Candida albicans is a dimorphic fungus capable of growing as a budding yeast and as a filamentous hypha. We have used the technique of immunofluorescence to study the changes in the microtubule cytoskeleton during the cell cycle in both growth forms. This approach has revealed the presence of an extensive system of microtubules, including cytoplasmic microtubules and a rod-like intranuclear spindle. We have provided a complete description of the arrangement of cytoplasmic and spindle microtubules at each phase of the yeast cell cycle. A novel and characteristic feature of the yeast phase of Candida is the presence of an array of short microtubules at the neck of the doublet cell. This neck-associated array (NAA), is apparently organized independently of the main microtubuleorganizing centre, the spindle pole bodies, at late anaphase. Analysis of microtubule patterns in the hyphal state reveals that the general arrangements of microtubules are similar to those seen in the yeast phase. These patterns suggest a role for the cytoplasmic microtubules in the nuclear migration that occurs during hyphal growth. A major finding is that the mitotic spindle in hyphae is considerably longer (12–20μm) than the spindle in yeast cells (7–8 μm). This may reflect the role of the hyphal mitotic spindle not only in nuclear division but also in the positioning of the daughter nuclei at the centres of hyphal compartments.

Acquisition of shape in animal cells is clearly dependent on a complex internal cytoskeleton. This serves to integrate internal organelle positions and functions with those of the cell surface. In walled cells, however, the form is ultimately a function of the surrounding cell wall. Despite this, evidence from many sources suggests that in these walled cells internal organelle functions are still integrated with surface phenomena via the cytoskeleton. In the fungi there are two distinct growth forms: spherical isometric growth and polarized tip growth exemplified by the budding yeasts and filamentous fungi, respectively. During the budding growth of yeasts, nuclear division must be integrated with cell wall growth, whereas in filamentous fungi nuclear division and segregation must also be coupled with nuclear migration to maintain the position of the nucleus relative to the growing hyphal tip. Recent evidence suggests that this integration of organelle dynamics with cell wall expansion is a function of the cytoskeleton (Pringle et al. 1986; Novick & Botstein, 1985; Marks et al. 1986; Anderson & Soli, 1986). Immunofluorescence studies of the dynamics of microtubule organization have so far been provided for two yeasts, Saccharomyces (Kilmartin & Adams, 1984) and Schizosaccharomyces’ (Hagan & Hyams, 1988), together with the filamentous fungi Uromyces (Hoch & Staples, 1985) and Schizophyllum (Runeberge et al. 1986).

Candida albicans is a dimorphic yeast, capable of growth in two different forms: as a budding yeast and a filamentous hypha, depending on external conditions (Mitchell & Soil, 1979). Thus it presents an ideal opportunity to study the role of the cytoskeleton in the morphogenesis and organelle behaviour of the two distinct morphologies and to relate observations on the cytoskeleton from budding and filamentous organisms. The development of a successful immunofluorescence method for Candida reveals a complex pattern of cytoplasmic and mitotic microtubules in both cell forms. Two particular features are emphasized in this study; first, the presence of an array of cytoplasmic microtubules at the neck of the budding cell typically late in the cell cycle, raising the possibility of a cytoplasmic microtubule organizer. Second, we show a pronounced cell-type-specific variation in mitotic spindle length between the yeast and hyphal cells, indicating the influence of cell morphology on microtubule dynamics.

Strains and culture

Candida albicans 124, a wild-type strain, was used throughout this study and was originally obtained from the Sandozforsch-ungsinstitut, Vienna. This strain was maintained on malt agar plates at 30°C and subcultured routinely every 2 weeks.

Yeast phase cells were grown by inoculating 25 ml of a modified synthetic amino acid medium (Lee et al. 1975; Soil et al. 1981) containing arginine but without added zinc in a 250 ml Erhlenmeyer flask from a plated colony or a previous liquid culture. These cultures were incubated at 30°C on a rotary shaker at 130 revs min−1.

To induce hyphal germ tube outgrowth, cells were centrifuged and resuspended in a starvation medium (Soil & Herman, 1983) for 30 min. A 200 μl sample of this cell suspension was plated out onto serum agar (starvation medium, supplemented with 1·5% agar and 20% newborn calf serum (Gibco BRL)), overlaid with sterile cellophane film. Plates were then incubated at 37 °C and germ tube outgrowth was monitored directly by phase-contrast light microscopy.

Immunofluorescence

The immunofluorescence protocol was developed from those of Kilmartin & Adams (1984) and Hagan & Hyams (1988).

Yeast cells

An exponential phase culture of 8×106 to 4× 107 spheres ml−1 (using the definition of spheres of Mitchell & Soil, 1979) was fixed for 90min in 3·7% formaldehyde in PEM (0·1 M-piperazine-N,N′bis[2-ethane-sulphonic acid] (Pipes), 1 mM-ethyleneglycol-bis-(β-aminoethyl ethyl) N,N,N′,N′-tetraacetic acid (EGTA) and 0·1mM-MgSO4, pH 6·9) freshly made up from paraformaldehyde. Cells were fixed by addition of concentrated formaldehyde-PEM to the flasks to a final concentration of 3·7% formaldehyde and were maintained on a rotary shaker at 30°C during fixation. Alternatively, the culture was centrifuged at 1·1×103g for 5 min and the pellet was immediately resuspended in 3·7% formaldehyde-PEM for 90 min. Both methods produced similar reproducible results. Despite considerable attempts to vary fixation and subsequent steps, glutaraldehyde, with or without formaldehyde, was never successful as a fixative. Careful observation of cells suggests that later protoplast formation steps are inhibited by chemical modification of the cell wall by glutaraldehyde fixation. This agrees with the observations of Kilmartin & Adams (1984), who found that glutaraldehyde fixation of Saccharomyces cells reduced antibody permeability in yeast cells.

