Photo morphogenesis and photo response of the blue-light receptor gene Cmwc-1 in different strains of Cordyceps militaris


  • Tao Yang,

    1. State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, Beijing, China
    2. University of Chinese Academy of Sciences, Beijing, China
    Search for more papers by this author
  • Caihong Dong

    Corresponding author
    1. State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, Beijing, China
    • Correspondence: Caihong Dong, State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, NO.3 1st Beichen West Road, Chaoyang District, Beijing 100101, China. Tel.: +86-010-64806138; fax: +86-010-64807468;


    Search for more papers by this author


Light is a necessary environmental factor for stroma formation and development of Cordyceps militaris, a well-known edible and medicinal fungus. In this study, photo morphogenesis and the blue-light receptor gene were studied using five representative strains of C. militaris. The results suggest that light was essential for colony pigmentation and could promote conidia production. Cmwc-1, the homologe of the blue-light photoreceptor of Neurospora crassa, was cloned from the genome of C. militaris by Hi-tail PCR. The protein CmWC-1 was characterized by the presence of the LOV and PAS domains and a GATA-type Znf domain. Genetic variation analysis of Cmwc-1 in different strains showed that 15-bp deletions occurred in three strains that resulted in 5-Gln deletions in the transcription activation domain. Phylogenetic analysis based on the Sordariomycetes WC-1-like proteins suggested that the sequence of WC-1 could be used as a candidate marker for phylogenetic analysis in fungi. Cmwc-1 mRNA was light inducible and the expression level increased significantly after irradiation in all tested strains. The sequence of CmWC-1 and the relative expressions responding to irradiation in degenerate and albino strains were similar as the cultivated one. This report will help to open the still-unexplored field of stroma development for this fungus.


Cordyceps militaris (L.) Link, a well-known edible and medicinal fungus, is the type species of Cordyceps that generally parasitizes the larva or pupa of lepidopteron insects. Recent studies have shown its multiple pharmacological effects, including antitumor (Jin et al., 2008), anti-influenza virus (Ohta et al., 2007) and immunomodulatory effects (Kim et al., 2007). The use of this fungal product in medicinal treatment and health foods has become popular in many countries, and it has been listed as a Novel Food by the Ministry of Health of the People's Republic of China.

Currently, the fruit bodies of this fungus have been successfully cultivated and commercialized. Light is the most important environmental factor affecting C. militaris stroma production (Sato & Shimazu, 2002). After exposure to light, the color of the colony changes from white to yellow or orange, and then, the primordia begin to develop. No stromata are produced in darkness.

Similar to other mushrooms, C. militaris frequently degenerates during continuous maintenance in culture by showing a loss in the ability to reproduce sexually or asexually (Lin et al., 2010). Characteristics of degenerate strains included the loss or reduction of culture pigmentation and fruit body formation after light irradiation. Strain degeneration is the main problem in C. militaris cultivation and can lead to great commercial losses. These problems prompted us to study the influence of light on the growth and development of this fungus, as well as its light transduction at the molecular level.

Several fungi sense light as a signal of morphogenesis. The best-characterized model is Neurospora crassa, whose blue light receptor, White Collar-1 (NcWC-1), has been identified (Ballario et al., 1996) and characterized (Froehlich et al., 2002). NcWC-1 contains a light sensor domain called LOV (Light, Oxygen, or Voltage) and is a member of the PAS (Per-Arnt-Sim) super family (Ponting & Aravind, 1997; He et al., 2002). As a transcription factor, NcWC-1 is an essential component of all known blue light responses, including mycelia carotenogenesis, phototropism of perithecial beaks and circadian rhythm of conidiation (Ballario et al., 1998). The photoreceptor orthologs BLR1 and BLR2 are known to mediate nearly all known light responses in species of the genus Trichoderma (Schmoll et al., 2010). The blue light receptor complex, WC-1/2, is involved in the fruit body formation of Schizophyllum commune (Ohm et al., 2012), and similar results were obtained in Coprinopsis cinerea (Kamada et al., 2010) and Lentinula edodes (Sano et al., 2007).

Increased attention has been paid to C. militaris, a valued edible fungus. In this study, we analyzed the influence of light on the growth and conidiation of this mushroom using five representative strains. Then, we cloned the full-length gene of Cmwc-1 from C. militaris. CmWC-1 was possibly involved in the photoreaction and was characterized by the presence of the LOV and PAS domains and a GATA-type Znf domain. The variation of the different strains was subsequently analyzed. Finally, we also briefly described the expression of the Cmwc-1 transcript after irradiation in different strains.

