Immunosuppressive Compounds Affect the Fungal Growth and Viability of Defined Aspergillus Species

Immunosuppressive drugs are administered to a number of patients; e.g., to allogeneic hematopoietic stem cell transplant recipients. Immunosuppressive drugs impair the immune system and thus increase the risk of invasive fungal disease, but may exhibit antifungal activity at the same time. We investigated the impact of various concentrations of three commonly used immunosuppressive compounds—cyclosporin A (CsA), methylprednisolone (mPRED), and mycophenolic acid (MPA)—on the growth and viability of five clinically important Aspergillus species. Methods included disc diffusion, optical density of mycelium, and viability assays such as XTT. MPA and CsA had a species-specific and dose-dependent inhibitory effect on the growth of all Aspergillus spp. tested, although growth inhibition by MPA was highest in A. niger, A. flavus and A. brasiliensis. Both agents exhibited species-specific hyphal damage, which was higher when the immunosuppressants were added to growing conidia than to mycelium. In contrast, mPRED increased the growth of A. niger, but had no major impact on the growth and viability of any of the other Aspergillus species tested. Our findings may help to better understand the interaction of drugs with Aspergillus species and ultimately may have an impact on individualizing immunosuppressive therapy.


Introduction
Immunosuppressive drugs have to be administered in a number of clinical conditions; e.g., in patients with autoimmune diseases such as rheumatoid arthritis, psoriasis or autoimmune hemolytic anemia, in patients after renal, liver or heart transplantation to prevent graft rejection or in allogeneic hematopoietic stem cell transplant (HSCT) recipients in order to prevent or to treat graft-versus-host disease (GvHD). Depending on the indication, the compounds are administered at different dosages and schedules. Immunosuppressive drugs exhibit specific negative effects on the different arms of the immune system [1]. For example, glucocorticosteroids such as prednisolone or methylprednisolone (mPRED) not only suppress the phagocytic function of monocytes and neutrophils, but also impair antigen presentation, T cell function, and the expression of pro-inflammatory cytokines [2,3]. Mycophenolate mofetil (MMF), which is metabolized after administration to its active compound mycophenolic acid (MPA), impairs the recruitment of monocytes and lymphocytes into sites of inflammation and induces the apoptosis of activated T lymphocytes [4,5]. The calcineurin inhibitor cyclosporin A (CsA), another commonly used immunosuppressive agent, is a fungus-derived molecule and inhibits relatively selective T cell activation, whereas the compound has a minimal effect on phagocytic cells [6]. Although it is well described that the impairment of the host immune system by immunosuppressive drugs increases the risk of invasive fungal disease [7,8], there are data indicating that some immunosuppressive compounds exhibit activity against Aspergillus fumigatus, Candida albicans or Cryptococcus neoformans [9,10]. Aspergillus spp. is the most frequent cause of invasive fungal disease in patients with hematological malignancies or undergoing HSCT. As A. fumigatus, A. brasiliensis, A. terreus, A. flavus, and A. niger comprise up to 70% of Aspergillus species isolated in HSCT patients suffering from invasive fungal disease [11], and A. flavus and A. niger often have reduced susceptibility to amphotericin B, which may affect outcomes [12], we thought to evaluate the effect of three commonly used immunosuppressive compounds-mPRED, MPA, and CsA-on these distinct species of Aspergillus.

Immunosuppressive Agents Inhibit the Fungal Growth of Aspergillus Species-Disc Diffusion Assay for Screening
The disc diffusion assay was used as a screening method to determine whether any of the immunosuppressive compounds had an impact on the growth of the Aspergillus species included in the study. The assay revealed that MPA and CsA exhibited an inhibitory effect on the growth of all Aspergillus species tested, although the extent of inhibition differed between the species (Figure 1). Growth inhibition by MPA was considerably more pronounced for A. flavus strain 253 and A. brasiliensis compared to both strains of A. fumigatus, A. terreus and A. niger, respectively. Similarly, CsA inhibited the growth of A. niger, A. flavus and A. brasiliensis to a higher extent compared to A. fumigatus and A. terreus, respectively. In contrast, mPRED had a marginal inhibitory effect on the growth of A. niger, but no impact on the growth of any of the other Aspergillus species tested was observed. Growth inhibition by all the immunosuppressive agents was lower compared to that of posaconazole. Of note is that the immunosuppressive drugs used at very high concentrations inhibited the growth of the fungi on the agar plates, whereas no major effect in the disc assay was seen when the immunosuppressive compounds were tested in lower concentrations (data not shown).

