The OxrA Protein of Aspergillus fumigatus Is Required for the Oxidative Stress Response and Fungal Pathogenesis

ABSTRACT

An efficient reactive oxygen species (ROS) detoxification system is vital for the survival of the pathogenic fungus Aspergillus fumigatus within the host high-ROS environment of the host. Therefore, identifying and targeting factors essential for oxidative stress response is one approach to developing novel treatments for fungal infections. The oxidation resistance 1 (Oxr1) protein is essential for protection against oxidative stress in mammals, but its functions in pathogenic fungi remain unknown. The present study aimed to characterize the role of an Oxr1 homolog in A. fumigatus. The results indicated that the OxrA protein plays an important role in oxidative stress resistance by regulating the catalase function in A. fumigatus, and overexpression of catalase can rescue the phenotype associated with OxrA deficiency. Importantly, the deficiency of oxrA decreased the virulence of A. fumigatus and altered the host immune response. Using the Aspergillus-induced lung infection model, we demonstrated that the ΔoxrA mutant strain induced less tissue damage along with decreased levels of lactate dehydrogenase (LDH) and albumin release. Additionally, the ΔoxrA mutant caused inflammation at a lower degree, along with a markedly reduced influx of neutrophils to the lungs and a decreased secretion of cytokine usually associated with recruitment of neutrophils in mice. These results characterize the role of OxrA in A. fumigatus as a core regulator of oxidative stress resistance and fungal pathogenesis.

IMPORTANCE Knowledge of ROS detoxification in fungal pathogens is useful in the design of new antifungal drugs and could aid in the study of oxidative stress resistance mechanisms. In this study, we demonstrate that OxrA protein localizes to the mitochondria and functions to protect against oxidative damage. We demonstrate that OxrA contributes to oxidative stress resistance by regulating catalase function, and overexpression of catalase (CatA or CatB) can rescue the phenotype that is associated with OxrA deficiency. Remarkably, a loss of OxrA attenuated the fungal virulence in a mouse model of invasive pulmonary aspergillosis and altered the host immune response. Therefore, our finding indicates that inhibition of OxrA might be an effective approach for alleviating A. fumigatus infection. The present study is, to the best of our knowledge, a pioneer in reporting the vital role of Oxr1 protein in pathogenic fungi.

INTRODUCTION

Increasing numbers of invasive fungal infections are being reported recently, most of which are detected in hospitals (1). Typically, phagocytosis by the innate immune cells (including neutrophils and macrophages) is the first line of defense against pathogenic fungi in the healthy hosts (2). Inside the immune cells, the generation of the reactive oxygen species (ROS) via the respiratory burst process serves as the primary mechanism against microbial infections (2–4). The ROS produced within the phagosome damage key cellular components, such as nucleic acids, lipids, and proteins, inducing programmed cell death (2, 5–7). Therefore, an effective ROS detoxification system plays a key role in the survival of the pathogenic fungi in such adverse environments. However, the current understanding regarding the ROS detoxification system of pathogenic fungi is quite limited, and therefore, it is of great significance to study the ROS detoxification mechanisms in the pathogenic fungi to identify novel targets for antifungal drugs.

Since the intracellular ROS are produced mainly in the mitochondria (8), the synthesis of numerous proteins that can resist oxidative damage, such as superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPX), occurs within the fungal mitochondria (2, 9–12). SOD is responsible for converting the superoxide anion free radicals into hydrogen peroxide (H₂O₂), while CAT is involved in the detoxification of H₂O₂ into oxygen and water. In addition, the induction of these antioxidant proteins occurs in pathogenic fungi, such as Candida albicans, after exposure to neutrophils and macrophages, indicating that the defense mechanisms involving such antioxidants play an important role in the survival of C. albicans during the innate immune cell-mediated phagocytosis (13, 14). Indeed, mutations in these antioxidant proteins are reported to decrease fungal virulence (13, 15, 16). Therefore, ROS-scavenging proteins are increasingly being acknowledged as putative fungal virulence factors. Since numerous ROS-scavenging proteins are synthesized in the fungal mitochondria, identification and characterization of these ROS-scavenging proteins in this organelle will be crucial in understanding how pathogenic fungi avoid host ROS toxicity.

Oxidation resistance 1 (Oxr1) is a conserved gene family that only occurs in eukaryotes and mainly participates in the protection against deleterious ROS (17). The expression of endogenous Oxr1 protein is induced under oxidative stress conditions in human cells (18, 19). In many organisms, such as yeast (18, 20), worms (21), mammalian cells (21–23), mosquito (24), and silkworm (25), the knockout of oxr1 increases the sensitivity to oxidative stress, suggesting that Oxr1 is essential to defend against oxidative stress. Suppression of Oxr1 protein decreases transcriptional expression of some ROS detoxification enzymes, indicating that Oxr1 is a regulator of the ROS detoxification system (17, 22, 24, 26, 27). In conclusion, the antioxidant properties of Oxr1 have been confirmed in multiple animal and cell models. There are several reports that Oxr1 can maintain genome integrity (22). In human cells, the deletion of oxr1 increases H₂O₂-induced mitochondrial DNA damage (26). Ectopic expression of human Oxr1 is able to prevent DNA oxidative damage in the DNA repair-deficient Escherichia coli mutants (20). Taken together, Oxr1 can prevent the formation of oxidative DNA damage to protect nuclear and mitochondrial genome integrity (21, 28). Additionally, the roles of Oxr1 also include innate immune defense (27, 29), mitochondrial morphology maintenance (18, 23), and the regulation of aging (19, 21) and the cell cycle (30). Although the role of Oxr1 in many organisms has been studied extensively, there is little information regarding the role of this protein in fungi. It is well recognized that Oxr1 plays an important role in coping with oxidative stress, which implies that the Oxr1 protein might be a virulence factor in pathogenic fungi. However, to date, no homolog of Oxr1 has been identified in pathogenic fungi, including Aspergillus fumigatus and C. albicans. The genome-wide search for homologs revealed the presence of Oxr1 protein in most of the pathogenic fungi. Therefore, we explored the role of Oxr1 protein in oxidation stress response and fungal pathogenesis in the present study, with the findings highlighting the mechanisms of the oxidative stress response and fungal pathogenesis at the molecular level.

The present work was aimed at identifying and characterizing the Oxr1 homolog of Saccharomyces cerevisiae in A. fumigatus, the latter being a pathogenic filamentous fungus. The homolog was named OxrA. It was demonstrated that A. fumigatus OxrA shares a conserved TLDc domain and plays an essential role in the oxidative stress response. Therefore, oxrA deletion may render the fungal cells susceptible to H₂O₂-induced damage. Moreover, this phenotype was observed to be related to a modulation in catalase activity. Remarkably, a loss of oxrA attenuated the fungal virulence in a mouse model of invasive pulmonary aspergillosis. In addition, the oxrA mutant modulates activation of the immune system affecting cell influx into the lungs and production of inflammatory cytokine. Taken together, the present study revealed that OxrA is a core regulator of the oxidative stress response and fungal virulence in A. fumigatus.

