Introduction

The improvement in transplant medicine and the therapy of hematological malignancies is often complicated by the threat of invasive aspergillosis. Aspergillus fumigatus accounts for approximately 90% of invasive aspergillosis cases. Specific diagnostics are still limited, as are the possibilities of therapeutic intervention, leading to a high mortality rate of 30 to 98% for invasive aspergillosis (reviewed in [8, 12, 22]). An important question concerning A. fumigatus is the identification of pathogenicity determinants and their regulation. The group of J. Kwon-Chung and our group had identified a gene that encodes a pathogenicity determinant. It was designated pksP (or alternatively alb1) for polyketide synthase involved in pigment biosynthesis [14, 15, 19, 20, 38]. Conidia of a pksP mutant strain are white. Based on genetic and biochemical data the conidial pigment consists of dihydroxynaphthalene (DHN)-like melanin [6, 9, 21, 39, 40].

The complete absence of DHN-like melanin, as in the case of pksP mutants resulted in a severe reduction in virulence. PksP mutant conidia of A. fumigatus were significantly more sensitive to hydrogen peroxide and sodium hypochlorite than wild-type conidia. As in other cases, it was shown that melanin-containing conidia are able to quench ROS derived from human granulocytes [14, 15]. These results indicated that conidial DHN-like melanin of A. fumigatus is involved in protecting conidia from the host immune response in which ROS are important for eliminating fungal conidia (reviewed in [21, 22]). However, because A. nidulans conidia are also protected by a green pigment, resistance against ROS does not explain why A. fumigatus conidia can be pathogenic while this is rarely the case with A. nidulans conidia. One attractive hypothesis is that besides the pigment, the pksP gene product of A. fumigatus is involved in the production of another compound that is immunosuppressive [5]. This hypothesis was further supported by the notion that the presence of a functional pksP gene in A. fumigatus conidia is associated with an inhibition of the fusion of phagosomes and lysosomes in human monocyte-derived macrophages [13, 16, 33]. Other pathways involving polyketide synthases have been shown to synthesize two different active products (reviewed in [21]).

The pksP gene is part of a gene cluster which consists of six genes. One of the genes designated abr2 encodes a putative laccase [39]. As mentioned above, our previous data led to the hypothesis that only the pksP gene is involved in virulence and not the other genes of the DHN-like melanin pathway. On the other side, a cell wall bound laccase was found to be required for virulence of the human-pathogenic fungus Cryptococcus neoformans [27, 28, 30, 43, 44]. Therefore, we characterised the putative laccase Abr2 of A. fumigatus in order to study its impact on virulence.

Materials and methods

Fungal and bacterial strains, media and growth conditions

Fungal strains used in this study are listed in Table 1. A. fumigatus ATCC46645 was used to generate an abr2 knock-out strain. A. fumigatus KHΔpyrG is a uracil-auxotrophic mutant of strain ATCC46645. The uracil-auxotrophic strain A. fumigatus KHΔpyrG contains a deletion of 144 bp in the 3′-coding region of the pyrG gene and therefore, codes for a truncated, nonfunctional orotidine-5′-monophosphate decarboxylase (see below).

Table 1 Aspergillus fumigatus strains
Full size table

Strain KHΔpyrG was used to generate strains KHpksPp-lacZ and KHabr2p-lacZ. A. fumigatus strains were cultivated at 37°C in Aspergillus minimal medium (AMM) as previously described [41]. As solid media, malt extract medium (2% (w/v) malt extract, 0.2% (w/v) yeast extract, 1% (w/v) glucose, 5 mM ammonium chloride, 1 mM di-potassium hydrogenphosphate) or AMM containing 3% (w/v) agar were used. Uridine (5 mM) or uracil (5 mM) were added to the media when required. In case the hph or ble gene were used as a selection marker gene for transformation of A. fumigatus, 100 μg hygromycin B and 100 μg phleomycin per ml, respectively, were added to agar plates.

For induction of conidiophore formation of A. fumigatus (developmental cultures), conidia were used to inoculate liquid cultures which were grown for 24 h at 37°C. Mycelia were filtered and exposed to air as previously described for A. nidulans [41]. For transformation of Escherichia coli, XL1-Blue (Stratagene, USA), INVαF′ or TOP10F′ (Invitrogen, The Netherlands) were used. E. coli strains were grown at 37°C in LB medium supplemented, when required, with 100  or 50 μg per ml of ampicillin or kanamycin, respectively.

