The opportunistic fungus Candida albicans causes oral thrush and vaginal candidiasis, as well as candidaemia in immunocompromised patients including those undergoing cancer chemotherapy, organ transplant and those with AIDS. We previously found that the AMPs (antimicrobial peptides) LL37 and hBD-3 (human β-defensin-3) inhibited C. albicans viability and its adhesion to plastic. For the present study, the mechanism by which LL37 and hBD-3 reduced C. albicans adhesion was investigated. After AMP treatment, C. albicans adhesion to plastic was reduced by up to ~60% and was dose-dependent. Our previous study indicated that LL37 might interact with the cell-wall β-1,3-exoglucanase Xog1p, which is involved in cell-wall β-glucan metabolism, and consequently the binding of LL37 or hBD-3 to Xog1p might cause the decrease in adhesion. For the present study, Xog1p(41–438)-6H, an N-terminally truncated, active, recombinant construct of Xog1p and Xog1p fragments were produced and used in pull-down assays and ELISA in vitro, which demonstrated that all constructs interacted with both AMPs. Enzymatic analyses showed that LL37 and hBD-3 enhanced the β-1,3-exoglucanase activity of Xog1p(41–438)-6H approximately 2-fold. Therefore elevated Xog1p activity might compromise cell-wall integrity and decrease C. albicans adhesion. To test this hypothesis, C. albicans was treated with 1.3 μM Xog1p(41–438)-6H and C. albicans adhesion to plastic decreased 47.7%. Taken together, the evidence suggests that Xog1p is one of the LL37/hBD-3 targets, and elevated β-1,3-exoglucanase activity reduces C. albicans adhesion to plastic.
Candida albicans infections occur with an incidence of 1.1–24 cases per 100000 humans [1,2]. Over the past three decades, C. albicans infections have emerged as a significant cause of human morbidity and mortality [3,4]. In the U.S.A., ~3000–11000 individuals die annually from nosocomial candidaemia . Those at high risk of C. albicans infection include cancer patients undergoing immunosuppressive chemotherapy and patients who have undergone major surgery, are on supportive ventilation and/or have inserted central venous and/or urinary catheters [4,6]. Antifungal therapies are of limited effectiveness against systemic infection by C. albicans, as drug resistance and extreme toxicity result in treatment failure and, consequently, a mortality rate of >40% .
The pathogenesis of C. albicans infection requires several steps, i.e. adhesion to the host mucosal surface, cell-surface colonization, cell invasion and tissue disruption . Therefore, if adhesion can be prevented, pathogens cannot colonize mucosal surfaces. C. albicans cell-wall macromolecules are required for adhesion to host mucosal cells. Certain HSPs (heat-shock proteins) function as adhesins on the cell surfaces of pathogens such as Helicobacter pylori, Haemophilus influenzae, Mycobacterium avium and Histoplasma capsulatum [9–12]. Cell-surface HSP70 and HSP100 family proteins, which protect certain pathogens against host-induced stress [13,14], also interact with AMPs (antimicrobial peptides) produced by the host cells [15,16]. Moreover, the agglutinin-like sequence gene family of C. albicans encodes eight glycosylphosphatidylinositol-anchored cell-wall proteins that are adhesion molecules and bind to host cell surfaces [17–20]. Heterologous gene-expression and gene-deletion experiments demonstrated that the different agglutinin-like sequence proteins have different binding affinities towards different host cells, e.g. oral mucosa and buccal epithelial cells [19,21–23]. Antibody-neutralization experiments demonstrated that when the C. albicans surface protein complement receptor 3-related protein was blocked, C. albicans adhesion on plastic and cell surfaces was reduced . Therefore C. albicans cell-wall macromolecules are crucially involved in the first step of infection.
