| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Eukaryotic Cell, September 2007, p. 1570-1583, Vol. 6, No. 9
1535-9778/07/$08.00+0 doi:10.1128/EC.00085-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
,
Departamento de Genética Molecular, Instituto de Fisiología Celular, Universidad Nacional Autónoma de México, Apartado Postal 70-242, 04510, México, D.F., México
Received 16 March 2007/ Accepted 6 July 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
sskA), SrrA (srrA), or both RR (srrA sskA) are viable and differentially affected in osmotic and oxidative stress responses. Both RR are involved in osmostress resistance, but sskA mutants are more sensitive to this stress, and only SrrA is required for H2O2 resistance and H2O2-mediated induction of catalase CatB. In contrast, both RR are individually required for fungicide sensitivity and calcofluor resistance and for normal sporulation and conidiospore viability. The srrA and sskA sporulation defects appear to be related to decreased mRNA levels of the key sporulation gene brlA. In contrast, conidiospore viability defects do not correlate with the activity of the spore-specific catalase CatA. Our results support a model in which NikA acts upstream of SrrA and SskA to transmit fungicide signals and to regulate asexual sporulation and conidiospore viability. In contrast, NikA appears dispensable for osmotic and oxidative stress signaling. These results highlight important differences in stress signal transmission among fungi and define a phosphorelay system involved in oxidative and osmotic stress, cell wall maintenance, fungicide sensitivity, asexual reproduction, and spore viability. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Among eukaryotic cells, phosphorelay systems have been described for slime molds, fungi, and plants (45, 50, 55, 58, 64, 68). Fungal phosphorelay systems consist of one or several hybrid HKs, a histidine-containing phosphotransfer (HPt) protein, and two canonical RRs. For instance, Saccharomyces cerevisiae has a single membrane-associated HK, Sln1; one HPt protein, Ypd1; and two RRs, Ssk1 and Skn7 (also called Pos9). Under normal osmotic conditions, Sln1 is an active kinase that autophosphorylates at the conserved histidine of the transmitter domain. The phosphoryl group is then successively transferred to the conserved aspartate of the Sln1 receiver domain, to the conserved histidine of Ypd1, and finally to the conserved aspartates of Ssk1 and Skn7 (58). Phosphorylation of Ssk1 prevents the activation of the partially redundant mitogen-activated protein kinase (MAPK) kinase kinases Ssk2 and Ssk22. High-osmolarity conditions promote dephosphorylation of phosphorylated Sln1, resulting in accumulation of unphosphorylated Ssk1. Unphosphorylated Ssk1 interacts physically with Ssk2 and Ssk22, leading to their activation and the subsequent phosphorylation of Pbs2 and Hog1 (57). Phosphorylated Hog1, in turn, activates various transcription factors, responsible for the induction of genes required for survival in a hyperosmotic environment (see reference 28 for a review). The other RR, Skn7, is a typical RR with a DNA-binding domain similar to that of the heat shock factor Hsf1 (11) and thus is directly involved in transcriptional regulation. The skn7 gene was identified in different screenings as a high-copy suppressor of mutations affecting cell wall assembly (11) and lethality associated with loss of the transcription factors SBF and MBF (47), as a mutation causing increased sensitivity to oxidative stress (37), and as a high-copy activator of a SLN1-dependent reporter gene (41).
In contrast to S. cerevisiae, the fission yeast Schizosaccharomyces pombe contains not one but three HKs, Mak1, Mak2, and Mak3, in addition to the Ypd1-orthologous HPt protein Mpr1 and the two RRs Mcs4 and Prr1. Although these phosphorelay pathways are similar to the Sln1-Ypd1-Ssk1 system, they are specialized in transmitting oxidative rather than osmotic stress signals. Mak2 and Mak3 (12) have been shown to transmit oxidative stress signals, through Mcs4, to the stress MAPK Spc1/Sty1 (12, 50), which is a Hog1 ortholog. Mak1 appears to regulate Prr1 (12, 54), the Skn7 ortholog. The prr1– mutant is sensitive to H2O2 and cadmium and is defective in the induction of the ctt1+ gene (53), which encodes the only catalase present in this fungus. The analysis of several fungal genomes indicates that the number of HKs in filamentous fungi can range from 11 to 21, and all of them appear to use a single HPt protein to relay signals to two conserved RRs (9, 14). How different environmental signals are perceived and integrated through these proteins is a matter of intense research.
Aspergillus nidulans is closely related to other aspergilli of medical or industrial importance and is a good genetic model with which to study the biology of fungi and other eukaryotes (2, 5, 16, 19, 44, 69, 70, 76). In this fungus, asexual reproduction (conidiation) is induced by environmental signals such as exposure to air (2, 16, 69), nutrient starvation (66), and self-generated signals (39, 43, 65, 67, 70). The mechanisms by which these signals are perceived and transduced are not well understood.