After fixation, the cell suspensions were transferred to 10 ml plastic centrifuge tubes centrifuged at 14×103g for 5 min and subsequently gently resuspended in 5 ml of PEM. Two subsequent washes with 1 ml of PEM were carried out in Eppendorf vials using an MSE microfuge at 1-16×104g for 4–6s to spin the cells down. The cells were then washed once in PEMS (PEM+1·2 M-sorbitol, pH 7·5) and resuspended in 1ml of PEMS. The cell walls were removed by adding 1 mg of lyticase (partially purified lyophilized powder, Sigma) and incubating on a slowly rotating wheel at room temperature for 30–60 min or until 20–50% of the cells appeared phase dark by phasecontrast light microscopy. After washing twice in PEMS the protoplasts were then resuspended in 100–500 μl PEMBAL (PEM, 1% bovine serum albumin (BSA), 0·1 M-lysine hydrochloride, 0·5% sodium azide) and 20–50 μl of the cell suspension applied to polylysine-coated slides and allowed to settle for 30 min in a moist chamber. The slides were then placed on ice and flooded with excess methanol at −20°C for 5 min and acetone at −20°C for 30 s. Excess solvent was tipped away and the slides were rehydrated by washing them in phosphate-buffered saline (PBS: 0·14M-NaCl, 2·5mM-KCl, 8mM-Na2HPO4, I1T1M-KH2PO4, pH7·4) for 10min. Slides were then processed for immunofluorescence using anti-tubulin monoclonal antibodies. A number of anti-tubulin monoclonal antibodies were used but only two gave consistently good staining patterns and these were then used extensively. These were YOL 1β4 (Kilmartin et al. 1982), a kind gift from Dr J. Kilmartin, and TAT-1 an antibody raised in this laboratory against trypanosome tubulin (T. Sherwin, unpublished). Both are specific for a-tubulin. First antibody incubations were carried out overnight (16—20 h) in a moist chamber at 25°C and excess antibody was removed by washing three times for 10 min in PBS. FITC-labelled second antibody was then applied to the slides: anti-rat, and anti-mouse IgG-FITC conjugate (Sigma) for YOL 1β4 and TAT-1, respectively. Second antibody incubations were carried out at 25 °C for 4–8 h before washing the slides three times in PBS. Finally, 25 μl of 1μgml diamidino phenylindole (DAPI) was applied to the slides and then washed off in PBS after a brief incubation. After rinsing in distilled water, and removing excess moisture, the slides were mounted in 15μ1 of a mowiol/p-phenylenediamine medium (Johnson & Nogueira Araujo, 1981). Slides were kept in the dark at 4°C for 48 h before observation and photography.

Hyphal germ tubes

Germ tubes were fixed by peeling the cellophane sheet from the serum agar plates and floating on 10ml of 3·7% formaldehyde-PEM in a sealed Petri dish for 90 min at room temperature. Adhering cells were gently washed from the cellophane and the cell suspension processed as above, except that protoplast formation required only 5 min; longer periods resulted in digestion of the cell wall to an unfavourable extent.

Microscopy and photography

Cells were observed using a Zeiss microscope equipped with Planapo ×100, ×63 and Neofluor ×100 lenses and filter sets appropriate for FITC and DAPI observation. Pictures were taken using an Olympus camera and lens and Ilford ×P1 film.

Quantitative analysis

An exponentially growing population of yeast phase cells stained for microtubules was grouped into classes based primarily on the pattern of the microtubule cytoskeleton. The percentages of cells in each class were determined from a total population of 3469. These values were corrected for their position in the cell cycle because of the bias that occurs as a result of the doubling of cell number at the end of the cell cycle (Williams, 1971).

To determine the length of germ tube at which nuclear division occurs yeast cells inoculated onto serum agar were fixed and DAPI stained at intervals of and 4h. At least 1000 cells at each stage were counted and the lengths of true, parallel-sided germ tubes (see Results for definition), from mother cell junction to hyphal apex were measured, using an eyepiece micrometer. For each cell the number of discrete DAPLstained nuclei was recorded.

Nuclear behaviour in yeast cells and genu tubes

A series of micrographs of DAPLstained C. albicans yeast and hyphal cells was obtained showing the nuclear behaviour during the budding cell cycle and germination. During budding growth of the yeast phase the nucleus divides and segregates into the daughter and mother cells (Fig. 1A-D), whereas during germ tube outgrowth, after the nucleus has divided, the distal daughter nucleus migrates along the germ tube (Fig, 1E-F). Reports in the literature on the length of germ tube at which nuclear division occurs are confused (Soli et al. 1978; Gow et al. 1986), probably because of the different conditions used to induce germination. Pseudohyphae, a cell type intermediate between yeasts and hyphae, tend to undergo nuclear division at a shorter germ tube length before reverting to budding growth. In this work we have dealt only with true germ tubes, defined as such by their appearance as parallel-sided, unbudded, short hyphae. In such cells mitosis occurred after germ tubes had reached a length of at least 9 μm. Counts revealed that 50% of germ tubes had undergone nuclear division at a length of 13–14μm (Fig. 2).