Colony radial growth rate determination

Colony diameters of A. fumigatus were measured twice a day on both malt extract and AMM agar plates over a period of 94 h. At least 10 colonies of each strain were analysed. Agar plates had been point-inoculated centrally with a 2.5 μl drop of a suspension of 1 × 106 spores per ml. Colony radial growth rates (C r) [37] were calculated from the slope of the line between 40  and 72 h from a plot of colony radius versus time starting from the time of inoculation. Data were processed by least square regression analysis.

Analysis of conidial germination

AMM (50 ml) were inoculated with 1 × 107 conidia. Cultures were incubated with 180 rpm at 37°C. Over a period of 16 h, samples were taken and deposited on microscope slides. To determine germination, at least 100 conidia of each sample were counted.

Standard DNA techniques

Standard techniques in the manipulation of DNA were carried out as described by Sambrook et al. [31]. Chromosomal DNA of A. fumigatus was prepared as previously described for A. nidulans [1]. For Southern blot analysis, chromosomal DNA of A. fumigatus was cut by different restriction enzymes, as indicated. DNA fragments were separated on an agarose gel and blotted onto Hybond N+ nylon membranes (Amersham Pharmacia Biotech, UK). Labeling of the DNA probe, hybridization and detection of DNA-DNA hybrids were performed using the DIG High Prime Labeling and Detection System (Amersham Pharmacia Biotech, UK) according to the manufacturer’s recommendations.

Sequence analysis

Plasmid DNA was sequenced on both strands by primer walking using the Big DyeTM Terminator Cycle Sequencing Kit (Applied Biosystems, UK). Sequencing reactions were separated on an Applied Biosystems ABI 310 sequencer. DNA sequence data were edited by the programs “Sequence Navigator” and “Auto Assembler” (Applied Biosystems, UK). The analysis of sequences was carried out using “Gene Works 2.2” (IntelliGenetics Inc., USA).

Generation of recombinant plasmids

For generation of plasmid pKH1pyrGhph an upstream flanking region (1020 bp) of the A. fumigatus pyrG gene with an introduced XbaI restriction site at its 3′end was generated by PCR using oligonucleotides PyrG1 and PyrG2XbaI (Table 2), and genomic DNA from the A. fumigatus wild-type strain as a template. The downstream fragment of pyrG with a size of 1032 bp with an introduced XbaI restriction site at its 5′end was synthesised by PCR, employing PyrG3XbaI and PyrG4 oligonucleotides (Table 2), and genomic DNA from the wild-type strain as a template. The upstream and downstream fragments of the pyrG gene were cloned independently into the pCR2.1 vector (TA cloning kit, Invitrogen, The Netherlands), yielding plasmids pTAFR1 and pTAFR2, respectively. After linearisation of pTAFR1 with XbaI, the hph selection marker gene conferring hygromycin B resistance, was introduced. A DNA fragment encoding the hph gene was obtained via double restriction of plasmid pUChph1 [25] with XbaI and SpeI. In the XbaI site of the resulting vector designated pTAFR1hph, the downstream fragment was transferred from the vector pTAFR2, by excision with XbaI restriction endonuclease.

Table 2 Oligonucleotide primers used in the study
Full size table

To construct the abr2 deletion plasmid pUCΔ abr2, a 4125 bp PCR product was generated using oligonucleotides Abr2del_for and Abr2del_rev (Table 2), and chromosomal DNA of the A. fumigatus wild-type strain ATCC46645 as a template. The resulting PCR fragment spanned the abr2 gene and its upstream and downstream flanking region with a size of 1,035 bp and 1,020 bp, respectively. This PCR fragment was cloned into the pCR2.1 TOPO vector (TA Cloning Kit, Invitrogen, The Netherlands) to give plasmid pCR2.1abr2. The PCR fragment was re-isolated by restriction digest of plasmid pCR2.1abr2 with EcoRI. The obtained DNA fragment of 4,125 bp was ligated into pUC18, also digested with EcoRI, yielding plasmid pUCabr2. As a selection marker, the hph gene was used. It was obtained from plasmid pUChph1 [25], digested with NsiI and XmnI. The resulting DNA fragment of 2,792 bp was cloned into plasmid pUCabr2 after restriction with NsiI, which allowed the replacement of the abr2-encoding region by hph. Thus, the resulting plasmid pUCΔabr2 contained the E. coli hph gene under control of the A. nidulans gpdA promoter, flanked by fragments encoding upstream and downstream sequences of the abr2 gene. Plasmid pUCΔabr2 was linearised by digestion with EcoRI. The resulting 4.8 kbp DNA fragment was used for transformation of the A. fumigatus wild-type strain ATCC46645.