Conidiation implicates the development of specialized structures called conidiophores that are able to produce numerous uninucleated conidiospores. As dormant structures, capable of resisting different types of stress and germinating under favorable conditions, conidiospores are critical for the dispersal of A. nidulans and many other fungi. The formation of conidiophores and conidia is completely dependent on the expression of the brlA gene (1, 3, 16, 66).
The A. nidulans genome (19) and our own analysis predict 15 HKs, 1 HPt protein, and 2 canonical RRs called SskA (17) and SrrA (this work; GenBank accession no. AY168636). Moreover, two additional proteins with putative receiver domains are predicted: AN7572.3, homologous to S. cerevisiae Rim15, which lacks the aspartate that is conserved in canonical RRs, and AN4134.3, which lacks a clear effector domain. The functions of the three HK-encoding genes (tcsA, tcsB, and fphA) have been investigated for this fungus. The tcsA gene encodes a PAS/PAC domain HK that belongs to the class IV HKs (14). Conidiation was reduced in tcsA mutants, a phenotype that was suppressed under high osmolarity or after successive propagation of the strains (73). Mutation of tcsB, the ortholog of yeast sln1, did not produce any clear phenotype (18). Finally, fphA was shown to encode a red-light sensor phytochrome involved in repression of sexual development (8).
A. nidulans SakA, also called HogA (24), is a Hog1 and Spc1/Sty1 ortholog that can replace Spc1 functions in S. pombe and is activated by both oxidative and osmotic stress signals. However, mutants lacking SakA are not sensitive to osmotic stress (34). Recently, it was found that the RR SskA mediates the activation of SakA in response to oxidative and osmotic stress signals (17). We have proposed that oxidative stress plays a central role in determining cell differentiation in fungi and other eukaryotes (5, 25). In this context, we are interested in understanding how filamentous fungi perceive and respond to oxidative and other stress signals. Given the importance of phosphorelay systems in transmitting oxidative stress signals in S. pombe (29) and other fungi (4, 7, 15), we sought to determine the roles of the HK NikA, the HPt protein YpdA, and the RRs SrrA and SskA in A. nidulans stress signal transduction and cell differentiation.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Strains, media, and growth conditions.
The A. nidulans strains used in this work are listed in Table 2. Genotypes of progeny derived from sexual crosses were initially determined by colony morphology and growth tests on media containing high sorbitol, H2O2, or the fungicide fludioxonil and were confirmed by Southern blot and/or PCR analysis. The presence of the wild-type nkuA+ allele was determined by PCR using primers 5DignkuA and 3DignkuA. All strains were grown in glucose minimal nitrate medium (26) plus supplements. Menadione and fludioxonil were filter sterilized and, like H2O2, added to agar medium at 
50°C before solidification. H2O2-containing plates were either used the day they were prepared or stored at 4°C for no more than 24 h. Since H2O2 can react with medium components, the actual concentration in plates cannot be estimated. To ensure experimental reproducibility, the same batch of H2O2-containing medium was used for comparing different strains. To test heterokaryon sensitivity under osmotic and oxidative stress, mycelial plugs of the same area (diameter, 0.5 cm) were cut from the borders of growing heterokaryotic colonies and transferred to the testing medium. Induction of catalase CatB by H2O2 and in-gel catalase activity were determined as reported previously (33, 35, 48). Induction of asexual sporulation, sample processing, and brlA Northern blot hybridization were carried out as reported previously (3, 66).
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Deletion of genes encoding the HPt protein YpdA and the RR proteins SrrA and SskA.
To analyze the function(s) of the A. nidulans signal transduction pathways that are formed by the phosphotransfer protein YpdA (17) and RRs SrrA and SskA, we attempted to delete the corresponding genes and generate ypdA, srrA, sskA, and srrA sskA null mutants. The deletion constructs hptA-AfpyrG-hptA (Fig. 1A), srrA-AfpyrG-srrA (see Fig. S2A in the supplemental material), and sskA-AfriboB-sskA (see Fig. S2B in the supplemental material) were generated by double-joint PCR (75) and were used to transform protoplasts or electrocompetent conidia (62, 63) from the A. nidulans nkuA– strain 11035. In this strain, most DNA recombination events occur via homologous recombination (49).