During adhesion of C. albicans to cell surfaces, the cell-wall components of pathogens are remodelled to attach to various surfaces . C. albicans cell-wall β-glucan and chitin, which are associated with mannoproteins, form the main structural microfibrillar polymer and provide the cell wall with structural rigidity . Two metabolic enzymes, β-1,3-exoglucanase and chitinase, are responsible for cell-wall morphogenetic events as they hydrolyse β-glucan and chitin respectively . In C. albicans, the major β-1,3-exoglucanase is Xog1p, a homologue of Saccharomyces cerevisiae Exg1p . Knockout experiments demonstrated that in the null C. albicans strain 5314 xog1Δ/xog1Δ, exoglucanase activity decreased by 60% compared with that in wild-type C. albicans , which indicated that Xog1p functions in β-glucan metabolism. Moreover, an Xog1p-deficient mutant was more susceptible to antifungal agents that inhibit β-1,3-glucan biosynthesis, e.g. papulacandin B and cilofungin, and had a reduced capacity to colonize the brain during systemic infection . Therefore Xog1p may participate in C. albicans adhesion and colonization. The Xog1p-deficient strain was equally viable in minimal or rich medium at 30, 37 and 42°C, and no morphological differences were observed under a SEM (scanning electron microscope) and TEM (transmission electron microscope) .
Mammalian AMPs are secreted mainly by epithelial cells and neutrophils, and AMPs are the first line of defence against infectious micro-organisms [29,30]. AMPs have distinct functions in response to different pathogens. For example, hBD (human β-defensin)-1 and hBD-2 have substantial microbicidal activity against Gram-negative bacteria, but not against Gram-positive bacteria. Conversely, hBD-3 is a broad-spectrum AMP that kills many pathogenic bacteria and opportunistic pathogenic yeast, including C. albicans . Recently, Schroeder et al.  reported that, after reduction of disulfide bridges, hBD-1 becomes a potent AMP against a wide range of pathogens including C. albicans, and anaerobic Gram-positive commensals of Bifidobacterium and Lactobacillus species. The AMP LL37 is also a broad-spectrum antimicrobial that is active against Gram-positive and Gram-negative bacteria and pathogenic fungi . In addition to its antimicrobial activity, LL37 neutralizes the effect of bacterial LPS (lipopolysaccharide) and consequently reduced endotoxic shock in a murine model . We previously showed that LL37 interacts with C. albicans cell-wall carbohydrates and reduces C. albicans adhesion to plastic and mouse bladders . We also found that C. albicans cell-wall Xog1p is an LL37 receptor , which suggested that LL37 may prevent C. albicans–host cell interactions. The present study demonstrates that LL37 and hBD-3 elevate Xog1p activity by interacting with the enzyme and that elevated Xog1p activity is key to reduced C. albicans adherence.
MATERIALS AND METHODS
Candidacidal activity of LL37, hBD-3 and Xog1p(41–438)-6H
A CFU (colony-forming unit) assay was used to assess the antifungal activities of LL37 and hBD-3. C. albicans SC5314 was grown in liquid YPD medium (10 g of yeast extract, 20 g of peptone and 20 g of glucose in 1 litre of water) at 30°C with shaking. After 14 h, the cells were diluted into fresh YPD medium at an initial D600 of 0.1 and cultured for 3–4 h at 30°C until the D600 reached 1.0. Cells were harvested by centrifugation at 2000 g at room temperature (25°C), washed twice using PBS, and suspended in PBS at a concentration of 4000 cells/ml. To determine the lethal doses of the AMPs and Xog1p(41–438)-6H, samples of 400 cells were incubated with 0, 0.1, 0.3, 1, 3 or 10 μM of an AMP at 30°C for 30 min, or with 0, 0.1, 0.3, 1, 3, 10, 30 and 79 μM Xog1p(41–438)-6H at 30°C for 24 h. Then, the sample supernatants that contained the AMPs and Xog1p(41–438)-6H were removed by centrifugation at 2000 g at room temperature, the cells were plated on YPD agar and incubated at 30°C overnight. Finally, colonies were counted, and, for each sample, the relative survival value was calculated as the mean of CFUs for AMP samples/CFUs for the control samples. The assays were performed in triplicate and then the S.E.M. was calculated. To determine the growth-inhibition effects of Xog1p(41–438)-6H, 400 cells were treated with 1.3 μM Xog1p(41–438)-6H at 30°C at the times indicated. Similarly to previous procedures, the CFUs were counted and the relative survival value was calculated.
