Activities and genetic interactions of fission yeast Aps1, a Nudix-type inositol pyrophosphatase and inorganic polyphosphatase

ABSTRACT Inositol pyrophosphate 1,5-IP8 regulates expression of a fission yeast phosphate homeostasis regulon, comprising phosphate acquisition genes pho1, pho84, and tgp1, via its action as an agonist of precocious termination of transcription of the upstream lncRNAs that repress PHO mRNA synthesis. 1,5-IP8 levels are dictated by a balance between the Asp1 N-terminal kinase domain that converts 5-IP7 to 1,5-IP8 and three inositol pyrophosphatases—the Asp1 C-terminal domain (a histidine acid phosphatase), Siw14 (a cysteinyl-phosphatase), and Aps1 (a Nudix enzyme). In this study, we report the biochemical and genetic characterization of Aps1 and an analysis of the effects of Asp1, Siw14, and Aps1 mutations on cellular inositol pyrophosphate levels. We find that Aps1’s substrate repertoire embraces inorganic polyphosphates, 5-IP7, 1-IP7, and 1,5-IP8. Aps1 displays a ~twofold preference for hydrolysis of 1-IP7 versus 5-IP7 and aps1∆ cells have twofold higher levels of 1-IP7 vis-à-vis wild-type cells. While neither Aps1 nor Siw14 is essential for growth, an aps1∆ siw14∆ double mutation is lethal on YES medium. This lethality is a manifestation of IP8 toxicosis, whereby excessive 1,5-IP8 drives derepression of tgp1, leading to Tgp1-mediated uptake of glycerophosphocholine. We were able to recover an aps1∆ siw14∆ mutant on ePMGT medium lacking glycerophosphocholine and to suppress the severe growth defect of aps1∆ siw14∆ on YES by deleting tgp1. However, the severe growth defect of an aps1∆ asp1-H397A strain could not be alleviated by deleting tgp1, suggesting that 1,5-IP8 levels in this double-pyrophosphatase mutant exceed a threshold beyond which overzealous termination affects other genes, which results in cytotoxicity. IMPORTANCE Repression of the fission yeast PHO genes tgp1, pho1, and pho84 by lncRNA-mediated interference is sensitive to changes in the metabolism of 1,5-IP8, a signaling molecule that acts as an agonist of precocious lncRNA termination. 1,5-IP8 is formed by phosphorylation of 5-IP7 and catabolized by inositol pyrophosphatases from three distinct enzyme families: Asp1 (a histidine acid phosphatase), Siw14 (a cysteinyl phosphatase), and Aps1 (a Nudix hydrolase). This study entails a biochemical characterization of Aps1 and an analysis of how Asp1, Siw14, and Aps1 mutations impact growth and inositol pyrophosphate pools in vivo. Aps1 catalyzes hydrolysis of inorganic polyphosphates, 5-IP7, 1-IP7, and 1,5-IP8 in vitro, with a ~twofold preference for 1-IP7 over 5-IP7. aps1∆ cells have twofold higher levels of 1-IP7 than wild-type cells. An aps1∆ siw14∆ double mutation is lethal because excessive 1,5-IP8 triggers derepression of tgp1, leading to toxic uptake of glycerophosphocholine.

The present study is focused on the Schizosaccharomyces pombe Nudix enzyme Aps1, a 210-aa protein homologous to DIPP1 and Ddp1.Although Aps1 is inessential for vegetative growth, it plays a role in cellular phosphate homeostasis via its impact on IP 8 dynamics.The fission yeast phosphate acquisition (PHO) genes pho1 (cell surface acid phosphatase), pho84 (phosphate transporter), and tgp1 (glycerophosphodiester transporter) are repressed under phosphate-replete conditions by upstream lncRNAmediated transcriptional interference (21) and derepressed during phosphate starvation when synthesis of the interfering lncRNAs is turned off (22,23).Transcription of the upstream PHO lncRNAs interferes with the downstream PHO mRNA genes by displacing the activating transcription factor Pho7 from its binding site(s) in the mRNA promoters that overlap the lncRNA transcription units.PHO lncRNA 3'-processing and termination is a key control point in PHO mRNA repression; i.e., transcriptional interference can be tuned by increasing or decreasing the frequency with which Pol2 terminates lncRNA transcription prior to encounter with the mRNA promoter (21).Genetic maneuvers that enhance precocious termination of lncRNA transcription result in derepression of PHO mRNA expression in phosphate-replete cells, and those that reduce the probability of lncRNA termination prior to the mRNA promoter result in hyper-repression of the flanking PHO mRNAs relative to their basal levels.
The initial insights that inositol pyrophosphates are involved in fission yeast phosphate homeostasis were as follows: (i) deletion or active site mutations of the Asp1 kinase that synthesizes IP 8 hyper-repress pho1 under phosphate-replete conditions; (ii) inactivating mutations of the Asp1 pyrophosphatase domain or deletion of the Aps1 Nudix pyrophosphatase derepresses PHO genes under phosphate-replete conditions; and (iii) derepression of Pho1 by aps1∆ depends on synthesis of IP 8 by the Asp1 kinase (24)(25)(26).Simultaneous inactivation of the Asp1 and Aps1 pyrophosphatases is synthetically lethal (26), as is simultaneous deletion of aps1 + and siw14 + (12); i.e., we were unable to recover viable asp1-H397A aps1∆ or siw14∆ aps1∆ haploid progeny on YES medium after crossing the respective single mutants.These results signify that too much IP 8 is toxic to fission yeast.Multiple lines of genetic, biochemical, and transcriptomic evidence cohere to show that (i) IP 8 acts as an agonist of precocious PHO lncRNA transcription termination dependent on the 3' cleavage and polyadenylation factor (CPF) complex (26)(27)(28)(29)(30) and (ii) IP 8 toxicosis caused by Asp1 pyrophosphatase-inac tivating mutations (so-called asp1-STF alleles isolated in a genetic suppressor screen for relief of lncRNA-mediated transcriptional interference with pho1 expression) results from overexpression of the tgp1 gene and deleterious import of glycerophosphocholine (GPC) present in the growth medium (31).Moreover, pyrophosphatase-defective alleles aps1∆ and asp1-H397A display severe synthetic growth defects in combination with seb1-G476S, a mutation of the essential transcription termination factor Seb1, which results in derepression of pho1 and tgp1 mRNAs by enhancing precocious termination of the upstream interfering lncRNAs (32).
In this study, we present a biochemical and genetic characterization of S. pombe Aps1.We report that (i) recombinant Aps1 catalyzes magnesium-dependent hydrolysis of a linear inorganic polyphosphate chain (poly-P) to generate tripolyphosphate as an end-product and (ii) Aps1 cleaves the β-phosphate moieties from 5-IP 7 , 1-IP 7 , and 1,5-IP 8 to yield IP 6 as an end-product (Fig. 1).While an E89A-E93A mutation in the metal-binding site of the Nudix motif abolished Aps1 polyphosphatase activity in vitro, the catalytically defective aps1-(E89A-E93A) mutant retained partial biological activity in complementing the lethality of siw14∆ aps1∆ on YES medium.The growth defects of siw14∆ aps1∆, aps1∆ seb1-G476S, and asp1-H397A seb1-G476S on YES medium are alleviated by deletion of the GPC transporter Tgp1.These results fortify emerging evidence that Tgp1 overexpression is a major contributor to IP 8 toxicosis in fission yeast.

