Aspergillus nidulans grows by apical extension of multinucleate cells called hyphae that are subdivided by the insertion of crosswalls called septa. Apical cells vary in length and number of nuclei, whereas subapical cells are typically 40 μm long with three to four nuclei. Apical cells have active mitotic cycles, whereas subapical cells are arrested for growth and mitosis until branch formation reinitiates tip growth and nuclear divisions. This multicellular growth pattern requires coordination between localized growth, nuclear division, and septation. We searched a temperature-sensitive mutant collection for strains with conditional defects in growth patterning and identified six mutants (designated hyp for hypercellular). The identified hyp mutations are nonlethal, recessive defects in five unlinked genes (hypA-hypE). Phenotypic analyses showed that these hyp mutants have aberrant patterns of septation and show defects in polarity establishment and tip growth, but they have normal nuclear division cycles and can complete the asexual growth cycle at restrictive temperature. Temperature shift analysis revealed that hypD and hypE play general roles in hyphal morphogenesis, since inactivation of these genes resulted in a general widening of apical and subapical cells. Interestingly, loss of hypA or hypB function lead to a cessation of apical cell growth but activated isotropic growth and mitosis in subapical cells. The inferred functions of hypA and hypB suggest a mechanism for coordinating apical growth, subapical cell arrest, and mitosis in A. nidulans.
MULTICELLULAR organisms use a variety of molecular signals to establish patterns of cell growth, cell cycle arrest, cell differentiation, and cell death. For example, pattern formation in insect and animal embryos initially depends on morphological gradients of positional determinants that are laid down during oogenesis (Duffy and Perrimon 1996; Hammerschmidtet al. 1997). In plants, only a provisional body plan is established in the embryo, and asymmetric cell divisions and highly localized patterns of gene expression continually establish new organs as the plant grows (Meyerowitz 1997). One of the simplest patterns of multicellular growth occurs in the hyphae of filamentous fungi.
In filamentous fungi, growth is localized to the tips of hyphae, which are tubular, uni-, or multinucleate cells. Cells within hyphae are delimited by the insertion of crosswalls called septa (Fiddy and Trinci 1976; Gull 1978; Harriset al. 1994). In some fungi, nuclear division cycles are synchronous within cell compartments delimited by septa (Rosenberger and Kessel 1967; Robinow and Caten 1969). Septa may play a variety of roles, such as preventing loss of cytoplasm during hyphal damage (Richle and Alexander 1965) and partitioning cell growth and differentiation (Gull 1978). Although much is known about fungal cell wall synthesis (Gooday 1994) and cytoskeletal structure (Heath 1994; Ayscough and Drubin 1996), little is known about the molecular mechanisms that establish the apical growth patterns in fungal mycelia.
Cell morphogenesis in fungi is best understood in the polarized growth processes of budding and mating projection in the yeast Saccharomyces cerevisiae (Kron and Gow 1995; Pringleet al. 1995; Drubin and Nelson 1996). These processes resemble hyphal growth in filamentous fungi in that they involve polarization of the actin cytoskeleton and localized cell wall synthesis. In filamentous fungi, calcium gradients (Jackson and Heath 1993; Piersonet al. 1994), pH gradients (Kropfet al. 1995), cyclic AMP signaling (Kore-edaet al. 1991; Brunoet al. 1996), calcium signaling (Rasmussenet al. 1994; Dayton and Means 1996), and ras function (Som and Kolaparthi 1994) also play roles in tip growth, and some of these processes are essential for growth in S. cerevisiae (Cunningham and Fink 1994, 1996; Moschet al. 1996; Wardet al. 1995).
Aspergillus nidulans forms a multicellular mycelium with a highly regulated pattern of growth and nuclear division (Figure 1). In the vegetative mycelium, only tip growing cells contain actively dividing nuclei, which have synchronous mitotic cycles. Septa are formed with a uniform spacing along vegetative hyphae (Fiddy and Trinci 1976), and septum formation is dependent on mitosis, nuclear positioning, and attainment of a critical cell size (Wolkowet al. 1996). After septation, subapical cells cease growth and nuclei arrest in interphase. This state can continue indefinitely, unless branching reestablishes polarized growth and active nuclear division. Nuclear division, however, is not essential for maintaining polarized growth, since many conditional nuclear division mutants of A. nidulans are able to produce a polarized growing germ tube at restrictive temperature (Morris 1976). Conversely, overproduction of the activated form of the GTPase, A-ras, results in swollen spores or cells containing abundant nuclei (Som and Kolaparthi 1994). Thus, the mechanisms that restrict cell growth and mitosis to tip cells in A. nidulans are unknown.
