Stage Specific Assessment of Candida albicans Phagocytosis by Macrophages Identifies Cell Wall Composition and Morphogenesis as Key Determinants

Candida albicans is a major life-threatening human fungal pathogen. Host defence against systemic Candida infection relies mainly on phagocytosis of fungal cells by cells of the innate immune system. In this study, we have employed video microscopy, coupled with sophisticated image analysis tools, to assess the contribution of distinct C. albicans cell wall components and yeast-hypha morphogenesis to specific stages of phagocytosis by macrophages. We show that macrophage migration towards C. albicans was dependent on the glycosylation status of the fungal cell wall, but not cell viability or morphogenic switching from yeast to hyphal forms. This was not a consequence of differences in maximal macrophage track velocity, but stems from a greater percentage of macrophages pursuing glycosylation deficient C. albicans during the first hour of the phagocytosis assay. The rate of engulfment of C. albicans attached to the macrophage surface was significantly delayed for glycosylation and yeast-locked morphogenetic mutant strains, but enhanced for non-viable cells. Hyphal cells were engulfed at a slower rate than yeast cells, especially those with hyphae in excess of 20 µm, but there was no correlation between hyphal length and the rate of engulfment below this threshold. We show that spatial orientation of the hypha and whether hyphal C. albicans attached to the macrophage via the yeast or hyphal end were also important determinants of the rate of engulfment. Breaking down the overall phagocytic process into its individual components revealed novel insights into what determines the speed and effectiveness of C. albicans phagocytosis by macrophages.


Introduction
Invasive C. albicans infection can present a serious clinical complication, especially in patients with an impaired immune system. Host defence against systemic candidiasis relies mainly on the ingestion and elimination of fungal cells by cells of the innate immune system, especially neutrophils and macrophages [1][2][3]. Despite the clinical importance of phagocytosis, this process remains poorly understood at a mechanistic level.
The fungal cell wall is the first point of contact with the innate immune system and plays an important role in recognition and phagocytosis by host immune cells [2]. It is a dynamic, highly organized organelle that determines both the shape of the fungus and its viability. The core structure of the C. albicans fungal cell wall is composed of a skeleton of polysaccharide fibrils composed of b-(1,3)-glucan that is covalently linked to b-(1,6)-glucan and chitin (a b-(1,4)-linked polymer of N-acetylglucosamine), and is designed to function as a robust exoskeleton and a scaffold for the external glycoprotein layer [4]. This outer layer consists of highly glycosylated mannoproteins that are modified by N-linked and Olinked mannosylation and phosphomannosylation [5,6].
Another important feature of C. albicans biology thought to play a major role in host recognition is the fungus' ability to undergo reversible morphological changes between yeast, pseudohyphal, and hyphal forms in response to environmental signals [7,8]. Its morphological plasticity is considered to be the most important virulence attribute of C. albicans [9] and plays a major role in the fungus' capacity to successfully infect many different anatomical sites of the human host. Hyphae have invasive properties that can promote tissue penetration and escape from immune cells [10], whereas yeasts are suited to dissemination in the bloodstream [9].
Phagocytic clearance of fungal pathogens may be considered to consist of four distinct stages;(i) accumulation of phagocytes at the site where fungal cells are located; (ii) recognition of fungal pathogen-associated molecular patterns (PAMPs) through pattern recognition receptors (PRRs) (reviewed in [11]; (iii) engulfment of fungal cells bound to the phagocyte cell membrane, and (iv) processing of engulfed cells within phagocytes by fusion with lysosomal vesicles to form the phagolysosome [12].
There is very limited information on how alterations in C. albicans morphogenesis or cell wall composition affect phagocyte migration towards the fungus. In contrast, a significant body of literature has identified an increasing number of PRRs and downstream signalling pathways that contribute to the recognition of fungal cells by macrophages [11,13]. These pathways have described recognition of N-linked mannans by the mannose receptor (MR), O-linked mannans by Toll-like receptor 4 (TLR4), b-glucans by dectin-1/TLR2, and a-mannosides by galectin-3/ TLR2 complexes [14]. More recently, additional PRRs have been shown to contribute to C. albicans recognition, including the scavenger receptors CD36 and SCARF1 [15], TLR9 recognition of nucleic acids [16], dectin-2 [17] and the C-type lectin mincle [18].
