Enrichment Environment Positively Influences Depression- and Anxiety-Like Behavior in Serotonin Transporter Knockout Rats through the Modulation of Neuroplasticity, Spine, and GABAergic Markers

Round 1

Reviewer 1 Report

The manuscript by Sbrini et al., describes the impact of environmental enrichment on behaviour and molecular parameters in SERT rats. The results are of interest but a number of issues need to be addressed:

1. How comparable is the EE paradigm to those cited in the introduction and why was this particular paradigm chosen? Additionally, at what point after weaning did the EE begin?
2. Rats were single-housed for 3days for the sucrose preference test during the EE - how stressful was this for the rats and for the groups, especially EE when returning to their home-cages and reforming hierarchies?
3. Many of the findings are not actually significant between control housing and EE group, which reduces their strength. This should at least be discussed since it suggests that the effects are not only an effect of EE.
4. Where were brains collected on day 31 and why did the authors only chose to look at the PFC and not other brain regions of interest?
5. Similarly, why the selected markers and not others that are associated with spines for example?
6. Actin was used as a control in the WB studies, but if the authors expect plasticity changes then there may be changes in actin too. Why was this used and did its expression differ across the groups?
7. Did the authors also assess time spent in the center of the open field?

Author Response

The manuscript by Sbrini et al., describes the impact of environmental enrichment on behaviour and molecular parameters in SERT rats. The results are of interest but a number of issues need to be addressed:

1. How comparable is the EE paradigm to those cited in the introduction and why was this particular paradigm chosen?

Reply: Among the papers cited in the text we employed those reported by Thamizovia e Vanisree.

Indeed, the paradigm they used consisted of 28 days of EE in big cages with games, tunnels, and wheels and the exposure to EE was conducted in adult rats.

Differently, in Gong and colleagues, animals were exposed to almost one month of EE but in a different period of life (postnatal day 10 to adolescence).

Finally, Speisman and colleagues exposed adult or aged rats to 10 weeks of EE. However, access to this positive environment was limited during the day (2-3 hours).

Additionally, at what point after weaning did the EE begin?

In our experimental paradigm, rats were left undisturbed under standardized conditions until adulthood when the EE paradigm started. We now added the age of the animals in the text (around postnatal day 140) (line 92)

2. Rats were single-housed for 3 days for the sucrose preference test during the EE - how stressful was this for the rats and for the groups, especially EE when returning to their home-cages and reforming hierarchies?

Reply: We thank the reviewer for this comment and we changed the text to better explain this point (lines 103-109).

Indeed, to measure the amount of sucrose consumed by each rat, animals were singly housed only for 4 hours a day (not all day long) for 3 consecutive days: 2 days of habituation and 1 test day.

Actually, the sucrose consumption test is widely used in this field of research to assess anhedonia, and to obtain the measures for every single animal we do not have any other option but to isolate the rats. Therefore, to avoid any potential stress deriving by this procedure the other tests were conducted 5 and 7 days after the sucrose consumption test, and the molecular analyses were performed 17 days after this test to allow the “potential” stress-induced effects due to this procedure to resolve.

3. Many of the findings are not actually significant between control housing and EE group, which reduces their strength. This should at least be discussed since it suggests that the effects are not only an effect of EE.

Reply: We thank the reviewer for this question; however, we are not sure we completely understand the point. If you meant that we do not see any difference between SERT^+/+/NE and SERT^+/+/EE, this is not surprising to us. Indeed, our hypothesis is that SERT^-/- rats are more sensitive to external stimuli (both positive and negative), and in line we found that EE normalized the pathological behavior as well as the molecular alterations in SERT^-/- animals while did not alter the results in SERT^+/+. Actually, we think this point is well described in the discussion (see lines 300-307). If instead, you meant that we do not see an effect of the EE as independent factor in the ANOVA for all the markers we measured, this is probably because EE in SERT^-/- normalized the deficits due to the genotype thus resulting in comparable levels of expression when comparing SERT^+/+/EE and SERT^-/-/EE

4. Where were brains collected on day 31 and why did the authors only chose to look at the PFC and not other brain regions of interest?