A large collection of conditional mutants of A. nidulans has been generated (Morris 1976; Harriset al. 1994) and used to identify loci involved in various growth processes. We screened this mutant collection for strains with defective patterns of growth and septation. Here, we describe the characterization of six conditional mutants defining five genes that appear to play roles in mycelial pattern formation. We call these mutations hypA1-hypE2 for hypercellular because they have abnormally short subapical cells with abundant nuclei at restrictive temperature. Phenotypic analyses suggest that hypC controls the spacing of septa and may be involved in cell size control. Mutations in hypA, hypB, hypD, and hypE cause the proliferation of nuclei in subapical cells and cause defects in cell polarization. Intriguingly, hypA and hypB appear to play dual roles as positive effectors of tip cell growth and negative regulators of subapical cell growth. All hyp mutants are able to complete the asexual life cycle at restrictive temperature, suggesting that these mutants mislocalize growth cues that are required to establish wild-type mycelial growth patterns.
MATERIALS AND METHODS
The strains used in this study are shown in Table 1. Methods are described in Harris et al. (1994) and Kafer (1977). For phenotypic analyses, we used strains grown on complete medium (CM) with supplements as described by Kafer (1977). Where required, media contained 1 m sucrose or 0.01–0.1% Calcofluor (a gift of American Cyanamid, Princeton, NJ).
Mutagenesis and screening: Six hundred fifty temperature-sensitive mutants were generated by 4-nitroquinoline oxide mutagenesis of strain A28, extending the collection screened by Harris et al. (1994) to 1800 mutants. Temperature-sensitive strains were screened microscopically after 16 hr growth at 28° and 42° on thin agar plates (Harriset al. 1994). Strains were selected for having abnormally short subapical cells when grown at 42°. These strains were mated with GR5. Cleistothecia were cleaned of Hülle cells by rolling on 4% agar containing 1% diatomaceous earth (Sigma, St. Louis; Kaminskyj and Hamer 1996). Segregation of the temperature-sensitive phenotype was determined for ascospore progeny grown on replica plates at 28° and 42°. Six or more temperature-sensitive progeny from each cross were examined after growth on coverslips at 42° to confirm the mutant phenotype. Mutant progeny were crossed to strain AH12.
Pairwise crosses between all combinations of mutants were used to assess allelism. Diploids were generated from heterokaryons in these pairwise crosses (Kafer 1977) and used to test for genetic complementation. Diploids made from crosses between hyp and wild-type strains were used to test for dominance/recessivity. Haploid sectors showing segregation of parental spore colors were generated from these diploids using benlate-induced instability (Hastie 1970). Gene designations were assigned after Clutterbuck and Arst (1995).
Microscopy: Conidia were inoculated in liquid and allowed to attach to glass coverslips for microscopic examination. Conidia were inoculated into liquid media (105 conidia/ml) with nutritional supplements and 100 μg/ml ampicillin, and they were incubated at 28° and 42°. Germlings were fixed and stained for nuclei using Hoechst (Warrington, PA) and for walls using Calcofluor (Harriset al. 1994). Some germlings were stained with 1 mg/ml FITC-conjugated wheat germ agglutinin (FITC-WGA; Sigma), which binds to the chitinous cell wall. Slides were mounted in Citifluor (Ted Pella, Redding, CA) and examined with a BH-2 microscope (Olympus, Lake Success, NY) using phase contrast and UV or FITC epifluorescence. Germling features were measured with an ocular micrometer. Statistical analyses used Statview SE+Graphics (Abacus Concepts, Berkeley, CA).
