Review Reports - Prey Lysate Enhances Growth and Toxin Production in an Isolate of <i>Dinophysis acuminata</i>

Round 1

Reviewer 1 Report

Review of ‘Prey lysate enhances growth and toxin production in an isolate of Dinophysis acuminata’ submitted to Toxins.

D. acuminata is a HAB species that producers DSP. It is an obligate mixotroph and uses kleptoplasty for photosynthesis, it is therefore difficult to culture. However, it is important to understand the factors that promote its growth and toxin production to increase our understanding of how this species grows in the environment and improve prediction and mitigation of blooms. Many biotic and abiotic factors have been studied that contribute to increased growth and toxicity; including temp, light, prey availability, and ammonium. The present study illustrates that including prey lysate with prey also increases the growth and toxin production of D. acuminata. These results will potentially be useful for isolating and culturing D. acuminata under laboratory conditions and also suggest that the population dynamics between D. acuminata and its prey are complex.

The results are interesting and this growth study would appear to be a first step in necessary future research to tease apart the actual part of the lysate or associated bacterial fraction that promotes D. acuminata. This paper creates more questions than it answers and the discussion is a lot of speculation. It may perhaps be better presented as a note.

Specific comments.

2.1. Lysate characterization, would better be termed ‘lysate size determination’ I was expecting chemical characterization.

Line 215-217. Is 3 days growth really supported? Is it statistically different from a control with only f/6-Si medium? Were there any visible differences between the cultures? E.g phytopam to confirm cell viability and growth? From the figure it is difficult to see that growth actually occurred. Additionally, would 3 days growth be supported by internal nutrient stores? Is there a no nutrient control?

Line 243? Could an anti-bacterial treatment be used to confirm the effect of associated bacteria within the prey and prey/lysate feeds?

Line 275: what is the prey saturation threshold, please write it out here.

Line 276: what is meant by the ‘mixed treatment’ is this the prey+lysate treatment? Please use consistent terms.

Line 206: typo ‘surround’ should be ‘surrounding’.

Author Response

2.1 Lysate characterization, would better be termed ‘lysate size determination’ I was expecting chemical characterization.

· Done.

Line 215-217. Is 3 days growth really supported? Is it statistically different from a control with only f/6-Si medium?

· Yes the biomass (final cells/mL) was statistically different between the monoculture control and Lysate_ciliatetreatment. We have added text to the results and discussion to clarify that the interpretation is based on statistics and that all Dinophysiswere starved for 2 weeks prior to the experimental period.

· “The maximum biomass (cells/mL) of Dinophysisin the Lysate_ciliate³⁰⁰⁰treatmentwas significantlygreater than the maximum biomass measured in the monoculture control, showing that growth was supported for three days on materials liberated from the ciliate lysate (after two weeks of starvation).”

Were there any visible differences between the cultures? E.g phytopam to confirm cell viability and growth?

· No additional measurements were collected. Cell abundance, as determined by light microscopy, was quantified and growth rate calculated.

From the figure it is difficult to see that growth actually occurred.

· As requested, we made major adjustments to Figure 1, separating the one figure into three panels so that readers can more easily distinguish between the many treatments and two species represented.

Additionally, would 3 days growth be supported by internal nutrient stores?

· Previous studies, including some of our own published work, have shown 3 days of continued division after removal of prey, i.e., on internal reserves; however, in this experiment, this short-term division on internal reserves was not observed in the monoculture control during the experimental period due to the fact that dinoflagellates were starved, in the light, for 2 weeks prior to the beginning of the experiment. Our conclusion, therefore, remains the same, but we have added text to the results and discussion to clarify.

· See our response to your similar comment above to review our changes to the manuscript.

Is there a no nutrient control?

· The control chosen to support this large experiment was a Dinophysismonoculture in medium with no organic amendments as the ability of Dinophysisand Mesodiniumto utilize (or not utilize) the chemical forms in f/6 medium is well understood.

Line 243? Could an anti-bacterial treatment be used to confirm the effect of associated bacteria within the prey and prey/lysate feeds?

· We agree this would be a great way to tease apart bacterial vs ciliate or dinoflagellate roles in nutrient utilization and remineralization; however, trials were run extensively with a variety of antibiotics to create axenic cultures; none were successful to-date given the complex feeding regime for Dinophysis,Mesodiniumand the cryptophytes. We continue to pursue this topic.

Line 275: what is the prey saturation threshold, please write it out here.

· Done. Text added includes “…were below and above the prey saturation threshold: ~2,000 cells/mL…”

Line 276: what is meant by the ‘mixed treatment’ is this the prey+lysate treatment? Please use consistent terms.