Reply: As mentioned in lines 131-133, after decapitation, the brain was removed and immediately dissected, frozen on dry ice and stored at -80° for the following molecular analyses.

Actually, we are aware that several brain regions could contribute to the behavioral phenotype observed. However, we decided to focus on the prefrontal cortex to be consistent with our previous data and to try to add information to a field of research we are working on for years. Indeed, in our previous papers, we demonstrated that is the main brain region affected by the massive presence of serotonin in SERT^-/- rats (Guidotti et al, 2012; Calabrese et al, 2013; Brivio et al, 2019; Schipper et al, 2019) but also by low levels of serotonin (Brivio et al, 2018; Sbrini et al, 2020).

5. Similarly, why the selected markers and not others that are associated with spines for example?

Reply: Similarly to the answer above (4), we selected the markers to be investigated in this specific experimental condition based on what we observed in our previous studies (Guidotti et al, 2012; Calabrese et al, 2013; Brivio et al, 2019).

Indeed, our aim here was to investigate if and how environmental enrichment can rescue the behavioral phenotype and which molecular mechanisms may underly this beneficial effect. Consequently, we thought that the best option was to investigate molecular mechanisms we had already associated with the behavioral deficits.

6. Actin was used as a control in the WB studies, but if the authors expect plasticity changes then there may be changes in actin too. Why was this used and did its expression differ across the groups?

Reply: We partially agree with this comment. Indeed, even if it is true that actin is implicated in synaptic plasticity, we used isoform b as housekeeping, which is mostly involved in cell motility (see for example a review of Artman et al., 2014). On the contrary, F-actin and G-actin are more related to synaptic plasticity and they don’t represent good housekeeping in our conditions.

This do not happen for the levels of b actin.

7. Did the authors also assess time spent in the center of the open field?

Reply: We thank the reviewer for this comment. Actually, since it has been demonstrated that EE may affect locomotor activity (Nag et al., 2009 Behavioral Brain Research) the open field test was performed to investigate this parameter. So, we did not measure the time spent in the center (a measure of anxiety in the open field test). To measure anxiety levels, we used instead the elevated plus maze test.

Reviewer 2 Report

In the present manuscript of Sbrini et al., the authors investigated the effects of environmental enrichment (EE) on emotional behavior and biomarkers for plasticity in SERT-KO rats, a well-established model for genetic serotonine transporter (SERT) impairment in humans. The authors found out that EE rescues the behavioral phenotype (anhedonia, increased anxiety) of SERT-KO rats, paralleled by an improvement of several plasticity genes in the frontal cortex (mBDNF, PDS95, CDC42, GAD-65, GAD-67). They deduce from their findings that non-pharmacological measurements could improve the situation of depressive patients with respective genetic risk factors rather than pharmacotherapy with serotonergic agents. This might be an important notion. The study has an excellent design, the methods used are timely and solid, the manuscript is well written and the authors investigate and address an interesting and important question. They may have detected a relevant mechanism that could help to assign patients to specific forms of therapy. However, some open questions have to be addressed to improve the manuscript.

1. The authors do not mention, whether they used male and female rats or only one sex. Neither do they mention n-numbers for their experiments (please adhere to the ARRIVE guidelines). The former is relevant in several aspects. Depression is more common in women. Moreover, male and female rodents express different amounts of BDNF, and this amount (in the different sexes) is also differentially prone to the effects of EE (see a.o. Chourbaji et al., Neuroscience, 2012). These points have to be clarified and to be vigorously discussed. Please mention the sex of the animals in the Abstract.
2. In this respect it seems strange that EE does not increase BDNF levels in wildtype/control rats in the present study, as observed in a plethora of other studies before. If Fig. 2B shows indeed "representative" Western blots, there would be increased EE-induced BDNF levels in the SERT+/+ rats, because the BDNF band is much stronger and the ß-Actin band clearly smaller than for NE rats.
3. What is the threshold value for anhedonia in the sucrose consumption paradigm used by the authors. Specifically: did EE cause anhedonia in control rats?
4. With regard to the translational value of the model: does a SERT-knockout really mimic a short allelic variant in humans? Would it not be better modelled by SERT+/- animals with a potential intermediate phenotype? What is a “constructive” environment for humans with mood disorders (line 291), or in which way is EE an approach to treat major depression?

