Ncs2* mediates in vivo virulence of pathogenic yeast through sulphur modification of cytoplasmic transfer RNA

Abstract

Fungal pathogens threaten ecosystems and human health. Understanding the molecular basis of their virulence is key to develop new treatment strategies. Here, we characterize NCS2*, a point mutation identified in a clinical baker's yeast isolate. Ncs2 is essential for 2-thiolation of tRNA and the NCS2* mutation leads to increased thiolation at body temperature. NCS2* yeast exhibits enhanced fitness when grown at elevated temperatures or when exposed to oxidative stress, inhibition of nutrient signalling, and cell-wall stress. Importantly, Ncs2* alters the interaction and stability of the thiolase complex likely mediated by nucleotide binding. The absence of 2-thiolation abrogates the in vivo virulence of pathogenic baker's yeast in infected mice. Finally, hypomodification triggers changes in colony morphology and hyphae formation in the common commensal pathogen Candida albicans resulting in decreased virulence in a human cell culture model. These findings demonstrate that 2-thiolation of tRNA acts as a key mediator of fungal virulence and reveal new mechanistic insights into the function of the highly conserved tRNA-thiolase complex.

Graphical Abstract

Figure 1.

NCS2* affects yeast growth and tRNA 2-thiolation. (A) Chemical formula of 5-methoxycarbonylmethyl-2-thiouridine (mcm⁵s²U); modifications are indicated in red. (B) Structure of human Posttranscriptional chemical modifications are indicated as coloured spheres. Inset: mcm⁵s²U highlighted by spheres (sulphur: yellow; carbon: green; oxygen: red). (C–E) Spot-dilution assays on YPD using isogenic yeast strains. (C) Growth at different temperatures. (D) Growth under different chemical stress conditions at 30 °C and 37 °C (2.7 nM rapamycin; 1.5 mM diamide; 6 mM caffeine). (E) Growth assay on zymocin. (F) Northern-blot analysis of tRNA from wild type and NCS2* yeast grown at 25 °C, 30 °C and 37 °C in YPD. The probe targets ; the upper gel contains ([N-acryloylamino]phenyl)mercuric chloride (APM). (G) Quantification of the fraction of s²U-labeled tRNA of northern-blot replicates including (F). NCS2: 69 % ± 5 % versus NCS2*: 63 % ± 9 % at 25 °C, P = 0.385; NCS2: 65 % ± 9 % versus NCS2*: 69 % ± 12 % at 30 °C, P = 0.671; NCS2: 21 % ± 10 % vs. NCS2*: 53 % ± 3 % at 37 °C, P = 0.007 (n = 3). Data are represented as mean ± SEM. ** = P ≤ 0.01, ns = P > 0.05 (two-sided Student's t-test).

Figure 2.

NCS2* modulates tRNA functionality in a thiolation-dependent manner. (A, B) Spot-dilution assay of NCS2 and NCS2* strains in combination with gene deletions that affect wobble uridine modifications grown on YPD. (A) NCS2 and NCS2* in combination with elp4Δ and elp6Δ (3 mM caffeine). (B) NCS2 and NCS2* in combination with urm1Δ (6 mM caffeine). (C) Northern-blot analysis of total RNA from different yeast strains grown at 37 °C in YPD. The probe is against . The gel contains ([N-acryloylamino]phenyl)mercuric chloride (APM).

Figure 3.