Comparatively little is known about the engulfment process once the fungus is tethered to the phagocyte cell membrane. However, a series of studies have shed some light on how the overall phagocytic uptake process is affected by alterations in C. albicans cell wall composition, morphogenesis and macrophage activation state [10,19]. For example, we have recently shown that the glycosylation status of the C. albicans cell wall profoundly affected the rate of macrophage phagocytosis. Distinct patterns emerged in that phosphomannan deficient strains (mnn4D, pmr1D, and mnt3D mnt5D) were taken up at a lower rate than the wildtype or reintegrant controls, and that O-linked and N-linked mannan deficient strains are taken up at higher rates (mns1D and mnt1Dmnt2D) [10]. A study by Keppler-Ross et al. conducting competitive phagocytosis experiments suggested that macrophages displayed strong preferences for phagocytosis based on genus, species and morphology. Candida glabrata and Saccharomyces cerevisiae were taken up by J774 macrophage cells more rapidly than C. albicans, and C. albicans yeast cells were favoured over hyphal cells [20].
These studies are informative but are limited in that they assess phagocytosis in its entirety and do not break down any observed differences into individual stages of the process, such as migration, recognition or engulfment, which may be affected differentially. Furthermore, such studies assess uptake at selected time points, rather than as a continuous dynamic process, with the inherent disadvantage of ignoring temporal differences in individual stages of the phagocytosis process, which are likely to play a major role in determining the overall outcome of pathogen-host interactions in vivo.
Here we have conducted a comprehensive analysis of C. albicans phagocytosis by primary macrophages and macrophage cell lines, employing video microscopy, coupled with sophisticated image analysis tools. To assess the contribution of C. albicans cell wall glycosylation and the ability to switch from yeast to hyphal forms, we have taken advantage of a large collection of genetically and phenotypically characterized isogenic mutants of C. albicans, depleted in specific cell wall components or impaired in morphogenic switching. We show here for the first time a detailed minute by minute account of the specific effects of C. albicans viability, cell wall composition, morphogenesis and spatial orientation on two distinct stages (macrophage migration and engulfment of bound C. albicans) of the phagocytosis process. These analyses revealed that macrophage migration towards C. albicans was dependent on the glycosylation status of the fungal cell wall, but not cell viability or morphogenic switching from yeast to hyphal forms. Macrophages rapidly engulfed viable and UV-killed C. albicans, but engulfment was protracted for all glycosylation and morphogenetic mutants examined. Engulfment of hyphal C. albicans was determined by multiple components including hyphal length and spatial orientation.

Results
Macrophage migration towards C. albicans is affected by fungal cell wall glycosylation but not morphogenesis C. albicans phagocytosis by macrophages is dependent on the C. albicans cell wall glycosylation status [10], but the question remains whether differences observed in overall uptake are a consequence of changes in migration of macrophages towards C. albicans or alterations in the engulfment process itself. Live cell video microscopy enabled examination of the individual stages of the uptake process. Representative videos are available to view in Supporting Videos S1 and S2. First we addressed the question of whether alterations in C. albicans cell wall glycosylation and morphogenesis affect migration of macrophages towards C. albicans. Primary macrophages and macrophage cell lines were challenged with glycosylation and morphogenesis defective strains of C. albicans. The strains used in this study are shown in Table 1. Briefly, the mnt1Dmnt2Dstrain is deficient in O-glycosylation [21] and has only a single O-linked mannose sugar. The mns1D strain has an N-glycosylation defect due to curtailed a1, 2-mannosidase activity in the endoplasmic reticulum [22] and the mnn4D strain has a complete loss of phosphomannan [23]. Morphogenesis defective strains included the hgc1D strain, a G1 cyclin mutant that is unable to form true hyphae, and efg1D that lacks a specific transcription factor that regulates yeast-hypha morphogenesis pathways [24,25].
Migration of macrophages was assessed by live cell video microscopy using our standard phagocytosis assay [10,26], with track measurements taken at 1 min intervals over a 6 h period. Figures 1A, B and C show images derived from video microscopy depicting the track of a single macrophage migrating towards and engulfing live C. albicans (wildtype strain). Initially, the macrophage's movement appeared to be random ( Figure 1A). However, dynamic analysis suggested that macrophages sensed C. albicans, accelerated and homed in on their target ( Figure 1B), leading to cell-cell contact and engulfment ( Figure 1C). The corresponding video is available to view in Supporting Video S3. Visual inspection of the videos suggested enhanced macrophage migration towards C. albicans glycosylation mutants, in particular the mnt1Dmnt2Dstrain, compared to wildtype control.
