The Transcriptional Regulator HbxA Governs Development, Secondary Metabolism, and Virulence in Aspergillus fumigatus

The number of immunodepressed individuals is increasing, mainly due to the greater life expectancy in immunodepressed patients due to improvements in modern medical treatments. However, this population group is highly susceptible to invasive aspergillosis. This devastating illness, mainly caused by the fungus Aspergillus fumigatus, is associated with mortality rates reaching 90%. Treatment options for this disease are currently limited, and a better understanding of A. fumigatus genetic regulatory mechanisms is paramount for the design of new strategies to prevent or combat this infection. Our work provides new insight into the regulation of the development, metabolism, and virulence of this important opportunistic pathogen. The transcriptional regulatory gene hbxA has a profound effect on A. fumigatus biology, governing multiple aspects of conidial development. This is relevant since conidia are the main source of inoculum in Aspergillus infections. Importantly, hbxA also regulates the biosynthesis of secondary metabolites and the pathogenicity of this fungus.

ABSTRACT

Aspergillus fumigatus is the leading cause of invasive aspergillosis, which in immunocompromised patients results in a mortality rate as high as 90%. Earlier studies showed that HbxA is a global regulator in Aspergillus flavus affecting morphological development and secondary metabolism. Here, we determined its role in A. fumigatus, examining whether HbxA influences the regulation of asexual development, natural product biosynthesis, and virulence of this fungus. Our analysis demonstrated that removal of the hbxA gene caused a near-complete loss of conidial production in the mutant strain, as well as a slight reduction in colony growth. Other aspects of asexual development are affected, such as size and germination of conidia. Furthermore, we showed that in A. fumigatus, the loss of hbxA decreased the expression of the brlA central regulatory pathway involved in asexual development, as well as the expression of the “fluffy” genes flbB, flbD, and fluG. HbxA was also found to regulate secondary metabolism, affecting the biosynthesis of multiple natural products, including fumigaclavines, fumiquinazolines, and chaetominine. In addition, using a neutropenic mouse infection model, hbxA was found to negatively impact the virulence of A. fumigatus.

IMPORTANCE The number of immunodepressed individuals is increasing, mainly due to the greater life expectancy in immunodepressed patients due to improvements in modern medical treatments. However, this population group is highly susceptible to invasive aspergillosis. This devastating illness, mainly caused by the fungus Aspergillus fumigatus, is associated with mortality rates reaching 90%. Treatment options for this disease are currently limited, and a better understanding of A. fumigatus genetic regulatory mechanisms is paramount for the design of new strategies to prevent or combat this infection. Our work provides new insight into the regulation of the development, metabolism, and virulence of this important opportunistic pathogen. The transcriptional regulatory gene hbxA has a profound effect on A. fumigatus biology, governing multiple aspects of conidial development. This is relevant since conidia are the main source of inoculum in Aspergillus infections. Importantly, hbxA also regulates the biosynthesis of secondary metabolites and the pathogenicity of this fungus.

INTRODUCTION

The opportunistic human pathogen Aspergillus fumigatus is a known cause of a wide range of illnesses, including invasive aspergillosis (IA). Immunodeficient individuals are particularly susceptive to these diseases (1, 2). This population group includes organ transplant patients, individuals with genetic immunodeficiencies or receiving chemotherapy, and HIV patients (3–8). The main site of entry leading to A. fumigatus infections is the respiratory tract. Healthy individuals are able to eliminate the inhaled fungal conidia (asexual spores) through mucociliary clearance. The remaining fungal spores encounter epithelial cells or alveolar macrophages responsible for phagocytosis and killing of spores and for the initiation of a proinflammatory response that recruits neutrophils. Neutrophils are able to destroy hyphae from germinated conidia that evaded macrophages. However, patients who are neutropenic are at high risk of developing IA, with a mortality rate of up to 90% (9–13).

The small size of the A. fumigatus conidia contributes to its pathogenicity, reaching the lung alveoli and establishing an infection that can become systemic (2). In addition, A. fumigatus produces numerous secondary metabolites (14–21) that are considered part of the fungal chemical arsenal required for niche specialization (22), including host-fungus interactions. Some of these metabolites act as immunosuppressants, which may be in association with pathogenic processes. For most IA infections, early diagnosis is critical, followed by treatment with antifungal drugs such as azoles, typically, voriconazole (23, 24). Recently, strains of A. fumigatus have been shown to gain drug resistance; thus, it is necessary to identify new targets to control or prevent the potentially lethal infections caused by A. fumigatus (24).

