Small, But Mighty: The Role of the Gut Microbiome in Social Behaviors

This week's papers give insight into an emerging field in behavioral neuroscience- the role of the gut microbiome in mental health and social behaviors. Both papers acknowledge that stress can cause an imbalance in the gut microbiome, which can lead to gastrointestinal disorders and altered social behavior. However, these papers found in their own research that an existing imbalance in the gut microbiome can lead to susceptibility to stress. So which is it? This is truly a "chicken or egg" matter. To me, it appears to be a vicious cycle. Luckily, these issues can be attacked at the behavioral and gastrointestinal levels.

Buffington et al found that in their model of maternal high fat diets, reinstating only one bacterial population in the offspring was enough to relieve the neural effects of an imbalanced gut microbiome. If this were applied to humans, could a simple probiotic cocktail resolve dozens of cases of depression, autism, and other mental disorders? It seems far fetched considering that the bacterial treatment did not resolve some symptoms such as the repetitive and persevering behavior of marble burying. Then again, a little extra probiotic couldn't hurt to try. In combination with stress management techniques, could this be a new and highly efficient way to treat mental health disorders?

On a different note, I noticed one thing in the Reber et al paper of particular interest. In the "Stress Promotes Colitogenic Dysbiosis" section, they mention that beta diversity, or diversity between samples, increased with chronic subordinate colony (CSC) housing. They seem to attribute this to the effect of stress, but I thought that it would make more sense that this change in diversity across many sampled time points was actually from being exposed to different mice throughout the behavioral paradigm. Buffington et al says that families that co-habitate are known to share gut microbiota. Although the CSC housing in the Reber et al paper is hardly a loving family, isn't it possible that the mice are still sharing important gut microbiota? Furthermore, the paradigm involves changing the dominant mouse at several intervals so that the experimental mice don't habituate. Wouldn't this change introduce novel gut microbiota to their environment? And could this bring a minor beneficial effect to the experimental animals?

11/28: Buffington et al. and Reber et al.

The MHFD offspring have many qualities of ASD, it shouldn’t be thought of as a complete model of ASD. Although obese mothers are 1.5 times more likely to have an autistic child, not all children of obese mothers are autistic. Furthermore, not all of these children have dysbiosis of gut microbiota. So, this is model is relevant for some cases of ASD, but not all. Also, only the social deficits are rescued through restoring gut microbiota. A different treatment would be required to rescue the repetitive behaviors and anxiety associated with ASD.

Since L. reuteri administration to MHFD offspring can rescue the social deficits in the reciprocal social interaction test and the three chamber test, I was wondering if administering the bacteria to the HFD mothers would change the gut microbiota of the offspring. Since no negative effects were seen when MRD offspring were given L. reuteri, it doesn’t seem like there would be any negative effects for the mothers. It would be interesting to do an experiment comparing the social behaviors and oxytocin levels of offspring of RD mothers, HFD mothers, and HFD mothers that had been administered L. reuteri.

Reber et al. found that there was a decrease in reactive coping measures and anxiety when mice were administered their final m. vaccae immunization one week prior to the beginning of CSC housing. They found similar results when the interval between immunization and the start of CSC housing was two weeks. To see if the results are even more long-lasting, I think that a longer interval of time between immunization and CSC housing would be more convincing. In the introduction, they claim that the effects can last up to 12 weeks after administration, so I was confused about why they only extended the interval by one week for the second test.

When the stress coping behaviors were assessed on day 15, it seemed like the effects of m. vaccae seemed to be wearing off. The dominance status and submissive behavior scores were the same for vehicle and m. vaccae treated CSC mice. This made me think that the effects of the m. vaccae bacteria were not strong enough to rescue mice from the stress they were going through, since the results were similar for the one week and two week injection to CSC interval.

However, these results have good translational value for patients that suffer from colitis and stress-related disorders. M. vaccae administration prevented colitis in stressful situations. It would be viable to administer m. vaccae to a human patient suffering from these symptoms, and there have been other human studies in the past that have had promising results in immunoregulation studies.

11/28 - Microbiota

          I found this week's papers to be the most interesting ones we have read this entire semester. They discuss how critical microorganisms are in immunity, the gut, and stress-related behaviors (Reber et al., 2016), as well as in social behaviors (Buffington et al., 2016). I am currently taking a Microbiology course and obviously I was aware of how important microbes are throughout our human body and lifestyle, but I was not aware they were involved to this extent, how strongly they can affect our social lives in addition to our physical being. One of, if not the most important, aspect of these papers is how much this research can help humans in the clinical setting suffering from not only colon disorders like colitis, but also psychiatric diseases like PTSD and other fear/anxiety illnesses. Since microorganisms are found in all areas of the body, and since the brain has pathways projecting to all areas of the body, I wonder how many other diseases in different areas of the body can be treated by probiotics and how far researchers are in this.
          Both papers left me with a lot of questions regarding how exactly the brain modulates immunity and the gut. That being said, it appears to be extremely complex and thus would have made these papers even longer than they already are, I would like to see a whole nother paper (or few) discussing these processes. Overall, I thought both of them had a very well-rounded approach in testing their hypotheses, although they left me wondering why Reber et al. did not use any control groups that included live Mycobacterium vaccae (they just used heat-killed preparations). They mentioned that heat-killed is involved in dendritic function and anti-inflammatory secretions, but they did not mention any disadvantages (or advantages) of using live preparations.

