An evaluation of the performance of five extraction methods: Chelex® 100, QIAamp® DNA Blood Mini Kit, QIAamp® DNA Investigator Kit, QIAsymphony® DNA Investigator® Kit and DNA IQ™

Science and Justice 55 (2015) 200–208

Contents lists available at ScienceDirect

Science and Justice journal homepage: www.elsevier.com/locate/scijus

An evaluation of the performance of five extraction methods: Chelex® 100, QIAamp® DNA Blood Mini Kit, QIAamp® DNA Investigator Kit, QIAsymphony® DNA Investigator® Kit and DNA IQ™ Stephen C.Y. Ip ⁎, Sze-wah Lin, Kam-ming Lai ⁎⁎ Forensic Science Division, Government Laboratory, Homantin Government Offices, Kowloon, Hong Kong, China

a r t i c l e

i n f o

Article history: Received 28 October 2014 Received in revised form 9 January 2015 Accepted 20 January 2015 Keywords: DNA extraction Investigator DNA IQ Chelex Blood Mini Microcon

a b s t r a c t DNA left at a crime scene was often limited in amount and far from pristine. To maximize the chance of recovering as much information as possible from such compromised samples, an appropriate extraction method using the available technologies needs to be devised. In this study, we used human blood, buffy coat and a total of 76 simulated touch DNA samples to test the effectiveness of the following five common DNA extraction methods, namely, Chelex® 100, QIAamp® DNA Blood Mini Kit, QIAamp® DNA Investigator Kit, QIAsymphony® DNA Investigator® Kit and DNA IQ™ system, in the recovery of such DNA. We demonstrated that the QIAamp® and QIAsymphony® DNA Investigator® Kits, and the DNA IQ™ system, exhibited a better effectiveness in DNA recovery amongst these methods and yielded extracts with higher success rate in subsequent DNA profiling. These extracts also generated profiles with better intra-colour signal balance. The findings in this work allowed us to propose an extraction approach as follows: 1) casework samples shall be extracted with the QIAamp®/ QIAsymphony® DNA Investigator® Kits or the DNA IQ™ system, viz., QIAsymphony® DNA Investigator® Kit and DNA IQ™, due to their higher throughput, are for the touched DNA evidence from the volume crime, while QIAamp® DNA Investigator Kit is preferable for challenging bloodstain samples; and 2) control samples, such as buccal swab, with known identity can be extracted with the Chelex, due to their cheaper cost per sample. © 2015 The Chartered Society of Forensic Sciences. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction The success of DNA profiling of crime scene samples relies largely on the quality and quantity of the DNA obtainable from the DNA purification process, which is the first step of any DNA analysis. An appropriate DNA extraction method ensures not only the target DNA to be efficiently extracted from the substrate, but also allows the removal of any potential inhibitor(s) which will interfere with subsequent downstream processing. In the last two decades, a number of extraction methods, ranging from in-house developed manual methods to high-end commercial DNA extraction kits amendable to automation, have been developed in the forensic scientific field to cope with the vast variety of extraction substrates encountered in the scenes [1–24]. For instance, the most common methods include organic phenol chloroform extraction [9,21], Chelex® 100 extraction (referred to as “Chelex”; Bio-Rad Laboratories, Hercules, CA, USA) [1,2,14,15,18,22,25,26], QIAamp® DNA Blood Mini Kit (referred to as “Blood Mini”; QIAGEN, Hilden, Germany) [4,10,24,27] and DNA IQ™ system (referred to as “IQ”; Promega, Madison, WI, USA) [6,7,11,14,15,28]. Phenol chloroform ⁎ Corresponding author. Tel.: +852 2762 3799. ⁎⁎ Corresponding author. Tel.: +852 2762 3808. E-mail addresses: [email protected] (S.C.Y. Ip), [email protected] (K. Lai).

extraction, which has been widely used for many years, is particularly useful for the extraction of high molecular weight DNA, but the organic reagent is hazardous to health and the procedure is relatively time consuming and labour intensive. Chelex® 100 ion-exchange resins are composed of styrene divinylbenzene copolymers containing paired iminodiacetate ions, which act as chelating groups in binding polyvalent metal ions. Chelex extraction is relatively simple and its reagent is inexpensive; however, PCR inhibitors are frequently found in the extract. To minimize the inhibition problem, the extract has to be further processed with centrifugal filter devices, such as the Microcon® DNA Fast Flow centrifugal filter device (referred to as “Microcon”; Merck Millipore, Darmstadt, Germany) [1,29,30], for the concentration of DNA. In contrast to Chelex, extracts isolated by Blood Mini and IQ, respectively, are relatively clean and are almost free from inhibitors [6, 7,10,11,14,15,24,27,28]. Blood Mini involves the QIAamp® silica-gel membrane which binds DNA specifically in the presence of the guanidine-based lysis buffer, while PCR inhibitors such as polyvalent cations and proteins are removed during subsequent washes, leaving DNA in the eluent. IQ, on the other hand, employs a paramagnetic resin for DNA isolation and one particular advantage of this system is its possibility of large scale automation with high throughput for up to 96 samples on robotic systems, such as the TECAN robotic liquid handling workstations [31], the Beckman BioMek 2000 [6,7] and the

http://dx.doi.org/10.1016/j.scijus.2015.01.005 1355-0306/© 2015 The Chartered Society of Forensic Sciences. Published by Elsevier Ireland Ltd. All rights reserved.

