Cell Wall-Related Bionumbers and Bioestimates of Saccharomyces cerevisiae and Candida albicans

Abstract

Bionumbers and bioestimates are valuable tools in biological research. Here we focus on cell wall-related bionumbers and bioestimates of the budding yeast Saccharomyces cerevisiae and the polymorphic, pathogenic fungus Candida albicans. We discuss the linear relationship between cell size and cell ploidy, the correlation between cell size and specific growth rate, the effect of turgor pressure on cell size, and the reason why using fixed cells for measuring cellular dimensions can result in serious underestimation of in vivo values. We further consider the evidence that individual buds and hyphae grow linearly and that exponential growth of the population results from regular formation of new daughter cells and regular hyphal branching. Our calculations show that hyphal growth allows C. albicans to cover much larger distances per unit of time than the yeast mode of growth and that this is accompanied by strongly increased surface expansion rates. We therefore predict that the transcript levels of genes involved in wall formation increase during hyphal growth. Interestingly, wall proteins and polysaccharides seem barely, if at all, subject to turnover and replacement. A general lesson is how strongly most bionumbers and bioestimates depend on environmental conditions and genetic background, thus reemphasizing the importance of well-defined and carefully chosen culture conditions and experimental approaches. Finally, we propose that the numbers and estimates described here offer a solid starting point for similar studies of other cell compartments and other yeast species.

INTRODUCTION

In their paper “A feeling for the numbers in biology,” Phillips and Milo (1) present a convincing case for a more quantitative approach in biological research. An important advantage of moving from a qualitative to a more quantitative understanding of a biological process is that one learns to view one's observations from a different perspective. A feeling for the numbers involved assists in prioritizing hypotheses and selecting better experimental approaches and leads to surprising insights. For industrial purposes, genetic engineering, and synthetic biology, a quantitative approach becomes even more important, whereas for modeling of biological processes, accurate bionumbers and bioestimates are crucial. Numerous interesting bionumbers can be found at the website http://bionumbers.hms.harvard.edu/ (2). The cell wall of yeasts accounts for up to 30% of the cellular biomass on a dry weight basis and thus represents a substantial metabolic investment of the cell. Here we focus on cell wall-related bionumbers and bioestimates of two important fungi: the workhorse Saccharomyces cerevisiae and the human pathogen Candida albicans.

CELLULAR DIMENSIONS OF SACCHAROMYCES CEREVISIAE

The shape of the parent cell and the growing bud of the yeast S. cerevisiae (and of C. albicans when growing in the yeast form) approximates a prolate ellipsoid. This allows accurate estimation of the volume and surface area of parent cell and bud by measuring their length (major axis) and width (minor axis). For example, the volume V = πab²/6 (where a is the major and b is the minor axis) and is usually expressed in μm³ or fl (1 μm³ = 1 fl = 10⁻¹⁵ liter). The online calculator “Ellipsoid” at http://planetcalc.com/149/ allows rapid calculation of both volume and surface area of ellipsoids.

The cellular dimensions of parent cells of S. cerevisiae, which is usually grown at 30°C, have been extensively investigated. They vary widely and depend on cell ploidy, growth rate, and nutrient status and also on turgor pressure and the number of buds formed by the parent cell. Table 1 shows that in exponential-phase cultures growing in rich medium, the average volume of parent cells is proportional to ploidy, increasing from ∼44 μm³ for haploid cells to ∼244 μm³ for hexaploid cells (3). This agrees with the median values of 42 μm³ and 82 μm³ obtained with haploid and diploid cells, respectively (4); see also reference 5, in which average volumes of 56 and 95 μm³ are presented for haploid and diploid cells, respectively. Table 1 further shows that the ratio of surface area and volume is inversely correlated with cell ploidy, decreasing by ∼40% (from 1.38 μm⁻¹ for haploid cells to 0.79 μm⁻¹ for hexaploid cells). This is consistent with the observation that in diploid cells the transcript levels of genes that are involved in cell wall formation are generally lower than in haploid cells (6).

TABLE 1.

