Adh1‐Programmed SNF1 Phosphogradients Decrypt Morphogenesis in Candida albicans: Chemical Interrogation Unveils Hyphal Transition Thresholds

ABSTRACT

While Candida albicans alcohol dehydrogenase I (Adh1) conventionally functions as an alcohol dehydrogenase, this study builds upon previous work to redefine its novel role in regulating hyphal morphogenesis and elucidates the underlying mechanisms. Adh1 knockout strains exhibited hyperfilamentation, suggesting that Adh1 directly regulates true hyphal development, independent of its canonical metabolic activity. Leveraging this phenotype, a ‘reverse screening strategy’ identified 2‐hydroxyanthraquinone (HAQ) through high‐throughput screening as a potent inhibitor of biofilm formation and hyphal growth by targeting Adh1. Biochemical and structural analyses confirmed HAQ's direct binding to Adh1's F224/A254/Q257 interface. Mechanistically, affinity purification‐mass spectrometry revealed Adh1 modulates the SNF1 signalling axis by accelerating SNF1 dephosphorylation via interactions with Bmh1/Ssb1 regulators, thereby inhibiting hyphal conversion. HAQ disrupted these interactions, reducing SNF1 phosphorylation levels in an Adh1‐dependent manner. This work establishes Adh1 as both an endogenous SNF1 pathway suppressor and an exogenous drug target, while demonstrating the efficacy of phenotype‐driven discovery pipelines. The findings provide a novel antifungal strategy targeting virulence‐regulating metabolic enzymes and validate HAQ as a lead compound for therapeutic development against C. albicans pathogenicity.

Based on the ADH1 knockout strain, HAQ was identified through forward/reverse genetic screening of 114 natural products and analysis of protein ligand and protein–protein interactions. With the help of HAQ, a new mechanism for regulating the phosphorylation level of Adh1–SNF1 has been revealed, in which HAQ binds to Adh1 to release dephosphorylation factors Bmh1 and Ssb1, thereby reducing the phosphorylation level of SNF1 and inhibiting hyphal growth.

1. Introduction

Candida albicans is a pathogenic fungus commonly found in the human digestive tract, urinary system, and other ecological sites, usually triggering infections when the immune system is weakened or mucous membranes are damaged (Kumar and Kumar 2024). Although existing drugs such as azoles can control acute infections, their efficacy is fundamentally limited by dose‐dependent toxicity and fungal resistance (Villa et al. 2020). This is particularly evident in invasive candidiasis, underscoring the urgent need for adjunctive therapies targeting alternative mechanisms. Therefore, it is urgent to break through the bottleneck of the original fungicidal strategies to develop novel antifungal approaches.

The pathogenicity of C. albicans is closely related to its morphological switching. Studies have shown that C. albicans can switch between yeast and pseudohyphae or true hyphae, which can cause severe infections (Jacobsen et al. 2012). In this case, pseudohyphae have the dual characteristics of yeast and hyphae, with the progeny remaining attached to the parent after cytoplasmic division, causing the intercellular junctions of the filaments to be concave inwards (Malinovská et al. 2023). In contrast, true hyphae always maintains close end‐to‐end contact, with elongated tubular structures separated only by a septum that does not form a significant constriction (Talapko et al. 2021). During pathogenesis, the yeast form is more adept at spreading by blood diffusion, whereas the hyphal form penetrates tissues, evades the immune system and exacerbates systemic infections (Jacobsen et al. 2012; Kornitzer 2019). If the morphological transformation is artificially restricted (e.g., fixed to a single form), its virulence will be significantly reduced (Lo et al. 1997; Murad et al. 2001; Zheng et al. 2004; Cheng et al. 2025; Futamura et al. 2025; Gao et al. 2025). Although it has been reported that environmental factors such as temperature, pH and nutritional conditions can regulate the expression of morphology‐related genes through specific signalling pathways (e.g., cAMP‐PKA, Rim101, etc.) (Chen et al. 2020; Iracane et al. 2021), two major research gaps still remain: first, no drug targets have been identified to directly intervene in the transition between the three morphologies, and second, despite tremendous research efforts aimed at elucidating regulatory networks, the exact mechanisms driving morphological switching still are not fully resolved. One particular difficulty is that the filamentation induction networks are differ between C. albicans strains. Unravelling the molecular details of this process will provide an important breakthrough in the development of novel therapies against drug‐resistant Candida infections.

Screening of compound libraries through genetically modified fungal models can efficiently identify potential drugs that modulate the function of pathogenesis‐related proteins (Choi and Kim 2024; Gan et al. 2025). Based on this understanding, the present study focused on alcohol dehydrogenase I (Adh1), which is not only involved in basic metabolism such as glycolysis but has also been found to be closely related to the processes of fungal pathogenicity enhancement, drug resistance formation and immune escape (Wang et al. 2024b). Although previous studies have confirmed that Adh1 has a regulatory role in biofilm formation and hyphal development (Mukherjee et al. 2006; Chauhan et al. 2011; Song et al. 2019), there are two major barriers: firstly, the precise regulator for this protein has not yet been found, and secondly, the concrete mechanism by which it regulates biofilm formation still remains at the level of gene expression. Deciphering the Adh1‐mediated signalling will not only reveal the deeper mechanism driving the fungal polymorphic transition but will also provide a key breakthrough for the development of novel antifungal therapies targeting biofilms.

In addition, pathogenicity of C. albicans is closely linked to carbon metabolism regulation. The SNF1 complex functions as a metabolic sensor and regulator, serving as a critical switch that enables the fungus to adapt to varying carbon sources across different host ecological microenvironments (Mottola et al. 2020). During glucose deficiency, Thr208 of the SNF1 α‐subunit is phosphorylated by the upstream kinase Sak1, which promotes the export of Mig1 from the nucleus and initiates the utilisation of alternative energy sources (Lagree et al. 2020; Ramírez‐Zavala et al. 2021). In Saccharomyces cerevisiae (a commonly used research model), Reg1 will conduct the phosphatase Glc7 to switch off SNF1 activity when glucose is sufficient (Ramírez‐Zavala et al. 2021; Schnell et al. 2021). At this point, the auxiliary proteins Ssb and Bmh form a special composite structure, which function as an adaptor complex that bridges SNF1 and Glc7, facilitating their precise interaction. This regulatory module is essential for maintaining fungal energy homeostasis. (Hübscher et al. 2016). Studies have shown that SNF1 signalling not only regulates metabolic adaptation, but also is directly related to hyphal morphological transformation. It controls key pathogenic behaviours such as biofilm formation and tissue invasion through downstream factors (e.g., effector proteins including Flo11), serving as a bridge between metabolism and infectivity (Berkey et al. 2004; Vyas 2004). Although the relevance between SNF1 and the regulation of filamentation has been identified, the exact mechanism is still unanswered.

