Direct PCR amplification from saliva sample using non-direct multiplex STR kits for forensic DNA typing

Due to its proficiency to provide the most discriminating results for forensic applications, medical research and anthropological studies, multiplex PCR based STR analysis has been established as the most efficient technique in the forensic DNA analysis. Several multiplex amplification kits based on 4, 5 and 6 dyes chemistry are commercially available and used in forensic DNA typing across the globe. These multiplex PCR systems are routinely used for amplification of multiple STR loci (Autosomal, Y and/or X STR’s) in the DNA extracted from various biological samples. In the routine forensic DNA testing, DNA profile obtained is compared with the DNA profile of the reference sample, which takes a certain turnaround time and employs costly lab resources. Successive development in forensic DNA typing have resulted in advent of improved multiplex kits which have reduced the effective analysis time, cost and minimized the number of steps required in comparison to conventional forensic DNA typing. Specialized direct amplification compatible multiplex kits are also available nowadays. These kits are relatively costlier but still require few pre-processing steps, which does not make them worth the hefty cost. Herein, this study, we have used non-direct multiplex STR kits to assess their efficacy for direct amplification. In the present study, 103 saliva samples were directly amplified without any pre-treatment of the samples using thirteen non-direct multiplex kits (4 dyes, 5 dyes and 6 dyes chemistry based) for forensic DNA typing. Here, we report a validated direct PCR amplification protocol from the reference saliva samples by omitting DNA extraction and quantification steps, which resulted in 80% reduction of the turnaround time. The developed protocol is cost effective, time efficient and it does not compromise with the quality of DNA profiles. To the best of our knowledge, this is the first report for direct amplification of DNA with the most commonly used non-direct multiplex STR kits without any pre-treatment of the sample. Complete DNA profiles matching all the essential quality parameters were obtained successfully from all the tested samples.

www.nature.com/scientificreports/ per manufacturer's recommendation. DNA was isolated and extracted from 188 saliva samples using automated DNA extraction system 12 GC (Precision System Science Co., Ltd., Matsudo, Japan) following manufacturer's instructions and pre formulated kits 41 . Few samples were tried for direct amplification of saliva samples with the same multiplex kit, using the same protocol for amplification. Promising results in terms of good quality DNA profile from the pilot study resulted into this detailed study on the direct amplification of saliva samples using commercially available non-direct multiplex kits. Kits used in the study were from leading brands including Thermo Fisher Scientific (AMPFLSTR IDENTIFILER, AMPFLSTR IDENTIFILER PLUS, GLOBALFILER, AMPFLSTR Y FILER and AMPFLSTR YFILER PLUS), Promega (POWERPLEX 16HS SYSTEM, POWERPLEX 21 SYSTEM, POWERPLEX FUSION 6C SYSTEM and POWERPLEX Y 23 SYSTEM) and Qiagen (INVESTIGA-TOR IDPLEX PLUS, and INVESTIGATOR ARGUS X-12 MULTIPLEX KIT) were tested for direct amplification protocol. This novel protocol was also tested using VERIFILER PLUS PCR AMPLIFICATION KIT (Thermo Fisher Scientific) and SURE ID PANGLOBAL HUMAN DNA IDENTIFICATION KIT (Health Gene Technologies) as well newly launched but still not used in routine forensic DNA typing kits. The details of multiplex kits used in this study are mentioned in Table 1. This study was designed with the aim to test and validate the direct amplification protocol using commercially available 4, 5 and 6 dyes chemistry based non-direct multiplex kits of leading brands in forensics.