|
ypdA mutants by transforming a regular strain using protoplast fusion, a procedure involving the use of a high-osmolarity medium. Here we used an alternative strategy, aimed at maximizing gene-targeting efficiency (nkuA– strain 11035) and avoiding possible high-osmolarity counterselection of the desired mutants (transformation by conidium electroporation). This procedure generated eight PyrG+ transformants, all showing irregular growth and conidiation, characteristic of A. nidulans heterokaryons (Fig. 1B). Conidiospores isolated from these colonies failed to grow in selective medium, suggesting that deletion of ypdA is a lethal event. To confirm this, we extracted DNA from several heterokaryons and used it for PCR analysis. The results in Fig. 1C show that in contrast to wild-type DNA, heterokaryotic DNA generated products of 3 and 4.2 kb, consistent with amplification of both wild-type and ypdA alleles. The smaller amount of the ypdA PCR product suggests that heterokaryons contain a higher proportion of ypdA+ nuclei. Additional PCR analysis (data not shown) using a ypdA primer external to the deletion construct (Fig. 1A) and primer PyrGreverse generated the expected PCR product only when heterokaryotic DNA was used as a template, further confirming ypdA deletion in the heterokaryons. We conclude that the lack of the HPt protein YpdA very likely results in a lethal event in A. nidulans.
Nevertheless, we explored the sensitivities of ypdA–/ypdA+ heterokaryons to osmotic and oxidative stress conditions by transferring mycelial plugs from growing heterokaryons 1 and 4 to a medium containing either 1 to 3 mM H2O2 or a high sorbitol or NaCl concentration. The responses of both heterokaryons were very similar under all conditions tested. While similar growth was observed in minimal medium (MM) and in the presence of H2O2, growth was decreased in 1.2 M sorbitol and was totally inhibited in 0.6 M NaCl (Fig. 1D). These results indicate that the partial lethality of the ypdA deletion in heterokaryons is increased by high osmolarity but is not affected by high H2O2 levels. To determine the terminal phenotype of the ypdA mutant, spores collected from ypdA–/ypdA+ heterokaryons were germinated in selective liquid medium for 16 h and examined under the microscope. Although all spores failed to germinate under these conditions, two different morphologies were distinguished. About 80% of the conidia were swollen and bigger than the remaining spores (data not shown). In line with the low ratio of the ypdA/ypdA+ PCR product in Fig. 1C, this result suggests that the spores that did not swell include ypdA (PyrG+) spores and thus the possible ypdA terminal phenotype. In any case, it is clear that ypdA mutants can be maintained only as heterokaryons and that ypdA conidiospores appear unable to form a germ tube.
To delete most of the srrA ORF, including the DNA-binding and receiver domain regions (see Fig. S1 in the supplemental material), we used the srrA-AfpyrG-srrA replacement construct. Several PyrG+ transformants were analyzed by Southern blotting (data not shown), and transformants with the correct gene replacement event [TsrrA-pyrG(nkuA)7 and -8 and TsrrA-pyrG9] were selected for further experiments. srrA mutants obtained from sexual crosses were further analyzed by Southern blotting (see Fig. S2A in the supplemental material).
The sskA-AfriboB-sskA construct, designed to delete the entire sskA ORF, generated 11 RiboB+ transformants, 9 of which were screened by Southern blotting. Transformants TsskA-riboB(nkuA)7 and -8 were shown to contain the desired sskA gene replacement event (see Fig. S2B in the supplemental material) and were used in further experiments. When both constructs (srrA-AfpyrG-srrA and sskA-AfriboB-sskA) were used to transform strain 11035, eight PyrG+ riboB+ transformants were obtained. Strains TsrrA-pyrG/sskA-riboB1(nkuA)6 and -7 were shown to contain the deletions of the srrA and sskA genes and were selected for additional experiments (see Fig. S3A in the supplemental material).
The deletions of srrA and sskA resulted in phenotypes that were clearly distinguishable from the wild-type phenotype (Fig. 2 to 4; see also Fig. S4 in the supplemental material). srrA mutants formed colonies with irregular borders and with apparent reductions in conidiophore density and conidiospore pigmentation. In addition, the mycelia from srrA mutants accumulated a pink pigment. sskA mutants showed similar but less accentuated phenotypes, and their colony borders were more regular. srrA sskA double mutants had a phenotype similar to that displayed by sskA mutants. Sexual crosses between selected srrA and sskA mutants and wild-type strains showed that the mutant phenotypes segregated in a Mendelian fashion and allowed us to obtain a set of isogenic srrA, sskA, and srrA sskA mutants in an nkuA+ background (Table 2). Similar results were obtained using nkuA– mutants derived from strain 11035 and the nkuA+ strains derived from sexual crosses. We verified the lack of srrA and sskA transcripts in srrA and sskA mutants by Northern blot analysis using strains COSsrrA03 and COSsskA02 (Table 2), respectively. The srrA and sskA transcripts are present at relatively low levels during growth and asexual development, and it does not appear that the mutation of one gene affects the mRNA accumulation of the other (data not shown).
|
|
sskA and srrA sskA mutants are more sensitive to osmotic stress than srrA mutants.