microbiome - joe

Buffington et al vs. Reber et al

Seminar in BioPsych

Fall 2016

Professor Shansky

The most powerful implication of this week’s papers is the readily translatable conclusions, and moreover the fact that the authors addressed such a prominent issue in Western civilization. Normally, the abstract of papers includes a few sentences about the significance of this research. The great thing about these papers is that they were so different from the rest of papers we have read this semester; they spoke to the changes in Western culture (i.e. increases in obesity and detachment from certain bacterial environments in high-income families) and how those changes directly affected the gut — and indirectly the brain. It’s bizarre to think that the brain receives input from and is regulated by things as far detached as the gut. But indeed, the brain is an amazing piece of machinery.

I appreciated that both papers used very many measures to prove that this manipulation had far-reaching effects. Buffington et al used behavioral (social tests), biochemical (oxytocin and DA level characterization), genetic (gene sequencing for dysbiosis), and electrophysiological (recording firing patterns in reward centers) approaches, while Reber et al used similar approaches along with testing to see if this bacterium had other effects on the colon and immunity in general. This is an impressive set of experiments!

One thing I wondered was why Buffington et al showed that the effect they noticed was limited to reconstitution of live and not heat-killed bacteria, while Reber et al used heat-killed bacteria without addressing the potential effects of using live bacteria instead. I tried googling the difference in using either type of bacteria but I could not find a definitive answer. Furthermore, neither paper addressed the entire communication pathway between the gut and the brain, and so it was difficult for me to ascertain the gut’s modulatory effects on the brain, other than to see the implicit effects it was having on, say, LTP in the VTA (Buffington et al) or biosynthetic markers of 5-HT (Reber et al). The craziest — and most mysterious — element of these studies is that one change in the constitution of the microbiome can affect so many different things in the brain (from biogenic amine synthesis to inflammatory microglial response to firing pattern changes to downstream behavioral changes).

Reber et al remarked, when discussing the preventative effects of M. vaccae on stress-induced colitis, that “…immunization with M. vaccae, or similar bioimmunomodulatory approaches, may be useful for prevention of chronic stress/repeated trauma-induced inflammation and subsequent development of somatic and mental disorders,” to which I reply, “the future!” The papers we have gone through in the past few months have gradually opened my eyes to the magnitude of questions there still are to be addressed regarding the brain and its functions. Specifically, I have found that the multitude of ways that the brain can regulate itself as well as ways the environment can regulate it brings a plethora of questions that will keep the curiosity of scientists fueled for as long as my mind can conceive.

cocaine - joe

Holly et al vs Vassoler et al

Seminar in BioPsych

Fall 2016

Professor Shansky

I thought it was interesting that the intro of Holly et al alluded to a 1995 paper by Haney et al regarding sex differences in cocaine self administration, and the paper ended up following up on it by using different measures to assess sex and stress differences to cocaine administration. Effectively, they asked, “what is happening to the dopamine system to lead to these changes in cocaine use and what about the female biological system leads to higher cocaine administration?” I think that they’re really interesting questions and probe the infrastructure of addiction; however, the finding was modest and did not answer fully any fundamental questions. I thought it was super interesting that DA levels remained strikingly high for so much longer in stressed females. This says that there is something happening to the dopaminergic system in response to stress; however, this response only occurs in females. Isn’t it crazy the differences in phenotype that can occur from one chromosomal change? Even more bizarre is the fact that these differences are highly neglected in research. It was also interesting to see that despite the higher levels of dopamine in stressed females, those animals were also less satiated (as indicated by their long binges of cocaine infusion). Quick note: did the stressed females have a higher number of total infusions because they binged for longer? Or is that adjusted for binge time? Would it more correctly be titled infusion rate then?

Anyway, where is this dopamine going if so much more is needed for them? And is there desensitization occurring more quickly on the post synapse of DA neurons? Is there a broken negative feedback in these stressed mice? Is cocaine not being biotransformed as well in them? I’ve been told many times that good science addresses a question and is able to prompt a bunch of new questions that can be addressed because of that finding. For example, this paper took differences in cocaine administration in stressed vs non stressed and male vs female mice and asked if there might be some dopaminergic differences between them, and whether that precipitates the behavioral changes. One can follow up on this by addressing any of the above questions regarding cocaine drug action or DA modulation.

Regarding the second paper, I think epigenetics is the coolest thing since sliced ham; it adds a very interesting level of complexity to what the genetic code means. I thought it was interesting that they addressed the transgenerational effects of cocaine administration in sires only because of potential confounds of in utero and maternal behavior effects — I think these should both be addressed. I think this especially because they had to account for maternal behavior with their experimental design anyway (because of the differential allocation hypothesis)!