Effect of cell ploidy on dimensions, volume, biomass, surface area, and the ratio of surface area over volume of S. cerevisiae parent cells^a

Ploidy	Length (μm)	Width (μm)	Vol (μm3)	Biomass (dry wt, pg)	Surface area (μm2)	Surface/vol (μm−1)
1	4.76	4.18	44	16.5	60	1.38
2	6.16	5.06	83	31.2	92	1.12
3	7.66	5.97	143	53.9	134	0.93
4	7.97	6.20	161	60.5	144	0.90
5	9.42	6.64	218	82.1	179	0.82
6	10.1	6.80	245	92.3	194	0.79

^aThis table is based on data from plate-grown cells (3). The biomass (dry weight) was calculated by multiplying the volume with the density (1.11) (88) to obtain the biomass (wet weight) of the cell and multiplying the obtained value with the dry weight fraction (0.34) of the wet weight (89).

The mean volume of parent cells of S. cerevisiae not only increases with increasing ploidy but, as has been shown for diploid cells, also positively correlates with the specific growth rate (Table 2), increasing from 29 μm³ at a specific growth rate of 0.045 h⁻¹ (doubling time of ∼15 h) to 95 μm³ at 0.46 h⁻¹ (doubling time of ∼1.5 h) (5, 7); a similar trend has been observed for haploid cells (8, 9). Interestingly, reduced growth resulting from either nutrient limitation or gene mutation is accompanied not only by a decrease in cellular volume but also by increased tolerance toward various stress conditions such as heat and oxidative stress (10). Similarly, post-exponential-phase yeast cells become rapidly more resistant to the wall-degrading enzyme preparation Zymolyase (11). This raises the question of whether nutrient sensing pathways, such as the Ras-cyclic AMP-protein kinase A signaling pathway and the Snf1 and TORC1 pathways (12), control specific cell wall properties, depending on nutrient availability. Finally, the inverse relationship between surface area and cell volume probably favors the use of small (haploid) cells in cell surface engineering when the primary goal is to maximize the number of surface-located heterologous proteins per unit of biomass. Conversely, the use of large (polyploid) cells is probably preferred when one is interested in maximizing intracellular contents.

TABLE 2.

Cellular volume of diploid S. cerevisiae parent cells is positively correlated with the specific growth rate^a

Specific growth rate (h−1)	Vol (μm3)	Biomass (dry wt, pg)
0.045	29	11
0.10	31	12
0.15	35	13
0.25	48	18
0.30	61	23
0.35	69	26
0.46	95	36

^aThe first six data points are from reference 7, whereas the last point is from reference 5. Biomass (dry weight) was calculated as described for Table 1.

Turgor pressure is another, often overlooked factor that can strongly affect cell dimensions. The turgor pressure in exponential-phase yeast cells cultured under hypo-osmotic conditions is estimated to be ∼0.6 MPa (13), similar to the tire pressure of racing bikes. Yeast and hyphal walls are elastic and, owing to the turgor pressure exerted on the walls when the cells are cultured in hypo-osmotic media, do expand considerably, resulting in turgid cells (13). When yeast cells are transferred to hyperosmotic solutions, their volume immediately decreases by 40 to 60% (14, 15). Consistent with this, cells grown in media of increasing osmotic strength become smaller and the porosity of the wall decreases (16). Cell contraction in a hyperosmotic medium is reversible: when the cells are transferred back to hypo-osmotic conditions, they rapidly regain their original volume (14). Cell contraction in hyperosmotic solutions has also been observed in several other yeasts, including Candida spp., the fission yeast Schizosaccharomyces pombe, and the basidiomycetous yeast Cryptococcus laurentii (14, 17), indicating that cell wall extension caused by turgor pressure is a general phenomenon. Cells contract also during fixation. For example, the volume of S. cerevisiae yeast cells decreases overnight by about 30% upon fixation in buffered glutaraldehyde (18). This means that volume measurements based on fixed cells (19, 20) tend to (seriously) underestimate the original volume. These observations raise the intriguing question of whether the invaginations of the plasma membrane that are often observed in electron microscopy pictures might be (partially) attributed to inward folding of the plasma membrane that becomes too wide for the cell owing to cell contraction during fixation and subsequent processing steps. Finally, the volume of parent cells has been shown to increase during each budding cycle by about 20% (21–23).

