Naftifine inhibits pigmentation through down-regulation on expression of phytoene desaturase gene CAR1 in Rhodotorula mucilaginosa

Naftifine, an antifungal drug, inhibits pigmentation in Rhodotorula mucilaginosa . However, the relative mechanism is minutely understood. In this study, regulation of gene expression by naftifine was investigated to elucidate mechanism of yeast de-pigmentation. RNA-sequencing (RNA-seq) was used to screen differentially expressed genes (DEGs), followed by quantitative PCR (qPCR). The qPCR results showed that mRNA expression of phytoene desaturase gene CAR1 was reduced to 37% of its original level, after one day’s naftifine treatment. Since CAR1 acts at the immediate upstream of carotenoid biosynthesis pathway, it was concluded that naftifine involves in the process to inhibit the activity of phytoene desaturase, and that the down-regulation of gene CAR1 by naftifine contributes to de-pigmentation in R. mucilaginosa .


INTRODUCTION
Naftifine is a topical allylamine antifungal drug that is commonly used to treat dermatophytes infections (Carrillo-Munoz et al., 1999;Gupta et al., 2008;Ghannoum et al., 2013). Previously, it was known that naftifine increased the level of squalene and decreased that of ergosterol through inhibiting the activity of squalene epoxidase in fungi. Changes in the above mentioned steroid levels at opposite directions might increase permeability of fungal cells, thus triggered death of their cells (Paltauf et al., 1982). In Staphylococcus aureus, naftifine at low concentrations inhibited production of the virulence factor Staphyloxanthin, a carotenoid pigment, with a IC 50 = 0.088 mg/L and had no effect to inhibit bacterial growth. This inhibitory effect was not through regulating the expression of operon crtOPQMN or by inhibiting isoprenoid biosynthetic pathway, but by inhibiting CrtN enzyme directly (Chen et al., 2016).
The research aimed at identifying the targets of naftifine and understanding the mechanism of yeast decoloration through naftifine activations. Up to now, the effect of any antifungal drugs at gene expression level was rarely reported. In order to explore gene candidates that are regulated by naftifine as an antifungal drug, RNA-seq was used to screen the DEGs with focus on DNA replication and pigment synthesis pathways, as well as to quantify the mRNA levels of selected genes in qPCR assay.

Yeast strain
The R. mucilaginosa strain was isolated from the nails of a healthy 41-year-old Chinese man (Idris et al., 2019).

Culture of R. mucilaginosa
Culture media: YPD (2% glucose, 2% peptone, 1% yeast extract); SDA agar (4% glucose, 1% peptone, 2% agar). R. mucilaginosa was inoculated on SDA plates and incubated at 28°C. Single colonies were picked into YPD medium and incubated 160 rpm at 28°C overnight. Log growth phase yeast was then transferred into 10 ml YPD medium in 50 ml flasks with different concentrations of naftifine. The liquid cultured R. mucilaginosa was in the log phase before 36 h, and entered the stationary phase after 40 h (Landolfo et al., 2018). The culture was exposed to lab lights.

RNA-seq
R. mucilaginosa was streaked on SDA agar and cultured at 28°C. There were three samples: "Rh_ctrl" grown for 3 days, "Rh_+naftifine" grown with 200 mg/L of naftifine for 3 days, sample "Rh_-naftifine" grown with 200 mg/L of naftifine for 3 days, followed by naftifine-free for 3 days. The samples were crushed with liquid nitrogen. Total RNA was extracted using ESscience Tissue RNA Purification Kit (ESscience, Shanghai, China) according to the manufacturer's instructions. Nano Drop ND2000® spectrophotometry was used to measure RNA purity. RNA was enriched by oligo (dT) beads, fragmented and reverse-transcribed into cDNA, purified with QiaQuick PCR extraction kit, end repaired, poly (A) added, and ligated to Illumina sequencing adapters. The products were selected by size using agarose gel electrophoresis; PCR amplified, and sequenced using Illumina HiSeqTM2500 by Gene Denovo Biotechnology Co.

Bioinformatics analysis
RNA-seq data was submitted to the National Center for Biotechnology Information (NCBI) Short Read Archive (SRA) database with the accession number: PRJNA590855. Alignment tool, Bowtie2, was used for mapping (rRNA reads were removed). The reconstruction of transcripts was carried out with software Cufflinks together with TopHat2 (version 2.0.3.12). To identify new gene transcripts, all reconstructed transcripts were aligned to reference genome and were divided into twelve categories by using Cuff compare (a method of cufflinks, version 2.2.1). Genes with class code "u" (the transcripts was either unknown or in intergenic spacer region) were defined as novel genes. Novel genes were then aligned to the Nr and Kyoto Encyclopedia of Genes and Genomes (KEGG) database to obtain protein functional annotation. Gene abundances were quantified by software RSEM. The gene expression level was normalized by using FPKM (Fragments Per Kilobase of transcript per Million mapped reads) method. To identify differentially expressed genes (DEGs) across samples, the edgeR package (http://www.r-project.org/) was used. Genes with a fold change ≥ 2 and a false discovery rate (FDR) < 0.05 in a comparison as significant DEGs were identified. DEGs were then subjected to enrichment analysis of Gene Ontology (GO) functions and KEGG pathways.