Figure 1.
Effect of immunosuppressive compounds on the growth of Aspergillus spp. Three 6 mm diameter paper discs were impregnated with 35 µL of a 16 mg/mL methylprednisolone (mPRED) solution, a 50 mg/mL mycophenolic acid (MPA) solution, a 50 mg/mL cyclosporin A (CsA) solution, a 0.5 µg/mL posaconazole solution, or PBS, respectively, and placed onto inoculated agar plates. Shown are representative results of one test; the assay was performed three times for each fungus and for each immunosuppressive drug at each concentration investigated.

Immunosuppressive Agents Inhibit the Fungal Growth of Aspergillus Species-Quantitative Assessment of Fungal Cell Density
In order to further evaluate and quantify the observations made in the disc diffusion assay, we determined the impact of the immunosuppressive agents on the growth and germination of resting conidia of the different Aspergillus spp. in yeast nitrogen base (YNB) medium by the assessment of fungal cell density. In this assay, three different concentrations of the immunosuppressive compound were investigated, with a concentration reflecting common target serum levels, a significantly lower Figure 1. Effect of immunosuppressive compounds on the growth of Aspergillus spp. Three 6 mm diameter paper discs were impregnated with 35 µL of a 16 mg/mL methylprednisolone (mPRED) solution, a 50 mg/mL mycophenolic acid (MPA) solution, a 50 mg/mL cyclosporin A (CsA) solution, a 0.5 µg/mL posaconazole solution, or PBS, respectively, and placed onto inoculated agar plates. Shown are representative results of one test; the assay was performed three times for each fungus and for each immunosuppressive drug at each concentration investigated.

Immunosuppressive Agents Inhibit the Fungal Growth of Aspergillus Species-Quantitative Assessment of Fungal Cell Density
In order to further evaluate and quantify the observations made in the disc diffusion assay, we determined the impact of the immunosuppressive agents on the growth and germination of resting conidia of the different Aspergillus spp. in yeast nitrogen base (YNB) medium by the assessment of fungal cell density. In this assay, three different concentrations of the immunosuppressive compound were investigated, with a concentration reflecting common target serum levels, a significantly lower and a significantly higher concentration. Fungal cell density was assessed after 17 hours of incubation of resting conidia with the respective immunosuppressive drug. Compared to untreated controls, MPA at a concentration of 50 µg/mL significantly suppressed the formation of mycelium of A. fumigatus AF4215, A. brasiliensis, both A. flavus strains tested and A. niger strain BS for up to 50% (p < 0.05 each; Figure 2A,E,G-I), respectively, whereas A. fumigatus AF293, both A. terreus strains tested ( Figure 2C,D) and A. niger 715 ( Figure 2F) were not significantly affected. With the exception of A. flavus strain 253 ( Figure 2G), lower concentrations of MPA did not result in a significant reduction of the cell density of any of the species tested. Cyclosporin A inhibited the formation of mycelium of A. niger strain BS in a dose-dependent manner (mean ± SEM: 84.4% ± 3.8%, (not significant); 68.7% ± 9.1%, (not significant); 57.4% ± 8.4%, p = 0.037; Figure 2E), whereas no significant effect was seen on A. niger strain 715 ( Figure 2F). The highest concentration of CsA significantly reduced the cell density of A. brasiliensis (mean ± SEM: 79.0% ± 2.1%, p = 0.01; Figure 2I), whereas none of the other Aspergillus species tested was affected by the compound at any concentration. In contrast to MPA and CsA, mPRED increased the growth of A. niger strain BS (mean ± SEM: 104.9% ± 5.2%; ns, 137.9% ± 8.8%; p = 0.0191, 134.5% ± 10.2%; p = 0.0442; Figure 2E), but had no major impact on the growth of any of the other Aspergillus species analyzed.  Figure 2A,E,G-I), respectively, whereas A. fumigatus AF293, both A. terreus strains tested ( Figure 2C,D) and A. niger 715 ( Figure 2F) were not significantly affected. With the exception of A. flavus strain 253 ( Figure 2G), lower concentrations of MPA did not result in a significant reduction of the cell density of any of the species tested. Cyclosporin A inhibited the formation of mycelium of A. niger strain BS in a dose-dependent manner (mean ± SEM: 84.4% ± 3.8%, (not significant); 68.7% ± 9.1%, (not significant); 57.4% ± 8.4%, p = 0.037; Figure 2E), whereas no significant effect was seen on A. niger strain 715 ( Figure 2F). The highest concentration of CsA significantly reduced the cell density of A. brasiliensis (mean ± SEM: 79.0% ± 2.1%, p = 0.01; Figure 2I), whereas none of the other Aspergillus species tested was affected by the compound at any concentration. In contrast to MPA and CsA, mPRED increased the growth of A. niger strain BS (mean ± SEM: 104.9% ± 5.2%; ns, 137.9% ± 8.8%; p = 0.0191, 134.5% ± 10.2%; p = 0.0442; Figure 2E), but had no major impact on the growth of any of the other Aspergillus species analyzed.