Materials and methods

Fungal strains

The strains CGMCC 5.699, 3.16321, 3.16322, 3.16323 and 3.16324 that were used in this study were maintained on potato dextrose agar at 4 °C as stocks. The identification of the strains was confirmed by morphological and molecular methods.

Genomic DNA was prepared using the cetyltrimethylammonium bromide (CTAB) method (Doyle & Doyle, 1987). Primers MAT1-1-1F/R, MAT1-1-2F/R and MAT1-2-1F/R (Supporting information, Table S1) were used to identify the mating type by PCR.

Fruit body characteristics of different strains

The fruit bodies of the five strains were cultivated according to the method of Zhan et al. (2006). The colors were observed and the lengths of fruit bodies were measured after 60 days of cultivation. Six biological replicates were used for the determination of the height of fruit bodies. Microscopic observations were carried out and photographed using a Zeiss Axioscope microscope (Zeiss, Welwyn Garden City, UK).

Influence of light on the growth and conidiation of different strains

Vegetative mycelia of C. militaris on solid medium were exposed to alternating, 12-h intervals of dark and light (including white and blue light, defined here as WL and BL) or kept in the dark (D). The blue light was produced by Samvol power 12-W LEDs (Light Emitting Diodes, Zhongshan, China). The distance between the LEDs and the agar plates was 50 cm. The growth of each colony was measured after 2 weeks of culture. Three biological replicates were used.

Conidiation was assessed using cultures grown for 14 days on PDA medium. Mycelia were scraped from the plate, resuspended in 10 mL of Tween 80 solution (20% w/v) and filtered. The conidia were counted using a Thoma chamber following appropriate dilution. The presented data represent six biological replicates with two technical replicates each.

Amplification of the Cmwc-1 gene using degenerate primers and genome walking with Hi-tail PCR

The degenerate primer set Cmwc1d-F and Cmwc1d-R (Table S1) was designed based on the conserved regions of the known fungal blue light receptors. The PCR mixture contained 12.5 μL 2 × Es Taq Master Mix (CWBIO, Beijing, China), 2 μL of each degenerate primer (10 mmol L−1), 1 μL of genomic DNA and 7.5 μL ddH2O in a total volume of 25 μL. PCR amplification was performed using a thermo cycler (Bio-rad) as follows: initial denaturing at 94 °C for 3 min; followed by 30 cycles of 95 °C for 30 s, 45 °C for 30 s, and 72 °C for 1 min; and a final extension at 72 °C for 10 min.

Hi-tail PCR was used to characterize the sequences flanking the 3′ and 5′ ends of the partial Cmwc-1 gene. The cycling conditions were followed according to the protocol of Liu & Chen (2007).

Cloning, sequencing and analysis of PCR products

The resulting PCR products were cloned into vector Ultra Power pUM-T following recommendations of the supplier (Bioteke Corporation, China) and transformed into Escherichia coli DH5α cells. The nucleotide sequences were sequenced by Beijing Genomics Institute (Beijing, China) and were assembled using the ContigExpress program (Invitrogen, Carlsbad, CA). The nucleotide sequences were translated into protein by expasy. Conserved domain analysis was performed by smart ( The molecular weights and isoelectric points (pI) were calculated using the protparam tool (

Phylogenetic analysis

Deduced amino acid sequences were aligned using clustalw and manually refined using bioedit 7.0. Models of sequence evolution were chosen using prottest (Darriba et al., 2012), and maximum likelihood analyses were performed using phyml (Guindon et al., 2010) by running 500 bootstrap replicates. Bayesian inference was performed using mrbayes 3.1.2 (Huelsenbeck & Ronquist, 2001) for 1000 000 generations with running four chains, sampling every 100th tree and a burnin value of 2500. The phylogenetic tree was drawn using figtree v.1.4.0. The blue light receptors of Sclerotinia sclerotiorum (XP001586924.1) and Botryotinia fuckeliana (XP001547999.1) were used as the out groups. The protein sequences used are listed in Table S2.

Photo response of Cmwc-1 in Cordyceps militaris

Strain CGMCC 3.16222 was grown on PDA medium in the dark for 14 days, and irradiation was conducted for 15, 30, 45 and 60 min with an intensity of 500 lux. The mycelia were collected and kept at −80 °C before RNA extraction.