· Done. We now include the treatment name (Prey+Lysate_ciliate³⁰⁰⁰) instead of referring to this treatment as the ‘mixed treatment.’

Line 206: typo ‘surround’ should be ‘surrounding’.

· Done.

Reviewer 2 Report

the implication is that dsp is released /produced when lysed cells versus prey is provided. why?

well written clear expt and presentation

Author Response

the implication is that dsp is released /produced when lysed cells versus prey is provided. why?

· It is possible that the predator (Dinophysis) produces more toxin when prey are detected nearby through chemical signaling, i.e., chemicals produced/released by Mesodiniumrubrum. In this case, ciliate cell lysis could exacerbate this effect. Alternatively, the liberated materials may be in a form that is more readily incorporated into toxin synthesis pathways than particulate nutrients, i.e., from consumption of live prey. To avoid adding speculation to the manuscript (as chemicals besides toxins were not quantified in this study), we did not add this information to the discussion.

well written clear expt and presentation

· Thank you.

Reviewer 3 Report

The manuscript "Prey lysate enhances growth and toxin production in an isolate of Dinophysis acuminata" describes a well designed experiment that shows one single fact, that Dinophysis acuminata grows faster and produces more okadaic acid (OA) when fed not solely on the Mesodinium rubrum but also on the lysate of this ciliate. It is clear from the presented data that the growth rate increases about 20%, which results in significant increase in biomass during plateau phase.

However, the idea of the study is not very original because it was shown before that growth and toxin level of Dinophysis depend on the prey. Moreover in the culturing experiments as well as in the field it is hard to reach conditions when there is absolutely no prey lysate, because cells can die due to various factors. Authors need to prove the absence of ciliate lysate in the groups Preyciliate1500 and Preyciliate3000.

Even though the experiment is conducted properly, the interpretation of the results is not clear or even misleasding in some aspects.

The description of prey lysate is important, however not interpreted in the discussion part. The particles that express orange fluorescence are smaller in size of the plastids, that Mesodinium rubrum has. To develop this part more the photographs of Mesodinium cells can be taken and the size of the plastids can be measured or refer to previous publications. However, the explanation is needed what these particles can be if not plastids.

The result about decrease of bacterial abundance, that was assessed by counting particles that incorporated DAPI and were collected on filters. This result needs more explanation of how authors discriminated between bacterial cells and the debris of nuclei of ciliates which also can be stained with DAPI and trapped on 22um sieve.

Figure 1 contains a lot of information packed in a small size. It can be split in 3 figures of bigger sizes: ciliate dynamics, dynamics of high density Dinophysis growth, dynamics of low density Dinophysis growth. The markers should be smaller, so they don’t hide the error bars. In the prey+lysate3000 group D.acuminata run out of prey on the 12^th day of culturing. This fact is wrongly described in the text lines 112-114 “ciliates were completely consumed by day 15-18 in the three treatments to which they were added: Preyciliate1500, Preyciliate3000, and Prey+Lysateciliate3000” . In the group Prey ciliate3000 the prey was depleted on 15^th day. So the group prey+lysate3000 survived with no prey and grew faster for 3 days. How this can be explained? Can the fact of starvation influence the toxin dynamic? Refer to the recent paper Garcia-Potela et al., Marine drugs 2018.

Authors claim that "The addition of ciliate lysate to co-incubations also resulted in maximum toxin quotas and extracellular concentrations of okadaic acid and dinophysistoxin-1” true only in part. Authors compare the toxin quotas to the initial level of toxins, which is usually low in the beginning of culturing and increases toward the plateau phase. If authors compare toxin levels of prey+lysate3000 group with prey3000 group they will find no significant differences in DTX-1, PTX-2 intracellular concentration and extracellular concentration of all the toxins (Figure 2). So only OA concentration seems to increase significantly in these two group, so we can not confidently conclude that addition of lysate results in increase of all toxins. On figure 2 the asterisks indicating significance of the measurement in comparison with the initial level, should be substituted to the asterisks comparing groups between each other to make the conclusion about influence of lysate. It is also more informative to use SD instead of SE for error bars.

The additional experiment and observation can be:

Feeding the Dinophysis with specific fraction of lysate for example only plastids fraction, which can be characterized by certain size (plastids are bigger than 1.3 um as the size of fluorescent particles in presented study) and fluorescent pigments. The isolation of the fraction can be done using sorting technique based on fluorescence or on size.