Fungal regulatory genes could constitute some of these novel genetic targets to design antifungal therapies. In this work, we focused on studying the homeobox (Hbx) transcriptional regulator gene hbxA. Hbx proteins are a class of transcriptional regulators that govern development in many eukaryotes, including animals, plants, and other fungi (25). A homolog of this gene in Aspergillus flavus has been shown to regulate aspects of morphological differentiation, including asexual development (26). A recent hbx1-dependent transcriptome analysis revealed that this gene controls numerous genes involved in morphogenesis and secondary metabolism (27). We hypothesize that the homolog of hbx1 in A. fumigatus, hbxA, may play a similar role in the regulation of development and secondary metabolism in this important opportunistic human pathogen.

Our study revealed that in A. fumigatus, a lack of hbxA leads to a slight reduction in colony growth and a near-complete loss of conidial production. Furthermore, we showed that loss of hbxA decreased the expression of genes in the brlA central developmental pathway, as well as the expression of the “fluffy” genes flbB, flbD, and fluG. Other aspects of asexual development were affected in the absence of hbxA, such as the size and germination rate of conidia. In addition, metabolomics analysis of the hbxA mutant indicated that this gene is a master regulator that governs the biosynthesis of numerous natural products. With respect to the effect of hbxA on pathogenicity, infection of neutropenic mice showed an increase in virulence in the hbxA deletion mutant compared to the A. fumigatus wild-type strain.

RESULTS

hbxA is required for normal fungal growth and asexual development in A. fumigatus.

To determine the role of hbxA in A. fumigatus, two strains were generated, an hbxA deletion strain and a complementation strain (see Fig. S1A and B in the supplemental material). The deletion of hbxA was confirmed by diagnostic PCR, yielding the expected 2.8-kb PCR product. The complementation strain was also verified by PCR, producing a 2.7-kb DNA fragment. In addition, the strains were confirmed by assessing hbxA expression levels with quantitative reverse transcription-PCR (qRT-PCR) (Fig. S1C). hbxA transcripts were absent in the ΔhbxA mutant strain, while the complementation strain showed restoration of hbxA expression similar to that of the wild type. Next, the growth of all the strains was evaluated in point-inoculated cultures growing on solid glucose minimal medium (GMM) for 7 days. Our results indicated an approximately 1.4-fold reduction in the growth of the ΔhbxA mutant compared to the controls (Fig. 1).

FIG 1.

hbxA is required for normal colony development. CEA10 wild-type (WT), ΔhbxA mutant, and complementation (Com) strains were point inoculated on solid GMM and allowed to grow in the dark at 37°C. (A) Images of the colonies after 5 days of growth. (B) Quantification of colony growth measured as the colony diameter. Error bars are used to indicate standard error. Different letters on the columns indicate values that are statistically different (P < 0.05), as determined by ANOVA with the Tukey test comparison. This experiment was performed three times with three biological replicates in each experiment.

The absence of hbxA resulted in an almost aconidial strain. Specifically, a significant reduction of approximately 100-fold in conidial production with respect to the wild type was observed (Fig. 1A and 2A). Complementation of the hbxA deletion mutant with the hbxA wild-type allele restored normal conidiation. To gain insight into the hbxA mechanism that controls sporulation in A. fumigatus, we examined the expression of the genes in the conidiation central regulatory pathway, brlA, abaA, and wetA (28). Deletion of hbxA resulted in a significant reduction in the expression of all three genes compared to the wild type (Fig. 2B to D).

FIG 2.

Asexual development is strongly regulated by hbxA in A. fumigatus. Wild-type (WT), ΔhbxA mutant, and complementation (Com) strains were grown in liquid GMM stationary cultures for 72 h. (A to G) Cores from the mycelial mats were collected at 48 h and 72 h, and conidia were quantified (A). RNA was also extracted from mycelia to perform qRT-PCR. Relative gene expression levels of brlA (B), abaA (C), wetA (D), flbB (E), flbD (F), and fluG (G) are shown. All values were normalized to the wild-type 48-h samples. Error bars indicate the standard error. Different letters on the columns indicate values that are statistically different (P < 0.05), as determined by ANOVA with the Tukey test comparison. This experiment was performed two times with three biological replicates in each experiment. For qRT-PCR, technical replicates were also run for each biological replicate.