For complementation of the abr2 deletion mutant, a PCR fragment of 3,641 bp comprising the abr2 gene and 1 kb of promoter region, was generated by the use of primer pair Abr2P_for and Abr2.2_rev (Table 2). The resulting abr2-encoding PCR fragment was used for a co-transformation approach. Co-transformation was carried out by the use of the plasmid pAN8-1 which contains the ble gene, conferring resistance to phleomycin [29]. The ble gene is under control of the A. nidulans gpdA promoter. Transcription of ble is terminated by the A. nidulans trpC terminator. For the increase of transformation frequency plasmid HELP1 [19] was simultaneously added to the transformation mix. Transformants were selected on AMM agar plates containing phleomycin. Phleomycin-resistant transformants producing gray-green conidia were checked by PCR for the presence of the abr2 gene using primers Abr2disr_for and Abr2disr_rev (Table 2) and chromosomal DNA of the transformants as the template.

To measure abr2 expression, an abr2p-lacZ gene fusion was generated. For this purpose, a 1.23 kbp DNA fragment spanning the putative promoter region of abr2 was amplified by PCR using the oligonucleotides Abr2P_rev and Abr2P_for (Table 2), each of which encoding BamHI restriction sites. As the template, chromosomal DNA of A. fumigatus wild type was used. The PCR fragment obtained was cloned into the pCR2.1 TOPO vector (TA Cloning Kit, Invitrogen, The Netherlands). After restriction with BamHI, the DNA fragment spanning the promoter region was cloned into the BamHI site of plasmid pUCpyrG2lacZ [23], which carries the non-functional pyrG2 allele. The resulting plasmid pUCpyrG2abr2PlacZ encoded an in frame abr2p-lacZ gene fusion. This was checked by DNA sequence analysis across the junctions. All lacZ-containing plasmids carried the pyrG2 allele as the selection marker. It encodes a nonfunctional pyrG gene of A. fumigatus, which forced site-specific integration of plasmids into the chromosomal pyrG locus [42].

Transformation of A. fumigatus and generation of the pyrG deletion strain KHΔpyrG and of strain KHpksPp-lacZ

Transformation of A. fumigatus was carried out using protoplasts as previously described [42]. The pyrG deletion strain KHΔpyrG was generated as follows: the vector pKH1pyrGhph was cut with SpeI and BglII yielding a DNA fragment of 5.9 kbp, which was used for transformation of the A. fumigatus wild-type strain ATCC46645. The selection of pyrG deletion strains occurred on AMM agar plates containing uracil and hygromycin B. Uracil-auxotrophic, hygromycin B-resistant transformants were checked by Southern blot analysis (data not shown). One of the resulting uracil-auxotrophic, hygromycin-resistant transformants was designated KHΔpyrG. It contained a deletion of 144 bp in the 3′ coding region of the pyrG gene. When uracil-auxoptrophy was used, the A. fumigatus strain KHΔpyrG (Table 1) was applied. When selection for hygromycin B resistance was used, the wild-type strain ATCC46645 was employed. Strain KHpksPp-lacZ was generated by transformation of strain KHΔpyrG with plasmid pUCpyrG2pksP-lacZ [23] encoding a pksPp-lacZ gene fusion. By Southern blot analysis, a transformant strain designated KHpksPp-lacZ was identified encoding a single copy of plasmid pUCpyrG2pksP-lacZ integrated at the chromosomal pyrG gene locus (data not shown).

Field emission scanning electron microscopy (FESEM)

FEMES was carried out according to Maerker et al. [26]. In brief, the conidia were harvested with sterile water, containing 10 mM MgCl2 and 10 mM CaCl2 to a final concentration of 3 × 108 conidia per ml. For fixation of the conidia, 35% (v/v) formaldehyde was used in a final concentration of 5% (v/v). After incubation on ice for 10 min, glutaraldehyde was added to a final concentration of 2% (v/v). After washing with cacodylate buffer and subsequently with TE buffer, the samples were placed onto poly(L-lysine) coated glass cover slips.