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

Perkin Elmer JANUS® automated workstations (referred to as “JANUS”; Perkin Elmer, Waltham, MA, USA) [32]. Recently, QIAGEN released an improved version of Blood Mini, namely QIAamp® DNA Investigator Kit, specially designed for the purification of genomic DNA from a wide range of forensic samples [14,18,22,23,33]. Its principle resembles that of Blood Mini, with the use of silica-membrane-based QIAamp® MinElute spin column for purification. This column enables a smaller elution volume of between 20 to 100 μL (as compared to 200 μL of Blood Mini), leading to a more concentrated eluent, thereby saving time and money for the additional Microcon purification step. Similar to IQ, this extraction kit can be fully automated on the QIAGEN QIAcube extraction system (referred to as “QIAcube”; QIAGEN, Hilden, Germany) in small scale with up to 12 samples per batch of extraction. Alternatively, for larger scale high throughput run with up to 96 samples, the QIAsymphony® DNA Investigator® Kit (referred as “QIAsymphony”; QIAGEN, Hilden, Germany) could be used on another QIAGEN platform QIAsymphony® SP, which offers the silica-based DNA purification on magnetic particles [22,23]. In view of the different choices of extraction methods that will affect the success rate of the subsequent DNA profiling step, we carried out a comprehensive study on the performance of these extraction methods (Table 1), including Chelex, Blood Mini, DNA Investigator Kits on QIAcube and QIAsymphony, and IQ on JANUS, using human blood, buffy coat and a total of 76 simulated touch DNA samples as the extraction substrates. Human blood was chosen, as it is one of the most commonly encountered biological evidences in serious crime. Similar to other body fluids such as semen and saliva, it is a rich source of human DNA for STR typing analysis. In addition, the heme group of haemoglobin in blood, being a common PCR inhibitor affecting the effectiveness of subsequent typing analysis, allows a direct comparison of the effectiveness of the various extraction methods in the removal of such. On the other hand, buffy coat has a relatively “cleaner” matrix consisting of mainly white blood cells (leukocytes) and platelet layer of the whole blood, and the number of cells for extraction could be quantitated by direct haemocytometer count so that a more homogenous starting material could be prepared to mitigate the effects caused by the complex matrix present in the blood samples. Hence, the use of serially diluted blood and buffy coat samples, as well as the simulated touch DNA samples, could shed light on the effectiveness of these extraction methods on the DNA analysis.

201

2.2. Preparation of buffy coat sample A human buffy coat sample was purchased from Innovative Research, Minneapolis, MI. The number of cells in the sample was first estimated by direct haemocytometer count under phase-contrast microscopy. Dilutions of cell samples (undiluted, 10-, 100-, 1000- and 2000-fold; equivalent to approximately 27,000, 2700, 270, 27 and 13 cells per 10 μL respectively) were prepared with phosphate buffer saline. Ten μL of diluted cells was deposited on cotton bud swabs. The samples were air-dried and stored at −20 °C before analysis. 2.3. Preparation of simulated touch DNA samples A total of 76 items in the office commonly used by our laboratory staff were used. These included 18 individuals' personally used items such as keyboard, mouse, telephone set, date chop and pen, together with other communal items such as light switch, photocopier buttons, electronic keypads and door knobs. For the sampling of DNA from these items, their surfaces were firstly cleaned with 75% ethanol. Before the experiment was commenced, DNA swabs were taken of the surfaces of these items following the “double swab technique” as described by Pang et al. [34] as negative controls. After these items being used for 7 days, DNA swabs were taken of the surfaces of these items again and the swabs were stored at −20 °C before analysis. The above cleaning and sampling process were repeated for three more times. 2.4. Chelex extraction Extraction of DNA was performed using 5% Chelex® 100 in autoclaved deionized water following the protocol described by Walsh et al. [2]. A two hundred μL DNA extract was removed from the Chelex resin for further analysis. 2.5. DNA Blood Mini extraction Extraction of DNA was performed using the QIAamp® DNA Blood Mini Kit following the manufacturer's recommendations. DNA was eluted in approximately 200 μL of buffer AE. 2.6. Microcon concentration on Chelex and Blood Mini DNA extracts Microcon concentration was performed using Merck Millipore Microcon® Centrifugal Filters (Merck, Darmstadt, Germany) following manufacturer's protocols. Approximately 35 μL of concentrated DNA extract was collected after the DNA concentration process.