Total RNA was isolated from 100 mg of frozen mycelia using the TRIzol reagent (Invitrogen) and was then treated with RQ1 RNase-free DNase (Promega). cDNA synthesis was achieved using the ReverTra Ace qPCR RT Master Mix (Toyobo CO., LTD, Japan), and quantitative real-time PCR (qPCR) was performed using the Master cycler ep realplex (Eppendorf, Germany) real-time PCR System. The 25 μL qPCR reactions contained 5 ng cDNA, 0.1 μM primers and 12.5 μL QPCR SYBR Green Mix (Toyobo CO., LTD, Japan). The thermal cycling conditions were as follows: 95 °C for 1 min; followed by 40 cycles of 15 s at 95 °C, 15 s at 58 °C and 45 s at 72 °C; and 58 °C for 1 min, 58 °C to 95 °C over 20 min, and 15 s at 95 °C for the dissociation curve analyses. The relative gene expression was calculated using the 2–∆∆CT method (Livak & Schmittgen, 2001).

The expression levels of the different strains were compared under dark conditions and irradiation for 15 min. The presented data represent three biological replicates with two technical replicates each.

Statistical analysis

All the data obtained from this investigation were analyzed by one-way analysis of variance (anova). Tests of significant differences were determined by Tukey's HSD tests (P = 0.05) using spss 10.0 (SPSS Inc.).


Characterization of strains used in the study

Five strains were chosen for the study. Cordyceps militaris is heterothallic (Zheng et al., 2011) with two mating types, i.e., Mating 1-1 and 1-2. Among the five strains, CGMCC 3.16321 and 3.16324 contained only Mating 1-1, while the others contained MAT1-1 and MAT1-2 loci (Fig. S1).

The characterizations of fruit bodies from different strains were summarized in Table 1. CGMCC 5.699 was the degenerate strain that could not produce fruit bodies, and CGMCC 3.16324 was the albino mutant with white fruit bodies. The morphology of the fruit bodies varied distinctly among the other three strains. No perithecia formed in the stroma of strain CGMCC 3.16321, whereas distinct perithecia formed in those of strains CGMCC 3.16322 and 3.16323 (Fig. S2). The fruit body of strain CGMCC 3.16322 was significantly longer than that of CGMCC 3.16323.

Table 1. Characterization of strains used in the study
Strains (CGMCC)Mating typeFruit bodies
ColoraHeight (cm)bPerithecium
  1. a

    Color based on Kornerup & Wanscher (1978): light yellow (3A4, 5), orange (5A6, 7) and white (1A1).

  2. b

    Values are presented as the mean ± SD (n = 6), and values followed by the same letters are not significantly different by the anova and Tukey's HSD tests (P < 0.05).

  3. c

    No fruit body formation (degenerate strain)

5.699Both MAT1-1 and MAT1-2c
3.16321MAT1-1Orange7.13 ± 1.01 aNo perithecium
3.16322Both MAT1-1 and MAT1-2Light yellow6.25 ± 0.89 aDistinct perithecium, heads swell slightly
3.16323Both MAT1-1 and MAT1-2Light yellow3.06 ± 0.46 bDistinct perithecium, heads swell significantly
3.16324MAT1-1White3.12 ± 0.47 bNo perithecium

Influence of light on the growth and conidiation of Cordyceps militaris

As shown in Fig. 1a, changes in the morphology of the colonies occurred in different light conditions and strains. CGMCC 3.16324, an albino, was white in all light conditions. The upper and reverse of the plates of each strain were nearly white in the dark. However, in white or blue light conditions, five strains showed different colors, especially in the reverse of the plates. The color of the colony was much heavier in the blue-light condition than in white light except the albino. Under white- and blue-light conditions, CGMCC 5.699, a degenerate strain that could not form a fruit body, showed the lightest color. The aerial hyphae of all strains in the dark condition were much denser than those under light conditions.

Figure 1.

Photo morphogenesis of the five strains of Cordyceps militaris that were used in this study. (a) Effect of light on morphology. (b) Effects of light on growth rates (c) Effects of light on conidiation. 5.699, 3.16321, 3.16322, 3.16323 and 3.16324 are the CGMCC number of the strains. Error bars in b and c indicate the standard deviation of three independent experiments. Different letters above the bars for the same strain indicate significant differences (anova followed by Tukey's HSD tests, P < 0.05). WL, white light; BL, blue light; D, dark.

The growth rates of the five strains were determined by the diameters of their colonies (Fig. 1b). All strains presented the same trend under the three light conditions, white light > dark > blue light. When comparing the growth rates of the different strains under the same light conditions, CGMCC 5.699 exhibited the fastest rate.

