A fragment from the open reading frame of the cloned chsA gene from Aspergillus nidulans was deleted and replaced with the argB gene. The resulting construct was used to replace the wild-type chsA gene in an argB deletion strain. The growth and morphology of the vegetative hyphae from the resulting chsA disruptant strain were indistinguishable from those of a wild-type strain but the chitin content of the hyphae from the disruptant was reduced to approximately 90% of that of wild-type. The disruptant showed reduced ability to produce the asexual spores (conidia) that are formed by differentiated aerial hyphae called conidiophores. The ability to form undifferentiated aerial hyphae was not impaired in the disruptant. The conidiophores and conidia produced by the disruptant were indistinguishable from those of wild-type. Conidium formation by the disruptant grown on a variety of media was reduced to about 30% of the wild-type. A chsE null strain did not show a defect in conidiation but a strain in which both chsA and chsE were inactivated produced about 3% of the conidia of wild-type. That finding supports the hypothesis that chsA and chsE encode a partially redundant function necessary for conidiophore development.
Chitin is a linear polysaccharide of N-acetylglucosamine residues linked by β-1,4 glycosidic linkages. The polymer plays an important structural role in the cell walls of all true fungi [1,2]. Studies in the yeast Saccharomyces cerevisiae have shown that the synthesis of the polysaccharide is complex, requiring three related genes that encode chitin synthase isozymes . Chitin synthase genes form a divergent family that can be divided into five classes [4,5]. The C-termini from the enzymes are relatively conserved and are thought to encode the catalytic domain . The N-termini of the isozymes are more highly divergent and may contain sequences necessary for the correct localization and regulation of the enzymes. The three chitin synthase isozymes from S. cerevisiae are functionally distinct since each synthesizes a unique chitin subfraction at specific sites in the yeast cell at specific times in the cell cycle . In the developmentally more complex filamentous fungi, chitin synthesis is even more intricate and involves additional chitin synthase genes. Five chitin synthase genes have been identified in the filamentous ascomycete Aspergillus nidulans[5,7–11]. Seven chitin synthase genes have been identified in Aspergillus fumigatus[7,12].
The phenotypes of A. nidulans mutants carrying inactivated chitin synthase genes indicate that the isozymes encoded by chsB, chsD and chsE are each responsible for the synthesis of a functionally distinct chitin subfraction [5,7,10,11,13]. The chsB gene is responsible for the synthesis of a minor chitin subfraction in hyphae that are necessary for the normal growth, morphology and organization of hyphal cells [7,11]. The chsD gene , also referred to as csmA[8,13], is responsible for the synthesis of a chitin subfraction that is necessary for the structural rigidity of hyphal walls and of conidiophore vesicles. Inactivation of the chsE gene , also referred to as chsD, results in strains that lack about 30% of hyphal chitin but are morphologically and developmentally normal.
Strains carrying an inactivated copy of either the A. nidulans chsA or chsC genes [9–11] are reported to exhibit no morphological or developmental defects. A possible explanation for these results is that the isozymes encoded by the genes are functionally redundant with isozymes encoded by other chitin synthase genes. Motoyama et al.  showed that a strain in which both chsA and chsE (referred to as chsD in that paper) had been inactivated produced about 7% of the conidia of a wild-type strain. In this paper, we report that inactivation of chsA alone in fact leads to a defect in conidiophore formation and to a deficit in the chitin content of vegetative hyphae.
2Materials and methods
2.1Strains, media and culture conditions
The following strains were used in this study: PB 001 (yA2 pabaA1; veA1); RMS 010 (biA1 methG1; ΔargB::trpCΔB; trpC801 veA1); RMS 011 (yA2 pabaA1; ΔargB::trpCΔB; trpC801 veA1); PB 122 (yA2 pabaA1; ΔchsE::argB; ΔargB::trpCΔB; trpC801 veA1) ; PB 141 (yA2 pabaA1; ΔchsA::argB; ΔargB::trpCΔB; trpC801 veA1); PB 142 (biA1 methG1, ΔchsA::argB; ΔargB::trpCΔB; trpC801 veA1) and PB 146 (yA2 pabaA1; ΔchsA::argB; ΔchsE::argB; ΔargB::trpCΔB; trpC801 veA1). Strains were grown at 37°C on YG medium (0.5% yeast extract, 1% glucose) or on the minimal medium described by Clutterbuck . PB 141 was crossed with RMS 010 to produce PB 142. PB 142 was crossed with PB 122 to give PB 146.
A cosmid containing the entire chsA gene was identified from an ordered genomic cosmid library in vector pCosAX  using PCR and the primers GCGTCTAGATACCGAGACTCGTACGAAGA and GCCTCTAGAGTGATCTCAGATACAGGAGG. The cosmid was digested with SalI and XbaI and a 3.9-kb fragment containing the entire open reading frame was cloned into the SalI-XbaI site of Bluescript KS+ to yield plasmid pYM4. pYM4 was cut with ClaI and PstI to remove a 2-kb internal coding fragment from the chsA gene and a 1.7-kb ClaI-PstI fragment from plasmid pRP8 that contained the entire argB gene from A. nidulans was ligated into the site, yielding pYM5. Strain RMS 011 was transformed with pYM5 that had been digested with KpnI. Transformations were performed according to Yelton et al. .