Fig. 1.

Phase-DAPI micrographs of the overall cell morphology and nuclear position during the C. albicans yeast cell cycle (A-D) and germ tube development (E,F). The nuclear position is denoted by the very bright fluorescence seen against a phase image of the cell. In A, B, C the undivided nucleus is present in the mother cell whilst in D the mother cell and bud both possess a daughter nucleus. In E the nucleus is located in the original mother cell, whilst in F nuclear division has occurred; one nucleus being located in the mother cell and one in the germ tube. Bar, 5 μm.

Fig. 1.

Phase-DAPI micrographs of the overall cell morphology and nuclear position during the C. albicans yeast cell cycle (A-D) and germ tube development (E,F). The nuclear position is denoted by the very bright fluorescence seen against a phase image of the cell. In A, B, C the undivided nucleus is present in the mother cell whilst in D the mother cell and bud both possess a daughter nucleus. In E the nucleus is located in the original mother cell, whilst in F nuclear division has occurred; one nucleus being located in the mother cell and one in the germ tube. Bar, 5 μm.

Fig. 2.

Relationship between germ tube length and occurrence of nuclear division. The lengths of over 1000 germ tubes were measured and grouped into classes of 2-μm intervals. The midpoints of these classes were plotted against the percentage of cells in that length class that possessed a divided nucleus.

Fig. 2.

Relationship between germ tube length and occurrence of nuclear division. The lengths of over 1000 germ tubes were measured and grouped into classes of 2-μm intervals. The midpoints of these classes were plotted against the percentage of cells in that length class that possessed a divided nucleus.

Microtubule patterns in the yeast cell cycle

The monoclonal antibodies YOL 1β4 and TAT-1 both produced similar staining patterns of filamentous structures in yeast cells of C. albicans, which, because of the specificity of these antibodies for tubulin, were interpreted as microtubules. It is impossible to tell from immunofluorescence data whether such stained structures are single microtubules or bundles of microtubules. It is likely that most of the stained filaments are single microtubules or bundles of a few microtubules, and the term microtubule will be used to describe these structures. Staining of microtubules was in general satisfactory. However, at times some cells exhibited punctate staining patterns. This is likely to result from poor fixation within these cells. Images shown are representative and obtained from observations of thousands of cells stained in over 20 separate immunofluorescence experiments. Each cell was assigned to a particular point in the cell cycle by consideration of the microtubule staining pattern, the mother: bud size ratio and the nuclear configuration as revealed by DAPI staining. It must be noted that the mother: bud size ratio was the most approximate criterion used and it appeared that following bud emergence, the relationship between this ratio, and nuclear division and segregation was an imprecise one. Populations of yeast cells of C. albicans appear to be very heterogeneous in size; however, the pattern of the microtubule staining appeared to be consistent in cells of different sizes. Micrographs of individual cells representative of the major cell-cycle stage are shown in Figs 3, 4. Because the small size of the yeast cells necessitated use of a × 100 objective lens for their observation and photography, the relatively short depth of focus meant that only the single most informative plane of focus was photographed. However, interpretations of microtubule configurations throughout this study reflect information derived from through-focussing. Fig. 3A,B shows typical unbudded cells with a single nucleus and exhibiting a number of apparently randomly orientated, short microtubules. Fig. 3C shows a cell with a single brightly staining dot that colocalizes with the nuclear periphery in a cell with a small bud; the scattered, cytoplasmic microtubules are also present. Interestingly, the cytoplasmic microtubules appear focussed on the neck of the bud. The dot is interpreted as the spindle pole body (SPB) and it acts to organize cytoplasmic microtubules with at least one, occasionally two, directed into the bud (Fig. 3D). At about this stage the background of randomly orientated cytoplasmic microtubules disappears.

Fig. 3.

Immunofluorescence staining of microtubules during the yeast phase cell cycle of C. albicans. Three images of each individual cell are shown. The first (1eft-hand panel) is the immunofluorescence image showing the distribution of microtubules. The second (centre panel) shows the nucleus as defined by the DAPI fluorescence image and the third (right hand-panel) shows the overall morphology of the cell superimposed on the nuclear position using a phase-DAPI exposure (with the exception of Fig. 3E, I, which are simply phase images). Arrows in Fig. 3C,E indicate the unduplicated and duplicated SPBs, respectively. Asterisks in Fig. 4H-K indicate the focus of microtubules of the NAA. Bars, 5 μm.

Fig. 3.

Immunofluorescence staining of microtubules during the yeast phase cell cycle of C. albicans. Three images of each individual cell are shown. The first (1eft-hand panel) is the immunofluorescence image showing the distribution of microtubules. The second (centre panel) shows the nucleus as defined by the DAPI fluorescence image and the third (right hand-panel) shows the overall morphology of the cell superimposed on the nuclear position using a phase-DAPI exposure (with the exception of Fig. 3E, I, which are simply phase images). Arrows in Fig. 3C,E indicate the unduplicated and duplicated SPBs, respectively. Asterisks in Fig. 4H-K indicate the focus of microtubules of the NAA. Bars, 5 μm.

Fig. 4.

