Anthropogenic stressors impact fish sensory development and survival via thyroid disruption

Larval metamorphosis and recruitment represent critical life-history transitions for most teleost fishes. While the detrimental effects of anthropogenic stressors on the behavior and survival of recruiting fishes are well-documented, the physiological mechanisms that underpin these patterns remain unclear. Here, we use pharmacological treatments to highlight the role that thyroid hormones (TH) play in sensory development and determining anti-predator responses in metamorphosing convict surgeonfish, Acanthurus triostegus. We then show that high doses of a physical stressor (increased temperature of +3 °C) and a chemical stressor (the pesticide chlorpyrifos at 30 µg L−1) induced similar defects by decreasing fish TH levels and affecting their sensory development. Stressor-exposed fish experienced higher predation; however, their ability to avoid predation improved when they received supplemental TH. Our results highlight that two different anthropogenic stressors can affect critical developmental and ecological transitions via the same physiological pathway. This finding provides a unifying mechanism to explain past results and underlines the profound threat anthropogenic stressors pose to fish communities.

This is important since the results for the olfactory system in response to the pesticide are different from that for the eye and the lateral line system. It should be noted that the development of the nose, eye and the lateral line system are relatively gradual through the larval stage of fishes with some morphological transformations that occurr at metamorphosis (enclosure of nose, formation of lateral line canals, both of which are complete in the fishes used in this study, which is interesting). Comparisons with other well studied fish species (ideally coral reef fishes, or other marine fishes with a pelagic larva) would have been helpful in this context. Olfactory -the number of lamellae (please check method used in the literature for counting bilaterally symmetrical lamellae in the rosette; however, this should not affect the trends in the data as presented) is roughly correlated with the number of olfactory neurons that are the receptor cells. This reflects a process of neurogenesis. Histology (tissue sectioning) would have more directly revealed numbers of receptor cells, which would have been more indicative of a process of functional relevance that is being altered by thyroid and stressor treatments. In larval fishes, the olfactory epithelium is exposed on the surface of the "snout" and (typically at metamorphosis), the epithelium sinks into a blind fluid filled sac and nostrils (nares) form in that process. The fact that so many lamellae are present at the time of collection ("settlement stage" larvae, d0) suggests that olfactory development is advanced at this stage compared to other coral reef fishes (see Hu et al. 2019, in J. Fish Biol; Larawork on nose of wrasses -cited in Hu et al.) Eye/retina -maturation of the retina (Number and relative thickness of layers) should be compared to that any well-studied fish, for instance, zebrafish. Comparison to a goatfish (line 351) which is a benthic feeding the fish may not be appropriate when comparing the ventral versus the dorsal sides of the retina. Further, measuring densities at in the peripheral area of the retina may not be accurate for the reason that is stated -cell densities are higher in the central area of the retina -but densities in this area maybe more indicative of visual abilities. The literature should be checked on this point. Lateral line -unlike the measurements taken for the nose and the retina the morphological parameter used to evaluate lateral line maturation (the density of canal pores) is not a function of the process of neurogenesis (see line 81), rather it is a process of enclosure any elaboration of the canal itself. It would have been helpful to know if the canal is found with in the lateral line scales or if the canal is composed only of soft tissue. Thus, maturation in the three sensory organs as defined in the study are not equivalent. This needs to be acknowledged. That being said, the addition of pores is a legitimate measure of the "maturation" of lateral line canal on the trunk, as it would be for canals on the head. 6. More specific issues: Line 108 -change nostril to olfactory organ, and change lateral line to trunk canal and correct this throughout.
Line 114 -what study looked at olfactory and visual cues, and is it known that those cues were olfactory and not gustatory. If this is not known please use "chemical cue" or "chemical choice experiment" and not "olfactory cue" or "olfactory choice experiment." The same arguments can be used for the term "olfactory" on line 432 unless the cited papers have clearly demonstrated that the behavior is guided by olfaction and not gustation. Line 205 & 210 -"anthropogenised" is not a word. Please change this. Line 214 -the argument for the fitness consequences needs better support with citations. Line 218 -please define "replenishment potential". Line 247 -please more precisely define "settlement stage". What was the condition of the other markers of metamorphosis such as meristic counts, presence of scales, and transitioning or juvenile pigmentation. Line 335 -the use of the term histology is correct here, but because tissue sectioning (the more common definition of histology) is not involved please remove the term histology from lines 371 and 398. Line 401 -There are newer reviews of lateral line structure and function that should be cited here. For instance the 2014 volume called "the lateral line system" (Springer).
Reviewer #2: Remarks to the Author: In this manuscript, Besson and colleagues demonstrate that disrupted TH signaling leads to aberrant development of sensory structures in A. triostegus, and that blocking TH signaling impairs the ability of fish to avoid predators. The authors then show that increased temperatures or CPF exposure during metamorphosis lower T3 and T4 levels, impair development of sensory structures and decrease survival in a predation assay. Survival in the predation assay is rescued by treatment with T3, and the authors conclude that TH disruption is the causal mechanism underlying decreased survival in fish exposed to higher temperatures or CPF. Additional data would strengthen this causal link, and the experiment needs an additional control (CT injected with T3). Especially without data showing that the T3 treatment rescues sensory structure development or avoidance behavior, the authors should explore alternative explanations for their increase in survival in T3-rescued fish. However, overall, these findings are novel, timely and important, the manuscript is well written, and the data are presented beautifully.
My specific critiques, in no particular order: Some citations are repeated (9 and 53) Line 81 should include the citation Hu et al 2019 Dev Dyn, which shows that TH is essential for development of the lateral line system in zebrafish. The lateral line results should be compared to the results in Hu et al.
The introduction and discussion would be strengthened by mentioning that aquatic organisms are even less buffered against climate change than terrestrial organisms, citing work including Pinsky et al 2019.
In Fig 2, it is unclear to me why there seems to be a difference in the time spent at stimulus area with no cue between the control and T3 treated fish? Is this apparent difference significant and can it be addressed?
In lines 183-18, the authors say that temperature has a higher impact on T4 levels than CPF, and this is a bit unclear to me. It looks in the figure like the +3°C treatment results in a similar T4 level as CPF30. Do the authors mean that the ratio of T4/T3 is different between the different treatments? My eye is not detecting ratio differences either. Can this be clarified? If the goal is to highlight differences in T3/T4 ratios, it would be very helpful to show a graph with T3/T4 ratios for each individual in each treatment. , the authors say that exposure to both +3°C and CPF5 or CPF30 causes death. I do not see these data presented in the manuscript, and the citation listed (Wu et al 2017) does not contain that experiment. This needs to be clarified. Fig 3F that the survival of fish exposed to high temperatures or to CPF can be rescued by T3 injections. This is a very exciting finding, but it left me wondering about the actual behavior of these rescued fish and the development of their sensory organs. Is it possible to show data for the rescued fish in some or all graphs in Fig 1, 2 and Fig 3 C-D? Without showing these data, it is difficult to establish a mechanistic link between the T3 rescue and the increase in survival. Further, are there data for a T3 treatment on control fish? Do these show any difference in survivorship from euthyroid controls? (Also, do they show differences in development of sensory organs or predator avoidance?) This control is essential in interpreting the findings. My lab has found that hyperthyroid fish are generally hyperactive and tend to swim very fast and erratically (this is an unpublished observation). I wonder if the rescue you see in survival rates is due to the hyperthyroid fish swimming faster and more erratically, rather than actually having properly developed sensory structures and rescued appropriate predator avoidance. Alternatively, is it possible that the T3 treated fish smell or taste different to the predators, and could that explain the decrease in predation? These possibilities should be explicitly tested or at least addressed as possibilities.

