Potent Activities of Roemerine against Candida albicans and the Underlying Mechanisms

Roemerine (RM) is an aporphine alkaloid isolated from the fresh rattan stem of Fibraurea recisa, and it has been demonstrated to have certain antifungal activity. This study aimed to investigate the antifungal activity of RM and the underlying mechanisms in Candida albicans (C. albicans). The in vitro antifungal activity of RM was evaluated by a series of experiments, including the XTT reduction assay, confocal laser scanning microscopy assay, scanning electron microscope assay. Results showed that 1 μg/mL RM inhibited biofilm formation significantly (p < 0.01) both in Spider medium and Lee’s medium. In addition, RM could inhibit yeast-to-hyphae transition of C. albicans in a dose-dependent manner. The biofilm-specific and hypha-specific genes such as YWP1, SAP5, SAP6, HWP1, ECE1 were up-regulated and EFG1 was down-regulated after 8 μg/mL RM treatment. Furthermore, the toxicity of RM was investigated using C. elegans worms, three cancer cells and one normal cell. The date showed that RM had no significant toxicity. In conclusion, RM could inhibited the formation of C. albicans biofilm in vitro, but it had no fungicidal effect on planktonic C. albicans cells, and the anti-biofilm mechanism may be related to the cAMP pathway.

. Chemical structure of RM, which is an aporphine alkaloid isolated from the fresh rattan stem of Fibraurea recisa.

Effect of RM against C. albicans Yeast Cells
The in vitro antimicrobial activity of RM was evaluated by broth microdilution assay, and the antimicrobial spectrum is demonstrated in Table 1. RM alone exhibited weak effect against C. albicans (MIC80: 128-256 mg/L), while it showed more potent activity against Candida glabrata, Candida krusei, Candida tropicalis, Candida parapsilosis, Aspergillus fumigatus and Staphylococcus aureus (Table 1). Meanwhile, the effect of different concentrations of RM on growth of C. albicans SC5314 was further established by time-growth curves ( Figure 2). It showed that at concentrations ranging from 2 to 128 μg/mL, RM could not significantly inhibit the growth of C. albicans, the yeast cells growth was similar to that of the control group without RM treatment. By contrast, 256 μg/mL RM showed a significant effect on C. albicans growth ( Figure 2).

RM Inhibits the Formation of C. albicans Biofilms
To validate that RM has the anti-biofilm effect, we evaluated XTT reduction assay. In this study, we found that RM inhibited C. albicans biofilm formation in a dose-dependent manner ( Figure 3A). In addition, one microgram per milliliter of RM inhibited biofilm formation significantly (p < 0.01) both in Spider medium and Lee's medium. The anti-biofilm activity of 8 μg/mL RM in the Spider medium was the strongest and the C. albicans SC5314 biofilm formation was inhibited by 90% compared with the control. RM inhibited biofilm formation by 80% in the concentration in 8 μg/mL in Lee's medium ( Figure 3A). Furthermore, the anti-biofilm effect of RM was evaluated by CLSM ( Figure 3B) and SEM ( Figure 3C). C. albicans biofilm formation was disrupted by RM in a dose-dependent manner. In the drug free medium, C. albicans cells formed biofilm with a dense network of yeasts and true hyphae ( Figure 3(Ba-c,Ca-c)). When cells exposed to 8 μg/mL RM, the biofilm formation was seriously disrupted (Figure 3(Bg-i,Cg-i)). Moreover, cell density was further reduced by 16 μg/mL RM (Figure 3(Bj-l,Cj-l)).

