Background

Shiraia bambusicola P. Hennings is a bambusicolous fungus parasitized on bamboo tender twigs and its fruiting bodies have been used in traditional Chinese medicine to stimulate blood circulation, relieve expectoration, cure rheumarthritis and relax muscle rigidity [1, 2]. Hypocrellins, main perylenequinone pigments isolated from the hypha and fruit-bodies of Shiraia fungi, have being developed as new non-porphyrin and reactive oxygen species (ROS)-generating photosensitizers in photodynamic therapy (PDT) for clinical application of anti-microbes, -cancers and -viruses [1, 3, 4]. Due to the difficulty of chemical synthesis of hypocrellins and the scarcity of wild fruiting bodies [5], Shiraia mycelium culture has become a promising production process of this new PDT agent [6]. More process strategies were applied to enhance hypocrellin production in the culture, including the condition optimization for the cultures [7], and application of biotic [8, 9] or abiotic elicitors [10, 11]. Nitric oxide (NO), a small molecular signal, was found to be involved in the induced hypocrellin production by a fungal elicitor from Aspergillus niger at 50 µg/mL [12] and a fungal elicitor PB90 at 5 nM [8]. The endogenous NO in S. bambusicola was also observed in extractive culture of Shiraia by Triton X-100 [13].

NO is an important regulatory molecule in mammals [14]. Recently, NO has been proved to be an essential signal in the elicitation of plant defense and secondary metabolite biosynthesis [15]. NO released by a NO donor sodium nitroprusside (SNP) could elicit plant secondary metabolites such as marjoram essential oil [16], total phenols and flavonoids in Echinacea purpurea roots [17], artemisinin in Artemisia annua hairy roots [18], and taxol in Taxus yunnanensis cells [19]. However, there are few reports on the effects of NO on fungal growth and secondary metabolism. NO may mediate L-arginine-induced conidiation of Coniothyrium minitans [20]. Wang and Higgins (2005) reported that the germination and development of Colletotrichum coccodes conidia were significantly inhibited by SNP at 100 µM [21]. A NO-releasing compound diethylenetriamine-NoNoate at 1.5 mM increased drastically the formation of cleistothecia in A. nidulans, suggesting a positive regulator of NO on fungal sexual development [22]. Zhao et al. demonstrated that NO was involved in the co-culture of Inonotus obliquus with Phellinus morii and triggered the biosynthetical pathway of phenylpropanoids, leading to an increased production of styrylpyrone derivatives [23]. SNP at 5 mM increased yield of ganoderic triterpenoid by 40.94% in submerged fermenting Ganoderma lucidum [24]. Although the endogenous NO generation was observed in S. bambusicola treated by fungal elicitors [8, 12] and Triton X-100 [13], the regulation roles of NO on fungal hypocrellin biosynthesis are still undetermined. Therefore, as a follow-up to our efforts on enhancing hypocrellin production [9, 25] and elucidating the role of NO on secondary metabolite biosynthesis [13, 26], we examined the effects of SNP on Shiraia hypocrellin biosynthesis. We also investigated the relationship between NO and elicitation responses including ROS generation, the activation of antioxidant defenses and hypocrellin production in Shiraia mycelium cultures. In this study, de novo transcriptome sequencing for Shiraia sp. S9 was performed for better understanding the NO regulation on hypocrellin biosynthesis. This study presented a novel elicitor of hypocrellin production and shed light on the relational roles of NO on the biosynthesis of secondary metabolites.

Results

SNP application released NO and enhanced hypocrellin production

After the application of SNP at 0.10 mM to 3-day-old cultures, we observed a rise of green fluorescence of NO indicator 4, 5-diaminofluorescein diacetate (DAF-2-DA) in Shiraia mycelia, whereas the relative fluorescent ration was markedly decreased by NO scavenger 2-(4-carboxyphenyl)-4, 4, 5, 5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO), indicating the release of NO in hyphae (Fig. 1a, b). On the solid medium, the concentration of fungal conidia was decreased from 17.08 × 106 to 2.43 × 106 spores/mL by SNP treatment (Fig. 2a). The pycnidium formation was suppressed by SNP (Fig. 2b). Although SNP at higher concentration (1.00 mM) inhibited the fungal growth, the diameter of fungal colony was increased on day 6 by SNP at 0.01 or 0.02 mM (Fig. 2c, d). In the liquid culture, there were not any notable changes in fungal biomass (Additional file 1: Figure S1A), pH value and sugar consumption (Additional file 1: Figure S2) in medium after SNP application at 0.10 mM. However, SNP decreased fungal pellet diameter by 16.09–25.73% (Additional file 1: Figure S1B) and the pellets were colored with dark red (Fig. 2e). The influence of SNP on hypocrellins including hypocrellin A (HA) and HC, and elsinochrome (A, B and C) in the culture was detected by HPLC (Additional file 1: Figure S3). Both the intracellular HA (1.22-fold of control) and EA (5.88-fold of control) were enhanced by SNP at 0.10 mM (Table 1). The extracellular HA was increased by 87.56%, while HC, EB and EC were not detected in cultural broth with or without SNP treatment.