The suggestion that migration was enhanced in macrophages exposed to the mnt1Dmnt2D mutant strain was further supported by macrophage tracking diagrams ( Figures 1D and 1E). Tracking diagrams ( Figures 1D and E) illustrate the distances travelled,

Author Summary
Host defence against systemic candidiasis relies mainly on the ingestion and elimination of fungal cells by cells of the innate immune system, especially neutrophils and macrophages. Here we have used live cell video microscopy coupled with sophisticated image analysis to generate a temporal and spatial analysis in unprecedented detail of the specific effects of C. albicans viability, cell wall composition, morphogenesis and spatial orientation on two distinct stages (macrophage migration and engulfment of bound C. albicans) of the phagocytosis process. The novel methods employed here to study phagocytosis of C. albicans could be applied to study other pathogens and uptake of dying host cells. Thus, our studies have direct implications for a much broader community and provide a blueprint for future studies with other phagocytes/microorganisms that would significantly enhance our understanding of the mechanisms that govern effective phagocytosis and ultimately the innate immune response to infection. directionality and velocity of macrophages cultured with live wildtype and mnt1Dmnt2D, respectively. Due to the large number of macrophages tracked per video, the data were filtered to show only macrophages with a mean track velocity greater than that of inactive macrophages not pursuing fungal cells (1.80 mm/min). Tracks represent the movement of individual macrophages relative to their starting position, symbols indicate the location of macrophages at 1 min intervals and arrows represent directionality. These diagrams illustrate that although macrophages can migrate rapidly and for long distances when cultured with both live wildtype and the mnt1Dmnt2D mutant, when incubated with live mnt1Dmnt2D a higher number of macrophages have a mean track velocity of greater than 1.80 mm/min ( Figure 1E).
Quantitative analysis of average macrophage track velocity for the entire length of the observation period (6 h) showed no significant differences between wildtype (1.8 mm/min 6 0.02 SE) and yeast-locked morphogenetic mutants, but confirmed enhanced migration with UV-killed wildtype C. albicans (1.9460.02 SE, p,0.05) and the glycosylation mutants mnt1Dmnt2D (2.160.02 SE, p,0.001), mns1D (2.0960.03 SE, p,0.001) and mnn4D (1.9660.03 SE, p,0.01) ( Figure 2B). The macrophage average track velocity was highest for the first 30 min of the phagocytosis assay ( Figure 2B). The data for this period again showed increased average track velocity for the glycosylation mutant strains mnt1Dmnt2D (2.6860.04, p,0.001), mns1D (2.4760.05, p,0.05) and mnn4D (2.5260.07, p,0.01) when compared with wildtype (2.1960.07) (Figure 2A). In contrast, there was no significant difference in the mean track velocity of macrophages when incubated with morphogenesis defective mutants and UV-killed wildtype. Overall track lengths were measured for the first 30 min and 6 h, and not surprisingly the data reflected the average track velocity for the wildtype and mutant strains tested (data not shown). Enhanced macrophage migration in phagocytosis assays with C. albicans glycosylation mutants was not a consequence of alterations in maximal macrophage velocity (average max velocity in mm/min: wildtype, 3.760.2; mnt1Dmnt2D, 3.660.1; mnn4D, 3.660.2; mns1D (3.160.2) but rather a reflection of increased macrophage activity, particularly during the first hour of the interaction assay.
Experiments with primary murine peritoneal macrophages showed a similar pattern. We observed no differences between the yeast-locked mutant strain hgc1D and wildtype but signifi-cantly increased average track velocity for the glycosylation mutant strain mnt1Dmnt2D (p,0.001) for the first 30 min and 6 h of the phagocytosis assay (Table 2). Overall the mean track velocity of peritoneal macrophages was found to be significantly lower than for J774.1 macrophages (p,0.001). However, the maximum velocity of peritoneal macrophages (3.960.2 mm/min) was comparable to J774.1 macrophages (3.760.2 mm/min). We hypothesised that the difference in mean track velocity between peritoneal and J774.1 macrophages is due in part to the peritoneal macrophages being more spread out and covering a larger surface area, therefore, reducing the need to migrate to achieve close proximity with fungal cells in the phagocytosis assay. However, experiments using the same macrophage:C. albicans ratios but lower macrophage densities confirmed that this was not the case, as mean track velocities were unchanged (Table 2). Thus, changes in C. albicans cell wall composition but not hyphal morphogenesis markedly influenced macrophage migration in vitro.