Immunofluorescence staining of microtubules during the yeast phase cell cycle of C. albicans. Three images of each individual cell are shown. The first (1eft-hand panel) is the immunofluorescence image showing the distribution of microtubules. The second (centre panel) shows the nucleus as defined by the DAPI fluorescence image and the third (right hand-panel) shows the overall morphology of the cell superimposed on the nuclear position using a phase-DAPI exposure (with the exception of Fig. 3E, I, which are simply phase images). Arrows in Fig. 3C,E indicate the unduplicated and duplicated SPBs, respectively. Asterisks in Fig. 4H-K indicate the focus of microtubules of the NAA. Bars, 5 μm.

Fig. 4.

Immunofluorescence staining of microtubules during the yeast phase cell cycle of C. albicans. Three images of each individual cell are shown. The first (1eft-hand panel) is the immunofluorescence image showing the distribution of microtubules. The second (centre panel) shows the nucleus as defined by the DAPI fluorescence image and the third (right hand-panel) shows the overall morphology of the cell superimposed on the nuclear position using a phase-DAPI exposure (with the exception of Fig. 3E, I, which are simply phase images). Arrows in Fig. 3C,E indicate the unduplicated and duplicated SPBs, respectively. Asterisks in Fig. 4H-K indicate the focus of microtubules of the NAA. Bars, 5 μm.

Careful observation revealed three classes of SPB configuration. First, a single roughly spherical dot (Fig. 3C,D) interpreted as an unduplicated SPB body. Second, two such structures with a distinctly less intensely stained area between them, interpreted as defining the duplicated SPB (Fig. 3E); and third, a very brightly stained, slightly oval structure (Fig. 3F,G), which may represent an intermediate during SPB replication or a double SPB where the individual elements are too bright and/or too close to be resolved separately. The duplicated SPB organizes an often extensive system of cytoplasmic microtubules with at least one directed into the growing bud (Fig. 3FH). The appearance of the SPB is correlated quite closely with the appearance of the early bud or evagination as evidenced by counts of cells at this stage of the cell cycle. These counts reinforce the conclusion that a major change in microtubule organization accompanies bud formation, 97% of all cells with short, scattered microtubules and no SPB staining are unbudded, whilst over 87% of cells that show a single SPB have buds.

In cells that were further into the cell cycle, the two spindle pole bodies have moved apart, elaborating between them a short, rod-like, intranuclear spindle with cytoplasmic microtubules nucleated from either end. At least one of the cytoplasmic microtubules continues to be directed into the expanding bud (Fig. 3I-L). The cell in Fig. 3I exhibits a very short spindle yet the nucleus, as revealed by the DAPI staining, extends into the bud ahead of the spindle. This pattern was only rarely seen, but has been documented previously by electron microscopy in C. albicans (Tanaka et al. 1985) and Sacchar-oniyces cerevisiae (Peterson & Ris, 1976). The sequence of cells in Fig. 3I-L demonstrates how the early spindle progresses from an orientation orthogonal to the bud to line up parallel to this axis in preparation for spindle elongation into the bud. The spindle appears to elongate into the bud concomitant with segregation of the nuclear DNA. During these stages each spindle pole continues to be associated with around 1–3 cytoplasmic microtubules (Fig. 4AG).

Large doublets cells, where the daughter cell has attained at least 0-75 of the mother cell size and in which nuclear division is complete, exhibit an unusual but consistent microtubule configuration. The configuration is characterized by the focussed congression of a group of short microtubules in the neck region between the mother and daughter cell, representing a level of microtubule organization independent of that imposed by the SPBs. This neck-associated array (NAA) of microtubules is a consistent and novel feature of the C. albicans yeast cell cycle. The NAA is illustrated in post-anaphase cells in Fig. 4H-L. These late anaphase/telophase cells may also possess other cytoplasmic microtubules focussed onto the SPBs. The microtubules of the NAA are sometimes, however, observed in cells prior to DNA segregation, and Fig. 5 illustrates this phenomenon. Cells in Fig. 5A-D show an undivided nucleus still present in the mother cell with a brightly staining short mitotic spindle (or merely a duplicated or unduplicated SPB). However, each cell exhibits an array of short microtubules focussed on the neck region, with little evidence of involvement of the SPBs in their nucleation.

Fig. 5.

Examples of the NAA of microtubules present in cells prior to DNA segregation. Each cell possess a congression of microtubules at the neck (asterisk) and a brightly staining short spindle or SPB (arrow). Bar, 5 μm.

Fig. 5.

Examples of the NAA of microtubules present in cells prior to DNA segregation. Each cell possess a congression of microtubules at the neck (asterisk) and a brightly staining short spindle or SPB (arrow). Bar, 5 μm.

In order to obtain quantitative information on the various periods of microtubule organization occurring during the cell cycle, 3469 cells were assessed and assigned to one of eight cell cycle classes, as shown diagrammatically in Fig. 6. The percentage of cells in each class was obtained, and the values used to calculate the fraction of a unit cell cycle occupied by that form of microtubule organization (after adjusting for the biasing that occurs because of the doubling of cell number at the end of the cell cycle; Williams, 1971). The data obtained suggest that a C. albicans yeast cell spends nearly half of its cell cycle time in the unbudded state with only short, randomly orientated microtubules in its cytoplasm. A key event in the cycle is the appearance of a tubulin-positive SPB staining pattern that correlated closely with bud emergence. Fig. 6 also shows how the period of early spindle elongation occurs over a relatively short fraction of the cell cycle.

Fig. 6.