Naturally‐derived small molecule compounds play a pivotal role in the competitive or symbiotic relationship between plants and microorganisms (Ruotsalainen et al. 2022). With unique structural features and a wide range of molecular properties, natural products are often an important source of new lead compounds in drug discovery (Najmi et al. 2022). Unfortunately, the biological functions of numerous natural products have not yet been identified due to the limitations of existing activity screening platforms. In this study, in order to gain insights into the specific functions and mechanisms of Adh1 in C. albicans polymorphic transition and to try to obtain lead molecules from natural products that affect polymorphism by targeting Adh1, we first constructed a C. albicans hyphae ADH1 knockout strain. On the basis of clarifying that Adh1 blocks true development, 2‐Hydroxyanthraquinone (HAQ), a natural compound that inhibits polymorphic transformation in C. albicans relying on Adh1, was screened based on the difference in the effect of the chemical molecule on biofilm formation in wild‐type and ADH1 knockout strains. Subsequently, this study confirmed that HAQ was able to interact directly with Adh1 and explored the SNF1 module whose interaction with Adh1 is regulated by HAQ using protein affinity chromatography‐mass spectrometry. Finally, we modulated SNF1 activity with the help of knockdown technology and functional enhancers and demonstrated that low, medium and high levels of SNF1 phosphorylation directed C. albicans to maintain yeast, pseudohyphae, or true hyphae morphology, respectively, and that Adh1 mediated the development of C. albicans into pseudohyphae under biofilm‐inducing conditions by restricting the over‐activation of SNF1 signalling. In conclusion, HAQ targets C. albicans Adh1 and blocks hyphae formation by modulating Adh1‐SNF1 signalling. Our study suggests that Adh1‐SNF1 signalling is a valid regulatory element of the C. albicans filamentation process and that HAQ may serve as a precursor for the development of signalling‐axis inhibitors, with a view to developing potent drug molecules to inhibit C. albicans filamentation in the future.

2. Result

2.1. Adh1 Inhibits True Hyphae Formation Within Biofilms

Biofilms provide a stable protective ecology for C. albicans and are the primary cause of recurrent Candida infections (Desai et al. 2014). The construction of a mature biofilm involves four stages: adhesion, germination, maturation and dispersion (Figure 1a) (Wall et al. 2019), of which hyphae formation is a central aspect of invasive infection (Vila et al. 2017). Given the correlation between Adh1 and C. albicans biofilm formation, we constructed ADH1 knockout strain (Figure S1a,b) and ADH1 gene overexpression strain (Figure S1c) based on homologous recombination strategy. By microscopic observation, it was found that high levels of Adh1 inhibited hyphae from extending on the surface of the Spider medium or expanding into the interior of the medium. In addition to this, it also mediated the derivation of more branched and lateral yeast cells from pseudohyphae in RPMI‐1640 medium (containing 10% FBS). In contrast, the absence of ADH1 prompted the fungus to differentiate into true hyphae and invade deeper into the agar (Figure 1b). Consistent with the phenotype, mRNA levels of C. albicans hyphae‐associated genes (including ECE1, HWP1, BCR1 and TEC1) also changed with altered expression of Adh1 (Figure 1c). Thus, C. albicans Adh1 is an effective repressor of true hyphae formation, giving the strain ample polymorphic flexibility under biofilm‐inducing conditions.

FIGURE 1.

HAQ inhibits hyphal development within biofilms via Adh1. (a) Schematic diagram of the complete life cycle for C. albicans biofilm. (b) Hyphal extension and penetration of wild‐type, ADH1 knockout and ADH1 overexpression strains on the surface of Spider medium, as well as microstructures of hyphae from three strains on the surface of polystyrene material in RPMI‐1640 medium (containing 10% FBS). (c) Effect of Adh1 on hyphae‐specific gene expression. The three strains were inoculated in RPMI‐1640 medium (containing 10% FBS) and qRT‐PCR assay was performed after 6 h. (d) Schematic diagram of biofilm inhibitor screening with Adh1 as a functional target. (e) Scattered statistics of the inhibition efficiencies for 114 natural products (30 μg/mL) on biofilm formation of the wild‐type and ADH1 knockout strains. (f) Under biofilm‐inducing conditions, the total amount of biofilms formed by WT on the solid substratum was measured at sequential time points, both before and after treatment with HAQ. (g) After treatment with increasing concentrations of HAQ for 24 h in RPMI‐1640 medium (supplemented with 10% FBS), the total fungal burden of the WT in the medium and the biofilm amount on the substratum were quantified. (h) Crystal violet staining to quantify the total amount of biofilm from wild‐type, ADH1 overexpressing and ADH1 knockout strains under HAQ treatment. (i) Morphology of hyphae within the mature biofilm of the three strains on the polystyrene substrate surface after co‐incubation with HAQ (20 μg/mL) for 24 h was observed by ordinary optical microscope and scanning electron microscope. For Figures c, f, g, h, data are presented are means ± SD. Statistical significance was determined by two‐tailed unpaired Student's t‐test. n = 3; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared to control group.

2.2. Screening of Compounds Potentially Targeting Adh1

Subsequently, we screened a natural product library covering multiple types of compounds based on the quantification of mature biofilms from wild‐type and ADH1 knockout C. albicans (Figure 1d). Given the target‐dependent nature of drug action, the pharmacological efficacy of compounds is expected to alter when the protein target is absent. Therefore, monitoring changes in biofilm inhibition rates of natural products before and after ADH1 deletion represents a critical strategy for identifying Adh1 functional modulators. Primary screening of 114 natural products identified 15 hits that preferentially inhibited WT over ADH1‐KO biofilms. Among these, anthraquinones constituted the most enriched class (Figure S1d). The results of the assay showed that anthraquinones such as 1, 4‐Naphthoquinone (109), 2‐Methylanthraquinone (110), Aloe emodin (114) and HAQ (113) showed significantly reduced biofilm inhibition after ADH1 deletion (Figure 1e, Figure S1d, Table S1). When we further assessed the temporal and quantitative efficacy of the above compounds on biofilm inhibition in the WT, all four anthraquinones exerted significant potency in the full cycle of biofilm construction with well dose dependence (Figure 1f,g, Figure S1e–g). Among them, HAQ and 1, 4‐naphthoquinone exhibited the most excellent activities, downregulating the total biofilm biomass by 97.3% and 71.8% at 40 μg/mL, respectively. Notably, 1, 4‐naphthoquinone demonstrates significant fungicidal activity, but high concentrations of HAQ did not show significant inhibition of C. albicans growth, which is consistent with the fact that ADH1 is not involved in the regulation of growth kinetics of the strain (Song et al. 2019). Finally, we examined the Adh1 dependence of the biofilm inhibitory effect exerted by HAQ. Compared to controls, knockout of ADH1 weakened the restriction of mature biofilm formation by HAQ (Figure 1h), at which point HAQ was unable to prevent the development of intra‐biofilm hyphae into true hyphae (Figure 1i). In contrast, overexpression of ADH1 was an effective contributor to the inhibitory activity of HAQ, suggesting that Adh1 is a pharmacological target of HAQ. In conclusion, HAQ inhibits hyphal development within C. albicans biofilms via Adh1.