Author Response

1. The authors do not mention, whether they used male and female rats or only one sex. Neither do they mention n-numbers for their experiments (please adhere to the ARRIVE guidelines). The former is relevant in several aspects. Depression is more common in women. Moreover, male and female rodents express different amounts of BDNF, and this amount (in the different sexes) is also differentially prone to the effects of EE (see a.o. Chourbaji et al., Neuroscience, 2012). These points have to be clarified and to be vigorously discussed. Please mention the sex of the animals in the Abstract.

Reply: We thank the reviewer for this comment. In this study, we used 40 male rats (10 in each experimental group). We now added this information in the abstract and in the method section (lines 19 and 93).

As you already mentioned depression is mostly diffuse in women than in men, so we are aware that employ only males could be considered as a limitation of this work. However, since this work is a follow up of previous studies performed by our group, where male rats were used, we choose male to be consistent and to complete the past work.

However, we now discuss this point in the discussion (lines 309-312).

2. In this respect it seems strange that EE does not increase BDNF levels in wildtype/control rats in the present study, as observed in a plethora of other studies before. If Fig. 2B shows indeed "representative" Western blots, there would be increased EE-induced BDNF levels in the SERT+/+ rats, because the BDNF band is much stronger and the ß-Actin band clearly smaller than for NE rats.

Reply: As you said, the bands reported in the figure are supposed to be representative, while the quantitative effect result derived from the mean of the data obtained from 10 rats belonging to the same experimental group. We agree that maybe the bands chosen are not completely representative, therefore we have changed the figure in the text.

3. What is the threshold value for anhedonia in the sucrose consumption paradigm used by the authors. Specifically: did EE cause anhedonia in control rats?

Reply: We thank the reviewer for this comment. In this work, we considered a rat anhedonic when it did not show any preference for either the water or the sucrose solution resulting in a sucrose preference around 50% (equal amount of water and sucrose consumed). In line, we considered anhedonic SERT^-/-/NE rats that showed a sucrose preference at 55% (an effect that is also statistically significant). On the contrary, SERT^+/+/EE animals preferred the sucrose solution instead of water with a preference at 66% (not statistically significant).

4. With regard to the translational value of the model: does a SERT-knockout really mimic a short allelic variant in humans? Would it not be better modelled by SERT+/- animals with a potential intermediate phenotype?

Reply: we thank the reviewer for this comment, and we agree with the fact that using heterozygous rats could better resemble the human condition of those individuals carrying the short allele of the gene. However, SERT^+/- rats do not normally show a clear and well-defined pathological phenotype at basal level, while we previously demonstrated and confirmed in this work that SERT^-/- resemble different pathological features of depression in resting conditions. In line, we decided to use knockout rats to evaluate the positive impact of the EE on a well-established and solid pathological condition. We agree that further study employing SERT^+/- rats could further support our results.

What is a “constructive” environment for humans with mood disorders (line 291), or in which way is EE an approach to treat major depression?

Reply: Exposure to a EE could be translated in humans as a positive and stimulating condition both in the familiar context as well as at school/work. For example, psychological support with a dynamic component such as psychological group therapy, physical therapy or musicotherapy.

Round 2

Reviewer 2 Report

The authors have meanwhile addressed and answered all concerns and questions I had. Thereby they have improved the manuscript significantly, and I do not have any objections against publication