2. Materials and methods 2.1. Preparation of blood sample

2.7. DNA IQ™ on the JANUS® workstation Serial dilutions of blood samples (undiluted, 25-, 100-, 250- & 1000fold) were prepared with Tris–EDTA (TE) buffer. Ten μL of the diluted blood samples was deposited on 1 cm × 1 cm of sterile cotton gauze. The samples were air-dried and then stored at −20 °C before analysis. Table 1 Details of the five extraction methods tested in this study.

Extraction of DNA was performed using the Promega DNA IQ™ system following the manufacturer's recommendations [35]. The extraction process was performed on the Perkin Elmer JANUS® automated workstation (Perkin Elmer, Waltham, MA, USA). DNA was eluted in approximately 40 μL of Elution buffer. 2.8. QIAamp® DNA Investigator Kit on the QIAcube system

Method (platform)

Technology

Max. Eluant Microcon temp.

Chelex (manual) DNA Blood Mini (manual) DNA Investigator (QIAcube, QIAGEN) DNA Investigator (QIAsymphony, QIAGEN) DNA IQ (JANUS, Perkin Elmer)

Ion-exchange resin Silica-based membrane Silica-based membrane Silica-based magnetic particles

95 °C

200 μL 35 μL

85 °C

200 μL 35 μL

Extraction of DNA was performed using the QIAamp® DNA Investigator Kit following manufacturer's protocol on the QIAcube system. Approximately 60 μL of DNA extract was collected.

70 °C

60 μL n/a

2.9. QIAsymphony® DNA Investigator® Kit on the QIAsymphony platform

70 °C

60 μL n/a

70 °C

40 μL n/a

Paramagnetic resin

Extraction of DNA was performed using the QIAsymphony® DNA Investigator® Kit following manufacturer's protocol on the QIAsymphony® SP platform. Approximately 60 μL of DNA extract was collected.

202

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

2.10. DNA quantification Quantification was performed using the Quantifiler® Human DNA Quantification Kit (Life Technologies™, Carlsbad, CA, USA) and ABI PRISM® 7000 Sequence Detection System according to the manufacturer's specifications. The collected data were analyzed by Applied Biosystems® SDS Software v1.0 (Life Technologies™, Carlsbad, CA, USA). 2.11. PCR Amplification with AmpFℓSTR® Identifiler® PCR Amplification Kit Samples were amplified using the AmpFℓSTR® Identifiler® PCR Amplification Kit (Life Technologies™, Carlsbad, CA, USA) on the Applied Biosystems® GeneAmp® PCR System 9700 (Life Technologies™, Carlsbad, CA, USA) following manufacturer's recommendations. A maximum of 1 ng or 10 μL of the extracted DNA sample was added in a total reaction volume of 25 μL for amplification. 2.12. STR separation Amplified products of Identifiler® were analyzed on the ABI PRISM® 3100 Genetic Analyzer (Life Technologies™, Carlsbad, CA, USA) using POP-4™ polymer according to manufacturer's recommendations. Data obtained from the runs were collected using the Applied Biosystems® 3100 Data Collection Software v1.1. Data for the Identifiler® were analyzed with GeneMapper® ID-X v1.2 software using the panels and bins provided by Applied Biosystems (Life Technologies™, Carlsbad, CA, USA). 2.13. STR data analysis For the STR typing analysis of human blood and buffy coat samples, a reportable locus refers to the correct detection of a homozygous allele with peak height above the stochastic threshold or heterozygote alleles with peak height above the analytical threshold, whereas for the simulated touch DNA samples, the number of loci refers to the detection of allele(s) with peak height above the analytical threshold. 2.14. Statistical analysis The yield, concentration, number of reportable loci and average peak heights obtained from various extraction methods were evaluated using one-way ANOVA followed by Newman–Keuls multiple comparison tests or unpaired t-test using Prism 5.0 (GraphPad Software, San Diego, CA, USA). All values were expressed as the mean ± S.E.M. 3. Results and discussion 3.1. QIAcube and QIAsymphony offer better DNA recovery for the whole blood samples In order to compare the DNA recovery of each extraction method, the total DNA yield was compared using undiluted and 25- to 1000fold diluted blood samples as starting materials. As shown in Fig. 1A (top panel), QIAcube and QIAsymphony offered the highest DNA recovery for the three sets of triplicate undiluted blood samples amongst these methods, which yielded more than 500 ng on average, whereas Blood Mini, Chelex and IQ yielded significantly less DNA of 300 ng or below. We further tested the DNA recovery of these extraction methods with blood samples with various dilutions. Our results showed that QIAcube consistenly performed the best on these diluted samples and yielded more DNA than the other four methods (Fig. 1A). QIAsymphony performed the second best, followed by Chelex, Blood Mini and IQ, all of which offered similar DNA recovery for these diluted samples. The average DNA concentrations obtained from these samples

