Light Signaling Regulates Aspergillus niger Biofilm Formation by Affecting Melanin and Extracellular Polysaccharide Biosynthesis

ABSTRACT Light is an important signal source in nature, which regulates the physiological cycle, morphogenetic pathways, and secondary metabolites of fungi. As an external pressure on Aspergillus niger, light signaling transmits stress signals into the cell via the mitogen-activated protein kinase (MAPK) signaling pathway. Studying the effect of light on the biofilm of A. niger will provide a theoretical basis for light in the cultivation of filamentous fungi and industrial applications. Here, the characterization of A. niger biofilm under different light intensities confirmed the effects of light signaling. Our results indicated that A. niger intensely accumulated protective mycelial melanin under light illumination. We also discovered that the RlmA transcription factor in the MAPK signaling pathway is activated by light signaling to promote the synthesis of melanin, chitin, and other exopolysaccharides. However, the importance of melanin to A. niger biofilm is rarely reported; therefore, we knocked out key genes of the melanin biosynthetic pathway—Abr1 and Ayg1. Changes in hydrophobicity and electrostatic forces resulted in the decrease of biofilm caused by the decrease of melanin in mutants.

different concentrations of A. niger spores were quantified using crystal violet (CV) assays. The results showed that light with an increasing intensity from 1,000 lx to 4,000 lx could increasingly promote biofilm formation (Fig. 1A). There was a significant difference in optical density at 570 nm (OD 570 ) values when the light intensity measured above 2,000 lx. The difference was particularly obvious when the inoculum amount measured 10 5 and 10 4 spores/well. The biofilm showed a 9.4-fold and 13.5-fold increase under 3,000 lx and 4,000 lx (Fig. 1B), respectively, when the inoculum amount measured 10 4 . The morphology of A. niger under light or dark conditions was analyzed using a scanning electron microscope (SEM) to further verify this phenomenon. The result in Fig. 2A shows that the adhesion degree of the mycelium under light conditions is higher. In contrast, hyphae lines under dark conditions were clearer. Cryo-SEM was used to observe the samples more realistically to avoid the drying progress of conventional SEM. The ECM between the mycelium under light conditions was more abundant (Fig. 2B), which obviously differs from that under dark conditions. This finding also provides a foundation for the subsequent study of light signaling effects on biofilms.
In order to exclude the effect of light signaling on the growth of the mycelium, we conducted a normal growth experiment. As Fig. 2C shows, A. niger observed on the solid plate medium shows the same colony diameter size under light or dark conditions, but the phenotype showed darker and denser spores under light conditions. There is no significant difference in the dry weight of the mycelium collected at the end of the shake flask culture (results not shown), but there is a significant difference in the color of the mycelium (Fig. 2D). Thus, our evidence indicates that light exerts a minimal effect on the growth of A. niger. We also found that the aggregation degree of hyphae in the shake flask culture for 12 h increased under light conditions, as observed with a microscope (Fig. 2E).
The ECM was composed of galactomannan (GM), galactosaminogalactan (GAG), a-1,3-glucans, melanin, antigens, and hydrophobins (27). Polysaccharides are the main components. Therefore, we performed immunofluorescence assays on A. niger biofilm mycelium and extracellular polysaccharides and captured three-dimensional (3D) images using a confocal laser scanning microscope. Figure 3A shows the difference in polysaccharide content. Chitin, a-1,3-glucan, and b-1,3-glucan contents were niger wild-type spores were inoculated into a 24-well plate, incubated at 30°C in the dark, exposed to light intensity of 1,000 to 4,000 lx for 36 h, and then photographed after CV staining. (B) The corresponding OD 570 value in the 24-well plate. The values represent the means and standard deviations of three independent experiments. ***, P , 0.001; **, P , 0.01; *, P , 0.05; two-way ANOVA.
