Myxobacteria restrain Phytophthora invasion by scavenging thiamine in soybean rhizosphere via outer membrane vesicle-secreted thiaminase I

Public metabolites such as vitamins play critical roles in maintaining the ecological functions of microbial community. However, the biochemical and physiological bases for fine-tuning of public metabolites in the microbiome remain poorly understood. Here, we examine the interactions between myxobacteria and Phytophthora sojae, an oomycete pathogen of soybean. We find that host plant and soil microbes complement P. sojae’s auxotrophy for thiamine. Whereas, myxobacteria inhibits Phytophthora growth by a thiaminase I CcThi1 secreted into extracellular environment via outer membrane vesicles (OMVs). CcThi1 scavenges the required thiamine and thus arrests the thiamine sharing behavior of P. sojae from the supplier, which interferes with amino acid metabolism and expression of pathogenic effectors, probably leading to impairment of P. sojae growth and pathogenicity. Moreover, myxobacteria and CcThi1 are highly effective in regulating the thiamine levels in soil, which is correlated with the incidence of soybean Phytophthora root rot. Our findings unravel a novel ecological tactic employed by myxobacteria to maintain the interspecific equilibrium in soil microbial community.


Peer Review File
Myxobacteria restrain Phytophthora invasion by scavenging thiamine in soybean rhizosphere via OMV-secreted thiaminase I Reviewers' Comments: Reviewer #2: Remarks to the Author: The authors make a strong case for the importance of thiamine acquisition in the rhizosphere.The specific role of the EEB strain in inhibiting oomycete infection by limiting the availability of thiamine has been demonstrated in several experiments which are carefully described.The method of thiamin delivery to the environment is especially interesting.The effectiveness of this bacteria as in limiting disease development by oomycete pathogens, suggests that it might have a role as a biocontrol agent in either greenhouse or field applications.But the focus of this manuscript is rightly on the biological significance of competition for thiamine resources, since this is a topic that will be of interest to the largest reading audience.Auxotrophy for thiamine is a well-known property of phytophthora pathogens (Erwin and Ribeiro 1996), but this is the first manuscript that I am aware of that considers it biological significance.In fact, the enormity of the evidence provided by the authors forced me to think how oomycete soil pathogens have adapted to this selection pressure, a research direction not covered in this manuscript.The most important difference between fungal pathogens and oomycetes is that swimming zoospores of oomycetes travel at speeds of 120-150 um/sec and can migrate through soil water or on the surface to find new hosts (Erwin and Ribeiro 1995).Newly released zoospores can swim for up to 24 h before encysting.In the absence of appropriate signals, a second zoospore can emerge from the cyst to continue swimming.Orientation to host roots can come from host specific signals or stress compounds such as the release of ethanol from roots under flooded conditions.Early research suggested that swimming zoospores ran on existing stores and isotope tracing experiments indicated that they did not express transporters capable of acquiring sugars or amino acids.However swimming zoospores do uptake polyamines, and polyamines are excreted from plant roots (Chibucos and Morris 2006).Polyamines however do NOT function as chemotactic molecules.I think that this observation is significant for this work because in plants thiamine transporters are also polyamine transporters (Martinus et al 2016).Related transporter to PUTs are also found in Oomycetes.Polyamines like thiamine are also required for rapidly dividing cells.Zoospores that are generated and released from hyphae in a thiamine-poor environment have the opportunity to acquire additional thiamine resources in transit.The authors have also noted that thiamine degradation in widespread across bacteria (Supplementary Fig. 3C).Thus selection pressure to respond to thiamine depletion in the soil has existed for perhaps millions of years.It seems likely to me that swimming zoospores are also capable of scavenging thiamine while swimming, and if this can be demonstrated, this is an important and relatively simple adaptive strategy of oomycetes in response to this selection pressure, that fungal pathogens don't have.A second reason for asking the authors to explore this avenue of research is that the ability of Phytophthora zoospores to migrate in the soil may well limit the effectiveness of a biocontrol strategy employing thiamine-degrading microbes in a field soil, but not as a foliar application to leaves or in greenhouse pots.I believe that this can be convincingly demonstrated by the authors and contribute to strengthening the value of this manuscript.Suggested Experiments.The highest concentration of zoospores of Phytophthora sojae are produced by transferring several agar plugs to a fresh V8 agar plate.Repeated washing of the plate (every 20 min) is initiated twothree days later when most of the plate surface has almost covered the agar surface.Thiamine content of zoospores can be manipulated as follows.EGB colonization of hyphae producing sporangia may limit the production of zoosporangia or the thiamine stores delivered to zoospores.Here I suggesting to test what happens when EBG In contrast does thiamine supplementation of zoospores improve their infectivity on soybean leaves (Supplementary fig 3)?Does thiamine supplementation diminish the effectiveness of EGB as a foliar biocontrol agent ?Overall impression A significant omission of the abstract was that there was no mention that the thiaminase was excreted in vesicles.In the introduction especially, I found the phrasing of ideas to be awkward, I am not happy with the present title but with revisions, I still feel that this manuscript is a very good fit for Nature Communications Other comments : Lines 180-183 the results are interpreted as "leading to the death of P sojae.Lines 340-341 Later they point out that the mechanism is not killing but inhibition of growth aand also Fig. 2d I found the images of Fig. 2a difficult to interpret.It would be better if the legend included a sentence of explanation as I don't know what I am supposed to be seeing.Lines 283-284…recombinant proteins against strain P6497, indicating that lipoation and membrane association are not necessary for its growth-inhibitory effect against Phytophthora True but the efficiency of inhibition by direct contact supplementary 3F was much much higher and should have been memorably mentioned in the discussion.Lines291 -294 germination of zoospores (Supplementary Fig. 3e).Furthermore, CcThi1 effectively limited the infection of soybean leaves by strain P6497 in the biocontrol assay.The diameter of disease spots was significantly reduced with the increase of CcThi1 concentration (Supplementary Fig. 3f).So is this negated by thiamin supplementation of swimming zoospores?
Entirely by accident I found possibly the earliest mention of the possible importance of thiamine to phytophthora ERWIN DC, KATZNELSON H. Suppression and stimulation of mycelial growth of Phytophthora cryptogea by certain thiamine-requiring and thiamine-synthesizing bacteria.Can J Microbiol. 1961 Dec;7:945-50. doi: 10.1139/m61-119. PMID: 13890710.I only knew of the thiamin requirement from the text cited below.
My interest in zoospores within the oomycete research community is well known.Combining that with the references that I felt needed to be cited is a clear indication of my identity.