Real-time qPCR
The qPCR primers were designed using the NCBI primer designing tool (Table 1). Two internal controls were selected, ltv1 (a cytoplasmic protein) and DLD2 (D-lactate dehydrogenase), since their fluctuation in expression level between different samples of RNA-seq was insignificant and their FPKM values were moderate (Van et al., 2017). The qPCR results showed that the relative expression of the internal control genes was stable between the naftifine-treated and control groups.
Yeast was cultured in liquid medium and treated with 4 mg/L naftifine for various days. Total RNA was extracted the same way as in RNA-seq. The RNA samples were reverse-transcribed into cDNA by PrimeScript TM RT reagent Kit (Takara, Dalian, China), and qPCR was performed using SYBR Premix Ex Taq GC kit (Takara, Dalian, China). The cycles were: 95°C 2 min, 95°C 20 s, 60°C 20 s, 72°C 15 s, 39 cycles. Dissolution curve conditions were: 65°C 5 s, 95°C 5 s, 4°C 30 s. Each sample was processed in triplicates using the CFX-96 TouchTM Real-Time PCR Detection System (BioRad, USA). In calculating the relative expression level of CAR1 and CAR2 genes, the two internal controls were used to calculate their ΔΔct values, and average with 2 -ΔΔCt method.

Data analysis software
GraphPadPrism7 was used to calculate IC50. Bio-Rad CFX Manager and GraphPadPrism7 software were used for qPCR analysis and plotting.

Decoloration of R. mucilaginosa was induced by low concentration of naftifine
The drug sensitivity of the selected R. mucilaginosa strain was first examined. The 50% inhibiting concentration (IC 50 )   Figure 2B and C), which was far lower than its growth inhibitory concentration (about 1, 400-fold lower).
Remarkably, butenaphthol and terbinafine, the naftifine analogs, did not decolorize R. mucilaginosa at the same or even higher mass concentrations even though their growth inhibitory IC 50 is lower than naftifine's. Based on the dramatic difference in IC 50 , the mechanisms of growth inhibition and decoloration through naftifine treatments were likely independent. In addition, decoloration depends on the yeast growth phase. No decoloration was observed when 4 mg/mL of naftifine was added to the yeasts that already grew in liquid medium for 3 days, not even with much higher concentrations ( Figure 2D).

Naftifine did not accelerate pigment degradation in vitro
Next, the mechanism of decoloration induced by naftifine was studied through assessing the possibility that naftifine accelerated pigment degradation. The hypothesis is that whether pigments are prone to degradation when they interact with naftifine. Pigment extract from yeast was mixed with various concentrations of naftifine and monitored for 48 h for light absorbance by a spectrophotometer. Results showed that only at extremely high concentration of 1, 000 mg/L, naftifine could reduce light absorbance at a significant level ( Figure 2E). There was no significant acceleration of pigment degradation with concentration up to 100 mg/L, that is > 1, 000-fold of the decoloration IC 50 . The results suggested that the major yeast decoloration was not due to naftifine-facilitated pigment degradation.

RNA-seq identified the candidate genes in carotenoid biosynthesis pathway
It is reasonable to hypothesize that decoloration might be due to the decrease of carotenoid synthesis. In order to screen candidate DEGs, R. mucilaginosa was cultured on SDA plates for different treatment scenarios for 3 days, which yielded three samples for RNA-seq analysis (Rh_ctrl, Rh_+naftifine, and Rh_-naftifine, see Materials and Methods). By aligning with the reference (JGI Rhomuc1_GeneCatalog_20160519), a total of 7, 618 known genes and 98 new genes were identified from these three samples (Table 2). Results indicated that none of the new genes was among carotenoid, isoprene, and steroid biosynthesis pathways based on annotation.
Gene Ontology function analysis showed that DEGs were significantly enriched in terms of both DNA replication and protein-DNA complex ( Figure 3A). KEGG pathway analysis showed that DEGs between Rh_ctrl and Rh_+naftifine were enriched in DNA replication, material metabolism (especially fatty acid metabolism), and oxidation pathways ( Figure 3B). Out of the expectation, no DEGs were found in the steroid biosynthetic pathway. However, two genes of carotenoid synthesis, CAR1 and CAR2, were marginally at the higher expression levels after 3 days' naftifine treatments. Since CAR1 and CAR2 were found to play important roles in carotenoid synthesis in R. mucilaginosa (Figure 1) (Landolfo et al., 2018), their expressional regulations were further studied in qPCR assay.