Light signaling affected A. niger melanin biosynthesis pathways by affecting the MAPK signaling pathway. It was reported that light signaling could regulate the MAPK signaling pathway and affect melanin formation (23,28). The expression levels of key genes in the MAPK signaling pathway and melanin biosynthetic pathway were investigated when A. niger was grown under light or dark conditions. Expression levels of Hog1, MpkA, and RlmA increased by 2.4-fold, 2.8-fold, and 3.2-fold, respectively, under light conditions (Fig. 4C). The expression of Abr1 and Ayg1 increased significantly, increasing by 4.1-fold and 5.4-fold under light, respectively (Fig. 4B). Through melanin content detection and immunofluorescence experiments, we observed a noteworthy difference in melanin levels when the amount of mycelium remained constant (Fig. 5A). We also observed a 26.1% increase in melanin under light conditions (Fig. 5B). Therefore, Abr1 and Ayg1 were knocked out to verify the fundamental role of melanin in biofilms.
In order to further prove that light signaling affects the high osmolarity glycerol (HOG)-MAPK and cell wall integrity (CWI)-MAPK pathways, Western blotting was used to detect the protein expression of Hog1 and MpkA and corresponding phosphorylation level. The key kinases involved in HOG-MAPK and CWI-MAPK pathways appeared increased under light conditions, and light signaling induced the phosphorylation of HOG-MAPK and CWI-MAPK (Fig. 4D). In addition, quantitative determination of Hog1p and MpkAp and the corresponding phosphorylation level also displayed a positive effect of light signaling on A. niger MAPK pathways (Fig. 4E).
A. niger DAbr1 and DAyg1 strains increased hydrophobicity and reduced zeta potential. After observing the same amount of DAbr1 and DAyg1 spores on a solid plate, the color change of the spores can be clearly observed after 48 h (Fig. 6A). The aggregation of spores is driven by electrostatic and hydrophobic interactions (29), which derive from the carboxyl group of the melanin layer and hydrophobin on the Light Signaling Regulates A. niger Biofilm Formation ® spore wall, respectively. Spore aggregation determines the initial stage of biofilm formation (7). Hydrophobicity experiments indicate that DAbr1 and DAyg1 are less hydrophobic than wild type (WT) (Fig. 6C), and the RodA expression level was also consistent with the results in Fig. 6B. The zeta potential test showed that DAbr1 and DAyg1 exhibited reduced negative charges compared with WT ( Fig. 6D).
DAbr1 and DAyg1 affected biofilm formation and cell wall integrity. We conducted CV assays to analyze the ability of mutants to form biofilms. The results showed that the biofilm-forming ability of DAbr1 and DAyg1 became weak and that Abr1C and Ayg1C restored the ability of biofilm formation to a certain extent ( Fig. 7A and B). It has been reported that melanin plays a role in maintaining the integrity of the cell wall (25,30). Congo red and Calcofluor white can be widely used as indicators of cell wall   defects. The sensitivity of mutants and WT to these two cell wall indicators were detected. The results clearly showed that DAbr1 and DAyg1 are more sensitive to Congo red and Calcofluor white (Fig. 7C). This result indicated that the lack of melanin alters the integrity of the cell wall. We observed lower levels of mycelium in DAbr1 and DAyg1 than in WT, using SEM (Fig. 7D). These results suggested that the degree of melanin reduction was positively correlated with the degree of biofilm reduction.

DISCUSSION
Biofilms play important roles in the environment and medical and industrial fields (31, 32), which promotes interest in biofilms as a research topic. However, research on biofilm mechanisms in the industrial field is relatively scarce. Here, we researched the effects of light on Aspergillus biofilm. Light is an important signal molecule in the environment, which regulates the physiological cycle, morphological changes, and fungal metabolites (15,33). Among bacteria, it has been reported that Acinetobacter baumannii could form biofilms under dark conditions, while the presence of blue light will inhibit the formation of biofilms (34). In fungi, there were also relevant reports showing that red or blue light affected the development of biofilms and the production of extracellular polysaccharides in Candida albicans (34,35). Although extensive studies have reported that fungi can sense light signaling, there were no direct reports on white light signaling and biofilms in filamentous fungi. Therefore, an attempt was made to analyze how light signals affect A. niger biofilm.
