Candida albicans cell wall as a target of action for the protein–carbohydrate fraction from coelomic fluid of Dendrobaena veneta

The protein–polysaccharide fraction (AAF) isolated from the coelomic fluid of the earthworm Dendrobaena veneta destroys C. albicans cells by changing their morphology, disrupting cell division, and leading to cell death. Morphological changes in C. albicans cells induced by treatment with AAF were documented using DIC, SEM, and AFM. Congo Red staining showed that the fungal wall structure was changed after incubation with AAF. The effect on C. albicans cell walls was shown by AFM analysis of the surface roughness of fungal cell walls and changes in the wall thickness were visualized using Cryo-SEM. The FTIR analysis of C. albicans cells incubated with AAF indicated attachment of protein or peptide compounds to the fungal walls. The intact LC–ESI–MS analysis allowed accurate determination of the masses of molecules present in AAF. As shown by the chromatographic study, the fraction does not cross biological membranes. The Cryo-TEM analysis of AAF demonstrated the ability of smaller subunits to combine into larger agglomerates. AAF is thermally stable, which was confirmed by Raman spectroscopy. AAF can be considered as a potential antifungal antibiotic with activity against clinical C. albicans strains.

Scientific RepoRtS | (2020) 10:16352 | https://doi.org/10.1038/s41598-020-73044-w www.nature.com/scientificreports/ room temperature. The supernatant was discarded, and the fungal cells were suspended in 5 μL of sterile TBS buffer. Next, 3 μL of the fluorochrome were added to each sample and incubated for 3 min at room temperature. C. albicans cells were observed at excitation wavelength λ = 440 nm with the use of an LSM 5 Pa confocal laser scanning microscope (Carl Zeiss, Jena, Germany) with the magnification of 1000×. The experiment was repeated three times. After Congo red staining of the control C. albicans cells and cells treated with the active fraction, the fluorescence of the red glowing cells was measured. Approximately 100 microscopic images were analyzed for each sample, both the control sample and samples incubated with 25, 50, and 100 µg mL −1 of the fraction. The experiment was repeated three times. The fluorescence measurement results from individual samples were statistically analyzed using the Statistica 12.5 program and one-way ANOVA tests with analysis of variance.
SeM analysis of c. albicans cells. The preparation of the control and AAF-treated C. albicans cells for SEM analysis was started by suspending the fungal cells in a fixative, i.e. 4% glutaraldehyde in 0.1 M phosphate buffer, pH 7.0. The samples were then incubated with OsO 4 and dehydrated in increasing acetone solution concentrations (15%, 30%, 50%, 70%, 100%) by centrifugation at 3000 rpm for 30 min. Next, the samples were dried for 24 h in a desiccator using silica gel beads and then sputtered with gold using a K550X sputter coater (Quorum Technologies). The preparations with C. albicans cells were analyzed using a Vega 3 scanning electron microscope (Tescan) at 30 kV.
AfM analysis of c. albicans cells. The surface of both the control and AAF-treated fungal cells was measured using NanoScope V AFM (atomic force microscope) in the Peak-Force Quantitative Nanomechanical Mapping Mode (Bruker, Vecco Instruments Inc., Billerica, MA, USA) and NanoScope 8.15 software. The nominal spring constant of the RTESPA probe (Bruker, Billerica, MA, USA) (silicone tip on the nitride lever) was 40 N/m.

Cryo-SEM analysis of C. albicans cells.
The control and AAF-treated C. albicans cells were centrifuged at 6.000×g for 10 min. The supernatant was then withdrawn and 200 μL of a GH solution (glucose, Na-HEPES, sterile water) were added to the pellet containing fungal cells. The samples were centrifuged again for 10 min and the supernatant was discarded. The C. albicans cells prepared in a small amount of the GH solution were transferred and placed in a sublimation chamber. The process was carried out for 12 min at − 92 °C. After this time, the samples were transferred to the preparation chamber, where they were cut with a special blade and further analyzed under a scanning electron microscope at 5 kV.