Real-time qPCR showed down-regulation of CAR1 gene expression after naftifine treatment
The above RNA-seq analysis was from solid culture and the concentration of nafitifine was higher than growth inhibitory IC 50 . To focus on studying the effect of naftifine on decoloration and to investigate the effect at a time-dependent manner, we switched to liquid culture and used a much lower concentration (4 mg/ml vs 200 mg/ml) for real-time qPCR assay. At 4 mg/ml, nafitifine decolorized yeast, but had little effect on growth. After treatment for 1 day, the expression level of CAR1 gene decreased to 37% when compared to control ( Figure 4A, B). The t-test of three experiments showed the reduction was significant (p = 0.007). The expression was also reduced after treatment for 3 days, but not statistically significant. Similarly, the reduction of expression for CAR2 was also not statistically significance after treatment for 1 or 3 days. After treatment for 5 days, both genes showed no change in relative expression level. Since decoloration was not detectable anymore if naftifine was not introduced until the yeast had grown for 3 days and afterward ( Figure  2D), whether the CAR1 down-regulation was also dependent of yeast growth phase in a similar way of decoloration dynamics was further analyzed. Results of real-time qPCR assay indicated that expression levels of both CAR1 and CAR2 did not change after 1-day's naftifine treatment when yeast had reached stationary phase ( Figure 4C). This result further supported that decoloration correlated to CAR1 down-regulation.

Bioinformatical analysis on inhibition of phytoene desaturase activity by naftifine
In addition to analysis of the effect on gene expressions, further evaluation was done to know whether naftifine acted on their protein levels. Since naftifine was a potent inhibitor of CrtN (diapophytoene desaturase) in S. aureus, homologous proteins of CrtN in R. mucilaginosa was searched. Interestingly, phytoene desaturase encoded by CAR1 had the highest similarity with CrtN ( Figure 5). Functionally, both of them are a membrane-peripheral and  FAD-dependent oxidase/isomerase that catalyzes the formation of multiple unsaturated double bonds of carotenoids (Schaub et al., 2012). Based on their high sequence homology, the results suggested that naftifine involved into the process of inhibiting the activity of phytoene desaturase in yeast.

DISCUSSION
Many non-phototrophic bacteria and fungi rely on carotenoids for protection from harmful radicals (Chi et al., 2015;Llansola et al., 2017). In humans, carotenoids are precursors of vitamin A, an effective antioxidant supplied from food (Bohn et al., 2017). As a non-photosynthetic fungus, R. mucilaginosa is a carotenoid producer and is protected by carotenoids against oxidative damage from UVB (Moliné et al., 2009(Moliné et al., , 2010. This study focused on understanding the mechanism of yeast decoloration by naftifine. Pigmentation was reduced when low concentration of naftifine was added to early phase yeast ( Figure 2B). For the first time, it was shown further that the reduction was not due to faster degradation in the presence of naftifine. No significant change happened in degradation rate when pigment extract was mixed with up to 333 mg/L naftifine, > 1, 000 fold higher than IC 50 .
For decolorization in R. mucilaginosa, naftifine is much more potent (IC 50 = 0.30 umol/L) than drug diphenylamine (IC 50 = 20 umol/L) (Raisig and Sandmann, 2001;Ghannoum et al., 2013;Mot et al., 2017). Diphenylamine reduces carotenoid accumulation by inhibiting desaturation of phytoene. Naftifine may reduce carotenoid levels partially by inhibiting phytoene desaturase in similarity to diphenylamine. Among annotated proteins of R. mucilaginosa, the phytoene desaturase has the highest homology in sequence with bacterial CrtN, a desaturase inhibited directly by naftifine in bacteria. However, the change of carotenoid level under naftifine treatment was not measured directly in this study. The results suggested that de-pigmentation by naftifine was largely due to regulation at gene expression level of CAR1. It was expected that diphenylamine is unlikely down-regulating CAR1 gene expression since it is much less potent in decoloration.
The study indicated that RNA-seq was an effective method to successfully screen DEGs. Naftifine regulated the expression of genes related to DNA replication and metabolism ( Figure 3A and B). Interestingly, CAR1 and CAR2 had a slightly higher FPKM values, suggesting that naftifine might regulate gene expression in carotenoid synthesis pathway. To further study naftifine regulation in gene expression, real-time qPCR was used to specifically quantify relative gene expression levels of CAR1 and CAR2 with treatment of lower naftifine concentration in liquid culture. Naftifine down-regulated the relative expression of CAR1 to 37% of control level after one-day treatment. Apparently, the CAR1 down-regulation was not related to the effect of naftifine in yeast growth. We further suspect that the other antifungal drugs without inducing de-pigmentation have no regulating effect in CAR1 expression. In conclusion, yeast decoloration by naftifine might be large through down-regulation of CAR1 expression.