Expression of chsE encoding one of the five chitin synthases of Aspergillus nidulans was analyzed. Expression of chsE was moderate in conidiophores, but somewhat weaker in vegetative mycelia. During sexual development, chsE was expressed strongly in young cleistothecia and hülle cells, but little in mature sexual structures. Deletion of chsE caused a significant decrease in the chitin content of the cell wall during early sexual development. Expression of chsE was increased by substituting glucose with lactose or by addition of 0.6 M KCl or NaCl, but affected little by substituting glucose with sodium acetate. Consequently, chsE was shown to have a mode of expression distinct from those of the other chitin synthase genes, chsA, chsB and chsC.
Chitin, a β-1,4-linked polysaccharide of N-acetylglucosamine, is one of the major structural components of the fungal cell wall, and its biosynthesis is thought to be important for many fungi to maintain the physical strength of cell walls as well as to form their specific shape during cell growth and differentiation [1–5]. Chitin synthases (EC 126.96.36.199) are integral membrane proteins which catalyze polymerization of N-acetylglucosamine from UDP-N-acetylglucosamine. The primary structures of many fungal chitin synthases are mainly deduced from their gene structures, and they have been divided into five groups, classes I–V, on the basis of their structures in the conserved region.
Five chitin synthase genes, chsA, chsB[2,6], chsC, chsE (identical to chsD reported by Motoyama et al.) [3,4], and csmA/chsD[3,8,9] corresponding to classes II, III, I, IV, and V, respectively, have been isolated and identified in the filamentous fungus Aspergillus nidulans. Disruption of chsA, chsC, or chsE (CAL1 homolog) [3,4] does not cause any defect in cell growth or morphology during the asexual cycle, supporting the conclusion that none of the three genes is essential for hyphal growth. Double disruption of chsA and chsE, however, causes a remarkable decrease in the efficiency of conidia formation, indicating that chsA and chsE serve redundant functions in conidia formation [4,10]. Mutant strains in which both chsA and chsE are disrupted exhibit loss of integrity of hyphal wall and remarkable abnormalities during its asexual development . It has also suggested that the functional importance of chsE increases when the expression of chsB is limited . chsA is mainly expressed in the metulae, phialides, and conidia, whereas chsC is expressed in hyphae as well as conidiophores, which implies that ChsA and ChsC share critical functions in hyphal wall integrity and differentiation . It has been thus revealed that chsA, chsC, and chsE perform redundant functions in the process of conidiation. The disruptant of chsB grows as minute colonies without conidia, and produces hyphae with enlarged tips, a high degree of branching, and disorganized lateral walls, which is not remedied by osmotic stabilizers. However, its mycelium is not deficient in chitin content and shows no evidence of lysis. It is thus probable that chitin synthesized by the ChsB enzyme does not substantially contribute to the rigidity of the cell wall but is necessary for normal hyphal growth and organization . csmA, a part of which was cloned and designated as chsD by Specht et al. , encodes a novel protein (1852 amino acids, 206 kDa) consisting of a C-terminal class V chitin synthase domain and an extra N-terminal myosin motor-like domain . The csmA null mutants show sensitivity to a chitin-binding reagent, Calcofluor white or Congo red, and morphological abnormalities in tip growth and septum formation . The CsmA enzyme has important roles in polarized cell wall synthesis and maintenance of cell wall integrity, and its myosin motor-like domain is indispensable for these functions [12,13].
As an effort to understand the role of the chitin synthase genes in A. nidulans, we have analyzed the expressions of chsA, chsB and chsC. While chsB is expressed ubiquitously throughout the fungal body and quite independently of the change in developmental status of the cells, chsA is expressed specifically during asexual differentiation. The expression of chsC is moderate in young vegetative mycelia and in sexual structures, i.e., such as young cleistothecia and mature ascospores, but weak in old vegetative mycelia and in asexual structures. In the present study, we analyzed the mode of expression of the chitin synthase gene, chsE, both by Northern blotting and by a vital reporter system with sgfp encoding a modified version of green fluorescent protein, sGFP . Herein, we present some evidence indicating that chsE has a distinctive mode of expression from those of the other chitin synthase genes of A. nidulans previously reported, chsA, chsB and chsC, and also that the gene is subject to differential expression in response to developmental status and environmental factors.