Authors speculate in the discussion about the possible incorporation of plastids, which can be easily assessed though the fluorescent microscopy. If authors are able to show that cells of Dinophyis acuminata incorporated significantly more plastids from the lysate, it will bring more significance to the study as it would show the alternative way for Dinophysis to acquire the essential plastids. With no experimental data the speculations should be removed from the discussion part.

Authors try to explain the observed increase of growth rates and production of OA by the influence of bacterial community. This part is also highly speculative without the additional experiments about bacterial community characterization (the density of bacterial community and diversity of bacterial species, the differences in the bacterial community with or without lysate). The studies with addition of bacterial inoculums to the cultures of Dinophysis acuminata with the following measurements in changes in growth rates and toxin production can be added, to prove the influence of particular bacterial species on production of toxin.

Discussion is written is a very broad and not specific way, containing a lot of additional information and speculations not supported by presented data, which distracts the reader from the main idea of the manuscript. It can not be accepted in the present form.

Author Response

The idea of the study is not very original because it was shown before that growth and toxin level of Dinophysis depend on the prey.

· The fact that prey have been found to be important to Dinophysisgrowth and toxin production in multiple field and lab studies (cited in our manuscript) is exactly why this study was carried out. The previous study investigating lysate as a nutrient source did not include prey in any treatments.

· We thank the reviewer for pointing out that this needed clarification and so we have added sentences to the introduction explaining why the prey – nutrient dynamic needs to be considered in Dinophysistoxin production and growth.

· “Additionally, ciliates should be considered in this relationship as laboratory studies [7,34-36] have indicated prey abundance as an important controller of Dinophysisgrowth and/or toxin production, and Dinophysisspp. have been found to bloom immediately after and co-occur with ciliate prey in the field [37], suggesting that factors controlling ciliate abundance and distribution are important to down-stream DSP events. Further research is now needed to better understand how inorganic and organic compounds interact with prey and a variety of Dinophysisspp. and isolates if predictive capabilities are to be developed and management strategies created.”

Moreover in the culturing experiments as well as in the field it is hard to reach conditions when there is absolutely no prey lysate, because cells can die due to various factors. Authors need to prove the absence of ciliate lysate in the groups Preyciliate1500 and Preyciliate3000.

· Separating exudate from the live prey treatments is unrealistic in both logistics and practicality as this is never observed in the field (or culturing). We have added text to clarify this in the introduction.

· “By including prey in a treatment with possible nutrient sources, additional questions can be asked regarding the combined roles of ciliates and nutrients in dinoflagellate growth and toxin production. This pairing of prey cells and nutrients also seems more environmentally relevant, as the two are likely found in conjunction in a system. Co-occurrence of cells and lysate, for example, may occur throughout a bloom, due to such processes as sloppy feeding or cell division; however, the presence of extracellular toxins [52] and other cellular components likely increase in the surrounding waters near the end of a ciliate bloom when cells may be experiencing aging and membrane permeability, parasitic lysis, or cell death.”

Even though the experiment is conducted properly, the interpretation of the results is not clear or even misleading in some aspects. The description of prey lysate is important, however not interpreted in the discussion part. The particles that express orange fluorescence are smaller in size of the plastids, that Mesodinium rubrum has. To develop this part more the photographs of Mesodinium cells can be taken and the size of the plastids can be measured or refer to previous publications. However, the explanation is needed what these particles can be if not plastids.

· As recommended by another reviewer as well, these fluorescent pigment results, and their associated methods and discussion text, have been removed from the manuscript. Multiple reviewers found the data regarding fluorescent particle measurements to be preliminary and/or unfocused, and the conclusions derived from them speculative.

· Thank you for pointing out that we should include more information regarding discrimination of bacterial cells from other cells and/or debris.

· As such, we added the sentence to the methods “As DAPI can stain bacterial cells as well asthe nucleic debris fromciliates, bacterial cells were discriminated from debris by their consistent rod-like or catenulate shape.”

· As recommended, we changed the SE bars to SD bars.

· As recommended, we decreased the size of the markers.

In the prey+lysate3000 group D.acuminata run out of prey on the 12^th day of culturing. This fact is wrongly described in the text lines 112-114 “ciliates were completely consumed by day 15-18 in the three treatments to which they were added: Preyciliate1500, Preyciliate3000, and Prey+Lysateciliate3000”.

· Thank you for pointing out this inconsistency. Changed as recommended.

In the group Prey ciliate3000 the prey was depleted on 15^th day. So the group prey+lysate3000 survived with no prey and grew faster for 3 days. How this can be explained? Can the fact of starvation influence the toxin dynamic? Refer to the recent paper Garcia-Potela et al., Marine drugs 2018.