ß-Galactosidase (ß-GAL) activity assays

A. fumigatus strains were grown in AMM at 37°C. ß-GAL activities were measured in protein extracts obtained from three A. fumigatus cultures grown in parallel. Specific activities were calculated as previously described [25].

Laccase activity assay and determination of protein concentrations

Aspergillus fumigatus strains were grown in AMM at 37°C. After harvesting at different time points as indicated, mycelia were frozen in liquid nitrogen and ground to a fine powder. Mycelia were suspended in extraction buffer (Tris/HCl 0.1 M, pH 7.0). Samples were centrifuged at 4°C with 13,000 rpm for 10 min. The supernatant was retained. N,N-Dimethyl-p-phenylendiamine (DMP) was used as the substrate. Laccase activity was determined in 900 μl laccase buffer (37 mM citric acid monohydrate, 126 mM Na2HPO4, pH 6.0), 50 μl DMP (147 mM stock solution) and 50 μl enzyme solution. Extinction was followed at 550 nm at 25°C and the enzyme activities were calculated using a molar extinction coefficient of 1.8 mM−1 cm−1, as previously described [32]. Laccase activities were measured in protein extracts obtained from three A. fumigatus cultures grown in parallel. Protein concentrations were determined according to Bradford [3].

Sensitivity towards reactive oxygen species

1 × 108 A. fumigatus conidia were mixed with 10 ml AMM top agar and poured onto AMM agar plates. In the center of the agar plate, a hole with a diameter of 10 mm was created, which was filled with a solution of H2O2 or diamide. After an incubation of the agar plates at 37°C for 16 h, the diameter of the inhibition zone was measured.

Animal infection model

An optimised murine low dose model for invasive aspergillosis was applied [24, 35]. Mice were intranasally infected with a 25 μl drop of a fresh suspension containing 5 × 104 conidia. Survival was monitored daily, and moribund animals were sacrificed by intraperitoneal injection of 200 μl 3.2% (v/v) narcoren (Rhone Merieux, Germany). The drinking water was supplemented with 0.5 mg of tetracycline (Sigma) per ml to prevent opportunistic bacterial infections. A control group (inhalation of PBS) remained uninfected to monitor the influence of the immunosuppression procedure on vitality.

Results

Deletion of the laccase-encoding gene abr2 of A. fumigatus

To analyse the importance of Abr2 for the overall laccase activity of A. fumigatus, the abr2 gene was deleted. For this purpose, plasmid pUCdelabr2 (Fig. 1a) was generated (see Materials and methods). A DNA fragment obtained by digestion of plasmid pUCdelabr2 with XmnI which encodes the hygromycin B resistance gene hph flanked by upstream and downstream sequences of the abr2 gene (Fig. 1a), was used for transformation of the A. fumigatus wild-type strain ATCC46645. Forty two hygromycin B resistant transformants were isolated. Twenty of them were tested by PCR for the presence of the hygromycin resistance gene. Seven of those, which showed the presence of the selection marker gene, were analysed by Southern blot analysis. Four of these transformants exhibited the expected gene replacement (Fig. 1b, lanes 3, 4, 5, 7, Fig. 1c). One of the mutant strains (lane 3) was designated AfΔabr2 and used for further studies. Growth of strain AfΔabr2 on agar plates revealed that the abr2 deletion affected the pigment formation of conidia (Fig. 2a). The colonies showed a brown color. A similar finding was previously reported for an abr2 mutant by both Tsai et al. [38] and Krappmann et al. [17]. Complementation of the abr2 deletion mutant using the wild-type abr2 gene was carried out. For this purpose, the abr2 gene was amplified by PCR. The generated PCR fragment was applied to a co-transformation experiment, using plasmids pAN8-1 and pHELP1 (see Materials and methods). After transformation of the Δabr2 mutant, 17 phleomycin-resistant transformants were isolated. Six of them produced gray-green conidia, i.e., were complemented to the wild type (data not shown). The presence of the abr2 gene in the transformants producing wild-type conidia was shown by PCR analysis (data not shown).