Immunosuppressive Agents Damage Aspergillus Species
In addition to a reduction in the formation of mycelium observed in the disc diffusion assay and the assessment of fungal cell density, both MPA and CsA impaired the viability of developing Aspergillus hyphae in a dose-dependent manner ( Figure 3). In contrast, for mPRED, only negligible effects were observed. The lowest cell viability, equivalent to the highest damage of developing hyphae, was seen when MPA at the highest concentration was co-incubated with conidia of both strains of A. niger, both strains of A. flavus, and A. brasiliensis, respectively. Similar results were observed for the damage by CsA at a concentration of 750 ng/mL on developing hyphae of both strains of A. fumigatus, both strains of A. niger and A. brasiliensis, respectively ( Figure 3A

Immunosuppressive Agents Damage Aspergillus Species
In addition to a reduction in the formation of mycelium observed in the disc diffusion assay and the assessment of fungal cell density, both MPA and CsA impaired the viability of developing Aspergillus hyphae in a dose-dependent manner (Figure 3). In contrast, for mPRED, only negligible effects were observed. The lowest cell viability, equivalent to the highest damage of developing hyphae, was seen when MPA at the highest concentration was co-incubated with conidia of both strains of A. niger, both strains of A. flavus, and A. brasiliensis, respectively. Similar results were observed for the damage by CsA at a concentration of 750 ng/mL on developing hyphae of both strains of A. fumigatus, both strains of A. niger and A. brasiliensis, respectively ( Figure 3A  When comparing the effect of MPA and CsA co-incubated with Aspergillus conidia, less hyphal damage was observed when the immunosuppressive agents were added to Aspergillus mycelium for 6 hours. For example, the hyphal damage of A. niger BS by MPA and CsA was significantly lower When comparing the effect of MPA and CsA co-incubated with Aspergillus conidia, less hyphal damage was observed when the immunosuppressive agents were added to Aspergillus mycelium for 6 hours. For example, the hyphal damage of A. niger BS by MPA and CsA was significantly lower compared to the hyphal damage of the respective agent towards germinating Aspergillus niger conidia (mean ± SEM: 18.5% ± 6.2% versus 67.2% ± 10.0% for MPA at a dosage 50 µg/mL (p = 0.0144), and 23.6% ± 6.8% versus 71.4% ± 4.0% for CsA at a dosage of 750 ng/mL (p = 0.0038); Figure 4E). Similar results were seen for A. brasiliensis (mean ± SEM: 13.4% ± 3.6% versus 99.8% ± 0.03% for MPA at a dosage 50 µg/mL (p = 0.0005), and 5.7% ± 5.8% versus 82.3% ± 4.6% for CsA at a dosage of 750 ng/mL (p < 0.0001); Figure 4I). With the exception of A. niger, mPRED showed minimal to no damage towards all other Aspergillus species tested (Figure 4). In order to directly compare the fungal damage of resting conidia incubated with immunosuppressive compounds for 17 hours, we extended the incubation time of the mycelium with immunosuppressive compounds from 6 to 17 h. The extended incubation time did not result in an increased hyphal damage compared to 6 hours (Supplementary Figure S1). compared to the hyphal damage of the respective agent towards germinating Aspergillus niger conidia (mean ± SEM: 18.5% ± 6.2% versus 67.2% ± 10.0% for MPA at a dosage 50 µg/mL (p = 0.0144), and 23.6% ± 6.8% versus 71.4% ± 4.0% for CsA at a dosage of 750 ng/mL (p = 0.0038); Figure 4E). Similar results were seen for A. brasiliensis (mean ± SEM: 13.4% ± 3.6% versus 99.8% ± 0.03% for MPA at a dosage 50 µg/mL (p = 0.0005), and 5.7% ± 5.8% versus 82.3% ± 4.6% for CsA at a dosage of 750 ng/mL (p < 0.0001); Figure 4I). With the exception of A. niger, mPRED showed minimal to no damage towards all other Aspergillus species tested (Figure 4). In order to directly compare the fungal damage of resting conidia incubated with immunosuppressive compounds for 17 hours, we extended the incubation time of the mycelium with immunosuppressive compounds from 6 to 17 h. The extended incubation time did not result in an increased hyphal damage compared to 6 hours (Supplementary Figure S1).