References
Chibucos MC, Morris PF. Levels of polyamines and kinetic characterization of their uptake in the soybean pathogen Phytophthora sojae.Appl Environ Microbiol. 2006 May;72(5):3350-6. doi: 10.112872(5):3350-6. doi: 10. /AEM.72.5.3350-3356.2006Reviewer #3: Remarks to the Author: This impressive manuscript describes an extensive investigation into the importance of thiamine for oomycete disease of soybean, and how thiaminases secreted by myxobacteria can protect against disease by inhibiting oomycete growth.This is a very important piece of work for the field.It is very thorough and provides a wealth of data that deliver a mechanistic understanding of interactions between bacteria, oomycetes and plants mediated by a specific metabolite.The findings are broadly relevant to soil ecosystems, with potential applications in biocontrol and sustainable agriculture.
The manuscript is generally well-written, but would benefit from English language editing as many sentences had minor grammatical problems.
The taxonomy of the myxobacteria as presented in line 86 and around line 513 is wrong.Myxobacteria were recently reclassified as a distinct phylum from Deltaproteobacteria called the Myxococcota -see Waite et al (2020) and Oren and Garrity (2021), which also affected the number of myxobacterial suborders.
I found the introduction a bit confusing.It read like a list of statements about microbiomes and there wasn't a clear logic running through it.Does the microbiome dictate which metabolites are found in the public commons, or vice-versa?If public goods drive microbiome resilience, how does mono-cropping render microbiome structure unstable, and why is the accumulation of pathogens an indicator of instability?I was also confused by the presentation of the earliest results sections, which seemed unnecessarily complicated.Why use a bioassay for thiamine, when later it is detected directly by UPLC-MS?Why use strain T-1 to put thiamine into the system, when pure thiamine could have been added directly?When using T-1 (Fig 1c), a better control than no E. coli, would have been an E. coli which doesn't secrete thiamine.In Fig 1d, why is there as much growth with water for two of the species as there is with thiamine -only one of the species seems to respond as expected to the water negative control.In I think it is difficult to claim that EGB's effect on P sojae is not predatory, as the co-incubation experiments only seem to last 24h.I assume that predation is dismissed as there is no significant growth of EGB, but 24h is not a long time for myxobacteria to start showing signs of active predation -in our experience many predatory strains take 2-3 days to start exhibiting significant growth when preying upon susceptible organisms.
Can the addition of thiamine stop the inhibitory effect of adding EGB? Line 203: Without showing that at this point in the paper it is difficult to conclude that it is specifically thiamine and not some other secretion of EGB that is inhibiting oomycete growth, which makes the conclusion in line 213-214 difficult to justify.
What is the significance of thiaminase being secreted in OMVs?If it is periplasmic in the cell, would it not be in the lumen of the OMV?How could it access thiamine substrate?Is the OMV permeable to thiamine, or do the OMVs lyse to release the enzyme?Are myxobacteria generally prototrophic for thiamine?
Minor comments Line 91 -in what way is it novel?Line 92 -it is not strictly increasing the plant's resistance to Phytophthora, but reducing the amount of pathogen.Line 95 -The statement about HGT seems throwaway.What do you mean by this statement?How is it relevant?Line 96-97.In what way is it unique?And what do you mean by fine-tuned?Fine-tuned implies that the concentration of thiamine and thiaminase in soil are tightly-regulated, but that hasn't been shown -rather that small changes in thiamine can have substantial impacts on oomycete growth and pathogenesis.Line 172 -words like 'plunder' and 'greedily' seem excessively poetic/emotive.(also 'pivotal' line 235, 'loopholes' line 557 and 'raiding' -several places including abstract).Line 178 -why would suppression of P sojae growth be unexpected when EGB is a known predator?Fig 2