Several experiments were conducted to investigate the role of light signaling to determine whether it affected the biofilm of A. niger. First, a CV assay was performed. The effect of light signaling on the formation of A. niger biofilm was confirmed using Light Signaling Regulates A. niger Biofilm Formation this classic semiquantitative experiment. Second, ECM formation was observed under light conditions using cryo-SEM. Results showed that light signaling promotes the production of A. niger ECM and promotes A. niger biofilm formation.
In Aspergillus sp., the effect of GAG and a-1,3-glucan on biofilm has been widely reported (36,37). It is clearly pointed out that these polysaccharides are similar to cell wall components. a-1,3-Glucan plays a vital role in mycelial aggregation in biofilm, and GAG mediates many virulence-related characteristics, including adhesion to host cells and other substrates and the formation of biofilms. Immunofluorescence experiment results indicate that extracellular polysaccharide production increased under light conditions. a-1,3-Glucan content increased by 45.9%, and b-1,3-glucan and chitin appeared slightly increased. In addition, the expression of extracellular polysacchariderelated genes in biofilms increased by at least 1.3-fold and up to 2.75-fold. Simply put, light signaling promoted the production of polysaccharides in the ECM of A. niger biofilm.
Spores grown on the plate and the mycelium grown on the shake flask exhibited different colors under light or dark conditions. We estimated that melanin may play an important protective and component role in A. niger biofilm. Therefore, melanin content was determined, and genes playing key roles in the process of melanin biosynthesis were quantitatively analyzed. It was found that light can indeed promote the expression of genes related to the melanin biosynthesis pathway, resulting in elevated production of melanin. Therefore, we concluded that light signaling promotes melanin production.
This study mainly proves the positive effect of light signaling on biofilms, while also focusing on melanin. It has been reported that melanin is essential for the correct assembly of the conidial cell wall of A. fumigatus (25). Moreover, melanin not only was a virulence factor but also affected the normal functioning of other virulence factors, such as adhesins and hydrophobin (38). Reports regarding the relationship between melanin biosynthesis and biofilm formation in Aspergillus sp. are lacking. Here, we studied the effects of Abr1 and Ayg1 in the melanin biosynthesis pathway on biofilm. The results showed that the charge and hydrophobicity of DAbr1 and DAyg1 were reduced, correlating with the decreasing biofilm trend. Since melanin is an important component of the A. niger cell wall (25), we analyzed the susceptibility of mutants to Congo red and Calcofluor white and found that DAbr1 and DAyg1 are more sensitive to these two cell wall disrupters and that DAyg1, which produces less melanin, displayed higher sensitivity. The results of CV and SEM assays show that DAyg1 contains less biofilm than DAbr1. Therefore, our results indicate that a lack of melanin will severely affect cell wall integrity and biofilm formation.
Amplifying the external pressure signal through the MAPK signaling pathway may also enable light sensing in A. niger because light signaling represents an external pressure on fungi. In order to prove that light signaling governs MAPK signaling pathways, qRT-PCR and Western blotting were carried out. We found that the expression levels of Hog1 and MpkA were increased under light conditions by 2.4-fold and 2.8-fold, respectively. Next, the phosphorylation status of Hog1 and MpkA were investigated, and the results demonstrated that both protein expression and corresponding phosphorylation levels were increased. Furthermore, the proportion of phosphorylated MpkAp and Hog1p under light is significantly increased. Therefore, we proved that light signaling could activate HOG-MAPK and CWI-MAPK signaling pathways in A. niger. It was reported that the lack of RlmA leads to reduced biofilm formation in A. fumigatus (39). As a transcription factor downstream of the HOG-MAPK and CWI-MAPK signaling pathways, RlmA could mediate the synthesis of chitin and glucan, as well as the melanin biosynthesis pathway (40,41). It was also reported that RlmA could induce the expression of Ags1 in A. niger (22). To test the hypothesis that light promotes ECM production by increasing RlmA expression, we used qRT-PCR and observed that the RlmA transcript increased 3.2-fold. Overall, this study confirmed that light signaling could activate the HOG-MAPK and CWI-MAPK signaling pathways and, secondly, could activate the expression of RlmA transcription factors to promote the production of melanin and polysaccharides in A. niger. Therefore, we have drawn a schematic diagram indicating the light signaling-mediated MAPK signaling pathway that regulates A. niger biofilm (Fig. 8).