Macrophages rapidly engulfed viable and UV-killed C. albicans but engulfment was delayed for glycosylation and yeast-locked mutants Effective migration of macrophages towards C. albicans is necessary to establish cell-cell contact, which is a prerequisite for initiation of the engulfment process. Next we addressed the question of whether alterations in C. albicans cell wall glycosylation and morphogenesis affected the ability and speed by which macrophages engulfed C. albicans after cell-cell contact was established. Live cell video microscopy coupled with image analysis generated a detailed minute by minute account of the engulfment process ( Figures 3A, B and C). Wildtype C. albicans was shown to be rapidly engulfed by macrophages once cell-cell contact was established ( Figure 3D). A three dimensional projection image confirming C. albicans phagocytosis is available in Supporting VideoS4.The average time taken for engulfment of wildtype C. albicans is 6.760.3 min, and the vast majority (95%) of fungal cells were engulfed within 15 min. UV-killed C. albicans yeast cells were engulfed even more swiftly, with all cells taken up within 15 min and engulfment taking an average of 4.260.1 min ( Figure 3E).
Interestingly, the rate of engulfment of all glycosylation mutant strains ( Figures 4B-D) was significantly slower than that of wildtype C. albicans (mnt1Dmnt2D) (p,0.001); mns1D (p,0.01; mnn4D (p,0.001) ( Figure 4A). The delayed engulfment was most marked for the mnt1Dmnt2D (13.560.7 min) and mnn4D (14.460.9 min) mutant strains ( Figures 4B and D), as macrophages on average took twice as long to engulf these mutants than wildtype C. albicans ( Figure 4A). Control strains mnt1Dmnt2D::MNT1 and mnn4D::MNN4, containing a single reintegrated copy of the corresponding deleted genes, partially restored the ability of macrophages to swiftly engulf C. albicans (data not shown). Experiments using primary thioglycollate elicited murine peritoneal macrophages and human monocyte derived macrophages also showed a significant delay for the engulfment of the glycosylation deficient mutant mnt1Dmnt2D and this was partially restored in the corresponding reintegrant control mnt1Dmnt2D::MNT1 (Table 3). Engulfment of yeast-locked morphogenetic mutant strains was delayed and ultimately impaired, relative to wildtype controls in J774 macrophages (p,0.001). Firstly, the average time taken for engulfment of the hgc1D ( Figure 5A and B) and efg1D ( Figure 5A and C) mutant strains was significantly greater than for the wildtype control (16.261.4 min, 21.962.4 min and 6.760.3 min, respectively). Engulfment of approximately 1.5% of wildtype C. albicans took longer than 30 min, compared with approximately 11% and 17% for hgc1D and efg1D, respectively. Secondly, we observed that approximately 2% of hgc1D and 41% of efg1D mutants that established contact with macrophages  were not internalised, even after prolonged cell-cell contact. In contrast, all wildtype C. albicans yeasts were successfully engulfed following recognition. However, 66% of the efg1D mutant cells were eventually engulfed by neighbouring phagocytes after detachment from the macrophage they were originally in contact with. Experiments using peritoneal macrophages showed a non significant delay in the engulfment of the yeast locked mutant strain hgc1D (Table 3) and little evidence of detachment once cellcell contact was established. Thus, macrophages rapidly engulfed viable and UV-killed C. albicans after cell-cell contact was established, but engulfment was markedly slower for all glycosylation (in all macrophage subsets  Figures 3A, B, and C are snapshots taken from live cell video microscopy capturing the engulfment process. Figure 3A shows a macrophage (*) and C. albicans prior to cell-cell contact, Figure 3B shows the same cells during contact and Figure 3C shows C. albicans within the macrophage post engulfment. Scale bar, 10 mm. Figures 3D and 3E show the time taken for J774.1 macrophages to ingest live and UV-killed wildtype C. albicans following cell-cell contact verses the percentage of uptake events, respectively (n = 6). The majority of live C. albicans were engulfed rapidly by macrophages once cell-cell contact was established ( Figure 3D). However, UV-killed C. albicans were engulfed more swiftly, with all cells internalised within 15 min ( Figure 3E). doi:10.1371/journal.ppat.1002578.g003 studied) and yeast-locked (solely in the J774 macrophage cell line) morphogenetic mutants examined.