C. albicans adhesion assay
C. albicans cells were prepared as described in the previous section. Then, the samples were centrifuged at 2000 g for 10 min, and the cells were washed with PBS three times. Cells were then diluted with PBS to a final density of 3000 cells/ml and mixed with LL37 (LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES) or hBD-3 (GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRKCCRRKK) or one of the control peptides, CYC3-3 (GWFWADKPS), TAT (YGRKKKRQRRR) or 3×FLAG (MDYKDHDGDYLDHDIDYLDDDDL) at a 1:1 (v/v) ratio to obtain the desired peptide concentration. The peptides were synthesized chemically by MDBio in Taiwan. The C. albicans/peptide mixtures (250 μl each) were each added into a well of a 24-well plate and incubated at 37°C for 30 min. Wells were then washed three times with PBS. The cells were scraped from each well, each cell sample was plated on the YPD agar and incubated at 30°C for 24 h, and then the colonies were counted. The relative adhesion was normalized to CFUs without peptide treatment. Assays were performed in triplicate.
To determine the effect of Xog1p(41–438)-6H on C. albicans adhesion, samples containing 300 cells were each treated with 0.33, 0.65 and 1.3 μM Xog1p(41–438)-6H at 30°C for 1.5 h. Then, 250 μl of each sample was added into a well of a 24-well plate and incubated at 37°C for 30 min. Wells were washed three times with PBS, and the cells were then scraped from the wells, spread on the YPD agar and incubated at 30°C overnight. The CFUs were counted and the relative adhesion was normalized to the CFUs without Xog1p(41–438)-6H treatment. Assays were performed in triplicate.
To characterize cell-wall morphology, C. albicans cells that had been treated with or without 1.3 μM Xog1p(41–438)-6H at 30°C for 2 h were visualized by scanning electron microscopy. Briefly, C. albicans cell suspension was prepared and pipetted dropwise on to the shiny side of a polycarbonate membrane with a 1 μm pore size (Nucleopore), allowed to settle for 5 min without drying and then immersed in 2% (w/v) aqueous OsO4 (osmium tetroxide) for 12 h at 4°C in the dark. The fixed material was washed in distilled water for 15 min to remove excess OsO4, and dehydrated in a 10% graded ethanol series, 15-min in each step from 10 to 90% ethanol. The membrane was washed in 95% ethanol followed by rising three times in absolute ethanol for 15 min each. The membrane was then immerse in ethanol/acetone (2:1), ethanol/acetone (1:2) and finally immerse in absolute acetone for three times (15 min each). The dehydrated sample was further dried in a Hitachi HCP-2 Critical Point Dryer and coated with platinum (10 nm thickness) in a Hitachi E-1045 ion sputter. The morphology of C. albicans was examined using a Hitachi S-4700 SEM at 3.0 kV.
Expression, purification and refolding of recombinant Xog1p-6H truncated fragments
XOG1(41–438) DNA fragments (GenBank® accession number XM_716123) were PCR amplified using C. albicans genomic DNA as the template and the primers 5′-ATATCATATGGGACATAATGTTGCTTGG-3′ and 5′-ATATCTCGAGGTGAAAGCCACATTGGTTTG-3′ (NdeI and XhoI sites are singly and doubly underlined respectively). By using the forward primer, the first 120 nt of XOG1, which encoded a highly hydrophobic N-terminus, were deleted so that the gene for N-terminus truncated Xog1p was synthesized. Four fragments of XOG1 were amplified using the primers 5′-ATATCATATGGGACATAATGTTGCTTGG-3′ and 5′-ATATCTCGAGTTGACCTTGGACGTATGGAT-3′ for XOG1(41–150), 5′-GCATCATATGGTTCAGTATTTGGAAAAGGC-3′ and ATATCTCGAGTTGGAAAGCATCGTGAATGA for XOG1(151–268), 5′-GCGCCATATGGTCTTTGGCTATTGGAATAA-3′ and 5′-ATATCTCGAGCTCATAACGTGCTCCTCTGT-3′ for XOG1(269–352), and 5′-GCATCATATGGGTGCTTACGATAATGCTCC-3′ and 5′-ATATCTCGAGGTGAAAGCCACATTGGTTTG-3′ for XOG1(352–438) (NdeI and XhoI sites are singly and doubly underlined respectively). XOG1(41–438), XOG1(41–150), XOG1(151–268), XOG1(269–352) and XOG1(352–438) were isolated by digestion with NdeI and XhoI, ligated into pGEM-T Easy vectors and sequenced. The genes were then each cloned into a pET23a(+) vector to generate the plasmids pET23-XOG1(41–438), pET23-XOG1(41–150), pET23-XOG1(151–268), pET23-XOG1(269–352) and pET23-XOG1(353–438). All of the constructs contain a C-terminal hexahistidine sequence (6H) derived from the pET23a(+) vectors.