Recombinant S. pombe Aps1 is a metal-dependent inorganic polyphospha tase
We produced recombinant full-length Aps1 (amino acid sequence shown in Fig. 2) in Escherichia coli as a His 10 Smt3 fusion and isolated His 10 Smt3-Aps1 from a soluble bacterial extract by Ni-affinity chromatography.The tag was removed by treatment with the Smt3 protease Ulp1, and the Aps1 protein was recovered free of His 10 Smt3 after a second round of Ni-affinity chromatography.In parallel, we produced and purified N-terminally truncated versions: N∆10 [Aps1- ] and N∆25 [Aps1-(26-210)] (Fig. 2).These truncations were made in light of the AlphaFold tool prediction that the N-terminal 25-aa segment of Aps1 does not adopt a definite secondary structure (https://alphafold.ebi.ac.uk/entry/Q09790).The tag-free Aps1 proteins were subjected to a final gel filtration step, during which they eluted as single peaks consistent with monomeric native size.SDS-PAGE revealed comparable purity of the Aps1 full-length and N∆ proteins (Fig. 2).
To interrogate Aps1 enzymatic function, we reacted full-length Aps1 and the N∆ truncations with 5 mM MgCl 2 and 0.2 mM inorganic polyphosphate (poly-P 45 ) with an average linear polymer chain length of 45.The products were analyzed by electrophore sis through a 36% polyacrylamide gel, and the polyphosphate chains were visualized by staining the gel with toluidine blue.All three recombinant proteins elicited an Aps1 concentration-dependent conversion of poly-P 45 to a ladder of progressively shorter poly-P species, culminating in the production of tripolyphosphate (PPP i ) as the apparent end-product (Fig. 3A).This initial experiment established that (i) fission yeast Aps1, like its budding yeast Ddp1 and human DDIP homologs, has poly-P endopolyphosphatase activity and (ii) the N-terminal 25-aa peptide is dispensable for catalysis.All subsequent characterizations were performed with full-length Aps1.

Characterization of the Aps1 inorganic polyphosphatase
The temporal profile of the inorganic poly-Pase reaction underscores that Aps1 progressively shortened the polyphosphate chains, until virtually all of the input substrate was converted to tri-and tetrapolyphosphate (Fig. 3B).Poly-Pase activity was optimal at pH 6.5 to 7.0 and declined steadily as the pH was reduced to 5.0 or increased to 9.0 (Fig. 3C).Poly-P hydrolysis depended on exogenous magnesium and was optimal at 1 mM MgCl 2 (Fig. 3D).To affirm that the observed poly-Pase activity is inherent to the recombinant Aps1 protein, we produced and purified a full-length mutant, Aps1-(E89A-E93A), in which two of the essential metal-binding glutamates of the Aps1 Nudix motif (17) were changed to alanine.The Aps1-(E89A-E93A) mutant failed to hydrolyze poly-P 45 when assayed in parallel with wild-type Aps1 (Fig. 4A).Lonetti et al. (18) reported that the poly-Pase activity of S. cerevisiae Ddp1 was inhibited by µM concentrations of sodium fluoride.In this study, we found that µM levels of sodium fluoride inhibited poly-P hydrolysis by Aps1 (Fig. 4B).

Aps1 hydrolysis of inositol pyrophosphates
Aps1 (10 µM) was reacted for 30 minutes with 0.25 mM 1-IP 7, and varying concentra tions of magnesium and the products were analyzed by electrophoresis through a 36% polyacrylamide gel.The phosphorylated species were visualized by staining the gel with toluidine blue (Fig. 5A).Aps1 effected quantitative conversion of 1-IP 7 to IP 6 when the Mg 2+ concentration (0.25 mM) was equivalent to that of the 1-IP 7 substrate.Activity was maintained at 0.5 mM Mg 2+ , but was significantly inhibited at 1, 2, and 4 mM Mg 2+ (Fig. 5A).The inositol pyrophosphatase activity of mammalian DIPP proteins is inhibited by fluoride (13,16).In this study, we found that Aps1 hydrolysis of 1-IP 7 is sensitive to fluoride inhibition (Fig. 5B).
The extent of hydrolysis of 1-IP 7 to IP 6 in 0.25 mM Mg 2+ was proportional to input Aps1 (Fig. 6A).Based on the observation that 10 pmol of Aps1 sufficed to convert virtually all the input 1-IP 7 (2.5 nmol) to IP 6 , we estimated a turnover number of ~8.3 min −1 .Parallel titrations performed with 0.25 mM 5-IP 7 as the substrate showed that Aps1 specific activity with 5-IP 7 was approximately half of the activity observed with 1-IP 7 (Fig. 6B).Aps1 converted 1,5-IP 8 to IP 6 with specific activity similar to that with 1-IP 7 (Fig. 6C).These activities are comparable to the k cat value of 10.2 min −1 reported by Safrany et al. (14) for Aps1 hydrolysis of 5-IP 7 .The Aps1-(E89A-E93A) mutant failed to hydrolyze 1-IP 7 , 5-IP 7 , and 1,5-IP 8 when assayed in parallel with wild-type Aps1, which sufficed to convert all of the input substrate to IP 6 (Fig. 7).

