Introduction

Screening of collected samples from crime scenes is a crucial first step in efficient DNA profiling. In sexual assault cases, it is often necessary to identify male-derived samples from a number of collected samples. Semen samples are easily screened by morphological analyses using a microscope or an immunochromatographic assay targeting prostate-specific antigen or semenogelin [1]. However, when the collected samples are blood or saliva stains, it may require time, effort, and cost to detect male-derived samples among numerous samples. Identifying human male DNA is useful for conducting subsequent genetic typing efficiently. Numerous methods for DNA-based sex identification have been described, including PCR or quantitative PCR targeting Y-chromosome DNA [2,3,4,5] or commercial kits such as the Quantifiler™ Trio DNA and Quantification Kit PowerQuant™ systems [6, 7]. However, these tests require several steps, including DNA extraction and amplification, and are limited to well-equipped laboratories. A simple and rapid screening method for male-derived samples from among many collected samples is therefore desirable.

Loop-mediated isothermal amplification (LAMP) is a rapid nucleic acid amplification method with high specificity and efficiency under isothermal conditions [8]. LAMP primers consist of two outer primers (F3 and B3) and two inner primers (FIP and BIP). Adding loop primers increases the sensitivity of reaction and accelerates the reaction speed [9]. Amplification can be observed with the naked eye or monitored by measuring turbidity caused by formation of magnesium pyrophosphate. Because special devices and expertise are not required for a LAMP assay, it is widely used for detecting viruses, bacteria, or plants [10,11,12,13,14].

With regard to forensic applications, it has been reported that the LAMP assay detecting amelogenin X and amelogenin Y makes it possible to identify sex from dental pulp [15]. However, because those LAMP assays were performed on 5 ng or more of DNA template and the amplification started at around 30 min, there is still room for improvement in sensitivity and reaction speed. Recently, it has been reported that the LAMP assay confirmed human DNA presence with high sensitivity and specificity using a colorimetric gold nanoparticle hybridization probe for visualization [16]. This technique made the evaluation of the LAMP results clearer than conventional LAMP detection methods using white turbidity, but required opening the tubes after the LAMP reaction to add the gold nanoparticle hybridization probe. Because of the high amplification efficiency of the LAMP reaction, opening the reaction tubes might cause contamination and should be performed carefully. In general, the LAMP reaction is visualized by adding calcein, a hydroxynaphthol blue dye, or an intercalating dye such as SYBR [13, 14, 17]. A recent report showed an improved LAMP detection system in which a reagent containing pyrophosphatase shortened the reaction time and the amplification was visualized using a colorimetric intercalating dye without requiring opening of the tubes after the LAMP reaction [18]. Thus, LAMP techniques are expected to become a useful method for rapid on-site screening in forensic investigations. In this study, we developed a LAMP assay targeting multi-repeat sequences of the Y chromosome for detection of male DNA and examined its application to an on-site screening method.

Materials and methods

Sample preparation and DNA quantification

Six blood samples from three men and three women were obtained prior to this study and kept frozen for approximately 10 years in collection tubes. Saliva samples were collected from three men and three women. Purified genomic DNA was extracted from blood samples with an EZ1 Investigator Kit (Qiagen, Germany) according to the manufacturer’s instructions. Purified genomic DNA was quantified using real-time PCR targeting the human-specific D17Z1 sequence [19]. Real-time PCR was conducted on a Smart Cycler II System (Cepheid, USA), and standard curves were generated using a known concentration of human genomic HeLa DNA (TaKaRa, Japan) or male DNA extracted from male blood samples. Reactions were carried out in 25 μL, using 12.5 μL 2× SYBR Premix Ex Taq II, 0.4 μM each of forward and reverse primers, and 2 μL DNA. Real-time PCR conditions were as follows: initial denaturation at 95 °C for 30 s, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 15 s, and extension at 72 °C for 15 s. Male DNA was diluted to 0.25 ng/μL and used for optimization of the LAMP reaction. Female DNA was diluted to 5 ng/μL and used as negative control. LAMP sensitivity was evaluated by using DNA template at concentrations of 5 ng/μL, 500 pg/μL, 50 pg/μL, 5 pg/μL, 500 fg/μL, and 50 fg/μL purified DNA, which was extracted from male blood samples.

