Influences of photic stress on postsubicular head‐directional processing

Abstract The stress response serves vital adaptive functions. However, acute stress episodes often negatively impact cognitive processing. Here, we aimed to elucidate whether stress detrimentally affects the head‐direction cells of the postsubiculum, which may in turn impair downstream spatial information processing. We recorded neurons in the rats’ postsubiculum during a pellet‐chasing task during baseline non‐stress conditions and after a 30‐min acute photic stress exposure. Based on their baseline firing rate, we identified a subpopulation of head‐direction cells that drastically decreased its firing rate as a response to stress while preserving their head directionality. The remaining population of head‐direction cells as well as other neurons recorded in the postsubiculum were unaffected. The observed altered activity in the subpopulation might be the basis for spatial processing deficits observed following acute stress episodes.

Specifically, one Reviewer and Section Editor raised serious concerns with the photic stress as model system. At present it is impossible to determine whether the observed effects are due to the stress response, or to the acute photodamage of retinal cells. We thus consider it necessary to provide data that would enable you to distinguish between systemic stress response and the acute retinal photodamage. Reviewer 1 raises issues of animal welfare. Please address each of their points carefully.
We also noted the following points that should be addressed in your revised version. -Ethical approval for the experiments needs to be explicitly stated.
-Include details of the surgical anesthesia & peri-operative care.
-Methods section is quite hard to follow and would benefit from sub-headings. -How were animals euthanised? -Please proof-read reference list to ensure it adheres to to EJN style.
Thank you for submitting your work to EJN.
Kind regards, Paul Bolam & John Foxe co-Editors in Chief, EJN Reviews: Reviewer: 1 (Enrico Alleva, Istituto Superiore di Sanità, Italy) Comments to the Author I only comment two aspects: 1) the bioethical aspects, since rats are in stressing and painful situation, therefore a paragraph detailig the 3Rs principle applied to this experiment is necessary (why uses only 6 animals, eg?) 2) the stress procedure and the stress data discussion. For the latters: lines 26-27, acute photic exposure is a reconisedly standardised procedure? 63: stress regulation, how? Details are necessary, as well more quotations. 83-103: painful and stressing procedures, see above. 153: non normality of the data needs to be measured, not simply assumed. 215 p=0,854 ? 252-3: a paragraph on eustress needs to be added.
Reviewer: 2 (Rafal Czajkowski, Nencki Institute of Experimental Biology, Poland) Comments to the Author The submitted manuscript describes changes in the properties of head direction neurons in the rat postsubiculum upon acute photic stress. The authors provide fairly exhausting comparison of neuronal population consisting of 65 HD cells from 6 animals. The head direction tuning properties remain stable under these conditions, but the average firing rate drops in a subpopulation of recorded cells. The high baseline firing rate is a good predictor of the susceptibility to the stress effect. This is a very interesting finding and the quality of electrophysiological data analysis is more than satisfying. There is one fundamental problem with the design of the experiment. The authors use photic stress as their model. Judging from the introduction and methods, the goal is to achieve a systemic stress response (lines 15-116). Unfortunately this type of stimulus influences not only the systemic pathways, but also one of the main inputs that drive the head direction neurons -the visual system. The Wistar rats are albinotic and their eyes are extremely sensitive to light stimulation. They do not possess melanin and even a relatively small light exposure causes irreversible damage to their retina. The albino rat has been a model system for retinal damage testing, starting from the classical paper by Noell (Noell 1966). A fairly recent review article neatly summarizes these studies (De Vera Mudry 2013). In the light of these findings it is impossible to determine whether the observed effect is due to the stress response, or to the acute photodamage of retinal cells. The authors provide only a two sentence comment to explain this possibility (lines 299-301). Also, they observe an adaptation effect upon consecutive sessions and explain it as adjustment to stress. An obvious alternative explanation is the decreased sensitivity of retina to the stimulus due to photodamage. The electrophysiological data deserve publishing, but the design of the experiments does not provide satisfactory explanation of the observed effect. It is absolutely necessary to perform experiments that would distinguish between systemic stress response and the acute retina photodamage. The very basic control would be the histological comparison between the retina of control animals (a parallel cohort never subjected to photic stress) and the experimental rats. The authors could use different stressors (for example auditory) and compare their effects with the photic stress. Another type of experiment would be to record HD activity in light vs dark conditions in order to identify visually driven HD cells, and then check whether this population is specifically affected by photic stress. If this was the case, the data would nicely support the concept that two populations of HD neurons may exist in parasubiculum, one responsible for the idiothetic input, and the other processing the allothetic stimuli provided by the visual system.