2Materials and methods
2.1Strains, media, cultivation and transformation
Aspergillus nidulans FGSC A26 (biA1) (Fungal Genetics Stock Center; Kansas City, KS, USA) was used for RNA preparation, and A. nidulans W × 24 (npgA1 biA1; sB3; chaA1 trpC801)  was used as a recipient strain for transformation. A. nidulans D3–2 (biA1 pyrG89; pyroA4; wA3; argB2; chsE::argB)  and ABPU/A2 (biA1 pyrG89; pyroA4; wA3), which were kindly provided by Dr. Horiuchi, were used for analysis of chitin content. FGSC A26 and the transformants from W × 24 were maintained on Aspergillus complete medium (CM) , W × 24 on CMW (CM + 4 mM tryptophan) medium, and D3-2 and ABPU/A2 on CMU (CM + 10 mM uridine + 10 mM uracil) medium.
Transformation of A. nidulans W × 24 was performed as described by Lee et al. , and the protoplasts of transformants were regenerated on 1% glucose minimal medium (MM)  supplemented with 0.1 μM biotin, 4 mM methionine, and 0.6 M KCl (MBMK). When necessary, the transformants were grown on MM supplemented with 0.1 μM biotin and 4 mM methionine (MBM), and W × 24 on MBM supplemented with 4 mM tryptophan (MBMW).
Preparation of vegetative mycelia and induction of asexual and sexual differentiation was performed as described by Lee et al. .
2.2Construction of plasmids
The vector pTchsE-p::sgfp used for analyzing the expression of chsE was constructed as follows. First, the presumptive promoter region of chsE (from nucleotide −1105 to −1) was amplified from the genomic DNA of A. nidulans by PCR using two primers, PE1 5′-GGTACCATGCGGTTGGTTGATCAG-3′ and PE2 5′-AGAATTCGATAACTATTCAATAGGCGAA-3′[3,4] (Fig. 1A). The resulting 1.1-kb PCR product was cloned into pT7Blue(R) vector to yield pT7/chsE-11. Then the 1.1-kb Kpn I–Eco RI fragment was excised from pT7/chsE-11 and inserted into Kpn I–Eco RI-digested pTsgfp vector (Fig. 1B)  upstream of sgfp to yield pTchsE-p::sgfp (6.5 kb).
DNAs from A. nidulans strains were prepared by using miniprep procedures . Southern blot analysis was performed according to the standard method  by using an ECL labeling and detection kit (Amersham; Buckinghamshire, UK). To confirm the correct integration of the vectors at the trpC loci of the transformants, a 0.4-kb Sal I-digested trpC fragment (from nucleotide −266 to 20) was used as a probe.
Total RNAs were prepared from the liquid nitrogen-frozen and ground mycelia of A. nidulans at the vegetative growth, asexual development, or sexual development stages by modified guanidine thiocyanate/CsCl density gradient ultracentrifugation . Poly(A)+ RNAs were isolated by an Oligotex mRNA purification kit (Quiagen; Valencia, CA, USA) according to the manufacturer's instructions. Northern hybridization was performed according to standard procedures , in which the DNA probes were labeled with 32P using a Random Primed DNA Labeling Kit (Boehringer Mannheim; Mannheim, Germany). A 1.1-kb chsE-specific probe for Northern blot analysis was prepared from gel-purified RT-PCR fragments amplified by using the primers, PEN1 (5′GACCTGCGATCCAGATGATT3′) and PEN2 (5′CCAACAGCACTCTTGACGAA3′). The nucleotide sequences of the cloned DNAs were determined by using an automatic DNA sequencer ABI 377 (Perkin–Elmer).
Confocal microscopic observations of A. nidulans cells were performed using a confocal microscopy system (TCS SP; Leica; Heerbrugg, Switzerland). To observe sGFP fluorescence, fungal cells were excited with the wavelength of 488 nm, and the fluorescent emission with the wavelength of 500–520 nm was detected. All sGFP images were generated under standardized conditions in which the main parameters are adjusted as follows: PMT gain, 690–710; PMT offset, −10; Pinhole, 150; number of sections, 8; number of scans for each section, 8. The intensity of green fluorescence developed in the confocal micrographs was analyzed by using an image analysis system (GAIA Material V188.8.131.52; Mirero Inc., Seoul, Korea).