LAMP primer design and optimization of LAMP reaction

LAMP primers (M-F3, M-B3, M-FIP, and M-BIP) were designed using PrimerExplorer v5 (https://primerexplorer.jp) and the loop primer was designed manually. Primer information is listed in Table 1. The LAMP reaction without pyrophosphatase was performed using a Loopamp DNA Amplification Kit (Eiken, Japan). Reactions were carried out in 25 μL reaction mixture, using 12.5 μL of 2× reaction mix, 8 units of Bst DNA polymerase, 0.5 μL of EvaGreen (Biotium, USA), 0.2 μM each of M-F3 and M-B3 primers, 1.6 μM each of FIP and BIP primers, 0.8 μM of M-loop primer, and 2 μL of each genomic DNA template. Reactions were incubated at 65 °C for 90 min. All the LAMP reactions other than those using the Loopamp DNA Amplification Kit were performed using Isothermal Master Mix (Optigene, UK) as a pyrophosphatase-containing LAMP reagent. In a 25 μL total reaction mixture, 15 μL of the Isothermal Master Mix, 0.2 μM each of M-F3 and M-B3 primers, 1.6 μM each of FIP and BIP primers, 0.8 μM of M-loop primer, and 2 μL of each genomic DNA were incubated at 65 °C for 20 min. The LAMP reaction was performed on a Smart Cycler II system to observe fluorescence signals with wavelengths between 510 and 527 nm. The LAMP products amplified under several temperature conditions were evaluated by electrophoresis on a 1.5% agarose gel. The amplification time was determined from the cycle threshold value, which was determined with the second derivative maximum method. All the LAMP reactions were performed in triplicate, and the results were expressed as means ± standard deviations (SDs).

Table 1 LAMP primer sequences used in this study
Full size table

Comparison of LAMP assay with real-time PCR

Purified DNA was extracted with the EZ1 Investigator Kit from three male blood stains, which were prepared by spotting 2.5 μL of blood on cotton and drying for several days. DNA was dissolved in 50 μL of water and then 2 μL of 10-fold diluted DNA was used as the DNA template. Standard curves were made from DNA template of 5 ng/μL, 500 pg/μL, 50 pg/μL, and 5 pg/μL male DNA. DNA concentrations of the blood stains were measured with the LAMP assay and real-time PCR.

On-site screening protocol

Dried stains were made from blood and saliva samples of males, females, and a mixture of the two. For mixed samples, 10-fold diluted male fluid was mixed with undiluted female fluid in a ratio of 1:9. Fluid samples of 2.5 μL blood or saliva were spotted onto filter paper as approximately 5 × 5-mm2 stains and were dried for several days. DNA extraction from blood and saliva stains was performed with an Easy DNA Extraction Kit version 2 (Kaneka, Japan) [14]. Samples were added to tubes containing 88 μL solution A and heated at 98 °C for 8 min; 12 μL solution B was then added. The crude DNA (2 μL) was used as template in the LAMP reaction. For visual detection of the LAMP reaction, a dried colorimetric intercalating dye, D-QUICK (Kaneka), was dissolved in the reaction mix [18]. After the LAMP reaction at 65 °C for 20 min, the results were evaluated with visual observation and also evaluated by electrophoresis on a 1.5% agarose gel.

Validation of on-site screening protocol

Blood stains were prepared by spotting 5 μL of unprocessed male blood samples and serially diluted (10-, 100-,1000-, and 10,000-fold) male blood samples onto filter paper and drying for several days. Purified DNA dissolved in 100 μL of water was obtained using the EZ1 Investigator Kit. Crude DNA in a total volume of 100 μL was prepared using the Easy DNA Extraction Kit version 2. The purified or crude DNA (2 μL) was used as a template in real-time PCR and LAMP reactions with D-QUICK. PCR amplification of Y-STRs was carried out in reaction volumes of 25 μL using the AmpFlSTR™ Yfiler™ PCR Amplification Kit (Applied Biosystems) according to the manufacturer’s instructions. PCR products (1 μL) were added to 9 μL of 17:1 Hi-Di™ Formamide:GeneScan™ 600LIZ Size Standard v2.0. Electrophoresis of amplified PCR products and standards was performed on a 3130xl Genetic Analyzer platform (Applied Biosystems), and data were analyzed using GeneMapper ID-X software v1.4 (Applied Biosystems).