RESULTS

In silico identification of S. cerevisiae Oxr1 homolog in A. fumigatus.

In order to identify the putative homolog of Saccharomyces cerevisiae Oxr1 in A. fumigatus, the BLASTP analysis of the amino acid sequences in the A. fumigatus database with S. cerevisiae Oxr1 as a query identified AFUB_040360 as the best hit (E value, 4e-43; identity, 37.1%). Moreover, the subsequent BLAST analysis against the S. cerevisiae genome database with the AFUB_040360 sequences as queries revealed Oxrl as the top hit. These findings suggested that AFUB_040360 sequences and Oxr1 are potential homologs. Thereafter, the AFUB_040360 protein was referred to as OxrA. The full-length sequencing revealed that OxrA contained 371 amino acid residues and displayed 37.1% sequence similarity to S. cerevisiae Oxr1. The SMART protein analysis revealed that the estimated topology of OxrA was the same as that of Oxr1, with a single putative TLDc domain starting and ending at positions 123 and 370, respectively, in the OxrA protein sequence. Previous studies have confirmed the TLDc domain as the protein motif with a high degree of conservation. The TLDc domain is present in certain mammalian proteins and is capable of resisting oxidative stress (31). In addition, similar to the S. cerevisiae Oxr1, OxrA also contains a cleavable N-terminal mitochondrial targeting sequence (MITOPROT; http://ihg.gsf.de/ihg/mitoprot.html) for targeting the inner mitochondrial membrane. A phylogenetic analysis was performed in the present study to compare the OxrA sequence with the sequences of homologs from several other fungi and mammals, and the results demonstrated the conservation of the OxrA amino acid sequence among the strains of Aspergillus spp. (Fig. 1A and B). However, this conservation in the sequence identity was lower than in the other fungal homologs and mammals and occurred mainly in the conserved TLDc domain (Fig. 1B and C). In the subsequent investigation of the role of OxrA in A. fumigatus, the entire coding region of the gene encoding this protein was deleted to generate the ΔoxrA mutant; the phenotypic characterization of the deletion mutant and that of a complemented strain are described below.

FIG 1.

Sequence analysis of Oxr1-encoded proteins in fungi. (A) Phylogenic tree of Oxr1 among selected eukaryotes. Amino acid sequences were aligned using ClustalW, and the phylogenetic tree was constructed via DNAman. (B) Amino acid sequence identity between Oxr1 of A. fumigatus 1163 and candidate homologs in selected eukaryotes. (C) Conservation of the TLDc domain among Oxr1 homologs.

OxrA localizes to mitochondria in A. fumigatus.

MitoProt was employed to predict the presence of the N-terminal mitochondrial targeting sequence in OxrA (32), and mitochondria were revealed as the probable location of OxrA. In order to verify the above finding, a strain containing green fluorescent protein (GFP)-labeled OxrA C terminus was constructed and analyzed. No significant difference was observed in growth between the OxrA::GFP strain and the corresponding parental strain in both minimal medium (MM) and complete medium (YAG) media (Fig. 2A). Moreover, there was no difference in the mRNA level of the constructed strain from that of the parental strain (Fig. 2B), indicating the full functioning of the OxrA-GFP fusion protein. At 18 h after germination when cultured in the MM medium, the microscopic observation of the OxrA protein revealed the predominant colocalization of the GFP-labeled OxrA with the mitochondrial marker MitoTracker (MitoTracker Red CMXRos; catalog no. M7512) (Fig. 2C). It has been reported that GFP-tagged MrsA was located in mitochondria in A. fumigatus (33). To further determine the mitochondrial localization of OxrA, we constructed two doubly labeled strains derived from OxrA-GFP strain by tagging the C terminus of MrsA (mitochondrial iron transporter) with red fluorescent protein (RFP). The dually labeled proteins were functional, and the strains expressing them showed normal growth phenotypes. Merged fluorescence microscopy studies indicated that MrsA-RFP mainly colocalized OxrA-GFP in mitochondria (Fig. 2D). Taken together, these findings confirmed the localization of OxrA in the mitochondria of A. fumigatus, which was consistent with the localization of Oxr1 in S. cerevisiae (18), suggesting that the OxrA protein possibly exerts a certain effect on A. fumigatus mitochondria. Since mitochondria function is required to initiate germination, we investigated the germination of ΔoxrA and the wild-type strain, respectively. Our results showed that the germination rate of ΔoxrA mutant is consistent with that of the wild-type strain, indicating that OxrA may not be required to initiate germination (Fig. S1 in the supplemental material).

FIG 2.

Subcellular localization of OxrA. (A) Colony morphologies of OxrA::GFP and the corresponding parental strains. The conidia were spotted on minimal medium (MM) and complete medium (YAG) media for 2 days at 37°C. (B) The relative expression of the oxrA gene was determined by quantitative PCR in OxrA::GFP and the corresponding parental strains. Gene expression was normalized to the endogenous reference gene tubA. Experiments were carried out in triplicate. Values are reported as the means ± standard deviation (SD). Statistical significance was calculated using the unpaired two-tailed t test (ns, not significant). (C) Localization pattern of OxrA proteins tagged by GFP. Colocalization of OxrA-GFP with a mitochondrial marker, as visualized by fluorescence microscopy; the merged yellow color shows colocalization. (D) Colocalization of OxrA-GFP with MrsA-RFP is visualized by fluorescence microscopy; the merged image (yellow) shows the high degree of colocalization. Scale bars, 10 μm.

OxrA is responsible for oxidative stress response in A. fumigatus.