· We agree with the reviewer that it is interesting that the mixed treatment (prey+lysate3000) consumed all their prey first, 5 days sooner than the treatment with the same number of live prey (prey1500). This was not necessarily expected, and should be a new line of research; more treatments and controls (e.g, ciliates at different initial cell concentrations with different levels of lysate) would need to be included to answer this question. This was not the goal of this manuscript, and so we suggest in the abstract that it should be a future research effort.

Authors claim that "The addition of ciliate lysate to co-incubations also resulted in maximum toxin quotas and extracellular concentrations of okadaic acid and dinophysistoxin-1” true only in part. Authors compare the toxin quotas to the initial level of toxins, which is usually low in the beginning of culturing and increases toward the plateau phase. If authors compare toxin levels of prey+lysate3000 group with prey3000 group they will find no significant differences in DTX-1, PTX-2 intracellular concentration and extracellular concentration of all the toxins (Figure 2). So only OA concentration seems to increase significantly in these two group, so we cannot confidently conclude that addition of lysate results in increase of all toxins.

· We identified that the reviewer is referring to a summary sentence in the abstract that refers to only OA and DTX1. We agree that the lysate treatment did not increase PTX2 values and never meant to imply that this amendment increased all toxins, but we recognize that PTX2 is excluded from the abstract and this can cause confusion to readers. To make this clearer, we have added a sentence describing PTX2 results to the abstract.

· “Pectenotoxin-2 values, intracellular or extracellular, did not show a clear trend across the treatments.”

· Thank you for also pointing out that our conclusion was not clear without indicating the significance between treatments on the bars in Fig.2. We have added that information (See next comment and response).

On figure 2 the asterisks indicating significance of the measurement in comparison with the initial level, should be substituted to the asterisks comparing groups between each other to make the conclusion about influence of lysate. It is also more informative to use SD instead of SE for error bars.

· As recommended, we changed the SE bars to SD bars.

· As recommended, we added lowercase letters to depict significant differences in toxin levels between treatments. We also kept the asterisks in the figure to show toxin increase from initial levels.

· To accompany the addition of treatment comparison to Fig. 2, we added a paragraph to the results section:

· “Total toxin concentrations, toxin quotas, and extracellular concentrations were also compared across treatments during plateau phase to look for effect of the different organic amendments (Fig. 2). Overall, the total toxin concentrations of OA and DTX1 were significantly greater in the Prey+Lysateciliate3000 treatment, as compared to the treatments with only live ciliates added (Preyciliate1500, Preyciliate3000). When evaluating intra and extracellular toxins separately, the same trend remained: the mean OA toxin quota and DTX1 extracellular concentration were significantly greater than treatments with live prey. Okadaic acid extracellular concentrations (Fig. 2d) and DTX1 intracellular toxin quotas (Fig. 2b), however, were indistinguishable between the treatments with lysate (Prey+Lysateciliate3000) and the higher prey abundance (Preyciliate3000). In all cases, Prey+Lysateciliate3000 and the higher prey abundance Preyciliate3000 contained more OA or DTX1 in the intra or extra cellular fractions than the treatment with the lowest prey abundance (Preyciliate1500). Pectenotoxin-2 levels did not follow a distinguishable trend across treatments or toxin measurements, but values were more than a magnitude greater than OA and DTX1 combined.”

In general this experiment is sound and shows important information about feeding behavior of Dinophysis acuminata. The presented data is very good as a preliminary result which can be developed to a full complete study, if other experiments intended to show the mechanism of increased growth rates are performed. The additional experiment and observation can be: Feeding the Dinophysis with specific fraction of lysate for example only plastids fraction, which can be characterized by certain size (plastids are bigger than 1.3 um as the size of fluorescent particles in presented study) and fluorescent pigments. The isolation of the fraction can be done using sorting technique based on fluorescence or on size.

· Thank you for your support of this work. We agree that some new hypotheses are raised by our study, as happens with all “sound experiments,” and as such, we suggest future avenues of research in the discussion.

· As recommended by another reviewer, these fluorescent pigment results, and their associated methods and discussion text, have been removed from the manuscript. Multiple reviewers found the data regarding fluorescent particle measurements to be preliminary and/or unfocused, and the conclusions derived from them speculative.

· We agree that this would be a great new avenue of research based on hypotheses derived from our study, and as such, we suggest this as a future avenue of research in the discussion.