To first dissect the roles of SrrA and SskA in stress signal transduction, we tested the growth responses of srrA, sskA, and srrA sskA mutants in the presence of 1.2 M KCl (data not shown), 0.6 M NaCl, or 1.2 M sorbitol (Fig. 2). Compared to the wild-type strain, the srrA mutant showed a slight reduction in radial growth under these conditions. In contrast, growth reduction was more evident for the sskA mutant, particularly in 1.2 M sorbitol, and was even more pronounced for the srrA sskA mutant (Fig. 2). Thus, although both RRs are required for full resistance to high-osmolarity stress, SskA plays a more prominent role in this response. Moreover, it was noticed that under moderately high osmolarity (0.6 to 0.8 M NaCl or sorbitol), the srrA mutants not only grew better than the sskA mutants, but asexual sporulation was also improved under these conditions (Fig. 2; see below). These results suggest that although activation of SskA by osmotic stress might compensate for the lack of SrrA, both RRs play significant roles in the process of conidiation.
SrrA is required to survive hyperoxidant stress conditions.
Next, we tested the growth responses of srrA, sskA, and srrA sskA mutants on plates containing either H2O2, tert-butyl hydroperoxide (tb-hprx), or the superoxide-generating compound menadione (23). We found that while sskA mutants were able to resist up to 2 mM H2O2 and 0.5 mM tb-hprx, srrA and srrA sskA double mutants were not (Fig. 3A). In fact, sskA mutants were capable of growing well in the presence of 4 mM H2O2 (data not shown). In the presence of 60 µM menadione, the growth of all strains was moderately reduced. However, srrA and srrA sskA mutants were more sensitive to this compound (Fig. 3B). These results indicate that the function of SrrA is essential for the cells to cope with the oxidative stress caused by H2O2 and, to a lesser extent, that provoked by menadione.
|
srrA and sskA mutants are sensitive to cell wall stress.
As mentioned earlier, srrA colonies show irregular borders, and when observed under the microscope, hyphae often show irregular shapes (i.e., not smooth and cylindrical). This and the fact that the SrrA ortholog Skn7 regulates cell wall biosynthesis in S. cerevisiae (42) prompted us to test the sensitivities of RR mutants to the cell wall-disturbing agent calcofluor. We included a sakA mutant in this test, because SakA is part of the SskA pathway and because its ortholog Hog1 has been shown to be needed for calcofluor action (20). The results in Fig. 4A show that the srrA and srrA sskA mutants were more sensitive to calcofluor than the sskA mutant. At a higher calcofluor concentration (5.4 µg/ml), the growth of the sskA mutant was completely inhibited, whereas the sakA mutant showed only moderate sensitivity compared to that of the wild type. These results suggest that although both RRs are involved in maintaining cell wall integrity under normal osmolarity, SrrA plays a major role in this process.
SrrA and SskA mediate fungicide sensitivity downstream of a HAMP (histidine kinase, adenylyl cyclase, methyl-binding protein, phosphatase) domain histidine kinase.
It has been shown that fungicides such as fludioxonil activate the osmosensitive MAPK pathway in filamentous fungi and that strains with null mutations in this pathway are resistant to fungicides (36, 77). The A. nidulans MAPK SakA is activated by high osmolarity and oxidative stress (34) in a process mediated by SskA (17). Therefore, to dissect the roles of SrrA, SskA, and SakA in fungicide sensitivity, we tested the growth of these mutants in the presence of fludioxonil. Although single srrA, sskA, and sakA mutants were slightly less sensitive to fludioxonil than the wild-type strain, only the double srrA sskA mutant was fully resistant to the fungicide concentration tested. Thus, in A. nidulans, the RR SrrA mediates fludioxonil sensitivity independently of the SskA-SakA pathway (Fig. 4B).
While this paper was in preparation, Izumitsu et al. (30) reported that RRs ChSsk1 and ChSkn7 in Cochliobolus heterostrophus control high-osmolarity adaptation and fungicide sensitivity, under the regulation of the histidine kinase Dic1. This led us to delete the orthologous HK (see Fig. S3B in the supplemental material) encoded by the AN4479.3 gene (19) and to evaluate its role in SrrA and SskA functions. Like Dic1, AN4479.3 encodes an HK with multiple HAMP domains (6) and belongs to class III HKs according to Catlett et al. (14). During the course of this work, Hagiwara et al. (22) published a report in which they named the AN4479.3 gene nikA and analyzed exclusively its role in fungicide resistance. In consequence, we use the name NikA when referring to the HK encoded by AN4479.3. When transformant TnikA-pyrG(nkuA)9 was crossed with strain CLK43 (Table 2; see also Fig. S3B in the supplemental material), we observed Mendelian segregation of the nikA mutant phenotype (see below). From the same cross, we obtained the nikA allele in a nkuA+ background.
