Ganoderma lucidum, (Curtis) P. Karst. (Curtis) P. Karst.

Lu, Xiaoxiao, Xie, Chunqin, Wang, Yunxiao, Liu, Yongnan, Han, Jing, Shi, Liang, Zhu, Jing, Yu, Hanshou, Ren, Ang & Zhao, Mingwen, 2020, Function of ceramide synthases on growth, ganoderic acid biosynthesis and sphingolipid homeostasis in Ganoderma lucidum, Phytochemistry (112283) 172, pp. 1-9 : 2-3

2.1. G. lucidum contains three ceramide synthases

On the basis of evidence of ceramide synthases in other fungi, we performed a bioinformatic search for putative ceramide synthases present in G. lucidum . Our analysis revealed that there are three putative ceramide synthases in G. lucidum by homologue alignment by comparing to the proteins from other fungi. They are referred in this study as LAG1, LAG2 and LAG3, respectively. The LAG1, LAG2 and LAG3 had significant homology to other ceramide synthase from Basidiomycetes including Lentinus tigrinus , Armillaria gallica , Cryptococcus depauperatus , Tsuchiyaea wing fi eldii, Cryptococcus gattii and Cryptococcus neoformans . The genomic DNA sequence of the lag1 gene is 1385 bp and is interrupted by three introns, whereas the spliced transcript of lag1 is 1089 bp and encodes a protein containing 362 amino acids. The genomic DNA sequence of the lag2 gene is composed of 1844 bp gene and contains nine introns, whereas the 1317 bp spliced transcript of lag2 encodes a protein consisting of 438 amino acids. The cDNA of lag3 contains a spliced transcript of 1431 bp encoding 476 amino acids, whereas the lag 3 genomic DNA sequence is 2083 bp and contains eleven introns.

We investigated the divergence of the three CERSs in G. lucidum from other known CERS proteins from animals, plants and fungi over the course of evolution. The entire phylogenetic tree can be divided into three major groups ( Fig. 2 ). Interestingly, the LAG1 protein from G. lucidum , the Cer1 proteins from Basidiomycetes and the Lac1 (or Bar1p) proteins from Ascomycetes are clustered together. The LAG2 and LAG3 proteins from G. lucidum , the Cer2 and Cer3 proteins from Basidiomycetes and the Lag1 proteins in Ascomycetes grouped into a separate cluster. In addition, LAG1, LAG2 and LAG3 from G. lucidum exhibited a high degree of sequence identity with other ceramide synthases from Basidiomycetes including Lentinus tigrinus , Armillaria gallica , Cryptococcus depauperatus , Tsuchiyaea wing fi eldii, Cryptococcus gattii and Cryptococcus neoformans . LAG2 and LAG3 shared high identity with each other (displayed 66.06% identity), as well as to ceramide synthases (Cer2 and Cer3) of L. tigrinus (displayed 44.10% identity), A. gallica (displayed 57.33% identity), C. depauperatus (displayed 49.88% identity), T. wing fi eldii (displayed 49.76% identity), C. gattii (displayed 66.19% identity) and C. neoformans (displayed 65.70% identity). The LAG1 amino acid sequence is greatly diverged from LAG2 and LAG3 (displayed 23.93% and 21.34% identity, respectively) in G. lucidum and showed 19.14–36.98% identity to the proteins denoted as LAG1 or Cer 1 in L. tigrinus (36.98%), A. gallica (26.88%), C. neoformans (35.84%), C. depauperatus (33.08%), T. wing fi eldii (32.59%) and C. gattii (36.90%), Lac 1 in Blastomyces parvus (20.11%), Emmonsia crescens (19.14%), Yarrowia lipolytica (30.26%), Wickerhamomyces ciferrii (26.12%), Kluyveromyces marxianus (27.98%), Candida albicans (24.25%) and Kluyveromyces lactis (32.29%), and Bar1p in Pichia pastoris (27.98%).

The conserved domain analysis for the CERS proteins was performed using the SMART database (http://smart.emblheidelberg.de). The domain analysis revealed that the three predicted G. lucidum CERSs contain one Tram-Lag-CLN8 (TLC) domain. The TLC domains of LAG2 and LAG3 are highly similar to each other and are different from that of LAG1. In addition, LAG2 and LAG3 have two transmembrane (TM) domains, whereas LAG1 has only one TM domain (Supplementary Fig. S1 ).