Additionally, the expression of the fluffy genes flbB, flbD, and fluG was also downregulated in the absence of hbxA (Fig. 2E to G). Interestingly, microscopic observations indicated that conidial size in the ΔhbxA mutant was significantly larger than those formed by the wild type (Fig. 3).

FIG 3.

The size of conidia is influenced by hbxA. Wild-type (WT), ΔhbxA mutant, and complementation (Com) strains were grown on GMM, and spores were collected after 72 h. Samples were observed under a microscope, and the spore diameter was measured. (A) Micrographs of conidia. (B) Measurements of spore diameter from 40 spores of each strain. Error bars indicate the standard error. Different letters on the columns indicate values that are statistically different (P < 0.05), as determined by ANOVA with the Tukey test comparison. This experiment was performed three times with three biological replicates in each experiment.

The absence of hbxA affects germination rates of A. fumigatus conidia.

To further examine the role of hbxA on conidia in A. fumigatus, an assay was performed to evaluate germination rates. As early as 2 h postinoculation, spores from the hbxA mutant started to produce germ tubes, while spores from wild-type or complementation strain remained ungerminated (Fig. 4A). Only after 6 h did the wild-type spores begin to germinate, but at that point, more than 70% of the ΔhbxA mutant spores were already germinated (Fig. 4B).

FIG 4.

The absence of hbxA results in earlier spore germination in A. fumigatus. Liquid GMM cultures of wild-type (WT), ΔhbxA mutant, and complementation (Com) strains were inoculated with 10⁶ spores/ml. Every 2 h postinoculation, an aliquot of 500 μl of culture was collected to observe conidial germination. (A) Micrographs of germinating conidia at different time points. (B) Percentage of germinated conidia. (C) Amount of biomass produced by each strain after 24 h under liquid shaking conditions. Error bars are used to indicate standard error. Different letters on the columns indicate values that are statistically different (P < 0.05), as determined by ANOVA with the Tukey test comparison. This experiment was performed four times with three biological replicates in each experiment.

Based on the rapid germination rate observed in the hbxA deletion mutant compared to that of the control strains, biomass yields were also evaluated (Fig. 4C). Cultures were grown for 24 h and then lyophilized to determine the dry weight of the fungal biomass. Our results indicated that the ΔhbxA mutant strain generated almost 2-fold the amount of biomass in comparison to the wild-type and complementation strains (Fig. 4C).

Deletion of hbxA causes an increase in sensitivity to the cell wall stressor SDS.

Since the absence of hbxA resulted in increased spore size and precocious germination, it is possible that concomitant hbxA-dependent alterations in cell wall integrity could occur. To test this hypothesis, the strains were grown on GMM containing different concentrations of SDS. While all strains demonstrated colony growth reduction when exposed to all SDS concentrations tested, the mutant exhibited greater sensitivity to that compound than did the control and was unable to grow at a concentration of 0.015%, a condition that still allowed the growth of wild-type and complementation colonies (Fig. S2).

Exposure to a high concentration of sucrose partially rescues the hbxA conidiation defect.

Osmotic stress has been shown to affect development of filamentous fungi, for example, increasing conidiation (29, 30). We examine whether hbxA still influences conidiation when exposed to osmotic stress. For most of the osmotic stressors tested, no significant difference was noticed in any of the strains except when the ΔhbxA mutant was grown in the presence of 1 M sucrose (Fig. S3). Under this condition, the ΔhbxA mutant strain produced more conidia than when grown on GMM alone.

Secondary metabolism in A. fumigatus is regulated by hbxA.

In A. flavus, hbx1 is a regulator of the production of numerous secondary metabolites, including aflatoxin, aflatrem, and cyclopiazonic acid (26). Similarly, we observed a broad regulatory scope of hbxA on secondary metabolism in A. fumigatus (Fig. 5). Our results revealed that the production of fumigaclavines, fumiquinazolines, and chaetominine was detected in the ΔhbxA mutant but at a significantly lower levels than in either the wild-type or complementation strain.