2.3. Adh1 Is a Direct Target of HAQ

Previous studies have reported the activity of natural anthraquinones in antifungal and anti‐biofilm activities (Tsang et al. 2012; Marioni et al. 2016; Manoharan et al. 2017), but the targets of the compounds and related mechanisms have not been proposed. In this study, the dependency of HAQ on Adh1 for the inhibition of hyphal‐yeast transition attracted us to investigate its potential targets. We flowed a whole protein solution of C. albicans through an epoxy‐activated column immobilised with emodin, an anthraquinone compound, and proteins capable of interacting with anthraquinone were sequestered by emodin. Emodin was used as a representative anthraquinone to capture proteins with potential affinity for this class of compounds. This was followed by competitive elution of HAQ‐specific binding proteins with HAQ solution (Figure 2a, Figure S2a). In the mass spectrometry identification results, the top three proteins in terms of abundance were transcription factor Tup1, enolase Eno1 and alcohol dehydrogenase Adh1, using MaxQuant (1.6.2.10) scoring of greater than 50 as a plausibility criterion (Figure S2b, Table S2). These results suggest that all three aforementioned proteins may interact with HAQ. As mass spectrometry cannot characterise direct compound‐protein binding, we subsequently performed molecular docking simulations followed by in vitro experimental validation.

FIGURE 2.

Adh1 is a direct target of HAQ. (a) Schematic of the epoxy‐activated column chromatography method used to search for HAQ‐interacting proteins. (b) Docking simulation using Autodock software to demonstrate the preferential binding mode of HAQ with Adh1. (c) Differential scanning fluorimetry assay to detect melting temperatures of recombinant Adh1 after co‐incubation with increasing concentrations of HAQ. (d) Tryptophan fluorescence spectroscopy to detect changes in the wavelength of tryptophan‐emitting fluorescence from recombinant Adh1 in the presence of HAQ. (e) MST curves and dose–response curves for the calculation of the affinity constants between HAQ and fluorescently labelled Adh1 by the microScale thermophoresis. (f) Effect of HAQ on Adh1 intracellular stability in a hyphae‐inducing environment (RPMI‐1640 with 10% FBS). Protein blotting of Adh1 at consecutive time points before and after treatment with 40 μg/mL HAQ in the presence of 400 ng/mL anisomycin.

First, we predicted the binding of HAQ to the top 10 scoring proteins using molecular docking. The results showed that Tup1 and Adh1 were able to bind to HAQ in a stable conformation with binding energies of −8.5 Kcal/mol and −8.0 Kcal/mol, respectively (Figure 2b, Figure S2c). Considering that Adh1, which is the focus of this study, is also a functional target of HAQ, Adh1 was initially identified as an intracellular target of HAQ for biofilm inhibition.

Then, we obtained recombinant Adh1 by heterologous expression and verified the interaction of HAQ with Adh1 in vitro. In the differential scanning fluorimetry assay (Gao et al. 2020), the presence of HAQ caused Adh1 to exhibit only one melting temperature in a temperature gradient field, and the Tm value increased with increasing drug concentration (Figure 2c). Similarly, in the tryptophan fluorescence spectroscopy assay (Sindrewicz et al. 2019), the wavelength of the emission fluorescence from tryptophan was progressively red‐shifted as the concentration of HAQ within the system was elevated (Figure 2d). The above results indicate that the co‐incubation with HAQ not only enhanced the polarity of the tryptophan surroundings in Adh1 but also improved the thermal stability of Adh1, and there was a direct interaction between them. In order to obtain further information on the affinity between HAQ and Adh1, the affinity constant between the two was then evaluated using microscale thermophoresis (Seidel et al. 2013). The binding of HAQ to Adh1 resulted in a change in the thermophoresis rate of Adh1 within the infrared laser heating field, and the affinity magnitude of the two was obtained from the fitting as 111.86 μM (Figure 2e).

Finally, we investigated the effect of HAQ on Adh1 stability in situ based on biophysics (Zheng et al. 2022). Using the ribosomal A‐site inhibitor Anisomycin (Zgadzay et al. 2022) to block protein translation within C. albicans , the hyphae‐induced environment would contribute to a time‐dependent decrease in the intracellular levels of Adh1 (Figure S2d). In contrast, under the same conditions, co‐incubation of the fungal culture with HAQ prevented the degradation of Adh1 that typically occurs in the hyphae‐inducing environment, resulting in stable levels of the protein. (Figure 2f). This result implies that HAQ entering C. albicans will interact directly with Adh1 and enhance the stability of Adh1. Taken together, these results suggest that C. albicans Adh1 is a direct target of HAQ.

2.4. F224, A254, Q257 Are Key Residues for HAQ‐Adh1 Binding

To thoroughly analyze the compound‐protein interactions, we simulated the direct binding conformation of HAQ to Adh1 by molecular docking. The results showed that a total of 11 amino acids in Adh1 could potentially bind to HAQ. Among them, the compounds form covalent hydrogen bonds with G180, S179 and Q257 via the carbonyl and hydroxyl groups, and form π‐π conjugation and σ‐π hyperconjugation with F224 and A254 via the thickened trisubstituted benzene ring, respectively (Figure 3a). Next, we selected the top 10 docking conformations and counted the frequency of the 11 interacting amino acids. F224 and A254 appeared with the highest frequency, followed by S179, Q257, G180, S249 (Figure 3b).

FIGURE 3.

F224, A254, Q257 are key residues for HAQ‐Adh1 binding. (a) Prediction of the amino acid sites where HAQ binds to Adh1 by molecular docking. (b) The top 10 ranked conformations of Adh1 docked with HAQ were selected and frequency distribution plots of amino acids interacting with the compounds were counted. (c) Schematic representation of the amino acid mutation site of the recombinant mutant Adh1. (d) In vitro detection of affinity between HAQ and Adh1 with mutated binding sites by microscale thermophoresis.

To make a validation of the above sites, we constructed a series of ADH1 heterologous expression vectors (Figure S3a) using F224 and A254 as the splitting point (Figure 3c) to obtain Adh1 mutated at the potential amino acid site for binding to HAQ (Figure S3b). On this basis, the interaction between HAQ and Adh1 mutants was examined using microscale thermophoresis. The results showed that F224A caused the binding of Adh1 to HAQ to be undetectable. Compared with wild‐type Adh1, A254T and Q257A resulted in a significant decrease in the affinity of Adh1 for HAQ, with affinity constants of 528.48 and 4990 μM, respectively. In contrast, the remaining mutant forms exhibited only a relatively modest effect (Figure 3d). The above results indicate that F224, A254 and Q257 of Adh1 are essential for the binding between HAQ and Adh1.

2.5. Adh1 Is a Key Intermediate in the Regulation of SNF1 Activity by HAQ

To analyse the downstream signalling network of Adh1 regulating morphological transitions in C. albicans , this study systematically compared the reciprocal proteomic differences of Adh1 under HAQ and DMSO treatment conditions using affinity purification‐mass spectrometry (AP‐MS) (Figure 4a; Figure S4a,b). By integrating bioinformatics analysis and literature mining (Berkey et al. 2004; Vyas 2004; Hübscher et al. 2016; Lagree et al. 2020; Ramírez‐Zavala et al. 2021; Schnell et al. 2021), it was found that Adh1 specifically co‐enriches the fungal carbon metabolism core regulator SNF1 kinase and key components of the complex regulating SNF1 dephosphorylation, while HAQ significantly reduced the strength of the interactions between these regulatory elements and Adh1 (Figure 4b). It is implied that Adh1 may block true hyphae development through SNF1 signalling associated with fungal biofilm formation, while HAQ regulates the activation of the SNF1 signalling pathway by targeting the destabilizations of the Adh1‐SNF1 complex.