were shown in Supplementary Fig. 1A, which showed that QIAcube generally yielded DNA of higher concentration than the others, especially for the whole blood with 100- to 1000-fold dilutions. We have demonstrated above that QIAcube and QIAsymphony both performed better than Blood Mini; hence, we repeated the above experiment on the buffy coat samples without further testing the latter. Our results showed that QIAcube, QIAsymphony and IQ all yielded more DNA than Chelex from the buffy coat samples with dilution up to 100fold (Fig. 1B). Not only the average amount of DNA obtained from these extracts were higher than those obtained from Chelex (approximately 2- to 5-fold; Fig. 1B), but these DNA extracts were also more concentrated (Supplementary Fig. 1B). 3.2. QIAcube and QIAsymphony yield more complete DNA profiles for blood samples In order to learn more about the quality of the DNA obtained from these methods, we amplified the blood and buffy coat DNA extracts with AmpFℓSTR Identifiler® PCR Amplification Kit to reveal if there is any difference in the DNA profiles generated by these extraction methods. Our amplification results showed that all extraction methods yielded complete profiles for the undiluted blood samples and nearly complete profiles for the 25-fold diluted samples (Fig. 2A). Consistent with the DNA quantification results (Fig. 1A), the extracts obtained from QIAcube and QIAsymphony yielded the best typing results amongst these methods for the diluted samples, in which complete profiles could be obtained from the 100-fold diluted samples. In addition, they both yielded nearly complete profiles even for blood samples with 250-fold dilution, and partial profiles with an average of 7 to 9 loci for the 1000-fold diluted blood samples. On the other hand, IQ yielded nearly complete profiles for blood samples with 100-fold dilutions only, and partial profiles with an average of 7 and 2 loci for the 250-fold and 1000-fold diluted blood respectively. Chelex and Blood Mini yielded an average of 13 to 14 and 5 to 6 loci for the 100fold and the 250-fold diluted blood respectively. No reportable loci were obtained for the 1000-fold diluted blood. To understand if Microcon has enhanced the typing efficiency, we have also examined the typing efficiency of the Chelex and the Blood Mini extracts prior to the Microcon concentration. We found that these extracts, prior to Microcon, yielded nearly complete profiles for the undiluted samples only, and yielded 12 to 14 loci and less than 4 loci, respectively, for the 25-fold and 100-fold diluted blood, whereas it yielded almost no profile for the 250-fold diluted blood (Fig. 3). All these data indicated that Microcon treatment increases the total number of reportable loci, and such enhancement is due to the increase of the ratio of DNA:PCR inhibitor(s), i.e., as a result of DNA concentration, more DNA template but less PCR inhibitor(s), could be added in the profiling reaction. Regarding to the buffy coat samples, our results revealed that QIAcube, QIAsymphony and IQ all yielded extracts with similar typing efficiency and all of them performed better than Chelex (Fig. 2B), which is consistent with the DNA quantification results (Fig. 1B) and the typing results of DNA extracts from the blood samples (Fig. 2A). These three methods all yielded complete profiles at 10-fold cell dilutions, which are approximately 2700 cells (Fig. 2B). Chelex, however, yielded only partial profiles with an average of 10 loci. For the 100-fold diluted samples with approximately 270 cells, an average of 6 and 10 reportable loci were detected from the QIAcube/ QIAsymphony and IQ, respectively, whereas no typing results were obtained from Chelex. For the 1000-fold dilution with approximately 27 cells and the 2000-fold dilution with approximately 13 cells, no typing results were obtained from all extraction methods. These data on the typing results indicated that QIAcube, QIAsymphony and IQ have a better extraction performance than Chelex, in which the two Investigator Kits yielded a more complete typing result for the blood samples, whereas IQ yielded profiles with more reportable loci for the relatively “cleaner” buffy coat samples. The performance difference between IQ

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

A

d

600

Undiluted

Key:

500

Chelex

ng

400 300

203

c

Blood Mini QIAcube

b

200

QIAsymphony

a

100

IQ

0

B

35

c

30

40

b

25

30

20 15 10

20

a

a

10

5 0

0

c

10

5

b

4

ng

ng

6 4

a

a

a

1 0

4

c

b

0.8

250-fold

100-fold

0.6

3

b

2

ng

ng

~5-fold

3 2

0

1

10-fold

b

6

100-fold

8

2

~2-fold

a

ng

ng

25-fold

Undiluted

b

50

a

a

~5-fold

0.4 0.2

a

0

0

2

0.20 1000-fold

1000-fold

b

0.15

c ng

ng

1.5 1

0.10

b 0.5

a a,b

a 0

0.05

a

0

Blood

Buffy coat

Fig. 1. The yield obtained from the undiluted and the serially diluted (A) blood samples (mean ± S.E.M., n = 9) and (B) buffy coat samples (mean ± S.E.M., n = 9) by different extraction methods. Different letters (a, b, c & d) indicate statistical significance (P b 0.05).

