接第一篇

2. Results

2.1. Species Identification of Fungal Materials and Analysis of the nrDNA-ITS Sequences

　　To confirm the species identities and verify sequence variabilities, the complete nrDNA-ITS regions were amplified from 18 samples of six species(Table 1). DNA fragments were abundantly amplified from all templates and they had expected sizes of approximately 600 bp. The lengths of these PCR amplicons, determined after sub-cloning into the pGEM-T Easy vector, were 567 bp for C. militaris, 583 bp for C. pruinosa, 587 bp for both I. cicadae and I. tenuipes, 580 bp for O. sinensis, and 569 bp for B. bassiana (Table 2). Taxonomic origins of these 18 samples were confirmed by performing Basic Local Alignment Search Tool (BLAST) searches of the obtained sequences in both GenBank and BOLD databases. The data revealed that four of the fungal strains had been misidentified. Specifically, two presumed I. tenuipes samples, KACC43335 and KACC43338, were identified asbeing I.cicadae and B. bassiana, respectively, and two presumed I. cicadae samples, KACC44476 and KACC43334, were identified as being I. tenuipes and B.bassiana, respectively (Table 1). However, all other sample sequences matched their corresponding expected species, sharing over 99% sequence identity with their expected species. We also further confirmed the species of the fungal materials by an analysis of phylogenetic trees using 31 samples of nrDNA-ITS sequences obtained in the current study,and 13 that were retrieved from NCBI GenBank (Table 1, Figure S1, and Materials andMethods).A neighbor-joining analysis was used to classify the 31 nrDNA-ITS sequences into six distinct clades (C. militaris, C. pruinosa, I. cicadae, I. tenuipes, O. sinensis, and B. bassiana) and four genus cluster groups (Cordyceps, Isaria, Ophiocordyceps, and Beauveria) (Figure S1). These results strongly demonstrated that KACC13335, KACC44476, and KACC43334 and KACC43338 were I. cicade, I. tenuipes, and B. bassiana,respectively (Figure S1).

　　The nrDNA-ITS regionswere compared using themultiple ClustalWtool in the BioEdit program[33].The aligned sequences were 611 bp in length and had 56.06–62.24% GC content. Intraspecific sequence variability was only observed in two species, C. militaris and O. sinensis, at 0.0021  0.0013 and 0.0023  0.0020, respectively (Table 2). Interspecific sequence variability was between 0.0931  0.0691 and 0.2136  0.0095, with I. cicadae and O. sinensis harboring the most similar and the most diverged nrDNA-ITS sequences among the species, respectively (Table 2).

2.2. Development of the SCAR Markers and the Real-Time PCR Assays

　　To identify the optimal species-specific SCAR primers, the entire nrDNA-ITS sequence was analyzed, comparing the positions of nucleotide substitutions and indels between the species.Based on these nucleotide variabilities, we prepared several candidate SCAR primers and verified their specificities. The primer sets CM F2/CM R2 and CM F3/CM R3, designed to amplify 339-bp and 102-bp length C. militaris-specific amplicons, respectively, yielded DNA products only in the four samples that had been identified as C. militaris. No PCR products were obtained with the remaining 12 templates of the four related species, C. pruinosa, I. cicadae, I. tenuipes, and O. sinensis (Tables 1 and 3, and Figure S2 and Figure 1A). These observations indicated that the two primer sets could be used to distinguish C. militaris, and medicinal and food materials containing C. militaris, from the four related species.

　　Two other primer sets, CP F2/CP R2 and CP F4/CP R3, yielded respective unique PCR amplicons of expected lengths (244 bp and 83 bp, respectively) from only the three C. pruinosa samples (Tables 1 and 3, and Figure S2 and Figure 1B). Similarly, the two primer sets for each of the other three species(I. cicadae, I. tenuipes, and O. sinensis) yielded unique PCR products of expected sizes only from their respective target species, with no cross-reactivity with other species. These were the primer sets IC F1/IC R1 and IC F3/IC R3 for I. cicadae (337 bp and 139 bp amplicons, respectively), IT F4/IT R3,and IT F3/IT R2 for I. tenuipes (132 bp and 107 bp amplicons, respectively), and OS F1/OS R2 and OS F3/OS R3 for O. sinensis (200 bp and 117 bp amplicons, respectively) (Tables 1 and 3, and Figure S2 and Figure 1C–E). These five sets of species-specific SCAR markers, each specific to two different target regions, can therefore be used to identify Cordyceps-related fungal species with high discriminability and stability, more efficiently and successfully than methods based on single SCAR markers. 

