Polyphosphate Discriminates Protein Conformational Ensembles More Efficiently than DNA Promoting Diverse Assembly and Maturation Behaviors

Reviewer #1 (Public review):

Summary:

In the article titled "Polyphosphate discriminates protein conformational ensembles more efficiently than DNA promoting diverse assembly and maturation behaviors," Goyal and colleagues investigate the role of negatively charged biopolymers, i.e., polyphosphate (polyP) and DNA, play in phase separation of cytidine repressor (CytR) and fructose repressor (FruR). The authors find that both negative polymers drive the formation of metastable protein/polymer condensates. However, polyP-driven condensates form more gel- or solid-like structures over time while DNA-driven condensates tend to dissipate over time. The authors link this disparate condensate behavior to polyP-induced structures within the enzymes. Specifically, they observe the formation of polyproline II-like structures within two tested enzyme variants in the presence of polyP. Together their results provide a unique insight into the physical and structural mechanism by which two unique negatively charged polymers can induce distinct phase transitions with the same protein. This study will be a welcomed addition to the condensate field and provide new molecular insights into how binding partner-induced structural changes within a given protein can affect the mesoscale behavior of condensates. The concerns outlined below are meant to strengthen the manuscript.

Strengths:

Throughout the article, the authors used the correct techniques to probe physical changes within proteins that can be directly linked to phase transition behaviors. Their rigorous experiments create a clear picture of what occurs at the molecular level with CytR and FruR are exposed to either DNA or polyP, which are unique, highly negatively charged biopolymers found within bacteria. This work provides a new view of mechanisms by which bacteria can regulate the cytoplasmic organization upon the induction of stress. Furthermore, this is likely applicable to mammalian and plant cells and likely to numerous proteins that undergo condensation with nucleic acids and other charged biopolymers.

Weaknesses:

The biggest weakness of this study is that compares the phase behavior of enzymes driven by negatively charged polymers that have intrinsic differences in net charge and charge density. Because these properties are extremely important for controlling phase separation, any differences may result in the observed phase transitions driven by DNA and polyP. The authors should perform an additional experiment to control for these differences as best they can. The results from these experiments will provide additional insight into the importance of charge-based properties for controlling phase transitions.

Reviewer #2 (Public review):

Summary:

In this study, Goyal et al demonstrate that the assembly of proteins with polyphosphate into either condensates or aggregates can reveal information on the initial protein ensemble. They show that, unlike DNA, polyphosphate is able to effectively discriminate against initial protein ensembles with different conformational heterogeneity, structure, and compactness. The authors further show that the protein native ensemble is vital on whether polyphosphate induces phase separation or aggregation, whereas DNA induces a similar outcome regardless of the initial protein ensemble. This work provides a way to improve our mechanistic understanding of how conformational transitions of proteins may regulate or drive LLPS condensate and aggregate assemblies within biological systems.

Strengths:

This is a thoroughly conducted study that provides an alternative route for inducing phase separation that is more informative on the initial protein ensemble involved. This is particularly useful and a complementary means to investigate the role played by protein dynamics and plasticity in phase transitions. The authors use an appropriate set of techniques to investigate unique phase transitions within proteins induced by polyphosphates. An alternative protein system is used to corroborate their findings that the unique assemblies induced by polyphosphates when compared to DNA are not restricted to a single system. The work here is well-documented, easy to interpret, and of relevance for the condensate community.

Weaknesses:

The major weakness of this manuscript is that it is unclear if the information on the initial protein conformational ensemble can be determined solely from the assembly and maturation behavior and the discrimination abilities of polyphosphates. In both systems studied (CytR and FruR), polyphosphate discriminates and results in unique assemblies and maturation behaviors based on the initial protein ensemble. However, it seems the assembly and maturation behavior are not a direct result of the degree of conformational dynamics and plasticity in the initial protein. In the case of CytR, the fully-folded system forms condensates that resolubilize, while the highly disordered state immediately aggregates. Whereas, in the case of FruR, the folded state induces spontaneous aggregation, and the more dynamic, molten globular, system results in short-lived condensates. These results seem to suggest the polyphosphates' ability to discriminate between the initial protein ensemble may not be able to reveal what that initial protein ensemble is unless it is already known.

Author response:

eLife Assessment

This manuscript offers important insights into how polyphosphate (polyP) influences protein phase separation differently from DNA. The authors present compelling evidence that polyP distinguishes between protein conformational states, leading to diverse condensate behaviors. However, differences in charge density between polyP and DNA complicate direct comparisons, and the extent to which polyP-driven phase transitions reveal initial protein states remains unclear. Addressing these concerns would strengthen the manuscript's impact for researchers interested in biomolecular condensates, protein dynamics, and stress response mechanisms.

We thank the editorial team for the favorable assessment. We, however, contend the specific point on the difference in charge density. We have already performed experiments wherein a higher concentration of DNA is used to match the overall ‘concentration of charges’ as in the experiments with polyP (see Figure S6), and we do not identify or observe any differences in the maturation behavior with DNA, i.e. we see only dissolution at both higher and lower concentrations of DNA. Charge density (i.e. the number of charges per unit volume of the polymer), on the other hand, is an intrinsic feature of the polymer which is naturally different between DNA and polyP. In fact, the primary result of our work is our observation that polyP can discern the starting ensembles more efficiently, likely through actively engaging and interacting with the ensemble while DNA appears to be a passive player.