FIG 5.

hbxA regulates the production of multiple secondary metabolites in A. fumigatus. Spores of A. fumigatus wild-type (WT), deletion (ΔhbxA mutant), and complementation (Com) strains were inoculated in liquid yeast extract-glucose-supplements (YES) medium. Cultures were incubated at 37°C, and supernatant was collected after 7 days for secondary metabolite extraction and analysis. (A to C) Analysis of ergot alkaloids fumigaclavine A, B, and C. (D and E) Analysis of fumiquinazoline A/B (D) and C/D (E). (F) Analysis of chaetominine. Error bars are used to indicate standard error. Different letters on the columns indicate values that are statistically different (P < 0.05), as determined by ANOVA with the Tukey test comparison. This experiment was performed two times with four biological replicates in each experiment.

Deletion of hbxA increases virulence of A. fumigatus in a murine model.

Due to the fact that hbxA plays an important role in regulating A. fumigatus development as well as its metabolome, we hypothesized that hbxA could also be relevant in virulence. To investigate this possibility, we used a neutropenic murine infection model. As shown in Fig. 6, the ΔhbxA mutant strain was significantly more virulent than were the controls.

FIG 6.

hbxA negatively regulates virulence in a mouse model. Six-week-old mice were rendered neutropenic by the administration of cyclophosphamide and Kenalog-10 treatments. Fifty mice divided into 5 separate groups were used (i.e, each group contained 10 mice). Mice were infected with 2 × 10⁶ conidia/mouse of A. fumigatus wild-type (WT), deletion (ΔhbxA mutant), and complementation (Com) strains and monitored daily for a total of 7 days. Two controls that did not received fungal spores were included in this analysis, with a group that was rendered neutropenic (uninfected) and another group not treated with cyclophosphamide or Kenalog-10 (untreated). Statistical analysis of survival was carried out by a Kaplan-Meyer pairwise comparison using a log rank test. d, days.

DISCUSSION

In humans, the primary route of A. fumigatus infections is through the inhalation of airborne conidia that can eventually germinate in the lungs of a host. An investigation of the genes that influence fungal development, as well as other aspects of A. fumigatus biology, could provide interesting targets to develop treatments against this opportunistic human pathogen. In the phylogenetically close and agriculturally important fungus A. flavus, the transcription regulatory gene hbx1, a homolog of hbxA, was found to regulate several aspects of morphological differentiation, including asexual development, as well as the synthesis of several secondary metabolites (26). Homologs of hbxA have also been identified in other fungi beyond the Aspergillus genus, for example, in Magnaporthe oryzae and in species of the genus Fusarium. In these species, while no connection of the possible role of hbx1 or hbxA homologs with virulence or secondary metabolism was reported, the studies indicated a role in asexual development (31–33). Our analysis of A. fumigatus hbxA demonstrated that this gene has indeed a conserved role in the regulation of conidial production. Additionally, other aspects of conidial formation and function were influenced by hbxA in this fungus. In the absence of hbxA, conidia appear enlarged and present a high germination rate.

The mechanism of action of hbxA on conidiation includes the activation of the central regulatory pathway brlA, abaA, and wetA genes (28). The expression of these three genes in this signaling pathway is significantly reduced in the absence of hbxA. This decrease in expression could lead to the observed reduction in conidial production in the hbxA mutant. In addition, examination of the effect of hbxA on the expression of other genetic regulatory elements upstream of the brlA central regulatory pathway showed that fluG, a well-known developmental regulator (34), is also hbxA dependent. In addition, the expression of flbB and flbD was also found to be positively regulated by hbxA. FlbB has been previously shown to promote asexual development in A. fumigatus (35). FlbD has not been characterized in A. fumigatus (28); however, in A. nidulans, flbD has been shown to also promote conidiation (36). It is possible that HbxA could regulate the expression of these fluffy developmental genes directly, which could consequently affect the brlA pathway. In addition, since flbD expression is dependent on the FlbB-FlbE protein complex (35, 36), it is possible that hbxA could affect the expression of flbD indirectly by controlling flbB transcription. Interestingly, the presence of highly concentrated sugars, particularly sucrose and fructose, resulted in a significant increase in conidial production in the hbxA deletion mutant. Currently, the mechanism that triggers conidiation under these conditions is unknown.