Fraction of the cell cycle occupied by cells with particular microtubule configurations; 3469 cells were scored into classes typified by the diagrammatic representations shown in the figure. Fraction times were calculated as described in Materials and methods. The first class includes all unbudded cells, of which 23% showed no staining pattern at all.

Fig. 6.

Fraction of the cell cycle occupied by cells with particular microtubule configurations; 3469 cells were scored into classes typified by the diagrammatic representations shown in the figure. Fraction times were calculated as described in Materials and methods. The first class includes all unbudded cells, of which 23% showed no staining pattern at all.

Immunofluorescence staining of microtubules in germ tubes

Yeast cells plated onto serum agar were allowed to develop germ tubes and were then fixed and processed for the immunofluorescence detection of microtubules. The location of the nucleus was revealed by staining with DAPI. Short germ tubes that were one to two times the length of the mother cell displayed a pattern of microtubule staining similar to that in the early stages of the yeast budding cycle. In general, cells with these short germ tubes exhibited a tubulin-positive duplicated SPB, which organized cytoplasmic microtubules directed into the outgrowing germ tube (Fig. 7A). However, in contrast to events in the budding cycle, the duplicated SPB moved out of the mother cell into the hypha and was positioned at the leading edge of the nucleus, with the associated microtubules extending either towards the hyphal tip and/or back towards the mother cell (Fig. 7B-C). These microtubules sometimes but not always extend to the very tip of the germ tube. The nucleus migrates out into the germ tube, often vacating the mother cell entirely (Fig. 7C), though in some cases a proportion of nuclear material remains there. The SPBs in most germ tubes are positioned at the leading end of the nucleus, though occasionally the SPB is on the trailing edge.

Fig. 7.

Microtubule immunofluorescence of C. albicans germ tubes during nuclear migration (1ayout as in Figs 3-4). A. A short germ tube with the nucleus and SPB (arrowed) in the mother cell. B. A germ tube where the nucleus has partially migrated out into the hypha with the SPB at the leading edge of the nucleus. C. A germ tube where the nucleus has completely migrated out of the mother cell into the hypha. Arrows mark the position of the SPB in B and C and are in equivalent positions on all three images. Bar, 5 μm.

Fig. 7.

Microtubule immunofluorescence of C. albicans germ tubes during nuclear migration (1ayout as in Figs 3-4). A. A short germ tube with the nucleus and SPB (arrowed) in the mother cell. B. A germ tube where the nucleus has partially migrated out into the hypha with the SPB at the leading edge of the nucleus. C. A germ tube where the nucleus has completely migrated out of the mother cell into the hypha. Arrows mark the position of the SPB in B and C and are in equivalent positions on all three images. Bar, 5 μm.

The duplicated SPBs move apart, forming between them an intranuclear spindle (Fig. 8A-B). This spindle in the hyphal phase of C. albicans has the general appearance described earlier for the spindle of the yeast cell. However, it has a unique characteristic in that it elongates to an extreme length. In yeast cells the spindle has a maximum length of around 7 pm (essentially almost the complete length of the telophase doublet cell). However, a clear switch occurs during the hyphal phase so that mitotic spindles lengthen to around 12-20 pm (Fig. 8A-B). This long spindle phenomenon associated with the hyphal growth phase is seen in both the first and subsequent mitotic divisions (Fig. 8C). During these mitotic periods in the hyphal cell compartments, spindlepole-associated cytoplasmic microtubules extend up and down the germ tube parallel to the long axis of the hypha. During the first division, after the proximal nucleus has segregated back into the mother cell, a septum forms at a variable position with respect to the mother cell-germ tube junction. At this point the spindle depolymerizes and the cytoskeleton is reduced to scattered remnants of the cytoplasmic and spindle microtubules.

Fig. 8.

Microtubule immunofluorescence of C. albicans germ tubes during mitosis (1ayout as in Figs 3-4). A,B. Germ tubes undergoing the first nuclear division; note the long spindles. C. A germ tube during the second mitosis. The mother cell contains a nucleus from the first division, a second mitosis is almost complete and the spindle has started to breakdown in the centre. The poles of the spindles are indicated by arrows. Bar, 5μm.

Fig. 8.

Microtubule immunofluorescence of C. albicans germ tubes during mitosis (1ayout as in Figs 3-4). A,B. Germ tubes undergoing the first nuclear division; note the long spindles. C. A germ tube during the second mitosis. The mother cell contains a nucleus from the first division, a second mitosis is almost complete and the spindle has started to breakdown in the centre. The poles of the spindles are indicated by arrows. Bar, 5μm.

Before this study, evidence of the microtubular cytoskeleton of C. albicans was confined to a handful of incidental electron-microscope studies of the yeast phase, which revealed an intranuclear spindle similar to that in S. cerevisiae (Suzuki et al. 1986; Soil, 1985) but shed little light on other aspects of this organelle and provided no information on the dynamics of the cytoskeleton during the cell cycle in either cell form. The protocol developed for the immunofluorescent staining of microtubules in C. albicans was based on that pioneered in S. cerevisiae and S. uvarum by Kilmartin & Adams (1984). Its extension to the C. albicans yeast cell cycle reveals some interesting and novel features of the microtubule organization in this organism. Singlet cells with no observable bud exhibit a pattern of randomly arranged short microtubules: no obvious organizing structure is evident in these cells. This class of cells does not appear to exist in Saccharo-ntyces, where a tubulin-positive SPB persists throughout the cycle (Kilmartin & Adams, 1984). Thus, Candida. appears to resemble more closely Schizosaccharoniyces pombe, where interphase cells possess an array of long, unconnected, cytoplasmic microtubules and do not show a tubulin-positive SPB until the start of mitosis (Hagan & Hyams, 1988).