I appreciated that they took into account the potential confound of learning deficits in the offspring with a sucrose learning task. But I also think that it was lucky that the differential male offspring effect that they found was well coupled with the BDNF exon IV transcript expression patterns. But then it was really cool that stopping BDNF signaling in the offspring was sufficient to abolish this tolerance, and furthermore that BDNF increase was accompanied by AcH3 level increases in the sperm of the sires.

11/21 Holly et al. and Vassoler et al.

Holly et al. examined the effects of cocaine in male and female rats after social defeat. In experiment 1, I would have liked to see data presented from more than just 5-10 minutes and 25-30 minutes. I think it would have been more interesting to see when the differences in walk duration occurred across the entire experiment. Also, statistical significance markers seemed confusing and crowded in the figure. It might have looked cleaner if they broke it up into multiple figures. In experiment 2, when tonic dopamine was being analyzed, estrous and non-estrous females weren’t separated. I think that separating the females by estrous cycle would strengthen the data because they talked about how their research is important because not much work has been done in the past about cocaine addiction and circulating levels of estradiol in females.

In the discussion, the authors addressed that the social defeat paradigm in males is different than females because of the body locations that the aggressors attacked. They said that there was no difference in stress response. However, I was wondering if the differences in cocaine response were due to the differences in social defeat. In the future, it would be interesting for another set of experiments like these to be done with a different stressor to see if the differences in response to cocaine were from the differences in social defeat stress.

Vassoler et al. examined the effects of paternal cocaine self-administration on offspring. The day following 60 days of cocaine use, rat fathers mated with females. The researchers investigated the differences in cocaine-sired male and female offspring’s response to cocaine. In the discussion, the authors mention that circulating gonadal hormones, including estrogen, may play a role in the differences. The Holly et al. paper showed that higher levels of estradiol enhance the effects of cocaine, so some of the sex differences found in this experiment may be because of that.

Also, it would be interesting to see what effects paternal cocaine use would have if the fathers had been abstinent from cocaine use for a period of time before mating. This would show if the harmful effects of cocaine on paternal sperm are chronic or acute.

Nature vs. Nuture- Transgenic and Environmental Factors in Cocaine use

 This week's papers address sex differences in both transgenic and environmental factors on cocaine usage. I found it interesting that while one paper found results consistent with human epidemiological studies, the other paper's results were counter intuitive.

Holly et al used episodic social defeat stress to explore the effects of stress on cocaine usage and sensitization. One issue that is addressed with the use of social defeat stress is that male and female aggressors attack in slightly different ways and therefore, the experimental animals may experience different levels of stress. The researchers claim that there were no differences in stress responses such as weight gain, corticosterone levels, and behavioral changes. However, this data was not shown and it appears to be anecdotal. An additional experiment that they could have included would be a social interaction test to see if the experimental rat did in fact learn to be afraid of the aggressor. In order to completely avoid the confound of sex differences in aggressive behavior, they could have also used chronic immobilization stress instead of episodic social defeat stress.

Vassoler et al took a transgenic approach to measuring the effects of cocaine resistance and sensitivity. They specifically focus on a pathway of increased BDNF in the mPFC which is at least partially a result of histone H3 acetylation. These transgenic changes cause male offspring, but not female offspring, to have resistance to cocaine addiction. However, because these rats are not part of a complex society, nor did they experience the environmental factors relevant to humans, the results were counter intuitive to what one may expect. In human epidemiology studies the offspring of cocaine users were actually more likely to become users themselves. I am curious to know if the increased BDNF levels found by Vassoler et al altered dopamine levels in the nucleus accumbens. Because dopamine is highly important for reward mechanisms, I wonder if the increased BDNF in the mPFC causes a decrease in dopamine that would cause cocaine to be less rewarding.

11/21 - Cocaine

          This week's papers discuss how some effects of cocaine differ between male rats and female rats. Holly et al. showed that in all rats, there was a significant increase in dopamine after cocaine administration, with greater levels in stressed rats than in non-stressed rats. However, there was a longer lasting DA increase in female rats. I would have liked to see functional effects of this increased DA with a DA antagonist injection, followed by a behavioral assay such as a locomotor test, which you would hypothesize that the stressed female rats would show the greatest decrease in locomotion compared to controls after a DA antagonist injection. They could perform further experiments to understand why there is a difference between male and female DA levels, as well as interactions between cocaine and estradiol/progesterone. Because cocaine affects females moreso than males, I wonder the current therapies for cocaine addiction in humans are geared more towards females than males. If there are not, then more research should be done on how to treat this in females, because this specificity could be more efficient.