In A. fumigatus and other fungi, development and secondary metabolism are genetically linked (22, 37, 38). As in A. flavus (26), in our study, we show that hbxA regulates not only development in A. fumigatus but also the production of multiple metabolites. Analysis of the hbxA-dependent metabolome revealed four different classes of compounds whose synthesis is under the influence of this regulator. Among them are ergot alkaloids (fumigaclavine) and fumiquinazoline. These compounds present bioactive properties; fumigaclavine has been shown to affect the nervous systems of the host and induce apoptosis, while fumiquinazoline presents cytotoxicity that inhibits neutrophils and aids A. fumigatus during infection (17, 39). Fumigaclavines and fumiquinazolines accumulate in asexual structures, and their production is linked to brlA expression (16, 17, 20, 21). Similarly, chaetominine is absent in the aconidial ΔbrlA mutant strain (misidentified as tryptoquivaline F) (20). It is possible that the hbxA regulatory role in the expression of genes involved in the synthesis of these compounds is indirect and mediated, at least in part, by its effect on brlA. In addition, the synthesis of chaetominine, a metabolite that is being tested to combat leukemia cells (40), is also controlled by hbxA. At the moment, it is unknown how this compound is produced or regulated in A. fumigatus. Chaetominine is structurally similar to fumiquinazolines; however, no evidence showing a link between their respective biosynthesis has been yet established. The present study reports a gene in A. fumigatus that controls the production of chaetominine.

Due to the fact that hbxA is important in A. fumigatus morphogenesis and secondary metabolism, we investigated whether this gene is relevant in virulence. To test this possibility, we used the neutropenic mouse infection model. Surprisingly, the lack of hbxA did not attenuate the virulence of this fungus; on the contrary, it enhanced it, resulting in higher mortality rates in the group of animals infected by the deletion strain than in the group inoculated with the wild-type strain. While the production of secondary metabolites was hampered in the mutant, and the conidial size was larger, these traits did not reduce virulence in the mutant strain. It is possible that the premature germination observed in the hbxA mutant could have accounted for the increase in virulence, allowing the deletion mutant to rapidly establish itself in the host. This could lead to an accelerated and enhanced fungal infection in neutropenic mice compared to that infected with the wild type. The increased amounts of biomass yield produced by the ΔhbxA mutant strain also suggest a greater capacity to colonize the host than with the wild type. In addition, the detected increased sensitivity to SDS in the mutant could reflect premature changes in cell wall composition during the early spore germination process which do not appear to be detrimental to infection.

In conclusion, we have established that the homeobox gene hbxA has a broad impact on the biology of A. fumigatus, affecting its development, secondary metabolism, and virulence. Specifically, we have shown that hbxA is necessary for normal asexual development, governing the expression of the fluffy genes fluG, flbB, and flbD, as well as the expression of those genes in the central developmental signaling pathway, brlA, abaA, and wetA. With respect to the role of hbxA on the A. fumigatus metabolome, our study revealed that the production of several fungal alkaloids, as well as chaetominine, is under its control. We also established that hbxA negatively affects pathogenicity, as a lack of hbxA increases virulence, possibly through the advantage of early germination and greater biomass production. For this reason, an hbxA loss-of-function strategy would not the suitable against A. fumigatus infection. Future studies will focus on elucidating whether forced overexpression of this regulator would have the opposite effect on virulence and on the production of beneficial secondary metabolites such as chaetominine. Additional studies will also provide further insight into the identification of additional hbxA-dependent genetic elements involved in the regulation of conidiation, triggering the germination and synthesis of fungal natural products.

MATERIALS AND METHODS

Culture conditions.

All strains used in this work are listed in Table 1. Strains were grown on glucose minimal medium (GMM) at pH 6.5 in the dark at 37°C, unless otherwise indicated. Agar at a concentration of 1% was used for solid cultures. Stocks of each strain were maintained at –80°C in 30% glycerol. The spore inoculum was generated on GMM plus 1 M sucrose and washed with sterile water before use.

TABLE 1.

List of strains used in this study

Strain	Genotype	Source
CEA10	Wild type	Gift from Robert Cramer
CEA17	pyrG1	Gift from Robert Cramer
TTRS8	pyrG1 ΔhbxA::pyrGA.parasiticus	This study
TTRS12	pyrG1 ΔhbxA::pyrGA.parasiticus hbxA::ptrAA.oryzae	This study

Strain construction.