RM Inhibits the Yeast-to-Hypha Morphological Transition of C. albicans
To evaluate the effect of RM on the yeast-to-hypha morphological transition of C. albicans, three hypha-inducing mediums (liquid and solid media)-Spider, Lee's, and YPD + 10% FBS-were used. RM showed strong hypha inhibition activity in different liquid medium ( Figure 4A-C, Figure S1), the strong-to-weak sequence of the effect of RM against C. albicans biofilms was Spider medium, Lee's medium and YPD + 10% FBS medium. In liquid Spider medium, one microgram per milliliter of RM could inhibit the growth of hypha, but the effect declined after 6 h, and this effect of RM on yeast-to-hypha transition enhanced with increasing RM concentrations. Four micrograms per milliliter of RM could inhibit the growth of hypha to 48 h, and 32 μg/mL RM could completely inhibit the hyphal growth ( Figure 4A). In liquid Lee's medium, one microgram per milliliter of RM could inhibit the growth of hypha, and the effect could continue after 48 h. Likewise, 32 μg/mL RM could completely inhibit the hyphal growth ( Figure 4B). The C. albicans hyphal growth in liquid serum-containing medium was inhibited by 32 μg/mL RM ( Figure 4C). Similarly, only smooth-edged colonies were observed on Spider (8 μg/mL) and Lee's (2 μg/mL) solid medium ( Figure 4D). When growing on serum-containing solid medium, colonies appeared smooth and round at 8 μg/mL, whereas, the colonies on the control plate were highly wrinkled and uneven ( Figure S5). The results supported our hypothesis that RM could inhibit hyphal growth.

RM Decreases CSH of C. albicans Biofilm
In this study, the results showed that RM decreased CSH in a dose-dependent manner in four mediums (Spider, Lee's, RMPI1640 and YPD + 10% FBS) ( Figure 5). The effect in Spider medium was the strongest and 4 μg/mL RM significantly decreased CSH from 0.75-0.17 (p < 0.001).

Figure 5.
Effects of different concentrations of RM on CSH of C. albicans SC5314. The water-hydrocarbon two-phase assay was used to evaluate CSH. Standard deviations are based on three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001.

Exposure to RM Alters C. albicans Gene Transcription
In order to understand the anti-biofilm mechanism of RM, we further investigated the expression changes of genes after RM treatment using real-time RT-PCR. The result showed that the biofilm-specific and hypha-specific genes such as YWP1, SAP5, SAP6, HWP1, ECE1 were up-regulated by 2.29, 5.99, 2.27, 7.63, 15.44-fold, respectively, and EFG1 was down-regulated by 0.4013-fold after 8 μg/mL RM treatment ( Figure 6). Figure 6. The expression changes of some important biofilm formation related genes after 8 μg/mL RM treatment. The C. albicans strain tested was SC5314. Gene expression was indicated as a fold change relative to that of the control group. 18S rRNA was used to normalize the expression data. Data are means ± standard deviations from three experiments.

Exogenous cAMP Reverts the Morphogenesis Defect Caused by RM
Because some changed genes after RM treatment, such as HWP1, ECE1 and EFG1, were regulated by Ras/cAMP pathway and cAMP/PKA pathway [17], we performed exogenous cAMP supplement assay to verify that the antibiofilm mechanism of RM was related to the cAMP pathway. The result showed that exogenous cAMP could revert the morphogenesis defect of C. albicans cells caused by RM ( Figure 7A,B). When 10 mM cAMP were added in RM treated cultures, true hyphae were observed both in liquid and on solid Spider media ( Figure 7A,B). This result indicates that RM may inhibit the biofilm formation by the cAMP pathway.

Cytotoxicity Studies
In this study, the results showed that RM had good cytotoxic effect against cancer cell lines SGC-7901, HT-29 and MGC-803 with IC50 of 0.844 mg/L (3.02 µM), 1.279 mg/L (4.58 µM) and 0.631 mg/L (2.26 µM), respectively. Besides, RM had no significant cytotoxicity against human umbilical vein endothelial cell with the IC50 of 43.047 mg/L (154.11 µM). This experiment confirmed that RM had lower cytotoxic effect for non-cancer cell lines.

The Toxicity of RM on C. elegans Worms
We investigated the toxicity of RM using health adult C. elegans worms. The results showed that at a range of concentrations, from 4-32 μg/mL, RM did not display toxicity, and all the worms looked healthy (Figure 8). At 64 μg/mL RM, only two worms died, and so high concentration of RM exhibits low toxicity. Nevertheless, RM has certain toxicity for the worms, the LD50 of RM on C. elegans worms is 4096 μg/mL. One gram per milliliter of RM can make all the worms die.