In humans, the Oxr1 protein provides protection against oxidative stress and regulates mitochondrial function (22, 23). It is well recognized that mitochondria are involved in adaptation to oxidative stress (33, 34), such as exposure to H₂O₂. Therefore, the next step was to verify whether the OxrA protein also played an essential role against oxidative stress. The oxrA null mutant was constructed through homologous recombination by substituting the open reading frames (ORFs) of oxrA with the selected marker of Neurospora crassa pyr4. The constructed oxrA deletion strain was named SJX01 (ΔoxrA). Diagnostic PCRs confirmed the correct insertion of the pyr4 disruption cassette in the OxrA deletion mutant and the absence of the original oxrA ORF (Fig. S2). Thereafter, the susceptibility to H₂O₂ of the deletion mutant was analyzed and compared to its parental strain. Subsequently, the spores of the mutant were diluted 10-fold in the plates containing the YAG complete medium comprising 3-mM and 5-mM concentrations of H₂O₂, and the colony growth phenotypes were observed. The ΔoxrA strain was observed to be more susceptible to H₂O₂ treatment than the parental strain (Fig. 3A). Similar to the colony growth phenotypes detected in the plates, the ΔoxrA strain was observed to be hypersensitive to H₂O₂ relative to the parental strain in the aqueous medium containing 5 mM H₂O₂ (Fig. 3B). Moreover, when the corresponding native gene was complemented into the ΔoxrA mutant, the SJX02 ensuring strains (OxrA reconstituted [OxrA-recon]) showed H₂O₂ susceptibility phenotypes similar to that of the parental strain (Fig. 3A), indicating that the aberrant ΔoxrA H₂O₂ sensitivity phenotypes were specific to the loss of oxrA gene. Next, the effect of oxrA overexpression on the susceptibility to H₂O₂ treatment was investigated, and we constructed a strain placing OE::oxrA in the wild-type background. Our results showed that WT^{OE:: oxrA} exhibited a similar H₂O₂ susceptibility profile to that of the parental strain (Fig. S3). To examine whether OxrA was involved in responding to other oxidative stressors such as menadione, we examined and compared menadione susceptibilities of the ΔoxrA mutant and its parental strain. As a result, the ΔoxrA strain displayed similar susceptibility to menadione to the parental strain under the same culture conditions (Fig. S4), indicating that OxrA has a distinct role in the response of A. fumigatus to oxidative stress generated by menadione and H₂O₂. In order to examine the role of OxrA in response to the other stress conditions such as thermal or salt stress, relevant experiments were conducted which revealed no significant differences in response to thermal (42°C) and salt stress (0.8 M NaCl) between the constructed ΔoxrA mutant strain and the corresponding parental strain (data are not shown). Collectively, the above findings indicated that OxrA mostly played a vital part in the oxidative stress response. Next, it was explored whether the observed hypersensitivity of ΔoxrA to oxidative stress was due to an increased level of ROS. The 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) oxidant-sensing probe was employed to detect the ROS contents, which were metabolized by the ROS into a highly fluorescent state. As presented in Fig. 3C, the ROS production in ΔoxrA was significantly higher than in the parental strain and the complemented strain, suggesting that higher intercellular ROS level in the tΔoxrA mutant may account for the increased sensitivity toward oxidative stress. To further test the possible relationship between H₂O₂ hypersensitivity and the increased ROS level in the ΔoxrA strain, antioxidant l-ascorbic acid sodium (Vc) was added to the medium. As shown in Fig. 3D, compared to treatment with H₂O₂ only, l-ascorbic acid sodium (10 mM) almost completely restored the colony phenotype of ΔoxrA to that of the parental wild-type strain in the presence of H₂O₂. This suggests that the abnormal ROS level in the ΔoxrA mutant is strongly related to its phenotype of hypersensitivity to oxidative stress.

FIG 3.

Deletion of oxrA results in increased sensitivity to oxidative stress. (A) Comparison of H₂O₂ susceptibilities in oxrA null mutant, oxrA-reconstituted mutant, and the parental wild-type A1160^C (WT). Different strains were inoculated as a series of 3-μl 10-fold dilutions derived from a starting suspension of 10⁷ conidia per ml onto solid YAG with or without H₂O₂ and cultured at 37°C for 2 days. (B) For comparison of the susceptibilities of the indicated strains in liquid YAG supplemented with H₂O₂, amounts of 10⁸ conidia of ΔoxrA mutant and the WT were inoculated into 50 ml of liquid YAG supplemented with H₂O₂ and cultured at 37°C with shaking at 220 rpm for 2 days. The cultures were poured into a new petri dish to be photographed. (C) Reactive oxygen species (ROS) production of the parental wild-type strain, Δoxr1 mutant, and Δoxr1-recon. The ROS contents of Δoxr1 and Δoxr1-recon were normalized to that of the parental wild type. The experiment was performed thrice with biological triplicates. The data are presented as the means and standard deviations of three biological replicates. Statistical analysis was performed using an unpaired two-tailed t test (**, P < 0.01). (D) Serially diluted conidia of each strain were spotted onto YAG plates containing the ROS scavenger l-ascorbic acid sodium (Vc, 10 mM) and H₂O₂ (5 mM). The plates were incubated at 37°C for 2 days.

OxrA contributes to oxidative stress resistance by regulating catalase function.

While OxrA is known to provide protection against oxidative stress in A. fumigatus, the underlying mechanism remains unclear to date. Our results revealed that the increased ROS contents in the ΔoxrA mutant might account for the increased sensitivity toward oxidative stress. Therefore, it was expected that the function of the ROS-scavenging proteins might decrease in the ΔoxrA mutant. In eukaryotes, cells have evolved numerous mechanisms to scavenge ROS. For example, antioxidant enzymes such as superoxide dismutase (SOD) and catalase (CAT) can scavenge ROS (35). SOD can convert superoxide anion into H₂O₂, and CAT can detoxify H₂O₂ into water and oxygen. Therefore, it was hypothesized that OxrA modulates these ROS-scavenging enzymes in A. fumigatus. In order to determine the effect of OxrA on the activity of these ROS detoxification enzymes, the activity of SOD and CAT was evaluated in the protein extracts isolated from the ΔoxrA strain, the parental wild-type strain, and the corresponding complemented strain, and they revealed that the total SOD activity in the ΔoxrA mutant was similar to that in the parental strain. However, the total CAT activity was decreased significantly in the protein extracts isolated from the ΔoxrA culture (Fig. 4A), and it was inferred that this reduced activity of the CAT antioxidant enzyme was the possible reason for the enhanced sensitivity to oxidative stress. The next question that arose was whether increasing the CAT activity in an overexpression experiment could rescue the phenotype associated with oxrA deficiency. In order to answer this, the full-length ORF of catA/B controlled by the gpdA promoter was transformed into oxrA deletion strain, yielding the strains SJX03 (ΔoxrA^OE::catA) and SJX04 (ΔoxrA^OE::catB). The results of the analysis revealed that in the ΔoxrA^OE::catB strain, the sensitivity to oxidative stress could be restored to levels similar to those in the corresponding parental wild-type strain, while the ΔOxrA^OE::catA strain demonstrated a partial restoration of the oxidative stress sensitivity of the ΔoxrA strain (Fig. 4B). Next, we further measured catalase and SOD activity in the ΔoxrA^OE::catA and ΔoxrA^OE::catB mutants. Our results showed that the total CAT activity in the ΔoxrA^OE::catA and ΔoxrA^OE::catB mutants was slightly higher than that of the parental strain (Fig. 4A). However, the total SOD activity in ΔoxrA, as well as ΔoxrA^OE::catA and ΔoxrA^OE::catB mutants, was similar to that in the parental strain (Fig. 4A). Additionally, the level of ROS in the ΔoxrA^OE::catA and ΔoxrA^OE::catB mutants could also be restored to levels similar to that in the parental strain (Fig. 4C). Moreover, catA or catB overexpression in the parental wild-type strain did not considerably enhance the oxidative stress sensitivity under the tested conditions (Fig. 4B). Altogether, these data indicate that OxrA may be required for the normal function of catalases during the regulation of ROS detoxification and that multiple copies of CAT genes could suppress the defect of OxrA during ROS detoxification. To further explore the molecular mechanism of OxrA regulating CAT function, we initially analyzed and compared the transcription levels of catA/B genes by reverse transcription-quantitative PCR (RT-PCR), using tubulin as a loading control. Our results showed that the transcriptional level of catA/B genes was almost normal in the oxrA deletion mutant compared to the level in the parental strain (Fig. S5). Next, we analyzed the expression level of CatA/B proteins by Western blotting. Our results showed that the protein expression level of CatB was also normal in the oxrA deletion mutant compared to the level in the parental strain (Fig. 4D). Interestingly, the molecular weight of CatB in the ΔoxrA mutant was lower than that in the parental strain (Fig. 4D). In humans, oxidation resistance 1 (Oxr1) can regulate posttranslational modifications of potent antioxidant enzymes (36). Thus, we speculate that the posttranslational modifications of CatB might be one of the reasons for the decrease in the molecular weight of the ΔoxrA mutant. Unfortunately, we were unable to detect the CatA protein band by Western blotting in this study. Taken together, these data suggest that OxrA might contribute to the regulation of the activity of catalases by regulating posttranslational modifications of CatB. Hence, it is necessary to further study and confirm our hypothesis.

