Figure 1.
Caspofungin-mediated unmasking in vitro is filament-biased.
(A,B) Wildtype strain CAF2-1 was grown overnight in RPMI-PS with vehicle (A) or sub-inhibitory doses of CF (B), then stained with anti-β-glucan antibody. Overlays between DIC and anti-β-glucan staining (in red) show that there is no β-glucan exposure on yeast-form cells, but there are low to high levels of exposure on filaments. (C) Covalent labeling of live yeast-form cells with Alexafluor 647-succinimidyl ester is homogeneous and bright. (D) Extensive growth in vitro of Alexafluor 647-labeled yeast-form cells in the presence of serum demonstrates that the label is long-lasting and remains confined to initial yeast-form cell wall. (E) Wildtype strain SC5314-GFP was pre-labeled with Alexa647, then grown overnight in RPMI-PS with subinhibitory doses of CF (½ MIC50). (F–H) Wildtype (F) or nrg1Δ/Δ (G) or edt1Δ/Δ (H) mutant cells pre-grown in YPD were washed and diluted into RPMI-PS plus different concentrations of CF; cells grown in ½ MIC50 of CF were stained with anti-β-glucan antibody and Cy3-labeled secondary antibody. Images are representative of three independent experiments. Scale bar shown in panel (A) is 20 microns long for panels (A) and (B) and 10 microns long for (C–H).
Figure 2.
Caspofungin is more effective against filament-locked mutants than against yeast-locked mutants.
Strains were pre-grown in YPD, washed, sonicated, diluted into RPMI-PS containing two-fold dilutions of CF, and grown overnight at 30°C to promote both yeast- and filamentous-form growth. RPMI has phenol red, so yellow color in the media indicates metabolic activity and acidification of the media. Changes in color and visual scoring of growth were used to identify MICs for each strain. Results are representative of four independent experiments run in duplicate.
Figure 3.
Candida β-glucan is masked early but unmasked late during disseminated infection.
BALB/c mice were infected i.v. tail vein with 1×106 (Day 1), 2.5×105 (Day 3), 1×105 (Day 5 or 7) organisms of strain SC5314-GFP. Animals were euthanized 1, 3, 5, or 7 days post-infection, kidneys were isolated, and organisms were stained by EVF for anti-β-glucan with or without primary antibody. (A) Anti-β-glucan staining of hyphae and yeast-form cells at Day 1 post-infection. Two representative examples are shown. (B) Anti-β-glucan staining of hyphae and yeast-form cells at Day 5 post-infection. Two representative examples are shown. (C) Representative filament from Day 5 post-infection, that the primary antibody was omitted. (D) Example of cell segmentation and quantification, with arrowhead pointing to yeast-form cell: Left panel is GFP image; center panel shows different cell segments in different colors; right panel shows the unmodified anti-β-glucan image used for quantification. (E) Categorization of exposure at different time points post-infection based on quantitative staining measure. No Staining is defined as within 2 standard deviations of the average fluorescence/cell of cells stained with no primary antibody; Low Staining is defined as cells with average fluorescence/cell greater than 2 standard deviations above, but less than 2-fold above, the average fluorescence of the no primary control; High Staining are cells with fluorescence/cell greater than two-fold higher than the average of the no primary control. These preparations have a significant amount of highly fluorescent debris from kidney tissue, which is the likely source of many of the 20% of cells that have “greater than background” levels of staining at the early time points. (F) Level of β-glucan exposure in hyphal cell wall vs. yeast-form cell wall at 5 to 7 days post-infection. Histogram shows the average fluorescence/cell for all cells quantified at 5 and 7 days post infection, in comparison to cells stained with no primary anti-β-glucan antibody, as measured in arbitrary units (a.u.). Error bars show standard error of average fluorescence/cell. All pair-wise comparisons were analyzed by two-tailed Student's T-test in Excel (Microsoft Corp.); asterisks indicate statistically significant differences (p<0.01). (G) Overall quantification of β-glucan exposure in all cells at Days 1, 3, 5, and 7 post-infection, as compared to background staining without primary antibody. All cells from a given day were pooled, averages and standard error for each group was calculated, and pairwise comparisons were made by two-tailed Student's T-test in Excel (Microsoft). Day 1 and Day 3 groups show exposure indistinguishable from the No Primary group (p>0.35 for all pairwise comparisons), and Day 5 and Day 7 groups are indistinguishable from each other (p>0.35), but all comparisons between the two groups show significant differences (p<0.000001). Comparisons where differences are not significant (p>0.35) are indicated with n.s. and significant comparisons (p<0.000001) are indicated with three asterisks. Data shown in panels A–G are derived from 12 experiments (Day 1) or 2 experiments each (Days 3, 5, and 7) which were internally consistent. Scale bar in (A) is 10 microns long and applies to all images.