-Sarah McMenamin
Reviewer #3: Remarks to the Author: Review of manuscript: Anthropogenic stressors undermine fish sensory development and survival via thyroid disruption. In this manuscript, the authors aim to investigate the physiological mechanism underlying observed effects of anthropogenic stressors on fish behaviour and survival. They tested two stressors: increased temperature and a chemical stressor (chlorpyrifos pesticide), and investigated the effect of these on the metamorphosis stage using the convict surgeonfish (Acanthurus triostegus). Metamorphosis is controlled by thyroid-hormones (TH), wherefore the effects of these stressors on TH levels was investigated. They also investigated the effect of TH levels and the two stressors on predator avoidance behaviour and survival. The authors report that both stressors decreased TH levels and increased predation vulnerability.
The manuscript is impressive in terms of containing a lot of data. The experimental design is thorough, with the authors first investigating sensory organ maturation and the role of TH during metamorphosis (with the TH signal disruption treatment, N3, showing repressed sensory organ maturation), then the effect of TH on behaviour (with N3 treated fish showing no response to chemical or visual cues from a predator, which led to lower survival). These experiments were followed by investigations of the effects of the stressors (separately and combined) on TH signalling and sensory development, and whether TH injections can restore survival rates. Several treatment levels were used, in combination with controls and solvent controls, as well as some experimental sampling during different days of the metamorphosis. This also means that there are many different treatments to keep track of for the reader. The authors did a pretty good job at presenting the data in a clear and concise manner, but due to the high number of treatments it is still sometimes difficult to follow all results.
The experimental design is quite well described, and I commend the authors for being clear, for example on number of replicates removed and which criteria that were used for removing those (e.g. page 18-19, line 544-464, line 480-482)). However, several important method aspects are missing, such as which year each experiment was performed (coupled with a lack of testing for year effect in the statistical analyses), how many in situ cages and aquaria that were used for housing and exposing the fish (coupled with a lack of tank-effect analysis), and in particular method description and data for how the warming and chemical treatments were maintained and measured. I did not find any description or data of analysis of water and fish samples from the chemical treatment to verify that the targeted levels were reached? See specific comments below. Also, the statistical methods used are rather simple, and while complicated statistical tools should not be a target in itself, more modern tools could be more appropriate, for example rather than using non-parametric tests, which was used in the manuscript for certain analyses.
Another general comment is the overall emphasis on the negative effects found. In particular in the title and abstract, but in general the paper gives the impression that these stressors had a negative impact on sensory development and behaviour, period. However most of the effects were seen in the most extreme treatments, i.e. +3°C and the CPF 30 μg/L treatment (e.g. as shown in Fig. 3), while the lower temperature (+1.5°C) and the lower levels of CPF (1 and 5 μg/L) only had effects on some of the measurements, effects that were not consistent. The fact that these treatments had no effect is mentioned in the result section, but largely ignored in the rest of the manuscript. The authors need to make it clear that the lower treatments had none, or a smaller and non-consistent effects.
I also have some concerns regarding the choice of the targeted chemical levels. The chlorpyrifos (CPF) levels used were 1, 5, and 30 μg/L. These levels are argued to be biologically relevant in the discussion (page 8, line 197-199). However, the cited report (reference 45, NRA Review of Chlorpyrifos) describes that chlorpyrifos is an occasional contaminant of surface waters, and that levels are usually below 1 μg/L. These values are from rivers, which would be relevant if tested on a freshwater organism inhabiting such areas. Since this manuscript use a marine species, levels measured in the sea where they occur are needed in order to say what is biologically relevant. If levels are usually below 1 μg/L in rivers, then those values would be much lower in the oceans due to the dilution effect. The report state that only a few high outliers have been measured, where levels reached 25-26 μg/L, this was again in rivers and irrigation drainage. Based on this, I would say that even the lowest level used here is not biologically relevant, and the highest level used is even higher than extreme outliers. The results are still interesting, knowing the effects of high levels of contaminants can be important, for example it is highly likely that levels will be higher in the future. However it should not be called biologically and/or environmentally relevant. Regardless the levels used here, any values of this contaminant measured closer to where the fish were collected, or at least in a coral reef environment, would be much more relevant to cite. The other references used by the authors to justify the levels used are for example reference 46 (Bigot et al. 2016), however they report levels of 0.18-0.54 pg/L, i.e. magnitudes lower than what was used here. Reference 57 (John & Shaike 2015) does not seem to include much data on environmentally relevant levels, but for example discuss toxicity and LC50 values. Reference 13 (Besson et al. 2017) do state that Australian reef surface waters can reach up to 1 μg/L, but I was unable to find the paper that is cited for this value (NRA. in National Registration Authority for Agricultural and veterinary Chemicals 1, 17 2000), meaning that the validity of this data point cannot be checked or confirmed. In relation to this, it could also be discussed how biologically relevant a 36h (page 13, line 308-309) chemical exposure period is?
Overall I think the manuscript has good potential to be of high value to many fields, but these concerns must be clarified first.
Specific comments: Was any data collected blind regarding treatment? I could not find such a statement so I assume it was not. This is fine, sometimes blind data collection is impossible, however this should be stated in the paper. In particular given the choice flume experiment was not video recorded despite only lasting rather short period of time (14 min for the choice flume experiment, out of which 10 min were used for data collection), and the apparent availability of a camera, which was used in the second behavioural experiment testing visual cues (page 19, line 477-478).
Throughout the manuscript, please change the word "olfaction", unless you are certain that this is the sensory system under investigation (which most of the time you are not, since the other sensory systems (chemoreception ant taste) were not blocked). Use the word "chemosensory cues", or similar.

Reviewer #1
General comment: This paper, which follows on prior work by several of the authors, looks at the effect of temperature and a common pesticide on thyroid hormone levels in settlement stage larvae of a coral reef surgeonfish. Treatment with two stressors revealed reduction in thyroid hormone levels. Then effects of high and low thyroid hormone levels (treatments for increased and decreased levels via injection) were shown to have an effect on the "maturation" of three sensory systems -olfactory, visual and lateral line and, in turn, effects on predation vulnerability. The study is well-constructed with appropriate controls (but no shams) and the statistical analysis appears to be appropriate. Use of a field site with the opportunity for in situ grow out of small fishes after treatments is unique.
R: We thank the reviewer for their considered and thorough comments, and are confident that they have greatly helped us to improve the manuscript.
Specific comments: 1. The results on manipulation of thyroid hormone levels on sensory organ maturation are presented first, followed by results on the effect of introduced stressors on thyroid levels. To tell the story in a logical manner, perhaps this sequence can be reversed. R: We appreciate the logic of this suggestion. However, as the discovery that thyroid hormones (TH) play a role in regulating recruitment processes in coral reef fishes is relatively new discovery (Holzer et al. 2017 -eLife 6:e27595), we believe that our message is best conveyed by first examining the role of TH signaling in sensory maturation and survival. We then extend on this to explore whether anthropogenic stressors can disrupt TH signaling with consequences for sensory maturation and survival.
2. In presenting the results and their interpretation, the origin of supporting data from prior studies (e.g., two stressors together caused death, line 187, etc.) need to be more obvious to make the discussion of the data from this study more clear (check throughout). R: We agree and have edited the text accordingly. For example, line 187 (now 231-233) now reads: "Given that stressors are rarely experienced individually, these results highlight the vulnerability of aquatic organism endocrine functions and the risk posed by anthropogenic stressors 7,9 " Concerning the presentation and interpretation of our results, we initially thought that referring to our figures in the Discussion was not necessary. However, we think this is a valuable point, and have followed the reviewer's remark and we have edited the Discussion section so that the origin of supporting data is now clearly referenced (either with a Figure call or a literature reference).

Metamorphosis (morphological transformation), settlement (behavioral transformation) and recruitment
(ecological process) are distinct from one another in fishes, but are often synchronized, or sequential (in rapid succession). However, it appears that these terms are used interchangeably in the manuscript -this needs to be fixed.
R: We thank the reviewer for pointing out that we mistakenly used the terms metamorphosis and recruitment interchangeably in the previous version of the manuscript. We have modified the text so that we now use the term "metamorphosis" to refer exclusively to the developmental process and we use the term "recruitment" to refer exclusively to the ecological process. Regarding "settlement", this term is widely used to refer to the precise period of time when larval fish move from the ocean to the reef (i.e. reef entry or reef colonization) rather than referring to the behavioral transformations that occur at this stage (see Atema et al. 2002 MEPS 241: 151-160;Leis et al. 2002 MEPS 232: 259-268;Wright et al. 2010 Coral Reefs 29: 235-243;Sponaugle et al. 2012 MEPS 453: 201-212). We have made the appropriate changes throughout the manuscript to make sure that the use of the term settlement is consistent with the above-mentioned meaning.
4. The morphology of the sensory organs (eye, nose) and system (lateral line; the organs are the neuromast receptor organs, which are not discussed at all) are described inaccurately or are not sufficiently described.
Please also check all figure captions including captions in the supplementary materials for accuracy of terminology (see below). The olfactory organ is a fluid filled blind sac that contains the ciliated sensory epithelium that forms a rosette comprise of a number of lamellae (feel free to use this text). In the MS -the terms nostril, rosetta are not used correctly.
R: We are very grateful for the detailed comments provided by this reviewer, and we have modified the manuscript (including figures and captions) and supplementary material accordingly. For example, the text in our Methods section that describes the olfactory organ now reads (lines 442-446): "In fish, the olfactory organ is a fluid filled blind sac that contains the ciliated sensory epithelium that forms a rosette comprise of a number of lamellae 44 . In A. triostegus, the left and right olfactory organs can be found in two cavities on the dorsal surface, between the eye and the snout edge, with water circulating in each cavity from the anterior nostril to the posterior nostril ( Supplementary Fig. 3)." Concerning the lateral line and trunk canal, we have added the following information to the Methods regarding the neuromast receptor organs, the trunk canal and its pores (lines 468-482): "The lateral line system enables fish to detect water motions and pressure gradients, such as those caused by other fish (e.g. movement from other fish in the shoal or predator strikes). It is composed of superficial and canal neuromast receptor organs, which are the functional units of the lateral line system and are ciliary sensory structures located either on the skin or embedded in lateral line canals 43 . Canal neuromasts are found in the epithelium lining the bottom of the lateral line canals, and one canal neuromast is usually found between two adjacent canal pores (e.g. on the cranial lateral lines) or at the level of the canal pore (e.g. in trunk canals) 43 . Neuromast maturation and morphogenesis of lateral line canals and their pores initiate in late-stage larvae and continue through metamorphosis 43 . Counting the number of pores on the trunk canal is thus an appropriate way to rapidly characterize the maturation of the lateral line system when one cannot perform more advanced histological analyzes of the neuromasts.
In d0 to d8 A. triostegus at recruitment, a fully formed trunk canal corresponding to a complete arched canal can be observed on each of the fish body flanks ( Supplementary Fig. 4a). At this stage, A.
triostegus also only exhibits very thin calcified vertical plates but no scales 69 , and the trunk canal is therefore only composed of soft tissue ( Supplementary Fig. 4a). Following the same preparation protocol as used for olfactory organs, we investigated the maturation of the lateral line system of A. triostegus by counting the number of pores on the trunk canal (Supplementary Fig. 4a-b)." 5. Line 100 -"nostril lamellae" -use just "lamellae". R: Agreed. We have modified the text accordingly.
6. Line 375 -"each nostril is covered by an olfactory epithelium" should be "each olfactory organ contains an olfactory epithelium.
R: Agreed. This was addressed in response to comment #4 and changed accordingly. 7. Line 379 -"olfactory organ cavity" should be "olfactory organ". R: Agreed. We have modified the text accordingly.