FIG 4.

OxrA contributes to oxidative stress resistance by regulating catalase function. (A) Total SOD and CAT activities quantified from the protein extracts of the YAG cultures. Conidia of different strains were inoculated in liquid YAG and shaken at 37°C for 48 h, and then the proteins were extracted for enzymatic assays. The experiment was performed thrice with biological triplicates. The data are presented as the means and standard deviations of three biological replicates. Statistical analysis was performed using an unpaired two-tailed t test (**, P < 0.01). (B) Susceptibility comparison of different strains on YAG plates supplemented with H₂O₂ as described in the legend to Fig. 3. (C) Reactive oxygen species (ROS) production of the indicated strains. The ROS contents of the indicated strains were normalized to that of the parental wild type. The experiment was performed thrice with biological triplicates. The data are presented as the means and standard deviations of three biological replicates. Statistical analysis was performed using an unpaired two-tailed t test (**, P < 0.01). (D) Western blotting shows the protein expression of CatB in the ΔoxrA and the parental wild-type strains with or without the addition of 5 mM H₂O₂, and the protein expression level of CatB was also normal in the oxrA deletion mutant compared with the level in the parental strain. However, the molecular weight of CatB in ΔoxrA mutant is smaller than that in the parental strain.

To further characterize the relationship between OxrA and catalases, we constructed ΔcatA, ΔcatB, ΔoxrA ΔcatA, and ΔoxrA ΔcatB mutants, and then examined and compared the oxidative stress susceptibilities of these mutants. Unlike the ΔoxrA mutant, which was supersensitive to H₂O₂, the ΔcatA mutant exhibited slightly increased susceptibility to H₂O₂ compared to the susceptibility of the parental strain. However, the ΔcatB mutant showed susceptibility profiles similar to that of the parental strain (Fig. 5). Additionally, the ΔoxrA ΔcatA mutant was more susceptible to H₂O₂ than the ΔoxrA mutant, suggesting that OxrA and CatA might cooperate to cope with H₂O₂ stress. To determine the relationship between OxrA and the SOD antioxidant enzyme, as well as glutathione peroxidase (GPX), the strains SJX05 (ΔoxrA^OE::sodA), SJX06 (ΔoxrA^OE::sodB), and SJX07 (ΔoxrA^OE::gpxA) were constructed through the transformation of full-length sodA, sodB, and gpxA, respectively, controlled by the gpdA promoter into the oxrA deletion strain. The results demonstrated that the three constructed strains (SJX05, SJX06, and SJX07) presented H₂O₂ susceptibility profiles similar to that of the ΔoxrA strain (Fig. S6). Moreover, the total CAT activity and the level of ROS in ΔoxrA^OE::sodA/B and ΔoxrA^OE::gpxA were also similar to that in the ΔoxrA mutant. In summary, these results suggested that OxrA contributes to oxidative stress resistance only by promoting CAT function and not the function of the SOD and GPX enzymes.

FIG 5.

OxrA and CatA cooperate to cope with H₂O₂ stress. Comparison of the H₂O₂ susceptibilities of the parental wild-type A1160^C (WT), ΔcatA, ΔcatB, ΔoxrA ΔcatA, and ΔoxrA ΔcatB mutants. Different strains were inoculated as a series of 3-μl 10-fold dilutions derived from a starting suspension of 10⁷ conidia per ml on to solid YAG with or without 5 mM H₂O₂ and cultured at 37°C for 2 days, showing that ΔoxrA ΔcatA mutant was more susceptible to H₂O₂ than the ΔoxrA mutant.

OxrA deficiency decreased the virulence of A. fumigatus and altered the host immune response.

In the case of infection, ROS are generated in macrophages via the NADPH oxidase route (37), which is an efficient approach to kill pathogenic A. fumigatus (38, 39). Therefore, an effective ROS detoxification system plays an important role in A. fumigatus survival in such adverse environments. As stated earlier, the present report describes the involvement of OxrA in the oxidative stress resistance demonstrated by A. fumigatus. In this context, a mouse model was used to test the virulence of the parental wild-type strain, ΔoxrA strain, and the corresponding complemented strain. Conidia were inoculated into each group of mice, while the mice in the control group were inoculated with saline solution. The mice subjected to ΔoxrA mutant infection had a markedly decreased virulence in comparison to the mice subjected to infection with the parental wild-type strain, as revealed by the Kaplan-Meier log-rank analysis (P < 0.01) (Fig. 6A). After 6 days, no mouse in the parental wild-type strain infection group survived, which was significantly different from the survival rate of 100% observed with the mice infected with the ΔoxrA mutant. Next, quantitative RT-PCR (qRT-PCR) analysis was conducted on the 3rd and 4th days to examine the pulmonary fungal burdens. The mice that received ΔoxrA mutant inoculation had markedly lower pulmonary fungal burden than those inoculated with the parental wild-type strain (Fig. 6B). To confirm whether the dead mice were infected with A. fumigatus, histopathological examination of lung sections was carried out. Histopathological examinations by the Grocott's methenamine silver nitrate staining revealed that lung tissue from the parental wild-type-infected mice and ΔoxrA-recon-infected mice displayed aggressive fungal growth, whereas lungs from ΔoxrA-infected mice displayed fewer and shorter hyphae, indicating that the host immune system could inhibit the growth of mycelia (Fig. 6C). Moreover, the albumin and lactate dehydrogenase (LDH) levels in the bronchoalveolar lavage fluid (BALF) were measured to analyze the degree of tissue injury caused by the above three strains, respectively. The mice inoculated with the ΔoxrA mutant exhibited significantly decreased levels of LDH and albumin release on day 3 and day 4 postinoculation compared to the mice inoculated with the parental wild-type strain (Fig. 6D). Taken together, inoculation with the ΔoxrA mutant resulted in decreased fungal burden along with mitigated tissue injury, indicating that OxrA had a vital role in the pathogenic mechanism of invasive pulmonary aspergillosis.