For protein expression, the plasmids were each transformed into Escherichia coli BL21(DE3)pLysS cells and the transformants were plated on to LB (Luria–Bertani) agar plates. Single colonies, each of which contained one of the plasmids, were individually added into 15 ml of LB broth containing 100 μg/ml carbenicillin and 50 μg/ml chloramphenicol at 37°C, and the cultures were shaken at 200 rev./min overnight. The cultures were then subcultured in 500 ml of fresh LB broth that contained the same antibiotics at 37°C until the D600 of each culture was between 0.5 and 0.8. Protein expression was induced by addition of 0.5 mM isopropyl β-D-thiogalactoside at 37°C for 5 h. Cell pellets were harvested by centrifugation, suspended in 15 ml PBS, sonicated, and centrifuged at 10000 g at 4°C for 10 min. Inclusion bodies in the insoluble fractions were dissolved in 10 ml of 20 mM Tris/HCl (pH 7.9) containing 6 M urea and 0.5 M NaCl (binding buffer) and incubated at 4°C overnight. After centrifugation at 10000 g and 4°C for 30 min, the supernatants were each chromatographed through HisLink resin (Promega), and unbound proteins were removed first by elution with binding buffer and then by elution with 10 mM imidazole in binding buffer. 6H tagged proteins were then eluted in a 50–300 mM imidazole gradient in binding buffer. The purities of the recombinant proteins were assessed by SDS/PAGE (12% gel) that was subsequently stained with Coomassie Blue.
As preparation for experimentation, the purified proteins were each incubated with 80 mM glutathione (reduced form) at 25°C for 30 min for reducing the disulfide bonds, rapidly diluted 100-fold in 0.1 M Tris/HCl (pH 7.5) containing 10% glycerol, 1 mM EDTA, 0.5 M L-arginine, 1 mM PMSF, 40 μM benzamidine, 40 μg/ml aprotinin, 20 μg/ml leupeptin, 20 μM AEBSF [4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride] and 1:4 reduced/oxidized glutathione at 4°C, and then slowly stirred for 24 h. Then proteins were individually concentrated through Vivaflow 200 (Sartorius) and Centricon (molecular-mass cut-off 10 kDa, Millipore) modules at 4°C, and protein concentrations were determined using the BCA (bicinchoninic acid) assay (Thermo Scientific).
The relative binding affinities of LL37, hBD-3, CYC3-3, TAT and 3×FLAG for Xog1p(41–438)-6H were measured by ELISA. Each peptide (10 μg) was dissolved in 50 μl of 50 mM sodium carbonate (pH 9.6), added into a well of a 96-well ELISA plate (GeneDireX), and incubated at 4°C overnight. The wells were then blocked with 0.5% BSA (Gibco) in 100 μl of PBS at room temperature for 1 h. Each well was then washed three times with PBS. Xog1p(41–438)-6H (50 μl, 200 μg/ml) was added to each well, and the mixtures were incubated at room temperature for 2 h. After washing with PBS, 100 μl of histidine-tagged antibody (Santa Cruz Biotechnology, diluted 1:1000 in 0.5% BSA/PBS) was added into each well, and the samples were incubated at room temperature for 4 h. Subsequently, each well was washed three times with PBS, and the samples were incubated with 100 μl of HRP (horseradish peroxidase)-conjugated goat anti-mouse antibody (Jackson ImmunoResearch Laboratories; diluted 1:5000 in 0.5% BSA/PBS) at room temperature for 2 h. After a final wash with PBS, 100 μl of 1,1,3,3-tetramethoxypropane (KPL) was added into each well, and the samples were incubated for 2 min in the dark at room temperature. Reactions were stopped by addition of 100 μl of 1 M HCl (Sigma), and the absorbance of each sample at 450 nm was immediately measured using an ELISA microtitre plate reader (Thermo Scientific). The assays were performed in triplicate. Each value is reported as the mean±S.E.M. (Prism 5.0, GraphPad Software).