CELLULAR DIMENSIONS OF CANDIDA ALBICANS

C. albicans is a polymorphic fungus that can grow as budding yeast and as pseudohyphae, but it can also form authentic hyphae with an apical body and perforated septa (24, 25). It mainly occurs as diploid cells, but haploid and tetraploid cells have also been observed (26). As C. albicans is associated with warm-blooded animals, we preferentially use bionumbers obtained at 37°C. Similar to those of S. cerevisiae yeast cells, the cellular dimensions of C. albicans yeast cells vary widely (20). Although fixed cells were used in this study, the trends were similar to those obtained with S. cerevisiae cells, including a positive correlation between cell size and growth rate and a slight increase in parental cell volume with each budding cycle (20). Measurements of live, exponential-phase parent yeast cells grown in a defined salt medium at 25°C resulted in an average calculated volume of 88 μm³ (27) and in a slightly higher volume, 95 μm³, when the cells originated from an exponential-phase culture grown in rich medium at 37°C (28). Both values are similar to the values obtained with S. cerevisiae diploid parent cells originating from an exponential-phase culture grown in rich medium (3–5). Hyphal width of C. albicans can also vary widely (19). Under optimal growth conditions, a diameter of 2.6 μm has been observed for germ tubes and a diameter of 3.4 μm for mature hyphae (29).

QUANTITATIVE ANALYSIS OF THE CELL WALLS OF S. CEREVISIAE AND C. ALBICANS

The general compositions of the walls of S. cerevisiae and C. albicans yeast cells are similar (Table 3). On a more detailed level, however, substantial differences emerge. For example, their wall proteomes differ sharply and the N-linked carbohydrate side chains of mannoproteins of C. albicans contain β-mannosides (30), which are lacking in S. cerevisiae. The molecular organizations of the walls of both species, including hyphal walls of C. albicans, are also similar (31, 32). Their walls consist of an external mannoprotein coat shielding an inner polysaccharide network that consists of β-glucans and chitin and is responsible for the mechanical strength of the wall. It has been proposed that the external protein coat consists of a basal layer of small, glycosylphosphatidylinositol (GPI)-modified wall proteins interlinked by disulfide bridges that is interspersed with much longer, fibrillar GPI-modified proteins (31, 33) that measure up to ∼150 nm in the case of C. albicans (17). This could explain why yeast walls are sometimes described as having a three-layered structure (34, 35). A similar three-layered wall structure has been described for Exophiala dermatitidis (36, 37). For a more detailed description of the molecular architecture of the walls of both S. cerevisiae and C. albicans and their wall proteomes, the reader is referred to recent reviews (31–34, 38, 39).

TABLE 3.

Macromolecular compositions of the walls of haploid yeast cells of S. cerevisiae and diploid yeast cells of C. albicans grown in rich medium at 30°C (62)

Component	S. cerevisiaea	C. albicansa
Mannoprotein
Protein (%)	4.0	3.5
Mannan (%)	34.2	26.6
Mannan/protein ratio	8.6	7.6
β-Glucanb (%)	60.3	64.0
β-1,3-Glucan/β-1,6-glucan ratio	3.8	2.5
Chitin (%)	1.4	4.2

^aThe wall dry weight of S. cerevisiae is ∼21% of the total biomass (67). For C. albicans yeast cells, the same value is assumed.

^bβ-1,3-Glucan plus β-1,6-glucan.

The exact cell wall composition can vary considerably (32, 40). The cell rapidly responds to changing environmental conditions by adapting the composition of the wall proteome of newly formed walls, by changing the thickness of the skeletal layer of the new walls (17, 40), and by varying the incorporation of chitin (32, 38, 41, 42). There is no clear evidence that the composition of existing walls is drastically changed in such cases, except probably for increased chitin incorporation upon cell surface stress (32). The wall proteome of C. albicans has been extensively studied both qualitatively and quantitatively (39, 43). It shows an amazing adaptability that helps the cell to adequately respond to changes in, for example, pH, temperature, hypoxic conditions, carbon source, and iron availability and to plasma membrane and wall stress (43–49). The wall proteome also reflects morphotype and contains yeast- and hypha-specific wall proteins (50). In addition, the wall proteome is involved in controlling cell wall porosity and cell adhesiveness and hydrophobicity, n recognizing specific substrates, and in forming biofilms (11, 16, 17, 39, 49, 51).