　　To verify the sensitivity and detection limits of the SCAR marker-based conventional PCR assay for the five species, serial 10-fold dilutions of pure genomic DNA (gDNA) (15 fg/L to 15 ng/L) were amplified using the respective SCAR primer sets (Figure 2). The amplification products with SCAR markers specific to C. militaris, I. cicadae, and O. sinensis, and to I. tenuipes were successfully obtained by conventional PCR in reactions containing ca. 150 fg and 15 pg of gDNA, respectively (Figure 2).

Figure 1. Development of the SCAR markers based on sequence variations in the nrDNA-ITS region.

(A) Verification of primer specificities for C. militaris. 

(B) Verification of primer specificities for C.pruinosa. 

(C) Verification of primer specificities for I. cicadae. 

(D) Verification of primer specificities for I. tenuipes. 

(E) Verification of primer specificities for O. sinensis. 

The numbers 1–16 correspond to those listed in Table 1 in the “Gel lane” column. The precise lengths of the PCR products and DNA ladders are indicated to the right and left of the gel images, respectively. M, 100 bp DNA ladder.

Figure 2. Verification of the detection limit of the SCAR markers using serial dilutions of template DNA. The precise lengths of the PCR products and the DNA ladders are indicated to the right and left of the gel images, respectively. M, 100 bp DNA ladder.

　　Standard curves of the species-specific SCAR markers and serial dilutions of the respective templates were used to establish a real-time PCR assay. To confirm the primer specificities for the real-time PCR assay, 15 ng of gDNA from the other four species were tested as non-target DNA; no cross-amplification was observed in any of the real-time PCR assays (Figure S3). The standard curves revealed high amplification efficiency and data linearity (Table 4 and Figure S3). The slopes of the standard curves were between -3.343 and -2.757, with the correlation coefficients being between 0.9587 and 0.9999 (Table 4). The Ct values were 20, 24, 28, 24, and 21 cycles for C. militaris, C. pruinosa,I. cicadae, I. tenuipes, and O. sinensis, respectively (Table 4). The sensitivities (LOD, limit of detection) of the real-time PCR assays were below 1.5 pg for all species (Figure S3).

2.3. Verification of the SCAR Markers and the Real-Time PCR Assay Using Commercial Products

　　Verification of the reproducibilities and discriminabilities of the SCAR markers developed in the current study was performed using 17 commercial herbal medicines and dietary supplements (Table 5). Thirteen samples of Cordyceps-related material, including seven dried herbal medicines, two fresh fruiting bodies, two dried powders, two mixed pill-type dietary supplements, and four dried Cordyceps herbal medicines, were purchased at markets in Korea, China, and Bhutan (Table 5). Of these samples, most products were labeled only by a common name, namely, Dong Chung Ha Cho or Cordyceps, without any specification of origin at the species level. Using the duplex SCAR markers and a real-time PCR assay, five out of the 11 samples that lacked species specification on the product label were authenticated as C. militaris-, two as I. tenuipes-, and four as O. sinensis-derived products (Table 5 and Figure 3). Further, one herbal medicine sample (voucher no. 2-2016-F020 in Table 5), that was labeled as a Paecillomyces japonica (a synonym of I. japonica) product, was also identified as I. tenuipes by both conventional and real-time PCR (Table 5 and Figure 3). For the other five products, the identities of the species specified on the product labels were confirmed(Table 5 and Figure 3). Interestingly, the original species in pill-type dietary supplements (Voucher nos. 2-2016-F024 and 2-2016-F024) that was presented as a mixture of diverse plant materials and cultured mycelium powder, were also identified as I. tenuipes and C. militaris, respectively (Table 5 and Figure 3). These observations indicated that the two sets of SCAR markers and the real-time PCR assay established in the current study could be used for the identification of the five Cordyceps and the related species, as the method was able to distinguish economically motivated adulterants from authentic Cordyceps in both commercially processed products and herbal medicines.

Figure 3. Validation of SCAR marker discriminability, and authentication of the taxonomic origin of resources in commercial Cordyceps and its related products. 

(A) Outcome of PCR amplification with the C. militaris-specific SCAR markers. 

(B) Outcome of PCR amplification with the C. pruinosa-specific SCAR markers. 

(C) Outcome of PCR amplification with the I. cicadae-specific SCAR markers. 

(D) Outcome of PCR amplification with the I. tenuipes-specific SCAR markers. 

(E) Outcome of PCR amplification with the O. sinensis-specific SCAR markers.

(F) Outcome of PCR amplification with the nrDNA-ITS primers (ITS1 and ITS4) as control. Numbers 1–17 in the commercial products correspond to those listed in Table 5. 

The precise lengths of the PCR products and the DNA ladders are indicated to the right and left of the gel images, respectively. M, 100 bp DNA ladder.