In vitro pull-down binding assay
Streptavidin beads (20 μl, GE Healthcare) in PBS were mixed with 10 μg of N-terminal BA (biotinylated)-LL37 or BA-hBD-3. Xog1p(41–438)-6H, Xog1p(41–150)-6H, Xog1p(151–268)-6H and Xog1p(269–352)-6H (10 μg each) were individually added into a PBS solution containing BA-LL37 or BA-hBD-3 (final volume, 200 μl), and the solutions were incubated at 4°C for 3 h. Each solution was centrifuged at 10000 g for 1 min, and the pelleted beads were each washed eight times with PBS. Finally, the beads were mixed with 20 μl of 1× SDS loading buffer [50 mM Tris/HCl, 100 mM DTT (dithiothreitol), 2% SDS, 0.1% Bromophenol Blue and 10% glycerol (pH 6.8)] and boiled. The proteins in the loading buffers were separated by SDS/PAGE (15% gel). Because Xog1p(151–268)-6H migrated at the same position as streptavidin derived from the streptavidin beads, Western blotting was employed to detect Xog1p(151–268)-6H. After SDS/PAGE of the Xog1p(151–268)-6H sample, proteins were transferred on to a PVDF membrane (Pall). The membrane was then blocked with 3% BSA/PBST (PBS containing 0.1% Tween-20) and incubated with histidine-tagged antibody (Santa Cruz Biotechnology, 1:1000 dilution) at 25°C for 2 h. After washing with PBST, HRP-conjugated goat anti-mouse antibody (Jackson ImmunoResearch Laboratories, 1:10000) was incubated with the membrane for 1 h at 25°C. Immunoreactive bands were visualized using enhanced chemiluminescence (Millipore).
Kd values for Xogp1(41–438)-6H and the AMPs
The ADS (affinity detection system; Affinity-sensor New Technology Co.) employing quartz crystal microbalance was used for measuring the dissociation constants . The molecular interaction could be observed by the decrease in vibration frequency of the quartz chip, ΔF. Briefly, AT-cut quartz chips were first activated with 2.5% glutaraldehyde for 30 min, then washed with double distilled water, and coated with Xog1p(41–438)-6H (100 μl, 100 μg/ml), all at room temperature. After 1 h, Xog1p(41–438)-6H solution was discarded, and the chip was washed twice with PBS. Unbound aldehydes were then blocked using 1 M ethanolamine at room temperature for 30 min. After an additional wash with double distilled water, the chip was bathed in 100 mM sodium cyanoborohydride for 10 min and then washed with 50 mM sodium acetate (pH 5.5) (Amresco). The Xog1p(41–438)-6H chip was installed in the flow cell of the ADS, and 50 mM sodium acetate (pH 5.5) was pumped through the cell at 50 μl/min. After the frequency had stabilized, 300 μl of AMP (0.5, 1, 2, 5, 10, 15 or 20 μM), which had been passed through a 0.22-mm-pore-size filter, was injected into the cell and the frequency decrease was recorded. The Kd values were calculated using the specific binding model of Prism 5.0.
Xog1p(41–438)-6H exoglucanase activity
Each of the peptides, LL37, hBD-3, CYC3-3, TAT and 3×FLAG (final concentrations 0, 0.1, 1, 3, or 10 μM) was incubated with 0.5 nM Xog1p(41–438)-6H at 4°C immediately before use. After a 30-min incubation period, each Xog1p(41–438)-6H/peptide mixture was incubated with 8 mg/ml laminarin (Sigma) in 50 mM sodium acetate (pH 5.5). Glucose oxidase/peroxidase reagent (100 μl; o-dianisidine dihydrochloride, 50:1 dilution; Sigma) was added to each mixture, and all samples were then incubated at 37°C for 15 min. The reactions were stopped by adding 100 μl of 12 M sulfuric acid (J.T. Baker), and the absorbance of each mixture was immediately measured at 540 nm to determine the concentration of H2O2, which is a by-product of glucose oxidation and is produced at the same molar amount as glucose. The experiments were performed in triplicate. The glucose concentration was calculated as the mean±S.E.M. using Prism 5.0.