Aps1 genetic interaction with Siw14
None of the three known fission yeast inositol pyrophosphatase activities is essential per se for vegetative growth; i.e., the pyrophosphatase-defective asp1-H397A strain and the aps1∆ and siw14∆ null strains grow on YES agar at 20°C to 37°C (Fig. 8 to 10).How ever, asp1-H397A is synthetically lethal with aps1∆ (26), suggesting that simultaneous ablation of these two pyrophosphatases results in accumulation of toxic levels of 1,5-IP 8 .Moreover, we found that aps1∆ was synthetically lethal with siw14∆, as gauged by the inability to recover viable aps1∆ siw14∆ haploid progeny of a pairwise cross after plating a large population of spores on YES medium, thus signifying that Aps1 and Siw14 pyrophosphatases have essential but redundant functions in fission yeast (12,30).Multiple lines of genetic evidence indicate that the lethality of the aps1∆ siw14∆ strain arises from unconstrained precocious transcription termination caused by too much IP 8 .That is to say, (i) the synthetic lethality of siw14∆ aps1∆ depends on the synthesis of 1,5-IP 8 by the Asp1 kinase and (ii) siw14∆ aps1∆ lethality is suppressed by loss-of-function mutations of multiple components of the fission yeast 3'-processing/ter mination machinery, including CPF (cleavage and polyadenylation factor) subunits Ctf1, Dis2, Ppn1, Swd22, and Ssu72 and termination factor Rhn1 (12).Previously, we showed that Siw14 pyrophosphatase activity is pertinent to the synthetic lethality of siw14∆ with aps1∆, i.e., the catalytically inert siw14-C189S active site mutant phenocopied siw14∆ with respect to synthetic lethality with aps1∆ (12).The issue is whether the catalytic activity of Aps1 is necessary for viability in the siw14∆ background.To address this question, we constructed a kanMX-marked aps1-(E89A-E93A) mutant strain of fission yeast encoding a catalytically inactive Aps1 protein.aps1-(E89A-E93A) cells grew as well as wild-type cells on YES agar (Fig. 8A) and dis played a modest derepression of Pho1 expression (Fig. 8B), akin to that seen for aps1∆ cells (26).A genetic cross of aps1-(E89A-E93A) and siw14∆ strains followed by random spore analysis (35) and selection for the kanRand hygR-marked aps1-(E89A-E93A) and siw14∆ loci yielded double-mutant progeny that germinated and formed colonies on YES agar.After amplification in liquid YES medium at 30°C and spotting serial dilutions of wild-type, single-mutant, and double-mutant cells on YES agar, we found that the aps1-(E89A-E93A) siw14∆ strain grew slowly at 30°C, 34°C, and 37°C, as gauged by colony size, and failed to form macroscopic colonies at 25°C or 20°C (Fig. 8A).We conclude that the catalytically dead Aps1-(E89A-E93A) mutant retains partial biological activity in complementing the inviability of siw14∆ aps1∆ spores.Pho1 acid phosphatase activity in the aps1-(E89A-E93A) siw14∆ strain was fourfold higher than that of the aps1-(E89A-E93A) single mutant (Fig. 8B), consistent with an additive effect on pho1 expression of ablating the functionally redundant Siw14 and Aps1 pyrophosphatases.

tgp1 + deletion suppresses the synthetic growth defect of siw14∆ aps1∆
Various asp1 mutations that delete or inactivate the Asp1 pyrophosphatase domain elicit growth defects in YES medium, ranging from severe sickness to lethality.The lethal alleles asp1-STF6 and asp1-STF9 were found to be defective for outgrowth after spore germination on YES medium (31).A key observation was that the growth defects of these and other asp1-STF pyrophosphatase-defective mutants were manifest in medium containing yeast extract but not in a synthetic medium ePMGT.A suppressor screen revealed that IP 8 toxicosis of asp1-STF mutants is caused by (i) a > 40 fold increase in the expression of the inessential tgp1 gene encoding a glycerophosphocholine transporter and (ii) the presence of glycerophosphocholine in the growth medium, which recapitu lates the toxicity of yeast extract to asp1-STF cells (31).
The question here is whether the synthetic phenotype of siw14∆ aps1∆ on YES agar is similarly connected to derepression of tgp1.By plating spores from a cross of siw14∆ and aps1∆ strains on ePMGT medium, rather than on YES, we were able to recover viable siw14∆ aps1∆ haploids that grew on ePMGT agar at 30°C, albeit more slowly than a wild-type strain or the siw14∆ and aps1∆ single mutants, as gauged by colony size (Fig. 9).Spot-testing the siw14∆ aps1∆ strain for growth on YES agar revealed it to be sicker at 30°C vis-à-vis ePMGT.The siw14∆ aps1∆ cells did not form macroscopic colonies on YES at 25°C or 20°C, but did yield tiny colonies on YES at 34°C and 37°C (Fig. 9).The key point is that the severe growth defects of siw14∆ aps1∆ cells on YES agar at 20°C-34°C were suppressed by deletion of tgp1, as was the small colony phenotype on ePMGT at 30°C (Fig. 9).Note, however, that tgp1∆ did not reverse the siw14∆ aps1∆ growth defect on YES at 37°C and that siw14∆ aps1∆ tgp1∆ cells were slow growing at 20°C, as gauged by colony size (Fig. 9).We surmise that IP 8 -driven overexpression of Tgp1 in siw14∆ aps1∆ is responsible for most (but not all) of the growth defects observed.