The retina is composed of rods and cones, two types of retinal photoreceptors (not photocones, check throughout).
R: Thank you for this comment. We have replaced "photocones" with "photoreceptors" throughout the text.
9. The portion of the lateral line system examined in the study is limited to the fully-formed trunk canal (found in most post-metamorphic fishes = juveniles). It would have been helpful to know the morphological condition of the lateral line canals on the head as well -were they informative? R: Unfortunately, the lateral line canals on the head were not available for analysis as we used fish heads to study the retina and/or olfactory organ. We have modified our manuscript to fully mention that only trunk canal pores were observed. For example, we have replaced "lateral line pores" with "trunk canal pores" throughout the text. Please see also our response to comment #4 for further relevant text revision.
10. Counting the pores associated with the trunk canal (not a lateral line) is appropriate, however I suggest using a graphic icon that is more stylized. The icon used is not recognizable as a component of the lateral line system and was confusing.
R: The icon used is based directly on the SEM image provided in Fig. 1d. This is also the case for the graphic icon used for the bpc density (based on Fig. 1c) and the lamellae (based on Fig. 1b). Therefore, we cannot think of a better way to highlight this structure, and would prefer to keep the graphic as it is. That being said, we have modified our figure caption to more explicitly highlight the link between Fig. 1a, Fig. 1d and Supplementary R: We agree that we could have been clearer here, and have modified the text to be more precise. For example, lines 105-108 now read: "The olfactory, visual, and mechanosensory organs ( Fig. 1b-d) showed rapid maturation from d0 to d8 in control (CT) individuals, with the development of new lamellae, a 50% increase in bipolar cell (bpc) density in the retina, and a 240% surge in trunk canal pore density, respectively ( Fig. 1e-g)." Concerning the comparison with other species, lines 186-191 now read: "A. triostegus exhibits an enclosed nose ( Supplementary Fig. 3a), well-formed rosette lamellae ( Supplementary Fig. 3b) and a fully formed trunk canal ( Supplementary Fig. 4a) at the time of settlement, but relatively low cell densities in the retina ( Fig. 1F and Supplementary Fig. 6-8)  R: It would be very interesting to disentangle and decipher the mechanisms underlying the effects of thyroid hormones on different sensory systems. While this was not the focus of our study, we agree that this would make for a great (albeit challenging) follow-up study. We have highlighted this in the discussion, and have made comparisons with other coral reef fishes (lines 184-204): "Thyroid hormones are important for regulating sensory development in teleost fishes 11,14,36,38 , and our results provide insights into the role they play in sensory system maturation during recruitment. A. triostegus exhibits an enclosed nose ( Supplementary Fig. 3a), well-formed rosette lamellae Despite this advanced stage of development, we found that fish with pharmacologically promoted TH signaling experienced faster sensory organ maturation, those with disrupted TH signaling (either pharmacologically ( Fig. 1e-g) or environmentally (Fig. 3c,e)) experienced impaired sensory development, and those with pharmacologically disrupted TH that received supplemental T 3 experienced rescued maturation of their olfactory organ ( Supplementary Fig. 1). This development of new lamellae promoted by TH is consistent with neurogenesis 35 , as the number of lamellae roughly correlates with the number of olfactory neurons 44 . In contrast, the maturation of the trunk canal involves both the development of new canal neuromasts (i.e. neurogenesis) and the enclosure and elaboration of the canal itself 43 , which is more consistent with epithelial proliferation. The fact that lamellae development was not affected by increased temperature or CPF ( Fig. 3d) suggests that the endocrine disruption caused by these stressors may not be severe enough (e.g. in comparison with the N3 treatment) to affect this maturation process. These results highlight the sensitivity and complexity of the mechanisms underlying the actions of TH on sensory system maturation, offering an avenue for future research." 15. Histology (tissue sectioning) would have more directly revealed numbers of receptor cells, which would have been more indicative of a process of functional relevance that is being altered by thyroid and stressor treatments. In larval fishes, the olfactory epithelium is exposed on the surface of the "snout" and (typically at metamorphosis), the epithelium sinks into a blind fluid filled sac and nostrils (nares) form in that process.
R: We agree that tissue sectioning could have been a great complementary approach to study the maturation of the olfactory organ. That being said, our results do show that TH signaling affects the development of new lamellae, which we believe is sufficient for this particular study.

R:
We agree that our results should be compared to ecologically similar species, and we would like to highlight that both Upuneus tragula (goatfish) and post-larval/juvenile stage A. triostegus are benthic feeders. While there are differences between the species (e.g. post-larval and juvenile A. triostegus live in rubble-dominated environments, while U. tragula prefer sandy areas), both species are marine, tropical, and reef-associated fishes that present similar life cycles, undergo metamorphosis at recruitment, and spend the majority of their time at the bottom of the water column. We therefore believe that the comparison we make with U. tragula is more relevant than a comparison with Danio rerio (zebrafish) (freshwater species, occupying different habitat, not benthic, does not undergo important metamorphic changes as those undergone by reef fishes at the time of recruitment; typical laboratory model fish).
Regarding the second part of this comment #17, we are aware that cell densities could vary across the retina.
However, our goal was to look at the role of thyroid hormones in the maturation of the retina, and we were only interested in the relative -not absolute -maturation of this organ. That being said, a recent study on Naso brevirostris (another Acanthuridae species) showed very weak, if any, spatial specialization in the retina of settlement-stage individuals (Tettamanti et al. 2019 J. Exp. Biol. 222, jeb209916). This weak pattern of spatial specialization was actually only observed on the temporal/nasal axis, not on the dorsal/ventral axis. Therefore, while we maintain that our methods are appropriate for our aim, we have modified our Methods text to ensure that another reader does not have the same question (lines 426-432): "Also, we only examined the dorsal side (ds) of the retina, as the ventral side (vs) was shown to not undergo maturation at metamorphosis in another coral reef fish species, the goatfish Upeneus tragula 13 "In d0 to d8 A. triostegus at recruitment, a fully formed trunk canal corresponding to a complete arched canal can be observed on each of the fish body flanks (Supplementary Fig. 4a). At this stage, A.
triostegus also only exhibits very thin calcified vertical plates but no scales 69 , and the trunk canal is therefore only composed of soft tissue ( Supplementary Fig. 4a)." We also provide supplementary discussion concerning the maturation of sensory structures and acknowledge that their respective maturation is not equivalent (lines 184-204): "Thyroid hormones are important for regulating sensory development in teleost fishes 11,14,36,38 , and our results provide insights into the role they play in sensory system maturation during recruitment. Despite this advanced stage of development, we found that fish with pharmacologically promoted TH signaling experienced faster sensory organ maturation, those with disrupted TH signaling (either pharmacologically ( Fig. 1e-g) or environmentally (Fig. 3c,e)) experienced impaired sensory development, and those with pharmacologically disrupted TH that received supplemental T 3 experienced rescued maturation of their olfactory organ ( Supplementary Fig. 1). This development of new lamellae promoted by TH is consistent with neurogenesis 35 , as the number of lamellae roughly correlates with the number of olfactory neurons 44 . In contrast, the maturation of the trunk canal involves both the development of new canal neuromasts (i.e. neurogenesis) and the enclosure and elaboration of the canal itself 43 , which is more consistent with epithelial proliferation. The fact that lamellae development was not affected by increased temperature or CPF (Fig. 3d) suggests that the endocrine disruption caused by these stressors may not be severe enough (e.g. in comparison with the N3 treatment) to affect this maturation process. These results highlight the sensitivity and complexity of the mechanisms underlying the actions of TH on sensory system maturation, offering an avenue for future research." 19. Line 108 -change nostril to olfactory organ, and change lateral line to trunk canal and correct this throughout.
R: Changed as requested.

