Figure 1.
Pga7 plays a central role in heme and hemoglobin-iron utilization in C. albicans.
PGA7 and RBT5 deletion strains (A, C) or PGA7 re-integrant strains (B, D) were grown in iron-free RPMI medium in the presence of increasing concentration of either human hemoglobin (A, B) or hemin (C, D) as a sole source of iron. Optical density was measured after 3 days at 30°C. Error bars represent standard deviations of triplicates. The following strains were used: A, C: CAI4 (KC79, KC590)), KC589 (rbt5−/−), KC626 (pga7−/−), KC594 (rbt5−/− pga7−/−). The two wild-type starting strains from different sources were highly concordant. B, D: KC645 (WT<vector>), KC646 (pga7−/−<vector>), KC647 (pga7−/−<PGA7>).
Figure 2.
The pga7−/− strain is less virulent in a mouse model of systemic candidiasis.
Survival of intravenously challenged mice with Candida albicans strains is indicated. Groups of 5 mice were inoculated with 2 independent clones of pga7−/− (KC646) or re-integrant (KC647) and were grouped together and compared to the wild type (KC645). Mice infected with the pga7−/− strains survived significantly longer that mice infected with the wild type strain (p = 0.0024), whereas mice infected with the PGA7 re-integrant strains did not survive significantly longer (p = 0.2310).
Figure 3.
Rbt5 is more abundantly expressed than Pga7.
(A) Western blot detection of cell envelope fractions of cells expressing FLAG-tagged Pga7 (KC605) or native Rbt5 under iron starvation conditions, vs. a control lacking both RBT5 and PGA7 (KC594). Left panels: SDS extract of the cell pellet after mechanical lysis. Right panels: 4% ß-ME extract of the cell pellet remaining after SDS extraction. Top panels: detection of the membrane with α-Rbt51, which cross-reacts with Rbt5 (Rbt51 itself is very weakly expressed under these conditions [14]). Bottom panels: detection of the membrane with monoclonal α-FLAG. (B) Expression of FLAG-tagged Rbt5 (KC706) or Pga7 (KC605), or FLAG-tagged Pga7 expressed under the RBT5 promoter, (KC712) in C. albicans cells grown under iron-limiting conditions (YPD +1 mM ferrozine), was measured by labeling the cells with 35S-methionine/35S-cysteine, followed by immunoprecipitation using a rabbit anti-FLAG antiserum, and separation of the proteins by SDS-PAGE. C = control cells with no FLAG.
Figure 4.
Different localization of Pga7 and Rbt5 expressed at similar levels.
(A) Cells expressing either FLAG-tagged Rbt5 (KC713) or FLAG tagged Pga7 under the RBT5 promoter (KC711), and a control wild type strain (KC68), were grown under iron limiting condition and fixed with formaldehyde. The fixed cells were either washed in PBS (“no treatment”), incubated with 250 mM NaOH (“NaOH treatment”), or fully permeabilized with zymolyase, β-ME and methanol (“Zymolyase treatment”). The cells were then incubated with anti-FLAG antibody followed by anti-Mouse IgG-Cy3 coupled antibody. red = FLAG, blue = DAPI. Bars = 10 µm. (B) To investigate the membrane or cell wall attachment mode of Rbt5 and Pga7, sequential fractionation and extraction of the cell culture shown in A was performed. An equal relative amount of each fraction was loaded in each lane, and the proteins were detected by immunoblotting with a rabbit anti-FLAG antibody. The FLAG-Rbt5 or FLAG-Pga7 signal was quantitated in each lane. The total amount of each protein was obtained by combining the signals of all the fractions together. The relative amount of protein in each given fraction, compared to the total for that protein, is indicated below the gel. The total signal of FLAG-Pga7 was about 15% less than the total signal of FLAG-Rbt5.
Figure 5.
Interaction of Pga7 and Rbt5 with heme.