In conclusion, we found that 3,000 lx light intensity has little effect on the normal growth of A. niger, but it will cause visible differences in the color of spores and hyphae. A series of biofilm characterization experiments confirmed that light could promote the formation of ECM of biofilms, and protective mycelial melanin played a role in resisting light stress. This was the first time that the connection between light signaling and biofilm has been established in A. niger. To our knowledge, the promotion of biofilm formation in fungi by light signaling is also a novel finding. Nevertheless, it is a pity that this discovery has not been applied to fermentation, but it will also be the goal of subsequent research. In the long-term evolution process, the fungus has formed an extremely fine and complete photoreceptive system, which can sense the presence or absence of light, the direction and quality of light, the intensity of light, and the length of the photoperiod, in order to better adapt to the environment. Therefore, it will be interesting to observe the effects of other factors, such as light wavelength, light intensity, light direction, and light period, on the formation of A. niger biofilm.

MATERIALS AND METHODS
Strains and media. In the present study, A. niger SJ1 (WT) was preserved in China Center for Type Culture Collection (Wuhan, China) under the deposit number CCTCC M201911. A. niger SJ1 was cultivated on defined minimal medium (MM) (42), yeast extract-peptone-dextrose (YPD) medium, or peptone-dextrose agar (PDA) medium. For the knockout gene, 60 mg/ml of hygromycin B (31282; Sangon Biotech, China) was added to PDA medium as a hph selection marker, and 10 mM acetamide and 15 mM cesium chloride were used in the MM instead of NaNO 3 for the amdS selection marker for the Light Signaling Regulates A. niger Biofilm Formation ® complement gene. Abr1 and Ayg1 knockout fragments were constructed using the hph gene as a selection marker. A complementing plasmid was constructed using A. nidulans amdS as a selection marker. The original plasmid was pAN-amdS, constructed by replacing the hph gene in pAN7-1 with the amdS gene, which was preserved in our laboratory. The gpdA gene was used as a promoter and the trpC gene was used as a terminator to construct a complementary plasmid expressing Abr1 and Ayg1, respectively. Gene knockout fragments and complementary plasmids were transformed into mutant strains by polyethylene glycol (PEG)-mediated protoplast transformation. The mutant strains used in subsequent experiments were stably passaged more than five times and were verified by colony PCR and qRT-PCR analysis. A schematic diagram of DNA fragments and plasmid construction is shown in Fig. S1 and S2 in the supplemental material. The PCR primers and verification results are indicated in Tables S1 to S3 and Fig. S3 in the supplemental material.
Biofilm formation on plastics and normal growth on medium. A. niger biofilm formation was evaluated using the classic CV assay, performed as previously described with minor modifications (43). Fresh A. niger conidia (10 6 , 10 5 , 10 4 , and 10 3 ) were inoculated on a 24-well microtiter plate (Corning, USA) with 1-ml MM and cultured for 36 h at 30°C under 1,000-lx, 2,000-lx, 3,000-lx, and 4,000-lx light intensity, with dark as a contrast. Subsequently, the plate was washed three times with phosphate-buffered saline (PBS) to remove free mycelium. Thereafter, the biofilm was stained with 1 ml 0.1% crystal violet solution at room temperature for 10 minutes, and wells were repeatedly washed with PBS and air dried. Acetic acid (1 ml; 100%) was added to each well, and wells were gently shaken at room temperature for 30 minutes to elute crystal violet. Finally, a microplate reader (SpectraMax Paradigm) was used to read the absorbance at 470 nm. The CV assay was also used to analyze differences in biofilm-forming ability between mutants (DAbr1, DAyg1, Abr1C, and Ayg1C) and WT A. niger. Next, the OD 470 was determined, as mentioned above. The light intensity used in subsequent light experiments was 3,000 lx.