FIGURE 4.

Adh1 is a key intermediate in the regulation of SNF1 activity by HAQ. (a) Schematic of the process for the detection of Adh1‐interacting proteins by protein affinity chromatography‐mass spectrometry. (b) Information related to the SNF1 complex in Adh1‐interacting proteins. Mass spectrometry quantified the abundance of the corresponding proteins in the HAQ administration group and in the control group, normalised to Adh1 and the ratio calculated. (c) Effect of Adh1 on SNF1 phosphorylation. Immunoblotting of intracellular SNF1 and its phosphorylation levels in the three strains after treatment with biofilm‐inducing conditions for 4 h. (d) Co‐IP detection of interactions between Adh1 and Ssb1 or Bmh1 before and after HAQ treatment. Flag‐Ssb1 or Flag‐Bmh1, His‐Adh1 were expressed in S. cerevisiae , respectively, and the target proteins were pulled down by anti‐His or anti‐Flag antibodies. (e) Immunoblotting of intracellular levels of SNF1 and its phosphorylation after treatment of wild‐type or ADH1 knockout strains for 6 h under biofilm‐inducing conditions with increasing concentrations of HAQ. The band at the arrow‐marked molecular weight is the phosphorylated SNF1 Thr208, and the asterisk denotes a band from other phosphorylation sites antibody binding.

To clarify the regulatory role of Adh1 on SNF1 phosphorylation, functional validation was systematically carried out by overexpressing and knocking out ADH1. The results showed that under biofilm‐induced conditions, the phosphorylation level of SNF1 in the ADH1 knockout strain was significantly elevated compared with that in the wild‐type and overexpression strains (Figure 4c), confirming that Adh1 is a negative regulator of SNF1 activity. To further elucidate the molecular interaction network, this study constructed His‐Adh1 and Flag‐Bmh1/Flag‐Ssb1 co‐expression system in S. cerevisiae BJ5464‐NpgA (Figure S4c), and HAQ was found to specifically weaken the binding strength of Adh1 to SNF1 dephosphorylating regulatory elements (including Bmh1 and Ssb1) (Figure 4d). Notably, HAQ significantly reduced the level of SNF1 phosphorylation in a dose‐dependent manner, and this effect was significantly attenuated in ADH1‐deficient strains (Figure 4e), suggesting that Adh1 is an indispensable target molecule for drug action. In summary, HAQ blocked SNF1 signalling activation in an Adh1‐dependent manner by competitively inhibiting the binding of Adh1 to the SNF1 phosphatase regulatory complex, thereby accurately regulating the process of fungal morphological transformation.

2.6. HAQ Suppresses the Hyphal Transition Through Adh1‐SNF1 Signalling

Sak1 is the only dedicated SNF1‐activated kinase in C. albicans , and when Sak1 is absent, SNF1 retains basal activity to maintain strain survival (Ramírez‐Zavala et al. 2017). To examine the regulatory effect of SNF1 phosphorylation on hyphal morphogenesis in the experimental strain CBS562, we knocked out the SAK1 (Figure S5a). Upon detection, the SAK1 knockout strain maintained extremely low levels of phosphorylation (Figure S5b). In Spider solid medium and RPMI‐1640 (containing 10% FBS) liquid medium, SNF1 phosphorylation‐deficient organisms were unable to form hyphae for osmotic extension and three‐dimensional spatial structural organisation. At this point, the biofilm consisted only of a simple tandem stack by yeast‐type cells (Figure 5a). Consistent with the phenotype, the expression of hyphae‐related genes also decreased significantly after SAK1 knockout (Figure 5b), suggesting that the strain was unable to progress to the hyphal form when the intracellular SNF1 was in a state of low‐level phosphorylation, and thus remained stable in the yeast form. In the quantitative results of crystal violet staining, although SAK1 knockout did not significantly affect the biomass of mature biofilms, the biofilm inhibitory effect of HAQ was subsequently decreased (Figure 5c). At the same time, HAQ was not able to morphologically adjust substantially to the blocked yeast or pseudohyphae morphology (Figure 5d), suggesting that HAQ may restrict hyphae formation with the help of SNF1 signalling and thus exert a biofilm inhibitory effect.

FIGURE 5.

HAQ alleviates filamentation through Adh1‐SNF1 signalling. (a) Hyphal extension and penetration of wild‐type and SAK1 knockout strains on the surface of Spider medium, as well as the microstructure of hyphae from both strains on the surface of polystyrene substrate in RPMI‐1640 medium (containing 10% FBS). (b) qRT‐PCR to detect the expression of hyphae‐specific genes before and after SAK1 knockout under biofilm‐inducing conditions. (c) Crystal violet staining to quantify the total amount of mature biofilm formed by the wild‐type and SAK1 knockout strains under the treatment of incremental concentrations of HAQ. (d) Using ordinary optical microscope and scanning electron microscope to observe the morphology of hyphae within the mature biofilm of wild‐type and SAK1 knockout strains on the surface of polystyrene substrates after 24 h of co‐incubation with 20 μg/mL HAQ. (e) Quantification the effect of ADH1 knockout or overexpression on the biomass of mature biofilms formed by the strains in the presence of Dorsomorphin using wild‐type strains as a control. (f) Ordinary optical microscope or scanning electron microscope observation of the morphology of hyphae in the mature biofilm formed by three strains in the 0 and 8 μg/mL Dorsomorphin administration groups in Fig. (e, g) Crystal violet staining to quantify the total amount of mature biofilm formed by the wild‐type strains under the co‐administration of 20 μg/mL HAQ and 8 or 16 μg/mL Dorsomorphin. (h) Ordinary light microscopy or scanning electron microscopy observation of the morphology of hyphae within mature biofilms at the indicated concentrations of HAQ co‐administered with Dorsomorphin.