FIG 6.

The ΔoxrA mutant is attenuated for virulence in mice in the cyclophosphamide and corticosteroid mouse model. (A) The cyclophosphamide and corticosteroid mouse model was used for virulence analysis. The mice were immunosuppressed as described in Materials and Methods, infected, and observed for mortality for 14 days. PBS was used for the mock infection control group. The mice infected by strain ΔoxrA displayed attenuated virulence compared to those infected by the parental wild-type A1160^C (WT) as determined by Kaplan-Meier log-rank analysis (P < 0.01). (B) Decreased fungal burden in ΔoxrA-inoculated mice. To assess fungal burden in lungs, the cyclophosphamide and corticosteroid model was utilized as described in Materials and Methods. Mice were sacrificed on days 3 and 4 postinoculation. Fungal burden in the lungs was determined by quantitative real-time PCR based on the 18S rRNA gene of A. fumigatus. Data are presented as total fungal genomic DNA normalized to input DNA. The mean and standard error are presented (n = 3 mice for the control group and n = 5 mice for inoculation groups). Statistical analysis was performed using analysis of variance (ANOVA). Statistical significance was accepted at a P value of <0.01. (C) Histopathological analyses for related strains-infected lung tissues conducted using Grocott s methenamine silver nitrate (GMS) staining. Scale bar, 50 μm. (D) Decreased levels of tissue damage in ΔoxrA-inoculated mice. LDH (lactate dehydrogenase) and albumin release in bronchoalveolar (BAL) fluids in cyclophosphamide and corticosteroid immunosuppressed mice were determined on days 3 and 4 postinfection in the WT-, ΔoxrA-, and ΔoxrA-recon inoculated mice. Data are presented as the means and standard deviations of three biological replicates (*, P < 0.05).

In order to better understand the mechanism underlying the lower virulence of the ΔoxrA mutant, the cell inflammatory profiles in the airways were determined. First, the cellular infiltrates in the BAL fluids were analyzed using differential cell counts. The differential cell counts in BALF at day 3 after inoculation suggested that macrophages, neutrophils in particular, significantly decreased in the ΔoxrA mutant-inoculated mice (Fig. 7A), which indicated a decreased inflammatory response in the mice infected with the ΔoxrA mutant. In order to quantify the different immune responses to the ΔoxrA mutant, the cytokine contents related to the recruitment of neutrophils in mice (such as MIP-2 and KC) were evaluated. It was revealed that the expression of cytokine KC was significantly decreased in the mice inoculated with the ΔoxrA mutant compared to the mice inoculated with the parental wild-type strain. However, the protein levels of cytokines MIP-2 were slightly elevated compared to BALF in the mice inoculated with the wild-type strain (Fig. 7B). Collectively, these findings indicated that, in the process of A. fumigatus infection, OxrA possibly serves as a vital signal transduction protein in A. fumigatus and is involved in the physiological processes of this fungus and that a loss of OxrA affects the infection outcomes due to the modulation of the innate immune response.

FIG 7.

OxrA deficiency alters the host immune response. (A) Characterization of cellular infiltrates in bronchoalveolar (BAL) fluids. Differential cell counts demonstrate that macrophage numbers and particularly neutrophil numbers were significantly decreased in mice inoculated with ΔoxrA mutant. Results are presented as mean and standard error of n = 5 mice. (**, P < 0.01). The experiment was repeated in duplicate with similar results. (B) Cytokine production in response to A. fumigatus, the parental wild-type A1160^C (WT), and ΔoxrA inoculation. KC and MIP-2 concentrations were determined in BAL fluids on day 3 postinoculation. Significant differences between the parental wild-type A1160^C and ΔoxrA inoculation groups could be observed for KC and MIP-2 protein levels. Results are the mean and standard error of n = 5 mice. (*, P < 0.05).

DISCUSSION

In the event of microbial infection, the host cells generate ROS to fight the invading microbes, such as pathogenic bacteria, viruses, and fungi (2, 40–44). Therefore, an effective ROS detoxification system plays a key role in the survival of the pathogen under such adverse environments. The present study proposes that OxrA plays a vital role in fungal pathogenic mechanism and ROS detoxification, which lays a certain molecular foundation for understanding the biological basis of oxidative stress response in pathogenic fungi. The present study is, to the best of our knowledge, a pioneer in reporting the vital role of Oxr1 protein in pathogenic fungi.

Intracellular ROS are produced mainly in the mitochondria (8), and therefore, numerous proteins involved in the processes of compensating for the adverse effects of ROS, including SOD, CAT, and GPX enzymes, are located in the mitochondria. Despite the seeming overabundance of these ROS detoxification functions in A. fumigatus, the ΔoxrA mutant remains sensitive to oxidative stress (Fig. 3A), which indicates that it plays a vital role in ROS scavenging. It is reported that the Oxr1 protein in humans and yeasts belongs to the highly conserved eukaryotic protein family involved in oxidative stress resistance (18). In A. fumigatus, the OxrA protein is located in the mitochondria and is involved in oxidative stress resistance (Fig. 2 and 3), which once again confirms that the location and function of the Oxr1 protein are conservative in eukaryotes.