          Vassoler et al. shows that paternal cocaine exposure decreases cocaine exposure in male offspring.  I thought these results were fascinating because I would have thought that offspring of a parent that used cocaine would be at higher risk of using cocaine, but this paper shows the opposite is true (in males, at least). But what about the studies done in humans that show that children of addicted parents have a higher susceptibility to become addicted as well? There must be some risk factors involved that this paper did not look into, such as environmental and other genetic influences. Also, how does the amount of cocaine self-administered by these male rats (about 0.7 mg per kg of body weight per infusion) compare to the high amount humans that are addicted to cocaine take in? Knowing this can be useful for translational studies. That being said, the results of this experiment alone do not appear to be too useful in the clinical setting. It doesn't really explain how someone currently addicted to cocaine can seek effective treatment. And, I highly doubt a man would self-administer himself cocaine so that his future son will inherit a protective effect of cocaine.

environmental enrichment - joe

Ashokan et al vs Lehmann et al

Seminar in BioPsych

Fall 2016

Professor Shansky

There was a stark difference in the delivery of these two papers. I found that Ashokan et al was rather dry (maybe due to the format of the journal — I’m not sure), although the finding was super interesting! Short term environmental enrichment is enough to lead to an inability to experience defeat! Not only is this true behaviorally (with regard to anxiety related and risk assessment behaviors), but they also found that there are distinct morphological changes (in spine density and dendritic morphology) as well as biochemical changes (in BDNF and corticosterone expression) in resilient states. Even still, the article sounded very little like a narrative, as I find scientific papers normally do, and more like a series of statistical findings.

Something interesting that I noted was that the statistical tests were very explicitly highlighted, which manifested that they conducted several statistical tests, of which some produced fruitful (statistically significant) results, while others did not. Why was this necessary? What’s a Sidak’s multiple comparisons test? And why is it important to conduct orthogonal planned comparisons along with ANOVAs? I think the density of statistical jargon clouded the message of the results section. Could this jargon have been shifted to the methods section? That’s where I go when I want to know the nitty gritty anyway.

Furthermore, I think it would’ve been interesting to know whether this short-term period of EE can lead to long-term resilience. Maybe the EE is just a distraction for a little while and once the transient euphoria of good experience fades, the dwindling road to depressive symptomatology ensues. It would have been an interesting question to probe.

Lehmann et al., on the other hand, addressed some really cool aspects of resilience — the induction of FosB as well as the role of the infralimbic cortex (ILCtx) in this phenomenon, a brain region largely implicated in limbic function. At first, they showed that EE indeed leads to a resilient behavioral state (as Ashokan et al. did). They then alluded to a really cool paper from 2010 by Vialou et al regarding the artificial induction of FosB and its antidepressant effects, as well as the physiological patterns of FosB induction after antidepressant administration (really cool paper). They found that environmental enrichment led to an increase in FosB, in line with the previously established role of FosB in resilience. However, they also found an opposite pattern of FosB expression in the HPA axis, which lends to a multifaceted role of FosB in different brain regions.

I found it interesting that they decided to address the question of whether ILCtx is crucial for resilience acquisition or for resilience expression, and more interesting that they found that its role lies in acquisition. FosB is known to degrade more slowly than many other proteins in the brain, and so it might make sense that gradual increase in FosB is concomitant with the gradual learning of resilience. Maybe if you selectively disabled FosB expression in the ILCtx during EE, resilience would not be able to be achieved. That would be a more targeted approach to answering the question of how FosB is important in resilience acquisition by the ILCtx. Furthermore, they showed that the ILCtx is necessary for resilience acquisition; however, it would have been interesting to see whether ILCtx activation is sufficient to enable a resilient phenotype. Considering how equivocal the data on ILCtx involvement in emotional regulation is, I imagine it’s more complicated than: increased ILCtx activity -> increased resilience. Perhaps, the opposite of the above proposed experiment (overexpressing FosB in the ILCtx instead of inactivating it) could provide some insight into the ILCtx’s involvement in the acquisition of a resilient phenotype.

Resilience is such an interesting phenomenon. I really wonder what allows some people to be more resilient to stress than others.

11/17 Environmental Enrichment

          Ashokan et al. (2016) asked whether or not environmental enrichment reduces basolateral amygdala dendritic complexes that were increased by stress and anxiety, as well as stress-induced anxiety behaviors as a result of these molecular changes. Overall I think they were sufficient in supporting their hypothesis, but there are a few aspects of their experiments I would like to criticize. They showed that short-term environmental enrichment was sufficient enough to produce stress resiliency by continuous EE exposure over a period of about 16 days, give or take. In each experiment, the mice were exposed to chronic immobilization stress for only 2 hours a day for the first 10 days. I would've liked to have seen a more diverse set of experimental designs, such as CIS immediately preceding the behavioral test (rather than a few days before), or EE occurring for just a few days before or after exposure to CIS, and even perhaps long-term environmental enrichment paradigms. I found it rather surprising in Figure 5 how stress did not induce a significant change in serum corticosterone levels since it involves excessive secretion of corticosterone (although, there did appear to be greater levels of corticosterone in stressed animals than in control animals in the absence of EE). The fact that EE did not show any significant difference between control and stress mice can hint that EE may not affect corticosterone levels.