Discussion
C. albicans is an opportunistic deep-infective fungi, still threatening our lives [18]. Most of established antifungal agents has been proved less than effective against C. albicans [19]. Therefore, the research and development of new antifungal agents is urgent. RM is a good therapeutic candidate that has been confirmed to have certain antifungal activity (MIC80: 160 mg/L) [13]. However, the effect of RM against C. albicans yeast cells and biofilm has not been elucidated. Consequently, in the current study, we focus on investigating the antifungal activity of RM against C. albicans biofilm and exploring the underlying mechanisms.
In this study, we found that RM had strong antimicrobial activity against common pathogens, such as A. fumigatus, S. aureus, C. glabratas, C. krusei, C. parapsilosis, and C. tropicalis. Unfortunately, RM showed weak activity against C. albicans (MIC80: 128-256 μg/mL). C. albicans forms biofilms easily, which results in C. albicans resistance to traditional antifungal agents [20]. Biofilm formation includes three stages: adhesion to biomaterial surfaces, growth to form an anchoring layer, and morphological transition to form a complex three-dimensional structure. To validate that RM has an anti-biofilm effect, we evaluated XTT reduction assay. Of note, we found that RM had strong effect against C. albicans biofilm. RM could inhibit the growth of biofilms in different liquid mediums, the strong-to-weak sequence of the effect of RM against C. albicans biofilms was Spider medium, Lee's medium and YPD + 10% FBS medium. Furthermore, CLSM assay and SEM assay were established to observe biofilm formation directly, and the results were highly consistent with that obtained by XTT reduction assay. In summary, RM has a strong antibiofilm effect against C. albicans in vitro.
It was demonstrated that biofilm formation and the yeast-to-hypha morphological transition contributed to the pathogenic potential of C. albicans [21]. Our data showed that RM significantly inhibited the yeast-to-hypha morphological transition in both liquid and solid medium. The results indicate that the anti-biofilm activity effect of RM may be involved the growth of C. albicans hypha rather than yeast cells. In this study, our results showed that 4 μg/mL RM decreased CSH significantly (p < 0.001). CSH is one of the major pathogenic attributes of C. albicans and showed a positive correlation with adhesion [22]. It is the indicator for adhesion ability. These results indicate that the anti-biofilm effect of RM seems attributable to its anti-adhesion and anti-morphological transition activities.
In order to understand the anti-biofilm mechanism of RM, real-time RT-PCR assay was further performed. The data showed that the biofilm-specific and hypha-specific genes such as YWP1, SAP5, SAP6, HWP1, ECE1 were up-regulated and EFG1 was down-regulated after 8 μg/mL RM treatment. ECE1 is a hypha-specific gene and it encodes a membrane protein dependent on the cAMP pathway [23]. HWP1 is a specific hyphal gene encoding a cell wall mannose protein, which is essential for normal growth of the mycelium. It also plays an important role in the pathogenesis of Candida infections. Mutation of the HWP1 gene results in incomplete biofilm formation [24]. ECE1 and HWP1 are regulated by Efg1. The protein is a transcription factor of the Ras/cAMP pathway and the cAMP/PKA pathway in hyphal development [17]. SAP5 and SAP6 encode secreted aspartic proteinase, which is expressed during hyphal growth. Expression of SAP5 and SAP6 also depends on Efg1 [25]. Based on this result, we speculated that the antibiofilm effect of RM might be related to the cAMP pathway. In this study, we fortunately found that exogenous cAMP could revert the hyphal morphogenesis defect of C. albicans cells caused by RM, the result further validated our speculation. However, the molecular mechanisms need to be further researched.
Fungal and mammalian cells belong to eukaryotic cells, thus, it is very difficult to research and develop a new antifungal drug without toxicity in mammals. In this study, we used healthy C. elegans worms to study the toxicity of RM, and the results indicated that at the final concentration as high as 64 μg/mL RM exhibited low toxicity. Cytotoxic activities of RM have been studied in lung adenocarcinoma cancer cells (A549), cervical adenocarcinoma cancer cells (HeLa), breast cancer cells (MCF-7), prostate cancer cells (LNCaP and PC3) and two human non-cancer cell lines: human embryonic kidney 293(HEK293) and neonatal foreskin fibroblast (NFF) cells [26]. Interestingly, RM had selective cytotoxicity against cancer cells, such as: A549 cells (IC50 of 3.4 µM) and HeLa cells (IC50 of 4.0 µM). However, RM had less cytotoxicity against non-cancer cell lines, HEK293 cells (IC50 of 15.9 µM) and NFF cells (IC50 of 15.2 µM), RM was selective against cancer cells because it had less inhibitory effect on the proliferation of non-cancer cells. It was noticable that the cells used in this study were not reported. The results revealed no significant cytotoxicity against human umbilical vein endothelial cell with the IC50 of 43.047 mg/L (154.11 µM). This experiment further confirmed that RM had lower cytotoxic effect for non-cancer cell lines although additional studies in vivo need to be carried out to determine the toxicity of RM in mammals.