Fig. 1
figure 1

Deletion of A. fumigatus abr2 gene. a Schematic map of the generation of the abr2 knock-out plasmid pUCdelabr2. Abbreviations: amp R, ampicillin resistance gene; hph, hygromycin B phosphotransferase gene used as the selection marker gene in A. fumigatus. b Southern blot analysis. Chromosomal DNA of the wild-type strain ATCC46645 (lane 1) and transformant strains (lanes 2–8) was cut by SacII. A 641 bp abr2-derived PCR fragment was used as the probe. In the AfΔabr2 mutant strains, the band characteristic of the wild type (lane 1) of 4,344 bp had disappeared. Instead, the band of 2,725 bp characteristic of gene replacement at the abr2 locus was detected (see c). c Schematic representation of the chromosomal abr2 locus of the wild type and the AfΔabr2 deletion mutant. Restriction endonuclease cleavage sites, the DNA fragments identified by Southern blot analysis (b) and the position to which the probe hybridises, are indicated

Full size image
Fig. 2
figure 2

Phenotypic characterisation of mutant strain AfΔabr2 and wild-type strain ATCC46645. a Growth and sporulation. Colonies were grown on AMM agar plates for 72 h at 37°C. b Kinetics of germ tube outgrowth for A. fumigatus conidia incubated in AMM at 37°C. The number of conidia showing a germ tube was recorded at different times of incubation and is presented as the percentage of the total number of conidia. The results are representative of the results of two independent experiments. c, d Growth of A. fumigatus strains on AMM agar plates and malt agar plates, respectively, at 37°C. The diameter of colonies was determined. Data for each strain represent the mean of at least ten colonies, grown independently. SDs were in the range of 0.2–0.7 mm

Full size image

The germination of conidia (Fig. 2b), the growth rate measured as diameter of colonies on AMM agar (Fig. 2c) and malt agar (Fig. 2d) were the same for both the abr2 deletion mutant and the wild-type strain, indicating that abr2 is not essential for vegetative growth. Radial growth rates (Cr) of colonies of A. fumigatus wild-type and Δabr2 mutant strain on AMM agar plates were 0.46 and 0.48 mm/h, respectively. Cr on malt extract showed values of 0.5 mm/h for A. fumigatus ATCC46645 and 0.53 mm/h for the abr2 deletion strain. SDs were in the range of 0.005–0.02.

Conidia of the strain AfΔabr2 were analysed by FESEM (Fig. 3). As previously reported [14] and shown here as a control, wild-type conidia display an ornamented surface (Fig. 3a) which is lacking in pksP mutant conidia (Fig. 3b). Conidia of the abr2 deletion strain displayed some ornamentation which was less pronounced than that observed on wild-type conidia. This finding indicates that formation of the DHN-like melanin intermediate present in the Δabr2 mutant is sufficient for production of at least some ornamentation on the surface of conidia.

Fig. 3
figure 3

Field emission scanning electron micrographs of conidia. Bar: 2 μm. a A. fumigatus wild-type strain ATCC46645. b A. fumigatus pksP mutant. c A. fumigatus abr2 deletion mutant

Full size image

In contrast to the pksP mutant the abr2 deletion mutant showed the same sensitivity against H2O2 and diamide in vitro as the wild type

Previously, we showed that mutation of the polyketide synthase gene pksP led to increased sensitivity of the respective mutant against ROS generated by immune effector cells [14]. To analyse whether the deletion of abr2 enhances sensitivity against ROS in the respective mutant or whether the compound produced by the DHN-like melanin pathway to the stage of the Abr2 laccase is sufficient to scavenge ROS, the sensitivity of the Δabr2 mutant against ROS was compared with that of both the pksP mutant and the wild-type strain. As shown in Fig. 4a, the pksP mutant showed increased sensitivity against ROS whereas there was no significant difference between the Δabr2 mutant and the wild-type strain. The analysis of the effect of diamide confirmed the results obtained for H2O2, but in case of the pksP mutant the effects were less prominent (Fig. 4b). Taken together, the Δabr2 mutant showed the same sensitivity against ROS as the wild type.

Fig. 4
figure 4

Sensitivity of AfΔabr2 towards H2O2 and diamide. The strains A. fumigatus ATCC46645 (Af wt), AfpksP and AfΔabr2 were analysed. Two concentrations of H2O2, 50 μl and 100 μl from a 6% (v/v) H2O2 solution, were examined. 100 μl and 200 μl from a 0.1M diamide solution were used in the experiment. Data for each strain and concentration represent the mean and SDs of five independently performed assays. a H2O2 b Diamide

Full size image

Abr2 activity was detectable during sporulation but did not contribute to laccase activity during hyphal growth under standard conditions

To determine the contribution of Abr2 to total laccase activity, cell extracts of both the wild type and the Δabr2 deletion strain were analysed for laccase activity. Laccase activity was detectable in both strains during vegetative growth after 24 h (Fig. 5). In contrast to the wild-type strain, laccase activity was reduced in the Δabr2 mutant strain in sporulating mycelia. These findings indicate that the Abr2 laccase is mainly active during sporulation under standard conditions. Furthermore, these results imply that additional laccases are active during vegetative growth of the fungus under the conditions tested and that their activities also increase during sporulation as seen from the abr2 deletion strain.