          I preferred the experimental design of Lehmann and Herkenham (2011)'s paper more than the previous one. Their designs were more diverse in both housing types (not only did they house mice in enriched and standard environments, but also they housed them in impoverished housing) and when the mice were exposed to the different types of housing (before or after lesions). I also like how they also included a wide variety of brain regions that are involved in abnormal activity during stress-related disorders, and how these brain structures are also found in humans as well. I wonder, since they only used short-term EE, if long-term EE would have the same effect as short-term EE or if the effects would be greater. Maybe that can be a future experiment.

11/17 Ashokan et al. / Lehmann and Herkenham

Ashokan et al. researched the effects of environmental enrichment on reducing the effects of stress through looking at changes dendritic complexity in the BLA and BDNF. They first identified the anxiolytic effects of EE on stressed animals, and then looked for the neuronal factors that caused the rescue effect. I was curious about why the authors mentioned that a running wheel was not provided. A wheel was provided in the Lehmann and Herkenham experiments. It would be interesting to see if their results changed by adding a running wheel.

Ashokan et al. concluded that environmental enrichment, even in a short period of adulthood, can support resilient behavior in mice. In the future, it would be interesting for them to test an enrichment paradigm that started after the stressful event for the research to be more translatable. Then, we would know if providing enrichment after a stressful time in a human’s life would foster resilient behavior in the future. It may be difficult to start enrichment while the human is going through stress.

Lehmann and Herkenham tested how environmental enrichment affects resilience of mice that were subjected to a social defeat paradigm and investigated the circuits involved in this process. From the cellular data, the researchers learned that the IL might be important to resilience. Lehmann and Herkenham ran behavioral tests after lesioning the IL. The EZM test yielded no significant results, and the L/D test showed a significant of infusion and an interaction between lesion and housing. Since these tests were both supposed to measure anxiety, I thought that they should have given similar results, especially since the results were similar when the mice were tested before the IL lesions. So, the lesion might affect a part of anxiety that is only relevant in the L/D test, not EZM.

I liked that it seemed like they focused on making their results as translatable as possible. They looked at brain regions in mice that were homologous to humans with depression, and the paradigm seems like it would be easy to translate humans. However, the lesion time point that I felt would most closely relate to a real world experience was before the mice got environmental enrichment. The results of the L/D, TST, FST, and SI showed that environmental enrichment after the lesion was not effective in rescuing the animals from anxiety or depressive behaviors. A lesion after environmental enrichment, however, did provide rescuing effects on these behaviors. These results suggest that in order for environmental enrichment to be an effective method of creating resilience, the human would have had to experience enrichment prior to a stressful event. This would not be useful when trying to treat a patient that experienced something stressful in the past.

Manufacturing Resiliency: An Investigation in Environmental Enrichment

The papers this week took very different approaches in designing experiments to test whether or not environmental enrichment (EE) can create resiliency to different forms of stress and stress effects.

The Ashokan et. al. paper used chronic immobilization stress (CIS) and countered it with EE. They chose to start the EE exposure at the same time that the CIS started. This kind of approach can be transferable to human models of chronic life stress. This approach would be analogous to an individual who is fortunate to have a wide variety of choice and enriching activities in their lives at the time that they begin to experience chronic stress. Perhaps this would not be representative of someone who comes from a low economic background and cannot afford such activities and might experience chronic stress from working multiple jobs. This group of researchers also made an interesting choice in using head dips as a measure of "active coping" behavior. To me, this seems like they are anthropomorphizing their lab animals to an inappropriate extent. Because of the non-specificity of the head dipping behavior, it seems to me that these animals could likely be exploring their environment.

The Lehmann and Herkenham on the other hand, used social defeat stress and countered it with EE. These researchers also chose to vary the time at which they gave the EE exposure, and the EE exposure never actually overlapped with the stressful experience. Because of this they were simulating a situation more similar to an individual who may have had a great experience a few days before experiencing trauma. However, this experiment is fundamentally different because the stress at hand is acute and not chronic. Although these two groups found that EE had similar effects in creating a resilient phenotype, the mechanism for this process was probably different.

On a final note, neither of these papers considered female animals in their experiments. I found this disappointing (as usual) but especially since both of these papers are fairly recent. At this point we understand that females are just simply different than males and furthermore, that many mental disorders occur more frequently in females than males.