As shown in Fig. 4C, mutants lacking NikA show decreased radial growth compared to that of wild-type and srrA sskA strains (40% reduction in colony diameter after 3 days of growth). Notably, nikA conidiospores became darkly pigmented with time, forming a distinctive circle in the center of the colony. As in the case of srrA sskA mutants, the growth of nikA mutants was barely affected by the concentration of fludioxonil tested. These results confirm those of Izumitsu et al. (30) and Hagiwara et al. (22), indicating that in different (but not all) fungi, a HAMP domain HK is required to relay fungicide signals to RRs such as SrrA and SskA.
In light of these results, it became important to analyze the responses of nikA mutants to high osmolarity, oxidative stress, and cell wall stress conditions. Somewhat unexpectedly, nikA mutants were not hypersensitive to high osmolarity. Accordingly, a similar reduction in colony diameter was observed when the wild-type and nikA strains were grown under high osmolarity. Furthermore, the nikA mutant grew better than the srrA sskA mutant under this condition (Fig. 4C). This indicates the involvement of NikA-independent pathways in high-osmolarity adaptation in A. nidulans. The results in Fig. 4C also show that in contrast to the srrA and double srrA sskA mutants, nikA mutants were resistant to 4 mM H2O2, as well as to 0.5 mM tb-hprx (data not shown). Likewise, in contrast to srrA, sskA, and srrA sskA mutants, all of which were sensitive to 5.4 µg/ml of calcofluor (Fig. 4A), nikA mutants were able to grow at this calcofluor concentration. This suggests that NikA does not act upstream of SrrA in response to oxidative stress signals and that this HK is not required for SrrA and SskA functions in the biosynthesis of putative cell wall components needed for calcofluor resistance.
Taken together, our results indicate that NikA is required to transmit fungicide signals to the RRs SrrA and SskA but is partially dispensable for transmission of osmotic and oxidative stress signals.
SrrA, SskA, and NikA are required for normal asexual sporulation and conidiospore viability.
As seen in Fig. 2 to 4, sporulating colonies from srrA and sskA mutants show reductions in conidiophore density and conidiospore pigmentation. In addition, we found that as with sakA mutants (34), srrA and sskA conidiospores lost their viability during storage. These observations led us to examine in more detail the conidiation process, as well as conidiospore viability, for these mutants. Since it was evident that srrA and sskA sporulation defects were partially suppressed under high-osmolarity and low-glucose conditions, respectively (see Fig. S4 in the supplemental material), we decided to quantify the number of spores formed under these conditions. The results in Fig. 5A show that compared to the wild-type strain, srrA and sskA mutants produce very low numbers of spores in MM. This sporulation defect was slightly more severe in srrA and srrA sskA mutants. Consistent with our previous observations, we found that sporulation of the srrA but not the sskA mutant was greatly improved at high sorbitol concentrations. In contrast, sskA but not srrA sporulation was notably improved at low glucose concentrations. These results demonstrate that both RRs regulate sporulation in A. nidulans, and they suggest that SskA stimulation by high osmolarity results in increased sporulation, while low glucose stimulates sporulation in a SrrA-dependent fashion. Unexpectedly, the sporulation of the srrA sskA double mutant was improved under both high-sorbitol and low-glucose conditions (Fig. 5A), suggesting that in the absence of both RRs, other sporulation pathways might become activated.
|
srrA and sskA spores was drastically reduced after only 3 to 6 days at 4°C. Furthermore, it was found that the viability of srrA spores produced under high-sorbitol conditions was greatly improved. In contrast, the viability of sskA spores produced under low-glucose conditions remained low. These results indicate that both RRs are required for full sporulation as well as for full spore viability and that SskA might replace SrrA functions in these two processes.
After we found that the HK NikA acts upstream of SrrA and SskA in fungicide signaling, it was interesting to evaluate its role in conidiation and conidiospore viability. We found that the nikA mutant produced 30 times fewer conidiospores per area than the wild-type strain. In comparison, sskA mutants produced 500 times fewer conidia than the wild type (Fig. 5A), and thus, the nikA conidiation defect was not as marked as that observed for the srrA or sskA mutant. In contrast, nikA conidiospores showed a decrease in viability that was similar to that observed for sskA spores (Fig. 5B and C). These results suggest that, acting through SskA or SrrA, NikA and other HKs appear to contribute to the sporulation process. In contrast, NikA appears to play a major role in regulating spore viability, possibly through the RR SskA.
SrrA regulates expression of the catalase gene catB.