Macrophages are more effective at engulfing yeast rather than hyphal C. albicans and engulfment is influenced by hyphal length The data above showed that UV-killed yeast cells were engulfed more rapidly than live wildtype cells that were able to form hyphae. The accelerated engulfment of UV-killed cells raised questions about how cell morphology affects engulfment of C. albicans by macrophages. We examined the engulfment of wildtype C. albicans cells that had established cell-cell contact with macrophages in either yeast or hyphal morphology. Hyphal C. albicans cells were engulfed at a slower rate than C. albicans yeast cells (10.860.9 min and 5.660.3 min, respectively). Furthermore, the vast majority (98%) of yeast cells of C. albicans were taken up within 15 min ( Figure 6A), whereas there was greater variability for hyphal cells of C. albicans, with 21% taking longer than 15 min to become engulfed ( Figure 6B).
Next we examined whether hyphal length influenced the speed of engulfment, perhaps explaining the variations observed for hyphal C. albicans engulfment. Macrophages were capable of ingesting C. albicans with hyphae of more than twice the average diameter of macrophages (the maximum observed length of ingested hyphae was 42.9 mm), but the mean hyphal length at time of engulfment was 8.760.7 mm. Intriguingly, and contrary to expectations, we found no correlation between hyphal length and speed of engulfment for hyphal cells of C. albicans of less than 20 mm length ( Figure 6C). However, when hyphal length exceeded 20 mm there was as significant impact on the macrophage's ability to engulf C. albicans ( Figure 6D). Although macrophages engulfed C. albicans with hyphae larger than 20 mm, uptake was markedly slower with 64% of uptake events requiring more than 15 min. It is worth noting that despite having difficulty engulfing large hyphae macrophages were nonetheless persistent in their attempt to do so. Thus, macrophages were more effective at engulfing yeast cells rather than hyphal cells of C. albicans and engulfment of hyphal cells was influenced in part by hyphal length, with a cut off of 20 mm, above which macrophage engulfment was markedly impaired.

The rate of engulfment of hyphal cells of C. albicans was influenced by spatial orientation
Finally, we took advantage of the large quantity of data amassed from live cell video microscopy phagocytosis assays to address previously unanswered questions relating to how spatial orientation of C. albicans may affect the efficiency of engulfment by macrophages. First, we established that hyphal cells of C. albicans could be taken up by macrophages independent of their spatial orientation ( Figure 7A). C. albicans germ tubes could be engulfed yeast-end on and germ tube apex-end on ( Figures 7B and C), sideon ( Figure 7D) and at an angle ( Figure 7E). However, it is noteworthy that although cell-cell contact could be initiated in any orientation, the rate of engulfment was affected; C. albicans that made contact in an end-on orientation were taken up more rapidly than those engulfed at an angle or where cell-cell contact was initiated side-on (5.560.7 min, 8.860.5 min and 9.562.3 min, respectively). The large SE observed when C. albicans makes contact side-on can be explained by the fact that macrophages had particular difficulty ingesting large hyphae (.20 mm) in the sideon orientation. The end initially encountered by the macrophage appeared to be random. Approximately equal numbers of encounters occurred that were yeast end-on or hyphal end-on, but there was a propensity for C. albicans to be taken up more rapidly yeast-end on (8.060.6 min) than hyphal end-on (9.861.0 min).
Initial C. albicans orientation when establishing cell-cell contact with macrophages influenced the rate of engulfment with end-on contact of the hyphal end resulting in the most rapid engulfment. Thus, engulfment of hyphal cells of C. albicans was influenced by multiple factors including hyphal length and spatial orientation, and whether the initial encounter was by the yeast or hyphal end.

Discussion
C. albicans is a major life-threatening human fungal pathogen. Host defence against systemic Candida infection relies mainly on phagocytosis of fungal cells by cells of the innate immune system. In this study, we analysed the contribution of distinct C. albicans cell wall components and yeast-hypha morphogenesis to specific stages of phagocytosis by macrophages.