Ayhan et al vs Burrows et al - Joe

Ayhan et al vs Burrows et al

Seminar in BioPsych

Fall 2016

Professor Shansky

Reading these two papers brought something to my attention that I remember being told during my first co-op, and I suppose I hadn’t thought about it thoroughly until now. During my first co-op, the post-doc I worked under once told me that he would never use mouse behavior to study complex mental disorders because it is inherently difficult to create an animal correlate of, say, schizophrenia (he was using electrophysiological techniques to understand how the brain processes innate olfactory valence). I shelved the idea until we delved into our readings about schizophrenia. Last week, and even more so this week, I’ve come to realize that these experiments using mouse models to probe the workings of schizophrenia are simply a small manipulation followed by a selection of tests (behavioral, histological, functional, etc.) to validate the small manipulation. The issue is that there is no standard for tests (each paper contains — in my eyes — an arbitrary selection of tests to attempt to validate the translatability of this model. Furthermore, the interpretations can get iffy because of the vast amount of things that are affected by the manipulation. Here’s what I mean:

Ayhan et al addressed the issue of introducing developmental abnormality to create a model for schizophrenia. It had been known previously that postnatal expression of DISC1 leads to developmental abnormalities resembling schizophrenia, but prenatal expression of DISC1 had not been addressed. While they successfully showed that expression during both time points leads to schizophrenic-like behavior (whatever that means), they did not address the comparability of Pre+Post vs Post groups. In many of the tests, they showed that these two groups showed significantly different responses than just the Pre group or the control did. But for some of the tests, the Pre+Post group was significantly different than the Pre group, indicating that something is going on when DISC1 is expressed during the prenatal period. It could be fruitful to delve deeper into the difference between these two groups. I also found it interesting that while DA levels were considerably lower in all groups exposed to DISC1 expression, DA turnover was not affected (as indicated by comparable DA/DOPAC ratios compared to control). In general, I’m not sure how much insight this paper provided.

Burrows et al, however, I thought was very interesting because it showed that a transgenic mouse that lacked a crucial receptor could show an amelioration in symptomatology with just a fuller and more eventful life. I’ve seen in the past environmental enrichment can have a plethora of positive effects, including neurogenesis and sociability. It was awesome to see that this could occur despite these mice lacking a receptor that’s been implicated in schizophrenia (is mGlu5 dysfunction sufficient to cause schizophrenia? who knows). While I thought it was great that they addressed the issue of male and female differences, I thought that this led to a difficulty to pinpoint definitive interpretations because there were so many groups and each test showed a different selection of groups responding. For example, let’s look at the prepulse inhibition test: there were four groups (SH WT, EE WT, SH KO, EE KO), and 2 sexes (males and females) and two frequencies (30ms ISI and 100ms ISI) — so 16 groups. There were no interactions at all for the 100ms groups, so just 8 groups. For the memory tasks, the mice showed better memory in the Morris water maze but not in the Y-maze (what does that mean? who knows). Moreover, the sex differences were not deeply discussed — just that there may be some differences between the two.

I’m not completely sold by these sets of experiments, but I also must admit that tracking down the etiology of such a complex mental disorder is a gargantuan task and many scientists have made valiant attempts at addressing it.

11/7 Schizophrenia Papers

          As a continuation of the schizophrenia model, this week's papers focused on the involvement of a DISC1 mutant (Ayhan et al., 2011) and a deficiency in metabotropic glutamate receptor 5 (Burrows et al., 2015) in schizophrenia. To begin with Ayhan et al., I thought they overall had a good argument.  I believe they included effective controls throughout their experiment. Since DISC1 is involved in both brain development and adult brain function, it was good that they tested the mutant gene both prenatally and postnatally. In addition, since schizophrenia is so complex and does not target just one neurotransmitter, I thought they did a good job testing the effects of a variety of monoamines that could all have links to schizophrenia. They could have been a little more consistent with including whether they used males or females in their figures, however. In Figure 3, they should have also included experiments done on female frontal cortices, rather than just males, as well as male hippocampuses, rather than just female ones. Finally, since schizophrenia typically affects human adults, this experiment should have been more specific about the postnatal expression in mice and done more tests or used more controls to clarify this.

          My favorite aspect of Burrows et al. was their inclusion of how the environment interacts with genetic predisposition in the schizophrenia model. The fact that environmental enrichment can improve schizophrenia-like behaviors in a variety of different behavioral tasks  (and, in some cases, improve so much that the KO matches the WT like in Figure 2b/c) in mice can be extremely useful in studying the onset, progression, and treatment of schizophrenia in humans. In this paper, they exposed the mice to EE prior to the different behavioral tests, so I wonder if the exact time you are exposed to EE makes a difference on behavior... is it just right before the test or can it be long before? What about exposure after a test, and then a re-test to measure if that had an effect? Maybe they can examine this in a future experiment.
(also, who thought it was a good idea to make the KO EE line in Figure 1 whitish gray? You can't even see it!!)

11/7 Ayhan and Burrows

Ayhan et al. used a genetic approach to model schizophrenia. They used a tTA transgenic mouse expressing the mutant hDISC1 gene. Genetic studies had previously identified DISC1 to be important to schizophrenia in both neurodevelopment and adulthood. In Ayhan et al.’s experiments, mice expressed hDISC1 prenatally, postnatally, pre- and postnatally, or not at all. I liked that they used experimental groups that accounted for problems with the gene that could occur developmentally or during adulthood, when schizophrenia symptoms occur.