tgp1 + deletion suppresses the synthetic growth defect of aps1∆ seb1-G476S
A prior genetic screen for relief of transcriptional interference with Pho1 acid phospha tase expression unveiled a mechanism by which lncRNA termination is enhanced via a mutation G476S in the RNA-binding domain of an essential termination factor, Seb1 (32).seb1-G476S derepressed the pho1 and tgp1 mRNAs.RNA analysis showed that the tgp1-interfering nc-tgp1 lncRNA and the pho1-interfering prt lncRNA were terminated precociously in seb1-G476S cells (32).The seb1-G476S allele was found to be syntheti cally lethal with aps1∆; viz., (i) we were unable to obtain viable double-mutants after screening large population of haploid progeny of the genetic cross for growth on YES medium and (ii) wild-type progeny and the differentially marked seb1-G476S and aps1∆ single-mutants were recovered at the expected frequencies (32).The synthetic lethality of seb1-G476S with aps1∆ was rescued by CPF/Rhn1 loss-of-function alleles (32).In the present context, it was of interest to establish whether tgp1 overexpression was causal for the aps1∆ seb1-G476S synthetic phenotype.By plating spores from a cross of seb1-G476S and aps1∆ strains on ePMGT medium, we recovered viable seb1-G476S aps1∆ haploids that grew on ePMGT agar at 30°C, but were unable to grow on YES agar at any of the temperatures tested (Fig. 10).The lethal growth defect on YES was fully suppressed by tgp1∆ (Fig. 10).

tgp1∆ suppresses the synthetic growth defect of asp1-H397A seb1-G476S
The Asp1 pyrophosphatase mutation H397A elicits derepression of the PHO genes (26) but does not adversely affect cell growth on YES agar (Fig. 10).We noted previously that combining asp1-H397A and seb1-G476S exerted a severe (virtually lethal) growth defect on YES agar.In this study, we reconstructed the asp1-H397A seb1-G476S doublemutant and found that it grew well on ePMGT at 30°C but failed to grow on YES agar at any temperature (Fig. 10).The lethality of asp1-H397A seb1-G476S on YES was fully suppressed by tgp1∆ (Fig. 10).

tgp1∆ does not suppresses the lethality of asp1-H397A aps1∆
Simultaneous inactivation of the Asp1 and Aps1 pyrophosphatases is lethal; i.e., we were unable to recover an asp1-H397A aps1∆ double-mutant after crossing the single-mutant strains and plating spores on YES agar.This synthetic lethality was rescued by loss-offunction mutations in components of the 3'-processing and transcription termination machinery, e.g., CPF subunits Ppn1, Swd22, and Ssu72, and the Pol2 CTD prolyl isomerase Pin1 (26,36).By contrast, null alleles of CPF subunit Dis2 and termination factor Rhn1 did not rescue the asp1-H397A aps1∆ lethality (26).We identified loss-of-function mutations in three SPX domain proteins (Spx1, Gde1, and Vtc4) as suppressors of lethality or sickness associated with asp1-STF mutations (29).When these suppressor alleles were tested for their ability to rescue the lethality of asp1-H397A aps1∆ in a genetic cross, we found that inactivation of Spx1 allowed for the growth of asp1-H397A aps1∆ cells on YES agar at all temperatures, whereas inactivation of Gde1 and Vtc4 did not alleviate the asp1-H397A aps1∆ synthetic lethality (29).Recent studies indicated that failure to grow out after germination is the root of the apparent lethality of certain inositol pyrophosphatase mutants (31).We found that we were able to recover a viable but sick asp1-H397A aps1∆ tgp1∆ triple-mutant haploid in a cross of asp1-H397A and aps1∆ tgp1∆ strains.The asp1-H397A aps1∆ tgp1∆ cells grew equally poorly at all temperatures on either YES or ePMGT medium (Fig. S1).Thus, we hypothesize that inactivation of the two main inositol pyrophosphatases in fission yeast elevates IP 8 to a very high level, which elicits toxic transcription termination events independent of the precocious lncRNA termination that leads to Tgp1 overexpression and deleterious uptake of GPC.