Response to review's suggestions:
Reviewer #2 (Remarks to the Author): The authors make a strong case for the importance of thiamine acquisition in the rhizosphere.The specific role of the EEB strain in inhibiting oomycete infection by limiting the availability of thiamine has been demonstrated in several experiments which are carefully described.The method of thiamin delivery to the environment is especially interesting.The effectiveness of this bacteria as in limiting disease development by oomycete pathogens, suggests that it might have a role as a biocontrol agent in either greenhouse or field applications.But the focus of this manuscript is rightly on the biological significance of competition for thiamine resources, since this is a topic that will be of interest to the largest reading audience.Auxotrophy for thiamine is a well-known property of phytophthora pathogens (Erwin and Ribeiro 1996), but this is the first manuscript that I am aware of that considers it biological significance.In fact, the enormity of the evidence provided by the authors forced me to think how oomycete soil pathogens have adapted to this selection pressure, a research direction not covered in this manuscript.
The most important difference between fungal pathogens and oomycetes is that swimming zoospores of oomycetes travel at speeds of 120-150 um/sec and can migrate through soil water or on the surface to find new hosts (Erwin and Ribeiro 1995).Newly released zoospores can swim for up to 24 h before encysting.In the absence of appropriate signals, a second zoospore can emerge from the cyst to continue swimming.Orientation to host roots can come from host specific signals or stress compounds such as the release of ethanol from roots under flooded conditions.
Early research suggested that swimming zoospores ran on existing stores and isotope tracing experiments indicated that they did not express transporters capable of acquiring sugars or amino acids.However swimming zoospores do uptake polyamines, and polyamines are excreted from plant roots (Chibucos and Morris 2006).Polyamines however do NOT function as chemotactic molecules.I think that this observation is significant for this work because in plants thiamine transporters are also polyamine transporters (Martinus et al 2016).Related transporter to PUTs are also found in Oomycetes.Polyamines like thiamine are also required for rapidly dividing cells.Zoospores that are generated and released from hyphae in a thiamine-poor environment have the opportunity to acquire additional thiamine resources in transit.The authors have also noted that thiamine degradation in widespread across bacteria (Supplementary Fig. 3C).Thus selection pressure to respond to thiamine depletion in the soil has existed for perhaps millions of years.
Response: Thank you for your recognition of the scientific significance of our work.Adaption of soil oomycetes pathogens to thiamine depletion pressure is a valuable issue for the adaptive evolution.This should inspire the interest of evolutionary biologist.Further experiments will be designed in our following research to investigate this issue.
It seems likely to me that swimming zoospores are also capable of scavenging thiamine while swimming, and if this can be demonstrated, this is an important and relatively simple adaptive strategy of oomycetes in response to this selection pressure, that fungal pathogens don't have.A second reason for asking the authors to explore this avenue of research is that the ability of Phytophthora zoospores to migrate in the soil may well limit the effectiveness of a biocontrol strategy employing thiamine-degrading microbes in a field soil, but not as a foliar application to leaves or in greenhouse pots.I believe that this can be convincingly demonstrated by the authors and contribute to strengthening the value of this manuscript.
Response: Although Phytophthora zoospores might escape thiamine limitation by migration, thiaminase I could diffuse into the environment to decrease thiamine availability.The key is the population of myxobacteria in the rhizosphere and the expression and secretion levels of thiaminase I.We will set field tests to evaluate the effectiveness of the thiamine scavenging strategy in control soybean Phtytophthora root rot this summer.This strategy indeed will sound more effective if Phytophthora zoospores need to absorb thiamine during mitigation.According to your kind suggestion, we checked the thiamine utilization by P. sojae zoospores.The best method to determine the absorption of thiamine is with isotop labelled thiamine.However, we could not get 14 C-labelled thiamine.Thus, we measured the content of thiamine in zoospores by bioassay.6×10 4 zoospores were added into the thiamine solution (0.1 nM), followed by incubation at 25°C for 0, 20 and 40 min.The zoospores were collected from 4 mL solution by centrifugation at 7000 g.The cells were re-suspended in 800 μL H2O and disrupted by Bioprep-24 Homogenizer (40 s, speed 6.00M/S, 3 times).4 mL extracellular supernatant was concentrated by freeze-drying and then dissolved in 800 μL H2O.The thiamine content was determiend by bioassay methods.
To verify the reliability of the method, we measured the detection limit of bioassay method with auxotrophic strain E. coli K-12 ΔthiE.A linear relationship is identified between E. coli K-12 ΔthiE growth and thiamine content (Response Fig. 1a) with a detection limit of <0.05 nM.Meanwhile, we found that zoospores keep active migration and swimming in the solution during detection process.Thiamine content inside the zoospores increased at 20 min, indicating the transportation of thiamine from outside.Response: In our study, we prepared the zoospores by incubation of the mycelia of Phytophthora in sterile water.According to your kind suggestion, we preformed the supplementary experiments about the zoospores preparation by suggested method (PF Morris et al., 1998), and 1.2×10 6 zoospores/plate (6-cm diameter) were obtained.By calculating the biomass of Phytophthora mycelia from two methods, high quantity of P. sojae zoospores are produced compared to liquid induction.Hence, zoospores used in all supplementary experiments were prepared according to the recommended method.
Revised Fig. S5e Growth of strain K-12 ΔthiE in 96 well-plates with the addition of thiamine (1 μM) and lysate of P6497 zoospores.The data represent the means ± SEM (n = 3).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).
In the revised manuscript, Fig. S3e  Response: From our previous study (results present as Fig. S2d in the revised manuscript), we observed that strain EGB cells adhered to the mycelia of P. sojae during co-culture.According to your suggestion, we provided supplementary results to show the production of zoospores during co-culture of EGB and P. sojae.Cells of strain EGB adhered to mycelia remarkably limits the production of zoospores (presented as Fig. S2e in the revised manuscript and below).These results showed that EGB adhering not only inhibits P. sojae growth, but also eradicates zoospores production, both of which are important for survival and plant infection of Phytophthora.
Revised Fig. S2e.Effect of strain EGB incubation on the production of zoospores of P. sojae P6497.Scale bars: 100 μm.
The result was added in the revised manuscript (line 177-178, Supplementary material Fig. 2e).
In contrast does thiamine supplementation of zoospores improve their infectivity on soybean leaves (Supplementary fig 3)?Does thiamine supplementation diminish the effectiveness of EGB as a foliar biocontrol agent?
Response: According to your kind suggestion, we conducted a supplementary experiment regarding the effects of thiamine supplementation on Phytophthora infection (mycelia and zoospores) and biocontrol efficiency of EGB.
Thiamine supplementation showed no obvious effects on the infection of the P6497 mycelia and zoospores from the in vitro leaf infection experiment.We deduced that the thiamine content in leaves is high enough to support Phytophthora infection and growth, because plants can synthesize thiamine (A Goyer et al., 2010).
However, thiamine supplementation obviously decreased the biocontrol efficacy of strain EGB and completely diminished inhibition effect of CcThi1, and increased the lesion length of P. sojae on soybean leaves (present as Fig. S7c, d in the revised manuscript, line 361-363).Otherwise, the effects of thiamine supplementation on zoospores growth were also evaluated.We found that 2.96 μM thiamine fully restores the growth of P6497 in the presence of CcThi1 (present as Fig. S7a, b in the revised manuscript, line 357-361).These results indicated that thiaminase, but not other factor secreted by strain EGB inhibits oomycete growth.
These supplementary results and related descriptions have been provided in the revised manuscript (present as Supplementary Fig. 7).
Thank you for your kind suggestion, which is very important for the conclusion that EGB inhibits Phytophthora growth by thiaminase I-mediated thiamine deficiency.