LL37 and hBD-3 kill C. albicans in a dose-dependent manner
Although LL37 and hBD-3 are both cationic peptides, they have different secondary structures, α-helical  and β-sheet  respectively. To assess the functional effect(s) of secondary structure on a cationic AMP, both LL37 and hBD-3 were used in the present study. Both AMPs were chemically synthesized and their candidacidal activities were assessed. After each AMP, at various concentrations, had been incubated with C. albicans at 30°C for 30 min, the cells were plated and the number of colonies was counted the next day (Figure 1). At 10 μM, hBD-3 reduced the number of colonies by 67.9% in comparison with the number of control colonies. Conversely, 3 μM hBD-3 had no apparent candidacidal activity. Only above a concentration of 10 μM did LL37 display candidacidal activity, and even then it was less effective than was hBD-3. Both AMPs at a concentration of 100 μM killed all C. albicans cells.
The antifungal activities of LL37 and hBD-3 and their abilities to reduce C. albicans adhesion to plastic
LL37 and hBD-3 inhibit the adhesion of C. albicans to plastic
To study the inhibition of C. albicans adhesion to plastic, yeast cells were treated with each of the AMPs at a concentration of 0, 3 or 10 μM at 37°C for 30 min. Then cells that had adhered to the wells were scraped and plated on to agar that contained the YPD medium. After overnight culture, the colony numbers were counted. At 10 μM, LL37 decreased C. albicans adhesion by 59%, and at 3 μM, hBD-3 decreased adhesion by 35% (both concentrations are non-lethal doses). At the concentration tested (10 μM), the control peptides CYC3-3, TAT and 3×FLAG, chosen because their net charges are neutral, positive and negative respectively, did not affect adhesion. Therefore the interaction of both LL37 and hBD-3 with the cell-wall components reduced the ability of C. albicans to adhere to plastic.
Xog1p-6H fragments interact with LL37 and hBD-3
Xog1p(41–438)-6H, which did not contain the N-terminal hydrophobic region of the full-length protein (residues 1–40), Xog1p(41–150)-6H, Xog1p(151–268)-6H, Xog1p(269–352)-6H and Xog1p(353–438)-6H were constructed, expressed, purified and prepared as described above (Figure 2A). However, Xog1p(353–438)-6H was not expressed in E. coli. After purification and SDS/PAGE analyses, the positions of the other fragments in an 12% polyacrylamide gel corresponded to 46 kDa [Xog1p(41–438)-6H], 14 kDa [Xog1p(41–150)-6H], 13 kDa [Xog1p(151–268)-6H] and 10 kDa [Xog1p(269–352)-6H], which are the expected molecular masses of the fragments (Figure 2B).
Recombinant Xog1p-6H fragments
ELISA was used to assess the interactions between Xog1p(41–438)-6H and LL37 or hBD-3. Both AMPs were coated on to ELISA plates with Xog1p(41–438)-6H serving as the probe. The interactions of Xog1p(41–438)-6H with LL37 and hBD-3 were 2.5- and 3-fold stronger respectively than were the interactions measured with the control sample (no peptide) or samples containing CYC3-3, TAT or 3×FLAG (Figure 3A). Furthermore, to identify the portions of Xog1p(41–438)-6H that interacted with the AMPs, the Xog1p-6H fragments were pulled down with the BA AMPs. Figure 3(B) shows Xog1p(41–438)-6H, Xog1p(41–150)-6H, Xog1p(269–352)-6H and Xog1p(353–438)-6H interacted with BA-LL37 and BA-hBD-3. Because Xog1p(151–268)-6H has the same molecular mass as monomeric streptavidin, we assessed the ability of Xog1p(151–268)-6H to interact with the AMPs using Western blotting (Figure 3C) and found that Xog1p(151–268)-6H also bound LL37 and hBD-3. Therefore Xog1p possesses multiple regions that interact with LL37 and hBD-3.