Effects of Aps1, Siw14, and Asp1 mutations on cellular inositol polyphos phate levels
The advent of sensitive capillary electrophoresis electrospray ionization mass spectrome try (CE-ESI-MS) methods to profile the pool of cellular IP 6 , 5-IP 7 , 1-IP 7 , and 1,5-IP 8 (33,34) enables us to gauge how mutations of enzymes that synthesize and catabolize these molecules affect inositol pyrophosphate dynamics in vivo.Whole-cell perchloric acid extracts from 10 A 600 units of wild-type and mutant fission yeast cells growing logarithmically in ePMGT medium were adsorbed to titanium dioxide beads to enrich for inositol polyphosphates, which were eluted with 3% ammonium hydroxide.CE-ESI-MS was performed as described (33) on three biological replicates (independent cultures) of each fission yeast strain.The amounts (in pmol) of IP 6 , 5-IP 7 , 1-IP 7 , and 1,5-IP 8 in each sample were determined, and the levels of 5-IP 7 , 1-IP 7 , and 1,5-IP 8 were normalized to that of IP 6 to correct for any variations in sample recovery during extractions and TiO 2 bead affinity enrichment.The data for 1,5-IP 8 , 5-IP 7 , and 1-IP 7 are plotted in Fig. 11.In wild-type fission yeast, the 1,5-IP 8 /IP 6 , 5-IP 7 /IP 6 , and 1-IP 7 /IP 6 ratios were 0.0238, 0.0713, and 0.0505, respectively.
The results obtained for the various fission yeast mutants affirm and extend those of earlier studies that relied on metabolic labeling of cells with 3 H-inositol and fractionation of 3 H-labeled inositol polyphosphates by ion exchange chromatography (3,4).The key conclusions are listed below: 1.The Asp1 kinase, which is specific for phosphorylation of the 1-phosphate position in vitro, is uniquely responsible for synthesis of 1,5-IP 8 and 1-IP 7 , neither of which was detected in asp1∆ cells lacking Asp1 protein or in asp1-D333A cells expressing a kinase-dead Asp1 enzyme.2. Absence of Asp1 kinase leads to accumulation of high levels of 5-IP 7 (asp1∆/WT = 6.2;P value 0.0037).3. Absence of the Aps1 Nudix-type pyrophosphatase, which removes either the 1 or 5 β-phosphate in vitro, leads to a modest increase in 1,5-IP 8 (aps1∆/WT = 1.46;P value 0.025) and 5-IP 7 (aps1∆/WT = 1.71;P value 0.0036) but has a bigger effect on accumulation of 1-IP 7 (aps1∆/WT = 2.06; P value 0.011).This suggests that Aps1 might have a preference for dephosphorylating 1-IP 7 in vivo.These in vivo data resonate with our finding that Aps1 specific activity with 1-IP 7 was approximately twice that of the activity seen with 5-IP 7 .4. Absence of the Siw14 cysteinyl-phosphate-type pyrophosphatase, which removes either the 1 or 5 β-phosphate in vitro (26), leads to modest increases in 1,5-IP 8 (siw14∆/WT = 1.72;P value 0.072) and 5-IP 7 (siw14∆/WT = 1.57;P value 0.16) but has no effect on accumulation of 1-IP 7 (siw14∆/WT = 1.0).This suggests that Siw14 prefers to hydrolyze the 5 β-phosphate in vivo.We noted previously that Siw14 was approximately twofold more active at hydrolyzing 5-IP 7 versus 1-IP 7 when in vitro enzyme titrations were performed in the presence of 1 mM magnesium (26). 5. We gained insights into how different mutations in the histidine phosphatase-type pyrophosphatase domain of Asp1 have a gradient of effects on fission yeast growth in YES medium, ranging from benign (H397A) to toxic (STF6 and STF9).By growing the mutants in ePMGT medium permissive for growth of asp1-STF strains, the toxicosis (or lack thereof ) appears to correlate with the degree of accumulation of 1,5-IP 8 in the mutant versus wild-type.To wit, the 1,5-IP 8 mutant/WT ratios were 1.44 for asp1-H397A (P value 0.016), 2.62 for asp1-STF6 (P value 0.042), and 2.41 for asp1-STF9 (P value 0.0035).6. Kcs1 is the kinase that converts IP 6 to 5-IP 7 .Kcs1 is essential for viability.We recovered the kcs1-R332T allele in a screen for suppression of the toxicity of asp1-STF mutants and proposed that the R332T mutation reduces IP 6 kinase activity and thereby limits the pool of cellular 5-IP 7 , which is the substrate for synthesis of 1,5-IP 8 by Asp1 kinase (31).The CE-ESI-MS results support this view, insofar as 1,5-IP 8 and 5-IP 7 levels are reduced in kcs1-R332T cells, both in "raw" levels and when normalized to IP 6 .The normalized kcs1-R332T/WT ratios are as follows: 1,5-IP 8 (0.21; P value 0.016) and 5-IP 7 (0.24; P value 0.007).