In Candida the appearance of a very brightly stained, single dot, the presumably unduplicated SPB, appears to coincide with the emergence of the early bud. The tubulin-positive double dot (duplicated) SPB pattern shown in Candida, is therefore a post-bud-emergence phenomenon. The duplicated SPB image has been detected in Saccharomyces (Kilmartin & Adams, 1984) and Sch. pombe (Hagan & Hyams, 1988). Byers & Goetsch (1975), however, found that SPB duplication in S. cerevisiae is correlated approximately with the initiation of bud outgrowth. This suggests that SPB duplication occurs later in the cell cycle in Candida than Saccharomyces.

Cytoplasmic microtubules extend into the bud from the SPB(s) as well as into the mother cell cytoplasm. However, in the light of evidence from Saccharomyces, an intact microtubular cytoskeleton appears not to be essential for events involved in yeast bud emergence (Byers & Goetsch, 1974; Quinlan et al. 1980; Pringle et al. 1986).

After SPB duplication, the next definable stage is the formation of a short rod-like spindle between the SPBs. Electron microscopy in S. cerevisiae has shown that a short spindle approximately 1 μm long is formed when the bud reaches 0·35 times the mother cell diameter. This spindle persists for 65% of the budded phase until the bud: mother cell ratio reaches unity, at which point the nucleus migrates to the neck and the spindle subsequently undergoes rapid elongation to 6–8 μm (Byers & Goetsch, 1975). In Candida, yeast cells, our immunofluorescence and DAPI staining observations suggest that in most cases the spindle extends to a length of up to 2·5 μm without the nuclear material being located actually in the neck. In most cells movement of the nucleus through the neck was closely correlated with further spindle elongation.

As the spindle enlarges its orientation changes from being perpendicular to being parallel to the bud axis in preparation for spindle elongation into the bud. The often long cytoplasmic microtubules associated with either end of the spindle at this time may play a role in orientating the nucleus along the critical mother/bud axis, so facilitating efficient nuclear segregation to each cell.

The appearance of an array of short cytoplasmic microtubules in the mother and daughter cell focussed in the neck region, the neck-associated array (NAA), seems to be a characteristic feature of C. albicans yeast cells. No equivalent structure has been observed in Saccharomyces in any of the numerous reports of yeast microtubule immunofluorescence. However, this pattern of NAA microtubules late in the cell cycle may correspond to the post-anaphase array (PAA) observed in Sch. pombe by Hagan & Hyams (1988). In Sch. pombe, the PAA appears after nuclear division has occurred, as a double dot structure, in the centre of the cell where a septum will develop later. The PAA serves to reorganize the interphase array of microtubules after nuclear division. Thus there are distinct similarities between the Candida NAA and the PAA of Sch. pombe. However, our observations suggest that in Candida the focussed array of cytoplasmic microtubules does not appear to be absolutely restricted to post-mitotic cells. Microtubule configurations similar to the NAA can be seen in budded cells with undivided nuclei well before anaphase. As this focussed array of microtubules may exist transiently in cells before mitosis, we have adopted the term neck-associated array, which merely describes its location. The Sch. pombe PAA arises de novo, since no cytoplasmic microtubules are present immediately prior to its appearance. The origin of the NAA in Candida is less clear, since cytoplasmic microtubules are present immediately prior to its formation. The NAA may act as a cytoplasmic organizing centre and actually nucleate the assembly of the microtubules. Conversely, the NAA may merely represent a congression of previously polymerized microtubules to the neck region. Actin granules are found at the neck during bud emergence and later during cytokinesis in Saccharomyces (Kilmartin & Adams, 1984), Schizosaccharomyces (Marks et al. 1986) and Candida (Anderson & Soli, 1986). Such positions correlate closely with sites of new cell wall synthesis. It is difficult to argue for a universal role for the NAA microtubules in septum formation, since they have not been detected in Saccharomyces. Moreover, the function of the PAA of Sch. pombe is independent of septum formation (Hagan & Hyams, 1988). Thus the role of these discrete microtubule arrays in cell cycle events remains to be ascertained.

Numerical analysis of microtubule staining patterns in an asynchronous population of Candida yeast cells, corrected for age distributions within a cell cycle, reveals that nearly half of all cells exhibit an interphase arrangement of their microtubular cytoskeleton. This is more than the 19% uncorrected estimate for Saccharomyces (Kilmartin & Adams, 1984), but much less than the corrected value for Sch. pombe of 82·2% (Hagan & Hyams, 1988). Thus, as in Saccharomyces, the mitotic spindle is present in Candida for a relatively large part of the cell cycle (≈ 0·5).

One of the characteristic features of germ tube outgrowth in Candida is the migration of the nucleus out of the mother cell into the hypha before nuclear division. This has been documented by Gow et al. (1986), but this contradicts other reports (Soil et al. 1978) that suggest that nuclear division occurs in the mother cell-germ tube junction when the hyphal length is approximately equal to the mother cell diameter. We here confirm the former view that the germ tube can grow to a considerable length (≈ 13–14 μm) before nuclear division occurs in the germ tube itself. This discrepancy of opinion is almost certainly due to differences in the mode of germ tube induction; the pH-regulated induction system used by Soli et al. (1978) results, in our hands, in a preponderance of what are more accurately called pseudohyphae.