Reviewer #1 (Public review):

Summary:

In the article titled "Polyphosphate discriminates protein conformational ensembles more efficiently than DNA promoting diverse assembly and maturation behaviors," Goyal and colleagues investigate the role of negatively charged biopolymers, i.e., polyphosphate (polyP) and DNA, play in phase separation of cytidine repressor (CytR) and fructose repressor (FruR). The authors find that both negative polymers drive the formation of metastable protein/polymer condensates. However, polyPdriven condensates form more gel- or solid-like structures over time while DNA-driven condensates tend to dissipate over time. The authors link this disparate condensate behavior to polyP-induced structures within the enzymes. Specifically, they observe the formation of polyproline II-like structures within two tested enzyme variants in the presence of polyP. Together their results provide a unique insight into the physical and structural mechanism by which two unique negatively charged polymers can induce distinct phase transitions with the same protein. This study will be a welcomed addition to the condensate field and provide new molecular insights into how binding partner-induced structural changes within a given protein can affect the mesoscale behavior of condensates. The concerns outlined below are meant to strengthen the manuscript.

Strengths:

Throughout the article, the authors used the correct techniques to probe physical changes within proteins that can be directly linked to phase transition behaviors. Their rigorous experiments create a clear picture of what occurs at the molecular level with CytR and FruR are exposed to either DNA or polyP, which are unique, highly negatively charged biopolymers found within bacteria. This work provides a new view of mechanisms by which bacteria can regulate the cytoplasmic organization upon the induction of stress. Furthermore, this is likely applicable to mammalian and plant cells and likely to numerous proteins that undergo condensation with nucleic acids and other charged biopolymers.

Weaknesses:

The biggest weakness of this study is that compares the phase behavior of enzymes driven by negatively charged polymers that have intrinsic differences in net charge and charge density. Because these properties are extremely important for controlling phase separation, any differences may result in the observed phase transitions driven by DNA and polyP. The authors should perform an additional experiment to control for these differences as best they can. The results from these experiments will provide additional insight into the importance of charge-based properties for controlling phase transitions.

We thank the reviewer for providing a positive review of our work. On the comment related to the final paragraph, we note that we have already conducted an experiment with a higher DNA concentration (11.24 µM) to explore if the concentration of charges plays any significant role. The results of this experiment are presented in Figure S6. We observe that even at a higher DNA concentration, the condensates dissolve over time. Therefore, the difference in the maturation behavior of condensates with varying initial protein ensembles is due to the nature of polyP (likely through its enhanced flexibility).

Reviewer #2 (Public review):

Summary:

In this study, Goyal et al demonstrate that the assembly of proteins with polyphosphate into either condensates or aggregates can reveal information on the initial protein ensemble. They show that, unlike DNA, polyphosphate is able to effectively discriminate against initial protein ensembles with different conformational heterogeneity, structure, and compactness. The authors further show that the protein native ensemble is vital on whether polyphosphate induces phase separation or aggregation, whereas DNA induces a similar outcome regardless of the initial protein ensemble. This work provides a way to improve our mechanistic understanding of how conformational transitions of proteins may regulate or drive LLPS condensate and aggregate assemblies within biological systems.

Strengths:

This is a thoroughly conducted study that provides an alternative route for inducing phase separation that is more informative on the initial protein ensemble involved. This is particularly useful and a complementary means to investigate the role played by protein dynamics and plasticity in phase transitions. The authors use an appropriate set of techniques to investigate unique phase transitions within proteins induced by polyphosphates. An alternative protein system is used to corroborate their findings that the unique assemblies induced by polyphosphates when compared to DNA are not restricted to a single system. The work here is well-documented, easy to interpret, and of relevance for the condensate community.

Weaknesses:

The major weakness of this manuscript is that it is unclear if the information on the initial protein conformational ensemble can be determined solely from the assembly and maturation behavior and the discrimination abilities of polyphosphates. In both systems studied (CytR and FruR), polyphosphate discriminates and results in unique assemblies and maturation behaviors based on the initial protein ensemble. However, it seems the assembly and maturation behavior are not a direct result of the degree of conformational dynamics and plasticity in the initial protein. In the case of CytR, the fully-folded system forms condensates that resolubilize, while the highly disordered state immediately aggregates. Whereas, in the case of FruR, the folded state induces spontaneous aggregation, and the more dynamic, molten globular, system results in short-lived condensates. These results seem to suggest the polyphosphates' ability to discriminate between the initial protein ensemble may not be able to reveal what that initial protein ensemble is unless it is already known.

We thank the reviewer for providing constructive comments on our work. On the final paragraph: we agree that the outcome does not provide information on nature of the starting ensemble. As of now, our experimental results are primarily observations on questions related to maturation outcomes when protein ensembles of varying structure, compactness and stability interact with polyP. if there are differences in the native ensemble due to mutations (which at times cannot be revealed by ensemble probes), polyP appears to discern it more efficiently than DNA.