We reported previously that the viability of sakA conidiospores decreased upon storage and that the activity of the spore-specific catalase CatA (48) was lower in the sakA mutant than in the wild type (34). Since SskA acts upstream of SakA and since sskA and srrA conidiospores show decreased viability (Fig. 5B), we analyzed catalase activity levels in these mutants to determine if there was a correlation between catalase activity, spore viability, and H2O2 resistance. The results show that the CatA activity levels of srrA and sskA conidiospores were comparable to those found for wild-type spores. CatA activity levels of srrA sskA and sakA conidiospores were lower (Fig. 6, left panel). These results indicate that SrrA and SskA are individually dispensable for CatA accumulation in conidia, and thus, CatA activity cannot explain the observed decreases in viability or H2O2 resistance of srrA and sskA mutants. However, for A. nidulans, catalase CatB constitutes the major inducible catalase activity in mycelia (33, 35). Therefore, we examined SrrA and SskA functions in CatB induction. The results show that H2O2 induces high levels of CatB activity in both the wild-type strain and the sskA mutant. In contrast, no induction of CatB activity was observed for the srrA mutant, indicating a critical role of SrrA in CatB induction by H2O2. In addition, it was noted that CatB induction was lower in the sakA mutant than in the wild-type strain. This is consistent with previous results and indicates that CatB induction seems partially dependent on the MAPK SakA, which in turn suggests some interaction between SrrA and SakA in catB transcriptional regulation (Fig. 6, right panel). Taken together, our results indicate that the H2O2 sensitivity of srrA mutants is related to their failure to induce CatB. Consistent with this, sskA mutants are able to resist H2O2 (Fig. 3) and to induce CatB (Fig. 6).
|
srrA and sskA mutants that were induced to conidiate. As reported previously, the two brlA transcripts (60) are not present before induction of conidiation and then gradually accumulate during wild-type conidiation. In the srrA mutant, brlA mRNA started to accumulate at the same time but then failed to reach wild-type levels. A similar result was observed for the sskA mutant, except that in this case brlA mRNA accumulation is delayed by 6 h relative to that for the wild type (Fig. 7). As in other cases (66), brlA mRNA appears somewhat diffused. However, probing of the same blot with the argB gene, whose expression remains relatively constant during conidiation, indicates that this is not due to general RNA degradation. In any case, brlA mRNA accumulation levels appear decreased in srrA and sskA mutants, suggesting that this might be related to the sporulation defects observed for these mutants.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ypdA allele could be maintained only in heterokaryons. Notably, the growth of ypdA–/ypdA+ heterokaryons, which appear to contain a higher proportion of ypdA+ nuclei, was reduced or eliminated under osmotic but not oxidative stress conditions (Fig. 1A to D). Although the interpretation of these results is complicated by the inherent genetic heterogeneity of heterokaryons, these results suggest that YpdA plays a more important role in osmoadaptation than in adaptation to oxidative stress. The fact that mutants lacking YpdA are very likely unviable contrasts with the fact that mutants lacking both RRs (srrA sskA) are viable, although clearly affected in stress signal transduction and asexual sporulation. This argues against a model in which YpdA simply exerts positive effects over both RRs and these in turn play positive roles for downstream components. By analogy with S. cerevisiae, in which ypd1 mutation results in lethal hyperactivation of the Hog1 MAPK (58), we propose that the lethality associated with YpdA elimination results from hyperactivation of SakA and/or MpkC, a second stress MAPK present in the aspergilli (17, 34), whose inactivation does not produce a clear phenotype in A. nidulans (K. Jahng, personal communication) (17). However, MpkC functionality is suggested by the facts that in Aspergillus fumigatus, mpkC deletion results in the inability to use polyalcohol sugars as a sole carbon source and that mpkC mRNA levels increase during oxidative stress (61). That constitutive activation of SakA and/or MpkC can lead to lethality is also indicated by the fact that the fungicide fludioxonil seems to kill A. nidulans (Fig. 4B and C) and other fungi through activation of the stress MAPK pathway (36, 77) and, as shown here, through the activity of the RR SrrA.
SrrA and SskA roles in osmotic and fungicide stress signal transduction.
We found that both SskA and SrrA are involved in osmotic stress signal transduction. Although the growth of the wild type and srrA mutants is similar under moderately high osmolarity (0.6 M sorbitol or KCl), the role of SrrA in osmotic stress resistance is inferred from the fact that srrA sskA double mutants are more sensitive to osmostress than single sskA mutants. On the other hand, the fact that sskA mutants are more sensitive to osmotic stress than sakA mutants (34) suggests that SskA activates SakA and MpkC and that both MAPKs might contribute to osmostress resistance. This is supported by the fact that PbsB, a MAPK kinase acting downstream of SskA, is required for phosphorylation of both SakA and MpkC (17). Testing of this simple hypothesis has been prevented by an apparent lethality of double sakA mpkC mutants (4, 17). If confirmed, this hypothesis will imply that A. nidulans uses two MAPKs as well as the RR SrrA to deal with high-osmolarity stress.