We show that macrophage migration towards C. albicans was dependent on the glycosylation status of the fungal cell wall, but not cell viability or morphogenic switching from yeast to hyphal forms. This finding was not a consequence of differences in maximal macrophage track velocity, but stems from a greater percentage of macrophages pursuing glycosylation deficient C. albicans cells during the first hour of the phagocytosis assay. The rate of engulfment of C. albicans by macrophages was significantly slower for glycosylation and morphogenesis deficient mutant strains, but enhanced for non-viable cells. Hyphal cells were engulfed at a slower rate than yeast cells, especially those with hyphae in excess of 20 mm, but there was no correlation between hyphal length and the rate of engulfment below this threshold. We show that spatial orientation of the hypha and whether hyphal C. albicans attached to the macrophage via the yeast or hyphal end were also important determinants of the rate of engulfment. This is the first study, to our knowledge, to show that individual stages of C. albicans phagocytosis by macrophages are differentially affected by changes in C. albicans cell wall composition. Our previous work, using assays that globally assess phagocytosis, have shown increased phagocytosis of O-linked and N-linked mannan deficient strains (mns1D and mnt1D mnt2D) [10]. Intriguingly, we show here that changes in cell wall glycosylation enhance macrophage migration towards C. albicans, but delay engulfment once cell-cell contact is established. This illustrates that standard assays do not differentiate between the individual stages of the phagocytosis process and are unable to detect significant temporal differences in migration or engulfment. For example, we show here that phosphomannan deficient cells of C. albicans were engulfed less efficiently. This effect was much less obvious in previous studies that simply evaluated phagocytosis efficiency by single time point measurements [27]. Macrophage migration was enhanced in all glycosylation mutants but most markedly in the mnt1Dmnt2D O-glycosylation mutant. This translated into much higher overall uptake compared to the phosphomannan deficient  Figures 4A, B, C and D show the time taken for J774.1 macrophages to ingest live wildtype (n = 6), mnt1Dmnt2D (n = 6), mns1D (n = 3) and mnn4D (n = 3) strains following initial cell-cell contact verses the percentage of uptake events. The rate of engulfment of all glycosylation mutant strains ( Figures 4B-D) was significantly slower than that of wildtype C. albicans ( Figure 4A). doi:10.1371/journal.ppat.1002578.g004 mmn4D mutant and is in keeping with our previous published results [10]. It is conceivable that enhanced macrophage migration in response to the absence of O-linked (mnt1Dmnt2D) or N-linked mannans (mns1D) is a consequence of unmasking underlying bglucans [28,29] or electrostatic signals as a consequence of alterations in surface charge following loss of phosphomannan (mnn4D) [30,31]. Observation of individual macrophage migration patterns indicated that macrophage movement was slow and random initially, but became directional towards a specific C. albicans cell, associated with a marked increase in macrophage velocity. Macrophage acceleration towards C. albicans occurred at distances in excess of 15 mm and, therefore, suggests the presence of a chemotactic signal. Key candidates are a number of glycolipids that are known to be shed by C. albicans and are  Figures 5A, B and C show the time taken for J774.1 macrophages to ingest live wildtype (n = 6), hgc1D(n = 6) and efg1D (n = 3) following initial cell-cell contact plotted versus the percentage of uptake events. The average time taken for engulfment of the hgc1D ( Figure 5B) and efg1D ( Figure 5C) mutant strains was significantly greater than for the wildtype control ( Figure 5A). doi:10.1371/journal.ppat.1002578.g005 potent inducers of macrophage cytokine synthesis in vitro and in vivo [32]. We are currently conducting detailed mathematical modelling of the macrophage tracking patterns to further elucidate the hypothesis that macrophage migration towards C. albicans is affected by differences in shedding of glycolipids between wildtype and glycosylation deficient strains.
We have shown that macrophage uptake of C. albicans is a multistep process, involving recognition and subsequent engulfment of C. albicans. C. albicans cell wall mannosylation is a key determinant in the rate of engulfment; the absence of specific PAMPs in the glycosylation mutant strains delays engulfment once cell-cell contact has been established, and this may be a consequence of differential activation of macrophage PRRs. This is in keeping with experiments in which C. albicans cell wall mutants were combined with specific macrophage receptor blocking methods that have been used to define the PAMP-PRR interactions required for cytokine induction [14,33]. These in vitro findings are relevant to C. albicans infections in vivo, since C. albicans mutants with defects in cell wall mannosyl residues are also less virulent in experimental models of disseminated candidiasis [21,22,[34][35][36].
Morphological plasticity is one of the hallmarks of the human fungal pathogen C. albicans [37], and its ability to switch between yeast and hyphal forms is thought to contribute to pathogenesis [24,25,38,39]. C. albicans mutants that are unable to form filaments are less virulent [40], although conversely, mutants that are unable to grow as yeast are also less virulent [41]. There are conflicting reports in the literature regarding the efficiency of macrophage phagocytosis for C. albicans yeast and hyphal forms [20,42,43].