20.
Line 114 -what study looked at olfactory and visual cues, and is it known that those cues were olfactory and not gustatory. If this is not known please use "chemical cue" or "chemical choice experiment" and not "olfactory cue" or "olfactory choice experiment." The same arguments can be used for the term "olfactory" on line 432 unless the cited papers have clearly demonstrated that the behavior is guided by olfaction and not gustation.
R: Good point. We now use "chemical" instead of "olfactory" as suggested.

Line 205 & 210 -"anthropogenised" is not a word. Please change this.
R: Changed as requested, we now use another wording (lines 246-255): "Acute exposure to increased temperatures of +1.5 and +3°C or CPF levels spanning from 1 to 30 µg L -1 therefore reflects the temperature fluctuations 49,50 and potential or future pesticide fluctuations 52 that larval fishes may experience when recruiting to coastal nurseries under high anthropogenic influence.
Larger and older fish are less vulnerable to TH signaling disruption and associated neurological defects than metamorphosing fishes 56 , indicating that exposure to acute stressors may alter fish predator-prey dynamics during this critical temporal window. As small declines in survival during recruitment can have dramatic consequences for population replenishment, our results raise concerns about the future of coastal fish nurseries, which, under the threats of climate change and anthropogenic stressors, could turn into ecological traps 57 ."

Line 214 -the argument for the fitness consequences needs better support with citations.
R: This sentence was intended to summarize our results and therefore does not require external references.
However, we can see that the term fitness was probably not appropriate and we have therefore changed our phrasing (lines 257-259): "Overall, this study highlights that short-term exposure to acute anthropogenic stressors has detrimental consequences for the development and survival of metamorphosing fish by affecting their TH endocrine function." 23. Line 218 -please define "replenishment potential". R: To make our point clearer, we have changed this term to "resilience" (line 261).
24. Line 247 -please more precisely define "settlement stage". What was the condition of the other markers of metamorphosis such as meristic counts, presence of scales, and transitioning or juvenile pigmentation? R: As mentioned in our reply to comment #3, settlement refers to the period when larval fish move from the ocean to the reef (i.e. reef entry or reef colonization). Our study species, A. triostegus, has a transparent body at settlement, and dark horizontal bars appear on the body flanks ~4 hours after settlement (Holzer et al., 2017).
Transparency is therefore a reliable criterion to attribute to settlement-stage in this species (Holzer et al. 2017).
We have added more detail to our Methods section to help any confusion by other readers (lines 299-301): "Settlement-stage A. triostegus (i.e. fully transparent individuals 11 , here define as day 0 (d0) individuals) were collected on the north-east coast of the island (S17°29'49.7362", W149°45'13.899") at night using a crest net 11 as they transitioned from the ocean to the reef." Please see our reply to comment #18 for discussion about the presence of scales.
25. Line 335 -the use of the term histology is correct here, but because tissue sectioning (the more common definition of histology) is not involved please remove the term histology from lines 371 and 398.
R: Agreed and changed accordingly.
26. Line 401 -There are newer reviews of lateral line structure and function that should be cited here. For instance the 2014 volume called "the lateral line system" (Springer).
R: Thank you for this suggestion. This citation has been added.

Reviewer #2
General comment: In this manuscript, Besson  3. The introduction and discussion would be strengthened by mentioning that aquatic organisms are even less buffered against climate change than terrestrial organisms, citing work including Pinsky et al 2019.
R: We have added this reference and made the suggested changes, e.g. in the Introduction (lines 57-59): "The negative impacts of anthropogenic stressors can be felt by all organisms at all life stages; however, aquatic species may be more vulnerable than terrestrial organisms 7 ." Again, in the Discussion (lines 231-233): "Given that stressors are rarely experienced individually, these results highlight the vulnerability of aquatic organism endocrine functions and the risk posed by anthropogenic stressors 7,9 ." 4. In Fig 2, it is unclear to me why there seems to be a difference in the time spent at stimulus area with no cue between the control and T3 treated fish? Is this apparent difference significant and can it be addressed? R: This difference is intriguing, indeed, but not significant (t = 1.982, df = 17.14, P = 0.064). We observed a lot of variability in our flume experiments (e.g. see how the data points range from 0 to 100 on Fig. 2a) and this difference could therefore just be a product of this variation. That being said, this comment raises a good point and we have removed the term "higher" when referring to the chemical preferences from the T3-treated fish.
The text now reads (lines 123-126): "In chemical choice experiments, d2 CT fish showed a clear avoidance of predator-cues (Fig. 2a). This response was similar in T3-treated fish, while N3-treated fish did not discriminate between water sources, similar to d0 fish (Fig. 2a)." 5. In lines 183-18, the authors say that temperature has a higher impact on T4 levels than CPF, and this is a bit unclear to me. It looks in the figure like the +3°C treatment results in a similar T4 level as CPF30. Do the authors mean that the ratio of T4/T3 is different between the different treatments? My eye is not detecting ratio differences either. Can this be clarified? If the goal is to highlight differences in T3/T4 ratios, it would be very helpful to show a graph with T3/T4 ratios for each individual in each treatment.
R: This is a good point, thank you for bringing it up. Initially we only looked at the effects of temperature and CPF on T4 and T3 levels, but not on the T3/T4 ratio. To address this, we have now added an additional analysis and added a supplementary figure (Supplementary Fig. 2). These results show that, in addition to CPF30 and +3.0°C exposures decreasing both T4 and T3 levels (Fig. 3a,b), CPF30 exposed fish also experienced a significantly lower T3/T4 ratio than +3.0°C exposed fish ( Supplementary Fig. 2). This adds support to our previously presented results and more correctly shows that temperature had a greater effect on T4, and CPF on T3. Specifically, we have added the following to the Results (lines 149-151): "Exposures to +3°C or CPF30 were therefore associated with comparably impaired maturation of sensory organs and TH disruption, but CPF30 fish experienced a lower T3/T4 ratio than +3.0°C fish ( Supplementary Fig. 2)." We have also added the following in the Discussion section (lines 222-227): "Temperature had a higher impact on T 4 -levels than CPF (Fig. 3a), as indicated by a higher T 3 /T 4 ratio ( Supplementary Fig. 2). This suggests a greater central effect, most probably at the neuroendocrine level, which may alter thyroid activity and thus explain the decreased levels of T 3 that we observed 45 . In contrast, CPF had a greater effect on T 3 levels than temperature (Fig. 3b), as indicated by a lower T 3 /T 4 ratio ( Supplementary Fig. 2), suggesting a more downstream or peripheral effect, possibly on T 3 metabolism 46,47 ." 6. the  This needs to be clarified.
We agree and we have removed this sentence. The text now reads (lines 233-237): "Given that stressors are rarely experienced individually, these results highlight the vulnerability of aquatic organism endocrine functions and the risk posed by anthropogenic stressors 7,9 ." Fig 3F that the survival of fish exposed to high temperatures or to CPF can be rescued by T3 injections. This is a very exciting finding, but it left me wondering about the actual behavior of these rescued fish and the development of their sensory organs. Is it possible to show data for the rescued fish in some or all graphs in Fig 1, 2 and Fig 3 C R: We thank Dr McMenamin for this valuable point. As there are many relevant aspects in this comment, and we want to make sure that they are addressed properly, we would first like to provide context for our response.

