The fungal strains, C. gattii R265 (does not produce 3-hydroxy C9:0), C. neoformans LMPE 046 (does not produce 3-hydroxy C9:0), C. neoformans UOFS Y-1378 (produces 3-hydroxy C9:0), and C. neoformans LMPE 101 (acapsular strain that does not produce 3-hydroxy C9:0), were used in the study. These strains were grown on yeast-malt-extract (YM) agar (3 g/l yeast extract, 3 g/l malt extract, 5 g/l peptone, 10 g/l glucose, 16 g/l agar; Merck, South Africa) plates at 30°C for 48 h. Cells (representing the respective strains) were separately standardized using a haemocytometer (Marienfield, Germany) to a final concentration of 1 × 10⁶ cells/ml in 10 ml of distilled water before use. For amoebapore (BioWorld, United States) sensitivity testing, five colonies (representing either R265, LMPE 046, UOFS Y-1378, or LMPE 101) were scraped off and suspended in 10 ml of distilled water. At the end, the cells were standardized to prepare final inocula of between 0.5 × 10⁵ and 2.5 × 10⁵ CFU/ml in RPMI 1640 medium (Sigma–Aldrich, South Africa) according to EUCAST guidelines (Arendrup et al., 2015).

The amoeba (Acanthamoeba castellanii) strain LMPE 187, used in the study, was grown on peptone-yeast extract glucose broth, PYG (ATCC medium 30234^TM) at 30°C for a week. The cells (in trophozoite state) were then standardized using a haemocytometer to a final concentration of 1 × 10⁵ cells/ml in 10 ml of sterile PYG broth and the viability of cells was determined to be 80% using trypan blue.

The 3-hydroxy fatty acid standard (3-hydroxy C9:0), was purchased from Larodan (Sweden). This compound was tested at a final concentration of 0.2 mM, which is the estimated physiological concentration secreted by C. neoformans UOFS Y-1378 (Madu et al., 2015).

Crude GXM was isolated in anticipation of experiments wherein it was used for comparison purposes. The isolation was done according to a protocol previously detailed by Zaragoza et al. (2008). In short, a loopful of scraped C. neoformans UOFS Y-1378 (grown for 48 h on YM plates) colonies was used to inoculate a 500 ml conical flask containing 250 ml of Difco-yeast nitrogen base (YNB) broth (6.7 g/l YNB and 40 g/l glucose; Becton, Dickinson and Company, United States) that was supplemented with 2% glucose (Merck, South Africa). The flask was placed on a rotary shaker (160 rpm/min) and cultivated for 48 h at 30°C. The cells were washed five times with distilled water and finally suspended in 40 ml of distilled water. The cells were then heat-killed at 55°C for 30 min before being irradiated (dosage of ∼200 grays) with a ¹³⁷Cs gamma-irradiator (kept at HEPRO Cape, South Africa) in order to make cells shed their capsule (GXM). After irradiation, the cells were centrifuged at 3500 × g to mobilize the shed GXM into the supernatant.

Next, the lipids were removed from the collected supernatant via using a modified Folch lipid extraction protocol. In brief, the supernatant (10 ml) was transferred to a 50 ml Falcon tube (Becton-Dickinson Labware, United States). Following which, 10 ml of methanol-chloroform (HPLC-grade) solution (Merck, South Africa; 1:1, v/v) was added. The suspension was vortex mixed and allowed to stand for 20 min. Thereafter, distilled water (10 ml) was added to the above solution and allowed to stand for a further 20 min. The chloroform fraction that contained 3-hydroxy fatty acids was disposed and the water fraction containing the GXM was kept. The isolated GXM was then lyophilised, weighed and reconstituted in sterile water to make a stock solution of 10 mg/ml.

To confirm the isolation, 100 μl of the delipidated stock solution was subjected to an agglutination (Meridian Bioscience, Inc., United States) assay and ELISA (IMMY, United States) assay. Both assays were performed according to manufacturer’s protocol. The optical density (OD) readings were measured at 450 nm using a spectrophotometer (Biochrom EZ Read 800 Research, United Kingdom).

A 100 μl suspension (in YPG broth) of standardized amoebae was seeded into designated wells of a sterile microtitre plate (Greiner Bio-One, Germany). Next, the cells were challenged with 3-hydroxy C9:0 (1:1, v/v) at a final concentration of 0.2 mM or 100 μl of isolated GXM to yield a final concentration of 2.5 mg/ml. Non-treated amoeba cells were included as a control. The plate was then incubated at 30°C for a 6 h period. At the end, the supernatant was aspirated and accordingly transferred to a sterile microtitre plate specific for the human-based fetuin A (Abcam, United Kingdom) while the cells were kept to harvest the lysate according to a protocol and materials provided in the NADPH oxidase-1 (human-based) kit (LifeSpan BioSciences, United States). The lysate was then transferred to a sterile microtitre plate specific for NADPH oxidase-1. The supernatant and lysate plates were, respectively, treated according to the manufacturer’s protocols. Each ELISA plate was then read at 450 nm using a Biochrom EZ spectrophotometer.