Overall impression
A significant omission of the abstract was that there was no mention that the thiaminase was excreted in vesicles.
Response: Thank you for your kind suggestion.We have revised the Abstract based on our key findings with concise statement, and the OMVs-mediated secretion of thiaminase I CcThi1 was also added (line 9-11).
In the introduction especially, I found the phrasing of ideas to be awkward,

Response:
Based on the research status of metabolic exchange in plant holobionts, we initially designed the Introduction into several sections: 1) Public goods mediated microbial interactions is important for the ecological function of community; 2) Root exudates serve as public goods and drive the assembly and functional maintenance of the microbial community; 3) B-vitamins are public goods that essential for oomycetes growth; 4) Predatory myxobacteria regulate the microbiome by predation or inhibition, and thereby highlight our key findings.Our logic is that the thiamine mediated microbial interaction occurred in plant holobiont.Since trophic interaction plays important roles in shaping the microbiome, we put our research in the background of plant microbiome.However, we realize that these descriptions are provided without sufficient consideration of our scientific statement and we did not display the scenario in microbiome level in this work.I am not happy with the present title but with revisions, I still feel that this manuscript is a very good fit for Nature Communications Response: Your scientific comments has made us more aware of the biological significance of CcThi1-mediated thiamine deficiency between myxobacteria and Phytophthora.According to the key findings and importance, we revised the manuscript title as: Myxobacteria restrain Phytophthora invasion by scavenging thiamine in soybean rhizosphere via OMV-secreted thiaminase I.

Other comments:
Lines 180-183 the results are interpreted as "leading to the death of P sojae.
Lines 340-341 Later they point out that the mechanism is not killing but inhibition of growth aand also Fig. 2d Response: In previous study, we preformed the co-cultural assay within 1 d, while we realized that 1 d may be not long enough for myxobacteria to start predation.Hence, we constructed the supplementary co-cultural assay of EGB and P. sojae within 3 d.
For the liquid co-cultural assay of P. sojae P6497 and EGB, the biomass of P6497 showed 25-fold decrease (Present as Fig. 2c and Supplementary Fig. 2c in the revised manuscript), and abundant cells of strain EGB were observed to adhere to the mycelia of P6497 (Supplementary Fig. S2d in the revised manuscript).Considering the cell wall integrity of P6497 in EGB treatment, we deduced that the physical hindering by myxobacterial cells limited substrates and oxygen transportation to P6497, which indirectly leads to cell death.
In the later, we identified that myxobacteria EGB inhibited P6497 growth by decomposing essential thiamine with a type I thiaminase CcThi1.By thiamine supplement experiment and microscopic observation, we concluded that direct interaction between EGB and P6497 is growth inhibition via thiamine competition, but not cell killing.
I found the images of Fig. 2a difficult to interpret.It would be better if the legend included a sentence of explanation as I don't know what I am supposed to be seeing.
Response: For a clear presentation, revised Fig. 2a with illustrations of normal (→) or inhibited (├) colony expansion of P6497 was provided, and the figure instruction was also added in the revised legend.
For visualization of the P. sojae during co-culture, we used a GFP-labelled P6497 in Fig. 2a, similar results were obtained that EGB suppresses the growth of P6497 mycelia in a contact-independent manner.Meantime, we also identified that no fluorescence quenching was detected from the co-culture assay on V8 plates, indicating that EGB inhibits the growth of P. sojae, but not kills the cells.
In the revised manuscript, we moved the results of co-culture of EGB and P6497 to the Supplementary material as Fig. S2a, and the complementary co-culture result of EGB and GFP-labelled P6497 was provided as revised Fig. 2a.
Thank you for your kind suggestion.
Lines 283-284…recombinant proteins against strain P6497, indicating that lipoation and membrane association are not necessary for its growth-inhibitory effect against Phytophthora True but the efficiency of inhibition by direct contact supplementary 3F was much much higher and should have been memorably mentioned in the discussion.
Response: Thank you for the helpful comment, which is important for the exploring the experimental results in depth.
As we verified, CcThi1 could inhibit mycelial growth of P. sojae and reduce the diameter of disease lesions in leaf infection experiments in vitro.In this experiment, strain EGB was applied directly on the mycelial discs.Direct adhesion of EGB cells to the mycelia allows myxobacteria to interact efficiently with Phytophthora, including physical hampering on the viability and zoospore producing of Phytophthora, and the high concentration of OMVs within local position.This may contribute to the improved prevention of Phytophthora infection in the in vitro leaf infection experiment.
According to your kind suggestion, we have supplemented this discussion in the revised manuscript (line 591-596).
Lines291 -294 germination of zoospores (Supplementary Fig. 3e).Furthermore, CcThi1 effectively limited the infection of soybean leaves by strain P6497 in the biocontrol assay.
The diameter of disease spots was significantly reduced with the increase of CcThi1 concentration (Supplementary Fig. 3f).

