Screening and Functional Verification of Selectable Marker Genes for Cordyceps militaris

.e selectable marker genes are necessary resistance genes for gene knockout, gene complementation, and gene overexpression in filamentous fungi. Moreover, the more sensitive the filamentous fungi are to antibiotics, the more helpful it is to screen the target transformants. In order to obtain the antibiotic (or herbicide) which can effectively inhibit the growth of Cordyceps militaris and verify the function of the corresponding resistance gene inC. militaris, the sensitivity of C. militaris to hygromycin and glufosinate ammonium was compared to determine the resistance gene that was more suitable for the screening of C. militaris transformants. .e binary vector of the selectable marker gene was constructed by combining the double-joint PCR (DJ-PCR) method and the homologous recombination method, and the function of the selectable marker gene in C. militaris was verified by the Agrobacterium tumefaciens-mediated transformation method. .e results showed that C. militaris was more sensitive to glufosinate ammonium than hygromycin..e growth of C. militaris could be completely inhibited by 250 μg/mL glufosinate ammonium..e expression cassette of the glufosinate ammonium resistance gene (bar gene) was successfully constructed by DJ-PCR. .e binary vector pCAMBIA0390-Bar was successfully constructed by homologous recombination. .e bar gene of the vector pCAMBIA0390-Bar was successfully integrated into the C. militaris genome and could be highly expressed in the transformants of C. militaris. .is study will promote the identification of C. militaris gene function and reveal the biosynthetic pathways of bioactive components in C. militaris.

ese ingredients endow C. militaris with anticancer [8], antitumor [9], antioxidation [10], and immunity enhancement [11] properties. Based on the above advantages, the demand for C. militaris has increased year by year. However, due to the low content of bioactive ingredients in C. militaris, it cannot meet the market demand, which leads to a high price of bioactive ingredients in C. militaris. erefore, scientists have been studying the biosynthetic pathways of C. militaris bioactive ingredients [2,12], identifying the functions of biosynthetic genes of bioactive ingredients [13,14] and aiming to improve the content of C. militaris bioactive ingredients by means of genetic engineering. Gene knockout [15,16], gene complementation [17], and gene overexpression [16,18] are often used to identify the gene function of filamentous fungi, but selectable marker genes are required in the above experiments. If strains are sensitive to antibiotics or herbicides, which is helpful for screening target mutants. e hygromycin resistance gene (hph gene) and the glufosinate ammonium resistance gene (bar gene) have been used as selectable marker genes for C. militaris [19,20]. At the same time, hygromycin and glufosinate ammonium have also been used to screen the transformants of C. militaris, but which antibiotic or herbicide is C. militaris more sensitive to? Up to now, the comparison of hygromycin and glufosinate ammonium on the growth inhibition of C. militaris has not been reported. erefore, the purpose of this study is to determine an antibiotic (or a herbicide) that is more effective in inhibiting the growth of C. militaris by comparing the inhibition of hygromycin and glufosinate ammonium on C. militaris. en, the double-joint PCR (DJ-PCR) method was used to construct an expression cassette of the selectable marker gene, and its binary vector was constructed by homologous recombination. Finally, Agrobacterium tumefaciensmediated transformation (ATMT) method was used to verify the effectiveness of the binary vector constructed in this study and the function of the related resistance gene in C. militaris. is study will contribute to the identification of C. militaris gene function and the elucidation of the biosynthetic pathways of bioactive components in C. militaris.

Strains and Vectors. Escherichia coli
DH5α was used as a host for the construction of the vector pCAMBIA0390-Bar. e C. militaris strain CM10 (GIM5.271) was used to verify the function of the bar gene. e vector pCAMBIA3301 contains the bar gene. e vector pAg1-H3 contains the trpC promoter (PtrpC) and the trpC terminator (TtrpC) derived from Aspergillus nidulans. e binary vector pCAM-BIA0390-Bar was reconstructed based on the binary vector pCAMBIA0390 (GenBank: AF234291.1).

Construction of the Bar Gene Expression Cassette.
According to the results of the concentration gradient test, glufosinate ammonium could completely inhibit the growth of C. militaris. Because the bar gene is the glufosinate ammonium resistance gene, the bar gene expression cassette needs to be constructed in this study. e primer pairs F1/R1 and F3/R3 were used to amplify the promoter PtrpC and the terminator TtrpC from the vector pAg1-H3, respectively. e primer pair F2/R2 was used to amplify the bar gene from the vector pCAMBIA3301. Primers F1 and R3 were used to fuse the promoter PtrpC, the bar gene, and the terminator TtrpC by the DJ-PCR method to obtain the fusion fragment PtrpC-bar-TtrpC (Supplementary Figure S1 and Supplementary Table S1).