Standardised cells (1 × 10⁶ cells/ml), in distilled water, of the strain LMPE 101 were used in this test. A 75 μl suspension of cells was separately transferred to wells of a sterile microtitre plate. These cells were then treated with 75 μl of hydrogen peroxide (60 μM). In other wells, cells (75 μl) were treated with 75 μl of hydrogen peroxide (240 μM) in the presence of either GXM (150 μl of 10 mg/ml) or 3-hydroxy C9:0 (150 μl of 0.8 μM). The plate was incubated for 3 h at 37°C. After incubation, the plate was gently agitated and the contents aspirated. Of this, 100 μl was used to make a 1:10 dilution in distilled water. Thereafter, 100 μl was used to create a uniform lawn of cells on a YM agar plate. The plates were incubated for 48 h at 30°C before colony forming units (CFU) were counted. The considered concentration of hydrogen peroxide was similar to that reported to accumulate inside phagosomes (Dupré-Crochet et al., 2013).

In addition, the levels of hydrogen peroxide in the medium containing hydrogen peroxide-treated cells and hydrogen peroxide-treated in the presence of GXM or 3-hydroxy C9:0 was measured as previously detailed by Noble and Gibson (1970). In brief, a microtitre plate was prepared as above-mentioned and at a specific time interval, i.e., 3-, 6-, and 24-h, an absorbance reading was taken at 360 nm. This reading was then multiplied by a molar extinction coefficient value of 43.6, in order to estimate the levels of hydrogen peroxide.

The effect of amoebapore (amoebal antimicrobial peptide) was independently assessed on biological duplicates of the strains R265, LMPE 046, UOFS Y-1378, and LMPE 101 at final concentrations (prepared in RPMI 1640 medium) of 3.25 μM or 7.5 μM. In short, a 100 μl of the standardized inoculum (0.5 × 10⁵ and 2.5 × 10⁵ CFU/ml) was aliquoted into designated wells of a microtitre plate. The cells, in the wells, were then treated with 100 μl of amoebapore (at twice the stated desired final concentrations). The plate was incubated for 48 h at 37°C. After 48 h, the OD of each well was measured at 562 nm using a Biochrom EZ spectrophotometer. The resultant effect of amoebapore on the growth of treated cells was compared to that of non-treated cells.

In order to assess if amoebapore killed cells by creating pores on cell walls, cells (i.e., UOFS Y-1378 and LMPE 101) treated with 7.5 μM, were prepared on a separate microtitre plate as stated above (for drug sensitivity testing) before being viewed by transmission electron microscopy (TEM). The cells were prepared for TEM viewing according to the method of van Wyk and Wingfield (1991). In short, the cells were chemically fixed with 1.0 M (pH 7) sodium phosphate-buffered gluteraldehyde (3%) for 3 h and then for 1.5 h in similarly buffered osmium tetroxide. The cells were next dehydrated in a graded acetone series, embedded in epoxy resin and polymerized at 70°C for 8 h. An LKB III Ultratome was used to cut 60-nm sections with glass knives. Uranyl acetate was used to stain sections for 10 min, followed by lead citrate for 10 min and the preparation viewed with a Philips EM 100 transmission electron microscope.

All experiments, unless otherwise stated, were performed in triplicate. Where appropriate, a student t-test was used to determine the statistical significance of data between the control and different experimental conditions.

In this study, fetuin A was assayed based on prior unpublished work in our group, which revealed that macrophages challenged with 3-hydroxy fatty acids produced this protein. This protein is a major globulin component of serum that has been shown to regulate the function of macrophages (Wang et al., 1998). More to the point, this glycoprotein has been reported to mediate the uptake of particulate material such the uptake of some bacteria (Escherichia coli and Staphylococcus aureus; both are not known to produce 3-hydroxy fatty acids) including apoptotic cells, by phagocytes (van Oss et al., 1974; Jersmann et al., 2003).

