Many prokaryotic and eukaryotic cells store inorganic phosphate in the form of polymers called polyphosphate (polyP). There has been an explosion of interest in polyP over the past decade, in part due to newly suggested roles related to diverse aspects of human health. The physical interaction of polyP chains with specific proteins has been proposed to regulate cellular homeostasis and modulate signaling pathways in response to environmental changes. Recently, several studies have challenged existing models for how polyP interacts with its protein targets, while identifying new motifs that are capable of binding to polyP. In this review, we summarize these findings, delineate the functional implications for polyP-protein interactions at the molecular level, and define open questions that should be addressed to propel the field forward.

The metabolism of polyP chains

Inorganic phosphates can be assembled into negatively charged and energy rich linear chains called polyphosphate (hereafter polyP) [1]. Recent years have seen a surge of interest in polyP biology based in part on discoveries highlighting tantalizing links to human disease from SARS-CoV-2 [2] to amyotrophic lateral sclerosis [3]. While connections to human health are exciting, it is noteworthy that the fundamentals of polyP metabolism are perhaps best understood in microorganisms.

In bacteria, polyP synthesis is achieved by the action of polyphosphate kinase (PPK) enzymes, which transfer the gamma phosphate of ATP to growing polyP chains [4]. While Escherichia coli have just one PPK enzyme [4], other bacteria have both PPK1 and PPK2 [5]. These enzymes have unique properties, but in some cases function redundantly to regulate polyP dynamics [6]. Bacterial polyP accumulation is countered by the action of the PPX exopolyphosphatase enzyme, which degrades polyP beginning at the end of the chain to release monomers of inorganic phosphate [7]. While most bacteria possess only one ppx gene, some bacteria such as Mycobacterium tuberculosis and Corynebacterium glutamicum express PPX1 and PPX2 proteins, both of which have exopolyphosphatase activity [8,9].

The PPK enzyme is not conserved in yeasts such as Saccharomyces cerevisiae. Here, the vacuolar transporter chaperone complex (VTC) is embedded in the vacuole membrane and uses ATP as a substrate to synthesize polyP chains of 3–300 units in length [10]. These chains are translocated into and stored within the vacuole lumen (i.e. inside the vacuole) at high concentration (>200 mM by some measurements) [11], serving in part as a reserve of inorganic phosphate that can be rapidly mobilized during starvation [12]. PolyP is also found in other cellular compartments (e.g. nucleus, cytoplasm, and mitochondria), but the processes by which these pools are made, or perhaps transported to these locations, are unknown [13]. Also unclear is the actual amount of polyP in these organelles, with estimates varying widely from study to study and no doubt complicated in part by the unintended rupture of polyP-rich vacuoles during fractionation procedures [13]. In fact, the forced accumulation of polyP outside the yeast vacuole, driven by ectopic expression of the bacterial PPK, is toxic [12,14]. This observation strongly suggests that there are upper limits to the polyP normally permitted to accumulate in some areas of the cell.

As in bacteria, the turnover of polyP chains in yeast also depends on the action of polyphosphatase enzymes, which include both exopolyphosphatases that cleave terminal phosphates, and endopolyphosphatases that cleave chains internally [15]. We can think of these enzymes as analogous to exo- and endo-nucleases that act on nucleic acid (RNA or DNA) substrates [16]. Some of these polyphosphatases are vacuolar (Ppn1, Ppn2, and possibly Pho8), and function to modulate overall chain length or to degrade polyP chains into free inorganic phosphate during phosphate starvation [10,17–19]. The Ppx1 and Ddp1 polyphosphatases are found predominately in the cytoplasm [17]. In contrast with the vacuolar enzymes, the function of these proteins in overall polyP homeostasis is unclear [17]. We speculate that they function to titrate the level of minor populations of cytoplasmic polyP in response to environmental changes.

In mammalian cells, there are lower levels of polyP (typically in the range of 25–120 µM) [20], although the enzymes involved in its metabolism are poorly understood [21]. The FoF1 mitochondrial ATPase has been suggested to synthesize polyP [22], but its contribution to total cellular levels of polyP remains unclear [23]. Several mammalian proteins have been proposed as polyphosphatases [17], with the best evidence for in vivo function resting with NUDT3 (a homolog of yeast Ddp1), which may be particularly important for regulation of the DNA damage response during oxidative stress [24].

The current landscape of polyP-protein interactions

PolyP impacts a wide variety of biological processes, from stress resistance in bacteria to blood clotting in mammals [25]. How does such a simple molecule impart its diverse functions at a molecular level? We don't expect there to be a single answer to this question. However, there are likely to be common mechanisms at play that can guide our understanding of polyP biology across evolution. One interesting possibility was proposed by Azevedo et al. [26], who in 2015 provided evidence that polyP chains could be covalently linked to lysine residues within two yeast proteins, Nsr1 and Top1, in a non-enzymatic reaction. Mutations that abrogate so called ‘polyphosphorylation’ also altered protein subcellular localization and resulted in changes to Top1's topoisomerase activity in vitro [26]. For both proteins, polyphosphorylation occurred in polyacidic serine and lysine rich (abbreviated as PASK) motifs [26].

The idea that polyP can bind to proteins to impact their function is not new. In bacteria, early work by Arthur Kornberg's lab showed that polyP can bind to the Lon protease and stimulate its role in degrading ribosomal proteins during starvation [27]. In mammals, polyP impacts blood coagulation by binding to various clotting factors [28]. Finally, the conserved histidine α-helical (CHAD) domain, conserved in prokaryotes and eukaryotes, has been characterized as a polyP-binding domain [29].

Yet, the thought that polyP could alternatively function as a new lysine-based post-translational modification (PTM) is particularly interesting, especially in light of the idea that it could directly compete with other modifications that occur on lysines, such as methylation, acetylation, or ubiquitylation [30]. The evidence for covalent attachment of polyP to proteins rested in large part on the intriguing observation that Top1 and Nsr1 isolated from wild-type cells display a dramatic decrease in electrophoretic mobility, relative to proteins isolated from polyP-deficient vtc4Δ mutants, when analyzed on Bis-Tris polyacrylamide gels (NuPAGE) [26]. The ‘polyP shift’ persisted when extracts were generated under harsh denaturing conditions (e.g. urea or SDS) [26], usually assumed to disrupt non-covalent interactions.

The following years saw an explosion in the identification of proteins that shifted via NuPAGE analysis in the presence of polyP. Our lab initially published 15 PASK-containing targets, with enrichment of proteins localized to the nucleolus and involved in various aspects of ribosome biogenesis and translation control [31]. We also provided the first evidence for polyphosphorylation of human proteins, with the identification of six targets [31]. Work with human proteins was extended by Azevedo et al. [32], who used a protein microarray to screen for polyP-binding proteins. They confirmed six targets that fit previously established criteria for polyphosphorylation, namely a polyP-dependent shift on NuPAGE gels [32]. In addition, they identified five proteins that did not undergo polyP shifts on NuPAGE gels, but still seemed to interact with polyP as judged by mobility shift on native PAGE or by chromatography, suggesting non-covalent binding [32]. The fact that some polyP-binding proteins behave differently on NuPAGE gels is consistent with alternative modes of polyP interaction.