204

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

A 15

Key: Chelex (-Microcon)

***

14

13

Chelex (+Microcon)

13

12

12

11

Blood Mini (-Microcon) Blood Mini (+Microcon)

b

a

10

No. of reportable loci

No. of reportable loci

***

14

11

9

a

8 7

10 9

7

5 4

3

3 2 1

2

0

1

a

0 Undiluted

25-

B

100-

250-

1-

25-

100-

250-

1000-

1000Fig. 3. Microcon concentration improved the profiling success of whole blood DNA extracts from Chelex and Blood Mini (mean ± S.E.M., n = 9). ***P b 0.005 versus without Microcon treatment.

b

15

Key:

14

Chelex Blood Mini QIAcube QIAsymphony IQ

a

13

c

12 11 10 9

b,c

8

b

7 6 5 4 3 2 1 0

***

6

5

a

***

8

6

4

No. of reportable loci

*** ***

15

b

a Undiluted

10-

100-

1000-

2000-

Fig. 2. Comparison of the number of reportable loci obtained from the undiluted and the serially diluted (A) blood samples (mean ± S.E.M., n = 9) and (B) buffy coat samples (mean ± S.E.M., n = 9) by different extraction methods. Different letters (a, b & c) indicate statistical significance (P b 0.05).

and QIAcube/QIAsymphony in these two sample types could be due to the fact that whole blood contains large quantities of haemoglobin which competes with DNA for binding to the IQ resin, whereas the buffy coat does not contain similar quantity of competing protein. 3.3. QIAcube, QIAsymphony and IQ yield profiles with better quality We further compared if there were any differences between the qualities of the DNA profiles amongst the extracts prepared by different methods. The presence of inhibitors in the extract could interfere with the multiplex PCR efficiency, leading to lower average peak height of a profile, heterozygous peak height imbalance and intra-colour signal imbalance, all of which could pose a challenge to subsequent data interpretation. To address this, we first compared the average peak heights and heterozygous peak height ratios of the DNA profiles obtained from the undiluted blood and buffy coat DNA extracts prepared by different extraction methods, as the input DNA was in the optimal level

for amplification, i.e. 1 ng. Fig. 4A and B showed the average peak heights and the heterozygous peak height ratios of the blood DNA and buffy coat DNA extracts obtained from these extraction methods respectively. Our data showed that Blood Mini/QIAcube/IQ and QIAcube/ QIAsymphony/IQ all yielded profiles with higher average peak heights than Chelex for the blood and buffy coat DNA extracts respectively. QIAsymphony, on the other hand, yielded profiles with similar average peak heights as Chelex for the blood DNA extracts, indicating the presence of inhibitor(s) in these extracts for the profiling analysis. We also compared the heterozygous peak height balance of the amplified profiles of these extracts. As shown in Fig. 4, all methods offered acceptable heterozygous peak height ratios with the majority above 0.70. We have shown in Fig. 3 that the application of Microcon could improve the typing efficiency of the whole blood DNA extracts; here we further showed that it could also improve both the average peak heights and the peak height balance, particularly for the Chelex extract (Fig. 5). We further compared the intra-colour balance of the profiles obtained from these extracts. Our data in Fig. 6 showed that Chelex blood DNA extracts (shown in yellow) displayed a bigger drop in peak heights across each electropherogram, whereas the blood DNA extracts obtained from QIAcube (shown in blue), QIAsymphony (shown in dark blue) and IQ (shown in pink) were subjected to less DNA degradation, as indicated by a more gradual decrease (see also Supplementary Fig. 2A). QIAcube appeared to yield profile with slightly better overall intracolour balance than QIAsymphony and IQ, as these two magneticbased kits performed well in most of the loci, except the vWA locus, in which lower peak heights were detected. It is interesting to point out that this drop in peak height was only detected in the blood DNA extract but not in the buffy coat extract (Figs. 6–7 & Supplementary Fig. 2B). Whether the lower peak height was due to the nature of magnetic beads-based purification or the presence of inhibitor(s) in the blood DNA extracts remained to be determined. In addition, we observed no noticeable difference between the profiles obtained from Chelex, QIAcube, QIAsymphony and IQ for the buffy coat DNA extracts (Fig. 7 & Supplementary Fig. 2B), further indicating that the presence of PCR inhibitor(s) in the blood DNA extracts, for example the haemoglobin, but not in the buffy coat DNA extract, might affect the quality of DNA profiling.

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

A

205

B

PHR 1.0

Key: Chelex Blood Mini QIAcube QIAsymphony IQ

0.9 0.8 0.7 0.6

PH (rfu) 1500

0.5

1250

b

b

PHR 1.0 0.9 0.8 0.7 0.6

PH (rfu) 2500

c

1000

a

a

750

1000

250 0

2000 1500

a

500

0

b

500 0

Blood

0

Buffy coat

Fig. 4. Comparison of the average peak heights (PH) of the profiles and the heterozygous peak height ratios (PHR) of the loci obtained from (A) the undiluted blood samples (mean ± S.E.M., n = 9) and (B) the undiluted buffy coat samples (mean ± S.E.M., n = 9) by different extraction methods. The rectangular box indicates median, upper and lower quarters of the peak height ratios. Different letters (a, b & c) indicate statistical significance (P b 0.05).