The GXM, which was used for comparison purposes in this experiment and others, was successfully isolated (Supplementary Figure S1). Our results suggest that amoebae also produce a fetuin A-like protein (Figure 1), which may have a similar function as that expressed in macrophages. This assertion is reasonable as macrophages have been theorized to have evolved from amoebae (Siddiqui and Khan, 2012) and in particular, this protein may have evolved from cystatin by gene duplication. Hence, it is conceivable that amoeba may have this protein. When considering the results, it was clear that there was a significant reduction (p < 0.01) in the levels of fetuin A when cells were challenged with either GXM (74% reduction) or 3-hydroxy C9:0 (68% reduction) compared to non-treated amoebae. Given the function of this protein, which (among others) is to promote macrophage phagocytosis (Jersmann et al., 2003), the GXM results are conceivable primarily because this molecule is said to be anti-phagocytic (Del Poeta, 2004; Zaragoza et al., 2009). Therefore, by extension, the data suggests that like GXM, 3-hydroxy C9:0 [which is intimately associated with cryptococcal capsules (Sebolai et al., 2007)] may also be anti-phagocytic. This finding may, in part, explain the findings of Madu et al. (2015), wherein they reported that the addition of 3-hydroxy C9:0 to the co-culture media resulted in reduced ability of amoebae to internalize cryptococcal cells compared to the absence of this molecule. Taken together, this points to impairment of the phagocytic process.

It was first sought to determine if 3-hydroxy C9:0 affected the functioning of NADPH oxidase-1, which could in turn alter the amount of oxidative radicals produced inside the amoebal food vacuoles (Figure 2). Here, no meaningful deductions could be made as there was no statistical significance (p > 0.05) between the non-treated amoebae and experimental conditions, i.e., 3-hydroxy C9:0-treated amoebae or GXM-treated amoebae. Therefore to compliment this experiment, it was then sought to determine if 3-hydroxy C9:0 could protect cells against the oxidative effects of hydrogen peroxide (when cells were directly challenged) (Figure 3). Previously, Zaragoza et al. (2008) showed that GXM protected cells against hydrogen peroxide. Our results confirm this finding. When comparing LMPE 101’s non-treated cells [an acapsular strain that also does not produce 3-hydroxy C9:0 (data not shown)] data to the corresponding hydrogen peroxide-treated cells’, GXM-treated cells’ and 3-hydroxy C9:0-treated cells’ data; significantly fewer CFUs were obtained under the treatment conditions. More importantly, this strain yielded significantly more CFUs (p < 0.05) when challenged with hydrogen peroxide in the presence of GXM or 3-hydroxy C9:0 compared to when this strain was challenged with hydrogen peroxide in the absence of the two test molecules. A similar response pattern was observed when examining the results obtained for the strain UOFS Y-1378, which has a capsule (Supplementary Figure S2). Also of interest, the levels of hydrogen peroxide in the medium did not dissipate over a 24-h period and remained more or less the same (Supplementary Figure S3). This suggests that the reported results specifically for hydrogen peroxide-treated cells and hydrogen peroxide-treated in the presence of GXM or 3-hydroxy C9:0, are not as a result of diminished levels of hydrogen peroxide in the test medium but rather speak to the protective quality of GXM and 3-hydroxy C9:0.

Amoebapore is an amoeba-specific digest peptide that creates pores in cell walls of targeted cells (Leippe et al., 1994). To test the effectiveness of amoebapore in killing cryptococcal cells, we considered its activity against four strains: (1) one that has a capsule and produces 3-hydroxy C9:0 (UOFS Y-1378), (2) two that have capsules but no 3-hydroxy C9:0 (R265 and LMPE 046), and (3) one without a capsule and 3-hydroxy C9:0 (LMPE 101) (Figure 4). Our data show that the strain UOFS Y-1378 was resistant to amoebapore at both test concentrations, and was rather stimulated by this protein. The same phenomenon was observed for the other two capsular strains (R265 and LMPE 046), although the level of growth stimulation was less than for UOFS Y-1378. However, the strain LMPE 101 (which lacks a capsule and 3-hydroxy fatty acids) was sensitive to amoebapore. Based on these results, it is reasonable to conclude that 3-hydroxy C9:0 in UOFS Y-1378 may have acted in concert with the capsule to shield cells hence resulted in the greatest level of resistance that were recorded. On the other hand, the absence of 3-hydroxy fatty acids (in applicable strains) led to a decrease in the levels of resistance. Logically, the strain without 3-hydroxy fatty acids and the capsule was the most susceptible. Nonetheless, an independent study is required to determine if addition of 3-hydroxy fatty acids would lead to strains showing resistance toward amoebapore.

Transmission electron microscopy was subsequently performed on UOFS Y-1378 and LMPE 101 in order to assess the effect of this amoebapore on the ultrastructure of cells. The cell wall of UOFS Y-1378 was without pores while that of LMPE 101 showed a nick in the cell wall (Figure 5). The latter could explain the resistance and susceptibility that was, respectively, expressed by UOFS Y-1378 and LMPE 101.