Cryo-TEM analysis of AAF. The AAF solution with a protein concentration of 1 mg mL −1 was placed in an ultrasonic chamber (Pol-Sonic, Poland) for 10 min at 40°. Next, the preparation of the AAF specimen consisted in vitrification of an aqueous suspension on the TEM grid with holey carbon film (Quantifoil R 2/2; Quantifoil Micro Tools GmbH, Großlöbichau, Germany). Prior to use, the grids were activated for 15 s in oxygen plasma using the Femto plasma cleaner (Diener Electronic, Ebhausen, Germany). The samples of AAF were vitrified by applying a droplet (3 μL) of the suspension to the grid, blotting with filter paper, and immediate freezing in liquid ethane using a fully automated blotting device Vitrobot Mark IV (FEI Company, Hillsboro, Oregon, USA). After preparation, the vitrified specimens were kept in liquid nitrogen until they were inserted into the Cryo-TEM-holder Gatan 626 (Gatan Inc., Pleasanton, USA) providing a sufficiently low temperature (− 178 °C) during the transfer of the samples to the microscope and during the TEM analyses 25 . Cryogenic Transmission Electron Microscopy (Cryo-TEM) images were obtained using a Tecnai F20 X TWIN microscope (FEI Company, Hillsboro, Oregon, USA) equipped with a field emission gun (FEG) operating at the acceleration voltage of 200 kV. Images were recorded with an Eagle 4k HS camera (FEI Company, USA) and processed with TIA software (FEI Company, USA).

Fourier transform infrared spectroscopy with attenuated total reflection (FTIR-ATR).
Information about the structure of organic material is provided by characteristic absorption bands describing a selected functional group of compounds. The position of the absorption bands in FTIR (Fourier transform infrared spectroscopy) is related to the change in the energy of particles resulting from stretching and deformation vibrations of interconnected atoms 26 . To investigate the molecular structure of C. albicans before and after the AAF treatment, FTIR-ATR spectra were recorded using a Thermo Scientific FTIR Nicolet 8700 spectrometer. They were obtained with the use of the ATR method with a diamond crystal in the wave number range of 4000-400 cm −1 and spectral resolution of 4 cm −1 . The spectra were recorded directly from the surface of the samples at room temperature. The spectra were subjected to ATR correction, baseline correction, and normalization.
Raman spectroscopy analysis. The effect of temperature on the secondary structure of proteins was evaluated using a Raman spectroscopy device (Renishaw, UK) equipped with a 785 nm diode laser and a 1200 l/ mm diffraction grating. At the beginning of the measurements, the spectroscope was calibrated using the Raman band of a silicon reference sample at 520.7 cm −1 . Temperature analysis was conducted using Linkam temperature control cells from 23 to 45 °C with 2 °C steps and from 45 to 165 °C with 10 °C steps. Raman spectra of the proteins were collected each time after temperature stabilization in the range from 200 to 3200 cm −1 . The intensity of Amide I band in the range between 1620 and 1715 cm −1 (labeled in Fig. 12a) was used to determine the percentage composition of particular secondary structures of proteins. The intensity of bands assigned to the Scientific RepoRtS | (2020) 10:16352 | https://doi.org/10.1038/s41598-020-73044-w www.nature.com/scientificreports/ alpha helix, beta sheet, beta turn, and random coil protein structures was obtained by the curve-fitting process, as described in our earlier study 24 . Size exclusion chromatography (SEC) analysis of AAF. 4  . The sample was loaded onto the column using the CTC Pal Autosampler (CTC Analytics AG, Zwinger, Switzerland) with the injection of 1µL of sample solution and separated on a 5CA-CL-300, 5 μm, 300 Å, size 0.5 × 150 mm column (Exigent) using a 77-min gradient (3-90% B, Solvent A 0.1% formic acid in H 2 O, Solvent B 100% Acetonitrile and 0.1% formic acid). The flow rate was 20 μl/min. All details of the LC gradient and intact MS method parameters as well as the final raw data for the intact analysis are available at the MassIVE repository (Computer Science and Engineering University of California, San Diego, Center for Computational Mass Spectrometry, https ://massi ve.ucsd.edu/Prote oSAFe /stati c/massi ve.jsp) under the DOI number https ://doi.org/10.25345 /C5M10 F. The intact LC-ESI-MS spectrum was analyzed in PeakView (Sciex), and all protein masses were reconstructed with the Bio Tool Kit micro-application based on a maximum entropy algorithm. Protein reconstruction was done using an output mass range between 3.000 and 50.000 Da with a step size of 1.