· “As such, interactions between Dinophysisand bacteria still need to be evaluated systematically if the link between the dinoflagellate, ciliate, and/or organic compounds or materials is to be realized.”

· As requested, we have removed speculation from the discussion. This included a whole paragraph focused on our interpretation of the fluorescent particles and their utilization without active kleptoplasty.

Reviewer 4 Report

The purpose of this study is very simple and clear. Experimental design is clear enough. However, frankly speaking I feel likely that this experiment is just repetition done by Nagai et al. (2011). I expected that the authors could provide more detail information on the effect of lysate on their growth and toxin productivity about how Dinophysis can utilize the lysate through the body surface or through the peduncle? This is the missing point of this study.

My concern is that the results of these experiments are changeable and not stable, it means when the authors conduct the same experiments again, if they can obtain the similar results or not. In this part, I want the authors to confirm the reliability, if the results are concrete. Why didn’t the authors utilize the C13 isotope method to clarify how Dinophysis uptake the lysate? Therefore, I was wondering if I should reject this paper or not at the beginning of reading this manuscript. However, overall this manuscript is well written. So, I decided to suggest editors the acceptance of minor revision.

Minor comment

L139, reported -> confirmed is better

Author Response

I feel likely that this experiment is just repetition done by Nagai et al. (2011).

· We thank the Editor and Reviewer 4 for pointing out the need for this clarification. We agree that our explanation of how this work is different from Nagai et al. (2011) was originally lacking. We have, as such, added a whole new paragraph to the introduction to show the clear distinction between these two studies, but also how our study builds on their valuable research (Lines 76-95). Please see our response to the editor’s comments for our full response.

I expected that the authors could provide more detail information on the effect of lysate on their growth and toxin productivity about how Dinophysis can utilize the lysate through the body surface or through the peduncle?

· We had originally attempted to address this more mechanistic question by considering the direct uptake of fluorescent particles to support Dinophysis growth, photosynthesis, and toxin production. These data, however, were considered by a reviewer to be too preliminary to be included in the study, and as such have been removed.

· We understand your concern, however, the fact that toxin values increased in the presence of lysate in two studies (ours and Nagai et al. 2011) using two different strains of D. acuminatasupported on two different prey lines, shows that this result is repeatable. This agreement in findings is highlighted in the discussion: “These novel findings build upon a previous study by Nagai et al., [18], whereby the authors showed that the addition of lysate alone increased total toxin levels in a monoculture of D. acuminata. While the current study cannot ask the same question, i.e., the lysate-only treatment did not produce enough biomass for toxin testing, it is interesting that both studies reported a toxin-promoting effect of lysate. The consistency in results between studies, two geographical isolates, and different prey lines, lends support to this result’s validity.”

Why didn’t the authors utilize the C13 isotope method to clarify how Dinophysis uptake the lysate?

· Stable isotopes were not considered in this work, but nitrogen and carbon isotopes could be an interesting tool in future studies to identify uptake mechanisms and chemical forms utilized by Dinophysisand the ciliate.

Minor comment:

L139, reported -> confirmed is better

· Toxin data are reported, not confirmed. We did not make the change.

Reviewer 5 Report

A very good publication that only requires minor text editing corrections
please check the italics of the species names
the internal legend of fig 1 could be improved
the growth rate of the control is not mentioned in Table 1 but is given in the abstract

75 mechanism

l 107 127 112 115 118 119 120 121 italics

Author Response

please check the italics of the species names

· Done. Thank you.

the internal legend of fig 1 could be improved

· Done. We modified the Figure legend by breaking it into three panels.

the growth rate of the control is not mentioned in Table 1 but is given in the abstract

· Thank you for pointing this out. We added the value to the results section for consistency. The value is not reported in Table 1 because that column is only for exponential growth rates, and the monoculture did not reach log growth.

L. 75 mechanism

· Done. Thank you.

L. 107 127 112 115 118 119 120 121 italics

· Done.

Round 2

Reviewer 1 Report

In revised manuscript the authors have satisfactorily addressed all issues and suggestions from the review.

Author Response

Thank you

Reviewer 3 Report

It is clear that authors worked on the manuscript and now it contains less speculations, but for my disappointment authors corrected mainly the text and pictures, however the manuscript requires repetition of some experiments and addition of new measurements. Main missing measurements are:

1. The baseline of lysate in the prey1500 and prey3000 treatments. It was pointed out during first round of revision that it is nearly impossible to reach the lysate free conditions in culture and in the environment. Authors agreed, however no alteration were done. The measurement can be done using sorting techniques and measuring the biomass of live cells and debris. The result might then be interpreted as the live/dead prey ration necessary to initiate higher growth and possibly toxin production.

