Effect of farnesol on a mouse model of systemic candidiasis, determined by use of a DPP3 knockout mutant of Candida albicans

Abstract

This work extends our previous observation that the fungus Candida albicans secretes micromolar levels of farnesol and that accumulation of farnesol in vitro prevents the yeast-to-mycelium conversion in a quorum-sensing manner. What does farnesol do in vivo? The purpose of this study was to determine the role of farnesol during infection with a well-established mouse model of systemic candidiasis with C. albicans A72 administered by tail vein injection. This question was addressed by altering both endogenous and exogenous farnesol. For endogenous farnesol, we created a knockout mutation in DPP3, the gene encoding a phosphatase which converts farnesyl pyrophosphate to farnesol. This mutant (KWN2) produced six times less farnesol and was ca. 4.2 times less pathogenic than its SN152 parent. The strain with DPP3 reconstituted (KWN4) regained both its farnesol production levels and pathogenicity. These mutants (KWN1 to KWN4) retained their full dimorphic capability. With regard to exogenous farnesol, farnesol was administered either intraperitoneally (i.p.) or orally in the drinking water. Mice receiving C. albicans intravenously and farnesol (20 mM) orally had enhanced mortality (P < 0.03). Similarly, mice (n = 40) injected with 1.0 ml of 20 mM farnesol i.p. had enhanced mortality (P < 0.03), and the onset of mortality was 30 h sooner than for mice which received a control injection without farnesol. The effect of i.p. farnesol was more pronounced (P < 0.04) when mice were inoculated with a sublethal dose of C. albicans. These mice started to die 4 days earlier, and the percent survival on day 6 postinoculation (p.i.) was five times lower than for mice receiving C. albicans with control i.p. injections. In all experiments, mice administered farnesol alone or Tween 80 alone remained normal throughout a 14-day observation period. Finally, beginning at 12 h p.i., higher numbers of C. albicans cells were detected in kidneys from mice receiving i.p. farnesol than in those from mice receiving control i.p. injections. Thus, reduced endogenous farnesol decreased virulence, while providing exogenous farnesol increased virulence. Taken together, these data suggest that farnesol may play a role in disease pathogenesis, either directly or indirectly, and thus may represent a newly identified virulence factor.

FIG. 1.

Strategy used to disrupt and then reconstitute C. albicans DPP3. It merges the fusion PCR and heterologous marker protocols of Noble and Johnson (31) with the SAT1 flipper protocols of Reuβ et al. (37). (A) Schematic diagram showing the method used to disrupt both alleles of DPP3. (B) Construction of disruption fragments for homologous recombination. Primers 1 and 3 and primers 4 and 6 were used to amplify the 350-bp flanking sequences upstream and downstream of DPP3, respectively. Primers 2 and 5 were used to amplify the two selectable markers. Red and green tails of primers are complementary sequences (31) needed for mutual primed synthesis in the fusion reaction. (C) 5′ and 3′ junctions of the selectable marker integration sites were confirmed using the primers mentioned. (D) Structure of the deletion cassette from pSFS2ADPP3 (top) in which the SAT1 flipper is inserted between two downstream fragments of DPP3. The genomic structure of KWN2 (middle) shows that the DPP3 allele has been replaced by one of the auxotrophic markers His1 and Leu2 (cross-hatched). Complementation of the DPP3 allele replacing the His1 marker (after flip-mediated excision of the nourseothricin marker) was used to create KWN3 (lower). dwstream, downstream.

FIG. 2.

Confirmation of the correct integration sites for the selectable markers used to knock out DPP3. His1dpp upstream check, the HIS1 Left and DPP3 upstream check primers were used to amplify the upstream region of the construct; His1dpp downstream check, the HIS1 Right and DPP3 downstream check primers were used to amplify the downstream region of the construct; Leu2dpp upstream check, the LEU2 Left and DPP3 upstream check primers were used to amplify the upstream region of the construct; Leu2dpp downstream check, the LEU2 Right and DPP3 downstream check primers were used to amplify the downstream region of the construct.

FIG. 3.

Southern analysis of ClaI-digested genomic DNA of SN152, KWN2, and KWN4 with two DPP3-specific probes. The two probes hybridize with the bp −350 and +456 regions of the gene, which has a single restriction site for ClaI at nucleotide 514. The sizes of the hybridizing fragments (in kb) are given on the left side of the blot. Fragments corresponding to hybridization at 3.7 kb are identical in size for the parent strain (SN152) and the complemented strain (KWN4). The KWN4 fragment corresponding to the other probe, hybridizing at 5 kb, is slightly different from that of the parent strain.

FIG. 4.

Altered pathogenicity for C. albicans strains with DPP3 removed and restored. Fifteen mice per group, each injected with 2 × 10⁶ cells. ○, SN152 parent; ▴, KWN2; •, KWN4.

FIG. 5.

Effect of orally administered mixed isomers of farnesol on mouse mortality caused by intravenous administration of wild-type C. albicans A72. ○, LD₅₀ (1.3 × 10⁶) of C. albicans-only control; □, LD₅₀ of C. albicans with farnesol in drinking water; each group contained seven replicates. Farnesol oral and i.v. control groups showed 100% survival.

FIG. 6.

Effect of i.p. administered E,E-farnesol on mouse mortality caused by intravenous administration of wild-type C. albicans A72. Each group of treatments had 40 replicates. All control groups contained five replicates. ○, C. albicans LD₅₀ (1.3 × 10⁶) dose, i.v. only; •, C. albicans LD₅₀ and 1 ml of i.p. 20 mM farnesol. Farnesol i.p. control and the 0.5% Tween 80 i.p. control never showed mortality.

FIG. 7.

Effect of i.p. administered E,E-farnesol on mouse mortality caused by sublethal levels of wild-type C. albicans A72. Each group contained 15 replicates. ○, sublethal C. albicans given i.v. with a single i.p. injection (1 ml) of 0.5% Tween 80 in sterile saline; •, sublethal C. albicans with a single i.p. injection (1 ml) of 20 mM farnesol. The farnesol i.p. and 0.5% Tween 80 i.p. controls (with no C. albicans) showed no mortality.

FIG. 8.

Time course of kidney infection by C. albicans A72 with or without i.p. farnesol. ○, C. albicans, 1.3 × 10⁶ cells, i.v. only; •, C. albicans, i.v., with 1 ml of 20 mM farnesol i.p. Each value is the average for three kidneys from three mice. Undiluted kidney homogenate (100 μl) was plated onto BiGGY agar plates (30). Values of 20,000 CFU/g of kidney correspond to actual plate counts of ≥800 CFU, deemed too numerous to count.