3.4. QIAcube, QIAsymphony and IQ improve DNA profiling on simulated touch DNA samples We have shown above that QIAcube, QIAsymphony and IQ yielded profiles with more reportable loci than Chelex from the buffy coat (Fig. 2B); here, we further used a total of 76 simulated touch DNA samples to compare these methods with Chelex. To safeguard the quality of the data and to avoid contamination, DNA swabs were taken from each of the items as negative controls before the experiment

PHR 1.0

Key: Chelex (-Microcon)

0.9

Chelex (+Microcon) 0.8

Blood Mini (-Microcon)

0.7

Blood Mini (+Microcon)

0.6

PH (rfu)

0.5

1250

0.4

1000

0.3

***

750

4. Conclusions

500

In summary, we demonstrated using serially diluted blood and buffy coat, and the 76 simulated touch DNA samples that QIAcube, QIAsymphony and IQ all yielded extracts with a higher success rate for the subsequent DNA typing analysis, as opposed to Chelex and Blood Mini even after their concentration with Microcon. Our data also showed that these three methods yielded extracts with better intracolour signal balance as compared with Chelex, and their improvement was most significant for the whole blood samples, which have a relatively “complex” matrix. Microcon could concentrate the DNA in Chelex extracts, thereby increased the profiling success; however, substantial loss of DNA was observed (data not shown) and we

250 0

commenced. We detected no DNA from these controls, showing that any DNA found on the items would be deposited by the users during the testing period. Compared with Chelex, both QIAcube/QIAsymphony and IQ resulted in a higher percentage of extracts with higher DNA concentration (Fig. 8A) and a higher number of loci being successfully amplified per sample (Fig. 8B). Up to 85% of the Chelex extracts yielded DNA with less than 0.023 ng/μL, compared with ~30%, ~39% and ~37% of QIAcube, QIAsymphony and IQ, respectively. Similarly, up to 37% of the Chelex extracts yielded no typing results, compared with less than 7% of the other three methods. The results were consistent with our findings with the buffy coat as shown above. All these data indicated that IQ, as well as QIAcube and QIAsymphony, are preferable to Chelex for the touch DNA samples. It is noteworthy that for this experiment, personal items yielded mostly single source profiles or mixed profiles with a major source deduced. We have compared these DNA profiles with those of the respective user, and all results were matched with the contributing sources. On the other hand, communal items produced mixed profiles with high complexity and no major source could be deduced from these mixed profiles. We also observed that a few users are better shedders than the others, and the DNA extraction methods of choice did not affect much the results for these shedders. In addition, we did not observe any differences in the performance of these extraction methods between the personal items and the communal items.

0

Blood Fig. 5. Microcon concentration improved the average peak heights (PH) and the heterozygous peak height ratios (PHR) of the Chelex whole blood DNA extracts (mean ± S.E.M., n = 9). The rectangular box indicates median, upper and lower quarters of the peak height ratios. ***P b 0.005 versus without Microcon treatment.

206

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

D8S1179

D21S11

D7S820

CSF1PO

D3S1358

TH01

D13S317 D16S539 D2S1338

1.2

1.0

1.0

0.8

0.8 0.6 0.6 0.4 0.4 0.2

0.2 0

0

D19S433

vWA

TPOX

D18S51

AmeI

1.2

1.2

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0

0

D5S818

FGA

Chelex Blood Mini QIAcube QIAsymphony IQ

Fig. 6. The average intra-colour signal balance obtained from the undiluted blood sample (n = 9) by different extraction methods. The signal strength of a locus was normalized against the first locus of its corresponding dye channel.

observed no improvement on the intra-colour signal balance (data not shown). The fact that Chelex yielded less DNA, and also with poor intra-colour balance in the blood DNA extracts, than the other extraction methods is mainly due to the presence of the co-extracted PCR inhibitors, which affects the efficiency of the PCR-based analysis, and/or

D8S1179

D21S11

D7S820

CSF1PO

the high temperature involved in the extraction process (Table 1) which could degrade the DNA to a greater extent. The findings in this work allowed us to propose an extraction approach as follows: 1) Casework samples shall be extracted with IQ or DNA Investigator Kit (on either QIAcube or QIAsymphony), viz., IQ and QIAsymphony, due to their

D3S1358

TH01

D13S317 D16S539 D2S1338

1.6

1.0

1.4

0.8

1.2 1.0

0.6

0.8

0.4

0.6 0.4

0.2

0.2 0

0

D19S433

vWA

TPOX

AmeI

D18S51

1.4

1.4

1.2

1.2

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0

0

D5S818

FGA

Chelex QIAcube QIAsymphony IQ

Fig. 7. The average intra-colour signal balance obtained from the undiluted buffy coat sample (n = 9) by different extraction methods. The signal strength of a locus was normalized against the first locus of its corresponding dye channel.