PHYSICAL PROPERTIES OF THE CELL WALL

Wall thickness varies considerably. Transmission electron micrographs suggest that the variations in thickness of the walls are largely due to variable thickness of the internal skeletal layer. Although a sharp demarcation is not always present between the inner and outer wall layers, the inner polysaccharide layer can usually be recognized because it is more electron transparent than the outer protein layer. When C. albicans yeast cells are transferred to a hyphal-growth-inducing medium, one or more germ tubes emerge from the parent cell. Interestingly, the wall of the emerging hyphae is considerably thinner than the walls of the parent cell (52). However, it is unknown if the thinner wall of the emerging hypha is an inherent quality of hyphal walls or resulted from the composition of the hyphal-growth-inducing medium used in this study. The dependence of the thickness of the inner transparent layer of the walls of C. albicans yeast cells on medium composition is illustrated by the observation that the thickness of the inner layer decreases from 75 nm to about 22 nm when the cells are transferred from glucose to the much poorer carbon source lactate (17). Recently, atomic force microscopy (AFM) measurements of wall thickness of live, haploid cells of S. cerevisiae have been introduced (53). Dupres and coworkers used a functionalized AFM tip, allowing it to bind to a carboxy-terminal His tag of plasma membrane-bound, elongated forms of the sensor protein Wsc1. As the external protein layer seems to consist of a basal layer of short GPI proteins interspersed with long, fibrillar proteins radiating from the cell surface, it is conceivable that the AFM tip can penetrate the protein layer to a considerable extent and recognize the His tag as soon as it becomes part of or emerges above the basal layer of the mannoprotein coat. This suggests that this method results in values that are mainly determined by the thickness of the inner skeletal layer. AFM measurements of the wall thickness of haploid, late-exponential-phase cells of S. cerevisiae resulted in values of about 115 nm (53). This corresponds closely to the electron microscopy-based estimation of the inner skeletal layer of haploid exponential-phase cells equaling ∼120 nm (54); interestingly, the inner transparent layer of stationary-phase cells of S. cerevisiae was about twice as high (54).

Atomic force microscopy also allows measuring of an important mechanical property of the walls of live cells, i.e., the elastic modulus (E), or Young's modulus. Young's modulus is defined as “stress/strain” and reflects the stiffness of a material; it has units of pressure and is often expressed in MPa. With haploid, glucose-grown, exponential-phase yeast cells of S. cerevisiae (55) and diploid yeast cells of C. albicans (17), a value of 1.6 MPa was obtained. The Young's modulus of haploid, stationary-phase S. cerevisiae cells was also 1.6 MPa (55). Dague and coworkers also included haploid wall mutants of S. cerevisiae in their study. For example, the walls of gas1 cells, which lack a GPI-modified protein involved in the processing of β-1,3-glucan (56, 57), and of crh1 crh2 cells, which lack two related chitin transglycosylases (57), show a strongly decreased stiffness (E = 0.80 and 0.26 MPa, respectively). The carbon source can strongly affect cell wall stiffness as well; for example, the Young's modulus of walls of C. albicans yeast cells grown on lactate as the sole carbon source increased more than 3-fold to 5.3 MPa (17). These two examples of the use of AFM in analyzing cell surface properties illustrate that it is a powerful technique to measure physical properties of the wall; for a recent review of how AFM can be used to measure other cell surface properties on a nanoscale level, the reader is referred to reference 58.