Oxr1 is reported to be involved in oxidative stress resistance in several eukaryotes (18, 24), although the precise mechanism underlying such protection is unclear, as the mechanisms appear to be different in different organisms. For instance, in zebrafish, Oxr1 regulates the expression of multiple antioxidant genes (i.e., gpx1b, gpx4a, gpx7, and sod3a) involved in the detoxification of cellular ROS (17), while in Anopheles gambiae, Oxr1 regulates the levels of GPX and CAT, the enzymes related to H₂O₂ detoxification (24). The results of the present study confirmed that OxrA serves as an antioxidant regulator in A. fumigatus and revealed the molecular mechanism of the OxrA-mediating oxidative stress resistance. As depicted in Fig. 4A, OxrA regulates the activity of catalase, a protective enzyme associated with H₂O₂ detoxification, which illuminates the mechanism of OxrA in A. fumigatus in a novel manner and corroborates the previous finding that the reduced activity of catalase antioxidant enzyme might be the reason for the increased sensitivity toward H₂O₂ in the ΔoxrA mutant. However, it was also revealed that OxrA was not involved in regulating the expression of catA/B genes encoding catalase (Fig. S5 in the supplemental material) and was not involved in regulating the expression of CatA/B proteins (Fig. 4D). However, OxrA contributes to regulation of the molecular weight of CatB, indicating that OxrA may regulate posttranslational modifications of CatB. Taken together, OxrA may contribute to regulation of catalase activity by regulating posttranslational modifications of CatB. Of course, it is necessary to further study and confirm our hypothesis. Furthermore, it was observed that, unlike zebrafish and A. gambiae, OxrA was not involved in activating the transcription of glutathione peroxidase (Gpx) and SOD antioxidant enzyme genes (Fig. S5). Moreover, the overexpression of GpxA, SodA, and SodB resulted in H₂O₂ sensitivity comparable to that of the ΔoxrA strain (Fig. S6), indicating that OxrA might only be regulating the catalase activity to affect the susceptibility of oxidative stress in A. fumigatus. Although there is evidence that Oxr1 provides protection against oxidative stress by regulating catalase activity (as discussed above), Oxr1 itself is reported to resist oxidative damage. A previous bioinformatics analysis confirmed that OxrA protein possesses a C-terminal TLDc domain with a high degree of conservation (Fig. 1). Typically, the C-terminal TLDc domain of mouse/human Oxr1 is reported to resist oxidative stress by directly interacting with H₂O₂ via a conserved cysteine residue (22). Another study reported that, in addition to the TLDc domain, the N-terminal domain in human Oxr1 plays an important role in oxidative stress resistance (21). Therefore, Oxr1 itself is capable of preventing oxidative damage in mammals. In view of the structural conservation of the Oxr1 protein in eukaryotes, it was hypothesized that the A. fumigatus OxrA may itself also prevent oxidative damage. Therefore, to reveal how the OxrA protein participates in the oxidative stress resistance in A. fumigatus, further biochemical and molecular biological studies illustrating the structure and functions of OxrA are necessary.

The ROS generated in the alveolar macrophages play an important role in the control over A. fumigatus conidia (39). An in vitro study examining the function of neutrophils revealed that H₂O₂ efficiently kills the A. fumigatus hyphae and that the addition of commercial catalase could protect the neutrophils from damaging fungal hyphae (45, 46). Furthermore, neutrophils from human patients with chronic granulomatous disease (CGD) caused by mutations in the NADPH oxidase complex, where the phagocytes are unable to produce ROS, were reported to be incapable of killing the A. fumigatus hyphae (38). Collectively, the above findings indicate that the host-generated ROS play an important role in killing A. fumigatus in vivo. Therefore, the ROS-scavenging proteins in A. fumigatus could be important virulence factors. The findings of the present study suggested that the ROS-scavenging capacity is defective in the ΔoxrA mutant and the OxrA deficiency significantly decreased the virulence of A. fumigatus, indicating that the OxrA protein might be an important virulence factor (Fig. 6). Indeed, the ΔoxrA mutant could not replicate in vitro or grow under oxidative stress conditions that mimic the intracellular environment within the host, suggesting that the markedly reduced ROS-scavenging ability of the ΔoxrA mutant fails to facilitate survival inside the phagosome with a high oxidative stress environment, causing the virulence of the ΔoxrA mutant within a mouse model (Fig. 6). The ΔoxrA strain is incapable of growing under the intracellular environment of the host, and, therefore, the host may mount an efficient immune response to eliminate the infection. Consistent with our working hypothesis, the fungal load in ΔoxrA strain-infected mice was markedly decreased compared to that in the parental strain-infected mice (Fig. 6B). Moreover, the albumin and LDH assays revealed that the level of tissue damage in the mice infected with ΔoxrA mutant was also reduced relative to the parental strain-infected mice (Fig. 6D). As depicted in the inflammatory profile in Fig. 7, infection with the OxrA-lacking A. fumigatus strain affected the recruitment of host cells and the generation of cytokines. Nonetheless, further investigation is warranted to elucidate the precise mechanisms underlying the regulation of OxrA in the A. fumigatus response within the host cells.

One of the most important experimental factors with aspergillosis in the mouse model is the immune status of the mice. Typically, the immunosuppressive regimens used in mouse models for invasive aspergillosis (IA) rely on cyclophosphamide and/or cortisone. Since the inhibitory effects of corticosteroids on the antifungal activities of phagocytes may be less efficient in vivo, corticosteroid- and cyclophosphamide-treated (CCT) mice are often used for virulence analysis of A. fumigatus mutants. Corticosteroid-treated (CT) or CCT models are selected for virulence analysis depending on the observed phenotype of the fungal mutant. CCT models are often used for virulence analysis of oxidative stress mutants (9, 47–49). Thus, in this study, we chose CCT models to analyze virulence of the oxrA mutant. However, cyclophosphamide can suppress neutrophils’ oxidative burst and reduce ROS production and release (50). Thus, if the mice are immunosuppressed with cyclophosphamide and corticosteroid in this study, it is unlikely the virulence defect of the oxrA mutant is due to the oxidative stress phenotype. Moreover, it has been reported that catalases are not essential for the virulence of pathogenic fungi. For example, in A. fumigatus, although the conidial catalase CatA can protect the spores against the deleterious effects of H₂O₂ in vitro, it does not play a role in protecting conidia against the oxidative burst of macrophages (51, 52). In another pathogenic fungus, C. albicans, the ectopic expression of catalase enhances resistance to oxidative stress, and Δcat1 cells are more sensitive to neutrophil killing. However, catalase inactivation did not attenuate C. albicans virulence (53). Thus, although Oxr1 participates in H₂O₂ resistance by regulating catalase activity, the reason for the virulence defect of ΔoxrA mutant may not be due to the oxidative stress phenotype. In this study, in order to confirm whether the oxidative stress-sensitive phenotype of ΔoxrA mutant is the main reason for the decrease of virulence, we tested this directly by using the ΔoxrA^OE::catA strain, which exhibits a similar oxidative stress-sensitive phenotype to wild type. Our result showed that the ΔoxrA^OE::catA strain had decreased virulence (Fig. S7), suggesting that the decreased virulence for ΔoxrA mutant may be caused by other reasons rather than oxidative stress-sensitive phenotype. Moreover, there are an increasing number of studies suggesting an apparent lack of importance of ROS-scavenging mechanisms to control A. fumigatus virulence. Such data came out of the analysis of other A. fumigatus mutants, such as Δcat mutants lacking either conidial or mycelial catalases and Δsod and Δyap1 mutants, where the hypersensitivity to H₂O₂ observed in vitro is not correlated with a reduction of fungal virulence in an experimental model of aspergillosis. Thus, non-ROS-scavenging mechanisms of OxrA might play a major role in the virulence of A. fumigatus. As the cell wall and virulence phenotypes share a striking correlation and a number of cell wall mutants have been found that are reduced in virulence, we tested the susceptibility of ΔoxrA mutant to wall-perturbing agents. Our results showed that ΔoxrA mutant exhibited slightly increased susceptibility to wall-perturbing agents SDS, calcofluor white (CFW), and Congo red (CR) (Fig. S8). Thus, we speculate that the phenotypes that correlate best with the decrease in virulence for ΔoxrA mutant are their cell wall defects. One explanation for the reduced virulence in some cell wall mutants is increased “unmasking” of β-1,3-glucan and greater immune response from the host. It is possible that this may be the case for the ΔoxrA mutant, but this remains to be tested. Alternatively, the ΔoxrA mutant may have defects in the cell wall that somehow compromise its ability to grow in the host.