Subapical cell characteristics: Germlings were grown on coverslips for 18–24 hr at 28° and 42° and prepared for microscopy (Harriset al. 1994). Unbranched subapical cells were chosen for having a complete septum at each end, with these septa being perpendicular to the long axis of the hypha. Cell length was defined as the distance between adjacent septa, measured at the junction between the septa and the lateral hyphal wall. Unlike other hyp strains, hypC4 germlings grown at 42° often had thick septa, so their cell lengths were measured from center to center of adjacent septa. At least 50 subapical cells were analyzed for each strain and growth temperature. Hyphal width was measured across septa that were complete and perpendicular to the long axis of the hypha. At least 30 width measurements were made for each strain and growth temperature.
Germling morphogenesis: Germination and growth characteristics were determined microscopically, as described below, for wild-type and hyp germlings grown for up to 16 hr at 28° and 42°. Only germlings growing from isolated spores (at least one spore diameter from all other spores) were chosen for this analysis. A. nidulans conidia adhere tightly to glass coverslips after 3–4 hr in complete media (CM) at either temperature, regardless of germination. As a result, ungerminated conidia were not lost from the original population.
Germination was defined as the production of a germ tube that was at least half the spore diameter. At least 100 spores were counted for each data point. Results are shown as percentage germination.
Total hyphal lengths were measured with an ocular micrometer and 20×, 40×, or 100× objectives, as appropriate. Total hyphal length was defined as the distance along the long axis of all hyphae in each germling plus all branches that were longer than half the hyphal diameter. At least 50 germlings were measured for each strain and growth temperature. Total hyphal length per germling is shown as the mean value ± SE of the mean.
The total number of nuclei in wild-type and hyp strains was determined by counting them in spores and germlings. Mitosis is synchronous in wild-type germlings (Bergen and Morris 1983), but not in some hyp strains grown at 42°. To compare the number of nuclei between hyp and wild-type germlings, these values are shown as the population mean ± SEM. This method is not intended to convey that germlings could contain partial nuclei. Data point markers obscured small error bars when these data were plotted.
Septa were counted in isolated wild-type and hyp germlings grown at 28°and 42°, using fixed germlings stained with Hoechst and Calcofluor. Unlike the cell length measurements described above, in this analysis, septa were counted whether they were perpendicular or oblique to the long axis of the hypha, or if they were incomplete. If incomplete, a continuous ring of Calcofluor-stained material had to extend more than half of the hyphal circumference. hypA1 and hypE2 strains frequently had double septa (morphologically normal septa <1 μm apart). Double septa usually defined anucleate cells, but some had single nuclei (e.g., Figure 3a), so they were counted as two septa. The broad septa characteristic of hypC4 strains were counted as single septa because an intervening space could not be resolved. Septation is asynchronous in hyp germlings growing at 42°, so the number of septa per germling was expressed as population mean ± SEM, as described above for nuclei.
Genetic analysis: Six mutant strains with abnormally short subapical cells in their vegetative hyphae were found by visually screening 1800 temperature-sensitive mutants (Harriset al. 1994). hyp mutants did not have terminal arrest phenotypes, and they grew indefinitely at restrictive temperature, albeit with abnormal morphologies. Table 2 summarizes the genetic analysis of the hyp mutants. The mutants defined six single-gene mutations in five complementation groups. Two alleles were recovered for hypA, and single alleles were recovered for the remaining mutants. All mutations were recessive to wild type.
hypA and hypE were mapped to chromosomes by fortuitous linkages to markers segregating in backcrosses. hypA is linked to pabaA (4%) on chromosome I. Because of the proximity of another hyphal morphology gene in this region (sepA; Harriset al. 1997), we investigated the linkage between hypA and sepA. hypA is tightly linked to sepA (2%) but is not allelic with it. The near-centromere chromosome 1R marker lysF is linked to both hypA (0.3%) and sepA (2.7%). hypE2 is linked to chaA (8%) on chromosome VIII. hypC4 and hypD3 were mapped to chromosomes by parasexual genetic crosses, as described by Kafer (1977), using strain A104.