Next, to further explore the relationship between SNF1 signalling and HAQ or Adh1 in phenotypic regulation, we up‐regulated the phosphorylation level of SNF1 using activators to further confirm that HAQ or Adh1 depends on SNF1 to exert biological effects. The results showed that Dorsomorphin, which promotes SNF1 phosphorylation in a dose‐dependent manner (Figure 5e, Figure S5c), significantly down‐regulated the biomass of C. albicans mature biofilm and exhibited enhanced efficacy in the ADH1 knockout strain (Figure 5f). Considering that ADH1 knockout would stimulate SNF1 activation under biofilm‐inducing conditions in the pre‐experiment, it is speculated that the superposition of high levels in SNF1 phosphorylation may limit the morphological plasticity of the strain during the biofilm construction cycle. These findings indicate that elevated SNF1 phosphorylation induces excessive hyphal elongation in C. albicans while impairing its capacity to form branched hyphae and undergo morphological transitions—abilities that are essential for biofilm assembly. Consequently, although SNF1 phosphorylation promotes hyphal extension, it simultaneously suppresses biofilm formation. In addition, Dorsomorphin promotes the development of wild‐type and ADH1 overexpressing C. albicans into true hyphae, which are morphologically convergent. It is therefore suggested that high levels of SNF1 phosphorylation mediated by ADH1 knockout or activator stimulation will promote the development of strains toward true hyphae compared to pseudohyphae within the biofilm of wild‐type strains. When the level of SNF1 phosphorylation was maintained at a constant state, Adh1 was unable to make some adjustments to hyphae morphology on this basis, suggesting that Adh1 exerts phenotypic regulatory effects through adjusting SNF1 phosphorylation (Figure 5g). Since the biofilm biomass of the ADH1 knockout strain was too low in high concentrations of Dorsomorphin, it was infeasible to characterise its hyphae morphology. Finally, we co‐treated the strain with Dorsomorphin and HAQ under biofilm‐inducing conditions. The results showed that the two drugs synergistically inhibited biofilm formation (Figure 5h). In this case, the fungus in the co‐administered group showed a true hyphae morphology. That is, HAQ was unable to exert significant inhibition of hyphae formation in the presence of Dorsomorphin in the system (Figure 5i). The above results suggest that HAQ blocked C. albicans in the yeast morphology by promoting the dephosphorylation of SNF1 and thus down‐regulating the level of SNF1 phosphorylation. In summary, HAQ alleviates C. albicans filamentation through Adh1‐SNF1 signalling.

3. Discussion and Conclusion

Phenotypic plasticity is the most typical pathogenic feature of C. albicans (Lu et al. 2014), and biofilms consisting of cells with different morphologies and extracellular matrix are the major challenge in clinically resistant candidiasis. Here, we noted the multifunctional glycolytic enzyme Adh1 in C. albicans associated with pathogenicity (Wang et al. 2024a, 2024b). Existing studies have shown that this enzyme regulates virulence processes such as adhesion, hyphal development and biofilm formation through alcohol‐aldehyde metabolism (Mukherjee et al. 2006; Chauhan et al. 2011), but there is still no definitive conclusion on its specific effects or mechanisms. Thus, we knocked out or overexpressed the ADH1 gene of the experimental strain CBS562. Differing from literature reports in which ADH1 knockout disrupted hyphal development in the clinical standard strain SC5314 (Song et al. 2019), CBS562 instead formed over‐extended hyphal as a result. We suggest that the two strains have ontogenetic differences in hyphal developmental events. It was observed that the hyphae within the biofilm formed by CBS562 existed mainly in the form of pseudohyphae, with a large number of yeast‐type cells scattered around it, suggesting that CBS562 possesses a stronger polymorphism. On this basis, knockout of ADH1 mediated further extension of pseudohyphae into over‐elongation true hyphae, whereas overexpression of ADH1 facilitated the conversion of the strain to the yeast state. The flexibility shown by CBS562 provides an opportunity to understand the function of Adh1 in polymorphic regulation.

Target‐based drug discovery is a common strategy for new drug development (Paananen and Fortino 2020). In this study, we found anthraquinones based on Adh1 to exert biofilm inhibitory effects in C. albicans through biofilm models of wild‐type and ADH1 knockout strains. While reports of active anthraquinones in previous studies were limited to biofilm phenotypes or specific gene expression, the present study attempted to explore the effect mechanism of the compounds in terms of their targets. We were concerned that among the four anthraquinones with Adh1 as their functional target, only HAQ was able to exert the most significant biofilm inhibitory activity without affecting the growth of C. albicans . Considering that the knockout or overexpression of ADH1 also did not affect the growth kinetics of the strain, HAQ was considered to be the most plausible regulator of Adh1 in the direction of hyphal morphological adjustment. Subsequently, we explored the target proteins of HAQ in reverse. The much less active emodin was immobilised as a weakly binding drug in epoxy‐activated agarose microspheres, and the proteins were then competitively eluted by HAQ. The mass spectrometry identification resulted in a high score for the target protein Adh1. Following experiments such as differential scanning fluorimetry and microscale thermophoresis also confirmed the direct interaction between HAQ and Adh1.

Next, we examined Adh1‐interacting proteins affected by HAQ by coupling affinity purification with mass spectrometry detection to explore the downstream signals of Adh1 regulating C. albicans biofilm hyphae formation. In previous studies, anthraquinones with C. albicans biofilm or hyphae inhibitory function often exhibit activities such as regulating the expression of genes related to glycolysis or the tricarboxylic acid cycle and inhibiting metabolism (Huang et al. 2019; Nain‐Perez et al. 2022). Similarly, deletion of ADH1 regulates mitochondrial metabolism and ATP production within fungi while inhibiting biofilms (Guo et al. 2013; Song et al. 2019). Therefore, we focused on SNF1 and its multiple dephosphorylation regulators associated with energy metabolism in our mass spectrometry data. It has been shown that SNF1 is involved in the regulation of biofilm formation in S. cerevisiae . Upon stimulation with excess glucose, inactivation of SNF1 will mediate the translocation of the DNA‐binding protein Mig1 into the nucleus, preventing the expression of alternative energy utilisation genes and negatively regulating the intracellular level of the flocculation‐associated protein Flo11 by binding to Tup1, Ssn6, etc. and ultimately antagonising biofilm formation (Wang et al. 2024a). In addition, the negative regulators of hyphal growth, Nrg1 and Nrg2, can also act as direct or indirect targets of SNF1 kinases, thus participating in hyphal differentiation and invasive growth in S. cerevisiae (Kuchin et al. 2002). It is implied that SNF1 is a potential effector molecule of Adh1 in regulating C. albicans biofilm formation. Due to the lack of commercially available antibodies specifically targeting C. albicans proteins, we resorted to the expression of proteins with His or Flag tags in S. cerevisiae to review the mass spectrometry information by Co‐IP, and here we highlight the down‐regulation of the interactions between Adh1 and Ssb1 or Bmh1 by HAQ. It is noteworthy that our clues are not sufficient to interpret the mechanism of HAQ as well as Adh1 in the regulation of SNF1 phosphorylation. Regarding whether Adh1 regulates SNF1 activity with the help of Ssb1 and Bmh1, and whether HAQ affects SNF1 activation by regulating the interaction between Adh1 and Ssb1 or Bmh1, further corroboration is required by intervening in the interaction between Adh1 and regulatory proteins. In the SNF1 phosphorylation assay results, Adh1 alleviated the activation of C. albicans SNF1. When ADH1 was knocked out, SNF1 maintained high levels of phosphorylation under biofilm‐induced conditions (Figure 4c). Furthermore, HAQ not only down‐regulated the interaction between Adh1 and SNF1 dephosphorylation regulators but also depended on Adh1 to function as an inhibitor of SNF1 phosphorylation. This result likewise provides more sufficient evidence for the involvement of Adh1 in the regulation of SNF1 phosphorylation.