Construction of the Binary Vector pCAMBIA0390-Bar.
In order to construct the binary vector pCAMBIA0390-Bar, the restriction endonuclease Hind III and EcoR I was used to digest the vector pCAMBIA0390, and the linear vector pCAMBIA0390 was obtained. e 5′ and 3′ ends of fusion fragment (PtrpC-bar-TtrpC) are homologous to the two ends of the linear vector pCAMBIA0390, respectively. e One Step Cloning Kit (Vazyme Biotech Co., Ltd, Nanjing, China) containing homologous recombinase was used to homologous recombine the fusion fragment (PtrpC-bar-TtrpC) and the linear vector pCAMBIA0390 to obtain the recombinant vector pCAMBIA0390-Bar (Supplementary Figure S2). e reaction system and reaction conditions of homologous recombination were carried out according to the manufacturer's protocol. e fusion fragment (PtrpCbar-TtrpC) was inserted into the transfer DNA (T-DNA) region of the vector pCAMBIA0390.

Functional Verification of the Bar Gene in C. militaris.
To verify the effectiveness of the vector (pCAMBIA0390-Bar) constructed in this study and the function of the bar gene in C. militaris, the vector pCAMBIA0390-Bar was firstly transformed into the competent cells of the Agrobacterium tumefaciens AGL1, and the Agrobacterium tumefaciens strain AGL1-pCAMBIA0390-Bar was obtained.
en, the conidia of the wild-type C. militaris strain CM10 were collected according to the method described by Lou et al. [27]. According to the ATMT method reported by Zheng et al. [21], the conidia of the wild-type C. militaris strain CM10 and the Agrobacterium tumefaciens strain AGL1-pCAMBIA0390-Bar were cocultured in order to integrate the bar gene expression cassette into the genome of the wild-type C. militaris strain CM10. e selective medium contains 300 μg/mL glufosinate ammonium and 300 μg/mL cefotaxime. e transformants of C. militaris on the selective medium were selected and inoculated to a new PDA medium containing 300 μg/mL of glufosinate ammonium for resistance screening. If transformants of C. militaris can grow on the PDA medium containing 300 μg/mL of glufosinate ammonium, they are considered to be resistant to glufosinate ammonium. e genome of C. militaris transformants with resistance to glufosinate ammonium was extracted. Primers F4 and R4 were used to amplify an 836 bp fragment of the bar gene expression cassette from the genome of C. militaris transformants. e PCR products were sequenced to determine whether the bar gene expression cassette was integrated into the genome of the wild-type C. militaris. e function of the bar gene in C. militaris was further verified by quantitative real-time PCR (qRT-PCR) using the tef1 gene (GenBank: DQ070019) as the internal control gene [28]. e PCR-positive mutant of C. militaris and the wildtype C. militaris were inoculated on the PDA medium and cultured at 25°C. e mycelia of the PCR-positive mutant and the wild-type C. militaris were obtained by culturing for 20 days under dark conditions and then culturing for 7 days under light conditions. e Fungal RNA Kit (Omega, Stamford, CT, USA) was used to extract the total RNA of C. militaris mycelia, and the TransScript All-in-One First-Strand cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China) was used for reverse transcription. e primers tef1-F/tef1-R and JCBar-F/JCBar-R were used to detect the expression levels of the tef1 gene and the bar gene, respectively (Supplementary Table S1). e qRT-PCR experiment was performed using the TransStart Tip Green qPCR SuperMix (TransGen Biotech, Beijing, China) with a QuantStudio 3 Real-Time PCR System ( ermo Fisher Scientific, MA, USA) according to the method described by Lou et al. [13]. e relative expression level of the bar gene relative to the tef1 gene was calculated using the 2 −∆∆CT method [29]. e qRT-PCR experiment was repeated three times.

Inhibition of Hygromycin and Glufosinate Ammonium on
the Growth of C. militaris. C. militaris cultured on the PDA medium without hygromycin and glufosinate ammonium grew well, and the colony diameter of C. militaris was larger (Figures 1 and 2). e colony diameter of C. militaris cultured on the PDA medium containing hygromycin was smaller than that of C. militaris cultured on the PDA medium without hygromycin. With the increase of hygromycin concentration, the colony diameter of C. militaris decreased gradually (Figure 1), which indicated that hygromycin could inhibit the growth of C. militaris. Previous studies showed that 400-800 μg/mL hygromycin could completely inhibit the growth of C. militaris [21][22][23], but 2000 μg/mL hygromycin still could not completely inhibit the growth of the C. militaris strain CM10 in this study, which might be caused by the difference of C. militaris strains. e C. militaris strain CM10 used in this study was less sensitive to hygromycin. erefore, hygromycin was not suitable as an antibiotic to screen the transformants of the C. militaris CM10.
With the increase of glufosinate ammonium concentration in the PDA medium, the colony diameter of C. militaris decreased significantly (Figure 2). When the concentration of glufosinate ammonium in the PDA medium reached 250 μg/mL, the growth of C. militaris was completely inhibited, which indicated that C. militaris was sensitive to glufosinate ammonium. erefore, glufosinate ammonium was suitable as a herbicide for screening C. militaris transformants.