Another physical property of cell walls is wall porosity. Early measurements of wall porosity of live cells of S. cerevisiae and C. albicans have been carried out with stationary-phase cells. In both organisms, only small polyethylene molecules (S. cerevisiae, M_r ≤ 760, hydrodynamic radius ≤ 0.89 nm; C. albicans, M_r ≤ 620, hydrodynamic radius ≤ 0.81 nm) could penetrate the cell wall (59, 60). However, exponential-phase S. cerevisiae cells can internalize 70-kDa dextrans (but not 250-kDa dextrans) (61), suggesting that walls of exponential-phase cells have much wider pores than stationary-phase cells. This was confirmed by De Nobel and coworkers, who introduced a rapid assay for relative measurements of wall porosity in live S. cerevisiae cells and other yeasts based on the cell lytic activity of the polycation DEAE-dextran (16). Not only is the porosity of the walls of live, exponential-phase cells grown in rich medium indeed much higher than in stationary-phase cells (11), but also it is much higher than the wall porosity of cells growing in minimal medium (16). Interestingly, cell wall porosity increased strongly after pretreatment of the cells with the reducing agent dithiothreitol, pointing to an important contribution by (presumably) intermolecular disulfide bridges between wall proteins in limiting wall porosity (15). Recently, Ene et al. (17), using the same assay, showed that lactate-grown C. albicans had a 2-fold higher relative cell wall porosity than glucose-grown cells. In summary, the earlier measurements of wall porosity expressed as exclusion limits in daltons or maximal hydrodynamic radius of the penetrating molecules are probably correct but are valid only for stationary-phase cells, whereas the walls of exponential-phase cells are much more porous depending on growth conditions, although exact values of the corresponding maximal hydrodynamic radii are currently unknown.

QUANTITATIVE ANALYSIS OF THE WALL PROTEOME

The polypeptide part of the walls of both S. cerevisiae and C. albicans accounts for ∼4% and 3.5% of the wall dry weight, respectively (35, 62) (Table 3). In view of the high mannose content of the walls, this indicates that their wall proteins are generally highly glycosylated. It seems likely that in particular very small GPI-modified, N-glycosylated wall proteins such as ScCcw12 (63) and CaPga59 and CaPga62 (35) contribute to this high mannose content. Their corresponding genes each have an exceptionally high codon adaptation index (0.87, 0.95, and 0.91, respectively; Saccharomyces Genome Database at http://www.yeastgenome.org/ and Candida Genome Database at http://www.candidagenome.org/) and thus probably encode highly abundant proteins (64).

The total number of wall proteins in a haploid parent cell of S. cerevisiae from an exponential-phase culture growing in rich medium is estimated to be ∼1.6 × 10⁶ (based on an average molecular mass of 52,728 Da) (2) (http://bionumbers.hms.harvard.edu/; BN 105224), which corresponds to a surface density of about 2.7 × 10⁴ μm⁻² (Table 4). For diploid C. albicans yeast cells the corresponding estimates are 2.9 × 10⁶ wall proteins per parent cell and a surface density of 3.0 × 10⁴ μm⁻² (Table 5). The minimal radius of a spherical 53-kDa protein equals 2.48 nm (65). Efficient (hexagonal) packing of such proteins on a surface would then result in a surface density of ∼4.7 ×⁴ μm⁻². This suggests that the outer protein layer of the wall can be approximately viewed as a monolayer, especially when taking into consideration that many wall proteins carry bulky N-linked carbohydrate side chains, resulting in increased protein dimensions and thus a lower surface density. The wall proteomes of both S. cerevisiae and C. albicans consist of >20 different wall proteins under any growth condition studied. These are not necessarily uniformly distributed over the cell wall, as several wall proteins are known to be cell cycle regulated (66). To obtain absolute copy numbers of individual wall proteins in S. cerevisiae, Yin and coworkers (67) used the iTRAQ (isobaric tags for relative and absolute quantitation) reagents (68) for isobaric tagging of tryptic peptides originating from wall proteins on the one hand and synthetic peptides that were added as internal standard on the other hand. Currently, the absolute copy numbers of five cell wall proteins in haploid, exponential-phase cells growing in rich medium have been determined (67): (i) the GPI-modified protein Cwp1, which localizes specifically to the birth scar (66), 6.7 × 10⁴ copies per parent cell; (ii) the GPI-modified protein Crh1, which is a chitin transglycosylase, 3.9 × 10⁴ copies per parent cell; (iii) the non-GPI-modified wall protein Scw4 and predicted β-glucanase, 3.4 × 10⁴ copies per parent cell; (iv) the GPI-modified protein Gas1, a β-glucan transglycosylase, a major plasma membrane protein but also present in the wall, 1.0 × 10⁴ copies per parent cell; and (v) the GPI-modified protein Ecm33, a plasma membrane protein involved in an unknown way in cell wall construction but also consistently identified in the wall, 5.8 × 10³ copies per parent cell (67); note that except for Cwp1, which is targeted to the birth scar, the original numbers were slightly corrected, using a correction factor of 0.89, to compensate for the fact that in the original paper the presence of buds was not taken into account. The correction factor was obtained as the ratio of the calculated dry weight of a haploid parent cell (16.5 pg [Table 4]) and the dry weight of a cellular unit in a haploid budding culture (18.6 pg) (67).

