study constitute a proof of concept that the filing method described here is a viable method for extracting dentine powder from teeth exposed to various environmental insults, when tooth morphology must be retained. In this study, greater DNA yields were obtained using the filing method than grinding the entire tooth root, which in turn led to improved STR profiles. These results support a previous study which also showed poorer STR profiles after crushing the entire tooth or root compared with a form of endodontic access to the root canal (via the crown) for removing pulp and dentine [14]. The manual process of removing dentine with endodontic files was laborious and more time consuming than grinding the entire root (2–3 h vs 40 min). The significantly longer time required for extracting dentin powder from teeth may not be a major concern for museum and anthropological specimens, casework or research samples when the demand for DNA extractions from teeth is infrequent. However, when large numbers of tooth samples are required for processing and genotyping (such as in a mass disaster situation or missing persons cases), the extra time required per sample would not make this method viable. Nevertheless, the combined benefits of retaining tooth morphology and more complete and evenly balanced STR profiles may outweigh the disadvantages of the additional time and skill required to perform the filing process. Acknowledgements The authors would like to thank Dr. John Cosson, a maxillofacial surgeon for providing the teeth used in this study. In addition, Mr Tim Connolly, the Director of New Haven Funerals, kindly provided the use of a cremator oven and the plot of land required for the fieldwork. The authors would also like to thank Dr Patrick Warnke for his valuable assistance with the initial editing of this manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.legalmed.2015. 11.008. References [1] P.C. Malaver, J.J. Yunis, Different dental tissues as source of DNA for human identification in forensic cases, Croat. Med. J. 44 (3) (2003) 306–309. [2] A. Alvarez Garcia, I. Munoz, C. Pestoni, M.V. Lareu, M.S. Rodriguez-Calvo, A. Carracedo, Effect of environmental factors on PCR-DNA analysis from dental pulp, Int. J. Legal Med. 109 (3) (1996) 125–129.
36
S. Hughes-Stamm et al. / Legal Medicine 18 (2016) 31–36
[3] D. Marjanovic, A. Durmic-Pasic, N. Bakal, S. Haveric, B. Kalamujic, L. Kovacevic, et al., DNA identification of skeletal remains from the World War II mass graves uncovered in Slovenia, Croat. Med. J. 48 (4) (2007) 513–519. [4] C. Sosa, M. Baeta, C. Nunez, Y. Casalod, A. Luna, B. Martinez-Jarreta, Nuclear DNA typing from ancient teeth, Am. J. Forensic Med. Pathol. 33 (3) (2012) 211– 214. [5] I. Zupanic Pajnic, B. Gornjak Pogorelc, J. Balazic, T. Zupanc, B. Stefanic, Highly efficient nuclear DNA typing of the World War II skeletal remains using three new autosomal short tandem repeat amplification kits with the extended European Standard Set of loci, Croat. Med. J. 53 (1) (2012) 17–23. [6] Y.J. Deng, Y.Z. Li, X.G. Yu, L. Li, D.Y. Wu, J. Zhou, et al., Preliminary DNA identification for the tsunami victims in Thailand, Genomics Proteomics Bioinf. 3 (3) (2005) 143–157. [7] R. Gaytmenn, D. Sweet, Quantification of forensic DNA from various regions of human teeth, J. Forensic Sci. 48 (3) (2003) 622–625. [8] C.J. Adler, W. Haak, D. Donlon, A. Cooper, T.G. Consortium, Survival and recovery of DNA from ancient teeth and bones, J. Archaeol. Sci. 38 (2011) 956– 964. [9] A. Corte-Real, L. Andrade, M.J. Anjos, M. Carvalho, M.C. Vide, F. Corte-Real, et al., The DNA extraction from the pulp dentine complex of both with and without carious, Int. Congr. Ser. 1288 (2006) 710–712. [10] D. Sweet, D. Hildebrand, Recovery of DNA from human teeth by cryogenic grinding, J. Forensic Sci. 43 (6) (1998) 1199–1202. [11] N. Rohland, M. Hofreiter, Ancient DNA extraction from bones and teeth, Nat. Protoc. 2 (7) (2007) 1756–1762. [12] B.C. Smith, D.L. Fisher, V.W. Weedn, G.R. Warnock, M.M. Holland, A systematic approach to the sampling of dental DNA, J. Forensic Sci. 38 (5) (1993) 1194– 1209.
[13] R. Trivedi, P. Chattopadhyay, V.K. Kashyap, A new improved method for extraction of DNA from teeth for the analysis of hypervariable loci, Am. J. Forensic Med. Pathol. 23 (2) (2002) 191–196. [14] F. Tilotta, P. Brousseau, E. Lepareur, K. Yasukawa, P. de Mazancourt, A comparative study of two methods of dental pulp extraction for genetic fingerprinting, Forensic Sci. Int. 202 (1–3) (2010) e39–e43. [15] D.A. Bolnick, H.M. Bonine, J. Mata-Miguez, B.M. Kemp, M.H. Snow, S.A. LeBlanc, Nondestructive sampling of human skeletal remains yields ancient nuclear and mitochondrial DNA, Am. J. Phys. Anthropol. 147 (2) (2011) 293–300. [16] T. Kitayama, Y. Ogawa, K. Fujii, H. Nakahara, N. Mizuno, K. Sekiguchi, et al., Evaluation of a new experimental kit for the extraction of DNA from bones and teeth using a non-powder method, Legal Med. (Tokyo) 12 (2) (2010) 84–89. [17] J. Cobb, Ancient DNA recovered by a non-destructive method, Ancient Biomolecules 4 (4) (2002) 169–172. [18] M.J. Park, H.Y. Lee, U. Chung, S.C. Kang, K.J. Shin, Y-STR analysis of degraded DNA using reduced-size amplicons, Int. J. Legal Med. 121 (2) (2007) 152–157. [19] J. Davoren, D. Vanek, R. Konjhodzic, J. Crews, E. Huffine, T.J. Parsons, Highly effective DNA extraction method for nuclear short tandem repeat testing of skeletal remains from mass graves, Croat. Med. J. 48 (4) (2007) 478–485. [20] S.B. Seo, A. Zhang, H.Y. Kim, J.A. Yi, H.Y. Lee, D.H. Shin, et al., Technical note: efficiency of total demineralization and ion-exchange column for DNA extraction from bone, Am. J. Phys. Anthropol. 141 (1) (2009) 158–162. [21] S.R. Hughes-Stamm, K.J. Ashton, A. van Daal, Assessment of DNA degradation and the genotyping success of highly degraded samples, Int. J. Legal Med. 125 (3) (2011) 341–348. [22] C.Y. Shiroma, C.G. Fielding, J.A. Lewis Jr., M.R. Gleisner, K.N. Dunn, A minimally destructive technique for sampling dentin powder for mitochondrial DNA testing, J. Forensic Sci. 49 (4) (2004) 791–795.