(i) Generation of the ΔhbxA mutant strain. To obtain the hbxA deletion strain, an hbxA deletion cassette was first generated by fusion PCR, as described by Szewczyk et al. (41). The Afum_hbxA_P1 and Afum_hbxA_P2 primers were used to PCR amplify the 5′ untranslated region (UTR) of the hbxA locus in the A. fumigatus genome, while the Afum_hbxA_P3 and Afum_hbxA_P4 primers were used to amplify the 3′ UTR fragment. The middle fragment containing the selection marker was PCR amplified from plasmid pPG28 (42) using primers Afum_hbxA_P5 and Afum_hbxA_P6. The marker used was pyrG from Aspergillus parasiticus. The three fragments were then fused by PCR using primers Afum_hbxA_P7 and Afum_hbxA_P8. All primers utilized in this study are listed in Table 2. The fused PCR product was transformed into A. fumigatus CEA10 (pyrG⁻ ptrA⁻) by a polyethylene glycol-mediated transformation, as previously described (Cary et al. [26]). Transformants were selected on half-strength potato dextrose agar (PDA) without uracil. Potassium chloride (0.6 M) was used as an osmotic stabilizer in the regeneration medium. Transformants were confirmed by diagnostic PCR with primers Afum_hbxA_P0 and Apara_pyrG_R. A selected hbxA deletion transformant, TTRS8, was used in this study.

TABLE 2.

Primers used in this study

Primer	Sequence
Afum_hbxA_P0	GTGGTGACAGTGGTGGTTTCCC
Afum_hbxA_P1	GGTGCTTAGTTCCCTGGATGGACA
Afum_hbxA_P2	AAGGGACGGCGGAGAAGAAG
Afum_hbxA_P3	GTAGTGGTAGGCAGGAGGCATG
Afum_hbxA_P4	TGCTTGTAGTCACCGATCACCATTCC
Afum_hbxA_P5	CTTCTTCTCCGCCGTCCCTTGGATCCTATGGATCTCAGAACAATATACC
Afum_hbxA_P6	CATGCCTCCTGCCTACCACTACGTCGACATCACCCTTACCCA
Afum_hbxA_P7	CAAGACAGAATGACTGCCCAAACTGG
Afum_hbxA_P8	CAGTCACCCCATTCCACAGCT
Apara_pyrG_R	CAGGAGCAGCATAAATTCCACGACC
Afum_hbxA_Com_P1	CAAGACAGAATGACTGCCCAAACTGG
Afum_hbxA_Com_P2	GGCTCATCGTCACCCCATTTTGTTAAGAAGCGTTGCCATTGCGTGA
Afum_hbxA_Com_P3	TCACGCAATGGCAACGCTTCTTAACAAAATGGGGTGACGATGAGCC
Afum_hbxA_Com_P4	TCAATGGGCAATTGATTACGGGATCC
R_ptrA Check	CAGCTGCCATCTACGAACCCAC
AFUM 18SqPCR F	TAGTCGGGGGCGTCAGTATTCAGC
AFUM 18S qPCR R	GTAAGGTGCCGAGCGGGTCATCAT
AFUM hbxA qPCR F	GGAGGAGACCGATAAAGCCAACGA
AFUM hbxA qPCR R	GCCGTCATTCCGATCCTGCTC
AFUM brlA qPCR F	TGCACCAATATCCGCCAATGC
AFUM brlA qPCR R	CGTGTAGGAAGGAGGAGGGGTTACC
AFUM abaA qPCR F	CCGCCGCAGGAGACTAGTCAG
AFUM abaA qPCR R	CTGTCGTGAACGCTAACGCCG
AFUM wetA qPCR F	TTGACTCGCTGTCAAGTGATTGTGG
AFUM wetA qPCR R	TGGTGGATTTGTGGTGGGGAGTT
AFUM flbB qPCR F	GCCTTGACACGACGACAGGAAC
AFUM flbB qPCR R	CTGAGCGTCGTTCTGCCCT
AFUM flbD qPCR F	GGGAGCGATTCCACCAGAACC
AFUM flbD qPCR R	ATGGGGCTTCGACTCGGC
AFUM fluG qPCR F	GGGGTAGCTCTACAGAATGCGACT
AFUM fluG qPCR R	CCATTGGTACGGCTCGATGTCC

(ii) Generation of the complementation hbx1 strain.