2.2. Construction of G. lucidum lag-silenced strains

To investigate the function of CERS in physiological processes, we constructed lag1, lag2 and lag3 single-silenced mutants and lag2/lag3 double-silenced mutants as previously described by using an established dual-promoter system (Supplementary Fig. S2 ), which carries the hygromycin B (Hyg) resistance gene as a selectable marker. Real-time PCR analysis was used to examine the expression of the lag genes and confirm silencing in the transformants. The results showed that the expression of lag was down-regulated greater than 50% (P <0.01) in all lag -silenced strains compared to that in WT strains. The silenced strains lag1- (LAG1i-7 and LAG1i-8), lag2- (LAG2i-1 and LAG2i-7), lag3- (LAG3i-22 and LAG3i-26) and lag2/lag3- (LAG23i-4 and LAG23i-5) strains were selected for further analysis (Supplementary Fig. S3 ).

2.3. Ceramide synthases LAG2 and LAG3 are required for G. lucidum growth

To explore whether CERS impact the growth of G. lucidum , the SiControl and WT strains and the individual lag - and the lag2/lag3 double-silenced transformants were cultivated on solid CYM for 5 days to evaluate the resulting colony diameters. The colony diameters of the individual lag -silenced transformants LAG1i-7, LAG1i-8, LAG2i-1, LAG2i-7, LAG3i-22 and LAG3i-26 were not significantly different (P> 0.05) from those of the WT and SiControl strains ( Fig. 3 and Supplementary Fig. S4A ). For the l ag2 / lag3 double-silenced transformants LAG23i-4 and LAG23i-5, growth was decreased by approximately 22% and 45% (P <0.01), respectively, compared with that observed for the SiControl and WT strains. We further evaluated the effects of the loss of the lag genes on G. lucidum growth by analysing the biomass of the strains. Compared to the SiControl and WT strains, the dry weights of the individual lag -silenced transformants were not significantly different (P> 0.05); however, the dry weights were significantly (P <0.01) decreased by approximately 44 and 49%, respectively, in the LAG23i-4 and LAG23i-5 strains compared with the SiControl and WT strains (Supplementary Fig. S4B ). These results indicated that lag2 and lag3 are required and are functionally redundant with respect to the observed growth phenotype.

Figures

Fig. 2. G. lucidum harbours three ceramide synthases. Phylogenetic analysis of ceramide synthases in eukaryotes. The phylogenetic tree was conducted using MEGA 6 and can be roughly divided into three major groups: human, plant and fungi. The evolutionary history was inferred by the neighbour-joining method from 1000 replicates using MEGA 6.

Fig. 1. Overview of the sphingolipid biosynthetic pathway in yeast and filamentous fungi. The abbreviations used are as follows: Dihydroxy LCB: dihydroxy longchain (sphingoid) base; Trihydroxy LCB: trihydroxy long-chain sphingoid base; Long chain FA-CoA: long chain fatty acyl-coenzyme A; Very long chain FA-CoA: very long chain fatty acyl-coenzyme A; IPC synthase: inositol phosphorylceramide synthase; IPC mannosyl transferase: inositol phosphorylceramide mannosyl transferase; MIPCs: mannose inositol phosphorylceramides; M (IP)2Cs: mannose (inositol phosphoryl)2- ceramides.

Fig. 3. Effect of lag gene silencing on G. lucidum growth. Morphology of fungal colonies in lag-silenced, WT and SiControl strains after cultivation in the dark on CYM medium at 28 ̊C for 5 days. Three independent biological replicates in each column. Error bars represent standard deviations, and asterisks show significant differences from control (WT and Sicontrol) strains according to Student's t-test (**P <0.01, n = 3).

Fig. 4. Ganoderic acid biosynthesis in G. lucidum is influenced by the lag gene. A Systematic content analysis of ganoderic acid in the lag-silenced, WT and SiControl strains. B–D Relative gene expression of hmgr (B), sqs (C) and osc (D) in the WT, SiControl and lag-silenced strains. There are three independent biological replicates in each column. Error bars represent standard deviations, and asterisks show significant differences from control (WT and Sicontrol) strains according to Student's t-test (**P <0.01, n = 3).