2. The toxin production per day in the prey+lysate3000 treatment in comparison with prey1500 and prey3000 treatments. In the prey+lysate3000 treatment Dinophysis run out of food 3 days earlier than in prey3000 and 6 days earlier that in prey1500. Food depletion was shown to increase the toxin production of Dinophysis (described in Garcia-Potela et al., Marine drugs 2018). Authors should be very careful with the interpretation of their result about lysate influence toxin production as it might be the combined action of lysate and food depletion. As earlier food depletion is almost unavoidable taking into account that Dinophysis growth fasted in the presence of lysate the different feeding scheme may be helpful for the experiment. Adjustment of the prey concentration (several additions during the culturing) to keep predetor+prey ratio similar across the treatments. Alternatively, the measuring of toxin ratios per day while prey is still available for both treatments will be helpful to rule out the influence of starvation.

I would like to strongly point out that without this additional measurements authors conclusions are not fully supported by the results.

Moreover, authors did not make attempts to include experiment that might have explained the observer increase in growth and toxin quotas when Dinophysis is supplied by the mixed treatment (lysate+live prey). Without it the presented results (in case the additional measurements are done) are only good to prove that both lysate and prey are important for survival of Dinophysis, which was separately shown before, however do not contribute to the understanding of reasons of observed phenomenon.

Detailed answers and comments are below italisized and indicated by ***

The idea of the study is not very original because it was shown before that growth and toxin level of Dinophysis depend on the prey.

· The fact that prey have been found to be important to Dinophysis growth and toxin production in multiple field and lab studies (cited in our manuscript) is exactly why this study was carried out. The previous study investigating lysate as a nutrient source did not include prey in any treatments.

· “Additionally, ciliates should be considered in this relationship as laboratory studies [7,34-36] have indicated prey abundance as an important controller of Dinophysisgrowth and/or toxin production, and Dinophysisspp. have been found to bloom immediately after and co-occur with ciliate prey in the field [37], suggesting that factors controlling ciliate abundance and distribution are important to down-stream DSP events. Further research is now needed to better understand how inorganic and organic compounds interact with prey and a variety of Dinophysisspp. and isolates if predictive capabilities are to be developed and management strategies created.”

***This paragraph, in my opinion does not really clarify authors’ idea. Authors could have said it in more concise and clear way: the influence of the prey abundance have been shown as an important controller of Dinophysis growth and/or toxin production [7,34-36] as well as toxin production was shown to be altered by organic compound liberated from the prey [21]. However, the effect of combination of these factors which usually occurs in natural environment on growth and toxin production of D. acuminata has never been tested before this report.

***Yes, exactly! So authors should exclude the influence of lysate in the live prey treatments by calculating the amount of lysate in those treatments and including it as a baseline in the calculations.

***Accepted

· Thank you for pointing out that we should include more information regarding discrimination of bacterial cells from other cells and/or debris.

· As such, we added the sentence to the methods “As DAPI can stain bacterial cells as well as the nucleic debris from ciliates, bacterial cells were discriminated from debris by their consistent rod-like or catenulate shape.”

***The proposed method of discrimination is not very convincing. Staining for specific bacterial markers would be more preferable. However, the concentration of bacteria is decreasing which is in line with the lysing of bacterial cells due to sonication. In fact, I don’t see the reason to include this result in the article at all as in the discussion authors provide information that bacteria might be coming from different cultures and that increase of bacterial abundance may indirectly support the growth of Dinophysis (303-313). As the concentration of bacteria were not measured properly I recommend to avoid including and discussing this result.

· As recommended, we changed the SE bars to SD bars.

· As recommended, we decreased the size of the markers.

***Thank you for the made adjustments, however a bit more work is required. The concentration scales of a and c should be change to start from 100 or include the break of the scale so readers can see the difference between measurement more clear.

In the prey+lysate3000 group D.acuminata run out of prey on the 12th day of culturing. This fact is wrongly described in the text lines 112-114 “ciliates were completely consumed by day 15-18 in the three treatments to which they were added: Preyciliate1500, Preyciliate3000, and Prey+Lysateciliate3000”.

· Thank you for pointing out this inconsistency. Changed as recommended.

In the group Prey ciliate3000 the prey was depleted on 15th day. So the group prey+lysate3000 survived with no prey and grew faster for 3 days. How this can be explained? Can the fact of starvation influence the toxin dynamic? Refer to the recent paper Garcia-Potela et al., Marine drugs 2018.