To generate the complementation strain, a two-fragment fusion PCR method was utilized joining the hbxA locus and the selection marker gene ptrA from Aspergillus oryzae. First, the hbxA locus was amplified from the A. fumigatus genomic DNA using primers Afum_hbxA_Com_P1 and Afum_hbxA_Com_P2, while the ptrA marker was amplified from the plasmid pPTRI (TaKaRa Bio, Mountain View, CA, USA) with primers Afum_hbxA_Com_P3 and Afum_hbxA_Com_P4. The two PCR products were then fused in a similar manner as described above using primers Afum_hbxA_Com_P1 and Afum_hbxA_Com_P4. The fusion cassette was transformed into the ΔhbxA mutant strain TTRS8. Transformants were selected on Czapek-Dox (CZ) medium (Difco, Franklin Lakes, NJ, USA) containing 1 μg/ml pyrithiamine. Confirmation of the reinsertion of hbxA in the transformants was determined by diagnostic PCR with primers Afum_hbxA_qPCR_F and R_ptrA_Check. A selected hbxA complementation strain, TTRS12, was used in this study. The pertinent genotypes of all strains are listed in Table 1.

Morphological analysis.

(i) Colony growth. The wild-type, ΔhbxA mutant, and complementation strains were point inoculated on GMM. The colony diameter was measured after 5 days of incubation at 37°C. The experiment was carried out with three replicates.

(ii) Conidial production.

To assess whether hbxA regulates conidiation in A. fumigatus, 10⁶ spores/ml each of the wild-type, ΔhbxA mutant, and complementation strains were inoculated into 25 ml of liquid GMM. Cultures were grown under stationary conditions, allowing an air interphase to promote development. Cores (7-mm diameter) were collected from the mycelial mats to quantify conidia after 48 h and 72 h of incubation at 37°C. The cores were homogenized in water, and spores were counted under an Eclipse E-400 bright-field microscope (Nikon, Inc., Melville, NY, USA) using a hemocytometer (Hausser Scientific, Horsham, PA). The experiments were performed in triplicate.

(iii) Effect of osmotic stress on hbxA-dependent growth and conidiation.

To examine the possible role of hbxA on osmotic stress resistance, the wild-type, ΔhbxA mutant, and complementation strains were point inoculated on solid GMM and GMM plus 0.6 M KCl, 1 M sucrose, 0.7 M NaCl, or 1.2 M sorbitol. The cultures were incubated in the dark at 37°C for 7 days. Colony growth was assessed by the colony diameter. Under these conditions, conidia were also quantified. Cores were collected from the colonies, and spores were counted as described above.

(iv) Cell wall stress test.

To assess whether hbxA influences the sensitivity to cell wall stress, the wild-type, ΔhbxA mutant, and complementation strains were point inoculated on solid GMM containing 0%, 0.01%, 0.015%, or 0.02% SDS. The experiment was performed in triplicate. Cultures were incubated at 37°C for 72 h.

(v) Germination assay.

Flasks with 50 ml of liquid GMM were inoculated with conidia (10⁶ spores/ml) of wild-type, ΔhbxA mutant, and complementation strains. Every 2 h postinoculation, 500 microliters of culture was collected from each flask for spore quantification under the microscope using a hemocytometer. Micrographs were taken with a Nikon Eclipse E-600 microscope with a total magnification of ×400. The experiment was performed in triplicate.

(vi) Determination of dry weight.

Cultures of 50 ml of liquid GMM inoculated with 10⁶ spores/ml of wild-type, ΔhbxA mutant, and complementation strains were grown in a shaker incubator at 250 rpm at 37°C. After 24 h of incubation, the total content was transferred to a 50-ml Falcon tube and centrifuged at 3,500 rpm for 5 min. The supernatant was removed, and the biomass was lyophilized for 48 h and then weighed.

Gene expression analysis.