Binding assays for Xog1p(41–438)-6H and the peptides used in the present study
The binding affinities of Xog1p(41–438)-6H for the AMPs were next quantified using affinity detection, which relied on a reduction in the acoustical frequency (ΔF) after mixing immobilized Xog1p(41–438)-6H with a solution that contained an AMP at a specified concentration. The observed ΔFs indicated that the AMPs interacted with immobilized Xog1p(41–438)-6H. Using the values of ΔF/peptide and ΔF, and the specific binding function in Prism 5.0, binding isotherm and Scatchard plots were obtained and Kd values were determined. The Kd value for Xog1p(41–438)-6H and LL37 was 1.41±0.10 μM and for Xog1p(41–438)-6H and hBD-3 was 7.52±0.64 μM (Figure 4).
Binding isotherm and Scatchard plots for the measurement of the Xog1p(41–438)-6H/LL37 and Xog1p(41–438)-6H/hBD-3 dissociation constants obtained by affinity detection
Xog1p(41–438)-6H β-1,3-exoglucanase activity is elevated by LL37 and hBD-3
Because LL37 and hBD-3 bound Xog1p(41–438)-6H in vitro, the question of their possible biological function(s) in relation to Xog1p needed to be addressed. We hypothesized that the hydrolytic activity of Xog1p might be affected when the enzyme bound LL37 or hBD-3. Xog1p(41–438)-6H was incubated with each peptide at a specified concentration for 30 min at 4°C. Then, the exoglucanase activity in each mixture was measured using laminarin as the substrate (Figure 5). The concentration of H2O2, the by-product of glucose oxidation, was then measured. LL37 and hBD-3 both increased the concentration of H2O2 (reported as glucose concentration in Figure 5), and at a concentration of 1 μM or greater, LL37 and hBD-3 enhanced Xog1p(41–438)-6H activity more than 1.8- and 1.9-fold respectively compared with the control. Conversely, the control peptides did not enhance Xog1p(41–438)-6H activity.
Xog1p(41–438)-6H activity assay
C. albicans adhesion is reduced by treatment with exogenous Xog1p(41–438)-6H
Exoglucanase activity is crucial for maintaining and remodelling the C. albicans cell wall. Therefore an abnormal concentration or activity of Xog1p may damage the cell wall, thereby reducing the potential infectivity of C. albicans. We hypothesized that an elevated level of Xog1p activity would affect the normal metabolism of cell-wall glucan, and that the observed decreased adhesion of C. albicans to plastic was a consequence of abnormal remodelling of the cell wall. To investigate whether Xog1p(41–438)-6H was cytotoxic to C. albicans, cells were treated with various concentrations of Xog1p(41–438)-6H at 30°C for 24 h, and surviving colonies were then counted (Figure 6A). Xog1p(41–438)-6H killed cells in a dose-dependent and somewhat sigmoidal manner with an IC50 of 1.3 μM. To determine the effect of Xog1p(41–438)-6H on C. albicans adhesion, the candidacidal activity of the enzyme had to be avoided. To determine non-candidacidal conditions for Xog1p(41–438)-6H, a time-course experiment was conducted (Figure 6B). After 2 h at 30°C, C. albicans suspended in PBS and not treated with Xog1p(41–438)-6H began replicating, and replication of C. albicans treated with Xog1p(41–438)-6H was still inhibited after 2 h (Figure 6B). Therefore 1.3 μM Xog1p(41–438)-6H and an incubation time of 2 h comprised the non-candidacidal condition, which was then used to investigate reduction in adhesion to plastic by C. albicans caused by an increase in Xog1p activity. Figure 6(C) shows that 0.65 and 1.3 μM Xog1p(41–438)-6H reduced C. albicans adhesion to plastic by 24.1 and 47.7% respectively. Therefore up-regulation of Xog1p activity in vivo may interfere with adhesion. Notably, the cell-wall morphology of C. albicans that had been exposed to these conditions was apparently not destroyed (Figure 7). These findings suggested that the AMPs may elevate the β-1,3-exoglucanase activity, which subsequently results in abnormal cell-wall glucan metabolism that leads to the inhibition of C. albicans adhesion, even though cell-wall morphology appears to be unaltered.