The authors show in
The work that we conducted in this study was hypothesis-driven, and the successive experiments that we conducted provided support for our main hypothesis -anthropogenic stressor induced survival defects in juvenile coral reef fishes are a result of TH disruption. The logic behind the successive components of this study were as follows: 1) We investigated whether, and showed that, TH signaling regulates sensory organ maturation (Fig 1e-g) and affects anti-predator behaviours (e.g. diminished ability to perform in a chemical choice experiment, diminished ability to perform in a visual preference experiment (Fig. 2a,b) and diminished ability to perform in a predation experiment (Fig. 2c)).
2) Based off these results, we then investigated whether, and showed that, temperature and CPF comparably disrupt T3 and T4 levels (Fig 3a,b), caused similar developmental defects of their sensory systems (Fig. 3c-e) and comparably decreased survival (Fig. 3f). We consider these results key to this study, as they show that anthropogenic stressors and pharmacological disruption of TH processes are associated with comparable sensory maturation defects, and that disruption of TH is associated with diminished anti-predator behaviors and a decreased ability of recruiting fish to avoid predation.
To examine the causality of these results, we hypothesized that supplementing exposed fishes with T3 would reverse the negative effects of TH disruption. Given that we worked with a wild species and that our work required collecting larval fishes in situ as they settled to the reef at night, we were inherently limited in the number of individuals that we could collect, and thus needed to prioritize which experiments were most valuable to examine our hypothesis. We therefore decided that, based on the results of our preceding experiments, that examining whether the survival defects that we observed, which were associated with TH disruption, were reversed when supplemental T3 was administered was the most important test. Indeed, it provides support for our hypothesis and demonstrates the survival consequences of TH disruption, and survival is the ultimate ecological output during recruitment. Our results are consistent with this, and we therefore believed that it was reasonable to conclude that our results support the hypothesis that anthropogenic stressorinduced TH disruption is a causal mechanism underlying decreasing sensory system maturation, diminishing ecologically important behaviors and decreasing survival prospects.
This being said, we agree with Dr McMenamin that this point is important, and have taken the following steps to address it: A) Our results already demonstrate that T 3 injections promote sensory organ maturation and predator avoidance in comparison to control fish (Figs 1, 2), that TH disruption in stressor exposed fish is correlated with impaired sensory organ maturation (see Fig. 3a-e) and that stressor exposed fish that receive supplemental T 3 recover predator avoidance abilities in a predation experiment (Fig. 3f). To increase our evidence in support of supplemental T 3 increasing sensory system maturation, we now provide additional data ( Supplementary Fig. 1, which we also include in this reply (figure on the right)) that we had performed on fish that were treated with both N3 and T3 treatments. This experiment was conducted to test whether T 3 and NH3 compete (following Holzer et al. 2017 -eLife 6:e27595), but was not included in the previous version of the manuscript. In this experiment, we compared the number of lamellae at d2 in CT, T3, N3 and N3T3 treated fish, and found that the maturation of the sensory system was rescued in the N3T3 individuals. While these data only concern the nostril lamellae, it does provide further support in favor of supplemental T 3 rescuing sensory organ maturation in TH disrupted fish. To this end, we have added the following to the Results (lines 116-117): "Supplemental T 3 rescued olfactory organ maturation in N3-treated fish ( Supplementary Fig. 1)." We also added the following to the Discussion (lines 191-196): "Despite this advanced stage of development, we found that fish with pharmacologically promoted TH signaling experienced faster sensory organ maturation, those with disrupted TH signaling (either pharmacologically (Fig. 1e-g) or environmentally (Fig. 3c,e)) experienced impaired sensory development, and those with pharmacologically disrupted TH that received supplemental T 3 experienced rescued maturation of their olfactory organ ( Supplementary Fig. 1)." B) While our above response provides further support that sensory maturation is rescued when T 3 is administered, Dr McMenamin's comment does highlight that the manuscript would benefit from a more cautious approach in the language used. Therefore, following her suggestion, we have adjusted the text and highlighted that further investigation into the effects of TH on sensory system maturation should be an important focus for future research. Specifically, we have added the following to the lines 203-204: "These results highlight the sensitivity and complexity of the mechanisms underlying the actions of TH on sensory system maturation, offering an avenue for future research." Further, we have removed the words "the causal mechanism" and have replaced it with "a mechanism" (line 178).
In the abstract, we have also removed "Both stressors decreased fish thyroid-hormone levels, causally impairing sensory development and increasing predation vulnerability" and changed it as follows (making special note to remove the word "causally") (lines 41-45): "We then show that high doses of a physical stressor (increased temperature of +3°C) and a chemical stressor (the pesticide chlorpyrifos at 30 µg L -1 ) induced similar defects by decreasing fish TH levels and impairing their sensory development. Stressor-exposed fish experienced higher predation; however, their ability to avoid predation improved when they received supplemental TH." As suggested by the reviewer, we agree that other processes may also contribute to rescuing the survival capacity of stressor-exposed / T3-treated fish. We have therefore included the potential reasons brought by the reviewer in the discussion of our revised manuscript (lines 211-218): "While we found that T3-treated fish exhibited more rapid sensory organ maturation (Fig. 1e-f) and comparable responses to predator cues to control fish ( Fig. 2a-b), which is consistent with an accelerated sensory development facilitating more effective anti-predator abilities, we cannot rule out the possibility that supplemental T 3 may have also impacted their behavior in other ways. For example, supplemental T 3 may affect swimming behavior (e.g. as suspected in Danio rerio, McMenamin pers. comm.) and might alter the cues emitted by these individuals (e.g. smelling or tasting different to the predators), which may have contributed to their ability to avoid predation (Fig. 2c)."

Reviewer #3:
General comment: In this manuscript, the authors aim to investigate the physiological mechanism underlying observed effects of anthropogenic stressors on fish behaviour and survival. They tested two stressors: increased temperature and a chemical stressor (chlorpyrifos pesticide), and investigated the effect of these on the metamorphosis stage using the convict surgeonfish (Acanthurus triostegus). Metamorphosis is controlled by thyroid-hormones (TH),

wherefore the effects of these stressors on TH levels was investigated. They also investigated the effect of TH levels and the two stressors on predator avoidance behaviour and survival. The authors report that both stressors decreased TH levels and increased predation vulnerability. The manuscript is impressive in terms of containing a lot of data. The experimental design is thorough, with the authors first investigating sensory organ maturation and the role of TH during metamorphosis (with the TH signal disruption treatment, N3, showing repressed sensory organ maturation), then the effect of TH on behaviour (with N3 treated fish showing no response to chemical or visual cues from a predator, which led to lower survival). These experiments were followed by investigations of the effects of the stressors (separately and combined) on TH signalling and sensory development, and whether TH injections can restore survival rates. Several treatment levels were used, in combination with controls and solvent controls, as well as some experimental sampling during different days
of the metamorphosis. This also means that there are many different treatments to keep track of for the reader.
The authors did a pretty good job at presenting the data in a clear and concise manner, but due to the high number of treatments it is still sometimes difficult to follow all results. The experimental design is quite well described, and I commend the authors for being clear, for example on number of replicates removed and which criteria that were used for removing those (e.g. page 18-19, line 544-464, line 480-482)). R: We thank the reviewer for their positive appraisal of our manuscript. The comments provided were very helpful, and we believe that they have increased the overall quality of the manuscript. Regarding the question about whether experiments were conducted across multiple years, we appreciate that this would not have been clear in the previous version of the manuscript, but note that only one experiment occurred across multiple (i.e. two) years and that this was considered in our statistical analyses (as a random effect). This was initially not clearly indicated in our Methods, but is fixed now (lines 587-592): "Gamma generalized linear mixed effect models (GLMEM) were used to assess if anthropogenic stressor exposures influenced TH levels and T 3 /T 4 ratios 78 . TH level or T 3 /T 4 ratios were used as the dependent variable, and replicate was included as a random factor to account for differences in TH levels only due to the two different Cobas analysers that were used in the two different years, as well as for potential differences between years." Further, we investigated T 3 levels in d0 Acanthurus triostegus across two seasons and three lunar phases in 2015. This work reveals that there is no variation in T 3 levels across these seasons and lunar phases and this is why we did not take these parameters into account in this study. We thought this would be relevant to mention this following the reviewer's remark. Here is the figure associated with this result (see below), which has been added as a supplementary figure (Supplementary Fig. 14 Concerning the CPF exposure, the CPF concentrations that are indicated in the manuscript are nominal concentrations. We did not measure CPF concentrations in the water, but a previous study (Botté et al. 2012 -Mar. Pollut. Bull. 65, 384-393) using similar methods measured 3.53, 13.9 and 42.7 µg L -1 of CPF in seawater estimated that approximately 80% of nominal concentrations were measured after 24 hours, and we therefore expected a similar stability in our study. We have edited our Methods to more make this clearer (lines 360-374): "For CPF exposure, five different treatments were applied: unaltered seawater (CT, control treatment), seawater with acetone at a final concentration of 1:1.000.000 (CPF0, solvent control treatment, as CPF was made soluble using acetone), or seawater with CPF at a nominal concentration of either 1, 5, or 30 μg L -1 (CPF1, CPF5, and CPF30 treatments), based on the findings of recent studies of reef fishes exposed to CPF 11,14,65 . CPF was spiked in each tank from dilutions that were prepared in advance: 1 µg µL -1 , 5 µg µL -1 , and 30 µg µL -1 . From these dilutions, 12 µL were pipetted and spiked in the 12 L exposure tanks, therefore reaching nominal concentrations of 1 µg L -1 , 5 µg L -1 , and 30 µg L -1 .
Similarly, 12 µL of acetone was spiked in the tank for the CPF0 condition. Spike was allowed to mix for 2 minutes (water mixing due to the air stone) before fish were introduced in the tank. At the end of the 32-hour exposure, CPF concentrations in the water or in the fish tissues were not evaluated as we were only interested in the effects of CPF spikes on fish metamorphic processes. Nevertheless, a previous study using similar methods and nominal concentrations of similar magnitude (i.e. ranging from 4 to 64 µg L -1 ) measured CPF levels corresponding to 80% of nominal concentrations after 24 hours 66 , therefore suggesting a good stability of CPF levels in the condition of our study." 4. Also, the statistical methods used are rather simple, and while complicated statistical tools should not be a target in itself, more modern tools could be more appropriate, for example rather than using non-parametric tests, which was used in the manuscript for certain analyses. (GLMEM) were used to assess if anthropogenic stressor exposures influenced TH levels and T 3 /T 4 ratios 78 . TH level or T 3 /T 4 ratios were used as the dependent variable, and replicate was included as a random factor to account for differences in TH levels only due to the two different Cobas analysers that were used in the two different years, as well as for potential differences between years. As preliminary experiments provided no evidence that season and lunar phase affected T 3 levels in metamorphosing A.
triostegus, we did not include them in our analyses (Gamma GLMEM, Supplementary Fig. 14). For each model, diagnostic plots were examined and outputs compared to raw data to confirm goodness-offit and residual homoscedasticity, and, when applicable, residual normality was assessed using Shapiro-Wilk Normality Test. Paired t-tests or Wilcoxon signed rank tests were used to assess whether fish spent more time in the no cue choice area vs predator cue choice area, depending on residual normality (Shapiro-Wilk Normality Test)"