Biopartitioning micellar chromatography (BMc) analysis of AAf. The BMC analysis was carried out according to the procedure described in the previous paper 27 . The Shimadzu Vp liquid chromatographic system equipped with an LC 10AT pump, an SPD 10A UV-Viz detector, an SCL 10A system controller, a CTO-10 AS chromatographic oven, and a Rheodyne injector valve with a 20-μL loop was used to obtain chromatographic data. The Class-Vp program was used to acquire and archive the data.
A stainless-steel C18 endcapped packed reversed-phase column (5 m, 125 mm × 4 mm, I.D., Purospher, Merck, Darmstadt, Germany) was used. Buffered solutions (pH 7.4) of Brij35 (polyoxyethylene (23) lauryl ether; Merck, Darmstadt, Germany, p.a.) were used with the following concentrations: 0.04 M, 0.06 M, 0.08 M, and 0.15 M. 5% v/v isobutanol (POCH SA, Gliwice, Poland, p.a.) was added to each mobile phase as an organic modifier. AAF was prepared in methanol at a concentration of approximately 0.1 mg mL −1 (Merck, Darmstadt, Germany, p.a.). The buffer was prepared from 0.02 M Na 2 HPO 4 and 0.01 M citric acid mixed together (Merck, Darmstadt, Germany, p.a.). Before use, the buffer was vacuum-filtered through a 0.45 μm membrane filter. The flow rate of the mobile phase was 1 mL min −1 . All the measurements were performed at a temperature of 20 °C.
The maximum absorbance of the tested compounds was set at a wavelength λ = 254 nm. The Direct-Q apparatus (Millipore) provided distilled water. Two minutes before use, micellar mobile phases were degassed in an ultrasonic bath. Each measurement was repeated three times. The retention factors were calculated according to the formula: k = (tr − t 0 )/t 0 , where t 0 is the column dead time measured as the retention time of citric acid.

Results
DIC microscopy analysis of C. albicans. The AAF-treated C. albicans cells and the cells from the control culture were analyzed using the DIC technique. Changes in the C. albicans morphology after incubation with the different concentrations of AAF and cells in the control culture were observed (Fig. 1a). The control cells shown in photos A1-A3 were regular, round, with a smooth cell wall.
After incubation with AAF at a concentration of 25 µg mL −1 , cell aggregation, cell wall deformations, and changes in the shape were observed; they are marked with white arrows in pictures B1-B2. In picture B3, the white arrowheads show cells with enlarged vacuoles and the black arrowheads indicate pseudohyphae. Images C1-C3 show cells treated with the AAF at a concentration of 50 µg mL −1 . The white arrows in Fig. 1a C1 indicate cells whose wall has lost its integrity, whereas images C2-C3 show deformed cells with clearly enlarged vacuoles filling almost the entire cells. Figure  In the control culture, 95% of cells had a correct shape. After incubation with 25 µg mL −1 of AAF, the percentage of cells with normal morphology decreased to 79%, while hyphae and pseudo-hyphae represented 6%, cells with enlarged vacuoles accounted for 13%, and clearly deformed cells appeared. After the treatment with 50 µg mL −1 of AAF, the proportion of normal cells decreased to 65%, the number of hyphae and pseudo-hyphae and cells with enlarged vacuoles increased to 13% in each group, and the percentage of deformed cells in the total pool was 9%. After incubation with 100 µg mL −1 of AAF, 51% of the cells retained their correct shape, 29% were hyphae and pseudo-hyphae, while deformed cells with enlarged vacuoles accounted for 10% for each form (Fig. 1b). As a result of disturbed cell division, single chains of cells appeared after incubation with the fraction at each concentration used.   AfM analysis of c. albicans cell wall. Changes in the surface of C. albicans cells exposed to AAF were imaged with the AFM technique. The control culture cell was characterized by regular formation of the cell wall, and the height profile confirms the rounded symmetrical surface, as shown in Fig. 4 A1, A2. After incubation with AAF (100 µg mL −1 ), the cell wall was clearly deformed with a visible collapse in the central part of the cell. The height profile determined by the analysis of the cell surface has two height peaks forming folds with a distinct depression between them, as shown in Fig. 4 B1, B2.