Strains and Growth Condition
In this study, the international general laboratory strain: C. albicans SC5314 [27], Candida parapsilosis 90018, Staphylococcus aureus ATCC25913 [28], three clinical Candida spp. isolates: Candida glabrata 8535, Candida krusei 4996, Candida tropicalis 8915, one clinical Aspergillus fumigatus isolates: Aspergillus fumigatus 7544, (Changhai Hospital, Shanghai, China) and eight clinical MRSA isolates (Kunming General Hospital of Chengdu Military Region, Kunming, China) were used. All strains were identified according to standard morphological criteria. Prior to test, strains grew on sabouraud dextrose agar (SDA) at 30 °C for 24 h and later subcultured in YPD (1% yeast extract, 2% peptone and 2% dextrose) liquid medium at 30 °C in a shaking incubator for 16 h. RM (purity ≥ 96.8%) was provided by Yunnan College of Traditional Chinese Medicine. It was dissolved in dimethyl sulfoxide (DMSO) for in vitro study.

Antifungal Susceptibility Testing
The minimal inhibitory concentration (MIC) was determined by the Clinical and Laboratory Standards Institute methods (M27-A3 and M27-S4) [29,30]. Briefly, the initial concentration in RPMI 1640 medium was 10 3 CFU/mL for fungal suspension, and the final concentrations ranged from 4-256 μg/mL for RM. The 96-well microtiter plates were incubated at 35 °C for 24 h. The optical density at 630 nm (OD630) was measured to determine the growth inhibition. Each strain was tested in triplicate. The MIC50 was defined as the lowest concentration of the drugs that inhibited growth by 50% compared with the drug-free growth control.

Time-Growth Curves Assay
Exponentially growing C. albicans SC5314 were suspended in YPD medium to the starting OD600 of 0.2 (about 2.5 × 10 6 cells/mL). Then the suspensions were added in different concentrations of RM and cultured at 30 °C, 200 rpm. Cells were counted at the predetermined time points (0, 2, 4, 6, 8, 12, 24 h). Three independent experiments were performed [31].

Confocal Laser Scanning Microscopy (CLSM) Assay
CLSM was performed to determine the inhibitory effect of RM on biofilm formation. The confocal laser scanning special culture dishes were inoculated with C. albicans 5314 at 37 °C for 90 min to allow adhesion. Then biofilms were stained with 100 μg/mL of calcofluor white (Fluka, 18909) in the dark. After that, samples were washed with 2 mL PBS and observed under a Leica TCS SP2 confocal laser scanning microscopy (excitation wavelength 355 nm; emission wavelength 433 nm) [35].

Scanning Electron Microscopy (SEM) Assay
For SEM assay, the biofilms were formed on silicone pads in Spider medium at 37 °C for 90 min [36]. After removing non-adherent cells, and the fresh medium with different concentrations of RM was added. Then, the samples were incubated at 37 °C for 24 h. Subsequently, biofilms were placed in fixed liquid specific for electron microscopy (G1020, Goodbio Technology CO., Ltd, Wuhan, China) for 2 h and dehydrated in an ascending ethanol series and dried naturally. The dry specimens were coated with gold and observed through an EVO MA1O SEM (Carl Zeiss GmbH, Eppelheim, Germany).