2.5Determination of chitin content
One milligram of lyophilized powder of SDS-extracted cell walls was hydrolyzed in 0.2 ml of 6 N HCl (H-0636; Sigma; St. Louis, MO, USA) at 100°C for 17 h, together with 0–200°C of N-acetylglucosamine standards. After removing the HCl by using a speed vacuum concentrator (Savant; Ramsey, MN, USA) at 50°C, dried samples were resuspended in 1 ml of water and centrifuged to remove insoluble material. To 0.1 ml of sample, 0.1 ml of solution A [1.5 M Na2CO3 in 4% (w/v) acetylacetone] was added and the mixture was incubated at 100°C for 20 min. After cooling to room temperature, 0.7 ml of 96% ethanol and 0.1 ml of solution B (1.6 g p-dimethylaminobenzaldehyde in 30 ml concentrated HCl and 30 ml 96% ethanol) was added and incubated for 1 h at room temperature. Triplicate samples were incubated for 1 h at room temperature before the absorbance was read at 520 nm [21,22].
3.1Construction of A. nidulans transformants
To analyze the expression of chsE using sgfp cDNA as a vital reporter, pTsgfp and pTchsE-p::sgfp plasmids (Fig. 1B and D) were individually introduced into A. nidulans W × 24. Transformants which grew in the absence of tryptophan were expected to carry the wild-type trpC gene which was formed by homologous recombination of trpCΔ176 (5′ 1.8-kb segment of trpC) of the transforming vectors with the chromosomal trpC801 allele. To confirm correct integration of the vectors at the trpC locus of the transformants isolated on MBM plates, chromosomal DNAs from one of the transformants and the host strain were digested with Eco RI and analyzed by Southern blotting using a 0.4-kb Sal I-digested trpC fragment (from nucleotide −366 to +19) as a probe (Fig. 1C). The DNA from A. nidulans W × 24 showed a single 2.4-kb fragment hybridized with the probe. On the other hand, the transformants exhibited differently sized DNA fragments hybridized with the trpC probe: specifically, the pTsgfp transformant showed 3.2- and 4.5-kb fragments; and the pTchsE-p::sgfp transformant 3.2-, 5.6-, and 6.5-kb fragments. While the two positive fragments from the pTsgfp transformant showed similar intensity, the 6.5-kb fragment of the pTchsE-p::sgfp transformant showed intensity about twice as high as the other two. Thus, the integration of the targeting vectors at trpC loci were estimated to be single in the pTsgfp transformant and tandem triple in the pTchsE-p::sgfp transformant (Fig. 1D).
3.2Differential expression of the chsE gene in response to developmental status
To figure out the mode of differential expression of chsE gene during vegetative growth, asexual differentiation, and sexual differentiation, the level of sGFP in mycelia and differentiated structures of the pTchsE-p::sgfp transformant was monitored by confocal microscopy in comparison with the host strain W × 24 and the pTsgfp transformant.
When slide-cultured on minimal media which supported vegetative growth and asexual development, the host strain W × 24 and the pTsgfp transformant did not show any detectable level of sGFP fluorescence in any part of substrate mycelia or asexual structures (data not shown). Similarly, sexually differentiated structures, such as hülle cells and cleistothecia, of the two strains did not show any sGFP fluorescence (data not shown). These results suggest that autofluorescence of the fungal cells does not affect the reliability of the sGFP reporter system.
In the pTchsE-p::sgfp transformant slide-cultured on 1% glucose MBM, sGFP fluorescence was visible in most part of conidiophores and in conidia (Fig. 2, panels a–d). The intensity of sGFP fluorescence seemed to increase in accordance with the progress of asexual development, and to be stronger in vesicles, metulae and phialides than in other parts of conidiophores. However, a lower level of sGFP fluorescence was observed in vegetative hyphae and foot cells throughout the culture period.
sGFP fluorescence was also examined during sexual development, which was divided into three arbitrary stages: Stage I when hülle cells and young cleistothecia were formed (0–2 days after induction), Stage II when cleistothecia enlarged and ascospore formation occurred (3–5 days after induction), and Stage III when mature asci were liberated from the spontaneously rupturing cleistothecia (6–8 days after induction).