TABLE 4.

Cell wall-related bionumbers and bioestimates of live, haploid, exponential-phase parent cells and budding cells of S. cerevisiae grown in rich medium at 30°C

Parameter	Value(s)	Reference
Haploid parent yeast cells
Length and width	4.76 and 4.18 μm	3
Surface area	60.0 μm2
Vol	43.6 μm3
Density	1.11 g cm−3 (= pg μm−3)	88
Biomass (wet wt)	48.4 pg
Biomass (dry wt)	16.5 pg (34% of wet wt)	89
Cellular protein	8.7 pg (53% of dry wt)	74
Walls	3.5 pg (21% of dry wt)	67
Wall protein	0.14 pg (4% of walls)	62
β-Glucan	7.7 × 109 Glc	62
β-1,3-Glucan	6.1 × 109 Glc	62
	3.9 × 106 moleculesa	62
β-1,6-Glucan	1.6 × 109 Glc
	6.6 × 106 moleculesb
Cellular proteinsc	1.0 × 108
Wall proteins
No.c	1.6 × 106
Surface density	2.7 × 104 μm−2
Haploid budding cells
Biomass formation	0.18 pg min−1; dry wt
Surface expansion rated	0.67 μm2 min−1
Incorporation rate in bud wallsd
β-1,3-Glucan	6.8 × 107 Glc min−1
β-1,6-Glucan	1.8 × 107 Glc min−1
Wall proteins	1.8 × 104 min−1

^aBased on an estimated length of 1570 glucose residues (90).

^bBased on an estimated length of 247 glucose residues (91).

^cBased on an average protein mass of 52,728 (466 amino acids) (2; BN105224).

^dBased on a generation time of 90 min and linear growth during that period.

TABLE 5.

Cell wall-related bionumbers and bioestimates of live, diploid, exponential-phase parent yeast cells, budding cells, and hyphae of C. albicans grown in rich medium at 37°C

Parameter	Value(s)	Reference
Diploid parent yeast cells
Length and width	6.0 and 5.3 μm	27
Surface area	96.1 μm2
Vol	88.2 μm3
Densitya	1.11 g cm−3 (= pg μm−3)	88
Biomass (wet wt)	98.0 pg
Biomass (dry wt)a	33.3 pg (34% of wet wt)	89
Cellular proteina	17.7 pg (53% of dry wt)	74
Wallsa	7.0 pg (21% of dry wt)	67
Wall protein	0.24 pg (3.5% of walls)	62
β-Glucan	1.6 × 1010 Glc	62
β-1,3-Glucan	1.1 × 1010 Glc	62
β-1,6-Glucan	4.8 × 109 Glc	62
Cellular proteinsb	2.0 × 108
Wall proteins
No.b	2.9 × 106
Surface density	3.0 × 104 μm−2
Diploid budding cells
Biomass formationc	0.37 pg min−1; dry wt
Surface expansion ratec	1.1 μm2 min−1
Incorporation rate in bud wallsc
β-1,3-Glucan	8.5 × 107 Glc min−1
β-1,6-Glucand	4.3 × 107 Glc min−1
Wall proteins	3.2 × 104 min−1
Elongating hyphae
Diam
Germ tubes	2.6 μm	29
Mature hyphae	3.4 μm	29
Elongation rate
Germ tubes	0.32 μm min−1	29
Mature hyphae	0.76 μm min−1	29
Surface expansion ratee
Germ tubes	2.6 μm2 min−1	29
Mature hyphae	8.1 μm2 min−1	29
Wall proteins
Surface density	2.8 × 104 μm−2
Incorporation rate
Germ tubes	7.3 × 104 min−1
Mature hyphae	2.3 × 105 min−1

^aAssumed to be equal to the value obtained for S. cerevisiae (88, 89).

^bBased on an average protein mass of 52,728 (466 amino acids) as determined for S. cerevisiae (2) (BN105224).