Revisiting defining characteristics of polyphosphorylation

The collective work between 2015 and 2020 raised the exciting possibility that polyP is a global effector of protein function. Indeed, PASK motifs appear often across eukaryotic species [31]. Recent efforts, including contributions from our group, demonstrated that histidine and lysine repeats (polyHis and polyLys) can also interact with polyP molecules [33,34] (Figure 1). In contrast with what was originally proposed for lysine polyphosphorylation, the salt and pH sensitivity of polyP interaction with polyHis repeat proteins suggested non-covalent binding — despite the fact that the interaction slowed migration on NuPAGE gels [33]. Beyond expanding the scope of proteins that can interact with polyP, the polyHis study called into question the use of NuPAGE gels to discern covalent versus non-covalent interactions, and prompted a re-analysis of the biochemical characteristics of lysine polyphosphorylation. Neville et al. [34] showed that the NuPAGE shift for lysine polyphosphorylation targets was lost when reactions carried out on beads were subsequently washed with high salt or high pH buffers. Importantly, these salt and pH sensitivities were confirmed in separate experiments using size exclusion chromatography [34]. Together, these results challenged the assertion that polyphosphorylation is covalent, since we expect covalent modifications to be resistant to these conditions.

PolyP-binding sequence element motifs.

Figure 1.
PolyP-binding sequence element motifs.

PolyP has been shown to bind proteins containing polyacidic serine and lysine-rich (PASK) motifs, positively charged surfaces, histidine repeats (polyHis), and lysine repeats (polyLys). Examples (e.g.) of proteins with known modes of polyP binding that were discussed in the review are provided.

Figure 1.
PolyP-binding sequence element motifs.

PolyP has been shown to bind proteins containing polyacidic serine and lysine-rich (PASK) motifs, positively charged surfaces, histidine repeats (polyHis), and lysine repeats (polyLys). Examples (e.g.) of proteins with known modes of polyP binding that were discussed in the review are provided.

Close modal

We note that the findings of Neville et al. are seemingly at odds with a more recent experiment conducted by Azevedo et al. [35] where protein extracts adjusted to 500 mM salt concentration are permissive for polyP-induced shifts for yeast Nsr1 and human nucleolin proteins, previously described as polyphosphorylated targets. A key difference between these works is that in Neville et al. [34] excess polyP was washed away from the target protein prior to analysis. The ability of targets to shift on NuPAGE gels may depend on the absolute quantity of polyP in the sample. Regardless, we conclude that shifts on NuPAGE (or other types of gels for that matter), while indicative of interaction with polyP, are a poor proxy for distinguishing covalent versus non-covalent interactions.

A unified nomenclature for polyP-protein interactions

With these new findings, we now consider the evidence for covalent attachment of polyP to proteins to be lacking. Of course, we cannot say for certain that covalent attachment of polyP to lysine or other amino acids never happens. For example, it is possible that polyP interaction with some targets is covalent in vivo. While there is currently no evidence to support this idea, models are made to be broken. Despite a reinterpretation of the mode of polyP interaction, the original work from Azevedo et al. [26] remains an outstanding contribution.

At this time, the term polyphosphorylation — with -ylation almost exclusively denoting a PTM — seems somewhat confusing. Other terms for non-covalent polyP interaction with proteins have been suggested. For example, our colleagues have proposed the term ‘histidine and lysine polyP modification’ [36]. However, the introduction of new language seems largely unnecessary when the terms ‘polyP binding’ and ‘polyP-binding proteins’ suffice.

It appears that polyP-binding proteins are not exactly rare. To date, at least 100 targets have been validated across prokaryotic and eukaryotic species (Supplementary Table S1), with additional candidates identified via large scale analyses. PolyP-binding proteins include PASK, polyHis, and polyLys containing proteins, as well as those that interact with polyP via a collection of positive charges concentrated on (or across) a target's surface (Figure 1). Below, we highlight key molecular outcomes of polyP-protein interactions and provide examples for each case. For now, we consider these outcomes independent of the proposed mode of binding (i.e. covalent versus varied types of non-covalent interactions), but it is not impossible that distinct interactions could preferentially lead to certain fates. Importantly, outcomes are not necessarily mutually exclusive, and polyP's impact on a given target could certainly be context specific. A summary is presented in Figure 2.

Outcomes of polyP-protein binding.

Figure 2.
Outcomes of polyP-protein binding.

PolyP binding to proteins can disrupt or promote protein-protein interactions or protein-substrate interactions. PolyP binding can also stabilize partially unfolded proteins or play a role in promoting protein aggregation. Additionally, polyP binding events can modulate the enzymatic activity of proteins or contribute to the regulation of signaling pathways within cells. Overall, polyP has wide-reaching roles in the regulation of protein function and activity through direct binding.

Figure 2.
Outcomes of polyP-protein binding.

PolyP binding to proteins can disrupt or promote protein-protein interactions or protein-substrate interactions. PolyP binding can also stabilize partially unfolded proteins or play a role in promoting protein aggregation. Additionally, polyP binding events can modulate the enzymatic activity of proteins or contribute to the regulation of signaling pathways within cells. Overall, polyP has wide-reaching roles in the regulation of protein function and activity through direct binding.

Close modal

Protein folding and aggregation: two sides of the same coin?

PolyP performs a chaperone-like role in stabilizing intermediate structures of ‘client’ proteins including citrate synthase [37], luciferase [37], and lactate dehydrogenase (LDH) [38] when these proteins are partially unfolded. Size exclusion chromatography experiments conducted with LDH suggest that polyP behaves analogous to a protein chaperone, that binds the partially unfolded protein, to promote the formation of intermediate stabilizing structures composed of soluble beta-sheets [38]. Additionally, in the case of LDH, intermediate structures formed in the presence of polyP could be refolded by actual protein chaperones upon removal of the thermal stressor that was initially used to unfold it [38,39]. How the activities of polyP and protein chaperones are co-ordinated during refolding is deserving of further investigation. This type of protein stabilization may have physiological relevance during conditions of heat or oxidative stress [37,38]. As noted previously, the mammalian NUDT3 polyphosphatase is proposed to regulate polyP levels during oxidative stress [24], and we wonder if changes in the folding of nuclear proteins could explain increases in H2AX phosphorylation, a marker of DNA damage, in NUDT3 depleted cells [24]. On the other hand, polyP has been shown to bind amyloidogenic proteins that have a propensity to aggregate, oligomerize, and nucleate [40]. For these proteins, polyP binding results in a different fate. It accelerates the aggregation and nucleation of these proteins, causing them to form compact, insoluble structures [40]. An example is the fibrillation of the E. coli CsgA protein [41]. Here, polyP plays a role in stabilization of the protein to promote biofilm formation [41]. Other proteins that are structurally reorganized by polyP include those involved in the progression of neurodegenerative Alzheimer's and Parkinson's diseases such as Aβ1–40/42, Tau, β2-microglobulin, and alpha-synuclein proteins [41–43]. Interestingly, fibrillation of proteins can happen in the absence of polyP, at a slower rate, presumably resulting in less compact fibrils that are cytotoxic and contribute to faster disease progression [41]. Work in neuronal cell lines and Caenorhabditis elegans neurodegenerative models suggests that polyP-stabilized alpha-synuclein fibrils are less cytotoxic and more susceptible to proteolytic degradation compared with their unstabilized counterparts [41]. Two key residues (K43 and K45) within a lysine-rich region of alpha-synuclein may be critical for polyP binding [44]. An open question is whether polyP can stabilize fibrils as they come apart and act as a chaperone to modulate their refolding. Understanding the residues or regions of proteins that interact to promote aggregate formation will help better predict the outcome of polyP binding to partially unfolded proteins and how it could play a role in reversing protein aggregation.

Modulation of protein-protein interactions

Beyond its role in modulating protein folding, polyP can impact target interactions with other proteins.