Microscopy analysis of c. albicans cells after congo
Cryo-SEM and AFM analysis of C. albicans surface. The wall surface of the C. albicans control culture cells and those incubated with AAF (100 µg mL −1 protein concentration) was analyzed using Cryo-SEM and AFM. At high magnification, the surface of the cell wall in the C. albicans control culture was not quite smooth, but intensely lumpy, as evidenced by the relief pattern over the entire surface (Fig. 5 A1, B1). After the exposure www.nature.com/scientificreports/ to AAF, the cell wall smoothened. Lumps and granules occurred sporadically (Fig. 5. A2 and B2). In addition, the percentage of surface roughness was determined using AFM. After the incubation with AAF, the roughness increased threefold compared to the control value. The roughness R a of the cell wall surface was 2.98 nm in the control and 9.94 nm after the AAF treatment.

FTIR-ATR analysis of C. albicans cells.
To study the effect of AAF on the C. albicans cell wall, FTIR spectroscopic studies were performed after incubation of the C. albicans fungus with AAF at the protein concentrations of 25, 50, and 100 µg mL −1 . Control and AAF-treated cells were analyzed. The spectroscopic studies of the control cells showed the presence of a characteristic high-intensity band in the range of 3600-3200 cm −127 . This band corresponds to the asymmetrical and symmetrical stretching vibrations of the O-H and N-H groups (Fig. 6). The presence of a broad band corresponding to OH groups suggests the presence of fatty acids in the C. albicans control samples, while the presence of NH groups is most likely related to the presence of amines. The FTIR-ATR spectra also show characteristic bands in the 3000-2850 cm −1 wave number range derived from the stretching vibrations of the C-H aliphatic groups. The presence of amide bands in the range of 1650-1515 cm −1 corresponds to the bending vibrations of NH groups and C=O stretching. The peak at 1539.37 cm −1 corresponds to the bending vibrations of NH groups, while the intense peak at 1635.93 cm −1 corresponds to the stretching vibrations of C=O groups derived from secondary amides. The FTIR-ATR spectra also showed characteristic bands in the range of 1040-1156 cm −1 corresponding to the polysaccharides present in C. albicans.
The FTIR spectroscopic studies of AAF-treated C. albicans showed changes in the peak intensity in the range of wave numbers 1615-1500 cm −1 (Fig. 6). An increase in the intensity of the peak at 1635.93 cm −1 and the peak Biopartitioning micellar chromatography (BMc) analysis of AAf. The analysis of the obtained BMC chromatograms was carried out according to the procedure described in the previous paper 27 . Two evident peaks were noted in each of the tested systems, suggesting the existence of two substances in the sample. The relationships between the logarithm of the retention factor (logk) and the micellized surfactant concentration (C M ), i.e. the total surfactant concentration minus CMC 28 , are presented in Fig. 7. Excellent linearity of the relationships was found over the whole eluent composition range with the correlation coefficient R 2 above 0.99. It allowed extrapolation of logk values. Values extrapolated to pure water logk (denoted as logk w ) are considered an alternative to the logarithm of the n-octanol-water partition coefficient (logP o/w ) lipophilicity descriptor. The obtained logk w values were as follows: for peak 1: -0.216, for peak 2: -0.558. The analysis of the logk w values indicates that both substances are very weakly lipophilic. Moreover, taking into account the molecular weight of the analyte (24 kDa), it can be concluded that such a large molecule does not penetrate biological barriers. This is in line with Lipinski's rule of five 29 . However, to verify this, other physicochemical parameters of a molecule should be analyzed. According to the Hansch approach, the steric, electronic, and lipophilic characteristics of a molecule are the most important parameters governing the transport and drug-receptor interaction 30 . Moreover, as indicated in numerous studies, the hydrogen-bonding potential is also an important factor in predicting the ability of a molecule to cross biological barriers [31][32][33] .  Based on the determined retention times and calibration function for the column, molecular weights of the components of AAF were estimated ( Fig. 8B and Table 1). Based on the results obtained with the SEC method, it can be concluded that the main peak corresponds to a 24-kDa protein.