Another general comment is the overall emphasis on the negative effects found. In particular in the title and
abstract, but in general the paper gives the impression that these stressors had a negative impact on sensory development and behaviour, period. However most of the effects were seen in the most extreme treatments, i.e.
+3°C and the CPF 30 μg/L treatment (e.g. as shown in Fig. 3), while the lower temperature (+1.5°C) and the lower levels of CPF (1 and 5 μg/L) only had effects on some of the measurements, effects that were not consistent. The fact that these treatments had no effect is mentioned in the result section, but largely ignored in the rest of the manuscript. The authors need to make it clear that the lower treatments had none, or a smaller and non-consistent effect.
R: Agreed. We have placed greater emphasis on the fact that we observed effects only for high doses of anthropogenic stressors, and that lower exposure levels had apparently none, smaller and non-consistent effects.
First, in the Abstract (lines 41-43): "We then show that high doses of a physical stressor (increased temperature of +3°C) and a chemical stressor (the pesticide chlorpyrifos at 30 µg L -1 ) induced similar defects by decreasing fish TH levels and impairing their sensory development." Then, in the Discussion (lines 220-230): "While exposure to high levels of increased temperature and CPF both inhibit sensory development and reduce a fish's likelihood of avoiding predation, the two stressors affected endocrine signaling differently. Temperature had a higher impact on T 4 -levels than CPF (Fig. 3a), as indicated by a higher T 3 /T 4 ratio ( Supplementary Fig. 2). This suggests a greater central effect, most probably at the neuroendocrine level, which may alter thyroid activity and thus explain the decreased levels of T 3 that we observed 45 . In contrast, CPF had a greater effect on T 3 levels than temperature (Fig. 3b), as indicated by a lower T 3 /T 4 ratio ( Supplementary Fig. 2), suggesting a more downstream or peripheral effect, possibly on T 3 metabolism 46,47 . This variety of action modes may explain the synergistic effect of both stressors on TH levels with co-exposure to 1.5°C or CPF5 causing TH disruption (Fig. 4a-b), while these stressor levels did not affect TH levels when exposed separately (Fig. 3a,b)."

I also have some concerns regarding the choice of the targeted chemical levels. The chlorpyrifos (CPF)
levels used were 1, 5, and 30 μg/L. These levels are argued to be biologically relevant in the discussion (page 8, line 197-199 R: We agree that the discussion of these concentrations was not appropriate. We have substantially modified this discussion paragraph to comply with this remark (lines 238-249): "Under these circumstances, acute fluctuations such as temperature spikes of +1.5°C and +3.0°C are environmentally relevant, and CPF levels from 1 to 30 µg L -1 are informative in the context of decreasing water quality in coastal areas in response to increasing pesticide use and land clearing 51 .
Indeed, rapid temperature shifts are common in coastal surface waters and can reach up to 12°C following solar and tidal forcing 50 . Likewise, while CPF levels in contaminated surface waters are generally below 1 µg L -1 with limited persistence in the water column, these levels can spike up to 26 µg L -1 in rivers 52 . This suggests that following run-offs, and on a short time scale such as the 32-hour exposure of our study, CPF levels in coastal waters could largely exceed the pg/ng per liter concentrations usually reported in seawater 53-55 . Acute exposure to increased temperatures of +1.5 and +3°C or CPF levels spanning from 1 to 30 µg L -1 therefore reflects the temperature fluctuations 49,50 and potential or future pesticide fluctuations 52 that larval fishes may experience when recruiting to coastal nurseries under high anthropogenic influence." We have also removed the John & Shaike 2015 -Environ. Chem. Lett. 13:269-291, and Besson et al., 2017 Sci.
7. In relation to this, it could also be discussed how biologically relevant a 36h (page 13, line 308-309) chemical exposure period is?
R: It is not a matter of duration but rather a matter of sensitivity and vulnerability of this specific temporal window. As indicated in our reply to the previous remark, we have discussed how this acute exposure is relevant in the context of short-term environmental fluctuations. In the results, we emphasize how relevant this short term exposure is in term of the ecology of reef fishes at recruitment (line 120): "As predation is high in the two days following settlement in reef-fishes 16 " This exposure duration is also developmentally relevant, as most metamorphic processes occur within two days post-settlement (see Holzer et al., 2017 -eLife 6:e27595, which we refer to and cite numerous times within our current study). We would like to emphasize that endocrine disruption is common on such short temporal windows: see Newbold 2004 -Toxicol. Appl. Pharmacol. 199: 142-150;Parsons et al., 2019 Aquat. Toxicol. -209:99-112) 8. Overall, I think the manuscript has good potential to be of high value to many fields, but these concerns must be clarified first. Specific comments: Was any data collected blind regarding treatment? I could not find such a statement so I assume it was not. This is fine, sometimes blind data collection is impossible, however this should be stated in the paper. In particular given the choice flume experiment was not video recorded despite only lasting rather short period of time (14 min for the choice flume experiment, out of which 10 min were used for data collection), and the apparent availability of a camera, which was used in the second behavioural experiment testing visual cues (page 19, line 477-478).
R: We agree that we could have been clearer here. The data were not collected blind, which is stated in our NRreporting-summary. We would like to highlight, however, that the choice flume experiment was video recorded.
We have modified our Methods to state this more clearly (lines 523-526): "After releasing the fish in the choice arena, a 2 min acclimation period was observed, then fish position (left or right choice area, or drain area) was recorded every 2 sec for 5 min ( Supplementary Fig. 10), using a camera (GoPro Hero 2) located above the edge of the flume tank." 9. Throughout the manuscript, please change the word "olfaction", unless you are certain that this is the sensory system under investigation (which most of the time you are not, since the other sensory systems (chemoreception ant taste) were not blocked). Use the word "chemosensory cues", or similar.
R: We have removed the word "olfaction" throughout the manuscript (as also requested by reviewer 1). We now use the word "chemical" to refer to the chemical cues presented in the flume experiment, which we now refer to as "chemical choice experiment".
13. Page 13, line 300-303: Please add methods describing how the chemical treatments were obtained (e.g., was the chemical simply mixed into each tank together with solvent?). In particular, give methods for how the target concentrations were measured and controlled. Please include data on actual measured concentrations per tank and treatment. In ecotoxicology it is standard that not only the treatment water concentrations are analysed, but also fish tissue. This was not done here (and the way the methods are written, it is not even given that water concentrations were measured). This means that the actual concentrations in the fish are unknown, and also so in the water?
R: The actual concentrations in fish and in water were, as already mentioned, not measured, but we do not consider this an issue as we were not interested in fish bioaccumulation of CPF but rather to the effects of acute exposure to waterborne CPF onto fish metamorphic process. Also, a previous study demonstrated good stability of CPF in conditions similar to our study (Botté et al. 2012 -Mar. Pollut. Bull. 65, 384-393). Please see our reply to remark #3 from this same reviewer for a detailed answer regarding this. We have also added in our Methods section additional information regarding how CPF spikes were performed (lines 364-369): "CPF was spiked in each tank from dilutions that were prepared in advance: 1 µg µL -1 , 5 µg µL -1 , and 30 µg µL -1 . From these dilutions, 12 µL were pipetted and spiked in the 12 L exposure tanks, therefore reaching nominal concentrations of 1 µg L -1 , 5 µg L -1 , and 30 µg L -1 . Similarly, 12 µL of acetone was spiked in the tank for the CPF0 condition. Spike was allowed to mix for 2 minutes (water mixing due to the air stone) before fish were introduced in the tank."  [236][237][238]. This should at least be discussed.