Promoting interactions

One of the earliest descriptions of a polyP-binding protein was made by Kuroda et al. [27], who showed that polyP binding facilitates degradation of target proteins by the Lon protease in E. coli. In this scenario, polyP is suggested to bind both Lon and the target, serving as a molecular adaptor. This was first proposed to play a role in the targeting of Lon to ribosomal subunits and nucleoid proteins, during starvation, as part of a broader response that allows cells to adapt to the lack of nutrients [27,45,46]. Subsequent work found that this mechanism also contributes to the arrest of DNA replication initiation in E. coli by regulating the pool of replication initiator protein, DnaA, when bound to ADP [47]. An interesting proposal by Ropelewska et al. [48] is that most substrates degraded by Lon in a polyP-dependent manner also bind polyP themselves. As another example, two-hybrid assays suggest that polyP binding promotes formation of a heterodimer complex between the Francisella tularensis transcription factors MglA and SspA [49]. However, it is unclear if disruption of this complex in ppk-ppx mutant strains is due to the direct loss of polyP or a downstream effect of the mutations themselves [49].

The effect of polyP may be particularly impactful when it binds to multiple proteins functioning in the same pathway. PolyP's role in the blood coagulation contact pathway is an excellent example. PolyP binds to diverse coagulation factor proteins (FV [50], FXI [51], FXII [52], and thrombin [53]) and plays a role in activating them either directly, or indirectly by accelerating the rate of their interactions. We speculate that polyP may play a similar role in bridging transient interactions in other cases where it interacts with multiple proteins in close proximity. For example, we previously identified many polyP interacting proteins that localize to the nucleolus [31], and theorize that transient interactions mediated by polyP may be important for ribosome biogenesis.

Disrupting interactions

In addition to promoting interactions, polyP can also disrupt or prevent interactions between two or more proteins. For example, polyP competes with FVIII to bind von Willebrand Factor (VWF) [54], and instead promotes VWF interaction with glycoprotein Ib [55]. This mechanism plays a role in increasing VWF binding to platelets, which is required for platelet activation and aggregation [54,55]. PolyP has also been reported to disrupt the interaction between PASK-containing Nsr1 and Top1 in vitro [26]. Whether polyP impacts interaction of these proteins in vivo is not clear, but it is tempting to speculate that this mechanism serves to regulate localization of both proteins to the nucleolus, which is more pronounced in vtc4Δ mutant cells that cannot make polyP [26]. Notably, similar polyP-induced changes in localization have been observed for the mammalian kinase DYRK1A [33], which has wide reaching roles in neurogenesis [56], cell proliferation [57,58], and cell homeostasis [59]. Specifically, in the presence of excess polyP synthesized by ectopically-expressed bacterial PPK, DYRK1A failed to form nuclear speckles [33]. We surmise that this effect and other changes in localization ultimately stem from changes to protein-protein interactions. For example, occlusion of a nuclear localization signal (which often contain lysine residues [60]) by polyP binding could alter a protein's subcellular distribution.

Modulation of cell signaling

Notably, polyP also directly interacts with cell surface receptors to initiate intracellular signaling events. While these receptors are also proteins, this function of polyP is deserving of special attention since the source of polyP is extracellular (e.g. from microorganisms [61], damaged cells [3], or dense granules of platelets [62]). For instance, polyP binding to RAGE and possibly P2Y1 receptors of endothelial cells leads to the activation of pathways like mTORC2 signaling [63] and inflammatory responses involving H4 and HMGB1, which also bind polyP with high affinity [64]. Additionally, the predicted binding of polyP to P2Y1 is thought to activate calcium signaling and release of internal polyP stores by astrocytes [65]. Intriguingly, released polyP can also act as a neurotransmitter to influence activation of neighboring cells and their uptake of polyP [65]. Conversely, polyP can also block receptor binding sites for other ligands. For example, polyP binding to the ACE2 receptor is thought to block the binding of the SARS-CoV-2 Spike protein, thereby preventing viral entry [2,66]. In these examples, polyP is proposed to act at the cell surface. However, it is also plausible that polyP may actually enter the cell during receptor internalization [2]. For example, via ACE2 internalization, bound polyP could enter the cell and potentially serve as an intracellular signaling molecule or perform one or more of the functions listed above.

Regulation of enzymatic activity

PolyP binding may also inhibit the enzymatic activity of proteins. For example, in vitro assays show that in the presence of polyP, Top1 loses the ability to relax supercoiled DNA [26]. Similarly, polyP inhibited the in vitro kinase activity of DYRK1A [33]. PolyP also inhibits the catalytic activity of HAL, a histidase from Trypanosoma cruzi proposed to interact with polyP, via its lysine-rich intrinsically disordered region [67]. Of note, it is unclear how the ability of polyP to chelate ions contributes to in vitro enzymatic assays and this is deserving of additional consideration in future work. Conversely, polyP could promote enzyme activation. We note that Wang et al. [68] demonstrated that polyP stimulates the kinase activity of mTOR, the master regulator of cellular homeostasis, in a concentration-dependent manner in vitro. In vivo, ectopic expression of the yeast exopolyphosphatase Ppx1 reduced phosphorylation of mTOR target PHAS-1 and reduced cell viability [68]. Given this evidence, it would be important to establish if mTOR regulation occurs via direct binding to polyP.

Modulation of protein-nucleic acid interactions and protein-lipid interactions

PolyP binding can modulate the affinity of targets for other molecules besides proteins. For instance, polyP has been shown to compete with DNA for binding to the mammalian transcription factor MafB, which is unable to bind DNA when polyP-bound [33]. However, it is also possible for polyP to alter the RNA or DNA binding specificity of target proteins. This activity of polyP has been shown for the E. coli DNA- and RNA-binding protein Hfq [69], and a similar effect is proposed for the sigma factor σ80 of Heliobacter pylori [70]. Here we speculate that polyP binding stimulates changes in protein conformation or cofactor interactions that alter target binding specificity. In another example, polyP-stabilized lipopolysaccharide micelles bind to the TLR4 receptor of macrophages more readily than in the absence of polyP [71]. However, in this case, it is unclear if polyP directly interacts with TLR4 receptors or if its role is more indirect.

Characteristics of polyP-binding elements

PolyP-binding sequence elements are positively charged in nature, suggesting that polyP interaction might simply be determined by the charged state of proteins and the corresponding anionic nature of the polymer. However, it is unlikely that charge is the only consideration. Interestingly, mutation of lysine and histidine residues in PASK and polyHis motifs to arginine prevents polyP interaction [26,31,33,72]. Why does positively charged arginine not support binding in these contexts? Neville et al. [36] have previously discussed the idea that the guanidinium group in the arginine side chain results in a delocalized charge that prevents polyP binding. However, we also note that loss of polyP binding for arginine mutations has only been demonstrated using NuPAGE analysis [26,31,33,72]. It is possible that polyP still binds to arginine-rich sequences, but with a lower affinity that does not permit detection on NuPAGE gels. As such, we encourage the use of alternative techniques, described below, to investigate the binding properties of arginine-rich sequences. This same logic extends to investigating the contribution of neighboring amino acids such as acidic and serine residues found in the PASK motif [26,73]. Finally, it is curious that in contrast with PASK and polyHis motifs, arginines have been implicated in polyP binding for the endopolyphosphatase Ddp1 [74], the polyP kinase PPK2 [75], and the polyP-binding CHAD domain [29]. Therefore, it could also be true that context is important — PASK and polyHis sequences are shorter motifs often located in regions of little structure, whereas Ddp1, PPK2, and CHAD domains bind to polyP across folded protein surfaces. Finally, we do not discount the possibility that polyP binding to protein targets could be impacted by direct or indirect interactions with peptide backbones.