DISCUSSION
The results herein extend our knowledge of the biochemical activities and genetic interactions of the fission yeast Nudix pyrophosphatase Aps1.With respect to biochemis try, we find that (i) Aps1 is a magnesium-dependent endopolyphosphatase that converts linear poly-P into shorter oligo-P species, culminating in tripolyphosphate as an apparent end-product; (ii) Aps1 is an inositol pyrophosphatase that hydrolyzes the β-phosphates of 5-IP 7 , 1-IP 7 , and 1,5-IP 8 to yield IP 6 as an end-product; (iii) Aps1 displays a ~twofold preference for hydrolysis of 1-IP 7 versus 5-IP 7 ; (iv) Aps1 endopolyphosphatase and inositol pyrophosphatase activities are inhibited by fluoride and abolished by alanine mutations of the conserved Nudix-box glutamates; and (v) the N-terminal 25-aa segment of Aps1 is dispensable for catalysis.The demonstration of Aps1's endopolyphosphatase activity in vitro resonates with our previous finding that the polyphosphate content of an aps1∆ strain, as gauged by PAGE and toluidine blue staining, was slightly higher than that of an aps1 + wild-type strain, especially the population of shorter polymers running at the same rate or faster than the Orange G dye (29).These results suggest that Aps1 contributes to polyphosphate catabolism in vivo.
The genetic and CE-ESI-MS analyses presented here clarify the contributions of Aps1 and two other inositol pyrophosphatases (Siw14 and Asp1) to IP 8 agonism of precocious transcription termination of the lncRNAs that interfere with the mRNA promoters of the PHO regulon genes (21), which can result in cytotoxicity if severe (31).We affirm that the Asp1 kinase is the sole enzyme responsible for synthesis of 1,5-IP 8 in fission yeast and that absence of Asp1 kinase leads to overaccumulation of its substrate 5-IP 7 .In contrast, the absence of Aps1 or Siw14 or an H397A mutation of the pyrophosphatase active site of Asp1 elicits modest increases in cellular IP 8 levels.While mutational ablation of IP 8 synthesis results in hyper-repression of the PHO genes tgp1, pho1, and pho84 (26), mutations of the three inositol pyrophosphatases exert disparate effects on gene expression, notwithstanding their similar extents of IP 8 accumulation (aps1∆/WT = 1.46; siw14∆/WT = 1.72; asp1-H397A/WT = 1.44).To wit, transcriptome profiling of siw14∆ cells showed no changes in the PHO regulon and identified only two mRNAs that were upregulated and 10 mRNAs that were downregulated (12).Thus, Siw14 per se has little impact on gene expression.Transcriptome profiling of aps1∆ cells highlighted 19 mRNAs that were upregulated, including tgp1, pho1, and pho84 (26).In asp1-H397A cells, 65 mRNAs were upregulated, including tgp1, pho1, and pho84 (26).The fold derepression of the PHO genes was greater in asp1-H397A cells than in aps1∆ cells.The reasons for this disparate impact are unclear, but it might reflect differential intracellular localization of the fission yeast inositol pyrophosphatases, leading to differential effects of their mutation on IP 8 levels in discrete cellular compartments.Current knowledge is scant in this regard.
What is clear is that there are profound mutational synergies between the three inositol pyrophosphatases, whereby aps1∆ siw14∆ and aps1∆ asp1-H397A double mutations are lethal (12,26).Recent studies of the asp1-STF6/9 alleles, in which large segments of the Asp1 C-terminal pyrophosphatase domain are deleted (27), revealed that their severe growth defect on YES medium (or on ePMGT medium supplemented with glycerophosphocholine) was caused by IP 8 -driven derepression of the glycerophos phocholine transporter Tgp1 and could be suppressed by an inactivating tgp1 mutation or a targeted tgp1∆ deletion (31).Thus, it was suggested that a supra-threshold level of IP 8 can lead to Tgp1-dependent toxicity via the agonist effect of IP 8 on precocious termination of the nc-tgp1 lncRNA that interferes with tgp1 mRNA synthesis (31,37).Consistent with this threshold model, we find that asp1-STF cells accumulate higher levels of IP 8 than asp1-H397A cells.
In this study, we provide evidence that the synthetic lethality of aps1∆ siw14∆ is also caused by excessive Tgp1-driven uptake of GPC, insofar as we were able to recover a viable aps1∆ siw14∆ mutant on ePMGT medium (lacking GPC) and to suppress the severe growth defect of aps1∆ siw14∆ on YES at 30°C by deleting tgp1.The aps1∆ and asp1-H397A mutations are both synthetically lethal with seb1-G476S, and this lethality is manifest in YES medium (but not ePMGT) and suppressed by deleting tgp1.
A surprising finding was that the catalytically defective aps1-(E89A-E93A) allele was not lethal in the siw14∆ background, though the aps1-(E89A-E93A) siw14∆ strain was slow growing.Our inference is that the Aps1 protein retains some biological activity in the absence of catalysis, which sustains viability of siw14∆ cells.We speculate that the Aps1-(E89A-E93A) protein might retain the capacity to bind IP 8 and thus sequester some of the excess IP 8 that accumulates when Siw14 and Aps1 pyrophosphatases are missing or crippled.On the flip side, the pyrophosphatase-dead siw14-C189S allele cannot sustain the viability of aps1∆ cells (12).
The fact that the severe growth defect of aps1∆ asp1-H397A could not be allevi ated either by growth on ePMGT medium or deletion of tgp1 suggests that IP 8 lev els in this double-pyrophosphatase mutant exceed an even higher threshold beyond which overzealous 3'-processing/termination affecting genes other than tgp1 results in cytotoxicity.Consistent with this second threshold 3'-processing/termination model, we found that aps1∆ asp1-H397A lethality was suppressed by the ssu72-C13S allele encoding a catalytically inactive version of the Ssu72 CTD phosphatase subunit of CPF (26).Even taking into account that Ssu72 mutation will soften the derepression of the PHO genes by aps1∆ asp1-H397A, an instructive finding was that the steady-state level of tgp1 mRNA (assayed by primer extension) was much higher in the aps1∆ asp1-H397A context than in the asp1-H397A single mutant (26).
Finally, our CE-ESI-MS results, taken in light of in vitro studies of recombinant enzymes, provide clues to the substrate preferences of the fission yeast inositol pyrophosphatases, whereby Aps1 seems to favor hydrolysis of 1-IP 7, while Siw14 favors hydrolysis of 5-IP 7 , as gauged by which IP 7 species accumulates most in the respective null mutants.These preferences are more subtle for the fission yeast enzymes Aps1 and Siw14 vis-à-vis their budding yeast orthologs Ddp1 and Siw14, as revealed by the recent work of Chabert et al. (38) who conducted a CE-ESI-MS analysis of inositol pyrophos phates in wild-type and mutant strains of S. cerevisiae.The ratios of IP 8 to 5-IP 7 to 1-IP 7 are similar in wild-type S. pombe (0.33/1.0/0.7) and wild-type S. cerevisiae (0.47/1.0/0.71).A noteworthy distinction between the two species is that while a fission yeast asp1∆ strain had no detectable 1,5-IP 8 or 1-IP 7 , the corresponding budding yeast vip1∆ mutant retained 25% of the wild-type concentration of IP 8 and 10% of the wild-type level of 1-IP 7 , indicating that budding yeast has an alternative kinase capable of phosphorylating the 1-phosphate group of 5-IP 7 and IP 6 (38).Most pertinent to the present work is their finding that ddp1∆ cells had a tenfold increase in 1-IP 7 but no change in 5-IP 7 and only a 20% increase in 1,5-IP 8 , signifying that budding yeast Ddp1 is acting predominantly on 1-IP 7 in vivo (38).Conversely, budding yeast siw14∆ cells accumulate fivefold higher levels of 5-IP 7 versus wild-type, a 25% increase in 1,5-IP 8 , and no significant change in 1-IP 7 (38).The latter result resonates with biochemical evidence that purified S. cerevisiae Siw14 has vigorous activity in hydrolyzing the 5-β-phosphate of 5-IP 7 and 1,5-IP 8 but negligible activity against 1-IP 7 (10).