The migration of the nucleus into the hypha during germ tube outgrowth is correlated closely with the movement of the duplicated SPB at the leading end of the nucleus. The SPBs are associated with long cytoplasmic microtubules that run up and down the germ tube. This microtubular arrangement is broadly similar to the early stages of the yeast cell cycle and the most obvious difference is that the formation of a spindle between the two SPBs is delayed whilst the nucleus moves into the germ tube. Further events in the hyphal microtubule cytoskeleton then essentially mirror those that occur in the yeast cell cycle, though the direction of spindle elongation is along the long axis of the hypha and necessarily back into the mother cell.

Directed nuclear movements are common features of morphogenetic events in the fungi. A number of reports have produced evidence that microtubules play some part in these movements (Oakley & Morris, 1980; Herr & Heath, 1982; Heath, 1982; Oakley & Reinhart, 1985; McKerracher & Heath, 1987). The position of the SPB and its associated cytoplasmic microtubules at the proximal end of the migrating nucleus (mother cell to germ tube) provides circumstantial evidence for a force mediated by interaction of the microtubules with the cytoplasm, as has been suggested in Schizophyllum (Runeberg et al. 1986). In Candida hyphal forms, cytoplasmic microtubule nucleation seems to be mediated mainly via the cytoplasmic face of the SPB. However, other apparently free microtubules can sometimes be seen in the cytoplasm. We have found no evidence to support the notion of a cytoplasmic organizing centre for microtubules located at the hyphal tip, as seen in some fungi (Hoch & Staples, 1985).

An unusual finding from our comparative study is that the hyphal phase mitotic spindle is unexpectedly long, up to 20 gm in length compared to a maximum of 8 pm in yeast cells. These long spindles observed in germ tubes may be necessary to position the daughter nuclei in the centre of the daughter hyphal compartments. Septum formation, a convenient marker for the end of mitosis, occurs at an apical germ tube compartment length of 32–40 μm (Gow & Gooday, 1982); thus a spindle 16–20 μm in length in the centre of such a cell would position the daughter nuclei in the centre of each future compartment. Both divisions are mitotic and there is unlikely to be any genetic difference between the nuclei. Thus, spindle length appears to be open to influence by the cell type morphology. In this context it is interesting that an increase in spindle length has been observed in 5. pombe when the length of the cell is increased genetically (Hagan & Hyams, personal communication). It may be important, therefore, to view the mitotic spindle as functioning not only in chromosome segregation but also in positioning the two daughter nuclei.

We thank John Kilmartin for his generous gift of the YOL 1β4 antibody, and Iain Hagan, Jeremy Hyams and colleagues in Canterbury for many useful discussions. R.B. is a recipient of a SERC CASE award held in conjunction with Dr S. Watts, Wellcome Research Laboratories, Beckenham, Kent.