Methodological considerations for studying polyP-binding proteins

While NuPAGE analysis may be useful for identification of the strongest polyP binders, the fact that these gels are (largely) denaturing, presents a challenge for understanding how proteins might interact with polyP in vivo. Under denaturing conditions, polyP binding could be mediated by amino acids that are not normally exposed in natively folded proteins. Immobilization of natively folded proteins on beads, followed by washing prior to NuPAGE would help to address these concerns [33]. Still, other more precise methods than gel electrophoresis are needed to identify and study the kinetics of polyP-binding proteins under native conditions. For example, plate-based enzyme-linked immunosorbent assays have recently been used to measure the affinity of polyP for three different plasma proteins [52], as well as the SARS-CoV-2 ACE2 receptor [2]. Even more sophisticated tools such as surface plasmon resonance analysis [51,55,76] and grating-coupled interferometry [29] have also been used to accurately measure dissociation constants for polyP-protein interactions. Since polyP concentrations vary across different cell types and even within regions of a single cell, accurate measurements of polyP binding parameters are critical to establish whether an interaction is likely to occur in vivo.

Finally, there has been some success in solving the crystal structure of polyP-binding proteins in the presence of their polyP ligand. For example, the PPK2 enzyme from Francisella tularensis [75] and the Ddp1 endopolyphosphatase from Saccharomyces cerevisiae [74] have been crystalized in the presence of polyP. The polyP-bound structure of PPK2 revealed a ‘polyP channel’ composed of positively charged amino acids and a ‘lid loop’ containing an arginine residue that helps orientate polyP to the catalytic site [75]. Likewise, the Ddp1 structure revealed important arginine, histidine, and lysine residues involved in polyP binding [74]. Thus, structural studies may prove to be an invaluable asset in the pursuit of understanding how polyP interacts with proteins in their native folds and how it might impart its functions on proteins in vivo. Combining data from these studies with recent advances in artificial intelligence algorithms may allow for improved in silico determination of polyP binding to proteins. Molecular docking has already been used to predict polyP-binding domains of PPK related proteins [77], the ACE2 receptor [2], and the viral RNA polymerase RdRp [2], and a similar space filling model has been used to predict the polyP binding site of the SARS-CoV-2 spike protein [66]. However, only the ACE2-polyP interaction has been validated in vitro [2]. Importantly, it is possible that modes of interaction in vivo differ from those in vitro, and so stringent validation of these in silico studies is warranted.

Bridging the gap from in vitro binding to in vivo function

Despite the growing list of polyP-binding proteins, there is disproportionately less evidence supporting functions for polyP binding in vivo. When a polyP-binding protein is uncovered the most obvious next step is to mutate the residues required for binding and assess phenotypic changes. An important consideration is the type of mutation made. As discussed above, lysine or histidine to arginine mutations seem to prevent interaction of PASK and polyHis proteins with polyP as judged by NuPAGE analysis, but whether this would actually prevent interaction in vivo is less clear. However, more dramatic mutations, lysine to alanine for example, would be more likely to impact target stability and function, independently of polyP binding. How can we begin to test if the observed effects are really due to the loss of polyP interaction? We think the most convincing experiments would be genetic in nature. First, we would expect to see similar phenotypes for cells expressing mutant versions of the target proteins that cannot bind to polyP, and cells mutated for polyP synthetase enzymes (vtc4Δ in yeast, for example). Critically, the mutants should behave epistatically, meaning that the phenotype of the double mutant should not be more dramatic than that of the stronger single mutant. Of course, this represents an ideal scenario, and complex genetic interactions could occur if polyP impacts multiple proteins (directly or indirectly) feeding into the same biological readout. This type of genetic test is also impossible if the identity of the enzymes required for polyP synthesis are unknown [23]. In this scenario, overexpression of polyphosphatases could be used to artificially deplete polyP levels. Choice of model system also bears consideration when studying the fundamentals of polyP-protein interactions in vivo. In this regard, bacteria present themselves as an ideal model. First, the enzymes that produce and degrade polyP in bacteria are well characterized and the level of subcellular compartmentalization is greatly reduced compared with other species like budding yeast, where the majority of polyP is sequestered in the vacuole [13]. Second, the accumulation of polyP can easily be regulated through environmental stimulus (e.g. a switch from nutrient rich to minimal media), allowing for temporal control of polyP accumulation [78].

Regulation by PTMs

Another important consideration that has not been explored in detail is the potential interplay between PTMs and polyP binding. Direct modification of lysine residues, by acetylation or methylation for example [30,79], could prevent polyP from interacting with these residues. Azevedo et al. [26] also previously suggested the possibility of interplay between polyP binding and covalent pyrophosphorylation of eukaryotic PASK containing proteins. Pyrophosphorylation involves the magnesium dependent non-enzymatic transfer of the β-phosphoryl group from inositol pyrophosphate to a pre-phosphorylated residue [80]. This modification occurs primarily on serine residues within intrinsically disordered regions [81]. Indeed, recent work suggests that pyrophosphorylation occurs on a global scale and many of the proteins modified are localized to the nucleolus [81], as is the case with polyP-binding proteins [26,31]. It is important to note that polyP binding is not strictly dependent on pyrophosphorylation since Nsr1 extracted from yeast cells with no diphosphoinositol pentakisphosphate, the phosphate donor for pyrophosphorylation, can bind polyP [26], and proteins purified from E. coli can also bind polyP in vitro [34]. Nonetheless, we postulate that pyrophosphorylation might regulate polyP binding in vivo under specific conditions. As a first step, proteins known to bind polyP that are also pyrophosphorylated, such as Nsr1 [26,80] or DEK [31,81], could be used to test whether this modification can alter the extent of polyP binding in vitro. Alternatively, polyP binding could precede pyrophosphorylation, either by promoting the recruitment of kinases responsible for priming phosphorylations, or the subsequent reaction with inositol pyrophosphates [80].

Regulation by polyphosphatases

Polyphosphatases are ubiquitous enzymes, from prokaryotes to eukaryotes [17]. The ability of these enzymes to degrade polyP chains makes them candidate regulators that could turn off or dampen signaling cascades initiated by polyP-protein interactions. In vitro work has shown that recombinant yeast Ppx1 can cleave protein-bound polyP [26], and many studies have used exogenous Ppx1 expression to study the effects of polyP in mammalian cells [68,82]. Polyphosphatases were initially proposed to remove the covalent interacting polyP chains from proteins [26,83,84], although there is no reason to assume that the mode of interaction would have any bearing on the ability of these enzymes to act on polyP bound to protein targets. Understanding how these polyphosphatases are regulated could shed light on pathways mediated by polyP-protein interactions. Since most polyphosphatase enzymes are dependent on cations for their function [17], we speculate that local concentrations of these cofactors are important determinants for the reversal of polyP-protein interactions in vivo.

Indirect effects of polyP: cation binding and phase separation

Delineating between direct and indirect effects of polyP binding to proteins is complicated by its highly anionic nature and ability to form electrostatic interactions with molecules other than proteins [85–87]. For example, it is not surprising that enzymes requiring cationic cofactors might be inhibited by polyP that sequesters these in solution. Also, polyP in the presence of monovalent or divalent cations can induce liquid-liquid phase separation [88–90], which could also promote polyP-protein interactions. There is evidence that polyP can interact with the Hfq protein from E. coli and form phase separated condensates in vitro that may be functionally relevant for heterochromatin formation in vivo [69]. Moreover, the positively charged green fluorescent protein (+36GFP) colocalizes with polyP granules during nutrient starvation when expressed in Citrobacter freundii and forms phase separated condensates in the presence of polyP in vitro, although direct polyP binding has not been demonstrated [91]. Whether direct interaction of polyP and the proteins in these phase separated condensates is required for observed effects is deserving of careful consideration. An important control here may be to test if other conditions that promote phase separation yield similar outcomes in the absence of polyP.