So is this negated by thiamin supplementation of swimming zoospores?
Response: As above mentioned, we carried out supplementary experiments regarding the effects of thiamine supplementation on zoospores and mycelia infection with or without presence of CcThi1 and EGB.Indeed, it was negated by thiamine supplementation of swimming zoospores (Supplementary Fig. 7, line 357-363).This impressive manuscript describes an extensive investigation into the importance of thiamine for oomycete disease of soybean, and how thiaminases secreted by myxobacteria can protect against disease by inhibiting oomycete growth.This is a very important piece of work for the field.It is very thorough and provides a wealth of data that deliver a mechanistic understanding of interactions between bacteria, oomycetes and plants mediated by a specific metabolite.The findings are broadly relevant to soil ecosystems, with potential applications in biocontrol and sustainable agriculture.
The manuscript is generally well-written, but would benefit from English language editing as many sentences had minor grammatical problems.
Response: Thank you for the scientific evaluation for the importance of the manuscript, which is important for our future related research.
The writing of manuscript has been polished by a professional English editing agency to avoid grammatical problems, and we think the writing has been substantially improved.
The taxonomy of the myxobacteria as presented in line 86 and around line 513 is wrong.
Myxobacteria were recently reclassified as a distinct phylum from Deltaproteobacteria called the Myxococcota -see Waite et al ( 2020) and Oren and Garrity ( 2021), which also affected the number of myxobacterial suborders.
Response: Thank you for the kind suggestion regarding the taxonomy of the myxobacteria.These works (Waite DW et al., 2020;Oren A et al., 2021) provide new insights for the evolutionary route of myxobacteria, which have brought extensive attentions and discussions for us and myxo community when published.
Considering the importance of the reclassification (Waite DW et al., 2020;Oren A et al., 2021), we have revised the distribution of thiaminase I in myxobacteria taxa referring to the novel phylum-level lineages (phylum Myxococcota), and the corresponding Fig. 6a has been revised.Thanks.
I found the introduction a bit confusing.It read like a list of statements about microbiomes and there wasn't a clear logic running through it.Does the microbiome dictate which metabolites are found in the public commons, or vice-versa?If public goods drive microbiome resilience, how does mono-cropping render microbiome structure unstable, and why is the accumulation of pathogens an indicator of instability?
Response: Metabolites from microbes and rhizosphere are commonly regarded as public goods in ecosystems, and the metabolic cross-feeding is an important process that broadly shape microbial communities and promote complex interactions between microbes and plant (Mee M T et al., 2014;Zhao M et al., 2021).Recent researches about rhizosphere show that root exudates dictate which health-promoting soil microbes are found in the rhizosphere (Rolfe SA et al., 2019).However, pathogens are also able to utilize the public goods for growth by different strategies during coevolution (Gu S et al., 2020).Hence, in relatively homogenous agricultural exerted by predation or inhibition were important, and thereby highlight out our core findings.
I was also confused by the presentation of the earliest results sections, which seemed unnecessarily complicated.Why use a bioassay for thiamine, when later it is detected directly by UPLC-MS?
Response: As mentioned in Introduction section, B vitamins are common public goods in oceanic algae communities.Phytophthora, which is phylogenetically related to algae, is also known for the thiamine (vitamin B1) auxotrophic property.However, its thiamine resources in saprophytic stage is not clear in complex plant holobiont.We thought it was necessary to clarify this question.In the Results section 3.1, we aimed to identify if soybean root exudates and rhizosphere bacteria could provide thiamine for growth of Phytophthora.The source of thiamine could determine the interactions among soybean plant, P. sojae, and soil microbiome.
Since the detection limit (0.13 μM) of HPLC is not low enough to measure thiamine in the environments and medium, bioassay was used with E. coli K-12 ΔthiE as the indicator (thiamine detection limit 0.05 nM).UPLC-MS was used to identify thiamine decomposition product by SUPL, CcThi1 (1.19 μM) and high concentration of thiamine (0.29 mM) was incubated with thiaminase I at 25 °C for 6 h to detect the product.It is much convenient and precise to measure thiamine by bioassay method.
For clear presentation, we have adjusted the presentation of Fig. 1 and added a description (line 114-120) in the revised manuscript.
Why use strain T-1 to put thiamine into the system, when pure thiamine could have been added directly?When using T-1 (Fig 1c), a better control than no E. coli, would have been an E. coli which doesn't secrete thiamine.
Response: Here, our key point is to identify the thiamine supplier in rhizosphere.
We deduced that rhizosphere bacteria may provide thiamine for Phytophthora.Hence, we recovered 1267 bacterial isolates from rhizosphere soil, and 65 isolates demonstrated various degree of thiamine production.Among the 65 isolates, strain T-1 (Preliminary identified as Escherichia sp., not E. coli) was a higher thiamine producing isolate and was selected as a representative of thiamine producers in soil.According to the results (present as Fig. 1c, d in the revised manuscript), strain T-1 promoted the growth of P6497, indicating that the thiamine exchange exists in plant holobiont (present as Fig. 1e in the revised manuscript).
Meanwhile, pure thiamine was used as positive control to show that thiamine producers support the growth of Phytophthora by supplying thiamine.
In Fig 1d, why is there as much growth with water for two of the species as there is with thiamine -only one of the species seems to respond as expected to the water negative control.In