Yeast-to-Hyphae Transition Assays
Exponentially growing C. albicans SC5314 were incubated with different concentrations of RM in hypha-inducing liquid medium at 37 °C for 6, 12, 24, 48 h respectively and on solid medium(Spider, Lee's, YPD + 10% FBS) at 37 °C for 5 days. The cells were observed and photographed by inverted phase contrast microscope (AMG ® EVOS xl, Life Technologies, Carlsbad, CA, USA).

Cellular Surface Hydrophobicity (CSH) Assay
CSH assay was based on water-hydrocarbon interface assay as described by Stephen A et al. [37]. The biofilms were harvested and suspended in YPD medium at OD600 of 1.0. A total of 1.2 mL of this suspension plus 0.3 mL octane was placed in a sterile glass tube and vortexed for 3 min. OD600 of the aqueous phase was recorded after settled for 10 min. OD600 of the suspension without the octane overlay was determined as control. Relative hydrophobicity (%) was obtained as [(OD600 of the control minus OD600 of the treatment)/OD600 of the control] × 100.

Real-Time RT-PCR
Real-time RT-PCR was performed according to the method described previously [38,39]. The biofilms of C. albicans 5314 were harvested and used for the total RNA extraction. Triplicate independent experiments were performed for each sample. The total RNA was extracted using Fungal RNAout kit (TIANDS, Beijing, China). 1st-strand cDNA was synthesized by reverse transcription reaction using a reverse transcription kit (TaKaRa, Biotechnology, Dalian, China). SYBR Green I (TaKaRa) and LightCycler Real-Time PCR system (Roche diagnostics GmbH, Mannheim, Germany) were used for Real-time PCR. Forward primer and reverse primer were listed in Supplementary Table S1. 18S rRNA played the role of internal control. The thermal cycling comprised four steps, an initial step at 95 °C for 2 min, 40 cycles at 95 °C for 10 s, 60 °C for 20 s, and 72 °C for 30 s. Finally, the gene expression level was calculated using the formula 2 −ΔΔCт .

Exogenous Cyclic AMP (cAMP) Supplement Assay
Exponentially growing C. albicans SC5314 cells were resuspended in Spider liquid medium at 5.0 × 10 5 CFU/mL. Then, different concentrations of RM with or without cAMP (10mM) were added. After that, the samples were incubated at 37 °C for 4 h. Simultaneously, 5.0 × 10 2 CFU/mL C. albicans SC5314 cells were colonized on Spider solid medium at 37 °C for 5 days. The cells were observed and photographed by inverted phase contrast microscope (AMG ® EVOS xl).

Cytotoxicity Studies
Three cancer cells (human gastric cancer cell line SGC-7901, human colon cancer cell line HT-29, human gastric cancer cell line MGC-803) and one normal cell (human umbilical vein endothelial cell) were employed in cytotoxicity tests. The cytotoxicity of RM was evaluated as the concentration of 50% cellular cytotoxicity (IC50) [40].

Toxicity Evaluation using C. elegans Worms
C. elegans glp-4; sek-1 adult nematodes were used to evaluate the toxicity of RM. Briefly, the nematodes were moved from Escherichia coli OP50 to pathogen-free liquid medium that contained different concentrations of RM or 1% DMSO. Then, the worms were incubated at 25 °C for 6 days and observed daily.

Conclusions
In the present study, we found that RM obviously inhibits yeast-to-hypha transition and the C. albicans biofilm formation in vitro, but it has no fungicidal effect on planktonic C. albicans cells, and the anti-biofilm mechanism may be related to the cAMP pathway. In addition, RM is a safe alkaloid for human non-cancer cell lines. Whether the anti-biofilm effect of RM is applicable in clinical management requires further study, and in vivo activities of RM against C. albicans need to be investigated.