The study of polyP-protein interactions has emerged as a hot topic of research with broad implications for diverse areas of biomedical research. In the long term, insights gained from this work could be leveraged to better understand how changes in polyP homeostasis contribute to disease and potentially provide new opportunities for treatments. Success in this regard demands that we move beyond generating lists of polyP-interacting proteins to focus instead on mechanism and in vivo function. At this point we should expect the unexpected – continued re-evaluation of assumptions regarding polyP-protein interactions will be essential to move the field forward.

  • PolyP chains of varying length participate in diverse functions across prokaryotic and eukaryotic cells. Recent work suggests that polyP chains exert their functions in part by binding to specific proteins via motifs (polyacidic, serine, and lysine rich (PASK), polyHistidine, and polyLysine motifs) often found in regions of proteins that are predicted to be disordered. PolyP also interacts with positively charged surfaces on select folded proteins.

  • Recent work has challenged a model that polyP interaction with PASK motifs is covalent in nature. Nevertheless, there remains considerable interest in the function of polyP-protein interactions and the list of polyP binding partners continues to grow.

  • A complete understanding of polyP-protein interactions demands careful consideration of model systems and experimental approaches used to address open questions.

The authors declare that there are no competing interests associated with the manuscript.

Polyphosphate research in the Downey lab is funded by a Canadian Institutes of Health Research Project Grant (PJT-174987) to M.D. L.M. and K.B. were supported by Ontario Graduate Scholarships. Order for co-first authors, who contributed equally to the manuscript, was determined by a coin flip.