At the early stages of sexual development (Stage I), the pTchsE-p::sgfp transformant showed quite strong sGFP fluorescence both in young cleistothecia and hülle cells (Fig. 2, panels e and f). As the process of sexual development proceeded (Stage II; Fig. 2, panels g–j), the sGFP fluorescence in the cleistothecia declined strikingly, and thus only a faint fluorescent was visible in the cleistothecial shell. Furthermore, we could not find any sign of fluorescence in the cleistothecial contents, i.e., asci and ascospores. Accordingly, it is probable that the sGFP fluorescence was absent in the content of young cleistothecia at Stage I. At the late stages of sexual development (Stage III; Fig. 2, panels g–j), no sGFP fluorescence was observed in any part of the sexual structures.
The level of chsE transcript in the cells of three different developmental status was monitored by Northern blot analysis using the chsE-specific probe (Fig. 3). While only a little chsE transcript was detected in young vegetative mycelia grown for 9 or 12 h, quite a higher level of the transcript was observed in older ones grown for 15 h. During asexual development, the level of chsE transcript was roughly as high as that of later vegetative stage and remained almost constant. These results are considerably accordant with the result of confocal microscopic observation of the pTchsE-p::sgfp transformant.
During sexual development, Northern blot analysis was performed only for the cells of Stage I because we could not obtain adequate amounts of RNAs at the later stages. The expression of actA, a member of house-keeping genes and used as an internal control, was not constant but fluctuated during sexual development as reported previously . The level of chsE transcript reached to the highest point at the very early stage of sexual differentiation (10 h after unsealing) and rapidly decreased thereafter. Thus, the amount of chsE transcript was almost negligible at the later moments of Stage I.
The results of confocal microscopy and Northern blot analysis in sum indicate that chsE is expressed considerably during the later stage of vegetative growth and throughout the process of asexual differentiation, and thus that it contributes to the synthesis of chitin contained in the cell walls of vegetative mycelia, conidiophores and conidia. They also show that the expression of chsE during sexual development is not only temporally specific to the early stage (Stage I) but also spatially differentiated, i.e., strong in young cleistothecial shells and in hülle cells, but negligible in other parts of the sexual structures, such as asci and ascospores.
Chitin contents in the cells of chsE deletion mutant (D3–2) of three different developmental stages, i.e., vegetative growth, asexual development, and sexual development, were analyzed in comparison with those of relevant wild-type strain (ABPU/A2) (Table 1). During vegetative growth and asexual development, the chitin contents in the cell walls of the mutant was about 90% of those of wild-type strain, respectively. On the other hand, the chitin content of the mutant cells decreased to ? 65% of that of wild-type strain at the early stage of sexual differentiation (Stage I). The effect of chsE deletion on the chitin synthesis thus seems to be much more significant during early sexual development than during either vegetative growth or asexual development.
Table 1. Chitin contents of the wild-type and chsE mutant
aMeans ± SD were calculated from the results of triplicate samples. The ratio of chitin content of the mutant to that the wild-type strain is shown in parentheses.
bFor preparation of vegetative mycelia, conidia were inoculated in CMU broth and grown at 37°C for 15 h.
cAsexual differentiation was induced by spreading vegetative mycelia onto agar plates and incubation at 37°C for 12 h.
dFor induction of sexual differentiation, vegetative mycelia were transferred onto agar plates, incubated for another 24 h while the plates were sealed closely, and incubated thereafter under unsealed conditions for 24 h (Stage I).
281 (100) ± 35
276 (100) ± 14
359 (100) ± 10
255 (91) ± 28
239 (87) ± 16
233 (65) ± 17
3.3Effect of osmostress on the expression of chsE gene
The effect of osmostress on the expression of chsE gene was analyzed by observing sGFP fluorescence in the cells of the pTchsE-p::sgfp transformant slide-cultured on 1% glucose MBM plates containing 1.2 M KCl (or NaCl). The transformant showed increased levels of sGFP fluorescence throughout the whole fungal body including substrate mycelia, conidiophores, and conida in the presence of KCl (Fig. 4A, panels c and d) or NaCl (data not shown). Image analysis of the confocal micrographs showed ?54%, 66%, and 52% increment of green fluorescence in conidiophores, in substrate mycelia, and in the whole thallus, respectively (Fig. 4B). This result thus suggests that osmostress stimulates the expression of chsE in the whole fungal body.
3.4Effect of carbon source on the expression of chsE gene
The effect of three different carbon sources, i.e., glucose, lactose, and sodium acetate, on the expression of chsE gene was examined by monitoring sGFP fluorescence in the cells of the pTchsE-p::sgfp transformant slide-cultured on MBM plates containing 1% of each carbon source.