Recombinant S. pombe Aps1
The ORF encoding full-length Aps1 was PCR-amplified from S. pombe complementary DNA (cDNA) with primers that introduced a BamHI site immediately flanking the start codon and an XhoI site downstream of the stop codon.Two Aps1 N-terminal trunca tions-N∆10 and N∆25-were generated by PCR using forward primers that introduced a BamHI site preceding codons 11 and 26.The PCR products were digested with BamHI and XhoI and inserted between the BamHI and XhoI sites of pET28b-His 10 Smt3 to generate T7 RNA polymerase-based expression plasmids encoding the full-length Aps1, N∆10, or N∆25 polypeptides fused to an N-terminal His 10 Smt3 tag.A double-alanine mutation, E89A-E93A, was introduced into the full-length Aps1 expression plasmid by two-stage overlap extension PCR with mutagenic primers.All plasmid inserts were sequenced to exclude the presence of unwanted mutations.The pET28b-His 10 Smt3-Aps1 plasmids were transfected into E. coli BL21(DE3) cells.Cultures (1,000 mL) ampli fied from single kanamycin-resistant transformants were grown at 37°C in Luria-Bertani medium containing 50 µg/mL kanamycin until the A 600 reached 0.6 to 0.7.The cultures were chilled on ice for 1 hour, adjusted to 2.2% (v/v) ethanol and 0.5 mM IPTG, and then incubated for 18 hours at 17°C with constant shaking.Cells were harvested by centrifugation and stored at −80°C.All subsequent steps were performed at 4°C.Cells were thawed and resuspended in 25 mL of buffer A (50 mM Tris-HCl, pH 7.4, 500 mM NaCl, 20 mM imidazole, and 10% glycerol), containing one complete EDTA-free protease inhibitor cocktail tablet (Roche).Lysozyme was added to a concentration of 1 mg/mL.After incubation for 30 minutes, the lysate was sonicated, and the insoluble material was removed by centrifugation at 16,000 rpm for 30 minutes.The supernatant was mixed for 1 hour with 5 mL of Ni-nitrilotriacetic acid agarose resin (Qiagen) that had been equilibrated with buffer A. The resin was recovered by centrifugation and washed twice with 50 mL of buffer A. The resin was centrifuged again, resuspended in 20 mL of buffer A, and poured into a column.After washing the column with 20 mL of buffer A, the bound material was eluted with 10 mL of buffer A containing 300 mM imidazole, while collecting 5-mL fractions.The polypeptide compositions of the flow-through and eluate fractions were monitored by SDS-PAGE.The 300 mM imidazole eluate fractions containing His 10 Smt3-Aps1 were supplemented with Smt3-specific protease Ulp1 (Ulp1/ His 10 -Smt3-Aps1 ratio of 1:425 [w/w]) and then dialyzed overnight against 2,000 mL of buffer B (50 mM Tris-HCl, pH 7.4, 250 mM NaCl, 20 mM imidazole, 1 mM DTT, and 10% glycerol) containing 1 mM EDTA, during which time the His 10 Smt3 was cleaved.The dialysates were mixed for 1 hour with 3 mL of Ni-nitrilotriacetic acid agarose resin that had been equilibrated with buffer B without EDTA.Tag-free Aps1 proteins were recovered in the flow-through fractions.The Aps1 protein preparations were concentra ted by centrifugal ultrafiltration (Amicon Ultra-15; 10 kDa cutoff) to 5 mL volume and then gel-filtered through a 125 mL 16/60 HiLoad Superdex 200 column (GE Healthcare) equilibrated in buffer C (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM DTT, 1 mM EDTA, and 5% glycerol) at a flow rate of 0.5 mL/min while collecting 2-mL fractions.The peak fractions were pooled, concentrated by centrifugal ultrafiltration (Amicon Ultra-15; 10 kDa cutoff), and stored at -80°C.Protein concentrations were determined with Bio-Rad dye reagent using bovine serum albumin as the standard.The yields of full-length Aps1, N∆10, N∆25, and Aps1-(E89A-E93A) were 14, 18, 15, and 14.5 mg per liter of bacterial culture, respectively.

Inorganic polyphosphatase assay
Reaction mixtures (10 µL) containing 50 mM Tris-HCl, pH 7.4 or 7.0 (as specified), 0.2 mM poly-P45 (Sigma, Cat # S4379-500MG, Lot # SLBX2788), and MgCl 2 and Aps1 protein (WT or variants) as specified in the figure legends were incubated for 20 or 30 minutes at 37°C.Reactions were terminated by adjustment to 50 mM EDTA and then mixed with an equal volume of 2 x Orange G loading buffer (10 mM Tris-HCl, pH 7.0, 1 mM EDTA, 30% glycerol, and 0.05% Orange G).The products were analyzed by electrophoresis at 4°C through a 20 cm 36% polyacrylamide gel containing 80 mM Tris-borate (pH 8.3), 1 mM EDTA for 2.5 hours at 10 W constant power.The gel was washed briefly with water and then stained with a solution of 0.1% toluidine blue (Sigma), 20% methanol, 2% glycerol, followed by destaining in 20% methanol.

Inositol pyrophosphatase assay
Reaction mixtures (10 µL) containing 50 mM Tris-HCl, pH 7.0, 0.25 mM inositol pyrophos phates 1-IP 7 , 5-IP 7 , or 1,5-IP 8 (chemically synthesized as described; 39-41), 0.25 mM MgCl 2 , and Aps1 protein as specified in the figure legends were incubated for 30 minutes at 37°C.The reactions were terminated by adjustment to 50 mM EDTA and then mixing with an equal volume of 2 × Orange G loading buffer.The products were analyzed by electrophoresis at 4°C through a 20 cm 36% polyacrylamide gel containing 80 mM Tris-borate (pH 8.3) and 1 mM EDTA for 3 hours at 10 W constant power.The inositol polyphosphates were visualized by staining the gel with toluidine blue, as described above.

Fission yeast strains
A list of strains used in this study is provided in Table S1.

Spot tests of fission yeast growth
Cultures of S. pombe strains were grown in liquid ePMGT (enhanced pombe minimal glutamate with thiamine) (23) or YES (yeast extract with supplements) medium until A 600 reached 0.3-0.8.The cultures were adjusted to an A 600 of 0.1, and aliquots (3 µL) of serial fivefold dilutions were spotted to ePMGT or YES agar.The plates were photographed after incubation for 2 days at 34°C, 2 to 2.5 days at 30°C and 37°C, 4 days at 25°C, and 6 days at 20°C.

