Role of VTC4 in stress response and regulation of inorganic polyphosphate levels in yeast

Inorganic polyphosphate (polyP) is an important factor of stress tolerance in microbial cells. In yeast, the major enzyme of polyP biosynthesis is Vtc4, a subunit of the vacuole transporter chaperone (VTC) complex. In this study, we demonstrated that Vtc4 knockout in Saccharomyces cerevisiae not only decreased polyP content but also caused shifts in the composition of the intracellular polyP pool and changed the stress tolerance profile. In the mutant S. cerevisiae, the level of short-chain acid-soluble polyPs was decreased nearly 10-fold, whereas that of longer acid-insoluble polyPs was decreased only 2-fold, suggesting the existence of other enzymes compensating the production of long-chain polyPs. The Δvtc4 mutant showed inhibition of Mg-dependent phosphate uptake and decreased resistance to alkaline stress but increased tolerance to oxidation and heavy metal ions, especially Mn. Quantitative PCR revealed the upregulation of the DDR2 gene implicated in multiple stress responses and downregulation of PHO84 encoding a phosphate and Mn transporter, which could account for the effects on phosphate uptake and Mn-related stress response in the Δvtc4 mutant. Our study indicates that short-chain polyPs, plays an important role in the regulation of stress response in yeast.


Introduction
Inorganic polyphosphate (polyP), the linear polymer of orthophosphoric acid, is a universal regulatory biopolymer in living cells [1][2][3][4]. PolyP and enzymes of its metabolism are involved in various processes regulating vital activities of the cell. In bacteria, PolyPs are important for stress response and virulence [1,5,6], whereas in the human organism, polyPs play a key role in bone tissue growth and development [7,8], blood coagulation cascade, inflammatory response [9], and signal transduction in astrocytes [10]. Furthermore, PolyP is a component of a specific calcium channel in mitochondrial membranes regulating calcium level and stress response [11].
The discovery of the polyP synthase activity of Vtc4 solved the main problem of polyP synthesis in yeast, whose genome does not contain orthologs of bacterial polyphosphate kinases. The mechanism of Vtc4 polyphosphate polymerase activity has been clarified using X-ray crystallography, which has revealed that the Vtc4p 189-480 fragment contains a long-chain electrondense domain winding through the tunnel, suggesting that this module generates polyPs from ATP [18]; the catalytic domain faces the cytoplasm and the polymer must pass through the membrane. This Vtc4 fragment demonstrated phosphotransferase activity and could synthesize polyPs in solution from ATP, releasing ADP in the presence of Mn 2+ . Yeast Δvtc4 deletion strains lack the entire vacuolar polyP pool, while Δvtc1 point mutations targeting the conserved basic residues of transmembrane domains drastically reduce cellular polyP levels [18]. It was proposed that Vtc transmembrane domains of the other proteins in the VTC complex participated in the transport of polyPs across the vacuolar membrane [18]. The VTC exists as two sub-complexes: Vtc4/Vtc3/Vtc1 and Vtc4/Vtc2/Vtc1; the first is located mostly on the vacuole membrane and the second can also be observed in the endoplasmic reticulum and nuclear envelope but is found in the vacuolar membrane under phosphate starvation conditions [19]. The two differently regulated sub-by the amount of Pi produced after biomass hydrolysis in 0.5 M HClO4 at 90 °C for 20 min.

PolyP electrophoresis
The chain length of polyPs was assessed by electrophoresis as described previously [33] in 24% polyacrylamide gels with 7 M urea; commercial polyP15, polyP25, and polyP75 (Sigma-Aldrich, St. Louis, MO, USA) were used as standards (the numbers indicate the average amount of phosphate residues in the polyP chain). PolyPs were detected by staining the gels with toluidine blue.

Enzymatic assay of polyPs
For the enzymatic assay, the samples of polyP fractions polyP2 and polyP3 were neutralized to pH 7,0 by HCl aliquotes and incubated with S. cerevisiae exopolyphosphatase Ppx1 obtained as described earlier [34]. The reaction mixture containing 0.5 mL of 50 mM Tris-HCl (pH 7.2), 2.5 mM MgSO4, 0.02 mL (~5 U) of Ppx1 preparation, and 0.1 mL of polyP extracts was incubated at 30 °С for 2 h with shaking, and the released Pi was assayed as previously described [32].