ROS production is a central element of the host immune response, and a severe hereditary defect such as CGD represents a high risk for IA. Although non-ROS-scavenging mechanisms of OxrA play a major role in the virulence of A. fumigatus in the corticosteroid- and cyclophosphamide-treated mice model, this kind of immune suppression mimics neutropenia, i.e., results in the absence of any relevant functional immune cells. Moreover, cyclophosphamide also can suppress neutrophil oxidative burst and reduce ROS production and release if the mice are neutropenic, as is the case with mice immunosuppressed with corticosteroid and cyclophosphamide. OxrA may not be as important for A. fumigatus to survive inside the host, and therefore, in order to further investigate the role of OxrA during infection, we further analyzed virulence of ΔoxrA mutant in a nonneutropenic (cortisone model) mouse model. In this study, we found that the ΔoxrA mutant is significantly less virulent than the wild type (Fig. S9), indicating that the fungal ROS hypersensitivity contributed to reduced virulence. This conclusion is consistent with ROS-mediated damage being attributed to the killing capacity of innate immune cells. Therefore, it will be interesting to further elucidate the role of OxrA in host-pathogen interactions. Future research may consider targeting the OxrA activity for therapeutic purposes.

MATERIALS AND METHODS

Strains and culture conditions.

All strains used in this study are described in Table 1. A. fumigatus strain 1160 was purchased from the Fungal Genetics Stock Center (FGSC) and was used to generate ΔoxrA null mutant strain. The media used in this study included YAG (2% glucose, 0.5% yeast extract, and trace elements), MM (1% glucose, trace elements, and salts), and 100 ml trace elements (2.20 g ZnSO₄·7H₂O, 1.10 g H₃BO₃, 0.50 g MnCl₂·4H₂O, 0.16 g FeSO₄·7H₂O, 0.16 g CoCl₂·5H₂O, 0.16 g CuSO₄·5H₂O, 0.11 g (NH₄)₆Mo₇O₂₄·4H₂O, and 5.00 g Na₄EDTA). A. fumigatus strain A1160 was cultured on YUU (0.5% yeast extract, 2% dextrose, trace minerals,1.2% uracil, and 1.1% uridine) at 37°C. Conidia were harvested and collected from plates with 48 h of incubation at 37°C. Next, conidia were diluted and counted in a Neubauer chamber.

TABLE 1.

Aspergillus fumigatus strains used in this study

Strain	Genotype	Source
A1160	ΔKU80 pyrG1	FGSC
A1160C	ΔKU80 pyrG, pyr4	This study
SJX01	ΔKU80 pyrG1, ΔoxrA::pyr4	This study
SJX02	ΔKU80 pyrG1, ΔoxrA::pyr4; oxrA (p)::oxrA::hph	This study
SJX03	ΔKU80 pyrG1, ΔoxrA::pyr4; gpd (p)::catA::hph	This study
SJX04	ΔKU80 pyrG1, ΔoxrA::pyr4; gpd (p)::catB::hph	This study
SJX05	ΔKU80 pyrG1, ΔoxrA::pyr4; gpd (p)::sodA::hph	This study
SJX06	ΔKU80 pyrG1, ΔoxrA::pyr4; gpd (p)::sodB::hph	This study
SJX07	ΔKU80 pyrG1, ΔoxrA::pyr4; gpd (p)::gpxA::hph	This study

Construction of the A. fumigatus strains.

Fusion PCR was used to construct the oxrA knockout cassette as previously described. In brief, approximately 1-kb sections of regions flanking the oxrA gene were amplified using the primers P1/P3 and P4/P6. The selection marker pyr4 from the plasmid pAL5 was amplified with the primers Pyr4 F/R. Next, the three PCR products were used as the template to generate the oxrA deletion cassette using the primers P2/P5 and then transformed into the parental A. fumigatus strain A1160 as previously described. Transformants were verified by diagnostic PCR using the primers P8/P9, P1/P7, and P6/P10, respectively.

To complement the ΔoxrA strain, the cassette containing the oxrA gene plus the two 1.5-kb flanking regions was PCR amplified using primers P11/P12. This fragment was subsequently cloned into the pAN7-1 plasmid, which contains the hygromycin B resistance gene hph, to generate the oxrA complementation plasmid. The plasmid was then transformed into the oxrA deletion strain, and transformants were selected on YAG medium supplemented with 200 μg/ml hygromycin. The primers used in this study are shown in Table 2.

TABLE 2.

Primers used in this study

Primer name	Sequence(s) (5′–3′)
P1	ACCGAGAAGGAGCGCGGGAG
P2	TGTCTGACGGACCAGCGGTC
P3	GGCTTGTCTGCTCCCGCAGCGATGGGAAGAGATTCA
P4	ACGCCAGGGTTTTCCCGTTTGAGGAAAGGAAC
P5	AGCCCATTTCGTTCACTCCG
P6	AGCCTCGCCGTTTGGGTC
P7	GGCCGATCCTCCGGCCGAGG
P8	AATGAACGCCAGAAACAACC
P9	GGACGATGAGCACGTAGCC
P10	GGATTGGGAAAGTTGAGAGG
P11	AGCGCCACTCAAGTCCGTA CAGCATACATTACATATTTT
P12	GTGTTGAGTCTCAGGAGA
Pyr4 F	CGGAGCAGACAAGCC
Pyr4 R	GGGAAAACCCTGGCGT

To overexpress OxrA in the parental strain, a 1.1-kb DNA fragment, including the oxrA coding sequence, was PCR amplified and cloned into the plasmid pBARGPE, which contains a gpd promoter. Strain SJX03 (WT^OE::oxrA) was generated by transforming the plasmid into the parental strain. The transformants were screened on MM containing 200 μg/ml of hygromycin B. The same method was used to generate strains SJX04 (ΔoxrA^OE::catA), SJX05 (ΔoxrA^OE::catB), SJX06 (ΔoxrA^OE::sodA), SJX07 (ΔoxrA^OE::sodB), and SJX08 (ΔoxrA^OE::gpxA).