Fig. 5
figure 5

Laccase activity of mutant strain AfΔabr2 and wild-type strain ATCC46645 (Af wt). Enzyme activity was monitored during vegetative growth in AMM after 24 h (mycelium) and sporulation after 24, 34 and 48 h (sp)

Full size image

Sporulation-dependent expression of an abr2p-lacZ gene fusion

To analyse the regulation of abr2 expression and correlate this data with the Abr2 activity during sporulation, A. fumigatus strains were generated carrying an abr2p-lacZ gene fusion integrated in single copy at the pyrG gene locus. For this purpose, plasmid pUCPyrG2abr2PlacZ was used (Fig. 6a). It encodes the abr2 gene promoter fused in frame with the E. coli lacZ gene. Transformation of A. fumigatus strain KHΔpyrG using this plasmid resulted in the isolation of eight transformants. Southern blot analysis indicated the presence of the gene fusion in single copy at the pyrG gene locus (Fig. 6b, c). The 8 kbp band characteristic of the wild-type pyrG gene (Fig. 6a, lane 1) had disappeared in the transformants (Fig. 6a, lanes 2 and 3). Instead, they showed a 10 kbp band due to the integration of the plasmid at the pyrG gene locus. The transformants shown in Fig. 6c were designated KHabr2placZ-2 and KHabr2placZ-4 and used in further studies.

Fig. 6
figure 6

Integration of an abr2p-lacZ gene fusion in single copy at the A. fumigatus chromosomal pyrG gene locus. a Schematic map of plasmid pUCPyrG2abr2PlacZ. Abbreviations: abr2P, promoter region of the abr2 gene; AmpR, ampicillin resistance gene; lacZ, E. coli lacZ gene; PyrG, orotidine 5′-monophosphate decarboxylase gene of A. fumigatus, used as the selection marker gene. The asterisks indicate mutations. b Schematic representation of the chromosomal pyrG gene locus of strain KHΔpyrG carrying a deletion of part of the pyrG gene and of strain KHabr2placZPyrG carrying the abr2p-lacZ gene fusion integrated at the pyrG gene locus. Restriction endonuclease cleavage sites and the position, to which the probe hybridizes, are indicated. c Southern blot analysis. Chromosomal DNA of the A. fumigatus wild-type strain KHΔpyrG (lane 1) and strains KHabr2placZPyrG-2 and KHabr2placZPyrG-4 (lanes 2 and 3) was digested by BglII. A 450 bp PCR fragment encoding a part of the A. fumigatus pyrG gene (b), was used as the probe

Full size image

The expression of the abr2p-lacZ gene fusion was determined. Results are shown in Fig. 7a. As expected, the expression of the gene fusion was detectable during sporulation and increased up to 48 h. There was hardly abr2p-lacZ expression during vegetative growth of the fungus. These data well agree with the results on the laccase specific activity (Fig. 5) indicating that Abr2 mainly contributes to overall laccase activity during sporulation but not during vegetative growth under the conditions applied. Interestingly, the expression pattern of the pksPp-lacZ and abr2p-lacZ gene fusions apparently differed. The abr2p-lacZ expression increased seventeen-fold during sporulation after 48 hcompared with the expression during vegetative growth, whereas for the pksPp-lacZ gene fusion, this increase during sporulation only was 2.5-fold (Fig. 7b).