Construction of the Bar Gene Expression Cassette.
e extracted vectors pAg1-H3, pCAMBIA3301, and pCAM-BIA0390 are shown in Figure 3. e promoter PtrpC (399 bp), the bar gene (582 bp), and the terminator TtrpC (773 bp) were obtained by PCR amplification. e promoter PtrpC, the bar gene, and the terminator TtrpC were successfully fused by the DJ-PCR method, and the fused bar gene expression cassette (1694 bp) was obtained (Figure 3).

Construction of the Binary Vector pCAMBIA0390-Bar.
e restriction endonuclease Hind III and EcoR I were used to digest the vector pCAMBIA0390, and the linear vector pCAMBIA0390 was obtained (Figure 3). e bar gene expression cassette and the linear vector pCAMBIA0390 were successfully homologous recombined by the homologous recombinase, and the binary vector pCAMBIA0390-Bar was obtained (Figure 3).

Functional Verification of the Bar Gene in C. militaris.
e conidia of C. militaris were round or oval (Figure 4(a)), which was consistent with previous studies [27,30]. Many white colonies of C. militaris appeared on the selective medium containing 300 μg/mL glufosinate ammonium and 300 μg/mL cefotaxime (Figure 4(b)). e colony of the C. militaris transformant picked from the selective medium grew well on the PDA medium containing 300 μg/mL of glufosinate ammonium, while the growth of the wild-type C. militaris cultured on the PDA medium containing 300 μg/ mL of glufosinate ammonium was completely inhibited, which indicated that the transformant of C. militaris picked from the selective medium had glufosinate ammonium resistance (Figure 4(c)). e results of amplification of the bar gene from C. militaris transformants with resistance to glufosinate ammonium are shown in Figure 4(d). e sequencing results of PCR products confirmed that the bar gene was successfully integrated into the genome of C. militaris. e wild-type C. militaris and its transformants grew well on the PDA medium without glufosinate ammonium (Figures 5(a) and 5(b)). However, the colony diameter of the C. militaris transformant was smaller than that of the wildtype C. militaris, which indicated that the insertion of the bar gene reduced the growth rate of the C. militaris transformant. In addition, the color difference between the wildtype C. militaris and its transformant might be due to the mutation of genes related to pigment synthesis caused by the insertion of the bar gene. e results of qRT-PCR suggested that the expression of the bar gene were not detected in wildtype C. militaris, which indicated that there was no bar gene in the wild-type C. militaris. However, the relative expression level of the bar gene in the C. militaris transformant was relatively high, which demonstrated that the bar gene expression cassette and the binary vector pCAMBIA0390-Bar constructed in this study could be effectively applied to C. militaris ( Figure 5(c)). Furthermore, the bar gene can be used as a selectable marker gene for C. militaris.