Fig. 1
figure 1

NO generation in SNP-treated mycelia of Shiraia sp. S9. a Bright-field image (above) and fluorescence microscopy of DAF-2-DA-stained mycelium (below) in cultures. SNP was added at 0.10 mM on day 3 of the culture. cPTIO (0.10 mM) was added 30 min prior to SNP treatment. The photos were taken after 2 h of SNP treatment. b NO accumulation (relative intensity of fluorescent ratio) in mycelium after SNP treatment. Values are mean ± SD from three independent experiments (**p < 0.01 vs. control, ##p < 0.01 vs. SNP group)

Full size image
Fig. 2
figure 2

Effects of SNP on the growth and development of Shiraia sp. S9. a Effect of SNP (0.01–1.00 mM) on the generation of conidia of S9 strain on day 8 in solid medium culture. b The morphologic characteristics of S9 strain was kept on PDA with or without SNP (0.10 mM) treatment for 8 days. c The diameter of S9 colony treated by SNP at 0.01–1.00 mM for 2–6 days. d The colony morphology of S9 strain with SNP (0.01–1.00 mM) treatment for 6 days. e Pellet morphology (15 ×) in submerged culture of S9 strain under SNP treatment at 0.10 mM on day 3. The culture was maintained in 150-mL flask containing 50 mL of the liquid medium at 150 rpm and 28℃. The red arrow indicates pycnidium. The black arrow indicates addition time. Values are mean ± SD from three independent experiments (*p < 0.05 and **p < 0.01 vs. control)

Full size image
Table 1 Effects of SNP on the individual hypocrellin production in submerged cultures of Shiraia sp. S9
Full size table

As HA is a major bioactive hypocrellin constituent in Shiraia [9], we optimized the conditions (concentration and addition time) of SNP application to obtain higher HA production. SNP at lower concentrations (0.01–0.20 mM) had no obvious impacts on fungal biomass (Additional file 1: Figure S4A), but increased HA contents in mycelium and in cultural broth. The higher HA in mycelium (9.72 mg/g DW) and in cultural broth (4.64 mg/L) were obtained at 0.02 mM and 0.20 mM, respectively (Additional file 1: Figure S4B, C). When SNP at 0.02 mM was applied on different days, the higher contents of intracellular HA (9.97 mg/g DW) and the released HA in cultural broth (4.42 mg/L) were achieved on day 3 (Additional file 1: Figure S5). Hence, SNP application at 0.02 mM on day 3 of the culture was then used for subsequent culture. The time course of the effect of SNP on HA production in 9-day-old cultures was shown in Fig. 3. Under this optimal condition, SNP application did not suppress mycelial growth (Fig. 3a), but promoted the intracellular HA contents by 73.31–178.96% (Fig. 3b) and extracellular HA accumulation by 26.09–119.26% (Fig. 3c). The total HA production was enhanced to 110.34 mg/L on day 7, a 2.65-fold increase over the control without SNP addition (Fig. 3d).

Fig. 3
figure 3

Time profiles of fungal biomass a, HA content in mycelia b, the released HA in cultural broth c and total HA production d in submerged culture of Shiraia sp. S9 under 0.02 mM SNP treatment on day 3. The culture was maintained in 150-mL flask containing 50 mL of the liquid medium at 150 rpm and 28 °C. The arrow represents the time of SNP addition. Values are mean ± SD from three independent experiments (**p < 0.01 vs. control)