We observed two immunoblots in the phosphorylation assay results of C. albicans SNF1. Sak1 has been documented as the major activator of SNF1 in C. albicans (Ramírez‐Zavala et al. 2017). The Lambda protein phosphatase treatment experiment demonstrated that both observed bands are specific phosphorylated forms of SNF1 (Figure S5d). Knockout of SAK1 attenuated the lower band to 15% of the wild‐type, and thus the band was used to characterise SNF1 phosphorylation (Figure S5b). We unexpectedly found that the commonly used AMPK inhibitor Dorsomorphin (Wang et al. 2022; Zheng et al. 2022) promoted phosphorylation of SNF1 in the experimental strain in a dose‐dependent manner. Considering that Dorsomorphin promotes the development of true hyphae in wild‐type and ADH1 overexpressing strains, constraining the modulation of hyphae morphology by HAQ and Adh1, whereas SAK1 knockout strains showed only a yeast‐like state, and that high levels of SNF1 phosphorylation caused by ADH1 knockout similarly mediated the formation of true hyphae in the strains, we conclude that the morphology of the organism within the biofilm of C. albicans is directly controlled by the degree of SNF1 phosphorylation, and that HAQ and Adh1 regulate the formation of hyphae with the help of SNF1 phosphorylation. The immunoblotting results of Dorsomorphin‐promoted SNF1 phosphorylation in C. albicans are hereby endorsed, and the discrepancy with literature reports is attributed to differences in metabolic regulation between species. Since Dorsomorphin inhibits the phosphorylation of the upper band while promoting that of the lower band in the SNF1 phosphorylation Western blot assay, we speculate that Dorsomorphin may preferentially inhibit the kinases responsible for the phosphorylation at the upper band (other sites). This inhibition could potentially lead to a relative increase in the availability of phosphate groups or kinase activity toward the Thr208 site, manifesting as an apparent overall promotion of the functional (lower band) phosphorylation in our assay. Furthermore, the experimental results demonstrate that HAQ, which promotes SNF1 dephosphorylation, inhibits hyphal elongation, whereas dorsomorphin, which enhances SNF1 phosphorylation, induces excessive hyphal extension. These results are consistent with the hypothesis that SNF1 phosphorylation promotes hyphal elongation. However, HAQ and dorsomorphin exhibit synergistic inhibitory effects on biofilm formation. This phenomenon may be attributed to the requirement of complete hyphal polymorphism for proper biofilm development. Both the suppression of hyphal growth and the impaired dispersal capacity associated with over‐elongation would compromise biofilm formation.

In conclusion, the present study reveals that under biofilm‐induced conditions, HAQ significantly impairs the interactions between Adh1 and SNF1 signalling modules by specifically targeting Adh1 protein, and then promotes the dephosphorylation of SNF1 kinase in an Adh1‐dependent manner, which ultimately prevents the pathogenic hyphal transition of C. albicans (Figure 5). At the mechanistic level, Adh1, as a multifunctional factor with the characteristics of ‘moonlighting protein’, not only participates in the reprogramming of carbon metabolism, but also mediates the development of fungal virulence by dynamically regulating the balance of SNF1 phosphorylation (Figure 6). It is noteworthy that the precise intervention of HAQ on Adh1‐SNF1 interactions can synchronously affect the key pathogenic features of C. albicans , such as host adhesion, biofilm formation and the evolution of drug resistance. The research results not only provide new ideas for the development of novel drugs for treating Candida infections, but also provide a rational reference for the optimization of drug formulations in the context of extensive drug resistance.

FIGURE 6.

Proposed mechanism diagram of HAQ targeting ADH1‐SNF1 phosphorylation to regulate hyphal growth. The purple arrow represents that SNF1 regulates the hyphae and biofilm pathway after phosphorylation under normal conditions. The blue arrow represents that HAQ binds to ADH1 to release dephosphorylation factors Bmh1 and Ssb1, thereby reducing the phosphorylation level of SNF‐1 and inhibiting fungal growth.

4. Methods

4.1. Experimental Strains and Plasmids

The C. albicans CBS562 ADH1 knockout strain was constructed by the SAT1‐flipper strategy (Sasse and Morschhäuser 2012) using the pRB895 plasmid. The ADH1 overexpression strain was constructed using the pENO1‐iRFP‐NATr plasmid. All of the above plasmids were purchased from Hangzhou Baosai Biotechnology Co.

The ADH1 gene was inserted into plasmid pXW06 by homologous recombination, and the BMH1 or SSB1 genes were inserted into plasmid pXK30, respectively. The above plasmids were transformed into S. cerevisiae BJ5464‐NpgA (MATα ura3‐52 his3‐Δ200 leu2‐Δ1 trp1 pep4::HIS3 prb1 Δ1.6R can1 GAL) for fermentation and expression of His‐Adh1, Flag‐Bmh1, or Flag‐Ssb1.

All strains were conserved in the Department of Marine Pharmacology, China Pharmaceutical University. C. albicans was maintained on the surface of YPD solid medium (2% peptone, 2% glucose, 1% yeast extract, 2% agar) and grown at 30°C.

4.2. Sources of Compounds

All 114 natural products were purchased from Chengdu Lemeitian Pharmaceutical Technology Co. Emodin and Anisomycin were purchased from Shanghai Aladdin Biochemical Technology Co. All the above compounds were dissolved in DMSO and the amount of DMSO added did not exceed 1.25% (v/v) in all experiments.

4.3. Biofilm Formation Assay

Single colonies were inoculated in YPD liquid medium and incubated at 30°C, 200 rpm until late logarithmic growth. The seed solution was aspirated and diluted in RPMI‐1640 medium (KeyGEN, KGL1501‐500) supplemented with 10% Normal fetal bovine serum (Sangon, E510002) to 1 × 10⁶ CFU/mL. 200 μL per well was added to 96‐well plates (Labselect, 11510), and incubated statically at 37°C for 24 h. Unless otherwise stated, the RPMI‐1640 medium used for biofilm production in this study contained 10% fetal bovine serum.

After biofilm formation, the absorbance at 595 nm for each well of the 96‐well plate was directly detected as ‘Biomass’ using an enzyme‐linked immunosorbent assay detector AMR‐100 T (Allsheng). The wells were then washed with ddH₂O to remove the planktonic organisms and stained with 0.1% crystal violet solution (Yuanye, B26890) for 20 min. The wells were washed again with ddH₂O to remove the excess staining solution, and then anhydrous ethanol (Greagent, G73537BK) was added to dissolve the crystal violet. The absorbance of each well at 595 nm was detected to characterise the total amount of biofilm. Three parallel wells were set up for each group.

4.4. Observation of Biofilm Microstructure

For hyphal samples that are observed with an ordinary optical microscope, scrape the stained mature biofilm from the bottom of a 96‐well plate. Mix well and apply 20 μL of suspension to the centre of the slide, cover with a coverslip, and observe microscopically. Three replicate wells were set up and observed for each group.