Anderson
,
J. M.
&
Soll
,
D. R.
(
1986
).
Differences in actin localization during bud and hypha formation in the yeast Candida albicans
.
J. gen. Microbiol
.
132
,
2035
2047
.
Byers
,
B.
&
Goetsch
,
L.
(
1974
).
Duplication of spindle plaques and integration of the yeast cell cycle
.
Cold Spring Harbour Symp. quant. Biol
.
38
,
123
131
.
Byers
,
B.
&
Goetsch
,
L.
(
1975
).
Behaviour of spindles and spindle plaques in the cell cycle and conjugation of Saccharomyces cerevisiae
.
J. Bad
.
124
,
511
523
.
Gow
,
N. A. R.
&
Gooday
,
G. W.
(
1982
).
Growth kinetics and morphology of colonies of the filamentous form of Candida albicans
.
J. gen. Microbiol
.
128
,
2187
2194
.
Gow
,
N. A. R.
,
Henderson
,
G.
&
Gooday
,
G. W.
(
1986
).
Cytological interrelationships between the cell cycle and duplication cycle of Candida albicans
.
Microbios
47
,
97
105
.
Hagan
,
I. M.
&
Hyams
,
J. S.
(
1988
).
The use of cell division cycle mutants to investigate the control of microtubule distribution in the fission yeast Schizosaccharomvces pombe
.
J. Cell Sci
.
89
,
343
353
.
Heath
,
I. B.
(
1982
).
The effect of nocodazole on the growth and ultrastructure of the fungus Saprolagnia ferax. Evidence against a simple mode of action
.
In Microtubules in Microorganisms
(ed.
P.
Capuccinelli
&
N. R.
Norris
), pp.
275
311
.
New York, Basel
:
Dekker
.
Herr
,
F. B.
&
Heath
,
M. C.
(
1982
).
The effects of antimicrotubule agents on organelle positioning in the cowpea rust fungus, Uromyces phaseoli var
.
vignae. Expl Mycol
.
6
,
15
24
.
Hoch
,
H. C.
&
Staples
,
R. C.
(
1985
).
The microtubule cytoskeleton in hyphae of Uromyces phaseoli germlings: its relationship to the region of nucleation and to the F-actin cytoskeleton
.
Protoplasma
124
,
112
122
.
Johnson
,
G. D.
&
Nogueira Araujo
,
G. M. Dec
. (
1981
).
A simple method of reducing the fading of immunofluorescence during microscopy
.
J. Immun. Meth
.
43
,
349
350
.
Kilmartin
,
J. V.
&
Adams
,
A. E. M.
(
1984
).
Structural rearrangements of tubulin and actin during the cell cycle of the yeast Saccharomyces
.
J. Cell Biol
.
98
,
922
933
.
Kilmartin
,
J. V.
,
Wright
,
B.
&
Milstein
,
C.
(
1982
).
Rat monoclonal antitubulin antibodies derived by using a new nonsecreting rat cell line
.
J. Cell Biol
.
93
,
576
582
.
Lee
,
K. L.
,
Buckley
,
H. R.
&
Campbell
,
C. C.
(
1975
).
An amino acid liquid synthetic medium for the development of mycelial and yeast forms of Candida albicans
.
Sabouraudia
13
,
148
153
.
Marks
,
J.
,
Hagan
,
I. M.
&
Hyams
,
J. S.
(
1986
).
Growth polarity and cytokinesis in fission yeast: the role of the cytoskeleton
.
J. Cell Sci. Suppl.
5
,
229
241
.
Mckerracher
,
L. J.
&
Heath
,
I. B.
(
1987
).
Cytoplasmic migration and intracellular organelle movements during tip growth of fungal hyphae
.
Expl Mycol
.
11
,
79
100
.
Mitchell
,
L. H.
&
Soll
,
D. R.
(
1979
).
Commitment to germ tube or bud formation during release from stationary phase in Candida albicans
.
Expl Cell Res
.
120
,
167
179
.
Novick
,
P.
&
Botstein
,
D.
(
1985
).
Phenotypic analysis of temperature-sensitive yeast actin mutants
.
Cell
40
,
405
416
.
Oakley
,
B. R.
&
Morris
,
N. R.
(
1980
).
Nuclear movement is β-tubulin dependent in Aspergillus nidulans
.
Cell
19
,
255
262
.
Oakley
,
B. R.
&
Reinhart
,
J. E.
(
1985
).
Mitochondria and nuclei move by different mechanisms in Aspergillus nidulans
.
J. Cell Biol
.
101
,
2392
2397
.
Peterson
,
J. B.
&
Ris
,
H.
(
1976
).
Electron-microscopic study of the spindle and chromosome movement in the yeast Saccharomyces cerevisiae
.
J. Cell Sci
.
22
,
219
242
.
Pringle
,
J. R.
,
Lillie
,
S. H.
,
Adams
,
A. E. M.
,
Jacobs
,
C. W.
,
Haarer
,
B. K.
,
Coleman
,
K. G.
,
Robinson
,
J. S.
,
Bloom
,
L.
&
Preston
,
R. A.
(
1986
).
Cellular morphogenesis in the yeast cell cycle
.
In Yeast Cell Biology
(ed.
J.
Hicks
), pp.
193
209
.
New York
:
Alan R. Liss
.
Quinlan
,
R. A.
,
Pogson
,
C. I.
&
Gull
,
K.
(
1980
).
The influence of the microtubule inhibitor, methyl benzimidazole-2-yl-carbamate (MBC) on nuclear division and the cell cycle in Saccharomyces cerevisiae
.
J. Cell Sci
.
46
,
341
352
.
Runeberg
,
P.
,
Raudaskoski
,
M.
&
Virtanen
,
I.
(
1986
).
Cytoskeletal elements in the hyphae of the homobasidiomycete Schizophyllum commune visualized with indirect immunofluorescence and NBD-phallacidin
.
Eur.J. Cell Biol
.
41
,
25
32
.
Soll
,
D. R.
(
1985
).
Candida albicans
.
In Fungal Dimorphism vrith Emphasis on Fungi Pathogenic for Humans
(ed.
P. J.
Szaniszlo
), pp.
167
195
.
New York
:
Plenum
.
Soll
,
D. R.
,
Bedell
,
G.
,
Thiel
,
J.
&
Brummel
,
M.
(
1981
).
The dependency of nuclear division on volume in the dimorphic yeast Candida albicans
.
Expl Cell Res
.
133
,
55
62
.
Soll
,
D. R.
&
Herman
,
M. A.
(
1983
).
Growth and the inducibility of mycelium formation in Candida albicans: a single-cell analysis using a perfusion chamber. J. gen
.
Microbiol
.
129
,
2809
2824
.
Soll
,
D. R.
,
Stasi
,
M.
&
Bedell
,
G.
(
1978
).
The regulation of nuclear migration and division during pseudo-mycelium outgrowth in the dimorphic yeast Candida albicans
.
Expl Cell. Res
.
116
,
207
215
.
Suzuki
,
T.
,
Kanbe
,
T.
,
Kuroiwa
,
T.
&
Tanaka
,
K.
(
1986
).
Occurrence of ploidy shift in a strain of the imperfect yeast Candida albicans
.
J. gen. Microbiol
.
132
,
443
453
.
Tanaka
,
K.
,
Kanbe
,
T.
&
Kuroiwa
,
T.
(
1985
).
Three dimensional behaviour of mitochindria during cell division and germ tube formation in the dimorphic veast Candida albicans
.
J. Cell Sci
.
73
,
207
220
.
Williams
,
F. M.
(
1971
).
Dynamics of microbial populations
.
In Systems Analysis and Simulation Ecology
, vol.
1
(ed.
B. C.
Pattern
), pp.
247
262
.
New York
:
Academic Press
.