WT A. niger spores (10 6 ) were one-point inoculated on PDA solid medium, cultured at 30°C for 24 h under light or dark conditions, and then observed. WT A. niger spores (10 7 ) were inoculated into a 500ml conical flask containing 50 ml YPD liquid medium and cultured at 30°C and 220 rpm under light or dark conditions. Sampling was conducted after 12 h to observe the growth difference using an optical microscope. After 24 h, the mycelia were removed and washed thrice with PBS, and differences were observed. WT and DAbr1 and DAyg1 spores (10 6 ) were spot-planted on PDA solid medium, cultured at 30°C for 48 h, and observed.
Determination of melanin, a-1,3-glucan, b-1,3-glucan, and chitin content in mycelium. Samples cultured in 24-well plates for 36 h under light or dark conditions were removed and washed thrice with PBS to remove free hyphae. Biofilm hyphae were collected using physical methods and ground into a powder after liquid nitrogen quick freezing and vacuum freeze drying. Specimens were homogenized thoroughly using a homogenizer after adding 1-g hyphae powder to 9-g PBS (pH 7.2 to 7.4). Samples were centrifuged for 20 minutes (3,000 Â g/min), and supernatants were carefully collected. The Microorganism Melanin ELISA kit (Jiangsu Baolai Biotechnology, Nanjing, China) was used to determine melanin content, the microorganism a-glucans ELISA kit was used to determine a-1,3-glucan content, and the microorganism b-flucans ELISA kit was used to determine b-1,3-glucan content in the biofilm (44)(45)(46). Chitin was measured as previously described (43).
SEM analysis and cryo-SEM analysis. WT A. niger spores (10 6 ) were inoculated into 24-well plates containing 8-mm round coverslips and 1-ml MM and then cultured for 36 h at 30°C under light or dark conditions. Subsequently, the round coverslip was washed three times with PBS to remove free mycelium and then removed to obtain samples. Sample morphologies were observed using the FEI Nova nano-scanning electron microscopy 450 (Nova NanoSEM, Prague, Czech Republic) after liquid nitrogen quick freezing, vacuum freeze drying, and gold spraying.
In order to reduce the influence of drying and dehydration on the observation of samples, we also adopted cryo-SEM to observe the samples. Samples were obtained as mentioned above. Processed samples were loaded into the scanning electron microscope (SU8010) refrigerated transport system (pp3010T). After being precooled with liquid nitrogen for 5 minutes, the sample was transferred to the freezing sample preparation chamber, vacuumed, and sublimated at 270°C. After gold spraying, the sample was transferred to a scanning electron microscope for observation and photographing (47).
In order to compare the difference between the mutants (DAbr1 and DAyg1) and WT A. niger, we also used SEM to observe the samples, using the method mentioned above.