Results and discussion

LAMP primer design and optimization of LAMP reaction

LAMP primers were designed based on alphoid repeat regions (accession no. AF522078), consisting of approximately 5.7 kb with repeat sequences (Table 1) [20]. Using a Loopamp DNA Amplification Kit (LA) with 0.5-ng male DNA template, a typical amplification plot was observed, with amplification beginning at 55.3 ± 3.1 min (Table 2). To shorten the reaction time, a loop primer was added to the LAMP reaction and an Isothermal Master Mix (IM) was used instead of LA. The IM contains pyrophosphatase to remove pyrophosphates, which have inhibitory effects on DNA amplification. The optimized conditions with the IM and a loop primer allowed the amplification to start within 10 min (8.2 ± 0.2 min), which was shorter than other conditions, with the LA and a loop primer (27.7 ± 1.0 min) or with the IM without a loop primer (13.6 ± 0.3 min). Under these optimized conditions, we examined the LAMP reaction at different temperatures. Amplification of 0.5 ng male DNA occurred at all temperatures tested between 61 and 67 °C (Fig. 1). At 65 °C, the time to initial amplification (8.2 ± 0.2 min) was slightly shorter than at 61, 63, and 67 °C (9.7 ± 0.2, 8.7 ± 0.2, and 8.7 ± 0.2 min, respectively). No amplification occurred of 10 ng female DNA at any temperature. The specificity of the reaction was also confirmed by gel electrophoresis of the LAMP products. Ladder-like patterns, which are characteristic of LAMP products, were observed in male DNA samples but not in female DNA samples. In addition, amplification specificity was confirmed by detecting male DNA in the presence of excess female DNA (male/female DNA ratio was 1:2000; data not shown). These results showed that the LAMP reaction is highly specific to male DNA and the LAMP amplification occurred at a wide range of temperatures from 61 to 67 °C, eliminating the need for strict temperature control.

Table 2 Optimization of LAMP reaction
Full size table
Fig. 1
figure 1

Optimization of LAMP reaction temperature. a The LAMP assay was performed at 61, 63, 65, and 67 °C. Amplification profiles were obtained in samples with male DNA (0.5 ng). No amplification was observed in samples with female DNA (10 ng) at any temperature. b LAMP reactions were confirmed by electrophoresis. Positive reactions in samples with male DNA showed ladder-like patterns, which are typically observed in LAMP amplification products. In contrast, negative reactions in samples with female DNA showed no ladder-like patterns

Full size image

Sensitivity and evaluation of quantitative LAMP

To assess the sensitivity of the LAMP reaction, serial dilutions of male DNA (10 ng, 1 ng, 100 pg, 10 pg, 1 pg, and 100 fg) were used as templates (Fig. 2). Amplification occurred with 10 ng, 1 ng, 100 pg, 10 pg, and 1 pg of DNA templates; the times to initial amplification were 6.6 ± 0.1, 7.6 ± 0.1, 8.6 ± 0.0, 9.6 ± 0.1, and 11.2 ± 0.7 min, respectively. No amplification occurred with 100-fg DNA template. Standard curves were generated by plotting the means of initial amplification times against the log-transformed DNA template amounts. The correlation coefficient of this relationship in the range of 10 pg to 10 ng of template was 0.9998, indicating that the LAMP assay can be used as a quantitative method.

Fig. 2
figure 2

Sensitivity of LAMP reaction. Serial 10-fold dilutions of male DNA (10 ng, 1 ng, 100 pg, 10 pg, 1 pg, and 100 fg) were used as LAMP reaction templates. Amplification was observed with 10 ng, 1 ng, 100 pg, 10 pg, and 1 pg but not with 100 fg. In the range of 10 pg to 10 ng of DNA template, a correlation between the amount of DNA template and initial rise time was observed (R 2 = 0.9998)

Full size image

To evaluate the possibility of using the LAMP technique as a quantitative method, we quantified purified DNA extracted from three male blood stains by using the LAMP assay and real-time PCR (Fig. 3). DNA concentrations calculated based on the results of real-time PCR were 37 ± 7, 26 ± 4, and 24 ± 10 pg/μL. DNA concentrations calculated based on the results of the LAMP assay were 34 ± 19, 21 ± 9, and 27 ± 19 pg/μL. The quantification results from the LAMP assay were almost the same as those from real-time PCR, but the standard deviations of the LAMP assay were larger than those of real-time PCR. Further improvement in LAMP reagent and conditions may be required for increasing precision. The reaction time of the LAMP quantification was approximately 15 min, which was shorter than that of real-time PCR (approximately 40 min). Our results showed that the LAMP assay was able to approximately quantify DNA and was faster than conventional real-time PCR. These results indicate that the LAMP assay could be used as an ultra-rapid amplification-based DNA quantification method.