Result and discussion
Optimizing parameters for direct PCR amplification. For the pilot study, initially 10 saliva samples were directly amplified using both the recommended full reaction volume of 25 µl and a reaction mixture with reduced volume of 10 µl with the help of PCR using all the tested multiplex kits. The DNA profile obtained from both the reaction volumes was evaluated, considering amplification of all the markers to showcase a full DNA profile, peak heights including inter and intra marker balance, observation of stutter and/or other artifacts. In this preliminary test, all the tested samples produced quality DNA profiles with both the reaction volumes and showed concordance. Our few previous studies have also reported use of 10 µl PCR reaction volume for an efficient reaction product, which further supported the reduction of final PCR reaction volume 42,43 . Direct amplification protocol was further validated for 10 µl reaction volume PCR reaction as per recommended PCR conditions of the particular multiplex kits. Using the above said direct amplification protocol, all the tested samples were directly amplified using non-direct 4 dyes chemistry based autosomal STR marker multiplex kit POW-ERPLEX 16HS SYSTEM (now not commonly used in forensics), 5  be obtained with all the tested multiplex kits using this protocol. All the tested samples were directly amplified at all the loci with the all the above mentioned multiplex PCR kits used in this study. However, it should be noted that the performance evaluation by the peak heights between the kits is only an approximation. It is because the same loci are labeled with different fluorescent dyes and have different amplicon sizes in different kits 22 and for the proper assessment of DNA profile, the overall quality of the profile using profile quality measures is more important 44 .
In conventional PCR method, Tris EDTA (TE) buffer is used for the final volume adjustment, which has also been reported to inhibit PCR amplification to some extent 19,45 . In this study, amplification grade water was used for final volume adjustment, which reduced the chances of reaction inhibition due to TE. In the case of DNA degradation or inhibition, larger/high molecular weight loci (usually more that 300 bp loci) reflect low peak height, higher peak imbalance or allele drop outs. This effect leads to development of a DNA profile with a slope peak height, also termed as SKI slope effect 19,46 . Noticeably, in our study almost all the DNA profiles obtained were having balanced peak heights without SKI slope effect. Rarely, very few DNA profiles were observed with the SKI slope effect which was due to the degradation of the samples. This was confirmed by the usual automated extraction followed by RT PCR of those few samples (data not shown).

Conventional DNA typing versus direct amplification methods.
To determine the difference between the conventional DNA typing (including DNA extraction, quality and quantity check, amplification and Genotyping), and the present method of direct amplification, and to estimate the DNA quantity likely to be present in the direct amplification reaction, 10 saliva samples were used for amplification with both the methods. The mean quantity of DNA was found to be 0.45 ng for all the 10 tested samples and most of the multiplex kits claim to be more sensitive for more than 0.25 ng input DNA 47,48 . With conventional and direct amplification methods, all the alleles were observed at all the loci of respective multiplex kits and balanced high quality DNA profiles were obtained, thus establishing concordance. Further, all the tested samples of this study were directly amplified using commonly used non-direct multiplex kits used in forensic DNA typing (Table 1). Direct PCR amplification protocol from the reference saliva samples by omitting DNA extraction and quantification steps, which resulted in 80% reduction of the turnaround time ( Table 2).

Assessment of DNA profile quality.
To assure the quality of DNA profile, Total Peak Height (TPH), Peak Height Ratio (PHR), inter and intra locus balance (local and global balance) of the obtained DNA profiles were analyzed. www.nature.com/scientificreports/ TPH is the sum of peak heights of both the heterozygous alleles in the DNA profile. PHR for heterozygous alleles defines the ratio of lower peak height and higher peak height at the specific marker. This value ranges from 0 to 1, where 1 represents the two identical peaks with equal heights and 0 represents the situation where one of the peaks is not observed which might be because of allele drop out due to PCR inhibition or degraded DNA 44 .
Inter and Intra locus balance was measured and represented in the local balance and global balance respectively, which is mean of TPH for all the tested loci in the multiplex kits.
Overall TPH, PHR and global balance quality parameters of DNA profile were evaluated statistically using one way analysis of variance (ANOVA) non parametric applying Friedman test at 5 percent of significant level and Dunn's Multiple Comparison test using Prism GrapghPad v5 software 49 .