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

A 100 90

x ele Ch

QI

Ac

ub

e QI

As

ym

o ph

ny IQ

5

DNA conc.:

6

> 0.05 ng/µL

80

28

34

28

70

< 0.023 ng/µL

60 % 50 40

0.023 - 0.05 ng/µL

18

20

30

28

19

65

30 20

23

10 0

B 100 90 80 70 %

Ch

x ele

QI

Ac

u

be QI

As

y

h mp

on

y IQ No. of loci: 15

11 12

35

33 40

3

7 to 10

60 50

1 to 6 22

40

15

0

22

30 20

11 to 14

11 28

10 0

22

6 8 5

15 2

4 7 3

Fig. 8. (A) The DNA concentration, and (B) the number of loci with positive typing results obtained from the simulated touch DNA samples (n = 76) by different extraction methods.

higher throughput, are preferable for the touch DNA samples encountered in volume crime, whereas QIAcube is preferable for extracting challenging samples with “complex” matrix, e.g. bloodstains collected at scene; and 2) control known samples, e.g. buccal swabs, can be extracted with Chelex, due to its lower reagent cost per sample. Acknowledgements We thank Dr. C.M. Lau, the Government Chemist of the Government Laboratory, Dr. F.C. Kwok, the Assistant Government Chemist, and Mr. B.K.K. Cheung, the Chief Chemist for their support and the approval of the publication of this work. We thank Mr. W.N. Cheng and Ms. S.F. Lam for their outstanding technical support for this work. We also thank other members of our laboratory for their inputs and discussions. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scijus.2015.01.005. References [1] J.M. Jung, C.T. Comey, D.B. Baer, B. Budowle, Extraction strategy for obtaining DNA from bloodstains for PCR amplification and typing of the HLA-DQ alpha gene, Int. J. Legal Med. 104 (3) (1991) 145–148. [2] P.S. Walsh, D.A. Metzger, R. Higuchi, Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material, Biotechniques 10 (4) (1991) 506–513. [3] C.A. Scherczinger, M.T. Bourke, C. Ladd, H.C. Lee, DNA extraction from liquid blood using QIAamp, J. Forensic Sci. 42 (5) (1997) 893–896.