The high heterogeneity of the samples suggested by previous analysis was further confirmed using the UHPLC approach. The UHPLC analysis of the biologically active fraction in standard conditions revealed the presence of the main component at around 10 min (Fig. 9). The peak is wide at the base and not entirely symmetrical, suggesting that it may contain more than one protein/compound. Additionally, other peaks are visible on the chromatogram before and after the main peak. The multiple peaks present on the chromatogram confirm this hypothesis. The longer retention time for the main component suggests higher hydrophobicity of this molecule or higher molecular weight.
intact Lc-eSi-MS analysis of AAf. To analyze the peptides and proteins present in AAF, we used electrospray ionization mass spectrometry coupled with a chromatographic system (LC-ESI-MS). The sample of the lyophilized fluid was dissolved in 50 mM ammonium acetate and prepared for MS analysis (see Materials and Methods). The application of over an hour-long LC gradient allowed separation of peptides and proteins (Fig. 10) contained in the AAF and determination of the initial mass of most of these compounds (see Tables 1  and 2 and Supplementary materials). The original spectrum and data on the method have been deposited in the MassIVE repository (see Materials and Methods). The analysis of the total ion chromatogram first shows a group of intense signals with a retention time between 8 and 15 min (Fig. 10B). It is obvious that it does not contain a single compound, but rather a group of molecules with similar mass and properties. Reconstruction of masses in this group is presented in Table 2 and additional drawings are shown in Supplementary materials   Figures 2S-6S). In this range, we successfully determined the masses of peptides at 3.6 kDa and 5.1-5.4 kDa as well as small proteins.
Mainly proteins began to elute after 15 min. The LC gradient with the increasing organic solvent content yielded a well-separated signal for some proteins, but the proteins eluted in groups in most cases due to the complex composition of AAF. Even in narrow time windows, it was sometimes difficult to extract m/z signals for one protein (see Supplementary materials, Fig. 7S-20S). The most confident masses of proteins are presented in Table 3. The most intense signal at the TIC chromatogram is the peak at 35.86 min, which mainly represents proteins with masses from 12.2 to 12.5 kDa (Fig. 16S). As demonstrated by the analysis, all protein mass values reconstructed above 15 min focus in ten main ranges of 9. Cryo-TEM analysis of AAF. The Cryo-TEM analysis showed two morphological forms of the preparation of AAF: a round compact structure with a dark color in the microscopic image (Fig. 11 A1-A3) and a round loose structure consisting of smaller subunits (Fig. 11 B1-B3). These forms were characterized by light tint in the image. The smaller forms had a marked tendency to combine into larger structures, which can be clearly seen in images B1-B3.
Raman spectroscopy analysis of the temperature effect on the AAF protein structure. The Raman spectroscopy method was effectively used to determine the secondary structure of proteins in our earlier investigations 24 . Therefore, in this study, this spectroscopic method was also used for evaluation of the temperature effect on the protein secondary structure. Figure 12a shows an example of a Raman spectrum of proteins with a labeled Amide I band. The intensity of bands in the Amide I region was used to estimate the alpha helix, beta sheet, beta turn, and random coil content in the studied material. The curve-fitting process was applied to estimate the percentage amount of alpha helix, beta sheet, beta turn, and random coil conformations. The curve-fitting process in the Amide I band region and the position of bands assigned to particular secondary structures of proteins was shown and described in our earlier study 24 . Figure 12b presents the Raman spectra of proteins at different temperatures (from 25 to 165 °C). In these spectra, there are no significant changes in the protein structure with the increasing temperature. The changes are shown in detail in Fig. 12c,d, which indicate the percentage changes in the protein secondary structure with the rise in temperature. In Fig. 12c, the content of proteins in the alpha helix conformation slightly decreases (from about 23% at 23 °C to about 18% at 45 °C), whereas the content of proteins with the random coil structure slightly increases (from about 14% at 23 °C to about 17% at 45 °C) with the temperature rise (Fig. 12c). No changes in the content of proteins in the beta turn and beta sheet conformation are visible between 23 and 45 °C. In a wider temperature range (Fig. 12d) from 25 to 165 °C, the content of proteins with the beta turn and beta sheet structure is maintained as well. In turn, the content of proteins in the alpha helix conformation decreases (from about 23% at 25 °C to about 5% at 165 °C), whereas the content of proteins in the random coil structure increases (from about 15% at 25 °C to about 28% at 165 °C) with the rise in temperature. Moreover, the content of proteins in the alpha helix and random coil conformation changes abruptly at about 100 °C (Fig. 12d).