Authors' Response 24 January 2018
Detailed response to the reviewers in blue: Reviewer: 1 Comments to the Author I only comment two aspects: 1) the bioethical aspects, since rats are in stressing and painful situation, therefore a paragraph detailig the 3Rs principle applied to this experiment is necessary (why uses only 6 animals, eg?) 2) the stress procedure and the stress data discussion.
We thank the reviewer for pointing this out and we now provide more details of bioethical aspects which are of course a key element of the procedures. As outlined in more detail (see Reviewer 2 section), our photic light stress was mild with approx. 330 lux of exposure. This was based on previous experiments in-house where we had reliable defecation and urination, partly freezing behavior and erection of the hair and observed cognitive deficits (refinement). As the photic stressor was stressful and as such we aimed to reduce the number of exposures and number of animals while maintaining meaningful results (recorded neurons, and in particular recorded HD cells) to allow valuable statistical analysis. As the prediction of neurons to be recorded per animal cannot be reliably estimated, we recorded and implanted rats sequentially to allow for a better estimation (reduction). Also we increased the tetrode bundles from 4 to 8 during the experiment to allow for a higher yield per animal (reduction & refinement). We consider the model system rat already as a replacement when one wants to study head-directional tuning in the postsubiculum. We have added a paraphrased version as paragraph to the methods section.
For the latters: lines 26-27, acute photic exposure is a reconisedly standardised procedure?
We are uncertain to what specifically the reviewer refers so we try to answer two interpretations: (1) The acute photic stress protocol is not incorporated into the IACUC standard procedures, but we are not aware that any acute stress procedure applied in the literature is. (2)  We have now expanded this section of the discussion and included more references to support that statement. See lines64 to 68 and respective references 83-103: painful and stressing procedures, see above.
153: non normality of the data needs to be measured, not simply assumed.
We apology for the inconsistency, by assuming non-normality based on our sample size we wanted to be conservative per se. In fact, all non-parametric tests were run in Sigmastat and were automatically tested first, for normality which they have failed. To be sure we have rerun all tests again and confirmed the nonnormality of the data and have now included an additional sentence in the Methods section. 215 p=0,854 ?
We thank the reviewer for spotting the mistake and have corrected it. 252-3: a paragraph on eustress needs to be added.
An important point addressed by the reviewer, we have now added a section on the positive effects of stress including additional references.
We hope that the added sections and details are satisfying and we believe they are important and improve the manuscript and the understanding of thereof.
Reviewer: 2 Comments to the Author The submitted manuscript describes changes in the properties of head direction neurons in the rat postsubiculum upon acute photic stress. The authors provide fairly exhausting comparison of neuronal population consisting of 65 HD cells from 6 animals. The head direction tuning properties remain stable under these conditions, but the average firing rate drops in a subpopulation of recorded cells. The high baseline firing rate is a good predictor of the susceptibility to the stress effect. This is a very interesting finding and the quality of electrophysiological data analysis is more than satisfying. There is one fundamental problem with the design of the experiment. The authors use photic stress as their model. Judging from the introduction and methods, the goal is to achieve a systemic stress response (lines 15-116). Unfortunately this type of stimulus influences not only the systemic pathways, but also one of the main inputs that drive the head direction neurons -the visual system. The Wistar rats are albinotic and their eyes are extremely sensitive to light stimulation. They do not possess melanin and even a relatively small light exposure causes irreversible damage to their retina. The albino rat has been a model system for retinal damage testing, starting from the classical paper by Noell (Noell 1966). A fairly recent review article neatly summarizes these studies (De Vera Mudry 2013). In the light of these findings it is impossible to determine whether the observed effect is due to the stress response, or to the acute photodamage of retinal cells. The authors provide only a two sentence comment to explain this possibility (lines 299-301). Also, they observe an adaptation effect upon consecutive sessions and explain it as adjustment to stress. An obvious alternative explanation is the decreased sensitivity of retina to the stimulus due to photodamage. The electrophysiological data deserve publishing, but the design of the experiments does not provide satisfactory explanation of the observed effect. It is absolutely necessary to perform experiments that would distinguish between systemic stress response and the acute retina photodamage.