Fig. 3
figure 3

Evaluation of quantitative LAMP. Purified DNA extracted from three male blood stains was quantified using real-time PCR and the LAMP assay, each in triplicate. The means measured by the LAMP assay, albeit with large standard deviation values, were almost the same as those measured by real-time PCR

Full size image

On-site screening protocol for male-derived samples

To apply the LAMP assay to on-site screening for male-derived samples, we simplified two procedures, the DNA extraction and the detection of the LAMP reaction. DNA extraction using alkaline lysis and neutralization reagents was used to generate DNA template for the LAMP reaction. The extraction method required three steps (alkaline treatment, heating, and neutralization) and took 10 min. Detection of the LAMP reaction was performed using a colorimetric intercalating dye. In general, the LAMP reaction is visualized by the appearance of white turbidity due to magnesium pyrophosphate. Our optimized LAMP reaction cannot be evaluated by this detection method because of the reagent containing pyrophosphatase. Therefore, a colorimetric intercalating dye was added to the LAMP reaction and the results were evaluated visually. The LAMP reaction with a colorimetric intercalating dye exhibited a sensitivity equal to that observed with an intercalating dye (Online Resource 1). Using the simplified protocol, the LAMP reaction was carried out with DNA from male, female, and mixed blood stains (Fig. 4a). After a 20-min LAMP reaction, the solutions of male-derived and mixed samples turned blue, whereas those of female-derived samples remained colorless. The LAMP products were analyzed by agarose gel electrophoresis and showed typical ladder-like patterns in samples with male and mixed template, but not in samples with female template. The same results were obtained from saliva stains (Fig. 4b).

Fig. 4
figure 4

Simplified LAMP reaction protocol. Simplified protocol was carried out using blood stains (a) and saliva stains (b). The LAMP results were evaluated with the naked eye and by electrophoresis. Nos. 1–3 male-derived stains, nos. 4–6 female-derived stains, nos. 7–9 mixed stains of male and female. Mixed stains consisted of 0.25 μL of 10-fold diluted male fluids and 2.25 μL of crude female fluids. The LAMP solutions of male and mixed samples (nos. 1–3, nos. 7–9) turned blue after the LAMP reaction. In contrast, the LAMP solutions of female samples (nos. 4–6) remained colorless. The LAMP reaction was also confirmed by electrophoresis. Positive samples showed ladder-like amplification products but negative samples did not

Full size image

To obtain further validation of our on-site LAMP reaction protocol, the LAMP assay, real-time PCR, and Y-STR analysis were performed (Table 3). Purified DNA and crude DNA were extracted from serially diluted male blood stains using the EZ1 Investigator Kit and the alkaline lysis and neutralization reagents, respectively. Extraction efficiency of crude DNA was slightly higher than that of purified DNA when examined by both real-time PCR and LAMP assays. The results indicated that the crude DNA extraction technique was able to obtain sufficient quantities of DNA as a DNA-based screening method. On-site optimized LAMP reactions using crude DNA detected blood stains diluted up to 100-fold with reproducibility, and Y-STR analysis using purified DNA detected a full or partial profile of the 17 loci of the Y-filer DNA profile from up to 100-fold diluted blood stains. These results showed that the on-site LAMP assay could be useful as a screening method prior to conducting Y-STR analysis.

Table 3 Validation of on-site LAMP reaction protocol
Full size table

Our on-site protocol could be performed in 30 min from sampling to detection, with little hands-on time and employing a simple device. Because an on-site system to detect short tandem repeats and amelogenin has already been developed, the technique for on-site identification of male-derived samples is not by itself novel [21]. However, our LAMP assay has advantages in being able to be performed with a simple device and little hands-on time in a cost-effective manner.

In summary, we have established a 20-min LAMP assay targeting the Y chromosome that is more rapid than conventional LAMP and real-time PCR methods for sex identification. The LAMP assay is highly sensitive, as amplification occurred with as low as 1-pg DNA template. Our LAMP assay can be used quantitatively, because there was a correlation between the LAMP reaction time and DNA concentration. Furthermore, we established a protocol for on-site screening for male-derived samples by modifying DNA extraction and detection methods. The on-site protocol can be used in forensic cases in which numerous samples must be handled. This method could be applicable to forensic investigations of criminal cases such as sexual assaults.