Restricted substrate specificity for the geranylgeranyltransferase-I enzyme in Cryptococcus neoformans: implications for virulence

Abstract

Proper cellular localization is required for the function of many proteins. The CaaX prenyltransferases (where CaaX indicates a cysteine followed by two aliphatic amino acids and a variable amino acid) direct the subcellular localization of a large group of proteins by catalyzing the attachment of hydrophobic isoprenoid moieties onto C-terminal CaaX motifs, thus facilitating membrane association. This group of enzymes includes farnesyltransferase (Ftase) and geranylgeranyltransferase-I (Ggtase-1). Classically, the variable (X) amino acid determines whether a protein will be an Ftase or Ggtase-I substrate, with Ggtase-I substrates often containing CaaL motifs. In this study, we identify the gene encoding the β subunit of Ggtase-I (CDC43) and demonstrate that Ggtase-mediated activity is not essential. However, Cryptococcus neoformans CDC43 is important for thermotolerance, morphogenesis, and virulence. We find that Ggtase-I function is required for full membrane localization of Rho10 and the two Cdc42 paralogs (Cdc42 and Cdc420). Interestingly, the related Rac and Ras proteins are not mislocalized in the cdc43Δ mutant even though they contain similar CaaL motifs. Additionally, the membrane localization of each of these GTPases is dependent on the prenylation of the CaaX cysteine. These results indicate that C. neoformans CaaX prenyltransferases may recognize their substrates in a unique manner from existing models of prenyltransferase specificity. It also suggests that the C. neoformans Ftase, which has been shown to be more important for C. neoformans proliferation and viability, may be the primary prenyltransferase for proteins that are typically geranylgeranylated in other species.

Fig 1

Ggtase-I involvement in thermotolerance and fungal cell morphology. (A) The cdc43Δ mutant has a growth defect at 37°C and 39°C. Five-fold serial dilutions of the indicated strains were spotted onto YPD medium and incubated for 48 h at the indicated temperatures. (B) The cdc43Δ mutant displayed apparent cytokinesis defects when grown at 37°C. Cells were incubated in YPD medium to mid-logarithmic phase at 30°C and shifted to preheated medium at 37°C for 18 h. Cell morphology was assessed by photomicroscopy.

Fig 2

Effect of cdc43Δ mutation on virulence. (A) The cdc43Δ mutant has a relative growth defect compared to wild-type and reconstituted strains in murine macrophages. C. neoformans cells were opsonized with anticapsular antibody (18B7) and coincubated with J774.1 murine macrophage-like cells that had been activated in PMA. After 1 h, unphagocytosed Cryptococcus cells were washed away, and phagocytosed cells were coincubated with macrophages for 24 h. Macrophages were lysed, and surviving fungal cells were quantitatively cultured. The viable output/input cell number was calculated for each strain and normalized to that of the wild type. ***, P < 0.001. (B) The cdc43Δ mutant has a delayed virulence in mice. Ten A/J mice were intranasally infected with 1 × 10⁵ cells for each strain and monitored for survival. WT, wild type.

Fig 3

Localization of Cdc42, Rac2, and Ras1. Wild-type and cdc43Δ strains expressing the indicated GFP fusion proteins were grown in YPD medium at 30°C and imaged by DeltaVision microscopy. The amino acid sequence of the C-terminal -Caax motif encoded by each allele is indicated under each image. Scale bar, 5 μm.

Fig 4

Cdc42 requires its prenylation motif for functionality. The WT, cdc42Δ mutant, cdc42Δ Cdc42 reconstituted strain, and the prenylation defective cdc42Δ Cdc42^C190A strains were serially diluted and incubated on YPD medium for 48 h at 30°C or 37°C.

Fig 5

Ggtase-I function is required for proper Rho10 localization but not for cell wall stress tolerance. (A) Wild-type and cdc43Δ cells expressing GFP-Rho10 were cultured in YPD medium at 30°C and imaged using DeltaVision microscopy. The amino acid sequence of the C-terminal -Caax motif encoded by each allele is indicated under each image. Scale bar, 5 μm. Ten-fold serial dilutions of each strain were spotted onto YPD medium containing the indicated cell wall stressor. Plates were incubated at 30°C for sufficient time to visualize colonies. CFW, calcofluor white.

Fig 6

Role of Ggtase-I in mating. (A) The indicated strains were coincubated in the dark on MS medium. The same location on each plate was imaged for 4 and 5 days. (B) Basidia and spore structures were imaged from matings shown in panel A. (C) Mating plugs were cut, permeabilized, and stained with calcofluor white to visualize the cell wall and with SYTOX green to visualize nuclei. Scale bar, 20 μm.

Fig 7

cdc43Δ mutant is hypersensitive to Ftase inhibitors. Wild-type and cdc43Δ cells were incubated at 37°C on yeast nitrogen base plates with sterile cotton disks containing 10 μl of 20 μM tipifarnib or 20 μM manumycin A.