Discussion
Natural sources have been very important candidates for the development of new drugs for many years. Compounds of natural origin are a source of various chemical structures often showing the desired biological activity 16 . They form a structurally privileged group in the process of binding to specific enzymes, receptors, or other binding sites and in this way show high affinity for structures found in living organisms. www.nature.com/scientificreports/ Fungal diseases are a major medical problem, and some types of Candida are resistant to the antifungal antibiotics used to date. C. albicans is the most common fungus causing infection in humans. AAF has been identified as a protein-carbohydrate fraction from D. veneta earthworm CF with antifungal activity against C. albicans, C. krusei strains 24 , and A549 lung cancer cells with approximately 90% cytotoxicity in vitro 34 . It is important that AAF showed no cytotoxicity effect on normal bronchial epithelium BEAS 2B cells. In addition, AAF exhibited selective cytotoxic action against colorectal cancer cells 35 . In both Candida and tumour cells, AAF caused cell death via apoptosis and necrosis. Our previous research on the earthworm D. veneta showed that the symbiotic bacterium associated with the intestine of earthworms Raoultella ornithinolytica was capable of producing metabolites with antifungal activity against C. albicans 36 and had anti-cancer activity against ovarian cancer line TOV-112D and breast ductal carcinoma T47D 37 . The effect of the AAF fraction on C. albicans cells was stronger than that of a polysaccharide-protein complex from R. ornithinolytica metabolites, with induction of a visible process of apoptosis that had not been observed earlier 24 . Therefore, this trend in the search for a potential pharmaceutical with anti-fungal and anti-cancer effects seems to be relevant. There is evidence that C. albicans infection increases the risk of cancer and metastasis. In particular, this opportunistic pathogen exploits the state of immunosuppression in chemotherapy patients. Candida can cause cancer through such mechanisms as production of carcinogenic by-products, induction of inflammation, induction of Th17 responses, and molecular mimicry 38 . C. albicans is able to elicit an inflammatory response that increases the adhesion of cancer cells to the liver endothelium, which was found in in vitro studies 39 . Oral candidiasis is the most common opportunistic fungal infection and has been associated with pre-cancer and cancer. Candida is involved in lung cancer, which is one of the most prevalent neoplastic diseases in the world. Thus, the search for a compound with a bi-directional effect on candidiasis and cancer without a clear cytotoxic effect on normal cells is well founded and worth developing and exploring.
Our research indicates that AAF acts on the C. albicans cell wall. The structural integrity of the cell wall of C. albicans is necessary for the survival and reproduction of yeast. The damaged cell wall causes osmotic disorders in the fungal cell, rupture of the cell membrane, and, consequently, outflow of cytoplasmic content and cell death 40 . The C. albicans cell wall is a very important cell element playing a key role in determining the balance between commensalism and disease. Cell wall proteins represent many virulence attributes of pathogens, and cell wall carbohydrates act as PAMPs, which induce both immune defence and potential overactivation of the inflammatory response 41 . C. albicans is able to reduce its detection by the body's immune system through masking the β-(1,3)-glucan present in the inner cell wall by an outer layer made of glycosylated mannoproteins. The β-(1,3)-glucan layer can be unmasked by mutations, drugs, or bioactive compounds that destroy the cell wall. This leads to detection of Candida by immune cells by the Dectin-1 receptor and C-type signalling lectin 42,43 . Glucan together with chitin are structural components of the fungal cell wall responsible for the integrity and physical strength of this structure. The production and assembly of glucan in C. albicans requires a number of enzymes and mechanisms that are characteristic only for fungi. The process of building the fungal cell wall is an interesting target for antifungal therapies 44 . Caspofungin and micafungin are the best-known antifungal antibiotics inhibiting β-(1-3)-glucan synthesis; however, mutations in β-(1,3)-glucan synthesis that confer resistance to caspofungin have already been observed. The mechanism of action of AAF is probably based on the exposure of β-(1,3)-glucan in C. albicans cells, which was demonstrated using the Congo Red staining in the present study. This property predisposes this compound for assessment of its potential as an antifungal antibiotic.