Phenotype analysis: Figure 2 shows typical morphologies of wild-type and hypA1 hyphae grown for 18 hr at 28° and 42°. Wild-type cells grown at 28° are shown in Figure 2a. hypA1 hyphae grown at 28° (Figure 2b) closely resemble those of the wild-type strain. This is typical of all hyp strains grown at 28°. Figure 2c shows the dramatic difference between hypA1 and wild-type cells grown in mixed culture at 42°. The hypA1 germlings were shorter and wider, and their walls and septa stained more brightly with Calcofluor. The hypA1 germlings had made one or two closely spaced septa despite their limited length, compared to ~10 septa per germling for the wild-type strain. Despite these abnormalities, septum placement was always asymmetric, producing a relatively long apical cell. Figure 2d shows older hypA1 germlings that were labeled with FITC-WGA. The arrows indicate the base of hemispherical hyphal tips, behind which the apical cells continued to increase substantially in width, distinctly unlike cylindrical wild-type hyphae (Figure 2, a and c) or hyp strains grown at 28° (Figure 2b). hypA1 germlings were pulse labeled with FITC-WGA and allowed to grow at 42° for another 3 hr. Walls formed during this interval were relatively faint (Figure 2e).
Figure 3 shows hyp germlings grown for 24 hr at 42°, then fixed and stained for walls and nuclei. Quantitative analysis of the hyp phenotypes is summarized in Table 3, and subapical cell characteristics are shown graphically in Figure 4. With the exception of hypC4, hyp germlings grown at 42° were significantly wider than wild type (see also Figure 2c), and all hyp mutants had significantly shorter subapical cells. The increased hyphal width in hyp mutants might compensate for the decrease in subapical cell length and thus conserve subapical cell volume. To test this hypothesis, average subapical cell volumes were calculated using the data in Table 3, A and C. Average cell volumes for wild-type and hyp strains varied considerably at 28°, suggesting that these calculations represent modest approximations. At 42°, the majority of the hyp mutants showed only modest changes (±50%) in average cell volume, with hypD3 and hypE2 increasing and hypB5 and hypC4 decreasing. In contrast, hypA1 underwent a dramatic increase in subapical cell volume. This result suggests that changes in cell width in hypA1 mutants result in increased cell volumes.
Wild-type hyphae typically had three to four nuclei per cell, regardless of growth temperature, as did hyp strains grown at 28° (Figure 4; Table 3B). In contrast, the short subapical cells produced by hypA1, hypB5, hypD3, and hypE2 strains at 42° had significantly more nuclei. hypA1 produced the most dramatic changes, with some cell compartments containing up to 20 nuclei. hypA1 nuclei appeared smaller and more compacted than wild-type nuclei. The dramatic increase in cell volume in hypA1 mutants may activate additional rounds of nuclear division.
Other defects observed in hyp mutants included a high frequency of dichotomous apical branches in hypA1 and hypE2 germlings (Figure 3a). hypB5 and hypD3 germlings had near-apical, lateral branches (Figure 3, b and d). About 25% of hypA1 and hypE2 germlings had double, oblique, or incomplete septa, unlike wild-type or other hyp strains. About 5% of hypB5 germlings had anucleate subapical compartments. hypC4 germlings had cell compartments that were half the size of wild-type at 42°, and ~20% of the septa were thickened or appeared as double septa (Figure 3c, arrow). All hyp mutants could be grown indefinitely at 42°, and all mutants were able to complete the asexual life cycle and produce viable conidia at restrictive temperature (data not shown).
Figure 5 shows plate phenotypes for wild-type and hyp mutants growing at 28° and 42°. Wild-type and hypA1-hypD3 colonies grew well at 28°, but growth was restricted for the hypE2 strain (Figure 6a). All hyp strains were clearly restricted for growth at 42°, compared to wild-type (Figure 5b) or to their growth at 28° (Figure 5a). The hypD3 defect was osmotically remedial at 42°, shown by increased colony growth on medium containing 1 m sucrose (Figure 5c). hypD3 germlings were grown for 18 hr at 42° in CM or CM plus sucrose and were prepared for microscopy. Figure 5e shows that the restrictive phenotype of hypD3 germlings grown in CM at 42° was fully remediated by growth on CM plus sucrose (Figure 5f). Under these conditions, the hypD3 hyphae resembled wild-type strains in width, branching pattern, and nuclei per cell.
hyp strains might be hypersensitive to growing in the presence of Calcofluor, which binds to cell walls and interferes with crosslinking between chitin and nonfibrillar components. Growth of the hyp strains was reduced in the presence of Calcofluor, but none of the strains were notably sensitive compared to wild type (Figure 5, a vs. d).
