Re-emerging Aspartic Protease Targets: Examining Cryptococcus neoformans Major Aspartyl Peptidase 1 as a Target for Antifungal Drug Discovery

Abstract

Cryptococcosis is an invasive infection that accounts for 15% of AIDS-related fatalities. Still, treating cryptococcosis remains a significant challenge due to the poor availability of effective antifungal therapies and emergence of drug resistance. Interestingly, protease inhibitor components of antiretroviral therapy regimens have shown some clinical benefits in these opportunistic infections. We investigated Major aspartyl peptidase 1 (May1), a secreted Cryptococcus neoformans protease, as a possible target for the development of drugs that act against both fungal and retroviral aspartyl proteases. Here, we describe the biochemical characterization of May1, present its high-resolution X-ray structure, and provide its substrate specificity analysis. Through combinatorial screening of 11,520 compounds, we identified a potent inhibitor of May1 and HIV protease. This dual-specificity inhibitor exhibits antifungal activity in yeast culture, low cytotoxicity, and low off-target activity against host proteases and could thus serve as a lead compound for further development of May1 and HIV protease inhibitors.

Figure 1

Diagram of constructs used to identify May1 architecture and optimize expression strategy. Residue numbers indicate construct boundaries. The first 17 N-terminal amino acids, forming a putative native secretion signal, were replaced with a secretion signal compatible with the S2 expression system (not shown for clarity). Beveled rectangles denote positions of affinity tags used for purification.

Figure 2

Analysis of May1(17-434)-Avi expression and purification. (a) Silver-stained SDS-PAGE of samples collected throughout the purification process. MED, crude S2 media; PER, permeate from media concentration; LOAD, concentrated filtrate; FT, flow-through; W1-3, wash fractions; E1-4, elution fractions; POOL, pooled concentrated elution fractions; MTX, matrix binding control. (b) Western blot analysis with the same layout as in a visualized using Streptavidin-800CW (LI-COR, 1:15,000). (c) Silver-stained SDS-PAGE of concentrated pooled fractions after size exclusion chromatography.

Figure 3

Substrate specificity profile of May1 at the P4–P1′ positions presented as a heat map. L and D rows indicate the amino acid enantiomer at a given position. Dashed lines indicate substrate cleavage sites detected by LC–MS, with the most abundant N-proximal Phe*Leu site selected as a residue numbering reference.

Figure 4

(a) Crystal structure of recombinant May1 in free form (PDB 6R5H). The catalytic Asp residues are highlighted as magenta sticks. (b) Superposition of May1 (green) with human renin (magenta, PDB code 2REN(42)).

Figure 5

PepA binding to May1 (PDB 6R6A). (a) Top view of the binding pose of PepA in the May1 active site. The 2Fo-Fc map (contoured at 1.5σ) is shown in blue; catalytic aspartates (sticks with carbons colored cyan) interact with the central statine residue. Residues contributing to PepA binding in the S4–S3 subsites are indicated; those forming polar interactions are underlined. (b) Ligand–protein interaction diagram of the May1–PepA complex generated by LigPlot. Residues engaged in polar ligand interactions (green dashed lines) are shown; numbers represent distances between hydrogen bond donor and acceptor in Å. (c) Model of N-terminally carboxybenzylated phenylstatine superimposed with PepA. Points of attachment for the R1 and R2 substituents are indicated by arrows. The protein is represented by its solvent accessible surface area colored by electrostatic potential (red for negative, blue for positive). Residues available to form polar interactions with the P1′ substituent are shown as sticks and labeled, and possible interactions are indicated by dashed lines.

Figure 6

Linear statin inhibitor structures and residue preference heatmaps. L and D rows indicate the amino acid enantiomer at a given position. (a) Initial screening of the N-proximal site with a degenerate C-terminal residue. A final concentration of 500 nM per individual compound was used in library screening. (b) Deconvolution of selected libraries with favorable N-proximal residues based on initial screening. Single-compound deconvolution experiments were conducted with a 1 μM final concentration of the test compound.

Figure 7

Evaluation of May1 inhibitor activity against cultured C. neoformans yeast. (a) Terminal saturation densities of yeast cultured for 48 h in YNB minimal medium containing May1 inhibitors or a 0.5% DMSO vehicle. Dashed lines represent terminal OD₆₀₀’s of the DMSO-treated H99 and may1Δ controls. Error bars represent standard deviation of the mean from triplicate cultures. (b) Growth curves of H99 and may1Δ yeast treated with lead compound 13a, negative control compound 1b, or a 0.5% DMSO vehicle control in either standard YNB medium (left) or YNB medium buffered to pH 6.5 (right). Error bars represent standard deviation of the mean from triplicate cultures.

Figure 8

Comparison of May1 (PDB code 6R5H) with HIV-1 Pr (PDB code 2hvp(62)). (a) Overall superposition of May1 (green) with HIV-1 Pr (maroon). Catalytic aspartates are highlighted in stick representation. (b) Top view into the May1 active site. The protein is represented by the solvent accessible surface area colored by electrostatic potential; PepA is represented as sticks with carbon atoms colored green; the inhibitor pose in HIV-1 Pr is shown with black lines. (c) Top view into the HIV-1 Pr active site. The protein is represented by solvent accessible surface area colored by electrostatic potential; acetyl-pepstatin is represented as sticks with carbon atoms colored yellow; the PepA pose in May1 is shown with black lines. In b and c, flap regions covering the active site are omitted for clarity.