A finding in this work that might have practical consequences in terms of fungicide resistance mechanisms is the fact that both SrrA and SskA are required for sensitivity to the widely used fungicide fludioxonil. There is strong evidence showing that the osmosensing MAPK kinase pathway mediates the killing effects of this broad-spectrum fungicide. Indeed, Neurospora crassa mutants with null mutations of the MAPK OS-2 (77), the RR RRG-1 (31), or the HK OS-1 (52) are both osmosensitive and resistant to fludioxonil. This is in contrast to our results, which show that a lack of SakA, the OS-2 ortholog, does not result in complete resistance to fludioxonil and that, in fact, the RR SrrA mediates fludioxonil sensitivity independently of the SskA-SakA pathway (Fig. 4). We found that like os-2 mutants, nikA mutants are resistant to fludioxonil. This indicates that out of the 15 predicted histidine kinases in A. nidulans, NikA plays a major role in relaying fungicide stress signals to SrrA and SskA. A similar fungicide sensitivity function has recently been found for NikA orthologs BOS1 (72) and Dic1 (30). However, some of these HK orthologs seem to have species-specific functions. For example, BOS1 inactivation in Botrytis cinerea results in lack of macroconidiation and plant virulence (72). In A. nidulans, nikA but not srrA sskA mutants show a clear reduction in radial growth, suggesting that NikA serves functions that are not mediated by SrrA and/or SskA. In addition, NikA is involved in conidiation and is required for spore viability (Fig. 5C).
We also found that the role of NikA in the osmoadaptation of A. nidulans does not appear as important as in other fungi such as N. crassa. While os-2-null mutants are hypersensitive to osmostress, nikA mutants show partial or no sensitivity to this stress. The fact that srrA sskA mutants are more sensitive to osmostress than nikA mutants suggests that another HK(s) is able to transmit osmostress signals to SrrA and/or SskA in A. nidulans. TcsB, the ortholog of the yeast HK Sln1, is a likely candidate to mediate osmostress signals in different fungi. Although TcsB inactivation did not produce a visible phenotype (18), it would be interesting to evaluate the osmostress response of double nikA tcsB mutants.
SrrA, SskA, and NikA roles in oxidative stress signal transduction and calcofluor sensitivity.
We found that SrrA but not SskA is required for H2O2 resistance and, consistent with this, for the induction of catalase CatB in response to H2O2. This is coherent with the roles that the SrrA orthologs Skn7/Pos1 and Prr1 play in antioxidant responses in S. cerevisiae and S. pombe, respectively. However, the disruption of skn7 in Cryptococcus neoformans did not affect H2O2 sensitivity (7), suggesting that the antioxidant response through this RR might not be conserved in all fungi. Notably, the Skn7 function in oxidative stress response does not require the conserved phosphoaccepting aspartate present in its receiver domain (41, 46). The fact that nikA mutants are resistant to H2O2 indicates that either another HK(s) is responsible for transmitting oxidative stress signals to SrrA or the SrrA function in the antioxidant response does not involve phosphorelay components. Likewise, the resistance of nikA mutants to calcofluor suggests that another HK(s) can regulate SrrA and/or SskA functions in cell wall integrity.
Since SskA is required for transmission of oxidative stress signals to the MAPK SakA, it was unexpected that sskA mutants were not sensitive to H2O2 and were able to induce CatB in response to H2O2. However, a role for the SskA-SakA pathway in the oxidative stress response cannot be ruled out. In fact, H2O2-mediated induction of a catB::lacZ reporter gene fusion is reduced in a sakA background (32), and CatA activity levels are reduced in sakA conidiospores (34) (Fig. 6).
The roles of SrrA, SskA, and NikA in asexual development and conidiospore viability.
We have shown that two RRs and one HK differentially contribute to the development of asexual spores as well as to the viability of these spores. Indeed, the inactivation of srrA, sskA, or nikA led to the production of fewer spores that showed decreased viability upon storage. Two observations indicate that, although related, spore formation and spore viability are separable processes. First, sakA and wild-type strains produce similar amounts of spores, but only the sakA spores lose their viability (34). Second, the sporulation of sskA mutants is greatly improved under low-glucose conditions, but the spores produced maintain their decreased viability. In contrast, under high-osmolarity conditions, srrA mutants clearly improve both sporulation and conidiospore viability (Fig. 5A and B). The fact that srrA and sskA mutants are sensitive to calcofluor (Fig. 4A) suggests that their cell wall biosynthesis is affected, which in turn could be related to cell wall-defective spores and decreased viability. However, sakA mutants are rather resistant to calcofluor (Fig. 4A), yet sakA spores show decreased viability (34).
The results showing that srrA and sskA sporulation defects are clearly more drastic than those seen in nikA mutants suggest that SrrA and/or SskA functions in sporulation might be regulated by other upstream HKs. This is supported by the facts that inactivation of the HK TcsA resulted in reduced conidiation and that this defect was remedied under high osmolarity (73).