CHAD

conserved histidine α-helical

LDH

lactate dehydrogenase

PPK

polyphosphate kinase

PTM

post-translational modification

VWF

von Willebrand Factor

1
Kornberg
,
A.
(
1999
)
Inorganic polyphosphate: a molecule of many functions
.
Prog. Mol. Subcell. Biol.
23
,
1
18
2
Ferrucci
,
V.
,
Kong
,
D.Y.
,
Asadzadeh
,
F.
,
Marrone
,
L.
,
Boccia
,
A.
,
Siciliano
,
R.
et al. (
2021
)
Long-chain polyphosphates impair SARS-CoV-2 infection and replication
.
Sci. Signal.
14
,
eabe5040
3
Arredondo
,
C.
,
Cefaliello
,
C.
,
Dyrda
,
A.
,
Jury
,
N.
,
Martinez
,
P.
,
Diaz
,
I.
et al. (
2022
)
Excessive release of inorganic polyphosphate by ALS/FTD astrocytes causes non-cell-autonomous toxicity to motoneurons
.
Neuron
110
,
1656
1670 e1612
4
Akiyama
,
M.
,
Crooke
,
E.
and
Kornberg
,
A.
(
1992
)
The polyphosphate kinase gene of Escherichia coli. Isolation and sequence of the ppk gene and membrane location of the protein
.
J. Biol. Chem.
267
,
22556
22561
5
Zhang
,
H.
,
Ishige
,
K.
and
Kornberg
,
A.
(
2002
)
A polyphosphate kinase (PPK2) widely conserved in bacteria
.
Proc. Natl Acad. Sci. U.S.A.
99
,
16678
16683
6
Neville
,
N.
,
Roberge
,
N.
,
Ji
,
X.
,
Stephen
,
P.
,
Lu
,
J.L.
and
Jia
,
Z.
(
2021
)
A dual-specificity inhibitor targets polyphosphate kinase 1 and 2 enzymes to attenuate virulence of Pseudomonas aeruginosa
.
mBio
12
,
e0059221
7
Akiyama
,
M.
,
Crooke
,
E.
and
Kornberg
,
A.
(
1993
)
An exopolyphosphatase of Escherichia coli. The enzyme and its ppx gene in a polyphosphate operon
.
J. Biol. Chem.
268
,
633
639
8
Chuang
,
Y.M.
,
Bandyopadhyay
,
N.
,
Rifat
,
D.
,
Rubin
,
H.
,
Bader
,
J.S.
and
Karakousis
,
P.C.
(
2015
)
Deficiency of the novel exopolyphosphatase Rv1026/PPX2 leads to metabolic downshift and altered cell wall permeability in Mycobacterium tuberculosis
.
mBio
6
,
e02428
9
Lindner
,
S.N.
,
Knebel
,
S.
,
Wesseling
,
H.
,
Schoberth
,
S.M.
and
Wendisch
,
V.F.
(
2009
)
Exopolyphosphatases PPX1 and PPX2 from Corynebacterium glutamicum
.
Appl. Environ. Microbiol.
75
,
3161
3170
10
Gerasimaite
,
R.
and
Mayer
,
A.
(
2017
)
Ppn2, a novel Zn(2+)-dependent polyphosphatase in the acidocalcisome-like yeast vacuole
.
J. Cell Sci.
130
,
1625
1636
11
Auesukaree
,
C.
,
Homma
,
T.
,
Tochio
,
H.
,
Shirakawa
,
M.
,
Kaneko
,
Y.
and
Harashima
,
S.
(
2004
)
Intracellular phosphate serves as a signal for the regulation of the PHO pathway in Saccharomyces cerevisiae
.
J. Biol. Chem.
279
,
17289
17294
12
Gerasimaite
,
R.
,
Sharma
,
S.
,
Desfougeres
,
Y.
,
Schmidt
,
A.
and
Mayer
,
A.
(
2014
)
Coupled synthesis and translocation restrains polyphosphate to acidocalcisome-like vacuoles and prevents its toxicity
.
J. Cell Sci.
127
,
5093
5104
13
Denoncourt
,
A.
and
Downey
,
M.
(
2021
)
Model systems for studying polyphosphate biology: a focus on microorganisms
.
Curr. Genet.
67
,
331
346
14
McCarthy
,
L.
,
Abramchuk
,
I.
,
Wafy
,
G.
,
Denoncourt
,
A.
,
Lavallee-Adam
,
M.
and
Downey
,
M.
(
2022
)
Ddp1 cooperates with Ppx1 to counter a stress response initiated by nonvacuolar polyphosphate
.
mBio
13
,
e0039022
15
Gerasimaite
,
R.
and
Mayer
,
A.
(
2016
)
Enzymes of yeast polyphosphate metabolism: structure, enzymology and biological roles
.
Biochem. Soc. Trans.
44
,
234
239
16
Yang
,
W.
(
2011
)
Nucleases: diversity of structure, function and mechanism
.
Q. Rev. Biophys.
44
,
1
93
17
McCarthy
,
L.
and
Downey
,
M.
(
2023
)
The emerging landscape of eukaryotic polyphosphatases
.
FEBS Lett.
597
,
1447
1461
18
Sethuraman
,
A.
,
Rao
,
N.N.
and
Kornberg
,
A.
(
2001
)
The endopolyphosphatase gene: essential in Saccharomyces cerevisiae
.
Proc. Natl Acad. Sci. U.S.A.
98
,
8542
8547
19
Kizawa
,
K.
,
Aono
,
T.
and
Ohtomo
,
R.
(
2017
)
PHO8 gene coding alkaline phosphatase of Saccharomyces cerevisiae is involved in polyphosphate metabolism
.
J. Gen. Appl. Microbiol.
62
,
297
302
20
Kumble
,
K.D.
and
Kornberg
,
A.
(
1995
)
Inorganic polyphosphate in mammalian cells and tissues
.
J. Biol. Chem.
270
,
5818
5822
21
Desfougeres
,
Y.
,
Saiardi
,
A.
and
Azevedo
,
C.
(
2020
)
Inorganic polyphosphate in mammals: where's Wally?
Biochem. Soc. Trans.
48
,
95
101
22
Baev
,
A.Y.
,
Angelova
,
P.R.
and
Abramov
,
A.Y.
(
2020
)
Inorganic polyphosphate is produced and hydrolyzed in F0F1-ATP synthase of mammalian mitochondria
.
Biochem. J.
477
,
1515
1524
23
Baijal
,
K.
and
Downey
,
M.
(
2021
)
The promises of lysine polyphosphorylation as a regulatory modification in mammals are tempered by conceptual and technical challenges
.
Bioessays
43
,
e2100058
24
Samper-Martin
,
B.
,
Sarrias
,
A.
,
Lazaro
,
B.
,
Perez-Montero
,
M.
,
Rodriguez-Rodriguez
,
R.
,
Ribeiro
,
M.P.C.
et al. (
2021
)
Polyphosphate degradation by Nudt3-Zn(2+) mediates oxidative stress response
.
Cell Rep.
37
,
110004
25
Schoeppe
,
R.
,
Waldmann
,
M.
,
Jessen
,
H.J.
and
Renne
,
T.
(
2024
)
An update on polyphosphate in vivo activities
.
Biomolecules
14
,
937
26
Azevedo
,
C.
,
Livermore
,
T.
and
Saiardi
,
A.
(
2015
)
Protein polyphosphorylation of lysine residues by inorganic polyphosphate
.
Mol. Cell
58
,
71
82
27
Kuroda
,
A.
,
Nomura
,
K.
,
Ohtomo
,
R.
,
Kato
,
J.
,
Ikeda
,
T.
,
Takiguchi
,
N.
et al. (
2001
)
Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli
.
Science
293
,
705
708
28
Travers
,
R.J.
,
Smith
,
S.A.
and
Morrissey
,
J.H.
(
2015
)
Polyphosphate, platelets, and coagulation
.
Int. J. Lab. Hematol.
37
,
31
35
29
Lorenzo-Orts
,
L.
,
Hohmann
,
U.
,
Zhu
,
J.
and
Hothorn
,
M.
(
2019
)
Molecular characterization of CHAD domains as inorganic polyphosphate-binding modules
.
Life Sci. Alliance
2
,
e201900385
30
Azevedo
,
C.
and
Saiardi
,
A.
(
2016
)
Why always lysine? The ongoing tale of one of the most modified amino acids
.
Adv. Biol. Regul.
60
,
144
150
31
Bentley-DeSousa
,
A.
,
Holinier
,
C.
,
Moteshareie
,
H.
,
Tseng
,
Y.C.
,
Kajjo
,
S.
,
Nwosu
,
C.
et al. (
2018
)
A screen for candidate targets of lysine polyphosphorylation uncovers a conserved network implicated in ribosome biogenesis
.
Cell Rep.
22
,
3427
3439
32
Azevedo
,
C.
,
Singh
,
J.
,
Steck
,
N.
,
Hofer
,
A.
,
Ruiz
,
F.A.
,
Singh
,
T.
et al. (
2018
)
Screening a protein array with synthetic biotinylated inorganic polyphosphate to define the human polyP-ome
.
ACS Chem. Biol.
13
,
1958
1963
33
Neville
,
N.
,
Lehotsky
,
K.
,
Yang
,
Z.
,
Klupt
,
K.A.
,
Denoncourt
,
A.
,
Downey
,
M.
et al. (
2023
)
Modification of histidine repeat proteins by inorganic polyphosphate
.
Cell Rep.
42
,
113082
34
Neville
,
N.
,
Lehotsky
,