We thank the reviewer for the words and important comments regarding the photosensitivity of albino rats. We realize we should have provided a far more detailed description of our photic stress procedure and put it in better context as also mentioned by reviewer 1. We have now extended not only the Methods section but also extended the discussion on that point. Importantly our previous light exposure definition was not accurate enough and may have been misleading. We stated 120cd but should have provided either distance or lux in addition. Light intensity as candela was measured on the bucket floor approx. 60 cm. in distance away from the light source resulting in approx. 330 lux. Our lux regime is thus in accordance with procedures employed for anxiolytic tests with similar exposure times of albino models ( The very basic control would be the histological comparison between the retina of control animals (a parallel cohort never subjected to photic stress) and the experimental rats. The authors could use different stressors (for example auditory) and compare their effects with the photic stress.
Another type of experiment would be to record HD activity in light vs dark conditions in order to identify visually driven HD cells, and then check whether this population is specifically affected by photic stress. If this was the case, the data would nicely support the concept that two populations of HD neurons may exist in parasubiculum, one responsible for the idiothetic input, and the other processing the allothetic stimuli provided by the visual system.
In light of our light stress we did and do not anticipate retinal damage and have no convincing evidence that we would have to expect one (see our comments above) as such additional control experiments would not be ethically justified. We acknowledge though the possibility as stated in the original and current extended discussion, that despite the evidence we cannot exclude retinal damage. The proposed experiment regarding the recording of HD activity in light vs. dark conditions would unfortunately not fully answer the questions either. Based on several lines of evidence it appears likely that the far lower head directional firing rates of neurons in the postsubiculum (compared to its HD input structures of the thalamus) present additional information modalities. A darkened state would also remove local cues (borders, walls) and peripheral cues (within the room) which may not be represented by pure visual inputs alone. We acknowledge and state in the manuscript that our conclusion would be stronger if one additionally tests alternative stressors and or stress protocols of varying lengths and strengths; however, we believe our novel data is presented in a precise and understandable manner and provides valuable information to the field.
We hope that our detailed responses are satisfying and provide more details on the procedure and methods applied. We apologize for inconsistencies and missing information in the original text.
2nd Editorial Decision 23 February 2018 Dear Dr. Passecker, Your resubmitted manuscript was reviewed by one of our original external reviewers as well as by the Section Editor, Dr. Marco Capogna, and ourselves. We are pleased to inform you that will be accepted for publication in EJN after a couple of minor revisions.
Please ensure that your reporting of statistical data adheres to EJN guidelines, in particular we request precise values of P rather than inequalities (see EJN 2008EJN , 28, 2363EJN -2364http://onlinelibrary.wiley.com/doi/10.1111/j.1460-9568.2008.06581.x/pdf). I would suggest also to remove 'its' from the title, it reads much better without it.
If you are able to respond fully to the points raised, we would be pleased to receive a revision of your paper within 30 days. Comments to the Author The explanations are satisfactory. Two new issues have been noticed upon second review. There is discrepancy between two descriptions of the photic stress paradigm. The authors state that control animals are kept in the home cage between recording sessions (lines 167, 168), but then when describing the timeline (lines 170, 171) they mention a black bucket.
There are problems with the axis labels in Fig 2 and 4. In Fig. 2, the X axis should be named "Control 1 1st half" and the Y axis "Ctrl 1 2nd half" or something along these lines. Otherwise there is confusion between Fig 2 and Fig 3. In Fig 4 the X and Y axis should be swapped (control session as X axis).