Germling development and morphogenesis: The hyp mutant defects were examined in detail by growing all strains on coverslips at 28° and 42° for microscopy and following the kinetics of germination, nuclear division, growth, and cell division (septation). At hourly intervals, coverslips were fixed and stained for walls and nuclei (Figure 6).
A. nidulans conidia are apolar (Bergen and Morris 1983) and must polarize to form a germ tube. The hyp phenotypes could result from defects in polarity establishment. Wild-type strains germinate synchronously after a short lag (Bergen and Morris 1983), and germination was faster at 42° than at 28°. In contrast, all hyp strains had delayed germination compared to wild type, even at 28° (Figure 6a). Despite delays, spore germination in hyp strains eventually exceeded 95% at both temperatures. The delay in germination was exacerbated at 42° for all but hypC4 strains. We conclude that hypA1, hypB5, hypD3, and hypE2 have delayed germ tube formation, perhaps because of defects in cell polarization.
hyp morphologies could result from abnormally rapid mitotic cycles, because nuclei were abundant in hypA1, hypB5, hypD3, and hypE2 germlings grown at 42° (Figure 3). Accumulation of nuclei in the wild-type strain was more rapid at 42° than at 28°, as shown in Figure 6b. The data for wild-type growth (Figure 6, open symbols) are plotted with the data for each of the hyp strains to facilitate comparisons. The nuclear cycle of hyp germlings was delayed compared to wild type when grown at 28° and 42° (Figure 6b). The length of the nuclear cycle was unaffected by growth at 42° for hypA1, hypB5, and hypE2 germlings, and like the parental strain, it was shorter at higher temperature for hypC4 and hypD3 strains. Most notably, the nuclear cycle of the hyp strains did not accelerate more than the wild type at 42°.
In the wild-type parental strain, the first nuclear division occurred before germination, and most germinating conidia had two nuclei, as did hyp conidia germinating at 28°. At 42°, hypC4 strains typically had two nuclei per spore when they germinated, but the hypA1, hypB5, hypD3, and hypE2 strains had four to eight nuclei. Therefore, germination was delayed with respect to accumulation of nuclei in hypA1, hypB5, hypD3, and hypE2 strains growing at 42°, consistent with a temperature-sensitive delay in germ tube emergence. These findings further demonstrate that hypA1, hypB5, hypD3, and hypE2 mutations affect cell polarization.
The hyp morphologies may also be caused by defects in the maintenance of cell polarity, that is, by reduced hyphal extension rates. Maximum hyphal length determines colony diameter (Figure 5, a–d), but this does not account for additional hyphal growth through branching. Hyphal growth was measured for individual germlings as the total hyphal length, including all branches longer than half the hyphal diameter. Figure 6c shows the average germling lengths for wild-type and hyp germlings growing at 28° and 42°. For wild-type strains, the total hyphal length per germling increased rapidly after germination, with the rate being faster at the higher temperature. Growth of all hyp strains was delayed because of slow germination. Growth of hypA1 and hypB2 strains was inhibited at 42° compared to 28° (Figure 6c). The colony growth of hypC4 and hypD3 strains was reduced at 42° (Figure 5, a and b). These strains, however, branched extensively at 42°, so the cumulative length of all branches increased more rapidly than at 28° (Figure 6c). The growth rate of hypE2 germlings was very slow (Figure 5a) and was relatively unaffected by growth temperature (Figure 6c), although hypE2 germlings were more branched at 42°. Therefore, hyp strains have a variety of defects in maintaining tip growth, all of which produce compact colonies at 42°.
Septation in hyp strains growing at 28° was qualitatively similar to wild type. All hyp germlings produced septa at 42° (Figures 2 and 3). For hypA1, hypB5, hypC4, and hypD3 strains, the onset of septation was delayed at 42° and occurred in short cells that contained ≥16 nuclei. Like wild type, septation in hyp strains appeared to correlate with mitosis (data not shown). Since tip growth was reduced in hyp strains at 42°, however, additional rounds of septation produced significantly shorter subapical cells (Figures 2, 3 and 4; Table 3).