Males were used in most experiments, but females were used in some tests, but not all. The tail suspension test and forced swim tests were done only on females. In other tests, like drug induced locomotion and open field interaction test, only males were used. Since they decided to include females in their experiment, it would have been better to see them used in all tests so the results could be more easily compared and interpreted.

Burrows et al. tried to alleviate the effects of knocking out the glutamatergic receptor mGlu5 by providing mice with an enriched environment rather than standard housing. On most of their behavioral tests, they saw that an enriched environment alleviated the schizophrenia related symptoms in mice. They saw many positive results with environmental enrichment, like reduced hyperactivity and less deficit in PPI.  However, environmental enrichment did not improve all aspects of schizophrenia tested; novelty preference in the Y maze was not affected. This research is important because environmental enrichment would be a viable translational approach to treating schizophrenia in humans. Mice began enriched housing at 4 weeks, which is around the time of adolescence in humans. Since symptoms of schizophrenia usually begin to occur shortly after adolescence, the results of this experiment may be useful to humans.

Something that bothered me about the methods was that drug-induced hyperlocomotion preceded prepulse inhibition. Even though the tests were about 3 weeks apart, I was wondering if there were any lasting changes in the brain after the drug was administered that could affect the way mice behaved in PPI.

There's a first time for everything: Exploring new topics in modeling schizophrenia

This week I had some questions and concerns about the papers. Both papers present  ideas that we have not previously encountered in our class discussions thus far. The Ayhan et al paper is the first paper that we have read that both utilized female animals and discussed sex differences in behavior and neuroanatomy. The Burrows et al paper is the first paper that we have read that not only presents a developmental model of schizophrenia, but proposed that an enriched environment might ameliorate schizophrenia related symptoms. 

I found it interesting and slightly confusing that in discussing differences between male and female mice, the Ayhan et al paper refers to these differences as "gender-differences." In other papers I have always seen these discussed as "sex differences." The difference in wording is subtle but was slightly off-putting since gender is a social construct. Because mice don't exist within a higher society, the idea of these differences being because of a socially constructed idea is inaccurate. However, it is probably just a simple mistake in semantics. 

I was also surprised at the way that the Ayhan et. al. paper presented their data. Although they claim to address sex differences, they never present data from male and female animals within the same graph. The only hint that they even completed their experiments in both sexes is hidden in the text. For example, when listing the results of the TST and FST they state that, "expression of mutant hDISC1 had no effects on these behaviors in male mice of either group (data not shown)." Despite the fact that they show that there are no differences between the male experimental groups, the paper still fails to actively compare them to the female groups. Without doing this, stating that there are significant sex differences seems like a bold statement. 

In regards to the Burrows et al paper I was pleasantly surprised to read a paper that proposed a solution that could temper the effects of an aggressive developmental model of schizophrenia. Although the model is by no means all-encompassing of every aspect of schizophrenia, the fact that there is some positive influence from an enriched environment makes me hopeful that there are some preventative measures parents can take in preventing schizophrenia and other mental illnesses. If an enriched environment can help reduce schizophrenia like symptoms in mice, I would be interested in knowing if there are studies in early education in humans and how it relates to schizophrenia later in life.

10/31 Moore et al. and Kellendock et al.

            Moore et al. administered MAM to pregnant rat dams. This model exhibited a neuropathology similar to that of schizophrenia, along with cognitive inflexibility and sensorimotor gating deficits, when the MAM was administered on E17. Previously, most work done with a MAM model of schizophrenia involved administering the drug on E15. They believe that the E17 model is better model of schizophrenia since there are less unrelated side effects, like microcephaly and motor impairments, when it is administered on E17 rather than E15.

I was curious about how E17 was chosen. I would be curious to see the results of their tests on other days, like E16 or E18. MAM-E17 brains were still about 7% smaller than control brains, so maybe administering it on E18 would have less of an effect on overall brain size.

            In the reversal learning task, the methods stated that females were used as controls. It didn’t seem like there were females used at any other time in the paper, so this may affect the results of the task. I would also like to see if they get the same results from using the MAM-E17 model on females. Also, the authors are unsure of how MAM preferentially affects cortical neurons. If more were known about how this model works, the model would be more convincing to me.

            Kellendock et al. used a transgenic mouse model of schizophrenia that overexpressed D2 receptors using a tetracycline transactivator (tTA). This expression was limited to the striatum. Overexpression of DA did not cause more locomotor activity, deficits in sensorimotor gating, or increased anxiety compared to control mice. In an attentional set-shifting task, DA overexpression mice had increased latency to choose between odors during reversal trials. This corresponds to executive function impairment in schizophrenia. The transgenic mice were tested in a DNMTS maze task off and on dox. The cognitive deficits remained even when the mice were put on dox so the DA receptor overexpression gene was turned off. The authors also found increased D1 expression in the mPFC, which is involved in executive function. They believe that the cognitive deficits are due to an imbalance in D1 activation in the mPFC. There is evidence that striatal dysfunction affects the PFC, so more research could be done on this pathway in the future. Since there are differences in circuits between rodents and humans, it is difficult for the rodent research to be translational. However, rodent models can still be used to test single aspects of schizophrenia. In the future, more research could be done on the additional cognitive deficits of schizophrenia.