For biofilm samples observed by scanning electron microscope, polystyrene slices were laid in 6‐well plates (Labselect, 11112), and fungal solution diluted in RPMI‐1640 medium was added to the wells to construct biofilms. After washing with ddH₂O to remove planktonic cells, 2.5% glutaraldehyde solution (Nanjing Reagent, C0670530735) was added, and the plates were fixed overnight at 4°C, protected from light. The biofilm was washed again with ddH₂O and dehydrated with 30%, 50%, 70%, 80%, 90% and 95% ethanol in a gradient; each gradient was processed for 10 min. The slices were taken out and dried naturally, and then sent to the Scientific Compass Experimental Centre for observation and photographing.

4.5. Evaluation of Hyphal Invasion

Fungal suspensions at concentrations of 1–5 × 10⁶ CFU/mL were prepared, and 0.5 μL was pipetted and added dropwise to the surface of Spider medium (1% nutrient broth, 1% mannitol, 0.2% K₂HPO₄ and 2% agar), which was allowed to stand at 37°C to induce hyphal growth. On the fifth day of incubation, 1/2 views of the colonies were observed and photographed through an inverted phase contrast microscope. On the seventh day of incubation, the agar was cut along the centre of the colony using a scalpel blade to obtain agar slices with a thickness of 0.5–1.0 mm. A stereomicroscope was used to observe and photograph at 20× magnification, and the width and depth of hyphal invasion were quantified using ImageView.

4.6. Quantitative Analysis by qRT‐PCR

The organisms were fragmented by cryomilling with liquid nitrogen. Total RNA from C. albicans was extracted using the FreeZol Reagent kit (Vazyme, R711). Subsequently, RNA samples were reverse transcribed using the HiScript III RT SuperMix for qPCR kit (Vazyme, R323) to generate templated cDNA. Primers specific to the target gene were synthesized from Sangon and mixed with SYBR Green (Vazyme, Q112) for qRT‐PCR. Samples were analysed in triplicate using a QuantStudio 3 rapid real‐time fluorescence quantitative PCR instrument (Thermo Scientific) and the results obtained were normalised according to 18S rRNA. The expression levels of the target genes in the test group samples compared to the control group were calculated by $2^{-\mathrm{\Delta \Delta }{C}_{\mathrm{t}}}$. The primers used for qRT‐PCR are shown in Table S3.

4.7. Protein Extraction and Western Blot Assay

Centrifugally harvested CBS562 was resuspended in RPMI‐1640 medium, treated with drug and incubated at 37°C, 200 rpm for the indicated times. The organisms were lysed by sonication in Triton X‐100 cell lysate (Yuanye, R21239) containing 1% PMSF, and the supernatant was obtained as protein samples after centrifugation (12,000 rpm, 15 min, 4°C).

The concentration of protein was quantified with a BCA kit (Yeasen, 20201ES86). After adding Loading Buffer to the solution, the samples were subjected to a boiling water bath for 5 min. 10 μg of proteins were separated by 10% SDS‐PAGE (Yamei, PG112) followed by transferring the target proteins to a PVDF membrane (Millipore, IPVH00010). The membranes were submerged in 5% skimmed milk and incubated at room temperature for 3 h. After washing, the membranes were incubated with antibodies specific for the indicated proteins at 4°C overnight. Then, the washed membranes were incubated with secondary antibodies at room temperature for 45 min. Protein blots were detected by enhanced chemiluminescence kit (Vazyme, E422‐01) and analysed by Tanon ImageJ. Antibodies used in the experiments: goat anti‐rabbit IgG (Zenbio, 511203), goat anti‐mouse IgG (Yeasen, 20201ES86), Adh1 antibody (Rockland, 30854), β‐actin antibody (Zenbio, 380624), anti‐His antibody (Proteintech, 66005‐1‐Ig), anti‐Flag antibody (Proteintech, 20543‐1‐AP).

4.8. Protein Immunoprecipitation

2% phosphatase inhibitor (Yuanye, R20208) and 1% N‐Ethylmaleimide (Sigma, MO‐L011) were added to Triton X‐100 cell lysate (containing 1% PMSF). The organisms were resuspended in the above solutions and lysed by sonication. After centrifugation (12,000 rpm, 15 min, 4°C), the supernatant was extracted, and the protein concentration was quantified by BCA assay. A portion of the protein solution was mixed with Loading Buffer and incubated in a boiling water bath for 5 min. Quantitative lysates were taken from the remaining supernatant solution, added with anti‐label antibody or Adh1 antibody and incubated overnight at 4°C with rotation. Subsequently, activated Protein A/G immunoprecipitation magnetic beads (Selleck, B23202) were added for pull‐down of the antibody bound with the target protein. The magnetic beads were washed with 1× TBS‐T solution, and then the Loading Buffer was added, boiling water bath for 5 min. Protein samples were separated by SDS‐PAGE and then stained with Coomassie Brilliant Blue, and the gel containing the target proteins was sent to APTBIO for LC–MS detection, or the binding level of interacting proteins was detected by Western blot.

4.9. SNF1 Phosphorylation Assay

The fungal suspension was heat‐treated in a boiling water bath for 3 min, followed by equilibration to room temperature. Fungal cells were collected by centrifugation (4000 rpm, 5 min) and resuspended in 1× TE buffer (10 mM Tris–HCl, 1 mM EDTA). An equal volume of 0.2 M NaOH was added to the suspension, followed by 5 min incubation at room temperature. After centrifugation (12,000 rpm, 1 min), the pellet was resuspended in 80 μL of 1× Loading Buffer and boiled for 5 min. The supernatant obtained after final centrifugation (12,000 rpm, 5 min) was collected as the Western blot sample. For the samples treated with lambda protein phosphatase, the lysate was incubated with lambda protein phosphatase (Beyotime, P2316S) at 30°C for 1 h prior to the addition of Loading Buffer and boiling. Protein expression levels of SNF1 and its phosphorylation status were analysed by Western blot using the following primary antibodies: anti‐AMPK α1 (Aifang, AFRM9099) and anti‐phospho‐SNF1 (Cell Signalling Technology, 2535).

4.10. Heterologous Expression of Recombinant Adh1

The target gene was amplified using the ADH1 gene sequence from the clinical standard strain SC5314 as a template and subsequently cloned into the pET‐28a(+) to generate the recombinant plasmid pET‐ADH1. The construct was transformed into Escherichia coli BL21 (DE3) and verified by sequencing. Single colonies were inoculated in LB medium supplemented with 50 μg/mL kanamycin and subcultured until the OD₆₀₀ to 0.6. Protein expression was induced with 0.1 mM IPTG (Yuanye, B23922) for 20 h at 16°C. Cells were harvested by centrifugation (5000 g, 15 min, 4°C), resuspended in tris buffer (pH 7.5) and lysed. The soluble fraction was obtained by centrifugation (12,000 rpm, 15 min, 4°C) and subjected to Ni‐NTA affinity chromatography (Qiagen, 30230). The target protein was eluted using an imidazole gradient and concentrated using a 30 kDa ultrafiltration tube (Millipore, UFC9030) via centrifugation (5000 g, 4°C). Protein expression and purity were assessed by SDS‐PAGE followed by Coomassie Brilliant Blue staining.