Temperature shift analysis: The hyp mutations appear to perturb aspects of hyphal morphology without blocking completion of the life cycle. To gain further insight into how the loss of hyp gene function affects cell morphogenesis, we grew hyp mutants on coverslips or agar at permissive temperature for 12–15 hr and then shifted the cells to restrictive temperature for varying lengths of time. Little or no change was observed for hypC mutants. Figure 7 illustrates changes associated with loss of hypA function. Similar results were obtained with hypB5 (data not shown). Within 3 hr of shifting to the restrictive temperature, apical cell growth ceased and only subapical cells began to enlarge isotropically (Figure 7, a–c). Subapical cells stained brightly with Calcofluor, indicative of cell wall synthesis (Figure 7b). Concomitant with their increased cell size, subapical cells accumulated abundant nuclei (Figure 7c). During this time, apical cells (arrows) underwent a progressive loss of nuclear staining and never resumed growth. This hypA1 defect was readily reversible (Figure 7d). If germlings were grown at restrictive temperature and then returned to permissive temperature, they quickly reestablished multiple new apical tip cells.
Loss of hypA1 overcomes the growth and nuclear cycle arrest of subapical cells. To test if the loss of hypA was capable of overriding nuclear cycle blocks, we constructed double mutants containing hypA1 and temperature-sensitive mutations in a variety of cell cycle genes (nimA5, nimE6, nimT23, and nimX1) that arrest cells at the G2-M boundary (Doonan 1992; Osmaniet al. 1994). In all cases, the double-mutant phenotypes were additive. Subapical cell compartments swelled, but nuclear division remained blocked (data not shown). Similarly, when germinated at restrictive temperature, double mutants produced pear-shaped germlings that contained single nuclei (data not shown). We conclude that hypA (and hypB) are required for apical cell growth but are also necessary to suppress growth in subapical cells until the onset of branch formation.
Temperature shift analysis with hypE2 is shown in Figure 8. Similar results were obtained with hypD3 (data not shown). When shifted to restrictive temperature, the apical and subapical cells widened and accumulated nuclei, suggesting that loss of hypD and hypE function causes a generalized perturbation in growth polarity.
Wild-type A. nidulans hyphae grow indefinitely by the extension of mitotically active apical cells. Septation results in the formation of subapical cells that are quiescent for growth and mitosis unless they branch (Figure 1). We have identified five genes, mutations in which cause markedly abnormal cell growth patterns. The hyp mutations define a class of temperature-sensitive mutations distinct from cell cycle mutants (Morris 1976), septation mutants (Harriset al. 1994), or conidiation mutants (Clutterbuck 1969; Champe and Simon 1992; Wieseret al. 1994). At restrictive temperature, all hyp mutants are able to complete the asexual life cycle. Cumulative data from phenotypic analyses suggest that at least some of the hyp genes play novel roles in controlling the pattern of cell growth in fungal mycelia.
The hypA1, hypB5, and hypE2 phenotypes are unlikely to be caused by defects in cell wall synthesis. Five chitin synthase genes have been cloned in A. nidulans, and none of the strains disrupted for these genes phenocopy the hyp mutants (Motoyamaet al. 1994; Borgiaet al. 1996a; Spechtet al. 1996). The genes gcnA, manA, orlA-D, and tsE encode additional components of the cell wall biosynthetic pathway (Borgia and Dodge 1992; Smith and Payton 1994; Borgiaet al. 1996b). Unlike the hyps, these mutants are chitin deficient and/or unable to grow without osmotic stabilization, which suggests that their walls do not mature properly. Although hyp mutants, particularly hypD3, have phenotypes suggestive of cell wall structural defects, all hyp strains were able to grow and complete their asexual life cycles at 42° without osmotic stabilization. Therefore, hyp mutants appear to define a class of genes that is phenotypically distinct from those directly involved in cell wall biosynthesis.
hypC4 is distinct from the other hyp mutations because its effect is largely restricted to septation. Like wild type, but unlike the other mutants, hypC4 germinates faster at 42° than at 28°, and it has normal hyphal width and nuclei per cell when grown at restrictive temperature. hypC4 strains branch profusely at 42°, so that total germling length increases even though colony diameter is restricted compared to growth at 28°. hypC4 has more septa per hyphal length and thus shorter subapical cells than wild-type strains. hypC4 hyphal walls are not stained abnormally with Calcofluor, and the defect is not osmotically remedial; however, about one-third of the septa in hypC4 strains grown at 42° are unusually thick. In A. nidulans, a threshold cell size controls the ability of nuclear divisions to activate septation (Wolkowet al. 1996). Thus, hypC4 may lower the threshold cell size required to activate septation. Alternatively, loss of hypC may alter nuclear distribution in a way that septa are formed in closer proximity. Experiments to distinguish between these possibilities are underway.