The most commonly used antibiotic acting on the Candida cell wall is amphotericin B, i.e. a polyene antibiotic from the group of heptaenes produced by Streptomyces nodosus. The mechanism of action of amphotericin B is based on binding the drug to sterol-containing fungal cell membranes and changing their permeability 44 . The   www.nature.com/scientificreports/ clinical use of amphotericin B has many side effects, such as nausea, vomiting, fever, hypokalemia, hypomagnesemia, and kidney or liver damage. Studies on the effects of amphotericin B on the same C. albicans strain as in the AAF study have demonstrated that the antibiotic is able to bind to groups of polar sugar monomers present in the cell wall, which leads to cell wall coating. The experiment has also shown that the drug caused an AAFlike effect in fungal cells, e.g. cell collapse, cell wall thickness and structure disorders, budding scars, and cell elongation 45 . Similar changes in the C. albicans cells have also been observed in our Cryo-SEM and AFM experiments following the AAF treatment. After the incubation with AAF, the fungal cells walls were irregular and had varying thickness. The cells collapsed and the wall surface roughness changed, which was evident in the AFM images. Morphologically changed cells were observed by light microscopy using DIC. After the application of the active fraction at the concentration of 100 µg mL −1 , changed cells constituted 49% of all imaged cells. While describing their morphology, it should be remembered that maintaining a regular shape is not synonymous with To check whether AAF binds to the C. albicans cell wall, FTIR analyses of C. albicans control cells and AAFtreated cells was performed. The increase in the intensity of the band at 1636 cm −1 and the band at 1539 cm −1 suggests an increase in the number of C=O groups and NH groups derived from amide compounds. This may be related to the attachment of the AAF-derived protein fraction to the C. albicans cell wall, as evidenced by SEM and Cryo-SEM microscopy. A local increase in the cell wall thickness was clearly visible in the microscopic images. Local layering of AAF-derived substances on C. albicans cells can be assumed. The FTIR spectroscopic studies suggest the protein or peptide nature of the substance attached to the outer cell wall of the fungi.
To confirm our assumptions about the mechanism of the AAF interaction with C. albicans cells, we decided to check the permeability of AAF through the biological barrier. Biopartitioning Micellar Chromatography (BMC) is one of the non-cell based in vitro methods. BMC systems are considered to mimic the biological environment due to their similarity to biological barriers and extracellular fluids; therefore, they are useful in describing e.g. intestinal absorption, skin permeability, blood-brain barrier penetration, etc. 33,[46][47][48] . Moreover, BMC is often used to determine the lipophilicity of various compounds 49 . The study results indicate poor lipophilicity of the tested fraction and suggest its impermeability through biological barriers; therefore, the cell wall is considered the target of AAF.
The observations correspond to the results obtained after Cryo-TEM and MALDI analysis indicating a complex structure of AAF. The images of AAF after the Cryo-TEM analysis suggest that the compound is polymeric in nature. One of the forms clearly forms larger agglomerates, creating a spherical structure consisting of many separate subunits. Most likely, such a complex structure causes impermeability through the cell membrane.