Full size image

SNP-induced transcriptional changes of genes for hypocrellin biosynthesis

To examine the transcriptional changes of Shiraia sp. S9 after SNP treatment, RNA-Seq experiment was subsequently performed. There were altogether 84,275 unigenes assembled with an average length of 1,011.81 bp (base pairs) and an N50 of 5,399 bp (Additional file 1: Table S1, Figure S6). 98.62%, 54.40%, 37.37%, 30.86% and 90.40% of the total unigenes were resemble to known genes reported in the current databases in Additional file 1: Table S2. A total of 571 differentially expressed genes (DEGs) were identified (Additional file 2: Table S3), including up-regulated unigenes (355, 62.17%) and down-regulated unigenes (216, 37.83%) under SNP treatment (Additional file 1: Figure S7). These DEGs were categorized into the independent classification group “biological process (BP)”, “molecular function (MF)” and “cellular component (CC)” (Additional file 2: Table S4). Compared with the annotated unigenes of BP (270 DEGs) and CC (169 DEGs) (Additional file 1: Figure S8A, B), more than 322 DEGs were categorized into MF group (Additional file 1: Figure S8C). Within the MF category, DEGs (129 unigenes) assigned to ‘catalytic activity’ (GO:0003824) were of the highest proportion and other DEGs were mainly involved in ‘oxidoreductase activity’ (GO:0016491), ‘transporter activity’ (GO:0005215) and ‘transmembrane transporter activity’ (GO:0022857) items.

Based on the reported gene clusters for hypocrellin biosynthesis [27,28,29], we explored the expression changes of DEGs associated with hypocrellin biosynthesis under SNP treatment. De novo sequencing and comparative analysis revealed 113 putative DEGs were enriched in 10 items related to fungal hypocrellin production (Additional file 2: Table S5 and S6), including ‘polyketide synthase’, ‘hydroxylase’, ‘probable metabolite transport protein’, and so forth. Among them, more than 75% DEGs which are associated with intracellular hypocrellin biosynthesis were up-regulated compared with control group (Additional file 2: Table S5 and Fig. 4a), such as laccase-like multicopper oxidase (MCO, TRINITY_DN31904_c0_g1_i4), conidial yellow pigment biosynthesis polyketide synthase (PKS, TRINITY_DN73250_c0_g1_i1), hydroxyindole-O-methyltransferase (Omef, TRINITY_DN28714_c0_g2_i1), FAD dependent oxidoreductase (FAD, TRINITY_DN33725_c0_g1_i8), phenol hydroxylase (Hyd, TRINITY_DN84020_c0_g1_i1), cytochrome P450 CYP2 subfamily (Mono, TRINITY_DN31508_c0_g2_i1), fasciclin (Fas, TRINITY_DN43991_c0_g1_i1) (Table 2). The expression levels of seven randomly selected unigenes related to hypocrellin biosynthesis were confirmed through qRT-PCR strictly (Fig. 4b), in which Hyd and PKS were more dramatically up-regulated by 59.35- and 14.59-fold, separately. The expression changes by qRT-PCR showed a similar tendency to those in transcriptome data (Table 2). In addition, it was found that more than 60% DEGs bound up with extracellular HA exportation were also up-regulated (Additional file 2: Table S6), such as major facilitator superfamily (MFS, TRINITY_DN33647_c1_g1_i8), ABC superfamily (ABC, TRINITY_DN73211_c0_g2_i1), secondary metabolites biosynthesis, transport and catabolism (MTP, TRINITY_DN30127_c0_g1_i1) (Table 2). Meanwhile, the SNP-induced up-regulation of transcriptional expression of selected unnigene MFS (TRINITY_DN33647_c1_g1_i8) from 2.40- to 7.31-fold was confirmed by qRT-PCR (Fig. 4c).

Fig. 4
figure 4

a Heat map of DEGs involved in hypocrellin biosynthesis of Shiraia sp. S9 by SNP. b Validation of the expression levels of unigenes related to hypocrellin biosynthesis of S9 strain by qRT-PCR on day 8. Hyd, TRINITY_DN84020_c0_g1_i1. PKS, TRINITY_DN73347_c0_g1_i1. MCO, TRINITY_DN31904_c0_g1_i4. FAD, TRINITY_DN33725_c0_g1_i8. Omef, TRINITY_DN28714_c0_g2_i1. Mono, TRINITY_DN33456_c1_g1_i4. Fas, TRINITY_DN43991_c0_g1_i1. c Validation of the expression levels of MFS (TRINITY_DN33647_c1_g1_i8) by qRT-PCR. The SNP treatment was the same as specified in Fig. 3. Values are mean ± SD from three independent experiments (**p < 0.01 vs. control)

Full size image
Table 2 Examples of DEGs involved in the hypocrellin biosynthesis, transport and transcription factors (TFs) of Shiraia sp. S9 by SNP
Full size table