Figure 4.
Treatment of C. albicans with caspofungin unmasks β-glucan.
BALB/c mice were injected i.p. with caspofungin (CF) or vehicle and then immediately infected i.v. tail vein with strain SC5314-GFP. 16 hours post-infection, mice were euthanized and kidneys were dissected then homogenized. (A) Kidney homogenates were assayed for viable colony forming units by broth dilution and plating to YPD agar plates. Bars represent the average of triplicate plates, and error bars represent standard deviations. (B–E) Kidney homogenates were stained with or without anti-β-glucan monoclonal antibody and then with Cy3-labeled secondary antibody as described in Materials and Methods. (B,C) SC5314-GFP isolated from mice treated with vehicle. (D,E) SC5314-GFP isolated from mice treated with 30 µg/kg CF. Images and CFU data are representative of more than 12 experiments using between one and four mice per treatment group. Scale bar in (B) is 10 microns long and applies to all images.
Figure 5.
Unmasking by caspofungin exposes epitopes recognized by the Dectin-1 β-glucan receptor.
Mice were infected and treated as in Figure 4. At 16 hr post-infection, kidney homogenates were stained with anti-β-glucan antibody followed by Cy3-labeled secondary antibody and Alexafluor 647-labeled recombinant Dectin-1-CRD. (A) Candida isolated from vehicle-treated mouse (B) Candida isolated from mouse pre-treated with 30 µg/kg CF. Images are representative of more than 6 experiments using between 1 and 3 mice per treatment. Scale bar in (A) is 10 microns long and applies to all images.
Figure 6.
Fluconazole treatment does not unmask β-glucan.
Mice were pre-injected with fluconazole and infected as in Figure 4. At 16 hr post-infection, kidneys were homogenized. (A) Homogenates were diluted and assayed for viable colony forming units or (B,D) homogenates were stained with anti-β-glucan monoclonal antibody and Cy3-labeled secondary antibody. Treatments of mice before infection were the following: (B) single dose of PBS (vehicle), (C) single dose of 2.5 mg/kg fluconazole in PBS, (D) single dose of 30 µg/kg caspofungin in PBS. Images and CFU data are representative of two independent experiments with 2 to 3 mice per group. Bars in (A) represent averages of triplicate counts and error bars represent standard deviations. Scale bar in (B) is 10 microns long and applies to all images.
Figure 7.
Caspofungin-mediated unmasking in vivo is filament-biased.
(A) Two examples in which yeast-form cells have very little exposed β-glucan, in contrast to the filamentous cell bodies. SC5314-GFP was isolated 16 hr post-infection and stained as above. (B) Quantification of exposure of hyphae and yeast-form cells in vehicle and CF-treated cases. Images are representative of more than 12 experiments. Quantification includes data from two independent experiments and includes 133 individual cell segments. Error bars represent standard deviations. All pair-wise comparisons were analyzed by two-tailed Student's T-test (Excel, Microsoft Corp.); asterisks indicate statistically significant differences (p≤0.0001). (C) Growth pattern in vivo is similar to in vitro, and proliferation leads to some unlabeled yeast-form cells already by 3 hours post-infection. (D) Quantification of labeling over time shows that, by 16 hr post-infection, no labeled cells are visible. Quantification includes data from two or three independent experiments per time point, although 100% labeling efficiency was always confirmed before infection. Scale bar in (A) is 10 microns long and applies to all images.
Table 1.
Strain details.