Here we show definitively data supporting the notion that macrophages are more effective at engulfing C. albicans yeasts. Furthermore, the use of video microscopy coupled with thorough analysis of large numbers of individual macrophage-C. albicans interactions provides a minute-by-minute account of the engulfment process, which offers detail that has not been previously available. A prime example is our observation that yeast-locked C. albicans cells were engulfed less efficiently than wildtype C. albicans. Not only was this not obvious in previous studies that simply evaluated phagocytosis efficiency by single time point measurements [10], but in addition, we observed here that delayed engulfment can result in detachment of the fungal cell and engulfment by a neighbouring macrophage. One may speculate that in vivo where phagocyte numbers are limited this may have a significant impact on pathogen clearance and that yeast locked mutant C. albicans cells have properties other than the induced phenotype that differ from wildtype yeast C. albicans cells. However, the observed delay in engulfment for yeast-locked mutant C. albicans in the macrophage cell line was almost completely abrogated in experiments using primary macrophages, underlining the importance of studying host-pathogen interactions in multiple phagocyte subsets.
The approach taken here further enabled us to dissect the complexity of engulfment of hyphal C. albicans by macrophages. We show that macrophages are capable of engulfing hyphal C. albicans in excess of 40 mm (approximately twice the diameter of macrophages) -in keeping with reports that macrophages are capable of ingesting apoptotic epithelial cells in the involuting mammary gland of similar or even larger size [44]. Hyphal length, however, does play a major role in the engulfment process of C. albicans, in that engulfment of C. albicans with hyphae in excess of 20 mm took significantly more time and phagocytosis was frequently frustrated. Interestingly, we show that below a 20 mm hyphal length threshold there was no correlation between hyphal length and the rate of engulfment. These observations are most likely related to difficulties associated with macrophages attempting to engulf very large particles. In addition to hyphal length, we have identified two other factors that influenced engulfment of hyphal C. albicans. We showed that the rate of engulfment was determined by the orientation in which C. albicans was encountered, with end-on being favourable to side-on orientation, suggesting that steric hindrance affects engulfment. We also showed that yeast end-on engulfment was more efficient than hyphal end-on encounters. This in turn may reflect differences in the wall chemistry of the hyphal tip compared to the mother cell, or be due to the efficiency of the assembly of proteins of the phagocytic cup for objects of different sizes and shapes [45].
Here we have conducted the most detailed analysis of the contribution of C. albicans viability, cell wall glycosylation and morphogenesis to phagocytosis by macrophages to date, to our knowledge. Our approach of combining live cell video microscopy with image analysis tools for the migration analysis, and minuteby-minute analysis of thousands of individual macrophage-C. albicans interactions, provides unique insight into the complexity of C. albicans phagocytosis by macrophages. The novel methods employed here to study phagocytosis of C. albicans could be applied to study other pathogens and uptake of dying host cells. Such studies would significantly enhance our understanding of the mechanisms that govern effective phagocytosis and ultimately the innate immune response to infection.

Ethics statement
All animal experiments have been conducted in strict accordance with UK Home Office guidelines. The appropriate project and personal licenses are in place PIL 60/6194 and approved by the UK Home office.

C. albicans preparation and staining with fluorescein isothiocyanate (FITC)
Single C. albicans colonies from plates stored at 5uC were cultured in 5 ml SC-Ura/YPD medium (recipes as above, excluding technical agar) and incubated overnight at 30uC, 200 rpm. In order to determine the impact of C. albicans viability on macrophage migration and engulfment, wildtype C. albicans were killed by UV-irradiation; 100610 6 fungal cells in 1 ml 16 PBS were exposed to 20 doses of UV irradiation at 100 mJ/cm 2 in 6 well plates. To aid visualisation of C. albicans during phagocytosis assays, 100610 6 live or UV-killed C. albicans were stained using 1 mg/ml FITC (Sigma, Dorset, UK) in 0.05 M carbonatebicarbonate buffer (pH 9.6) (BDH Chemicals, VWR International, Leicestershire, UK) for 10 min at room temperature in the dark. Fungal cells were washed 3 times in PBS to remove unbound FITC and resuspended in 16 PBS.