Response:
The different Phytohpthora growth responses to the additives is partially attributed to their different growth rates and observed mycelia structures.P. sojae mycelia expands much slower than P. capsi and P. parasitica when cultured in V8 medium and tends to form compact colony, which results in different biomass under same cultural condition.
During cultivation of P6497, weak growth of P6497 was observed in water treatment.We deduced that stored thiamine within mycelia may support growth of P6497 to some degree.However, the additives in the synthetic medium (P1 medium) can support the increase the Phytophthora biomass with magnitude order.The biomasses with additives were statistically higher than those in control (H2O).
To further check the effects of additives on growth of Phytophthora, we repeated the co-culture assay with optimal concentration of CaCl2 (0.01%, m/v) in P1 medium (Erwin DC et al., 1961).Under this condition, growth of P. sojae was much evident.
Phytophthora DNA was isolated from a mixture of 3 wells (regarded as a repetition) and biomass determined from three independent experiments (9 wells) by qPCR.The raw data are provided in the files of Raw Source Data, and the related method and results (present as Fig. 1a and b, Supplementary Fig. 1a and b  Response: Phytophthora biomass was quantified by qPCR using primers (PS-F1 and PS-R1) specific for the actin gene (Table S3), which is not existed in genome of Escherichia sp.strain T-1.
I think it is difficult to claim that EGB's effect on P sojae is not predatory, as the coincubation experiments only seem to last 24h.I assume that predation is dismissed as there is no significant growth of EGB, but 24h is not a long time for myxobacteria to start showing signs of active predation -in our experience many predatory strains take 2-3 days to start exhibiting significant growth when preying upon susceptible organisms.
Response: Considering our previous studies, predation of bacteria and fungi (prey) by strain EGB was obviously observed after 1 d of incubation, hence, we tested the EGB's effect on P. sojae growth with 24 h of co-culture.However, as you suggested, 1 d may be not long enough for myxobacteria to start predation.
Hence, according to your kind suggestion, we constructed the supplementary co-cultural assay of EGB and P. sojae within 3 d.Results showed that co-cultures of EGB results in 25-fold decrease in P6497 biomass, but the biomass of EGB remains unchanged.These results indicated that strain EGB restrains the growth of P. sojae by inhibition rather than predation, and the adhering of EGB results in decrease of P6497 biomass by indirect physical effect, which was consistent with the results after 1 d of incubation.Considering the importance of these results, the quantitative analysis of the biomass of strains EGB and P6497 from co-culture have been moved to the main text (Fig. 2c), and the growth state of co-culture was present as Supplementary Fig. 2c.
Meantime, the SEM observation of the P6497 mycelia from the co-culture assay with 3 d was also provided (revised Supplementary Fig. 2d).Related descriptions were also provided (line 172-175).Thanks.
Can the addition of thiamine stop the inhibitory effect of adding EGB? Line 203: Without showing that at this point in the paper it is difficult to conclude that it is specifically thiamine and not some other secretion of EGB that is inhibiting oomycete growth, which makes the conclusion in line 213-214 difficult to justify.What is the significance of thiaminase being secreted in OMVs?If it is periplasmic in the cell, would it not be in the lumen of the OMV?How could it access thiamine substrate?Is the OMV permeable to thiamine, or do the OMVs lyse to release the enzyme?
Response: Commonly, enzymes involved in extracellular macromolecules decomposition are entrapped on cell surface by lipo, while counterparts in Gram negative bacteria are secreted in periplasm.Such enzymes are retained closely to the cells by these approaches for effective nutrient acquisition.To our surprise, thiaminase I in Gram negative myxobacteria is generally lipoation and outer membrane located as revealed by signal peptides structure analysis.Gram negative bacteria have long been found producing outer membrane vesicles (OMVs).OMVs play roles in pathogenesis, cell-to-cell communication, and stress responses (Sartorio et al., 2021).OMVs have been considered playing roles in killing prey cells at distance (Whitworth et al., 2011;Evans et al., 2012), which translocate hydrolases or antimicrobial substances to the target cells.Here, we deduce that CcThi1 could disperse in the environment via OMVs blebbing.Otherwise, locating on OMVs might increase the stability of CcThi1 in harsh environment, and also allows myxobacteria to interact with Phytophthora by promoting adhesion to the cell surfaces.However, its biological significance needs further investigation.
We failed to discriminate its location on OMVs by protease digestion for its protease resistance property.However, intact OMVs and broken OMVs by ultrasonification showed similar thiaminase I activity (Response Fig. 1).No specific transporter has ever been identified on outer membrane, we deduced that OMVs is permeable to thiamine.Thus, its location on the surface or in the lumen may not restrict its enzymatic activity.Anyway, we think your comments raised interesting topics on the structure of bacterial OMVs.We will try to determine its location on the OMVs in To further verify this conclusion, we deleted the gene encoding ThiC in the model strain M. xanthus DK1622, generating strain CL1006.Strain DK1622 grew on CTT-1 medium (no thiamine), and addition of thiamine promoted its growth.However, thic mutant strain CL1006 lost the ability to grow on CTT-1 medium, which could be partially restored by thiamine addition (Response Fig. 2c).All these results show that myxobacteria are generally prototrophic for thiamine and reserve the ability to utilize exogenous thiamine.
However, considering the aim of the manuscript, these results were not presented with in-depth discussion in the revised manuscript.Response: The function of Ccthi1 homologs in myxobacterial genomes is not annotated before our work.It was conceptually annotated as a solute-binding protein.
Sequence analysis showed that CcThi1 from strain EGB shared rather low sequence identity (less than 20%) with known type-I thiaminases, and phylogenetic tree analysis showed that thiaminase I enzymes from myxobacteria formed a separate branch from those of other species, indicating its novelty on sequence level.Till now, all reported thiamine I are extracellular protein secreted into the environment, however, most homologs of Ccthi1 in myxobacteria are predicted to be outer membrane located as a lipoprotein.Furthermore, thiaminase-mediated metabolite interaction between myxobacteria and Phytophthora provides new insights for the ecological significance of thiamine, indicating functional novelty.Hence, we concluded that CcThi1 from myxobacteria is novel thiaminase that mediated new interaction of myxobacteria and

Phytophthora.
Line 92 -it is not strictly increasing the plant's resistance to Phytophthora, but reducing the amount of pathogen.Response: Previous research show that oomycetes feed by secreting depolymerizing enzymes (public goods) to process complex substrates in the extracellular environment, and taking up the resulting simple nutrients into their own cells (Richards TA et al., 2013).During the evolutionary process, horizontal gene transfer (HGT) may play major role in in shaping the repertoire of these enzymes.