Determination of yeast sensitivity to peroxide, alkali, and heavy metal ions
Yeast samples normalized by cell concentration (0.5 ·107 cell/ mL) were added to the wells of sterile plates containing the YPD medium and different concentrations of Cd(CH3COO)22H2O, MnSO4, CoSO4, NiSO4, hydrogen peroxide, or KOH, cultured for 24 h, and the optical density was measured at 594 nm using an EFOS photometer.

Pi uptake
Freshly harvested yeast cells (~ 55 mg wet biomass) were incubated in 0.75 mL of MiliQ water containing 110 mM glucose and 1 mM K2HPO4 and supplemented or not with 5 mM MgSO4, at 30 °C with shaking (850 rpm) in ThermoMixer (Eppendorf, Hamburg, Germany). After 60 min, the cells were centrifuged at 5,000 × g and Pi was assayed in supernatants by the colorimetric method with malachite green [34].

Quantitative PCR
Yeast were grown in 25 mL of YPD medium in 250-mL flasks at 28 °C, and biomass from 10 mL of culture was harvested after 36-h growth (stationary phase). Total RNA was extracted using the acid hot phenol method [35]; two biological replicates were used. RNA quality was assessed by electrophoresis in 1.5% agarose TBE gels. After treatment of RNA with DNase I (Thermo  Table S1) designed with Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primerblast/) and synthesized by Evrogen (Moscow, Russia). The reactions were performed with 2.5 ng of cDNA at the following cycling conditions: initial denaturation at 95 °C for 5 min and 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 40 s. To normalize gene expression levels, the S. cerevisiae ALG9 gene was used as reference [36]. The qRT-PCR results were statistically analyzed with Graph Pad Prism version 7.02 (GraphPad Software Inc., San Diego, CA, USA; https://www.graphpad.com/scientific-software/prism/) and gene expression levels were calculated relative to ALG9 expression using the 2 −ΔΔCT method [37].

Statistics
The experiments were performed in triplicate and the results are presented as the mean with standard deviation (Excel). The experiment with electrophoresis was repeated twice.

VTC4 knockout mutant contains polyPs
The question of whether S. cerevisiae cells lacking polyP synthase Vtc4 can still produce polyPs remains unresolved because of differences in polyP extraction methods [23-26]. We used a polyP extraction method that allowed obtaining separate fractions of polyPs with different chain length [22]. Comparison of the polyP content in the Δvtc4 and WT strains showed that the shortchain polyP1 fraction was decreased 10-fold, whereas longer chain polyP2 and polyP3 fractions decreased only 2-fold in Δvtc4 compared to parental cells, and polyP4 and polyP5 fractions containing the longest polyPs of yeast cells [22,38] did not differ between the strains ( Figure 1A).
Furthermore, the Pi level in the Δvtc4 strain was lower than that in the WT strain ( Figure 1A).
These results indicated that Δvtc4 cells still had a considerable amount of polyP2 and polyP3 with longer chains.
The presence of polyPs in the fractions was confirmed by electrophoresis and enzymatic assays. Electrophoregrams revealed characteristic bands stained by toluidine blue in polyP1, polyP2, and polyP3 fractions ( Figure 1B). The polyP chain length in these fractions was similar for WT andΔvtc4 cells ( Figure 1B) and corresponded to that observed earlier for S. cerevisiae [22,38]. PolyPs in polyP2 and polyP3 fractions were hydrolyzed by Рpx1 polyphosphatase; however, in both strains the hydrolysis was incomplete ( Figure 1C). Incomplete hydrolysis (88%) was also observed for commercial polyP with the average chain length of 188 phosphate residues. The inaccessibility of polyPs from biological sources for enzymatic hydrolysis has been reported earlier [30], but the reason is still unclear. One explanation can be the presence of Ca 2+ or Fe 2+ in polyP preparations obtained from the yeast cells, which could inhibit Ppx1 activity [39].
Overall, these results indicated that the Δvtc4 knock-out mutant could still synthesize a considerable amount of polyPs.