Protein extraction and Western blotting. As previously described, fresh conidia were incubated in 50-ml YPD medium at 30°C in a rotatory shaker (220 rpm) for 24 h under light or dark conditions. The mycelia collected by filtration were washed thrice with PBS and then quick frozen in liquid nitrogen. The lysis solution was added for protein extraction. The mycelial powder was obtained using a homogenizer and collected into centrifuge tubes. Protein extraction buffer (0.8 ml) containing protease inhibitors and 1 mM phenylmethylsulfonyl fluoride (PMSF) were added into each centrifuge tube and incubated on ice for 20 min. The supernatant was collected via centrifugation, and the protein content was measured using the bicinchoninic acid (BCA) assay. Subsequently, protein samples of the same concentration were loaded onto a 12% (wt/vol) SDS polyacrylamide gel and transferred to a nitrocellulose membrane. The phosphorylation of Hog1 and MpkA was examined using anti-phospho-p38 MAP kinase antibodies and anti-phospho-p42 antibodies (Cell Signaling Technology, USA; dilution, 1:1000). Anti-Hog1p C-terminal antibody (Santa Cruz Biotechnology, USA; dilution 1:500) and anti-p42 antibodies were used against Hog1 and MpkA, respectively. A b-Actin antibody was used as a control, and anti-rabbit IgG (whole molecular) peroxidase antibody (Sigma-Aldrich) was used to detect all primary antibodies, except anti-Hog1p C-terminal antibody. Goat anti-mouse IgG (H&L) horseradish peroxidase (HRP) was used to detect the anti-Hog1p C-terminal antibody (17,48). The experiment was repeated three times.
Immunofluorescence. Samples were fixed with 1.5% glutaraldehyde for 1.5 h and then stained with propidium iodide (PI; Sigma; excitation wavelength, 535 nm; emission wavelength, 615 nm) and fluorescein isothiocyanate-labeled concanavalin A (FITC-ConA; Sigma; excitation wavelength, 495 nm; emission wavelength, 515 nm) in the dark. PI embedded in double-stranded DNA fluoresces red, while FITC-ConA binds to glucose and mannose residues of cell wall polysaccharides and fluoresces green. Confocal laser-scanning microscopy graphics were captured using a Leica TCS SP5 II instrument.
SP-D, a C-type lectin, binds to melanin pigment on the surface of Aspergillus sp. (49,50). Glutaraldehyde-fixed A. niger biofilm samples were incubated with surfactant protein D (SP-D) (1 h; 37°C ). Next, samples were washed twice and incubated overnight with primary anti-human SP-D antibody (5 mg/ml; R&D Systems) at 4°C, then washed and incubated with secondary anti-mouse IgG-FITC (dilution 1:200; Sigma) on ice for 1 h, and finally observed using a fluorescence microscope.
Hydrophobicity assay and zeta potential determination. The hydrophobicity of A. niger spores was tested according to the method described previously. Glyceryl tributyrate (50 ml) was added to a 50-ml suspension containing 10 8 spores in a 1.5-ml centrifuge tube and then observed after incubation at room temperature for 24 h. The color depth in the hydrophobic layer indicates the hydrophobicity of the spores (43,51).
qRT-PCR analysis. The mycelium inoculated on a 24-well plate under light or dark conditions was washed with PBS thrice to remove floating hyphae, and mycelium was then collected using physical methods. Next, RNA was extracted from mycelia using the TaKaRa MiniBEST universal RNA extraction kit, according to the manufacturer's instructions. cDNA was obtained via reverse transcription of RNA according to the HiScript III RT SuperMix for qPCR (1gDNA wiper) (Vazyme, China) instructions. The qRT-PCR was performed on a StepOnePlus real-time PCR system (Applied Biosystems, USA) using 2Â ChamQ universal SYBR qPCR master mix (Vazyme, China). The reaction and calculations were performed according to standard protocols. Each sample had three parallel and negative controls. The expression level of the gene was calculated according to the comparative threshold cycle (2 2DDCT ) method. Actin was chosen as the internal reference gene (53), Primer 5 software was used to select primers, and primers used for analysis are listed in Table S3.qRT-PCR experiments were performed, as indicated above, to verify the difference in hydrophobicity between WT and DAbr1, DAyg1, Abr1C, and Ayg1C and to verify the expression of the hydrophobin RodA.
Statistical analysis. All experiments were done at least in triplicate. Statistical significance was analyzed by one-or two-way analysis of variance (ANOVA) or paired t test with Prism software (GraphPad software, San Diego, CA). The values are the means and standard deviations of three independent experiments and P values generated by the Student's t test (n.s., not significant at P . 0.05; ***, P , 0.001; **, P , 0.01; *, P , 0.05).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.