(A) Recombinant wild type Rbt523–219 and Pga718–195 or their mutants in a conserved aspartic acid residue were incubated with hemin agarose (Hm) or glutathione agarose (glut) beads. The beads were washed and bound proteins were released by heating to 100°C in SDS gel loading buffer, separated by PAGE and detected by Western blotting with an anti-Myc antibody. (B) ITC analysis of the interaction of hemin with Rbt523–219 (left panel), and with Pga718–195 (right panel). The proteins (60 µM) were titrated into 10 µM hemin (Pga7) or into 20 µM hemin (Rbt5). Representative experiments are shown, with the heat signal in the top half of each panel and the binding isotherm derived from this signal in the lower half. The average fitting of three independent experiments with Rbt5-heme gave n = 2.152±0.0706 sites, Ka = 1.852×105±3.679×104 M−1 and H = −7454±304.7 KJ mol−1. The average fitting of three independent experiments with Pga7-heme gave n = 1.930±0.00999 sites, Ka = 1.531×107±1.817×106 M−1 and H = −7002±56.69 KJ mol−1. (C) UV/visible spectra from the titration of 1 µl hemin aliquots (2 mM hemin stock) into 3 ml of 5.5 µM apoPga718–195 (1 cm path length cuvette, PBS). The inset shows the hemin-Pga7 absorbance minus hemin alone absorbance at 406 nm, at increasing hemin concentrations. The inflection point indicates saturation of hemin binding to Pga7 at close to 1∶1 concentration.
Figure 6.
Pga7 and Rbt5 can extract heme from hemoglobin.
(A) Hemoglobin covalently conjugated to CnBr sepharose beads was incubated with 50 µM recombinant Rbt523–219 or Pga718–195, or with PBS buffer alone. After 30 min incubation, the UV-Vis spectrum of the supernatant was measured with a Nano-Drop 2000 spectrophotometer. The peak at 280 nm represents the protein. The 406 nm Soret absorption peak visible in the supernatants containing Rbt5 and Pga7 is typical of protein-bound heme. The average reading of a triplicate experiment is represented. (B) Upfield areas of 1H NMR spectra recorded at 600 MHz and 298 K to monitor the transfer of heme to Pga718–195. Pga7 shows signals attributable to heme methyls in the range 53–67 nnm (top panel). Hemoglobin is characterized by broader unresolved features in the 50–100 ppm range. Titration of hemoglobin (middle panels) with increasing amounts of Pga7 gives rise to the transfer of the heme from hemoglobin to Pga7 with progressive disappearance of the broad features of hemoglobin and appearance of the characteristic signals of heme methyls in Pga7. Note that hemoglobin contains 4 heme groups per molecule.
Figure 7.
Transfer of heme between Rbt5 and Pga7.
The indicated proteins were separated by gel filtration and the protein-bound heme was detected by absorbance at 406 nm. Red curves: 50 µM of either apo-Rbt523–219 or apo-Pga718–195 in 100 µl was incubated for 5 min with 25 µM heme before loading on the column. Black and green curves: after pre-incubation of apo-Rbt523–219 or apo-Pga718–195 with heme, 50 µM of the second protein was added and incubated a further 5 min prior to loading on the column. The areas under the curves were similar (+/−10%) in all four runs.
Figure 8.
Holo-Pga7 interacts with apo-Rbt5 and with apo-Pga7.
SPR analysis was carried out by immobilizing apo-Pga718–195, apo-Rbt523–219, and the D72A mutant of apo-Rbt523–219, on a Biacore biosensor chip. 3 µM holo-Pga718–195 was injected at time zero for 60 sec.
Figure 9.
Schematic model of the heme relay network that extracts heme from hemoglobin and delivers it to the endocytic pathway.
Cell wall polysaccharide organization was modeled after [62], [63]. The identity of the membrane receptor that mediates internalization into the cell – Rbt5, Pga7, or another protein – is unknown.
Table 1.
List of C. albicans strains.
Table 2.
List of P. pastoris strains.