^cBased on a generation time of 90 min and linear growth during that period.

^dBased on the ratios of β-1,3-glucan and β-1,3-glucan as determined in reference 62.

^eSurface expansion rate (μm² min⁻¹) = π × diameter × (hyphal elongation rate).

QUANTITATIVE ANALYSIS OF THE SECRETOME

The mannoproteins of the wall (and the β-glucans and chitin) of exponential-phase cells of S. cerevisiae are metabolically stable and show only limited turnover or replacement (69, 70). This is consistent with the observations obtained with fluorescein-conjugated concanavalin A to label existing mannoproteins. Incorporation of new and thus unlabeled wall proteins took mainly place in the distal tip of the growing bud (71). Although some release of mannoproteins into the medium was observed (69), this is probably largely attributable to local dissolution of the wall in the neck region to allow cell separation (note that before cell separation, the walls of the parent cell and the bud form a continuous structure). These earlier observations are supported by analysis of the secretome of C. albicans (72, 73). The absolute amount of proteins recovered by filtration over a 10-kDa filter from the culture medium of late-exponential-phase yeast cells was relatively small (72): ∼0.8 mg (g biomass; dry weight)⁻¹, or ∼1.5 mg/g cellular protein, assuming that cellular proteins account for about 53% of the total biomass (74) (Tables 4 and 5). Mass spectrometric analysis of the medium proteins demonstrated that the majority of them (33 proteins) had a predicted extracellular location and included most of the experimentally identified wall proteins, whereas 13 predicted, cytosolic proteins were identified (72). As wall proteins in C. albicans yeast cells are estimated to account for ∼14 mg/g cellular protein (Table 5), it is clear that release of wall proteins either by turnover and replacement or by local enzymatic dissolution of the wall during cell separation is limited (<1.5/14, or <11%). A convincing example of wall protein stability in C. albicans is the GPI-modified adhesion protein Als1, which after its early, but temporary, incorporation into yeast or hyphal walls persists long into stationary phase (75). Finally, it has been proposed that transwall transport of vesicles plays an important role in fungal physiology (76). However, the relatively small amount of cytosolic proteins found in the medium of C. albicans (<0.15% of the total amount of cellular protein) under the growth conditions used suggests that if such an active mechanism exists in C. albicans, its role is limited (49, 72). Conceivably, the membranous vesicles isolated from the medium of C. albicans (74) originated from apoptotic or dead cells. Nevertheless, from a diagnostic point of view it is a relevant observation.

GROWTH RATES AND SURFACE EXPANSION RATES OF S. CEREVISIAE AND C. ALBICANS

Population growth can be accurately approximated by an exponential function: B_t = B₀ × e^kt where B is biomass, t is time, and k is the specific (relative) growth rate, which is usually expressed in reciprocal hours (h⁻¹) or reciprocal min (min⁻¹); one also frequently encounters the doubling time, which is related to the specific growth rate by the following equation: (doubling time) × k = ln 2. However, the biomass (dry weight) of individual budding cells of S. cerevisiae and the volume of individual C. albicans yeast cells increase linearly over at least the main part of the cell cycle (77, 78). This is further supported by the observation that the surface area of growing buds of diploid S. cerevisiae cells increases linearly over a large period with a constant surface expansion rate of 0.75 μm² min⁻¹ at 25°C and 1.17 μm² min⁻¹ at 37°C (79). Hyphae of C. albicans also elongate with a constant rate, resulting in a linear increase of the surface area in time (28, 29) similar to mycelial fungi (80); note, however, that on a more detailed timescale the elongation rates of the latter oscillate (80). Exponential growth of the population is thus primarily obtained by regular bud formation and hyphal branching or a combination of both (29). Assuming linear growth of individual cells and a doubling time of 90 min, biomass formation (dry weight) by a budding haploid cell of S. cerevisiae with an average volume of ∼44 μm³ equals 0.18 pg min⁻¹ on a dry weight basis (Tables 1 and 4), whereas biomass formation (dry weight) of a diploid budding cell with a volume of 83 μm³ (3) is 0.34 pg min⁻¹. For budding diploid yeast cells of C. albicans with a volume of 88 μm³ (25) biomass formation (dry weight) is estimated to be ∼0.37 pg min⁻¹ (Table 5).