Cytotoxic effect on and adhesion inhibition of C. albicans by Xog1p(41–438)-6H
C. albicans cell-wall morphology
LL37 and hBD-3 are highly cationic low-molecular-mass AMPs. Cell-membrane and cell-wall carbohydrates are often receptors or co-receptors for positively charged AMPs. For example, the AMP eosinophil cationic protein (pI 10.8) kills mammalian cells via its interaction with cell-membrane heparan sulfate [40,41] and kills bacteria via its association with LPSs [42,43]. The electrostatic affinity between positively charged AMPs and negatively charged membrane carbohydrates is probably a consequence of the fact that carbohydrates are often modified with negatively charged sulfate, phosphate and carboxy moieties . In addition to membrane carbohydrates, membrane (or cell-wall) proteins serve as receptors . We report in the present paper that the recombinant C. albicans cell-wall β-1,3-exoglucanase Xog1p bound LL37 and hBD-3, and, as a consequence, its exoglucanase activity increased. This up-regulated activity might abnormally enhance the hydrolysis of cell-wall β-glucan, thereby damaging the cell-wall integrity, and consequently reducing the ability of C. albicans to adhere to epithelial cells in vivo. However, our previous study showed that the activity of β-1,3-exoglucanase in C. albicans cell wall was decreased with the treatment of LL37 . The cell-wall components are composed of lipids, glycans and proteins to form molecular complexes for regulating biological functions. In addition to Xog1p, our previous study also indicated that LL37 preferentially binds mannan, the main component of the C. albicans cell wall, and partially binds chitin or glucan, which underlie the mannan layer . Therefore some unidentified cell-wall components might form large complexes with LL37, carbohydrates and Xog1p resulting in the reduction of Xog1p activity. In the present study, to clarify the direct effect, the pure recombinant Xog1p(41–438)-6H was incubated with synthetic LL37, and we found that Xog1p activity was up-regulated. The opposite result suggests that other cell-wall components such as glucans may also affect the activity of Xog1p.
Carbohydrates account for 80–90% of C. albicans cell-wall mass, with β-glucan, chitin and mannoproteins as the major components. β-Glucan and chitin maintain the structural skeleton of the cell wall, and β-glucan accounts for 47–60% of the cell-wall mass [45,46]. It has been thought that β-glucan is buried in the cell wall under a layer of mannoprotein . However, a recent study found that the anti-β-glucan monoclonal antibody IgG2b (mAb 2G8) specifically binds to β-1,3-glucan epitopes found on the outer surface of the C. albicans cell wall and, by doing so, inhibits fungal replication and adhesion to human epithelial cells . Therefore it is possible that at least some of the C. albicans cell-wall β-1,3-glucan is on the outer surface of the cell wall. In the present study, we treated Xog1p(41–438)-6H with 0.33, 0.65 and 1.3 μM for 2 h and observed reduced adhesion of C. albicans to plastic; therefore the enhanced activity of Xog1p(41–438)-6H may have damaged the cell wall.
Taken together, our results indicate that, at high concentrations, LL37 and hBD-3 kill C. albicans, but, at low non-cytotoxic concentrations, the AMPs prevent C. albicans adhesion to plastic by elevating the β-1,3-glucanase activity of Xog1p. Furthermore, the AMPs may be developed as peptide drugs for preventing the infection via the inhibition of the C. albicans adhesion.
Hao-Teng Chang conceived the study, designed the experiments and wrote the paper. Pei-Wen Tsai started the project, participated in the cytotoxicity and adhesion assay and assisted in drafting the paper. Hsin-Hui Huang conducted the β-glucanase activity assay and adhesion assay. Yu-Shu Liu measured the Kd of interaction between AMPs and Xog1p(41–438)-6H. Tzu-Shan Chien performed the pull-down assay. Chung-Yu Lan helped with experimental design and participated in discussions. All of the authors read and approved the final paper.
We thank Mr Ting-Jia Chang for assistance with adhesion assays.
This work was supported by the National Science Council, Taiwan, Republic of China [grant numbers NSC-98-2311-B-039-003-MY3 and NSC-100-2627-B-039-002 (to H.-T.C.) and NSC98-2311-B-007-010-MY3 (to C.-Y.L.)].