Discussion
Our data demonstrate for the first time that various immunosuppressive compounds exhibit specific activities on the growth and viability of different species of Aspergillus, the most common cause of invasive fungal infection. Although an early microscopic analysis revealed that CsA delays

Discussion
Our data demonstrate for the first time that various immunosuppressive compounds exhibit specific activities on the growth and viability of different species of Aspergillus, the most common cause of invasive fungal infection. Although an early microscopic analysis revealed that CsA delays the filamentation of A. fumigatus [13], which has also been described for other fungi such as Neurospora crassa and Cryptococcus neoformans [14,15], data on the effect of CsA on other important Aspergillus species such as A. terreus or A. niger were lacking. Our data assessing fungal growth and viability show that CsA exhibits antifungal activity across all Aspergillus species tested, although the effect was most pronounced in A. niger and A. brasiliensis. In addition, higher concentrations of CsA mostly had a stronger effect than lower concentrations, although clinically relevant concentrations of CsA resulted in the damage of approximately 20% of A. fumigatus, A. niger, and A. brasiliensis, respectively. It has been shown previously that CsA inhibits the growth and hyphal elongation of A. fumigatus in concentrations of 625 ng/mL and higher, which are higher than the recommended dosages for clinical use [16,17]. The antifungal activity of calcineurin inhibitors such as CsA is not surprising, as it is known that Cryptococcus neoformans requires calcineurin for hyphal elongation during mating [14], and the calcineurin-mediated signal transduction pathway impacts the fungal cell membrane via the regulation of the biosynthesis of ergosterol, chitin and β-glucan [18]. Similar to CsA, we observed an effect of MPA on growth and damage on all Aspergillus species tested, with a species-dependent magnitude of the effect. Our data clearly show that, with the exception of A. terreus, hyphal damage by MPA was more pronounced when the immunosuppressants were added to growing conidia than to mycelium, suggesting that fully developed hyphae might be less susceptible to immunosuppressants than growing conidia. Although the antifungal mechanism of MPA has not been revealed to date, animal data demonstrate an anti-Pneumocystis jirovecii effect of MPA, which is supported by clinical association studies [19][20][21]. In contrast to CsA and MPA, mPRED increased the growth of A. niger, but did not have a major impact on the other Aspergillus species tested. An early study described a species-specific effect of hydrocortisone, which increased the growth rate of A. fumigatus and A. flavus, but did not affect A. niger and A. oryzae [22]. Whether corticosterone-binding proteins-which have been described for various Candida species [23]-are involved in the growth-promoting effect of hydrocortisone in A. fumigatus and A. flavus is unclear to date. The fact that no impact of corticosteroids on the growth or cell density of R. oryzae has been found supports the existence of fungus-specific differences in the effect of corticosteroids on fungal biology [24].
We observed only minor differences between the strains of specific species, although results regarding growth inhibition and viability within one species were comparable. We acknowledge the fact that we tested only a limited number of Aspergillus species and strains, and it might be possible that other strains may show different results. It is also important to note that we used concentrations above the pharmacologically relevant dosages for the disc assay, in which the conidia are exposed to different concentrations of the drug, whereas in the experiments assessing OD, all conidia are exposed to an identical concentration of the respective compound. However, no in vitro experiment can exactly reflect physiological conditions, and under physiological conditions, even significantly lower concentrations of the compound may have the same biological effect as very high concentrations in an in vitro assay.
Although our data demonstrate that immunosuppressive agents may inhibit the growth and decrease the viability of defined species of Aspergillus, it is important to note that the data clearly show that the antifungal activity of the agents do not outweigh their immunosuppressive effects. In this regard, daily injections of CsA could not rescue cyclophosphamide-immunosuppressed mice suffering from lethal pulmonary aspergillosis [16]. Similarly, corroborating a report on MMF [25], the administration of CsA was associated with a four-fold risk of invasive fungal infection within the first 6 months after kidney transplantation compared to corticosteroid and azathioprine therapy [26]. On the other hand, potent synergisms in the activity against a variety of fungi has been described for CsA when combined with azoles, which affected even azole resistant isolates [9,10,27,28]. In conclusion, our data show that 1) MPA and CsA exhibit a species-specific inhibitory effect on the growth Aspergillus, 2) higher fungal damage is observed when MPA and CsA are added to growing conidia than to mycelium, and 3) mPRED increases the growth of A. niger, but has no major impact on the growth and viability of the other Aspergillus species tested. Further studies including microscopic and phenotypic analyses are warranted to reveal the mechanisms of how immunosuppressive agents affect different species of Aspergillus, which might not only increase our understanding of the complex interaction of drugs with various Aspergillus species but also of the unique characteristics of each immunosuppressive agent (immunosuppression versus antifungal activity), which ultimately may have an impact to individualize immunosuppressive therapy.