Page
R: We agree that we could have been clearer here. Flume experiments were conducted at night only for d0 fish.
The reason for this is that d0 fish actively settle to the reef at night (i.e. d0 fish are not diurnal), and this is therefore the most biologically relevant time to do this. We are not concerned about a light pollution issue, as the experiment was conducted under red light. We have now made this clearer in the text (lines 532-534): "d0 fish were tested immediately after collection (i.e. at night) as this is when they are actively moving from the ocean to the reef, and is thus the most biologically relevant time to do so." 16. Page 19, For the second behavioural experiment, testing visual cues, the experiment was video recorded (please add details on camera used, placement of camera etc., and how the videos were analysed, was some software used?) in order to limit disturbance by an observer. However, in the choice flume experiment, an observer was present (page 18, line 453). How did you ensure that the observer did not cause any disturbance in the first experiment, when this was a concern in the second behavioural experiment?
R: We agree that we could have been clearer here. We now provide the requested information about the camera that we used, how it was placed during the experiment, and how videos were analyzed (lines 555-558): "Fish position (i.e. choice area 1, no choice area, choice area 2; Supplementary Fig. 11) was then assessed every two seconds over a 10 min, using a camera (GoPro Hero 2) to limit any external visual disturbances such as an observer's presence. The camera was located above the choice tank." No software was used, as indicated in our NR-reporting-summary. We were not concerned about the observer causing visual disturbance during the flume experiment, as it was conducted in the dark under a red light. This was not the case for the visual choice experiment, which was conducted under lighted conditions, and therefore an observer was not present. R: Please see our replies to remarks #1 and #2, which answer to the same concerns.

Reviewers' Comments:
Reviewer #1: Remarks to the Author: This manuscript looks at the effects of higher temperatures and/or the presence of a chemical stressor on the development of three sensory systems (olfactory organ, retina, lateral line canal on the trunk). As such it is a novel study and especially because it used a coral reef species -experimental work on the early life history stages of coral reef fishes are rare. Further, it looks at environmental stressors and their effects on morphological development, hormonal pathways and behavior -a nice integration! The statistics appear to be done properly, and data is presented carefully in graphic form. with respect to reproducibility, the major barrier would be the availablility of study specimens, which for this study were collected in the field at a remote lab on an island in the Pacific.
The authors carefully considered the comments of the two reviewers of the first submission of this paper and made appropriate changes, which has improved the manuscript. Nevertheless, there are some issues that need to be resolved, with respect to word usage as they relate to the precise descriptions of the variables measured in the various experiments presented, and the presentation of the data in the figures.
The term "undermine" in the title is misleading for the same reason that the use of "impaired" (and "impairment"), "disrupt", "maturation" in other places in the manuscript. It is clear from the data that some aspects of the experimental treatments certainly "affected" the morphological development of the retina, olfactory organ (lamellae), and formation of pores of the lateral line canal on the trunk. And perhaps it is accurate to say that treatments affected the "timing" of the normal development of these structures. However, "maturation" suggests that there is some developmental endpoint (the mature condition), but that is not considered here since data on adults (sexually mature individuals) are not presented. Further, the structure-function relationship for density of bipolar cells in the retina, number of lamellae, and number of lateral line pores in the trunk canal are each unknown (although density of some retinal cells -the photoreceptors -not the bipolar cells, predicts visual acuity)various studies using various species have tried to find functional correlates but failed. The changes observed in this study cannot necessarily be considered "impairments", or the "undermining" of development without knowing the functional correlates of these specific changes. Finally, while the retina and olfactory rosette are composed of sensory receptor cells, the number of lateral line canal pores reflects the development of the non-sensory accessory structures of the lateral line canal (the sensory organs, neuromasts, are within the canal; these are not studied here). Thus, the language used to describe the results of the experiments presented need to be modified so that they are not overstated; such changes in word usage will not change the outcomes of the experiments or the conclusions reached, but will make the report of the outcomes more precise.
That being said, behavioral responses to a visual predator stimulus, and to a chemical predator stimulus ("chemical" not "olfactory"; see caption in Supplemental Figure __) were indeed affected by some aspects of experimental treatments, implying that changes in the timing of sensory development did indeed affect behavior. In addition, overall survival (# of fish surviving; not "survival rate" [line _____]) was affected by experimental treatments. However, in this case, behavior was not specifically observed -although overall behavioral alteration must have resulted in a change in the ability of predators to capture prey [and/or for prey to avoid predators]). Again, the language used to describe the results of the experiments presented need to be modified so that they are not overstated; such changes in word usage will not change the outcomes of the experiments or the conclusions reached, but will make the report of the outcomes more precise.
One issue with behavioral experiment design and data analysis -fish that did not show a preference are important -why were they deleted from the analysis (line 569)?
It should also be mentioned that the auditory system may be affected by the stressors…..It has been shown that larval fish respond to sound, so do not neglect this sensory system, which was not studied here.
More specific comments, for clarity: Line 36/41 -please differentiate between metamorphosis and recruitment -these are two transitions, but "transition" (singular) is used in line 36. Line 45/6 -suggestion -"Our results highlight the fact that two different anthropogenic stressors (physical, chemical) can alter critical developmental (morphological) and ecological transitions via the same hormonal pathway" Line 60 -do you mean "metamorphosis" here? Line 75-76 -I do not understand the logic in this sentence. Please check.
Line 85-87 -please delete "anthropogenic" -this is understood. Line 87 -here and throughout -do not use TH because in different parts of the MS you are talking about one form and in other places you are talking about two forms. Just spell it out. Line 106 -"rapid" -compared to what? Line 108 -"surge" -compared to what? And is it # or density that was measured? Further, the # of pores is initially determined by the number of neuromasts in the canal (one pore on either side of a neuromast). So, is there a maximum number of pores reached during development? Line 112 -"increased rate of maturation" -does a statistical test support this? And please see my comments about "maturation" above. Line 112/113 -pores are not sensory organs -see my comment about this above. Line 116 -did supplemental T3 rescue the retina or LL pores? If not, why not? Line 122 -instead of CT in the text, can you spell out "control"? This would make the text easier to read. Line 130-131 -"survival" -does this refer to the number of fish that survived, or the total % of fish that survived? Please clarify. Line 143 -bpc -would it be possible to spell this out, or put in caps throughout, and in graphs? Every time I see it, it looks like a typo to me. Line 154 -is it survival "rate" or the % of fish that survived? Check usage of "rate" throughout. Sline 164 -change "co-exposed" to "simultaneously exposed" Line 165 -"suboptimal" -how was this determined? Line 172 -"recruiting fishes" -perhaps say "settlement stage fishes"? Line 176 -it is "presumed" anti-predator behaviors -behavior was not measured. Line 178 -"apparently" recovered their ability to avoid….. Line 193 -"Promoted TH signaling", should be "enhanced TH signaling…" Line 196 -"rescued maturation _____ of their olfactory organ…." (but again, consider whether "maturation" should be used). Line 199 -development (not maturation) of the trunk canal does include differentiation of canal neuromasts (not studied here) and enclosure and elaboration of the canal. The epithelial proliferation that occurs, that is relevant here, is the formation and elongation of the pores in the skin covering the lateral line canal. Line 211 -"increased mortality" -OK, but refer to "higher numbers of fish captured by predators", which is presumably due to a change in the developmental timing of the sensory organs. Line 213 -217 -please break into two sentences. Line 261 -delete "detrimental" and add "changes behavior" at the end of the sentence. Line 265 -say "populations and communities". Line 285 -say "metamorphosis is quickly followed by settlement and recruitment" (not "coincides") -this is a good place to reinforce that these are three different processes.
Line 362 -This sentence should be in the Intro to the paper, where the relevance of the stressors are laid out. Line 409 -say "Retina" Line 427 -please spell out "ds" Line 429 -should be "another acanthurid species" Line 432 -"processing efficiency of the retina toward visual cues" -please clarify. Perhaps "the ability of the retina to form an image"? Line 439 -do you mean "thickness" instead of "width"? Line 448 -say "number of folds or lamellae" -to define lamellae. Line 450 -change "circulating" to "moving through each…." Line 454 -shorten to "2.5% glutaraldehyde in 1M sucrose and 0.1M sodium cacodylate (pH 7.4)…". Make same changes in line 456-7. Line 476 -should be "the lateral line system and are cilated sensory organs, composed of hair cells, like those in the inner ear, located…." Line 479 -"at the level of the canal pore" is incorrect -not sure what is being described here. The neuromasts are within the short canal segments between pores.
Line 486 -what are "vertical plates" -are these the canal walls such that the roof of the canal is still just soft tissue? Line 494 -"numbers" -yes, but please check throughout where "density" may have been used. Line 519 -"odor" suggests olfaction. Can you just say "chemical cues from predator"? Line 530 -arena or area -please check throughout and in figures and figure captions (main and supplementary). Line 533 -Not sure what immobility has to do with anything about side preference. Please clarify and check throughout. Line 537 -was this "in the dark with IR light" or was it "in the dark with red light"? Line 552 -should be "in the presence of a visual predator". Line 594 -should be "and the Fulton's K condition factor was used for pharm……treatments." Line 595 -varied with age, not varied across metamorphosis; was d8 post-metamorphic? Line 599 -should be "due to the fact that the analysis was carried out by two different people in two different years".