As shown in Additional file 2: Table S7, a total of 6440 unigenes were classified into 55 transcription factor (TF) groups, including C2H2, Trihelix, bZIP, bHLH, MYB related and so forth. Among all the TFs identified, bHLH (12.61%), ERF (7.47%), and MYB-related TF (7.08%) were of a higher proportion. Furthermore, there were altogether 189 DEGs enriched into 27 TFs differently expressed under SNP treatment (Additional file 2: Table S8), in which 24 TFs (110 DEGs) were up-regulated while 20 TFs (53 DEGs) were down-regulated in Shiraia mycelia. Among the up-regulated TFs, ERF ranked the highest (12 DEGs), followed by C3H (12 DEGs), and MYB (11 DEGs) (Table 2). Meanwhile, in the down-regulated TFs, NAC ranked the highest (8 DEGs), followed by C2H2 (6 DEGs), and bHLH (5 DEGs). Some prominent expression changes of unigenes associated with the high proportion of TFs were listed in Table 2, including several hypothetical proteins (TRINITY_DN31904_c0_g1_i4, TRINITY_DN33640_c0_g1_i50, TRINITY_DN33640_c0_g1_i25 and TRINITY_DN28413_c0_g1_i2), two aldehyde dehydrogenases (TRINITY_DN33276_c0_g1_i17 and TRINITY_DN33276_c0_g1_i28), hexose transporter-like protein (TRINITY_DN31580_c0_g1_i1), ATP binding (TRINITY_DN31301_c0_g1_i7).

SNP-induced transcriptional changes of genes involved in membrane permeabilization

On the basis of the analysis results from GO classification (Additional file 2: Table S4), a large number of unigenes were enriched into ‘membrane’ (61 DEGs of GO:0016020), ‘transporter’ (51 DEGs of GO:0006810), ‘integral to membrane’ (36 DEGs of GO:0016021), ‘intrinsic to membrane’ (36 DEGs of GO:0031224), ‘membrane part’ (36 DEGs of GO:0,044,425), ‘transmembrane transport’ (34 DEGs of GO:0055085), ‘transporter activity’ (33 DEGs of GO:0005215) and ‘transmembrane transporter activity’ (30 DEGs of GO:0022857) items after SNP application. Hence, we explored the alteration of hyphal cell membrane permeability and membrane lipid components of Shiraia sp. S9. As shown in Fig. 5, the fluorescence was strengthened in SYTOX Green-stained cells after SNP treatment, indicating the increased permeabilization of cell membrane. Simultaneously, we found that the composition proportions of some saturated fatty acids such as palmitic (C16:0) and stearic (C18:0) were decreased by 19.08% and 23.97%, respectively (Table 3). On the contrary, the proportions of four unsaturated fatty acids including palmitoleic (C16:1), oleic (C18:1), erucic (C22:1) and nervonic (C24:1) were raised significantly, up to 6.23-fold (Table 3). Hence, the ratio of unsaturated fatty acids with saturated ones was enhanced to 0.57 in the hyphal cells of S9 strain under SNP treatment, a 4.07-fold of the control group, suggesting the enhancement of cell membrane fluidity.

Table 3 Effect of SNP treatment on fatty acid composition (% total fatty acid) of Shiraia sp. S9
Full size table
Fig. 5
figure 5

Effect of SNP treatment on membrane permeability in Shiraia sp. S9 mycelia (400 ×). The 8-day-old fungal mycelium was incubated with 0.50 μM SYTOX Green for 30 min. The SNP treatment was the same as specified in Fig. 3. Panels represent bright-field images on the left and green fluorescence images on the right