The transformant showed higher levels of sGFP fluorescence on lactose than on glucose both in substrate mycelia and asexually differentiated structures, i.e., conidiophores and conidia (Fig. 4A, panels e and f). Image analysis of the confocal micrographs showed ?44%, 69%, and 38% increment of green fluorescence in conidiophores, in substrate mycelia, and in the whole thallus, respectively. Quite differently from the case of lactose, the transformant showed similar sGFP intensity on sodium acetate as on glucose (Fig. 4A, panels g and h). These results suggest that the expression of chsE is increased to some extent by substituting lactose for glucose as a sole carbon source, however, affected little by sodium acetate.
In the present study, we analyzed the expression mode of chsE in A. nidulans and found that this gene is subject to differential expression in response to developmental status and environmental stress in quite a distinctive manner from any other chs genes of this fungus previously reported .
As presented above, chsE was expressed at a considerable level during asexual differentiation and at a lower level during vegetative growth. During sexual development, the expression of chsE was quite strong in young cleistothecial shells and hülle cells, but was negligible in the asci and ascospores (Fig. 2).
The expression of chsB gene which is essential for cell growth [2,6] occurs strongly and ubiquitously throughout the whole fungal body, and is relatively independent of the change of developmental status of fungal cells . Thus, the manner of chsE expression is similar to that of chsB during vegetative growth and asexual differentiation, however, is clearly distinguishable from that of chsB during sexual development. The functional relationship between chsE and chsB has been analyzed by using double mutant strains of chsB and chsE, in which chsB was placed under the control of alcA promoter . The significance of chsE gene to the growth of mycelia under high-osmolarity conditions as well as to the process of conidiation increases when the expression of chsB is limited.
It has been well established by intensive mutation analysis that chsA, chsC, and chsE perform redundant functions in the process of conidiation [3–7,10]. However, differential expression of chsA and chsC, especially during sexual development, has been revealed only recently by Lee et al. . chsA is specifically expressed in conidiophores and conidia during asexual differentiation, but is not essential for either sexual differentiation or vegetative growth. chsE is thus clearly distinct from chsA in that it was expressed during sexual differentiation as well as vegetative growth. chsC is expressed moderately during the early phase of vegetative growth, but weakly in old vegetative mycelia and in asexual structures. It is also expressed during sexual development, i.e., relatively strongly in mature ascospores and young cleistothecia, but negligibly in hülle cells. Therefore, chsE can be discriminated also from chsC in that it was expressed more in asexual structures than in vegetative mycelia, and furthermore in that it was not expressed in asci and ascospores during sexual development.
Deletion of chsE affected the chitin content of cell wall quite significantly during early sexual development (Table 1). Although chsE was expressed during asexual differentiation and vegetative growth as well as early sexual differentiation, its deletion caused only a little difference in chitin content of cell wall during either vegetative growth or asexual development. Thus, it is suggested that the role of chsE in chitin synthesis cannot be substituted by any other chs genes during early sexual development while it can be during vegetative growth and asexual development.
It is of interest that expression of chsE reached its highest level on media containing lactose, rather than glucose or acetate, as sole carbon source (Fig. 4). On the contrary, the expression of chsA, chsB, and chsC, are stimulated by acetate which was supplied as a sole carbon source but not by lactose . It is thus suggested that the mode of regulation of chsE expression according to the carbon source is quite different from the other chitin synthase genes. In A. nidulans, genes encoding the enzymes committed to the catabolism of less preferred carbon sources , those involved in gluconeogenesis and glyoxylate cycle [25,26], and those involved in secondary metabolite synthesis  are subject to regulation by carbon sources. However, there have been there have been only limited reports of effect of carbon source on expression of chitin synthase genes .
We found in the present study that osmostress caused by high concentrations (up to 1.2 M) of KCl or NaCl stimulates the expression of chsE (Fig. 4). Similarly, osmostress stimulates expression of chsA mainly in conidiophores and neighboring substrate hyphae, and the expression chsC in the whole fungal body, while expression of chsB is not intimately responsive to osmostress . It is thus suggested that chsE, together with chsA and chsC, can contribute to the resistance of fungal cells to osmostress causing dehydration and shrinkage of cells, while chsB does not necessarily participate to this process.
This work was supported by the grant KRF-2002–070-C00079 from the Korea Research Foundation, Republic of Korea. We thank Dr. Horiuchi of the University of Tokyo for providing the A. nidulans strains.