All strains produced the fewest conidia under the dark condition and the most under white light, suggesting light could stimulate conidiation in C. militaris. Conidia were remarkably abundant in CGMCC 5.699 under white- and blue-light conditions, and the numbers were 100-times greater than those of the strains CGMCC 3.16321 and 3.16324 (Fig. 1c).

Amplification of the Cmwc-1 gene using degenerate primers and genome walking with Hi-tail PCR

Using degenerate primers, a 787-bp product was amplified from C. militaris genomic DNA. tblastx analysis showed the closest homology (97%) to WC-1 of Beauveria bassiana (EJP60778.1) followed by that of Hypocrea jecorina (AAV80185.1).

The partial sequence was used for genome walking in the 5′ and 3′ directions by Hi-tail PCR. In all, we obtained 4322 bp of Cmwc-1 sequence information. The nucleotide sequences have been deposited in GenBank under accession numbers KC702390, KC702391, KC790039, KC790040 and KC790041 for strains CGMCC 5.699, 3.16321, 3.16322, 3.16323 and 3.16324, respectively.

Analysis of the deduced amino acid sequences

The gene Cmwc-1 of strain CGMCC 3.16322 extended for 2936 bp from its start to stop codons and included a single, 59-bp intron that was deduced from genomic and cDNA sequence comparison. The open reading frame coded for a predicted, 958-aa polypeptide, which implied that this sequence was 209 aa shorter than that of the NcWC-1. The putative molecular weights and pI values of CmWC-1 were 104.53 KD and 8.56, respectively.

Database searches using blast revealed that the predicted CmWC-1 protein was highly similar (68%) to NcWC-1. CmWC-1 had the N-terminal Gln-rich activation region (6–40 aa), the PAS A domain (LOV domain 310–379 aa), PAS B domain (505–571 aa), PAS C domain (619–690 aa), a nuclear localization signal (NLS 819–827 aa) and a single, putative GATA-type Znf domain (830–882 aa, Fig. 2).

Figure 2.

Schematic representation of WC-1 from several mushrooms and Neurospora crassa. Conserved domains of all putative photoreceptors were determined using Pfam ( and smart ( The size of each protein is indicated on the right. The regions corresponding to the AD (Gln-rich region), LOV (Light-Oxygen-Voltage) and PAS (Per-Arnt-Sim) domains and the NLS (nuclear localization signal), CC (coiled-coil structure) and ZnF (zinc-finger DNA-binding motif) are shown.

Genetic diversity of Cmwc-1 among different strains and phylogenetic analysis

The five Cmwc-1 gene sequences from the tested strains were aligned. The complete sequences of the Cmwc-1 genes were 2951 bp in length for strains CGMCC 3.16321 and 3.16324 but was 2936 bp for the other three strains because 15-bp nucleotides were deleted from their 5′ ends. It was interesting that the deleted nucleotides were ‘GCAGCAGCAGCA’ and resulted in a 5-Gln deletion in the Amino-terminus of the amino acid sequence. The CmWC-1 protein sequence was blasted against the genome of C. militaris CM01 (Zheng et al., 2011) and it was found to exhibit 99% similarity with the GATA-type Zinc finger domain-containing protein (EGX96523.1). No evident variation in the Cmwc-1 sequences of the degenerate and albino strains, compared with the other strains, was observed.

A phylogenetic tree was constructed using the identified protein sequences and the Sordariomycetes WC-1-like proteins that were retrieved from GenBank. The Jones-Taylor-Thornton (JTT) model was chosen. The two methods of Maximum Likelihood (ML) and Bayesian inference generated phylogenetic trees with the same topologies. The tree (Fig. 3) showed four major clades with relatively strong supports, each with proteins from Hypocreales, Sordariales, Glomerellales and Ophiostomatales. In the Hypocreales clade, five families were involved, and every family formed a sub-clade. Specifically, Ophiocordycipitaceae and Cordycipitaceae were two independent branches with 100/100 of posterior probabilities/ML bootstrap value, which was consistent with the result of multi-gene phylogenetic analysis of Cordyceps and the clavicipitaceous fungi by Sung et al. (2007).

Figure 3.

Phylogenetic tree of WC -1 from Sordariomycetes fungi. The topology of this tree was generated using mrbayes. The Bayesian posterior probability values as obtained by mrbayes 3.1.1 (1000 000 generations) are shown first and are followed by the ML bootstrap values. When only one value is shown, that value represents the Bayesian posterior probability value. ML bootstrap values were obtained using phyml under the JTT + I + G + F model with 500 bootstrap replicates. The scale bar represents 0.2 substitutions per site.