Cell-surface acid phosphatase activity
Cells were grown at 30°C in YES medium.Aliquots of exponentially growing cultures were harvested, washed, and resuspended in water.To quantify acid phosphatase activity, reaction mixtures (200 µL) containing 100 mM sodium acetate (pH 4.2), 10 mM p-nitrophenylphosphate, and cells (ranging from 0.01 to 0.1 A 600 units) were incubated for 5 minutes at 30°C.The reactions were quenched by addition of 1 mL of 1 M sodium carbonate, the cells were removed by centrifugation, and the absorbance of the supernatant at 410 nm was measured.Acid phosphatase activity is expressed as the ratio of A 410 (p-nitrophenol production) to A 600 (cells).The data are averages (±SEM) of at least three assays using cells from three independent cultures.

Preparation of cell extracts and enrichment for inositol polyphosphates
Cultures of S. pombe strains were grown in liquid ePMGT medium at 30°C.Aliquots corresponding to 10 A 600 units of exponentially growing cells (A 600 between 0.6 and 0.8) were harvested by centrifugation.The cells were washed in ice-cold water and resuspended in 1 mL of 1 M perchloric acid.After snap-freezing in liquid nitrogen, the samples were stored at −80°C.The samples were thawed, mixed briefly by vortexing, and cell debris was removed by centrifugation at 16,200 g for 5 minutes at 4°C.Acidextracted inositol polyphosphates in the supernatants were then purified using titanium dioxide beads (GL Sciences 5020-75000), as described by Wilson and Saiardi (42).In brief, the cell supernatants were mixed with TiO 2 beads (4 mg per sample) that had been equilibrated in 1 M perchloric acid and incubated on a nutator for 20 minutes at 4°C.The TiO 2 beads (plus bound material) were recovered by centrifugation (5,000 g for 1 min at 4°C) and washed twice with 1 M perchloric acid.Inositol polyphosphates were eluted from the beads in two cycles of resuspension in 200 µL of 2.8% ammonium hydroxide, rotation of the samples for 5 minutes at 4°C, and centrifugation (5,000 g for 1 minute).The combined eluates (400 µL per sample) were evaporated to dryness in a vacuum centrifuge (Savant SpeedVac) at 42°C for ~3 hours.The dried samples were stored at −80°C and resuspended in 30 µL of water immediately prior to analysis by CE-ESI-MS.

Capillary electrophoresis electrospray ionization mass spectrometry (CE-ESI-MS)
The analyses were conducted as described (33,34).Inositol polyphosphate levels were determined with an Agilent 7100 capillary electrophoresis (CE) system coupled to a triple-quadrupole tandem mass spectrometry Agilent 6495 c system.Ionization was performed with the help of an Agilent Jet Stream (AJS) electrospray ionization (ESI) source and a liquid coaxial interface from Agilent.Ionization spray was stabilized with sheath liquid containing a water-isopropanol (1:1) mixture and pumped with an Agilent 1200 isocratic LC pump and a 1:100 splitter to a flow rate of 10 µL/min.Cell extract samples were spiked with heavy ( 13 C 6 ) inositol polyphosphate standards (43, 44; provided by Dorothea Fiedler), and 20-nl aliquots of the samples were injected by applying a pressure of 100 mbar for 20 seconds.A bare fused silica capillary (length 100 cm; inner diameter 50 µm) was used for measurements.Ammonium acetate (35 mM, adjusted to pH 9.75 with ammonia solution) was used as the background electrolyte (BGE).By applying a voltage of +30 kV, a stable CE current of 22 µA was received.The gas temperature of the nebulizer was set to 150°C with a flow rate of 11 l/min and a pressure of 8 psi.Sheath gas flow was set at 8 L/min with a temperature of 175°C.Capillary voltage was set to 2,000 V and nozzle voltage to 2,000 V. 70 V was the negative high-pressure radio frequency, and the low-pressure radio frequency was 40 V.The multiple reaction monitoring (MRM) transitions (shown in Table S2) were optimized with the MassHunter Optimizer software.

FIG 2 4 FIG 3
FIG 2 Recombinant full-length Aps1 and N∆ truncations.(Left panel) Aliquots (10 µg) of the Superdex 200 preparations of full-length Aps1 and the N∆10 and N∆25 truncated proteins were analyzed by SDS-PAGE.The Coomassie blue-stained gel is shown.The positions and sizes (kDa) of marker polypeptides are indicated on the left.(Right panel) Amino acid sequence of fission yeast Aps1.The margins of the truncations are indicated by arrows.The metal-binding glutamates E89 and E93 that were mutated to alanine are denoted by dots.

FIG 9
FIG9 Synthetic growth defect of siw14∆ aps1∆ is rescued by tgp1∆.Serial fivefold dilutions of fission yeast strains (as specified on the left) were spot-tested for growth on YES agar and ePMGT agar at the indicated temperatures.

FIG 10
FIG 10Synthetic growth defect of aps1∆ seb1-G476S and asp1-H397A seb1-G476S is suppressed by deletion of tgp1.Serial fivefold dilutions of S. pombe wild-type, single-mutant, and double-mutant strains with the indicated asp1, seb1, aps1, and tgp1 alleles were spot-tested for growth on YES agar and ePMGT agar at the temperatures specified.

FIG 11
FIG 11 Mutational effects on inositol pyrophosphate levels as gauged by CE-ESI-MS.The inositol pyrophosphate levels in extracts of the indicated fission yeast strains, normalized to that of IP 6 , are shown for 1,5-IP 8 (top panel), 5-IP 7 (middle panel), and 1-IP 7 (bottom panel).The bar heights depict the mean of three individual biological replicates (denoted by dots) ±SEM for all analyses, except for kcs1-R332T in the bottom panel, which shows the mean of two individual biological replicates ± the range of values.