Figures -
In ALL Graphs -please add the units of the variables on the Y-axis (e.g., # of lamellae, bpc density (#cells/____), # trunk canal pores), etc. In all captions, results of statistical texts with P-values and F values must be added, esp. since these are not in the text. Then they can be eliminated from the figures themselves (use ** instead), which will make them less cluttered. Figure 1 is quite information rich -Can the image be split from the graphs with the addition of the image in Supplementary Fig. 4a -for anatomical context. Similarly -can the image showing the nares in the Supplementary Figure be added here to accompany the rosette for context? These are likely not commonly understood since they are "fish-specific". However, the cross section of the retina should be generally recognizable. Suppl. Fig. 3 -"rosetta" should be "rosette" and add text to say how image b is related to image a (what was dissected away?). "SEM picture" should be "scanning electron micrograph" Suppl. Fig. 4 -Fig. a should be moved to Fig. 1 -see comment above, then this figure can be eliminated, since b is in Fig. 1. Supple Fig. 5 -change "ganglionar" to "ganglionic", "layer width" should be "layer thickness" -see other comment above and check throughout.
Suppl. Fig 10 -should be "chemical" not "olfactory" in caption and figure.
Suppl. Fig. 12 -please add scale bar or mention tank size in caption.
Reviewer #2: Remarks to the Author: All concerns have been adequately addressed. R: We have modified the language used in the manuscript to ensure that it is not misleading nor overstating. We have replaced "undermine" by "impact" in our title (line 2):

Response to Reviewer 1:
"Anthropogenic stressors impact fish sensory development and survival via thyroid disruption" Similarly, we have replaced "impairing" by "affecting" in our abstract (line 43-46): "We then show that high doses of a physical stressor (increased temperature of +3°C) and a chemical stressor (the pesticide chlorpyrifos at 30 µg L -1 ) induced similar defects by decreasing fish TH levels and affecting their sensory development." In the abstract, again, we have replaced "disrupt" by "affect" (lines 47-49): "Our results highlight that two different anthropogenic stressors can affect critical developmental and ecological transitions via the same physiological pathway." We have then replaced "sensory impairment" by "impacts on sensory development" in the final paragraph of the Introduction (lines 94-98): "Here, we use the coral reef-dwelling convict surgeonfish, Acanthurus triostegus, to investigate the importance of the TH endocrine pathway on sensory development during metamorphosis, whether exposure to two distinct anthropogenic stressors (increased temperature and the waterborne organophosphate pesticide chlorpyrifos) can cause TH signaling disruption, and whether the resulting impacts on sensory development increase vulnerability to predation." Again, we have replaced "impair" by "affect" in the title of one subsection of our Result section (line 135): "Increased temperature and chlorpyrifos affect TH levels and sensory development" We have also replaced "impaired" by "affected", "maturation" by "development", "sensory organs" by "sensory structures" and "TH disruption" by "TH levels" in the following sentence of the result section (lines 152-154): "Exposures to +3°C or CPF30 were therefore associated with comparably affected TH levels and development of sensory structures, but CPF30 fish experienced a lower T3/T4 ratio than +3.0°C fish ( Supplementary Fig. 2)." Then, we have also replaced "disrupt TH signaling" by "decrease TH levels" in another Results subsection title (line 165): "Increased temperatures and chlorpyrifos synergistically decrease TH levels" Similarly, we have replaced "maturation" by "development", removed the term "disruptive", replaced "TH disruption" by "decreased TH levels", and replaced "developmental impairments" by "affected sensory development" in the first paragraph of our Discussion section (line 175-178): "Our results highlight the key role an endocrine process (i.e. TH signaling) plays in regulating sensory system development in recruiting fishes, the effect anthropogenic stressors can have on this endocrine function, and the consequences of decreased TH levels and affected sensory development for determining the outcome of predator-prey interactions." In this same paragraph, we have also replaced "impaired sensory organ maturation" by "diminished sensory development" (lines 178-180): "Fish with pharmacologically disrupted TH signaling experienced diminished sensory development and anti-predator behaviors, and were more vulnerable to predation." Again, in this paragraph, we have replaced "disrupted" by "affected" (lines 180-182): "Sensory development and vulnerability to predation were comparably affected by two distinct anthropogenic stressors (increased temperature and CPF)." Later in the discussion, we have replaced "impaired" by "affected" (lines 215-216): "In addition to causing developmental defects, TH signaling disruption also affected the behavior and survivorship of metamorphosing fish." In the discussion, we have also replaced "causing TH disruption" by "decreasing TH levels" (lines 236-239): "This variety of action modes may explain the synergistic effect of both stressors on TH levels with co-exposure to 1.5°C or CPF5 decreasing TH levels ( Fig. 4a-b), while these stressor levels did not affect TH levels when exposed separately (Fig. 3a,b)." According to the reviewer's comment, we have also replaced "maturation" by "development" on 26 occurrences throughout the whole manuscript.
Similarly, we have complied with the reviewer's comment by replacing "sensory organs" by "sensory structures" (6 occurrences throughout the entire manuscript), to encompass both the sensory organs (retina and lamellae) and the sensory accessory structures (trunk canal pores).
Overall, the terms "undermine" and "maturation" are no longer used in the manuscript, and the terms "impair/impairments" and "disrupt/disruptions" are only used when referring to results from other studies or when referring to broader mechanisms/hypotheses, but no longer when describing the precise results of this study. However, in this case, behavior was not specifically observed -although overall behavioral alteration must have resulted in a change in the ability of predators to capture prey [and/or for prey to avoid predators]). Again, the language used to describe the results of the experiments presented need to be modified so that they are not overstated; such changes in word usage will not change the outcomes of the experiments or the conclusions reached, but will make the report of the outcomes more precise.
R: We have removed the terms "rate/rates" as we agree with the reviewer that we are referring to the number of fish surviving. Similarly, we thank the reviewer for pointing the "olfactory" term that we had forgot to replace by "chemical" in the supplementary figure. The change has now been made.
Regarding the reviewer's comment on the potential overstated results of our survival experiments, we would like to precise that we actually only refer to "anti-predator behaviors" in the manuscript when describing/discussing the results of the visual and chemical choice experiments, not the survival experiment. This is particularly evidenced on lines 178-180, as this sentence summarizes the sensory development results, the behavioral results, and the survival results: "Fish with pharmacologically disrupted TH signaling experienced diminished sensory development and anti-predator behaviors, and were more vulnerable to predation" Similarly, lines 216-217 concerns the visual and chemical preference results, not the survival experiments: "Fish that experienced pharmacological TH disruption presented anti-predator behaviors that were comparable to pre-metamorphosed larvae" In contrast, when referring to increased temperatures and chlorpyrifos treatments (for which we did not assess the visual and chemical preferences), we have been very cautious not to mention behavioral alterations as we did not test it (lines 229-230): "exposure to high levels of increased temperature and CPF both inhibit sensory development and reduce a fish's likelihood of avoiding predation" To further emphasis this, we have brought supplemental precisions in the first paragraph of our discussion (lines 180-182): "Sensory development and vulnerability to predation were comparably affected by two distinct anthropogenic stressors (increased temperature and CPF)." Comment 3: One issue with behavioral experiment design and data analysis -fish that did not show a preference are important -why were they deleted from the analysis (line 569)?
R: We agree that fish that did not show a preference are important, and those fish were included in the analysis, as stated in our Methods (lines 549-552): "Fish that did not make a clear choice between the two water sources but spent more than 50% of the time in the choice area were included in the analysis. This was done as we wanted to assess fish preference as well as the absence of preference." Actually, in both the visual and chemical choice experiments, the only fish that were removed from the analyses are the fish that remain immobile (to prevent side bias unrelated to a clear choice made by the fish) or which spent > 50% of the time in the "no choice area" (as our goal was to compare the time spent in the choice area 1 vs the time spent in the choice area 2).
Comment 4: It should also be mentioned that the auditory system may be affected by the stressors…..It has been shown that larval fish respond to sound, so do not neglect this sensory system, which was not studied here.
R: Agreed. In addition to the reference to the effects of sound pollution on the ecology of fish recruitment already made in the introduction, we now also refer to this system in our discussion section (lines 210-213): "These results highlight the sensitivity and complexity of the mechanisms underlying the actions of TH on sensory system development, offering an avenue for future research, in particular on the auditory system, as hearing was not studied here but is also used by larval fish during recruitment and impacted by stressors 23 ." Comment 5: Line 36/41 -please differentiate between metamorphosis and recruitment -these are two transitions, but "transition" (singular) is used in line 36.
R: Agreed. We have changed this sentence (lines 38-39): "Larval metamorphosis and recruitment represent critical life-history transitions for most