The intact LC-ESI-MS approach is the next step undertaken to obtain as much information as possible on the peptide-protein composition of AAF. In our previous study, intact MALDI analysis in three matrices (SA, sDHB, and DHB) gave a preliminary picture of peptides and proteins with masses from 5 to 44 kDa present in the examined sample. The intact LC-ESI-MS analysis presented in this study allowed accurate determination of With the use of the LC-ESI-MS analysis, we were able to identify masses below 5 kDa (3.6 and 4.0, 4.7 kDa, Table 1), which indicates the presence and potentially important role of the peptide components of AAF in the biological activity. It is, therefore, necessary to study further the peptides contained in this fluid, which may reveal new biologically active compounds. We identified a group of small proteins with a mass of 7-8 kDa. Their signals were well separated on the chromatographic column, and we were able to designate their masses with high accuracy ( Table 1). The next step will aim to establish whether they are separate molecules with different amino acid sequences or a group of two-three small proteins with modifications (acetylation/methylation/oxidation). This problem will be addressed in the planned Top-down analysis. Despite the complex nature of the preparation, we were able to determine successfully the mass of proteins in the range of 9 to 14 kDa. We easily assigned the charge states to specific m/z values in the spectrum, which allowed us to perform mass reconstruction ( Table 2, Figs. S6-S19). With great success, we were also able to separate proteins, which in the intact MALDI analysis occurred in the form of broad signals or completely unseparated groups with m/z from 15 to 44 kDa. We identified with good accuracy approximately 17 proteins present in the preparation with the use of the electrospray technique combined with chromatographic separation. Only proteins in the range of 30-35 kDa and above 50 kDa were not identified in the intact ESI analysis. Most likely, their concentration in the sample is low and they eluted at the end of the LC gradient, or other better ionizing molecules suppressed their m/z signals. In this case, fractionation of the examined fluid has to be performed to simplify the sample composition. It is also worth noting that some of the identified masses are represented by one clear signal in the spectrum showing the reconstructed masses. However, many of the proteins occur in the form of several masses, between which the differences are in the range of 16-100 Da, which indicates the presence of different isoforms of a given protein.
These may be methylation, formylation, or oxidation (e.g. methionine) modifications, as well as the presence of sugar residues, which were detected in the preparation previously. In the next investigations, we plan to fractionate the preparation and carry out accurate identification of protein and peptide components combined with Top-Down MS analysis and de-novo sequencing, which will allow identification of specific proteins and assignment to particular identified masses.
In the pharmaceutical industry, it is important that any exothermic or endothermic processes that occur during the manufacture of drugs should be examined using thermal analysis methods. During the analysis of the substance, it can be determined whether the compound undergoes changes during preparation of the formulation. These changes have a major impact on the later use of production methods and formulation of the drug. The Raman spectroscopy analysis showed that the AAF fraction did not change its chemical structure under the influence of elevated temperature, which indicates the applicability of the fraction as a preparation.
In the light of the analysis of both C. albicans cells after the exposure to AAF and the fraction itself, it can be concluded that the fraction from CF acts on C. albicans cells by lowering their metabolic activity and causing apoptotic cell death by a direct effect on the cell wall, which was demonstrated in earlier research 24 . The changes in cell morphology and disorders of cell division reported in this study, similar to those caused by the action of amphotericin B, are a result of attachment of the complex proteins to the cell wall and initiation of a series of phenomena leading to cell death.
Antibiotics acting on the fungal cell wall (echinocandins) and cell membrane (azoles, allylamines, or amorolfine) cause side effects that cover a wide range of symptoms. These undesirable effects limit the use of these antibiotics, and the current situation requires development of new drugs with less harmful effects. We hope that the AAF obtained from earthworm CF will not show such toxic properties; therefore, it will be able to replace the antibiotics mentioned above. Due to the high antifungal 50 and anti-cancer 51,52 activity of AAF and no signs of endotoxicity and cytotoxicity towards normal human cells, AAF has been patent pending in Poland, and research into the exact mechanism of this action will be continued.
In conclusion, mycoses are very serious and chronic diseases, and their treatment is difficult and long-lasting. The incidence of fungal infections has increased significantly in recent years. The development of surgical techniques and methods of intensive medical care creates situations favorable for the development of fungal infections, and C. albicans causes over 80% of fungal diseases. The difficulties in the treatment of candidiasis and the numerous side effects of the antibiotics used prompt the search for new compounds that do not exhibit endotoxicity and, at the same time, are effective against C. albicans cells. Exploitation of the potential of earthworms living in an environment rich in fungi seems to be a good way to obtain an effective drug. The obtained compound will be tested in a mouse model in order to analyze fully its immunological parameters and applied in clinical trials in the case of successful results.

Data availability
All data generated or analyzed during this study are included in this article (and its Supplementary Information files).