The hyphal elongation rate of C. albicans depends on growth conditions and can reach values up to ∼46 μm h⁻¹ (Table 5). Although hyphae of many filamentous fungi can elongate considerably faster (81), the elongation rate of hyphae of C. albicans is much higher than that of yeast cells. A single yeast cell has under favorable growth conditions an average length of ∼6 μm, which corresponds to a length increase of only ∼4 μm h⁻¹. This shows that hyphal growth permits C. albicans to cover much larger distances per unit of time than the yeast mode of growth, thus promoting tissue invasion and biofilm outgrowth, which both imply hyphal growth. Conceivably, the accelerated extension rate of hyphae versus yeast cells also facilitates the escape of C. albicans cells from macrophages upon their internalization and subsequent switch from yeast to hyphal growth (82). The accelerated extension rate associated with hyphal growth is not necessarily accompanied by a similar increase in the production of intracellular proteins because the subapical compartments of the hyphae of C. albicans tend to become strongly vacuolated (28). Surface expansion rates at 37°C of C. albicans yeast cells are ∼1.1 μm² min⁻¹ (Table 5). During hyphal growth of C. albicans the surface expansion rates increase considerably to ∼2.6 μm² min⁻¹ for germ tubes and ∼8.1 μm² min⁻¹ for mature hyphae (29) (Table 5), suggesting that wall protein synthesis in growing hyphae is much more prominent than in yeast cells. Table 5 shows that the estimated incorporation rate of wall proteins in C. albicans increases from 3.2 × 10⁴ min⁻¹ in yeast cells to 7.3 × 10⁴ min⁻¹ in germ tubes (a 2.3-fold increase) and to 2.3 × 10⁵ min⁻¹ in mature hyphae (a 7.2-fold increase). It seems likely that hyphal growth implies increased expression levels not only of wall protein-encoding genes but probably also of other genes involved in cell wall formation.

CELL WALL-ASSOCIATED BIONUMBERS AND BIOESTIMATES

Tables 4 and 5 present an overview of cell wall-associated measurements and related bioestimates of live, haploid, exponential-phase parent cells and budding cells of S. cerevisiae growing in rich medium at 30°C, and live, diploid, exponential-phase parent yeast cells (and budding cells) and exponential-phase hyphae of C. albicans growing in rich medium at either 30 or 37°C. Many questions with regard to cell wall synthesis remain unanswered or have only been addressed in a qualitative way. For example, how is cell wall synthesis adjusted to (i) changes in cell volume (and thus to changes in surface/volume ratios) that accompany changes in cell ploidy and growth rate; (ii) changes in morphotype, for example, upon the switch from yeast to hyphal growth in C. albicans; (iii) changes in medium composition and environmental conditions; or (iv) various stress conditions (12, 31, 41, 83)? We hope that the bionumbers and bioestimates discussed here will lead to a more precise understanding of these processes. The approach taken in this review will also be useful for the analysis of other cellular parts or compartments such as the plasma membrane or the cytoplasm. For example, assuming that the wall of haploid, exponential-phase yeast cells is 120 nm thick (53) and adjusting the cellular dimensions of live cells (Table 4) for this, the plasma membrane has a calculated surface area of 54 μm². In combination with an estimated 1.26 × 10⁶ Pma1 molecules per cell (64), this results in a surface density of ∼2 × 10⁴ μm⁻². Considering that Pma1 is an integral membrane protein with 10 predicted transmembrane domains, this indicates that the plasma membrane is packed with Pma1 (84). Our data can further be used as a starting point for a more quantitative analysis of other yeasts such as Candida glabrata (62) and other Candida spp., the industrial yeast Kluyveromyces lactis (85), and the fission yeast Schizosaccharomyces pombe (86, 87). Interestingly, the haploid yeast form of E. dermatitidis (exponential-phase cells growing in rich medium) has been studied in depth by serial sectioning followed by a quantitative three-dimensional structural analysis (36, 37). Although these two studies used freeze-substituted cells, presumably resulting in cell contraction, many of their measurements, such as the number of ribosomes/cell (∼2 × 10⁵) (36), probably predict the corresponding values in haploid, exponential-phase S. cerevisiae cells reasonably well.