While we do not know why spore viability decreases in mutants lacking SakA, SrrA, SskA, or NikA, the decreased sporulation observed for srrA and sskA mutants might be related to their lower levels of brlA mRNA during sporulation (Fig. 7). BrlA is a rate-limiting step function during conidiospore formation, required not only for initiation of conidiation but throughout the entire process. Because of the close link between decreased brlA mRNA levels and decreased sporulation, it is possible that SrrA and SskA functions affect brlA mRNA accumulation and that this in turn affects the initiation and/or maintenance of the conidiation process. Reduced sporulation has been related to reduced brlA expression in several "fluffy" developmental mutants (2, 39, 65, 67, 74). However, srrA and sskA mutants might represent a different class of sporulation mutants. First, srrA and sskA mutants do not show a "fluffy" phenotype, characterized by the proliferation of aerial hyphae. Second, srrA and sskA mutants are able to induce conidiation of fluG and tmpA "fluffy" mutants when grown next to them (data not shown). This suggests that srrA and sskA mutants are not defective in the production of the putative sporulation signals produced by FluG (39, 65) and TmpA (67).
Mutants with null mutations of the devR gene show conidiation defects similar to those found for srrA mutants, and these defects are remedied by high osmolarity (71). devR encodes a helix-loop-helix transcription factor proposed to be a downstream target of the phosphorelay system. This is because a DevR-green fluorescent protein fusion failed to show nuclear localization in a mutant in which the HK gene tcsA was affected (71, 73). We found that devR mutants are not as sensitive to H2O2 as srrA mutants (data not shown), suggesting that if DevR acts downstream of SrrA, it mediates sporulation functions. The role of DevR in conidiospore viability has not been evaluated. How conserved the SrrA function in regulating sporulation is among fungi has yet to be determined. Recently, it was reported that A. fumigatus mutants lacking the orthologous RR AfSkn7 are sensitive to oxidative stress but do not show growth or conidiation phenotypes, even under high-osmolarity conditions (38).
The function of SskA in spore viability is consistent with its function as an upstream component of the SakA pathway (17). Indeed, SakA is phosphorylated shortly after induction of asexual development, and its inactivation results in spores with decreased viability (34). The fact that sakA conidia contained lower levels of the spore-specific catalase CatA suggested a relationship between CatA content and spore viability. However, we found here that srrA, sskA, and wild-type spores contain comparable levels of CatA activity.
While this work was under review, Hagiwara and coworkers (21) reported the characterization of srrA, sskA, and srrA sskA mutants with regard to their resistance to osmotic and oxidative stress, conidiospore viability, and the mRNA accumulation of several genes, including catalase genes. Some of the results reported are consistent with ours, but others show important differences. First, Hagiwara and coworkers report that both srrA and sskA mutants are sensitive to H2O2; we find that only srrA mutants are sensitive to H2O2, within a 1 to 4 mM range. Second, the cited work shows that the srrA and sskA genes are both required for catB gene expression. By directly measuring catalase activity, we find that only srrA is fully required for CatB induction. Third, the report of Hagiwara et al. shows that only sskA conidiospores lose viability; we show that both srrA and sskA conidiospores show decreased viability. Fourth, Hagiwara and colleagues report that catA gene expression is downregulated in srrA and sskA conidiospores. By directly measuring catalase activity, we find that neither srrA nor sskA conidiospores show lower CatA activity. It seems unlikely that these differences are due to strain variations, because we have confirmed our results using strains with different genetic backgrounds. Although currently we do not have an explanation for these discrepancies, it will be important to consider them in designing future experiments.
Our results highlight important differences in the ways different fungi perceive and respond to stress signals and in the ways in which phosphorelay systems are connected to developmental processes. Taken together, these results support a model in which the A. nidulans HK NikA transmits fungicide stress signaling to the SskA-SakA pathway, as well as to the RR SrrA, resulting in cell death. NikA might also transmit osmostress signals to SrrA and/or SskA, but other HKs seem important for osmoadaptation. We propose that NikA is also involved in transmitting unknown signals to contribute to asexual sporulation and to determine conidiospore viability, probably via SskA. In addition, NikA appears to have SrrA/SskA-independent functions in the regulation of radial growth.
In our model, SrrA plays a major role in the oxidative stress response independently of NikA and perhaps independently of any other phosphorelay component. Both RRs are required for normal asexual sporulation and for maintenance of conidiospore viability, perhaps through regulation of brlA gene mRNA levels. Finally, SrrA and SskA are very likely involved in cell wall biosynthesis.
|
ACKNOWLEDGMENTS |
|---|
We are grateful to Anet Rivera for experimental support, to Fernando Lara for nkuA primer design, and to the IFC-UNAM molecular biology unit for DNA synthesis and sequencing. We thank Michael Hynes and Reinhard Fischer for providing very useful strains.
|
FOOTNOTES |
|---|
Published ahead of print on 13 July 2007.
Supplemental material for this article may be found at http://ec.asm.org/.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