Schizophrenia - Halloween

Schizophrenia is an extremely complex disease of the brain, which is discussed in both Moore et al. and Kellendonk et al. Due to the various different brain structures involved in this disorder, it takes a lot of research and experiments to show just how complex it really is. Moore et al. tested for ataxia, PPI, orofacial dyskinesias and ambulation following amphetamine distribution, among other things. While their results support the schizophrenia model, they do not make it very clear on why they included the amphetamine experiment in their schizophrenia paper, unless I am missing something. One particular overlap I noticed between the two papers was the PPI experiment, although they used different variables which yielded different results. The Moore et al. experiment used MAM E17 (useful for schizophrenia models) to show that there is a decrease in prepulse inhibition of startle in that experimental group, whereas the Kellendonk et al. experiment showed that mice overexpressing D2 receptors (an increased activity of D2Rs has been linked to schizophrenia) showed no deficits in prepulse inhibition of startle. I found these results to be interesting, and most likely demonstrate the complexity and selectivity of schizophrenia.

            I think these rodent models are relatively effective in studying schizophrenia due to obvious ethical reasons and overall difficulty in studying it in humans, but just how much of these animal models can we translate to humans? Schizophrenia involves a vast number of different symptoms such as hallucinations and delusions, which are not something researchers can study in animal models. I think these papers both do a good job supporting the schizophrenia model, but Kellendonk et al. does a more believable approach as they manipulate the D2 receptor, which is how many antipsychotics work.

Complexities in Modeling Schizophrenia

To begin, I want to talk about the two theories/ hypotheses behind the etiology of schizophrenia. Schizophrenia symptoms are typically categorized into three categories as follows: Positive (hallucinations, delusions, disordered thoughts), Negative (flat affect, reduced pleasure in activities, reduced speaking), and Cognitive (poor executive functioning, problems with working memory, trouble focusing). The two main theories of schizophrenia focus on the malfunctioning of two different neurotransmitter systems to explain the various symptoms implicated in schizophrenia. 

The Dopamine Hypothesis came into view in the 1970s after scientists observed that the psychosis caused by amphetamine drugs was similar to the positive symptoms of schizophrenia. These drugs are dopamine receptor (D2) agonists. A class of drugs now known as anti-psychotics were found to be helpful in relieving positive symptoms by taking a reverse action and antagonizing D2 receptors. Unfortunately, these drugs do almost nothing to alleviate the negative symptoms of schizophrenia, implying that the etiology of the disease is much more complicated than an increase of dopamine activity. 

The Glutamate Hypothesis emerged in the 1980s after scientists observed that the effects of NDMA antagonist drugs like PCP and ketamine can be similar to the symptoms of schizophrenia. It is possible that the activity at this receptor can cause a wider variety of schizophrenia symptoms via interactions with dopamine and GABA. So far, there has been mixed results in new treatments that involve an NDMA agonist.

  I found the difference in approach between the two papers to be particularly interesting this week since one paper embraced both the dopamine and glutamate hypotheses and the other paper focused on the dopamine hypothesis alone. 

The Moore et al paper seeks to create a model of schizophrenia by disrupting brain development at an important time point to create reduced brain volumes in areas specific to schizophrenia. Although this model seems viable, it requires much further testing from this initial study. Moore et al has sufficiently shown that this model behaviorally looks like schizophrenia, but they have done minimal investigation about the model's effects on neurotransmitter activity. They do include some experiments with both PCP and amphetamine to show that MAM E17 rats have and elevated response to both drugs. This implies that their model has developmentally altered the glutamate and dopamine pathways, however, more exploration is needed in this area. They could have stained for TH (a enzyme vital to making dopamine) or NDMA receptors to examine the specific alterations that MAM causes over the course of development to see if these changes are analogous to changes seen in post mortem brains of humans with schizophrenia.

Alternatively, the Kellendonk et al paper focused on a model that only concerned the dopamine hypothesis and transgenically altered the expression of D2 receptors in the striatum. This paper outlines some of the major issues that come with models of schizophrenia that only focus on the dopamine hypothesis. This model does demonstrate that a developmental change in D2 expression causes long-lasting changes in cognitive function, even after the restoration of normal D2 expression. It would be interesting to see if this developmental change has significant effects on other neurotransmitter pathways- primarily glutamate and NDMA receptors. 

Regardless of the approach to modeling schizophrenia in animals, great care must be taken to avoid over-anthropomorphizing lab animals. It can become very easy to interpret animal behavior as an emotionally analogous to humans. However, it is important to remember that despite many genetic similarities, humans are still very different from rodents. It is impossible to ask mice and rats if they are hallucinating or having delusions of grandeur, or even to know if rodents can experience such things. Therefore any models that are created will have to be researched with a grain of salt.