***The provided answer doesn’t answer the question if 3 more days of prey depletion in the prey+lysate3000 group might have been the reason for this group having maximum toxin quotas, which is one of the main results of the experiment and is claimed by authors to be consequence of added lysate. Need to repeat experiment to see if an early prey depletion is consistent (might be the consequence of faster growth) and to measure toxin concentration several times after prey depletion to identify the toxin production per day in presence and absence of prey. Most likely will be higher in the group where prey is depleted. The conclusion that solely addition of lysate influences toxin quotas is not fully supported by presented data (lines 331-333).

Authors claim that "The addition of ciliate lysate to co-incubations also resulted in maximum toxin quotas and extracellular concentrations of okadaic acid and dinophysistoxin-1” true only in part. Authors compare the toxin quotas to the initial level of toxins, which is usually low in the beginning of culturing and increases toward the plateau phase. If authors compare toxin levels of prey+lysate3000 group with prey3000 group they will find no significant differences in DTX-1, PTX-2 intracellular concentration and extracellular concentration of all the toxins (Figure 2). So only OA concentration seems to increase significantly in these two group, so we cannot confidently conclude that addition of lysate results in increase of all toxins.

· “Pectenotoxin-2 values, intracellular or extracellular, did not show a clear trend across the treatments.”

· As recommended, we changed the SE bars to SD bars.

· To accompany the addition of treatment comparison to Fig. 2, we added a paragraph to the results section:

***Thank you for correcting this result. It is now more clear. The panel 2 now looks a bit overwhelming with all different symbols. The lowercase letters that identify the graph and significant difference are the same, which causes confusion. I strongly recommend removing asterisk that show significant differences with initial level. It is clear from the initial level bar provided on graph. To remove letters that show significant difference between different treatments and show them with asterisk instead.

***Without additional attempt to show the mechanism of observed growth and toxin production the provided results do not explain the observed phenomenon.

***Accepted

· We agree that this would be a great new avenue of research based on hypotheses derived from our study, and as such, we suggest this as a future avenue of research in the discussion.

***Then the discussion about possible influence of bacteria should be removed.

· “As such, interactions between Dinophysisand bacteria still need to be evaluated systematically if the link between the dinoflagellate, ciliate, and/or organic compounds or materials is to be realized.”

***Accepted

I believe that lysate supports the growth and toxin production of Dinophysis when added together with prey in culture and it must be true for the environment as there is always some lysate availiable. I also hope that authors consider the comments and improve their manuscript to provide better supported conclusions and make attemts to explain the mechanism of observed phenomenon.

Best wishes

Author Response

1. The baseline of lysate in the prey1500 and prey3000 treatments. The measurement can be done using sorting techniques and measuring the biomass of live cells and debris.

2. The toxin production per day in the prey+lysate3000 treatment in comparison with prey1500 and prey3000 treatments. Adjustment of the prey concentration (several additions during the culturing) to keep predetor+prey ratio similar across the treatments. Alternatively, the measuring of toxin ratios per day while prey is still available for both treatments will be helpful to rule out the influence of starvation.

I would like to strongly point out that without this additional measurements authors conclusions are not fully supported by the results.

· While we appreciate the reviewer’s suggestion for new experiments targeting more mechanistic aspects of lysate utilization, we feel that the experiments conducted herein sufficiently answered the proposed question of whether lysate from ciliates or cryptophytes could support Dinophysis growth and toxin production. No additional experiments will be conducted; however, we have indicated future avenues of research throughout the document.

· Same comment as above. No additional experiments will be conducted.

As the concentration of bacteria were not measured properly I recommend to avoid including and discussing this result.

· These results have now been removed from the manuscript as recommended.

Fig 1. Thank you for the made adjustments, however a bit more work is required. The concentration scales of a and c should be change to start from 100 or include the break of the scale so readers can see the difference between measurement more clear.

· The additional change to Figure 1 has been made as recommended.

The provided answer doesn’t answer the question if 3 more days of prey depletion in the prey+lysate3000 group might have been the reason for this group having maximum toxin quotas, which is one of the main results of the experiment and is claimed by authors to be consequence of added lysate. Need to repeat experiment to see if an early prey depletion is consistent (might be the consequence of faster growth) and to measure toxin concentration several times after prey depletion to identify the toxin production per day in presence and absence of prey. Most likely will be higher in the group where prey is depleted. The conclusion that solely addition of lysate influences toxin quotas is not fully supported by presented data (lines 331-333).