Discussion
Selectable marker genes are indispensable for gene knockout, gene complementation, and gene overexpression. It is well known that the workload of screening target mutants is very large. If the strain is sensitive to antibiotics, the corresponding resistance gene can effectively reduce the workload of screening target mutants. If the strain is less sensitive to antibiotics, a lot of work is needed to screen target mutants. e hph gene and the bar gene have been used as selectable marker genes for C. militaris, but the comparison of the sensitivity of C. militaris to hygromycin and glufosinate ammonium has not been reported. In this study, the inhibitory effects of hygromycin and glufosinate ammonium on the growth of C. militaris were compared for the first time. e bar gene expression cassette was constructed by the DJ-PCR method, and then the binary vector pCAMBIA0390-Bar was constructed by homologous recombination. Furthermore, the bar gene was successfully integrated into C. militaris genome by the ATMT method and the transformant of C. militaris with resistance to glufosinate ammonium was obtained. e bar gene expression cassette constructed in this study could be efficiently expressed in C. militaris transformants.
Hygromycin is an aminoglycosidic antibiotic that inhibits protein synthesis by disrupting translocation and promoting mistranslation at the 80S ribosome [31]. e hygromycin phosphotransferase encoded by the hph gene can convert hygromycin into phosphorylated products that are not biologically toxic [31]. Based on this, hygromycin has been used as an antibiotic to screen the transformants of C. militaris, and the hph gene has also been used as a selectable marker gene for C. militaris [21,23,24]. However, in this study, 2000 μg/mL hygromycin still failed to completely inhibit the growth of the C. militaris strain CM10, which indicated that the C. militaris strain CM10 was less sensitive to hygromycin. Glufosinate ammonium strongly inhibits the activity of glutamine synthetase (GS) that can convert NH + 4 to glutamine. e presence of glufosinate ammonium causes cells accumulate NH + 4 , which leads to cell death or growth inhibition. Phosphinothricin acetyltransferase encoded by the bar gene can acetylate the free amino groups of glufosinate ammonium, which makes glufosinate ammonium unable to inhibit the activity of GS [32]. erefore, glufosinate ammonium has been used as herbicide for screening the transformants of C. militaris, and the bar gene has also been used as a selectable marker gene for C. militaris [26]. Furthermore, 250 μg/mL glufosinate ammonium could completely inhibit the growth of the C. militaris strain CM10 in this study. In addition, we also conducted a concentration gradient test on C. militaris strains CM11 and CM12 and found that 2000 μg/mL hygromycin could not completely inhibit the growth of C. militaris strains CM11 and CM12, while 250 μg/mL glufosinate ammonium could completely inhibit the growth of C. militaris strains CM11 and CM12. erefore, we believe that C. militaris is more sensitive to glufosinate ammonium than hygromycin. Xiong et al. believed that the concentration of glufosinate ammonium to completely inhibit the growth of C. militaris was 150 μg/mL [25], while Wang et al. suggested that the concentration of glufosinate ammonium to completely inhibit the growth of C. militaris was 400 μg/mL [23], but they were inconsistent with the results of this study, which might be due to the different sensitivity of different C. militaris strains to glufosinate ammonium. Based on this, we believe that it is necessary to conduct a concentration gradient test of glufosinate ammonium to inhibit the growth of C. militaris before the transformation experiment, so as to determine the minimum concentration of glufosinate ammonium that completely inhibits the growth of C. militaris. Furthermore, we believe that an appropriate increase in the concentration of glufosinate ammonium can reduce the appearance of false-positive transformants and the workload of screening target transformants. erefore, 300 μg/mL of glufosinate ammonium was used to screen the transformants of C. militaris in this study.
ree DNA fragments (such as the promoter PtrpC, the bar gene, and the terminator TtrpC) could be fused in two steps by DJ-PCR [33]. However, it was required that the 5′ end sequences of primers R1 and R2 were homologous with the 5′ end sequences of the bar gene and the terminator TtrpC, respectively, and the 5′ end sequences of primers F2 and F3 were homologous with the 3′ end sequences of the promoter PtrpC and the bar gene, respectively. e length of homologous sequence was about 15 bp. It was worth noting that PCR products needed to be purified after amplifying the three elements (the promoter PtrpC, the bar gene, and the terminator TtrpC). In this way, the influence of primers and DNA templates of the first round of PCR on DJ-PCR could be avoided. In order to avoid base mutation, it was necessary to use high fidelity DNA polymerase and reduce the cycle times of PCR amplification [34].
When restriction endonucleases were used to prepare linear vectors, it was necessary to purify the digestion products to reduce the occurrence of false positive vectors. In the construction of knockout vectors, two homologous sequences are often inserted on both sides of the resistance gene expression cassette. In the construction of a complementary vector or an overexpression vector, one side of the resistance gene expression cassette needs to be inserted with a complementary gene or an overexpression gene. We believed that it was necessary to retain the restriction sites on both sides of the resistance gene expression cassette during the construction of a vector. erefore, the Hind III and EcoR I sites were retained on both sides of the bar gene expression cassette when the vector pCAMBIA0390-Bar was constructed. Because the constructed vector pCAM-BIA0390-Bar could be effectively applied to C. militaris, so it can be used to construct target gene knockout vectors, complementary vectors, and overexpression vectors.

Conclusions
e sensitivity of C. militaris to hygromycin and glufosinate ammonium was compared for the first time. It was found that C. militaris was more sensitive to glufosinate ammonium, and glufosinate ammonium was more suitable as herbicide for screening the transformants of C. militaris. 250 μg/mL glufosinate ammonium could completely inhibit the growth of C. militaris. e bar gene expression cassette and the binary vector pCAMBIA0390-Bar were efficiently constructed by the DJ-PCR method and the homologous recombination method, respectively. e bar gene of the vector pCAMBIA0390-Bar was successfully integrated into the genome of C. militaris by the ATMT method, and C. militaris transformants with resistance to glufosinate ammonium were obtained. It was confirmed that the bar gene was more suitable as a selectable marker gene for C. militaris. is study will promote the identification of the functions of C. militaris genes and reveal the biosynthetic pathways of bioactive components in C. militaris.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
All authors declare no conflicts of interest.