Response
However, this introduction was not closely relevant to our findings.According to your kind suggestion, we have deleted these statements and reorganized the part of Introduction.Thanks.
Line 96-97.In what way is it unique?And what do you mean by fine-tuned?Fine-tuned implies that the concentration of thiamine and thiaminase in soil are tightly-regulated, but that hasn't been shown -rather that small changes in thiamine can have substantial impacts on oomycete growth and pathogenesis.
Response: Public goods in microbial ecosystem have been shown to regulate the relationships of microbial members.Microbes engage in the acquisition and protection of public goods by diversified strategies, including resource limitation and cell killing, such as siderophore-iron complexes in iron acquisition (Cordero O X et al., 2012) and cell killing for exploiting public goods (Lonergan Z R et al., 2019).Consequently, social behaviors such as cooperation, cheating or division of labor arise in microbial communities.However, mechanisms involved in regulation of public goods supplying such as thiamine in microbial communities remain underexplored.
To avoid confusion, we have revised the manuscript to remove such emotional expression in scientific presentation (line 167-170).Thank you for your kind suggestion.Response: For a clear presentation, revised Fig. 2a with illustrations of normal (→) or inhibited (├) colony expansion of P6497 was provided, and the figure instruction was also added in the revised legend.The colony radius of normal (→) or inhibited (├) growth of P. sojae was measured (revised Fig. 2a, b).By MTT assay, we identified that SUPL-treated mycelia are metabolically active, combing the results that SUPL treatment exhibits no fluorescence quenching, we concluded that secreted SUPL from strain EGB inhibited growth of Phytophthora, but not killed the mycelia.Hence, the measurement of metabolic activity was kept in the text and present as Fig. 2d in the revised manuscript, while the observation of MTT staining of P6497 mycelia was moved into supplementary material as Fig. S4a.
Otherwise, for visualization of the P. sojae during co-culture, we used a GFPlabelled P6497 in Fig. 2a, similar results were obtained that EGB suppresses the growth of P6497 mycelia in a contact-independent manner.Meantime, we also identified that no fluorescence quenching was detected from the co-culture assay on V8 plates, indicating that EGB inhibits the growth of P. sojae, but not kills the cells.
In the revised manuscript, we moved the results of co-culture of EGB and P6497 to the Supplementary material as Fig. S2a, and the complementary co-culture result of EGB and GFP-labelled P6497 was provided as revised Fig. 2a (below).The results of microscopic analysis of SUPL-treated GFP-labelled P6497 mycelia have been moved to Supplementary material as Fig. S4b.
Fig 1b, it doesn't look like P. sojae is being stimulated by any of the additives -it seems hard to reconcile the results in Fig 1b with the data in Fig 1d.In Fig 1c -how is the biomass of phytophthora distinguished from that of E. coli?
When the incubation time last to 40 min, the thiamine inside the zoospores decreased.Thiamine concentration in the solution decreased all the time (Response Fig.1b).These results indicate that swimming zoospores are able to absorb thiamine during swimming.Whereas, evidences with 14 C-labelled thiamine are needed to directly prove the thiamine utilization.Response Fig. 1. Assay for thiamine utilization of Phytophthora zoospores during migration.a, Relationship between E. coli K-12 ΔthiE growth and thiamine content.b, Determination of thiamine in and outside of Phytophthora zoospores during incubation of 0, 20 and 40 min.The data represent the means ± SEM (n = 3).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).
1. Suggested Experiments.The highest concentration of zoospores of Phytophthora sojae are produced by transferring several agar plugs to a fresh V8 agar plate.Repeated washing of the plate (every 20 min) is initiated two-three days later when most of the plate surface has almost covered the agar surface.Thiamine content of zoospores can be manipulated as follows.
was replaced with the supplementary result (present as Supplementary Fig. 5e in the revised manuscript) according to your suggested experiments.EGB colonization of hyphae producing sporangia may limit the production of zoosporangia or the thiamine stores delivered to zoospores.Here I suggesting to test what happens when EBG Revised Fig. S7.Effects of thiamine supplementation on zoospores growth and soybean infection of Phytophthora.a, b, Inhibition of mycelial growth on 10% V8 plates by CcThi1 (0.24 nM) with supplementation of various concentrations of thiamine (0, 2.96×10 -3 , 2.96×10 -2 , 2.96×10 -1 and 2.96 μM) (a), and microscopic observation of mycelia after CcThi1 treatment was performed to measure the hyphal length after 7 d of incubation (b).c, d, Effects of thiamine supplementation on infections of Phytophthora mycelia and zoospores toward soybean leaves.All the leaves were stained with lactophenol trypan blue and decolorized with chloral hydrate (c), and the corresponding lesion diameters were measured (d).The data represent the means ± SEM (n = 5 or 10).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).
According to your kind suggestion, we have reorganized the Introduction based on the research background and key findings of the manuscript.Part 1 (line 23-42): Public goods exchange is a crucial process driving the evolution of microbial ecosystems; Part 2 (line 43-59): Thiamine is regarded as public goods, consequences of thiamine auxotrophy in Phytophthora pathogens are largely unclear in territorial ecosystem; Part 3 (line 60-74): Root exudates-mediated root-associated bacterial communities determine pathogen invasion, and exploring the thiamine supply and regulation in the holobiont of Phytophthora-plant-rhizosphere microbes is meaningful.Part 4 (line 75-89): Functions of predatory myxobacteria in microbial ecosystems exerted by predation or inhibition were important, and thereby highlight out our core findings.
Entirely by accident I found possibly the earliest mention of the possible importance of thiamine to phytophthora ERWIN DC, KATZNELSON H. Suppression and stimulation of mycelial growth of Phytophthora cryptogea by certain thiamine-requiring and thiamine-synthesizing bacteria.Can J Microbiol.1961 Dec;7:945-50.doi: 10.1139/m61-119.PMID:    13890710.I only knew of the thiamin requirement from the text cited below.My interest in zoospores within the oomycete research community is well known.Combining that with the references that I felt needed to be cited is a clear indication of my identity.ReferencesChibucos MC, Morris PF. Levels of polyamines and kinetic characterization of their uptake in the soybean pathogen Phytophthora sojae.Appl Environ Microbiol.2006 May;72(5):3350-6.doi: 10.1128/AEM.72.5.3350-3356.2006.PMID: 16672477; PMCID: PMC1472313.D. C. Erwin and O. K. Ribeiro.1996.Phytophthora Diseases Worldwide 592pp.St The American Phytopathological Society, Paul, MN, USA: Martinis J, Gas-Pascual E, Szydlowski N, Crèvecoeur M, Gisler A, Bürkle L, Fitzpatrick TB.Long-Distance Transport of Thiamine (Vitamin B1) Is Concomitant with That of Polyamines.Plant Physiol.2016 May;171(1):542-53.doi: 10.1104/pp.16.00009.Epub 2016 Mar 22. PMID: 27006489; PMCID: PMC4854701.Mulangi V, Chibucos MC, Phuntumart V, Morris PF. Kinetic and phylogenetic analysis of plant polyamine uptake transporters.Planta.2012 Oct;236(4):1261-73.doi: 10.1007/s00425-012-1668-0. Epub 2012 Jun 19.PMID: 22711282.Response: Thank you for your kind suggestion regarding to thiamine requirement of Phytophthora and background of transporters.These references are crucial for us to deeply understand the biological significance of the CcThi1-medidated thiamine competition between myxobacteria and Phytophthora.The thiamine transportation in plant will provide guidance for our future research on the absorption of thiamine in Phytophthora.The related references have been cited in the Introduction and Discussion to highlight the importance of thiamine to Phytophthora.