Since strain A1160 harbors a nonfunctional pyrG gene and cannot be used in animal infection models as a wild type, the Neurospora crassa pyr4 gene was complemented into the strain A1160. To construct the strain A1160 harboring a functional pyr4 gene, the pyr4 gene was amplified from the pAL5 plasmid using the primer pair Pyr4F/4R. The PCR product was cloned into the pEASY-Blunt Zero cloning kit (TransGen Biotech) and used to transform the recipient strain A1160, yielding the strain A1160^C, which harbors a functional pyr4 gene.

Light microscopy.

To visualize localization of OxrA-GFP, conidia were incubated in 2 ml of liquid MM on coverslips at 37°C for 12 h. After that, the medium was removed from the culture dish and washed by phosphate-buffered saline (PBS) at least three times, and then the prewarmed (37°C) staining solution containing MitoTracker Red CMXRos probe (working concentration, 25 nM) was added and incubated for 5 min under growth conditions. After straining was complete, the staining solution was removed and washed with PBS, and cells were observed using a fluorescence microscope. All images were captured using the Axio Imager A1 fluorescence microscope (Carl Zeiss, Jena, Germany).

Measurement of reactive oxygen species.

We incubated 10⁷ spores in 100 ml YAG media at 37°C for 18 h with shaking at 220 rpm. Then, 20 μM 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA; Invitrogen) was added to the medium and incubated at 37°C for 1 h. After that, the mycelia were harvested and washed three times with the distilled water to remove extracellular H₂DCFDA. The filtered mycelia were then ground in liquid nitrogen and suspended in PBS. The resulting supernatant was collected by centrifugation at 15,000 × g and 4°C for 10 min. Fluorescence was measured using a SpectraMax M2 reader (Molecular Devices, USA), with an excitation wavelength of 504 nm and an emission wavelength of 524 nm. The fluorescence intensity was normalized to the protein concentration of the sample, which was measured using a Bio-Rad protein assay kit.

RNA extraction and RT-PCR.

Total RNA from the spores cultured in liquid YAG at 37°C and 200 rpm for 48 h was extracted using the TRIzol reagent (Invitrogen). One hundred milligrams of mycelia per sample was used as the starting material for the determination of total RNA. cDNA synthesis was performed with 1.5 μg of RNA using HiScript Q RT SuperMix (Vazyme; catalog no. R123-01), and then cDNA was used for the real-time analysis. Real-time PCRs were performed in triplicates, and the expression levels of all genes of interest were normalized to β-tubulin levels. The threshold cycle (ΔΔCT) method of analysis was used to determine fold changes of gene expression in the ΔOxrA mutant relative to the wild-type 1160 strain.

Enzyme activity assay.

Briefly, 1 × 10⁶ conidia of ΔoxrA, the parental wild type, and the corresponding complemented strains were inoculated in liquid YAG and shaken at 37°C for 48 h. The mycelia were harvested and ground in liquid nitrogen and suspended in ice-cold extraction buffer (50 mM HEPES [pH 7.4], 137 mM KCl, 10% glycerol, 1 mM EDTA, 1 μg/ml pepstatin A, 1 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride [PMSF]). After centrifugation, the supernatants were used for enzymatic assays. Protein concentrations were determined by using a Bradford protein assay kit (Sangon Biotech, China). Catalase activity and superoxide dismutase activity were quantified using commercial assay kits (Beyotime Biotechnology, China).

Virulence assay.

The immunocompromised mouse model for invasive pulmonary aspergillosis was described previously (54). Briefly, the virulence of the A. fumigatus strains was tested in two immunologically distinct murine models of invasive pulmonary aspergillosis. For the corticosteroid model, white female ICR mice (6 to 8 weeks old, 22 to 25 g) were immunosuppressed with a single dose of hydrocortisone acetate injected subcutaneously (s.c.) at 40 mg/kg 1 day prior to inoculation. For the cyclophosphamide and corticosteroid model, mice were given intraperitoneal injections of cyclophosphamide (150 mg/kg of body weight) on days 3 and 1 relative to infection and a subcutaneous injection of hydrocortisone acetate (40 mg/kg of body weight) on day 1. Bacterial infections were prevented by adding 2 g/liter neomycin to the drinking water. Infection inoculum was prepared by growing the A. fumigatus isolates on YAG agar plates at 37°C for 3 days. Conidia were harvested by washing the plate surface with sterile phosphate-buffered saline-0.01% Tween 80. The resultant conidial suspension was adjusted to the desired concentration by hemacytometer count. On day 0, mice were anesthetized with pentobarbital sodium. For survival studies, 10 mice were intranasally infected with 1 × 10⁸ conidia of the A. fumigatus wild-type or ΔoxrA strain in 40 μl of sterile PBS. Cyclophosphamide (75 mg/kg) was injected every 3 days to maintain immunosuppression. The mortality was monitored during 14 days in total after inoculation. Differences in survival between experimental groups were compared using the log-rank test. Statistical analysis of survival was performed using Kaplan-Meier log-rank analysis.

To assess fungal burden in lungs, the cyclophosphamide and corticosteroid model was utilized as described above. Mice were sacrificed on days 3 and 4 postinoculation, and lungs were harvested and immediately frozen in liquid nitrogen. Samples were freeze-dried and homogenized with glass beads on a Mini-Beadbeater (BioSpec Products, Inc., Bartlesville, OK, USA), and DNA was extracted with the E.N.Z.A. fungal DNA kit (Omega BioTek, Norcross, GA, USA).

For histopathology, the cyclophosphamide and corticosteroid model was utilized as described above, and mice were sacrificed on day 3 postinoculation. When mice were sacrificed, lungs were removed on that day. Lung tissue was fixed in 10% phosphate-buffered formalin. Pathological lung tissues were strained by Grocott's methenamine silver nitrate by using standard histological techniques. A total of 3 mice were examined.

For evaluation of pulmonary infiltrate, the cyclophosphamide and corticosteroid model was utilized as described above; after 3 days of incubation, 5 infected mice were euthanized with a solution of 180 mg/kg of body weight of ketamine and 24 mg/kg of xylazine. Subsequently, bronchoalveolar lavage (BAL) fluid was harvested by washing the lungs twice with 2 ml of PBS. Fluid was centrifuged, and cell pellets were used for differential cell counts. Supernatants were used for cytokine quantification.

For determination of lactate dehydrogenase (LDH) and albumin levels in BAL fluids, the cyclophosphamide and corticosteroid model was utilized as described above, and mice were sacrificed on days 3 and 4 postinoculation. In vivo lung tissue damage was determined by measurement of LDH and albumin levels in mouse BAL fluid samples by using an LDH assay (CytoTox 96 nonradioactive cytotoxicity assay, Promega, Madison, WI, USA) and an albumin assay (albumin [BCG] reagent set; Eagle Diagnostics, Cedar Hill, TX, USA) according to the manufacturers’ instructions.