Autosomal STR multiplex kits. AMPFLSTR IDENTIFILER AND AMPFLSTR IDENTIFILER PLUS PCR
AMPLIFICATION KIT. DNA profiles obtained with these multiplex kits were found to be complete and balanced. The average TPH and PHR ranged from 6553 to 7833 and from 0.875 to 0.984, respectively for both the kits. Out of 103 DNA profiles, only five DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles showed no significant variation in the global balance (Figs. 2, 3), (Supplementary Tables S1, S2).
POWERPLEX 16HS SYSTEM. DNA profiles obtained with this multiplex kit was also found to be balanced with an average TPH and PHR ranging from 6593 to 7889 and 0.863 to 0.993 respectively. Seven DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles exhibited no significant variation in the global balance (Fig. 4), (Supplementary Table S3).    Table S4).
INVESTIGATOR IDPLEX PLUS KIT. DNA profiles obtained with this multiplex kit were found to have an average TPH and PHR ranging from 6593 to 7883 and from 0.866 to 0.993, respectively. Out of the total studied DNA profiles, fourteen DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles exhibited no significant variation in the global balance ( Fig. 6), (Supplementary Table S5).
POWERPLEX FUSION 6C SYSTEM. DNA profiles obtained with this multiplex kit were found to have an average TPH and PHR values ranging from 6548 to 7927 and from 0.864 to 0.994, respectively. Out of the total studied DNA profiles, ten DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles accounted no significant variation in the global balance ( Fig. 7), (Supplementary Table S6).
GLOBALFILER PCR AMPLIFICATION KIT. DNA profiles obtained with this multiplex kit showed average TPH and PHR values ranging from 6546 to 7924 and from 0.861 to 0.994, respectively. Out of the total studied DNA profiles, ten DNA profiles showed ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles exhibited no significant variation in the global balance (Fig. 8), (Supplementary  Table S1).
VERIFILER PLUS PCR AMPLIFICATION KIT. DNA profiles obtained with this multiplex kit had an average TPH and PHR values ranging from 6549 to 7928 and from 0.867 to 0.995 respectively. Out of the total studied DNA profiles, nine DNA profiles were with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles exhibited no significant variation in the global balance ( Fig. 9), (Supplementary Table S8).   Table S9).

Y-STR multiplex kits.
All the male samples in this study were directly amplified using Y STR multiplex kits. The haplotype data of these samples were statistically evaluated on TPH and global balance parameters. The obtained profiles matched the standard criterion of profile quality index for most of the tested sample.
POWERPLEX Y 23 SYSTEM. DNA profiles obtained with this multiplex kit showed an average TPH range from 3224 to 4019. Out of the total studied DNA profiles, six DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles accounted no significant variation in the global balance (Fig. 11), (Supplementary Table S10).
AMPFLSTR YFILER PCR AMPLIFICATION KIT. DNA profiles obtained with this multiplex kit were found to have an average TPH values ranging from 3211 to 4015. Out of the total studied DNA profiles, eleven DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles did not show any significant variation in the global balance (Fig. 12), (Supplementary Table S11).
AMPFLSTR YFILER PLUS PCR AMPLIFICATION KIT. DNA profiles obtained with this multiplex kit were found to have an average TPH ranging from 3213 to 4051. Out of the total studied DNA profiles, seven DNA profiles had ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles showed no significant variation in the global balance (Fig. 13), (Supplementary Table S12). www.nature.com/scientificreports/ X-STR multiplex kit. Since male sample possesses only one X chromosome, it resulted in single peak at particular loci of X-STR marker, and female samples showed two peaks at particular loci due to the two chromosomes. Thus, the DNA profile obtained from the male samples were evaluated for TPH and global balance and the DNA profile obtained from the female samples were evaluated for TPH, PHR and global balance as well.
Here we used the mean value of THP, PHR for this evaluation. Conclusive outcomes have been represented in Fig. 14 and Supplementary Table S13 for both the male and female DNA profiles accordingly.
INVESTIGATOR ARGUS X-12 MULTIPLEX KIT. DNA profiles obtained with this multiplex kit were found to have an average TPH value ranging from 6593 to 7949 (for both the male and female DNA profiles) and the mean PHR ranged from 0.847 to 0.999 (for female DNA profiles). Out of the total studied DNA profiles, twenty three DNA profiles were observed with ski slope effect. In Dunn's Multiple Comparison test, at 5 percent significant level, all the DNA profiles showed no significant variation in the global balance (Fig. 14), (Supplementary  Table S13).
Overall data suggests that this novel protocol of direct amplification worked well with all the multiplex kits used in this study. The proposed method will not only save the effective cost, but will also curb down the turnaround time. Moreover, this novel direct amplification protocol will supposedly be useful for the speedy analysis of forensic DNA cases and ultimately will lead to the justice dissemination at the earliest. Quality control. All the authors have passed proficiency test and quality control exercise for DNA fingerprinting from GITAD, Spain (http:// gitad. ugr. es/ princ ipal. htm). Also internal laboratory control standards were followed and controls provided with multiplex kits were used.