· We added a sentence to the results, and a sentence to the discussion citing the work by Garcia-Potela et al., Marine drugs 2018. We provided the alternative hypothesis that the earlier depletion of prey may have contributed to the enhanced toxin quotas in the prey+lysate3000 treatment.

· Line 345: “Additionally, the earlier depletion of live ciliates from the Prey+Lysate_ciliate³⁰⁰⁰ treatment (Fig. 1, b) may have contributed to the enhanced toxin quotas in D. acuminata as toxin content has been shown to increase in cultures during the prey-limited phase (52).”

· Line 148: “More specifically, ciliates were completely consumed by days 15-18 in the two treatments to which they were added without lysate: Prey_ciliate¹⁵⁰⁰ and Prey_ciliate³⁰⁰⁰ (Fig. 1). The prey in the Prey+Lysate_ciliate³⁰⁰⁰ treatment were depleted from the co-incubation by day 12, however, transitioning Dinophysis into a prey-limited phase earlier than the other two treatments with live ciliates.”

Fig. 2 Thank you for correcting this result. It is now more clear. I strongly recommend removing asterisk that show significant differences with initial level. It is clear from the initial level bar provided on graph. To remove letters that show significant difference between different treatments and show them with asterisk instead.

· We added the letters to the last revision, at the reviewer’s request, to show significant differences between treatments. These differences cannot be shown using an asterisk as there are instances where some treatments are not different, but others in the same panel that are different. This is better depicted using shared and unshared lower case letters, respectively. No change was made.

Without additional attempt to show the mechanism of observed growth and toxin production the provided results do not explain the observed phenomenon.

· The goal of this work was not to characterize the mechanism, but to see if lysate utilization was possible in the presence and absence of live prey.

The discussion about possible influence of bacteria should be removed.

· Same comment as above. Text was removed as recommended.

Round 3

Reviewer 3 Report

1. The baseline of lysate in the prey1500 and prey3000 treatments. The measurement can be done using sorting techniques and measuring the biomass of live cells and debris.

I would like to strongly point out that without this additional measurements authors conclusions are not fully supported by the results.

***I am sorry, I can not accept it as a satisfying answer. The first suggested measurement, if done would show that probably even less than half of prey lysate (mass-to-mass ratio) is necessary to increase Dinophysis growth rate but if not done it does not change the significance of your result. The second measurement on the other hand is absolutely crucial and since you can not provide it, your experiment doesn't prove that lysate enhance the toxin production by Dinophysis. Therefore, the manuscript should be seriously re-written. and the part about toxin production should be seriously reconsidered.

· Same comment as above. No additional experiments will be conducted.

*** accepted

As the concentration of bacteria were not measured properly I recommend to avoid including and discussing this result.

· These results have now been removed from the manuscript as recommended.

***accepted

· The additional change to Figure 1 has been made as recommended.

***accepted

***the explanation of the result is now satisfying and takes into account the possibility of prey depletion to influence the toxin concentration. Unfortunatelly as authors are unable to provide additional experiments proving that lysate and not prey limitation enchance the toxins production by Dinophysis. In this light the result about toxin production enhancement now lost its significance and the part about toxin production should be seriously re-written in the manuscript. Therefore, the following changes (but not limited to) have to be made:

"toxin production" should be removed from the title as it is not proven,

Abstract line 17-20 should be paraphrased to add that food depletion might have led to toxin production increase or removed as it is not fully supported by results,

Results part 2.3 should be shortened and to report more concisely the cases when toxin increase was noticed between different treatments. For examples lines 185-189 can be moved to 5.4 toxin analysis, lines 190-199 can be omitted without losing much information and so on,

Discussion lines 256-262 all information about toxin production should be paraphrased to add that food depletion might have led to toxin production increase or removed as it is not fully supported by results (line 256 - "OA toxin quotas, DTX1 extracellular toxin concentrations, and DSP total toxins", line 258 - "toxigenicity", line 262 - "toxin production by the dinoflagellate")

Discussion part 3.2 Ciliate Lysate in Support of Toxin Production should be re-written as it is not completely proven by conducted experiment.

Conclusion line 361 "DSP toxin metrics" should be removed to avoid misleading interpretation of the results.

***I see your point, it is okay then.

Without additional attempt to show the mechanism of observed growth and toxin production the provided results do not explain the observed phenomenon.

· The goal of this work was not to characterize the mechanism, but to see if lysate utilization was possible in the presence and absence of live prey.

***accepted

The discussion about possible influence of bacteria should be removed.

· Same comment as above. Text was removed as recommended.

***304-314 is too long, better to shorten it to 1-2 sentences.