K.
,
Klupt
,
K.A.
,
Downey
,
M.
and
Jia
,
Z.
(
2024
)
Polyphosphate attachment to lysine repeats is a non-covalent protein modification
.
Mol. Cell
84
,
1802
1810 e1804
35
Azevedo
,
C.
,
Borghi
,
F.
,
Su
,
X.B.
and
Saiardi
,
A.
(
2024
)
On the covalent nature of lysine polyphosphorylation
.
Mol. Cell
84
,
1811
1815 e1813
36
Neville
,
N.
,
Lehotsky
,
K.
and
Jia
,
Z.
(
2024
)
Back on the chain gang: polyphosphate modification of proteins
.
Trends Biochem. Sci.
49
,
757
760
37
Gray
,
M.J.
,
Wholey
,
W.Y.
,
Wagner
,
N.O.
,
Cremers
,
C.M.
,
Mueller-Schickert
,
A.
,
Hock
,
N.T.
et al. (
2014
)
Polyphosphate is a primordial chaperone
.
Mol. Cell
53
,
689
699
38
Yoo
,
N.G.
,
Dogra
,
S.
,
Meinen
,
B.A.
,
Tse
,
E.
,
Haefliger
,
J.
,
Southworth
,
D.R.
et al. (
2018
)
Polyphosphate stabilizes protein unfolding intermediates as soluble amyloid-like oligomers
.
J. Mol. Biol.
430
,
4195
4208
39
Sasahara
,
K.
,
Yamaguchi
,
K.
,
So
,
M.
and
Goto
,
Y.
(
2019
)
Polyphosphates diminish solubility of a globular protein and thereby promote amyloid aggregation
.
J. Biol. Chem.
294
,
15318
15329
40
Guan
,
J.
and
Jakob
,
U.
(
2024
)
The protein scaffolding functions of polyphosphate
.
J. Mol. Biol.
436
,
168504
41
Cremers
,
C.M.
,
Knoefler
,
D.
,
Gates
,
S.
,
Martin
,
N.
,
Dahl
,
J.U.
,
Lempart
,
J.
et al. (
2016
)
Polyphosphate: a conserved modifier of amyloidogenic processes
.
Mol. Cell
63
,
768
780
42
Wickramasinghe
,
S.P.
,
Lempart
,
J.
,
Merens
,
H.E.
,
Murphy
,
J.
,
Huettemann
,
P.
,
Jakob
,
U.
et al. (
2019
)
Polyphosphate initiates tau aggregation through intra- and intermolecular scaffolding
.
Biophys. J.
117
,
717
728
43
Zhang
,
C.M.
,
Yamaguchi
,
K.
,
So
,
M.
,
Sasahara
,
K.
,
Ito
,
T.
,
Yamamoto
,
S.
et al. (
2019
)
Possible mechanisms of polyphosphate-induced amyloid fibril formation of beta(2)-microglobulin
.
Proc. Natl Acad. Sci. U.S.A.
116
,
12833
12838
44
Huettemann
,
P.
,
Mahadevan
,
P.
,
Lempart
,
J.
,
Tse
,
E.
,
Dehury
,
B.
,
Edwards
,
B.F.P.
et al. (
2024
)
Amyloid accelerator polyphosphate fits as the mystery density in alpha-synuclein fibrils
.
PLoS Biol.
22
,
e3002650
45
Kuroda
,
A.
(
2006
)
A polyphosphate-lon protease complex in the adaptation of Escherichia coli to amino acid starvation
.
Biosci. Biotechnol. Biochem.
70
,
325
331
46
Kuroda
,
A.
,
Nomura
,
K.
,
Takiguchi
,
N.
,
Kato
,
J.
and
Ohtake
,
H.
(
2006
)
Inorganic polyphosphate stimulates lon-mediated proteolysis of nucleoid proteins in Escherichia coli
.
Cell. Mol. Biol.
52
,
23
29
47
Gross
,
M.H.
and
Konieczny
,
I.
(
2020
)
Polyphosphate induces the proteolysis of ADP-bound fraction of initiator to inhibit DNA replication initiation upon stress in Escherichia coli
.
Nucleic Acids Res.
48
,
5457
5466
48
Ropelewska
,
M.
,
Gross
,
M.H.
and
Konieczny
,
I.
(
2020
)
DNA and polyphosphate in directed proteolysis for DNA replication control
.
Front. Microbiol.
11
,
585717
49
Wrench
,
A.P.
,
Gardner
,
C.L.
,
Siegel
,
S.D.
,
Pagliai
,
F.A.
,
Malekiha
,
M.
,
Gonzalez
,
C.F.
et al. (
2013
)
Mgla/SspA complex interactions are modulated by inorganic polyphosphate
.
PLoS One
8
,
e76428
50
Choi
,
S.H.
,
Smith
,
S.A.
and
Morrissey
,
J.H.
(
2015
)
Polyphosphate accelerates factor V activation by factor XIa
.
Thromb. Haemost.
113
,
599
604
51
Choi
,
S.H.
,
Smith
,
S.A.
and
Morrissey
,
J.H.
(
2011
)
Polyphosphate is a cofactor for the activation of factor XI by thrombin
.
Blood
118
,
6963
6970
52
Malik
,
R.A.
,
Zhou
,
J.
,
Fredenburgh
,
J.C.
,
Truong
,
T.K.
,
Crosby
,
J.R.
,
Revenko
,
A.S.
et al. (
2021
)
Polyphosphate-induced thrombosis in mice is factor XII dependent and is attenuated by histidine-rich glycoprotein
.
Blood Adv.
5
,
3540
3551
53
Mutch
,
N.J.
,
Myles
,
T.
,
Leung
,
L.L.
and
Morrissey
,
J.H.
(
2010
)
Polyphosphate binds with high affinity to exosite II of thrombin
.
J. Thromb. Haemost.
8
,
548
555
54
Montilla
,
M.
,
Atienza-Navarro
,
I.
,
Garcia-Cozar
,
F.J.
,
Castro
,
C.
,
Rodriguez-Martorell
,
F.J.
and
Ruiz
,
F.A.
(
2022
)
Polyphosphate activates von Willebrand factor interaction with glycoprotein Ib in the absence of factor VIII in vitro
.
Int. J. Mol. Sci.
23
,
14118
55
Montilla
,
M.
,
Hernandez-Ruiz
,
L.
,
Garcia-Cozar
,
F.J.
,
Alvarez-Laderas
,
I.
,
Rodriguez-Martorell
,
J.
and
Ruiz
,
F.A.
(
2012
)
Polyphosphate binds to human von Willebrand factor in vivo and modulates its interaction with glycoprotein Ib
.
J. Thromb. Haemost.
10
,
2315
2323
56
Martinez de Lagran
,
M.
,
Benavides-Piccione
,
R.
,
Ballesteros-Yanez
,
I.
,
Calvo
,
M.
,
Morales
,
M.
,
Fillat
,
C.
et al. (
2012
)
Dyrk1a influences neuronal morphogenesis through regulation of cytoskeletal dynamics in mammalian cortical neurons
.
Cereb. Cortex
22
,
2867
2877
57
Dirice
,
E.
,
Walpita
,
D.
,
Vetere
,
A.
,
Meier
,
B.C.
,
Kahraman
,
S.
,
Hu
,
J.
et al. (
2016
)
Inhibition of DYRK1A stimulates human beta-cell proliferation
.
Diabetes
65
,
1660
1671
58
Rammohan
,
M.
,
Harris
,
E.
,
Bhansali
,
R.S.
,
Zhao
,
E.
,
Li
,
L.S.
and
Crispino
,
J.D.
(
2022
)
The chromosome 21 kinase DYRK1A: emerging roles in cancer biology and potential as a therapeutic target
.
Oncogene
41
,
2003
2011
59
Barzowska
,
A.
,
Pucelik
,
B.
,
Pustelny
,
K.
,
Matsuda
,
A.
,
Martyniak
,
A.
,
Stepniewski
,
J.
et al. (
2021
)
DYRK1A kinase inhibitors promote beta-cell survival and insulin homeostasis
.
Cells
10
,
2263
60
Lu
,
J.
,
Wu
,
T.
,
Zhang
,
B.
,
Liu
,
S.
,
Song
,
W.
,
Qiao
,
J.
et al. (
2021
)
Types of nuclear localization signals and mechanisms of protein import into the nucleus
.
Cell Commun. Signal.
19
,
60
61
Roewe
,
J.
,
Stavrides
,
G.
,
Strueve
,
M.
,
Sharma
,
A.
,
Marini
,
F.
,
Mann
,
A.
et al. (
2020
)
Bacterial polyphosphates interfere with the innate host defense to infection
.
Nat. Commun.
11
,
4035
62
Ruiz
,
F.A.
,
Lea
,
C.R.
,
Oldfield
,
E.
and
Docampo
,
R.
(
2004
)
Human platelet dense granules contain polyphosphate and are similar to acidocalcisomes of bacteria and unicellular eukaryotes
.
J. Biol. Chem.
279
,
44250
44257
63
Hassanian
,
S.M.
,
Dinarvand
,
P.
,
Smith
,
S.A.
and
Rezaie
,
A.R.
(
2015
)
Inorganic polyphosphate elicits pro-inflammatory responses through activation of the mammalian target of rapamycin complexes 1 and 2 in vascular endothelial cells
.
J. Thromb. Haemost.
13
,
860
871
64
Dinarvand
,
P.
,
Hassanian
,
S.M.
,
Qureshi
,
S.H.
,
Manithody
,
C.
,
Eissenberg
,
J.C.
,
Yang
,
L.
et al. (
2014
)