207

[4] S.A. Greenspoon, M.A. Scarpetta, M.L. Drayton, S.A. Turek, QIAamp spin columns as a method of DNA isolation for forensic casework, J. Forensic Sci. 43 (5) (1998) 1024–1030. [5] B. Rerkamnuaychoke, W. Chantratita, U. Jomsawat, J. Thanakitgosate, N. Pattanasak, P. Rojanasunan, Comparison of DNA extraction from blood stain and decomposed muscle in STR polymorphism analysis, J. Med. Assoc. Thai. 83 (Suppl. 1) (2000) S82–S88. [6] S.A. Greenspoon, J.D. Ban, K. Sykes, E.J. Ballard, S.S. Edler, M. Baisden, et al., Application of the BioMek 2000 Laboratory Automation Workstation and the DNA IQ System to the extraction of forensic casework samples, J. Forensic Sci. 49 (1) (2004) 29–39. [7] C.A. Crouse, S. Yeung, S. Greenspoon, A. McGuckian, J. Sikorsky, J. Ban, et al., Improving efficiency of a small forensic DNA laboratory: validation of robotic assays and evaluation of microcapillary array device, Croat. Med. J. 46 (4) (2005) 563–577. [8] S.A. Montpetit, I.T. Fitch, P.T. O'Donnell, A simple automated instrument for DNA extraction in forensic casework, J. Forensic Sci. 50 (3) (2005) 555–563. [9] S. Kochl, H. Niederstatter, W. Parson, DNA extraction and quantitation of forensic samples using the phenol–chloroform method and real-time PCR, Methods Mol. Biol. 297 (2005) 13–30. [10] V. Castella, N. Dimo-Simonin, C. Brandt-Casadevall, P. Mangin, Forensic evaluation of the QIAshredder/QIAamp DNA extraction procedure, Forensic Sci. Int. 156 (1) (2006) 70–73. [11] N. Franke, C. Augustin, K. Puschel, Optimization of DNA-extraction and typing from contact stains, Forensic Sci. Int. Genet. (Suppl. 1) (2008) 423–425. [12] Z. Krnajski, S. Geering, S. Steadman, Performance verification of the Maxwell 16 Instrument and DNA IQ Reference Sample Kit for automated DNA extraction of known reference samples, Forensic Sci. Med. Pathol. 3 (4) (2007) 264–269. [13] M.G. Brevnov, H.S. Pawar, J. Mundt, L.M. Calandro, M.R. Furtado, J.G. Shewale, Developmental validation of the PrepFiler Forensic DNA Extraction Kit for extraction of genomic DNA from biological samples, J. Forensic Sci. 54 (3) (2009) 599–607. [14] V. Bogas, F. Balsa, M. Carvalho, M.J. Anjos, M.F. Pinheiro, F. Corte-Real, Comparison of four DNA extraction methods for forensic application, Forensic Sci. Int. Genet. (Suppl. 3) (2011) e194–e195. [15] T.J. Verdon, R.J. Mitchell, R.A. van Oorschot, Evaluating the efficiency of DNA extraction methods from different substrates, Forensic Sci. Int. Genet. (Suppl. 3) (2011) e963–e994. [16] M. Stangegaard, T.G. Froslev, R. Frank-Hansen, A.J. Hansen, N. Morling, Automated extraction of DNA from blood and PCR setup using a Tecan Freedom EVO liquid handler for forensic genetic STR typing of reference samples, J. Lab. Autom. 16 (2) (2011) 134–140. [17] C.P. Davis, J.L. King, B. Budowle, A.J. Eisenberg, M.A. Turnbough, Extraction platform evaluations: a comparison of AutoMate Express, EZ1(R) Advanced XL, and Maxwell(R) 16 Bench-top DNA extraction systems, Leg. Med. (Tokyo) 14 (1) (2012) 36–39. [18] K. Phillips, N. McCallum, L. Welch, A comparison of methods for forensic DNA extraction: Chelex-100® and the QIAGEN DNA Investigator Kit (manual and automated), Forensic Sci. Int. Genet. 6 (2) (2012) 282–285. [19] J.Y. Liu, C. Zhong, A. Holt, R. Lagace, M. Harrold, A.B. Dixon, et al., AutoMate Express forensic DNA extraction system for the extraction of genomic DNA from biological samples, J. Forensic Sci. 57 (4) (2012) 1022–1030. [20] K. Fujii, S. Inokuchi, T. Kitayama, H. Nakahara, N. Mizuno, K. Sekiguchi, A comparison of DNA extraction using AutoMate Express and EZ1 advanced XL from liquid blood, bloodstains, and semen stains, J. Forensic Sci. 58 (4) (2013) 981–988. [21] S. Ghatak, R.B. Muthukumaran, S.K. Nachimuthu, A simple method of genomic DNA extraction from human samples for PCR–RFLP analysis, J. Biomol. Tech. 24 (4) (2013) 224–231. [22] M. Stangegaard, B.B. Hjort, T.N. Hansen, A. Hoflund, H.S. Mogensen, A.J. Hansen, et al., Automated extraction of DNA from biological stains on fabric from crime cases. A comparison of a manual and three automated methods, Forensic Sci. Int. Genet. 7 (3) (2013) 384–388. [23] M. Stangegaard, C. Borsting, L. Ferrero-Miliani, R. Frank-Hansen, L. Poulsen, A.J. Hansen, et al., Evaluation of four automated protocols for extraction of DNA from FTA cards, J. Lab. Autom. 18 (5) (2013) 404–410. [24] M.C. Thomas, M.J. Shields, K.R. Hahn, T.W. Janzen, N. Goji, K.K. Amoako, Evaluation of DNA extraction methods for Bacillus anthracis spores isolated from spiked food samples, J. Appl. Microbiol. 115 (1) (2013) 156–162. [25] J.M. Butler, DNA extraction from forensic samples using chelex, Cold Spring Harb. Protoc. (6) (2009) (pdb prot5229). [26] Bio-rad laboratories, Chelex®-100 and Chelex®-20 Chelating Ion Exchange Resin Instruction Resin Instruction Manual, 1998.. [27] QIAGEN, QIAamp® DNA Mini and Blood Mini Handbook, Third edition, 06/2012. [28] Promega, DNA IQ™ System Small Sample Casework Protocol, Part# TB296,11/13 (2013). [29] W.R. Hudlow, R. Krieger, M. Meusel, J.C. Sehhat, M.D. Timken, M.R. Buoncristiani, The NucleoSpin(R) DNA Clean-up XS kit for the concentration and purification of genomic DNA extracts: an alternative to microdialysis filtration, Forensic Sci. Int. Genet. 5 (3) (2011) 226–230. [30] Merck Millipore Ltd, User Guide: Microcon® Centrifugal Filter Devices, 08/13 (2013). [31] C.J. Fregeau, C.M. Lett, R.M. Fourney, Validation of a DNA IQ-based extraction method for TECAN robotic liquid handling workstations for processing casework, Forensic Sci. Int. Genet. 4 (5) (2010) 292–304.

208

S.C.Y. Ip et al. / Science and Justice 55 (2015) 200–208

[32] L. Tack, G. Reznik, Application Note: Walk-away Magnetic Bead-based DNA Purification Using the JANUS Automated Workstation, 2009. [33] QIAGEN, QIAamp® DNA Investigator Handbook, 04/2010.

[34] B.C. Pang, B.K. Cheung, Double swab technique for collecting touched evidence, Leg. Med. (Tokyo) 9 (4) (2007) 181–184. [35] Promega, DNA IQ™ System — Database Protocol, Part# TB297, 11/13 (2013).