Fig. 7
figure 7

Expression of abr2p-lacZ (a) and pksPp-lacZ (b) gene fusions during vegetative growth and sporulation. The β-GAL activity was measured after 24 h of growth in liquid culture, and after 24 and 48 h after induction of sporulation. Data for each condition represent the mean and SDs of three independently grown cultures

Full size image

The Δabr2 mutant showed no reduction in virulence compared with the wild type

To assess a possible role of Abr2 in pathogenesis, the corresponding deletion mutant was tested in an intranasal mouse infection model of invasive aspergillosis. Groups of 10 immunosuppressed mice were infected by intranasal inhalation with 5 × 104 conidia of the wild-type or the Δabr2 mutant strain. Results of a representative experiment are shown in Fig. 8. In the groups infected with wild-type conidia (strain ATCC46645), mortality was 80% after 11 days (Fig. 8). When mice were infected with conidia of the Δabr2 mutant strain AfΔabr2, mortality was 80% after 13 days. Hence, the overall-mortality of the abr2 deletion strain was similar to that of the wild type. Moreover, at the beginning of the experiment more of the mice infected with conidia of the AfΔabr2 strain died compared with mice infected with wild-type conidia. Taken together, these data indicate that Abr2 does not contribute to virulence of A. fumigatus.

Fig. 8
figure 8

Virulence of A. fumigatus wild type (ATCC46645) and Δabr2 mutant strain in mice. Groups of ten BALB/c mice were analysed, each infected intranasally with 5 × 104 A. fumigatus conidia, as indicated. Survival was monitored for 17 days

Full size image

Discussion

Under the conditions applied, laccase activity of A. fumigatus was detected during vegetative growth, which strongly increased during sporulation. The analysis of the abr2 deletion mutant led to the conclusion that Abr2 only contributes to overall laccase activity during sporulation. It is worth to notice that the substrate for determinating laccase activity, DMP, can be oxidised not only by laccases but also by other enzymes as peroxidases [32]. However, because Abr2 clearly used DMP as the substrate and its amino acid sequence resembles that of a typical fungal laccase, it is very likely that Abr2 represents a laccase. The finding that Abr2 only contributes to overall laccase activity during sporulation well agrees with the analysis of an abr2p-lacZ gene fusion which was mainly expressed during sporulation. By contrast, there was expression of a pksPp-lacZ gene fusion during vegetative growth which also increased during sporulation, but the increase was less than that observed for the abr2p-lacZ gene fusion. In fact, the abr2p-lacZ expression increased steadily up to seventeen-fold during sporulation compared with the expression during vegetative growth, whereas for the pksp-lacZ gene fusion, this increase during sporulation only was 2.5-fold and reached its maximum after 24 h. This finding implies that differential regulation of genes from the same cluster occurs. A similar observation was made for genes of other clusters in filamentous fungi, e.g., genes belonging to the penicillin biosynthesis cluster in A. nidulans are, at least in part, differentially regulated [4, 7]. Consistently, the promoter regions of pksP and abr2 apparently contain different cis-acting elements. Computer analysis revealed that the pksP promoter region encodes a single putative stress response element (STRE) [11, 36], three putative AbaA binding sites [2], a potential cAMP-responsive element originally described in the S. cerevisiae SSA3 gene promoter and two putative stunted A (StuA) binding sites [10]. By contrast, in the promoter region of abr2 a single putative AbaA site, a single potential STRE and two putative copper signalling elements (CuSE) were detected. CuSE are defined by the consensus sequence 5′-DWDDHGCTGD-3′ (D = A, G, or T; H = A, C, or T and W = T, or A). They are bound by Cuf1 (copper-sensing transcription factor) and were described for Schizosaccharomyces pombe [18]. Whether these elements contribute to abr2 expression remains to be tested.

Even in liquid medium abr2p-lacZ expression increased during later stages of the cultivation (data not shown) when most likely sporulation was induced even in liquid medium [34]. Based on the analysis of the abr2 deletion mutant it can be concluded that the laccase activity measured during vegetative growth was due to other laccases of A. fumigatus. In the genome of A. fumigatus, there are at least three additional candidate genes (Afu1g15670, Afu4g14280, Afu2g17540; www.tigr.org). Abr2 activity is mainly, if not exclusively, required for the production of the gray-green spore pigment. Interestingly, in contrast to the pksP mutant, conidia of the Δabr2 mutant did not show increased sensitivity against ROS or diamide. Therefore, the intermediate of DHN-like melanin produced up to the stage of Abr2 is apparently sufficient to protect conidia against ROS. This assumption was further supported by the observation that in a low dose intranasal mouse infection model there was no difference in virulence of the abr2 mutant compared with the wild type. Furthermore, the surface of Δabr2 conidia looked more similar to that of wild-type conidia, i.e., it displayed at least some ornamentation, whereas the pksP mutant showed a smooth surface [14]. Taken together, these data support the model that mainly the pksP gene of the DHN-like melanin biosynthesis gene cluster is important for virulence.