Mechanism of NCS2* function. (A) Spot-dilution assay of diploid yeast strains containing combinations of NCS2 alleles grown on YPD (2.7 nM rapamycin). (B) Phyre-based structural model of the Ncs2/6 complex with tRNA in cartoon representation with close-up views of the L71 residue (left) and the catalytic site of Ncs6 (right). Putative binding sites of ATP and the coordination of the iron/sulphur cluster are indicated with ball-and-stick models. Zinc atoms (light brown) positioned in the N- and C- terminal domains of both Ncs2 and Ncs6 are enlarged for visibility (Ncs2: cyan, Ncs6: purple, tRNA: brown). (C) Spot-dilution assay of different NCS2 alleles grown on YPD (2.7 nM rapamycin). (D) Northern-blot analysis of total RNA from strains expressing different NCS2 alleles grown at 37 °C in YPD. The probe targets . The gel contains ([N-acryloylamino]phenyl)mercuric chloride (APM). The fraction of s²U-labeled tRNA is indicated below. (E) Yeast-two-hybrid assay of Ncs6 with Ncs2 variants. Ncs2: 133.5 ± 17.8 Miller units [MU], Ncs2*: 251.8 ± 12.7 MU, Ncs2^H71A: 130.2 ± 15.1 MU, Ncs2^H71I: 109.2 ± 12.8 MU. Ncs2/Ncs6 vs. Ncs2*/Ncs6: P = 0.0007; Ncs2/Ncs6 vs. Ncs2^H71A/Ncs6: P = 0.814; Ncs2/Ncs6 vs. Ncs2^H71I/Ncs6: P = 0.127 (n = 3 independent yeast-two hybrid assays for each construct). Data are represented as mean ± SEM. *** = P ≤ 0.001, ns = P > 0.05 (Two-sided Student's t-test). (F) Western blot analysis of yeast strains expressing Ncs6-HA and Ncs2-TAP or Ncs2*-TAP, respectively. Translation was blocked by adding 200 μg/ml cycloheximide to the culture at timepoint 0. Samples were taken every 30 min. (G) Insoluble protein aggregates were isolated from ncs2Δ, NCS2 or NCS2* yeast grown in YPD at 30 °C or 37 °C, respectively. Proteins were separated by SDS-PAGE and subsequently visualized by Colloid Coomassie staining (left: total extracts; right: aggregates).

Figure 4.

NCS2 modulates virulence in pathogenic baker's yeast. (A) Northern-blot analysis of total RNA from wild type and ncs2Δ yeast derived from clinical isolates grown at 30 °C in YPD. The probe is against . The gel contains ([N-acryloylamino]phenyl)mercuric chloride (APM). The fraction of s²U-labeled tRNA is indicated below. (B) Spot-dilution assay of NCS2 alleles generated in the clinical isolates YJM128 and YJM223 grown on YPD (YJM128: 5 nM rapamycin; YJM223: 4 nM rapamycin). (C) Northern-blot analysis of total RNA from YJM128-derived strains shown in (B) grown at different temperatures in YPD. The probe is against . The gel contains APM. The fraction of s²U-labeled tRNA is indicated below. (D) Survival curves of mice injected with NCS2/NCS2, NCS2*/NCS2* or ncs2Δ/ncs2Δ yeast; P < 0.0001; Mantel-Cox Log-rank test (n = 18 animals per group). The yeast strains used for in vivo infection experiments were derivatives of YJM128.

Figure 5.

NCS2 in Candida albicans. (A) Northern-blot analysis of total RNA from wild type and ncs2Δ/Δ C. albicans grown in YPD at 30 °C. The probes are against , and . The gel contains ([N-acryloylamino]phenyl)mercuric chloride (APM). The fraction of s²U-labeled tRNA is indicated below. (B) Spot-dilution assay of NCS2/ncs2Δ and ncs2Δ/Δ cells grown at different temperatures on YPD (10 nM rapamycin). (C) Spot-dilution assay of different NCS2 alleles grown on YPD at different temperatures. (D) Wild type and ncs2Δ/Δ C. albicans grown at 37 °C using conditions that induce hyphae formation (YPD and choco medium). Scale bar is 500 μm. (E) Damage to human kidney (A498) and colon (C2BBe1) cell lines as determined by release of host LDH. Damage caused by the ncs2Δ/Δ strains is shown compared to the host cell damage elicited by the isogenic wild-type strain SN87HL (100 %) (n = 4 or 5 biological replicates; significance levels compared to the wild type 100 %). ** = P ≤ 0.01, *** = P ≤ 0.001, **** = P ≤ 0.0001 (two-sided Student's t-test).

Figure 6.

Translation defects in Candida albicans. (A) Relative A-site codon-occupancy in ncs2Δ/Δ cells relative to wild type (mean ± SD; n = 3). Codons cognate for tRNAs with mcm⁵s²U₃₄ are marked in red. Dot size indicates the relative frequency of each codon in the wild-type A site (the larger the more frequent). (B) Principal component analysis (PCA) of gene expression of wild type (blue) and ncs2Δ/Δ (red) C. albicans grown in YPD at 30 °C (light colour), 37 °C (medium colour), and 42 °C (dark colour). Replicates are indicated as circles, triangles, and squares. (C, D) Gene ontology (GO) terms mis-regulated in ncs2Δ/Δ C. albicans relative to wild type at 37 °C. (C) Cellular component. (D) Biological process.