Full size image

SNP-induced transcriptional changes of genes involved in oxidative stress

According to the transcriptome analysis, there were 121 DEGs enriched into ‘oxidation–reduction process’ (GO:0055114), ‘oxidoreductase activity’ (GO:0016491), ‘electron carrier activity’ (GO:0009055) and ‘oxidoreductase activity, acting on paired donors, with incorporation or reduction of molecular oxygen’ (GO:0016705) items (Additional file 2: Table S4), suggesting an oxidative stress induced by SNP. By comparison between control and SNP groups (Fig. 6a), the green fluorescent signals of 2, 7-dichlorodihydroflurescein diacetate (DCFH-DA) in hyphal cells were brighter and more intense (4.72-fold, Fig. 6b), indicating the increase of ROS generation. When Shiraia sp. S9 was cultured in the presence of ROS scavenger vitamin C (Vc) and diphenyleneiodonium (DPI), an inhibitor of ROS generating enzyme NADPH oxidase (NOX) for 30 min prior to SNP application, the relative intensities of fluorescence in mycelia exhibited much more notable reduction of 72.39% and 72.14% (Fig. 6b) compared with the SNP group, separately. The content of superoxide anion (O2−) in mycelia was induced rapidly around 30 min of SNP treatment, reaching a highest value of 6.47 µmol/g FW (fresh weight) with time up to day 7, which was 59.14% higher than that of control (Fig. 6c). The change trends of hydrogen peroxide (H2O2) concentration were consistent with O2− production. The generation of H2O2 was strikingly increased from 9.12 to 11.75 µmol/g FW after 2–4-day treatment of SNP and then decreased on day 7–9, but it was still higher than those of the control group (Fig. 6d). As shown in Fig. 7, the activities and expression levels of three oxidoreductases were significantly stimulated by SNP. The most significant stimulation of enzyme activities of NOX, catalase (CAT) and superoxide dismutase (SOD) occurred on day 7, which were 157.71%, 85.49% and 64.71% higher than those of control, respectively. Accordingly, the transcriptional expression levels of NOX, CAT and SOD were activated on day 5 or 7, and the strongest induction effect appeared on day 7, about 3.71-, 2.90- and 4.21-fold of control, separately.

Fig. 6
figure 6

Effect of SNP treatment on ROS production in mycelium of Shiraia sp. S9. a Bright-field images (left) and fluorescence microscopy (right) of DCFH-DA-stained mycelia (400 ×). b ROS accumulation in mycelium after SNP treatment. Vc (0.01 mM) and DPI (5.00 μM) were added 30 min prior to SNP treatment. The contents of O2− c and H2O2 d in mycelium of S9 by SNP. The photos and contents of ROS were taken and detected after 2 h of SNP treatment. The SNP treatment was the same as specified in Fig. 3. The arrow represents the time of SNP addition. Values are mean ± SD from three independent experiments (*p < 0.05, **p < 0.01 vs. control and ##p < 0.01 vs. SNP)

Full size image
Fig. 7
figure 7

Effects of SNP treatment on activities and expressions of NOX (a, b), CAT (c, d) and SOD (e, f) in Shiraia sp. S9 mycelium on day 1–9. NOX (comp2367_c0_seq1), CAT (comp15524_c0_seq1), SOD (CL8477Contig1). The SNP treatment was the same as specified in Fig. 3. The arrow represents the time of SNP addition. Values are mean ± SD from three independent experiments (*p < 0.05 and **p < 0.01 vs. control)

Full size image

Discussion

SNP is a potent vasodilator used clinically to treat hypertensive emergencies and heart failure. In addition, it is widely used as nitric oxide donor in pharmacologic studies to investigate on the physiological roles of NO [14]. In plant cells or root cultures, SNP has been used as NO donor to elicit the production of important bioactive secondary metabolites such as anthocyanin and flavonol glycoside in Pisum sativum [30], and tropane alkaloids in hair roots culture of Hyoscyamus reticulatus [31]. More recently, the signal roles of NO in fungal developmental and metabolic biosynthesis have drawn more attention [32]. With the help of using NO donor SNP and NO scavengers, previous studies indicated that NO modulated the germination of C. coccodes conidia [21], hyphal elongation of Magnaporthe oryzae [33] and biosynthesis of fungal secondary metabolites such as styrylpyrone polyphenols, flavonoids and phenolics [62]. Meanwhile, the significance of gene expression differences was evaluated using the cut-off criteria of |foldchange|≥ 2 and p value < 0.05.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted using RNAprep pure Plant Kit (Tiangen, Bei**g, China) according to the instruction. The primer sequences of 18S ribosomal RNA (internal reference gene) and selected unigenes from NCBI database of SRR7293200-7,293,205 and PRJNA323638 were listed in Additional file 1: Table S9. The qRT-PCR was performed according to the method in our previous study [61]. And, the transcriptional expression levels of selected genes were calculated from cycle threshold values by using the 2−△△CT method described in detail by Zhang et al. [63].

Statistical analysis

Data analyses were carried out using Microsoft Excel and expressed as Mean ± Standard Deviation (SD). Student's t-test was applied for the comparison of the means between two groups. One-way analysis of variance (ANOVA) test was used for the comparison of the means among multiple groups. p < 0.05 is considered statistically significant.