Assessment of the Antifungal Activity of Immunosuppressive Agents
The antifungal activity of immunosuppressive agents was tested by assessing growth inhibition by means of disc diffusion assay and by determination of the cell density. The viability of the fungus was assessed by using the XTT assay.

Disc Diffusion Assay
For the assessment of growth inhibition by the immunosuppressive compounds, disc diffusion assays were performed on agar plates as described elsewhere with some modifications [9]. In brief, 200 µL of a suspension containing 5 × 10 6 conidia/mL of the fungal strain were spread uniformly onto Sabouraud glucose agar plates (BD Bioscience). Paper discs of 6 mm in diameter (Becton Dickinson) were impregnated with 35 µL of a 16 mg/mL mPRED solution, a 50 mg/mL MPA solution, a 50 mg/mL CsA solution or a 0.5 µg/mL posaconazole (MSD/Merck, Whitehouse Station, NJ, USA) solution, respectively, and placed onto the inoculated agar plates. Plates were then incubated at 37 • C and evaluated for the degree of growth inhibition after 24 h.

XTT Assay
The effect of the immunosuppressive compounds on the viability of Aspergillus hyphae was analyzed by means of the colorimetric XTT assay using (2,3-bis[2-methoxy-4-nitro-5sulphenyl]2Htetrazolium-5-carboxyanilide) sodium salt (XTT; Sigma, Steinheim, Germany) plus coenzyme Qo (2,3-dimethoxy-5methyl-1,4-benzoquinone; Sigma), as described previously [31]. Although the XTT assay assesses the metabolic activity of cells, the results of the XTT assay are often used as a measure of cell damage [32,33]. In brief, immunosuppressive compounds were added in different concentrations either directly to 2 × 10 4 dormand conidia (co-incubation for 17 hours) or to hyphae developed from 2 × 10 4 conidia of the respective Aspergillus spp. (co-incubation for 6 or 17 hours), respectively. Fungi incubated alone served as control. Thereafter, hyphae were washed once with sterile aqua dest. and were incubated in XTT solution (0.25 mg/mL) supplemented with coenzyme Q 0 (40 µg/mL) at 37 • C for another hour. The absorbance of the supernatant was assessed spectrophotometrically at 450 nm. Antifungal activity was calculated as follows: percentage of hyphal damage = (1−X/C) × 100, where X is the absorbance of experimental wells and C is the absorbance of control wells with hyphae only [32,33].

Assessment of Fungal Cell Density
In order to evaluate the effect of the immunosuppressive compounds on the fungal growth of Aspergillus, fungal cell density was assessed spectrophotometrically [34][35][36][37].

Statistical Analyses
Data were analyzed using GraphPad Prism (version 5.04; GraphPad Software, La Jolla, CA, USA). Each concentration of the respective immunosuppressant was compared with the control only and therefore Student´s t-test was used. A two-sided P value of less than 0.05 was considered to be statistically significant.

Conclusions
The present data show that 1) MPA and CsA exhibit a species-specific inhibitory effect on the growth of Aspergillus, 2) higher fungal damage is observed when MPA and CsA are added to growing conidia than to mycelium and 3) mPRED increases the growth of A. niger, but has no major impact on the growth and viability of the other Aspergillus species tested. This study might increase the understanding of the complex interaction of immunosuppressants with various Aspergillus species and of the unique characteristics of each immunosuppressive agent (immunosuppression versus antifungal activity), which ultimately may have an impact to individualize immunosuppressive therapy.