Petri dishes containing 25 ml of liquid GMM were inoculated with conidia (10⁶ spores/ml) of A. fumigatus CEA10 wild-type (WT), ΔhbxA mutant, and complementation strains. Cultures were incubated under stationary conditions at 37°C in the dark. Total RNA was extracted from lyophilized mycelial samples using TRIsure reagent (Bioline, Taunton, MA, USA), according to the manufacturer’s instructions. cDNA was synthesized with Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega, Madison, WI, USA). qRT-PCR was performed with the Applied Biosystems 7000 real-time PCR system using SYBR green dye for fluorescence detection. cDNA was normalized to A. fumigatus 18S ribosomal gene expression, and the relative expression levels were calculated using the 2^−ΔΔCT method (43). The primer pairs used are indicated in Table 2.

Liquid chromatography and mass spectrometry analysis.

Sample analysis was performed using high-performance liquid chromatography (HPLC) coupled to an LTQ Orbitrap XL high-resolution mass spectrometer (HRMS; Thermo Fisher Scientific, Les Ulis, France). Extracts were resuspended in 500 μl of acetonitrile-water (50:50 [vol/vol]) and then mixed with 250 μl of acetonitrile, and 10 μl of this suspension was injected into a reversed-phase (150 mm by 2.0 mm) 5-μm Luna C₁₈ column (Phenomenex, Torrance, CA, USA) operated at a flow rate of 0.2 ml/min. A gradient program was performed with 0.1% formic acid (phase A) and 100% acetonitrile (phase B) with the following elution gradient: 0 min of 20% B, 30 min of 50% B, 35 to 45 min of 90% B, and 50 to 60 min of 20% B. HRMS acquisitions were achieved with electrospray ionization (ESI) in the positive and negative modes, as follows: spray voltage of +4.5 kV, capillary temperature of 350°C, sheath gas (N₂) flow rate of 40 arbitrary units (AU), and auxiliary gas (N₂) flow rate of 6 AU in the positive mode; and spray voltage of −3.7 kV, capillary temperature of 350°C, sheath gas (N₂) flow rate of 30 AU, and auxiliary gas (N₂) flow rate of 10 AU in the negative mode. Full MS spectra were acquired at a resolution of 60,000 with a range of mass-to-charge ratio (m/z) set to 50 to 800. The chaetominine standard was purchased from BioAustralis Fine Chemicals (Smithfield, Australia).

Pathogenicity analysis.

Pathogenicity studies using a neutropenic mouse model were carried out as previously described (44), with minor modifications. Briefly, 6-week-old female outbred ICR Swiss mice weighing approximately 25 g were used for this experiment. Fifty mice divided into 5 separate groups were used (each group contained 10 mice). Animals were rendered neutropenic by intraperitoneal injection of cyclophosphamide (150 mg/kg of body weight) on days −4, −1, and 3 postinfection and triamcinolone (Kenalog; 40 mg/kg) on the day of infection. The immunosuppressed mice were infected with fungal spores of A. fumigatus CEA10 wild-type, ΔhbxA mutant, and complementation strains. Sedated mice (10 mice per strain) were infected by nasal instillation of 2 × 10⁶ spores/40 μl phosphate-buffered saline (PBS). Postinfection mice were observed three times daily. Mice that survived to day eight were euthanized.

This study was carried out in strict accordance with the Guide for the Care and Use of Laboratory Animals of the National Research Council. The protocol was approved by the Institutional Animal Care and Use Committee of Northern Illinois University (permit no. 12-0006). All efforts were made to minimize suffering. Humane euthanasia by CO₂ inhalation was performed when mice met criteria indicating a moribund state; these endpoints include behaviors of unresponsiveness to tactile stimuli, inactivity, lethargy, staggering, anorexia, and/or clinical signs of bleeding from the nose or mouth, labored breathing, agonal respirations, purulent exudate from the eyes or nose, abnormally ruffled fur, or greater than 20% weight loss. The method of euthanasia by CO₂ inhalation is consistent with recommendations of the Panel on Euthanasia of the American Veterinary Medical Association.

Statistical analysis.

Statistical analysis was applied to analyze all of the quantitative data in this study utilizing analysis of variance (ANOVA) in conjunction with a Tukey multiple-comparison test using a P value of <0.05 for samples that are determined to be significantly different. The exception was for pathogenicity assays, in which a Kaplan-Meyer survival test was applied.