4.11. Microscale Thermophoresis Assay

The recombinant Adh1 protein was fluorescently labelled using the Monolith TM RED‐NHS Second‐Generation Protein Labeling Kit (NanoTemper, MO‐L011). Binding experiments were conducted in 1× PBS‐T buffer (pH 7.5, 0.05% Tween‐20).

A serial dilution of the ligand (HAQ) was prepared and mixed with a constant concentration of fluorescently labelled Adh1. The mixture was loaded into glass capillaries. Prior to measurement, we confirmed the absence of capillary wall adhesion and sample aggregation. Microscale thermophoresis measurements were performed under the following conditions: Monolith NT.115 Standard Treated Capillary (K2002); 20% Excitation power; Medium MST power. Binding affinity was analysed using MO. Affinity Analysis software (x86 version).

4.12. Tryptophan Fluorescence Spectroscopy

A serial dilution of HAQ and Adh1 recombinant protein solution at a concentration of 1 mg/mL was prepared using 1× PBS solution. Various concentrations of the drug solution and protein solution were mixed in equal volumes in the wells of a 96‐well plate, centrifuged (200 g, 1 min) to allow the reaction solution to aggregate at the bottom of the wells, and the tryptophan emission spectra were detected by a multifunctional microplate reader Infinite M200 Pro (Tecan). The experimental conditions were set as fluorescence channel with 280 nm as excitation wavelength and 2 nm as emission interval, and the emission wavelength was set as Start: 305 nm; Stop: 500 nm; Step: 2. The results were analysed by Orgin 2024.

4.13. Differential Fluorescence Scan

Compound and SYPRO Orange dye were diluted using Glycine buffer to obtain a 50× dye solution, and compound solutions at concentrations of 0.5 and 1.0 mM, respectively. The compound dilutions and the recombinant protein solution were added to the qPCR 96‐well reaction plate together in a total volume of 25 μL per well. Centrifugation (200 g, 1 min) was performed to accumulate the reaction solution at the bottom of the wells, and the melting temperature of the proteins was detected by a real‐time quantitative PCR detecting system. The experimental programme was set as ‘melt curve’, 25°C (2 min) to 99°C (2 min), increment 1°C, and the ROX fluorescence channel was selected. The results were analysed by Orgin 2024, and the protein defolding temperature (T _m value) was calculated as the highest first‐order derivative fluorescence value.

4.14. Epoxy‐Activated Affinity Chromatography

The Emodin coupling solution was mixed with 8 mL of Epoxy‐Activated Beads 4FF (Smart‐Lifesciences, SA040025) microspheres and incubated at 25°C in the dark for 24 h. Subsequently, it was transferred to a syringe‐type affinity chromatography vacutainer column (ϕ13 mm × 66 mm) and washed sequentially with coupling solution (0.1 M Na₂CO₃, pH 8.5–10.0) and blocking solution (1 M ethanolamine, pH 8.0). The packing was transferred to a conical flask and incubated with shaking at 37°C for 4 h. The system was again transferred to an empty column and washed twice sequentially using 15 mL of ddH₂O, wash solution 1 (0.1 M acetic acid–sodium acetate, 0.5 M NaCl, pH 3.0), ddH₂O, wash solution 2 (0.1 M Tris–HCl, 0.5 M NaCl, pH 8.0), ddH2O. The original protein solution was passed through a emodin‐agarose column and incubated overnight at 4°C after three repetitions. The protein solution was flowed out and non‐specific binding proteins were removed using 50 mL of binding buffer (20 mM Tris–HCl, 0.25 M NaCl, 5 mM MgCl₂), and finally a specific elution was performed using 50 mL of HAQ saturated solution. The eluent was diluted five times by adding ddH₂O, and the samples were concentrated by ultrafiltration using an ultrafiltration tube with a pore size of 10 kDa. The protein samples were separated by SDS‐PAGE and stained with Coomassie Brilliant Blue staining, and the lanes containing the target proteins were cut off and sent to Biotec Biotechnology in Beijing, China for LC–MS analysis.

4.15. Molecular Docking

Binding simulations of HAQ to Adh1 were performed using Autodock vina 1.1.2. The 3D structure of C. albicans Adh1 was obtained from Uniprot with reference to the clinical standard strain SC5314. The 2D structure of HAQ was drawn by ChemBioDraw Ultra 14.0 and optimised to obtain the 3D structure by the MM2 method using ChemBio3D Ultra 14.0 software. Docking input files were generated using AutoDockTools 1.5.6 software package. During molecular docking, water molecules were removed, and polar hydrogen atoms and Coleman's joint atom‐type charges were added. The search grid for Adh1 was set to center_x: 7.309, center_y: 0.998, center_z: −7.556, size_x: 74, size_y: 73 and size_z: 99. The value of exhaustiveness is set to 10 and other parameters are default parameters. The best score docking obtained by Vina docking was selected and visually analysed using PyMoL 1.7.6 software.

4.16. Growth Curve

To evaluate the impact of the Adh1 protein on C. albicans growth, a fungal seed suspension was diluted approximately 100‐fold in YPD medium to an initial concentration of 10⁵ CFU/mL. The diluted cultures were incubated at 37°C with continuous shaking at 200 rpm for 26 h. The growth kinetics of the WT, ADH1‐KO and ADH1‐OE strains were monitored by measuring the optical density at 595 nm (OD₅₉₅) at 1‐h intervals. The data are presented as the mean values of at least three independent biological replicates (n ≥ 3).

4.17. Statistical Analysis

Graphs and statistical analyses were generated using GraphPad Prism 8.0. All biological experiments were performed at least three times. Data were analysed for significance using the two‐tailed unpaired Student's t‐test, with p < 0.05 indicating a statistically significant difference in the data. The band densities in Western blot results were measured using the ‘Lane Analysis ‐ Semi‐Quantitative Analysis’ function of the Tanon Image software. The values were normalised to the corresponding internal reference bands. The band intensity of the control group was set to 1.00, and further normalisation was performed accordingly.

Author Contributions

Ziqi Wang: investigation, methodology, writing – original draft, visualisation. Ziran Wang: investigation, methodology. Qi Zhang: investigation, methodology. Yuanyuan Song: investigation, methodology. Haoying Zhang: investigation, methodology. Qin Xu: investigation, methodology. Jianmin Liao: supervision, writing – review and editing. Yuanyuan Lu: conceptualization, supervision, writing – review and editing, funding acquisition.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: HAQ inhibits mycelial development within biofilms via Adh1.

Figure S2: Adh1 is a direct target of HAQ.

Figure S3: F224, A254, Q257 are key residues for HAQ‐Adh1 binding.

Figure S4: Adh1 is a key intermediate in the regulation of SNF1 activity by HAQ.

Figure S5: HAQ alleviates mycelialisation through Adh1‐SNF1 signalling.

Table S1: Biofilm inhibition efficiency of natural products.

Table S2: Proteins specifically eluted by HAQ.

Table S3: Primers for PCR in strain construction.

Table S4: Primers for qPCR.

Data Availability Statement

The mass spectrometry proteomics data generated in this study have been deposited to the ProteomeXchange Consortium via the iProX partner repository with the dataset identifier PXD073598 (http://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD073598) and PXD073258 (http://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD073258).