Polyphosphate amplifies proinflammatory responses of nuclear proteins through interaction with receptor for advanced glycation end products and P2Y1 purinergic receptor
.
Blood
123
,
935
945
65
Holmstrom
,
K.M.
,
Marina
,
N.
,
Baev
,
A.Y.
,
Wood
,
N.W.
,
Gourine
,
A.V.
and
Abramov
,
A.Y.
(
2013
)
Signalling properties of inorganic polyphosphate in the mammalian brain
.
Nat. Commun.
4
,
1362
66
Neufurth
,
M.
,
Wang
,
X.
,
Tolba
,
E.
,
Lieberwirth
,
I.
,
Wang
,
S.
,
Schroder
,
H.C.
et al. (
2020
)
The inorganic polymer, polyphosphate, blocks binding of SARS-CoV-2 spike protein to ACE2 receptor at physiological concentrations
.
Biochem. Pharmacol.
182
,
114215
67
Mantilla
,
B.S.
,
Azevedo
,
C.
,
Denny
,
P.W.
,
Saiardi
,
A.
and
Docampo
,
R.
(
2021
)
The histidine ammonia lyase of Trypanosoma cruzi is involved in acidocalcisome alkalinization and is essential for survival under starvation conditions
.
mBio
12
,
e0198121
68
Wang
,
L.
,
Fraley
,
C.D.
,
Faridi
,
J.
,
Kornberg
,
A.
and
Roth
,
R.A.
(
2003
)
Inorganic polyphosphate stimulates mammalian TOR, a kinase involved in the proliferation of mammary cancer cells
.
Proc. Natl Acad. Sci. U.S.A.
100
,
11249
11254
69
Beaufay
,
F.
,
Amemiya
,
H.M.
,
Guan
,
J.
,
Basalla
,
J.
,
Meinen
,
B.A.
,
Chen
,
Z.
et al. (
2021
)
Polyphosphate drives bacterial heterochromatin formation
.
Sci. Adv.
7
,
eabk0233
70
Yang
,
Z.X.
,
Zhou
,
Y.N.
,
Yang
,
Y.
and
Jin
,
D.J.
(
2010
)
Polyphosphate binds to the principal sigma factor of RNA polymerase during starvation response in Helicobacter pylori
.
Mol. Microbiol.
77
,
618
627
71
Ito
,
T.
,
Yamamoto
,
S.
,
Yamaguchi
,
K.
,
Sato
,
M.
,
Kaneko
,
Y.
,
Goto
,
S.
et al (
2020
)
Inorganic polyphosphate potentiates lipopolysaccharide-induced macrophage inflammatory response
.
J Biol Chem.
295
,
4014
4023
72
Gauthier
,
C.M.
,
LeGallais
,
J.
,
Savic
,
N.
,
Moradi-Fard
,
S.
,
Grew
,
A.
,
Loe
,
M.
et al. (
2024
)
Intrinsic disorder of a nucleoplasmin-like histone chaperone specifies its discrete nuclear and nucleolar functions
.
FEBS Lett.
598
,
187
198
73
Azevedo
,
C.
,
Desfougeres
,
Y.
,
Jiramongkol
,
Y.
,
Partington
,
H.
,
Trakansuebkul
,
S.
,
Singh
,
J.
et al. (
2019
)
Development of a yeast model to study the contribution of vacuolar polyphosphate metabolism to lysine polyphosphorylation
.
J. Biol. Chem.
295
,
1439
1451
74
Marquez-Monino
,
M.A.
,
Ortega-Garcia
,
R.
,
Shipton
,
M.L.
,
Franco-Echevarria
,
E.
,
Riley
,
A.M.
,
Sanz-Aparicio
,
J.
et al. (
2021
)
Multiple substrate recognition by yeast diadenosine and diphosphoinositol polyphosphate phosphohydrolase through phosphate clamping
.
Sci. Adv.
7
,
eabf6744
75
Parnell
,
A.E.
,
Mordhorst
,
S.
,
Kemper
,
F.
,
Giurrandino
,
M.
,
Prince
,
J.P.
,
Schwarzer
,
N.J.
et al. (
2018
)
Substrate recognition and mechanism revealed by ligand-bound polyphosphate kinase 2 structures
.
Proc. Natl Acad. Sci. U.S.A.
115
,
3350
3355
76
Wijeyewickrema
,
L.C.
,
Lameignere
,
E.
,
Hor
,
L.
,
Duncan
,
R.C.
,
Shiba
,
T.
,
Travers
,
R.J.
et al. (
2016
)
Polyphosphate is a novel cofactor for regulation of complement by a serpin, C1 inhibitor
.
Blood
128
,
1766
1776
77
Wei
,
Z.Z.
,
Vatcher
,
G.
,
Tin
,
A.H.
,
Teng
,
J.L.
,
Wang
,
J.
,
Cui
,
Q.H.
et al. (
2015
)
Positively-charged semi-tunnel is a structural and surface characteristic of polyphosphate-binding proteins: an in-silico study
.
PLoS One
10
,
e0123713
78
Kuroda
,
A.
,
Tanaka
,
S.
,
Ikeda
,
T.
,
Kato
,
J.
,
Takiguchi
,
N.
and
Ohtake
,
H.
(
1999
)
Inorganic polyphosphate kinase is required to stimulate protein degradation and for adaptation to amino acid starvation in Escherichia coli
.
Proc. Natl Acad. Sci. U.S.A.
96
,
14264
14269
79
Wang
,
Z.A.
and
Cole
,
P.A.
(
2020
)
The chemical biology of reversible lysine post-translational modifications
.
Cell Chem. Biol.
27
,
953
969
80
Bhandari
,
R.
,
Saiardi
,
A.
,
Ahmadibeni
,
Y.
,
Snowman
,
A.M.
,
Resnick
,
A.C.
,
Kristiansen
,
T.Z.
et al. (
2007
)
Protein pyrophosphorylation by inositol pyrophosphates is a posttranslational event
.
Proc. Natl Acad. Sci. U.S.A.
104
,
15305
15310
81
Morgan
,
J.A.M.
,
Singh
,
A.
,
Kurz
,
L.
,
Nadler-Holly
,
M.
,
Ruwolt
,
M.
,
Ganguli
,
S.
et al. (
2024
)
Extensive protein pyrophosphorylation revealed in human cell lines
.
Nat. Chem. Biol.
20
,
1305
1316
82
Abramov
,
A.Y.
,
Fraley
,
C.
,
Diao
,
C.T.
,
Winkfein
,
R.
,
Colicos
,
M.A.
,
Duchen
,
M.R.
et al. (
2007
)
Targeted polyphosphatase expression alters mitochondrial metabolism and inhibits calcium-dependent cell death
.
Proc. Natl Acad. Sci. U.S.A.
104
,
18091
18096
83
Bentley-DeSousa
,
A.
and
Downey
,
M.
(
2019
)
From underlying chemistry to therapeutic potential: open questions in the new field of lysine polyphosphorylation
.
Curr. Genet.
65
,
57
64
84
Azevedo
,
C.
and
Saiardi
,
A.
(
2016
)
The new world of inorganic polyphosphates
.
Biochem. Soc. Trans.
44
,
13
17
85
Cramer
,
C.L.
and
Davis
,
R.H.
(
1984
)
Polyphosphate-cation interaction in the amino acid-containing vacuole of Neurospora crassa
.
J. Biol. Chem.
259
,
5152
5157
86
Toso
,
D.B.
,
Henstra
,
A.M.
,
Gunsalus
,
R.P.
and
Zhou
,
Z.H.
(
2011
)
Structural, mass and elemental analyses of storage granules in methanogenic archaeal cells
.
Environ. Microbiol.
13
,
2587
2599
87
Onaka
,
T.
,
Miyajima
,
T.
and
Ohashi
,
S.
(
1981
)
On the binding of magnesium to long-chain polyphosphate ions
.
J. Inorg. Nucl. Chem.
43
,
3323
3327
88
Chawla
,
R.
,
Tom
,
J.K.A.
,
Boyd
,
T.
,
Tu
,
N.H.
,
Bai
,
T.
,
Grotjahn
,
D.A.
et al. (
2024
)
Reentrant DNA shells tune polyphosphate condensate size
.
Nat. Commun.
15
,
9258
89
Borghi
,
F.
,
Azevedo
,
C.
,
Johnson
,
E.
,
Burden
,
J.J.
and
Saiardi
,
A.
(
2024
)
A mammalian model reveals inorganic polyphosphate channeling into the nucleolus and induction of a hyper-condensate state
.
Cell Rep. Methods
4
,
100814
90
Furuki
,
T.
,
Togo
,
A.
,
Usuda
,
H.
,
Nobeyama
,
T.
,
Hirano
,
A.
and
Shiraki
,
K.
(
2024
)
Monovalent Ion Effect on Liquid-Liquid Phase Separation of Aqueous Polyphosphate-Salt Mixtures
.
J Phys Chem B.
128
,
11435
11440
91
Wang
,
X.
,
Shi
,
C.
,
Mo
,
J.
,
Xu
,
Y.
,
Wei
,
W.
and
Zhao
,
J.
(
2020
)
An inorganic biopolymer polyphosphate controls positively charged protein phase transitions
.
Angew. Chem. Int. Ed. Engl.
59
,
2679
2683

Author notes

*

These authors contributed equally to this work.

This is an open access article published by Portland Press Limited on behalf of the Biochemical Society and distributed under the Creative Commons Attribution License 4.0 (CC BY-NC-ND).

Supplementary data