Strains were inoculated at the center point of solid media contained in 90-mm Petri dishes. The plates were incubated at 37°C until the colonies had reached a diameter of 80 mm. Each dish was flooded with 10 ml of 50% glycerol and the conidia were harvested by scraping the surface of colonies with a nichrome wire scraper. For each strain, the conidia from three plates were pooled and the number of conidia present in the suspension was counted in a hemocytometer. Following dilution of the suspensions, the number of viable conidia from each strain was determined by colony count on YG medium supplemented with 0.6 M KCl.
The chitin content of hyphae was assayed as previously described . The data presented represent the means for five independent experiments. In each experiment, three cultures were grown for each strain and the mycelia from each culture were pooled for determination of amino sugar and protein content.
Strain RMS 011 was transformed with plasmid pYM5 that had been linearized and Arg+ transformants were selected. Fig. 1A shows a restriction map of the chsA region of the wild-type A. nidulans genome and that predicted for a transformant produced by homologous integration of the fragment. We screened transformants for individuals with a Southern blot pattern consistent with the disruption event. Several transformants showed the blot pattern predicted for a disruption (Fig. 1B). PB 141 was chosen for characterization in this study.
Wild-type A. nidulans strains form copious numbers of brightly colored asexual spores (conidia) that are produced by differentiated aerial hyphae known as conidiophores. On YG-based media, PB 141, a yellow spored strain containing the disrupted chsA gene, could be visually distinguished from an isogenic wild-type strain, PB 001, since the yellow color of the mutant colony was not as intense as that of PB 001 (Fig. 2). Microscopic examination of the colony surface of PB 141 grown on YG showed numerous aerial hyphae but a decreased number of conidiophores compared to the wild-type strain. The conidiophores and conidia produced by PB 141 were indistinguishable from the wild-type (not shown).
We compared the number of conidia produced by chsA::argB and chsA::argB, chsE::argB strains with a wild-type grown on a variety of media (Table 1). On the four media, the chsA::argB disruptant strain produced an average of 31% of the conidia of the wild-type strain. The chsA::argB, chsE::argB strain produced an average of 3.1% of the conidia of wild-type. The viability of the conidia from the wild-type and mutant strains produced on all media was comparable and averaged about 80% of the microscopic count (data not shown). The reduced viable count was due to the fact that conidia frequently occur in chains that could be distinguished in the hemocytometer but give rise to single colonies in the viable count experiments.
Table 1. Spore formation by a wild-type and chitin synthase null mutants
YG+0.6 M KCl
Minimal+0.6 M KCl
Data are presented as the number of conidia per ml. The percentages of conidia produced by the wild-type control are given in parentheses.
The chitin content of hyphae of the chsA::argB strain (PB 141) was compared with that of a wild-type strain (PB 001). Five independent experiments were performed in which each strain was grown and processed for chitin content. In each experiment, the chitin content of the mutant was lower than that of the control strain. The average chitin content of PB 001 in the five experiments was 1.20 (±0.16) mg N-acetylglucosamine equivalents mg−1 protein while PB 141 averaged 1.05 (±0.09) mg N-acetylglucosamine equivalents mg−1 protein. Using the paired t-test for the five experiments, the mean difference in chitin content (0.15 mg N-acetylglucosamine equivalents mg−1 protein) was statistically significant with a probability of 0.0082.
Previous work has shown that inactivation of the chsB, chsD or chsE genes leads to strains with distinguishable phenotypic defects [5,7,11,13]. These observations indicate that each gene is responsible for the synthesis of a unique chitin subfraction that serves a distinct organizational, structural or developmental role. The data presented in this paper demonstrate that inactivation of chsA produces strains that are deficient in conidiation and in hyphal chitin content. Motoyama et al.  reported that the disruption of chsA resulted in no deficit in conidiation. The discrepancy in results could be due to differences in the disruption strategies used in the two studies. In our work, a substantial fraction of the open reading frame of chsA was replaced with the argB gene while in the previous study, a simple insertion of a disrupting marker was made into the 3′ region of chsA. In the latter case, it is possible that the gene retained partial activity.
Strains carrying an inactivated copy of chsA can be distinguished phenotypically from chsB and chsD disruptants and consequently chsA is functionally distinct from chsB and chsD. The hyphal chitin deficiency and the defect in conidiation of chsA disruptants might be related. The hyphal chitin synthesized by the chsA-encoded enzyme could be necessary for the initiation of conidiophore formation. Alternatively, the enzyme might also synthesize chitin in conidiophores and the lack of that chitin is responsible for the conidiation defect in chsA disruptants.
The chsA disruptants are only partially blocked in conidium formation (30% of the conidia of wild-type are produced) and the conidiophores and conidia that are produced have the wild-type morphology. Inactivation of chsE does not result in a defect in conidiation (this study, [5,10]). Strains carrying inactivated copies of both chsA and chsE have a severe block in conidiation, producing only 3–7% of the conidia of a wild-type strain (this study, ). These observations are consistent with the proposal that the chsA and chsE-encoded enzymes are capable of synthesizing the same chitin subfraction. The reduction in conidiation caused by inactivation of chsA and the finding that chsE disruptants conidiate at wild-type levels could be explained by assuming that, in wild-type strains, the chsA-encoded activity is present in excess while the chsE-encoded activity is present in limiting amounts. The fact that double disruptants are capable of 3–7% of the conidiation of wild-type suggests that chitin synthases encoded by other genes may be capable of fulfilling the chsA/chsE function in conidiation to a limited extent.
The work in this paper was supported by grants to P.T.B. from the American Lung Association of Illinois and by the NIH (Grant GM 54380).