Live cell video microscopy phagocytosis assays
Our standard phagocytosis assays were performed as previously described [10]. In brief, 3610 6 FITC-stained C. albicans were added to 1610 6 macrophages in glass based Iwaki dishes (VWR, Leistershire, UK) immediately prior to imaging. Video microscopy experiments were performed using a DeltaVision Core microscope (Applied Precision, Washington, USA) with an environmental control chamber set at 37uC. Images were captured at 1 min intervals for 6 h using an EMCCD camera. At least two independent experiments were conducted for each C. albicans strain, and at least 3 movies were analysed from each experiment. One hundred macrophages were randomly selected from each movie and their phagocytic activity determined (as below).

Analysis of live cell video microscopy movies
Volocity 5.0 imaging analysis software was used to track macrophage migration at 1 min intervals throughout the 6 h phagocytosis assay. The software enabled high throughput analysis of macrophage migration, providing detailed information on the distances travelled, directionality and velocity of thousands of individual macrophages. Data were subsequently displayed in tracking diagrams and used to calculate the mean track velocity and track length of macrophages cultured with C. albicans. These analyses enabled assessment of the affects of C. albicans viability, glycosylation status and morphology on migration. Figure 6. Role of hyphal morphogenesis in the engulfment of C. albicans by macrophages. Figures 6A and B show the time taken for J774.1 macrophages to ingest live C. albicans in yeast and hyphal morphology plotted against the percentage of uptake events (n = 6). Hyphal C. albicans was engulfed at a significantly slower rate than the yeast form of C. albicans. Figures 6C and D show the time taken for macrophages to engulf C. albicans with hyphae less than and greater than 20 mm, respectively (n = 6). There was no correlation between hyphal length and speed of engulfment for hyphal C. albicans of less than 20 mm length ( Figure 6C). However, when hyphal length exceeded 20 mm there was significant impact on the macrophage's ability to engulf C. albicans ( Figure 6D). doi:10.1371/journal.ppat.1002578.g006 Figure 7. The role of C. albicans spatial orientation in engulfment by J774.1 macrophages. Figure 7A plots the time taken for engulfment of individual hyphal wildtype C. albicans in relation to the exact angles at which cell-cell contact was established (0u, side-on; 90u, end-on). C. albicans can be engulfed in any spatial orientation, including end-on, side-on and at an angle ( Figure 7A One hundred macrophages from each movie were analysed individually at 1 min intervals throughout the 6 h phagocytosis assay. Measurements taken include the time points at which initial cell-cell contact occurred and at which C. albicans was fully enclosed, the number of C. albicans taken up and their morphology, the orientation of hyphal C. albicans relative to the macrophage and hyphal length. The rate of engulfment of live and UV-killed wildtype C. albicans, and glycosylation and yeast-locked morphogenetic mutant C. albicans was calculated by subtracting the time point at which initial cell-cell contact occurred from the time point at which the fungus was fully phagocytosed. This enabled accurate assessment of the affects of C. albicans viability, glycosylation status and morphology on the speed of engulfment. C. albicans spatial orientation, morphology, hyphal length and the end of hyphal C. albicans recognised were determined to assess whether these factors impact on the rate of engulfment. This strategy enabled in depth analysis of individual C. albicans-macrophage interactions in real time.

Statistical analysis
Mean values and standard errors were calculated. One-way analysis of variance (ANOVA) and Tukey-Kramer Multiple Analysis Comparison Tests were used to determine statistical significance.

Supporting Information
Video S1 Phagocytosis of live C. albicans by J774.1 macrophages. Shows a representative 6 hour live video microscopy of live C. albicans being ingested by macrophages. It further illustrates hyphal growth within macrophages and macrophage killing by hyphal C. albicans.

(MP4)
Video S2 Phagocytosis of live C. albicans by J774.1 macrophages. Shows a high magnification (680) representative 6 hour live video microscopy of live C. albicans being ingested by macrophages. It further illustrates hyphal growth within macrophages and macrophage killing by hyphal C. albicans.

(MP4)
Video S3 Macrophage tracking of C. albicans. Shows an example of a J774.1 macrophage tracking a UV-killed C.albicans yeast cell. It illustrates random macrophage movement followed by directional tracking of the fungal cell and ultimately uptake of the pathogen. (MP4) Video S4 Projection movie of C. albicans phagocytosis by macrophages. Shows a 3D projection of C. albicans ingested by macrophages. It also confirms co-localisation of green FITC stained fungal cells with lysotracker red in macrophage phagosomes. (MOV)