Pi uptake by WT and Δvtc4 strains
Considering the reduction of Pi and polyP levels in the mutant strain, we compared Pi uptake in the Δvtc4 strain with that in the WT strain in the presence of Mg2 + ions, which should stimulate the process through the effect on high-affinity phosphate transporter Pho84 involved in both Pi and manganese uptake in S. cerevisiae [40,41]. The results indicated that there was no stimulation of Pi uptake by Mg 2+ in the Δvtc4 strain (Figure 2A). The same experiment performed in the СRN strain and the СRN/PPN1 strain overexpressing polyphosphatase Ppn1 [31] revealed that Ppn1 overproduction reduced Pi uptake ( Figure 2B). strain [14]. These results suggest that the decrease in short-chain polyPs may serve as a signal to decrease Pi transport, possibly through downregulation of PHO84 gene expression or Pho84 transporter activity.

Stress tolerance
PolyP is an important factor in stress resistance of microorganisms. Vtc4 is involved in the function of vacuoles [16,17], which play a significant role in stress resistance of yeast [42].
Therefore, we compared the sensitivity of WT and Δvtc4 strains to stressful conditions such as high pH, hydrogen peroxide, and heavy metals. The Δvtc4 strain was more sensitive to alkaline stress, as evidenced by total growth cessation in the presence of 60 mM KOH, whereas the growth of the WT strain at this concentration was inhibited only by 50% ( Figure 3A). However, unexpectedly the mutant strain appeared to be more resistant to the other tested stressful conditions ( Figure 3B-F). The most pronounced difference between the strains was observed in the presence of Mn 2+ . Thus, the growth of the WT strain was already markedly inhibited at 2 mM MnSO4, whereas in the Δvtc4 strain the same level of growth inhibition was observed at 7 mM MnSO4 ( Figure 3E).

Real-time PCR
The resistance to manganese and peroxide stresses observed in the Δvtc4 strain was also characteristic for the Ppn1-overexpressing CRN/PPN1 strain [14], in which it was accompanied by changes in the transcription of genes associated with response to external stimulus, plasma This activity was found in the membrane fraction of yeast cells extracted with Triton X-100 [43].
The specific activity of the solubilized preparation was 20-fold higher than that in the protoplast 3-phospho-D-glyceroyl-1-phosphate + polyPn → 3-phosphoglycerate + polyPn+1 This activity was first observed in the adenine-deficient Neurospora crassa mutant, where concentrations of ATP and other adenyl nucleotides were sharply reduced [44], and was also detected in the WT N. crassa strain and some other bacterial species, but at a much lower level.
The Δvtc4 mutant is a convenient model for the search of alternative polyP biosynthesis enzymes and their cellular localization in future studies.
The second feature is the increased resistance of the Δvtc4 mutant to stresses caused by oxidation and heavy metal ions. In this respect, the Δvtc4 mutant is similar to the Ppn1overexpressing strain, which also exhibits a reduced content of short-chain polyPs (acid-soluble polyP1 fraction). The observed increase in the expression of the DDR2 gene encoding one of the important stress response regulators and a slight decrease in that of PHO8, the transporter of Pi and bivalent metal ions, indicate a possible similarity of stress response mechanisms in the Vtc4deficient and Ppn1-overexpressing strains. It is suggested that short-chain polyPs serve as signaling molecules and their decrease leads to the activation of stress response genes, which can be associated with the antioxidant properties of polyPs [45]. With regard to the reduced resistance of the Δvtc4 mutant to the alkaline stress, it is consistent with the notion of the important role of short-chain polyPs in cellular pH homeostasis. Thus, it was shown that a strain overexpressing endopolyphosphatase Ppn2, which cleaves long-chain polyPs into shorter ones, had a significantly higher content of short-chain polyPs and was more resistant to the alkaline stress than the parental strain [46].
In conclusion, our results indicate that Vtc4 polyphosphate synthase is responsible for the level of short-chained polyP and a part of long-chained polyP, suggesting the presence of