Reviewer # 3 (
Remarks to the Author): Fig 1b, it doesn't look like P. sojae is being stimulated by any of the additivesit seems hard to reconcile the results in Fig 1b with the data in Fig 1d.
in the revised manuscript) are provided in the revised manuscript.Revised Fig. 1 Available thiamine from plant hosts and soil microorganisms promotes the growth of Phytophthora.The results showed that the additives promote the growth of Phytophthora in the synthetic medium (P1 medium, 0.01% CaCl2), and the results in revised Fig 1a is reconciled with the data in revised Fig 1b.In Fig 1c -how is the biomass of phytophthora distinguished from that of E. coli?

Fig
Fig. 2c and Supplementary Fig. 2c, Co-culture assay of P6497 (0.1 g, wet weight) and strain EGB (10 5 cells/mL) or its secreted SUPL (10 mg/mL) in 25 mL of TV medium.The data represent the means ± SEM (n = 3).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).
Revised Fig. S7.Effects of thiamine supplementation on zoospores growth and soybean infection of Phytophthora.a, b, Inhibition of mycelial growth on 10% V8 plates by CcThi1 (0.24 nM) with supplementation of various concentrations of thiamine (0, 2.96×10 -3 , 2.96×10 -2 , 2.96×10 -1 and 2.96 μM) (a), and microscopic observation of mycelia after CcThi1 treatment was performed to measure the hyphal length after 7 d of incubation (b).c, d, Effects of thiamine supplementation on infections of Phytophthora mycelia and zoospores toward soybean leaves.All the leaves were stained with lactophenol trypan blue and decolorized with chloral hydrate (c), and the corresponding lesion diameters were measured (d).The data represent the means ± SEM (n = 5 or 10).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).
our future work.Response Fig. 1 Measurement of thiaminase activity of OMVs and ultrasonic OMVs.The data represent the means ± SEM (n = 3).Means within columns followed by a different letter are significantly different (p < 0.05, ANOVA, Duncan's multiple range test).Discussion regarding the significance of thiaminase secretion via OMVs was supplemented in the revised manuscript (line 586-596).Thanks.Are myxobacteria generally prototrophic for thiamine?Response: Many bacteria can synthesize thiamine, while thiamine auxotrophs must obtain it or its precursors from the environment.Commonly, the pyrimidine moiety is derived from the purine intermediate 5-aminoimidazole ribonucleotide (AIR), which is converted to hydroxymethyl pyrimidine phosphate (HMP-P) by ThiC (HMP synthase).Then, HMP-P is phosphorylated to HMP-PP by ThiD (kinase).The G3P and glycine could be combined to synthesize HET-P in a multistep process by Dxs (1deoxyd-d-xylose-5-phosphate synthase), ThiO (glycine oxidase) and ThiG (thiazole synthase).Thiamine phosphate synthase (ThiE) combines HET-P and HMP-PP to form thiamine monophosphate (TMP), and typically ThiL phosphorylates TMP to produce the active cofactor TPP(Response Fig. 2a, below, Sannino DR et al., 2018).The strains M. xanthus DK1622 and EGB used in this research are thiamine prototrophs.Genome analysis of myxobacteria with the available genome sequences showed that the key enzymes ThiC, ThiD, thiG and thiE and homologs are widely distributed in most members of myxobacteria taxa (2 classes, 4 orders, 7 families and 18 genera)(Response Fig. 2b, below,), indicating that myxobacteria are generally prototrophic for thiamine.

:
According to your suggestion, we have reorganized the Introduction based on the research background and our key findings.The related description has been revised as: Myxobacteria release a novel thiaminase to scanvage the public thiamine in the soil, which in turn regulate the thiamine content and promote plant protection (line 83-85).Thanks.Line 95 -The statement about HGT seems throwaway.What do you mean by this statement?How is it relevant?

Fig 2 -
Fig 2 -I found it difficult to relate panel b to panel a.Panel c didn't seem to add much to the manuscript.In panel d, rather than comparing with/without EGB extract, it would have been more interesting to see what the mycelia looked like at the edge of where growth was being inhibited by EGB in the experiments of panel a.