In contrast to the hypC4 mutation, hypA1, hypB5, hypD3, and hypE2 mutations seem to primarily affect hyphal morphogenesis and, consequently, affect septation. Kinetic studies of germ tube polarization and nuclear division suggest that these hyp mutations delay cell polarization. Even more unusual, hyp germlings continue to get wider behind the growing tip and in subapical cells. The lateral walls of hypA1, hypB5, hypD3, and hypE2 germlings stain more brightly with Calcofluor than do those of wild type, suggesting that hyp wall thickness may be excessive. The tips of hyp germlings stained less brightly with Calcofluor than their lateral walls, suggesting that hyp cell walls may not be evenly thick (Figure 3, a, b, and d). This growth pattern implies that wall deposition in hyp mutants may not be restricted to hyphal tips. This is consistent with colony growth (hyphal extension) rates being reduced compared to wild type at 42°. Thus, at restrictive temperature, hypA1, hypB5, hypD3, and hypE2 germlings may grow by both expansion and extension.
A variety of genes affect growth polarity in filamentous fungi. These include components of the actin cytoskeleton (McGoldricket al. 1995), components of the cAMP signaling pathway (Kore-edaet al. 1991), and small GTPases of the ras family (Som and Kolaparthi 1994). In addition, some of these genes also affect patterns of septation. Mutations in the Neurospora crassa regulatory subunit of protein kinase A (mcb-1) cause major defects in growth polarity, actin localization, and septation (Bruno et al. 1997). In particular, temperature shift experiments with the mcb-1 mutants resulted in phenotypes similar to hypD3 and hypE2 (Figure 8). hyp phenotypes also resemble the effects of moderate overproduction of the constitutively active Aspergillus ras homolog A-ras. In this case, extension but not growth was limited, so the germlings were wide and had thickened walls and septa and abundant nuclei compared to wild type. Germination was not affected by moderate A-ras overproduction, but it was inhibited by high A-ras. Nuclear division and wall deposition were largely unaffected by A-ras activity (Som and Kolaparthi 1994). Molecular cloning of hyp genes may provide clues to their function in growth polarity.
Temperature shift analysis revealed that hypA and hypB have cell-specific functions in hyphal morphogenesis. Unlike other mutations affecting growth polarity in fungi, loss of hypA function resulted in the rapid onset of subapical cell growth and nuclear division and a complete cessation of apical growth. Subapical cell growth occurred isotropically, with cells becoming swollen and rounded. In contrast, the apical cell underwent vacuolization, and cytoplasmic and nuclear staining were rapidly lost (data not shown). Thus, the hypA gene product must play two roles in distinct cell types. In apical cells, hypA is required for continued growth polarity. After septation, hypA must activate a growth arrest in subapical cells. It is not currently known how a single gene product could play contrasting roles in different cell types; however, the apparent dual function of hypA (and hypB) provides the mycelium of A. nidulans with a simple regulatory mechanism for patterning growth. We have recently cloned hypA by virtue of its close linkage to sepA (Harriset al. 1997). hypA encodes a 5.3-kb gene that has no significant similarity to any gene or expressed sequence tag in current databases (S. Kaminskyj and J. E. Hamer, unpublished data). Thus, hypA1 and perhaps other hyp mutations will define novel genes involved in cell polarity and cell pattern formation.
We thank Drs. Lis Hamer, Kelly Kinch, Michelle Momany, John Marhoul, and Tom Wolkow for critical reading of the manuscript. This research was funded by the National Institutes of Health (GM-49867; J.E.H.), the Indiana Elks Cancer Research Fund (J.E.H. and S.G.W.K.). S.G.W.K. is a recipient of a postdoctoral fellowship from the Natural Sciences and Engineering Research Council of Canada.
Communicating editor: R. H. Davis
- Received September 18, 1997.
- Accepted October 20, 1997.
- Copyright © 1998 by the Genetics Society of America