Cmwc-1 mRNA is light inducible

Levels of the Cmwc-1 mRNA were ascertained by real-time PCR using CGMCC 3.16322. After the selection of reference genes using the bestkeeper algorithms (Pfaffl et al., 2004), the tef1 gene was used as an internal control for constitutive expression (Fig. S3). A single amplification product was formed from each primer (Fig. S4). The expression level of the Cmwc-1 gene increased significantly after irradiation for 15 min (Fig. 4a) and the same occurred in the other strains (CGMCC 5.699 and 3.16324, Fig. 4b). However, the expression level decreased as the irradiation time increased. When the irradiation time lasted for 45 min, the expression level was nearly the same as that in the dark (Fig. 4a).

Figure 4.

Light-regulated gene expression of Cmwc-1 revealed by quantitative real-time PCR. (a) Cmwc-1 gene expression of strain CGMCC3.16322 after being irradiated for different time. (b) Comparison of Cmwc-1 gene expression among different strains under dark and white light irradiation for 15 min. Error bars indicate the standard deviation of three independent experiments. Different letters above the bars for the same strain indicate significant differences (anova followed by Tukey's HSD tests, P < 0.05).


Light is essential for the development of stromata of C. militaris. In the study, the influence of light on the growth and conidiation of this mushroom are analyzed and the gene of Cmwc-1, the homologe of the blue-light photoreceptor of Neurospora, is cloned and analyzed.

Five strains are used according to their mating types and characterization of the fruit bodies. There are degenerate, albino and normal strains. The stromata are with distinct or without perithecium. The strains with single or double mating types are also included. Therefore the strains used in the study are representative.

Light is the essential factor for the pigment production (Fig. 1A) and can also stimulate conidiation (Fig. 1C). Blue light can inhibit the growth rate (Fig. 1C). So light is not only essential for the stroma development but also important for the asexual growth of this fungus. In fact, light can regulate the development of many ascomycete fungi such as Paecilomyces fumosoroseus (Sánchez-Murillo et al., 2004), Tuber borchii (Ambra et al., 2004) and Trichoderma atroviride (Casas-Flores et al., 2004).

Motif Scan analysis shows that the WC-1 proteins from the Ascomycetes C. militaris, Ophiocordyceps sinensis, T. borchii and N. crassa have similar functional regions of highly conserved co linearity. CmWC-1 lacks the C-terminal glutamine-rich domain and a coiled-coil structure compared to NcWC-1. However, the WC-1 proteins from the Basidiomycetous mushrooms C. cinerea, L. edodes and S. commune are quite different. The Znf and NLS domains are deficient. The PAS C domain is also missing in C. cinerea and S. commune. It seems that some of the domains of WC-1 have been lost during evolution.

The 5-Gln deletions in CmWC-1 proteins of three strains occur in the N-terminal Gln-rich region, which might serve as an activation domain for a transcription factor. For Gln-rich transcription activation domains, the proportion of Gln residues seems to be more important than the overall structure (McLennan et al., 2012). It is unclear whether the reduction in the amount of polyglutamine repeats of CGMCC 5.699, 3.1322 and 3.16323 affects the activity of Cmwc-1. Phylogenetic analysis based on the Sordariomycetes WC-1-like proteins suggests that the sequence of WC-1 can be used as a candidate marker for phylogenetic analysis in fungi.

There is no relevant effect of light on the amounts of wcoA (white collar protein) mRNA in Fusarium fujikuroi (Estrada & Avalos, 2008), bwc1 in the basidiomycete Cryptococcus neoformans (Idnurm & Heitman, 2005) and madA in the zygomycete Phycomyces blakesleeanus (Idnurm et al., 2006). However, the results in this study suggest that the Cmwc-1 mRNA is light inducible, as are the genes for other WC-1-like proteins, such as Tbwc-1 from T. borchii (Ambra et al., 2004) and Oswc-1 from O. sinensis (Yang et al., 2013). The importance of this finding is that it demonstrates a light-dependent response at the molecular level.

Compared to the cultivated strain, the color of the colony changes slowly for the degenerate strain and no change occur for the albino one after exposure to light (Fig. 1). It's unexpected that the sequences and the relative expressions respond to irradiation of Cmwc-1 in degenerate and albino strains are similar as those of the cultivated one. It is suggested that the phenotype changes are not caused by light acceptor Cmwc-1 directly.

The proteins of the WC family govern many light-controlled processes in fungi. We believe this study will help to open the still-unexplored field of stroma development of this important edible fungus.


This work is supported by National Natural Science Foundation of China (31100014) and the project of State key laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences.