“… rs6741949in a DPP4 intron on chromosome 2q24, where the G allele was associated with smaller hippocampal volume (β=−52.8 mm3, p=2.9×10-7).”
The association with DPP4 sheds light on a fascinating connection between diabetes and hippocampal (memory) function.
“Further, DPP4 is an intrinsic membrane glycoprotein and a widely expressed serine exopeptidase. It is also an adipokine over-expressed in visceral adipose tissue of obese persons and those with diabetes, conditions associated with smaller hippocampal volume. A novel class of antidiabetic medications (sitagliptin, and related incretin compounds) inhibits DPP4 to improve insulin sensitivity and glucose tolerance through increased levels of glucagon like proteins-1 and 2 (GLP-1, -2). Interestingly, endogenous incretin GLP-1 is also heavily expressed in some hippocampal neurons and has neuroprotective properties.”
——————-
note: 23andMe does not cover rs6741949, but they do cover 2 flanking SNPs that are in pretty good linkage disequilibrium with rs6741949 … so, um, I’m trying to figure out how I might impute/infer my genotype here … hmmm.
rs3788979 (bp162900889) CC D’=0.81 strand + forward
rs6741949 (bp162910223) A?G? strand + forward
rs4664446 (bp162910403) AG D’=0.86 strand + forward
You already know this, but when you are stressed out (chronic stress), your brain doesn’t work very well. That’s right – just when you need it most – your brain has a way of letting you down!
Here are a few things that happen to the very cells (in the hippocampus) that you rely on:
Thanks brain! Thanks neurons for abandoning me when I need you most! According to this article, these cellular changes lead to, “impaired hippocampal involvement in episodic, declarative, contextual and spatial memory – likely to debilitate an individual’s ability to process information in new situations and to make decisions about how to deal with new challenges.” UGH!
Are our cells making these changes for a reason? Might it be better for cells to remodel temporarily rather than suffer permanent, life-long damage? Perhaps. Perhaps there are molecular pathways that can lead the reversal of these allostatic stress adaptations?
Check out this recent paper: “A negative regulator of MAP kinase causes depressive behavior” [doi 10.1038/nm.2219] the authors have identified a gene – MKP-1 – a phosphatase that normally dephosphorylates various MAP kinases involved in cellular growth, that, when inactivated in mice, produces animals that are resistant to chronic unpredictable stress. Although its known that MKP-1 is needed to limit immune responses associated with multi-organ failure during bacterial infections, the authors suggest:
“pharmacological blockade of MKP-1 would produce a resilient of anti-depressant response to stress”
Hmmm … so Mother Nature is using the same gene to regulate the immune response (turn it off so that it doesn’t damage the rest of the body) and to regulate synaptic growth (turn it off – which is something we DON’T want to do when we’re trying to recover from chronic stress)? Mother Nature gives us MKP-1 so I can survive an infection, but the same gene prevents us from recovering (finding happiness) from stress?
Of course, we do not need to rely only on pharmacological solutions. Exercise & social integration are cited by these authors as the top 2 non-medication strategies.
In his undergraduate writings while a student at Harvard in the early 1900’s E. E. Cummings quipped that, “Japanese poetry is different from Western poetry in the same way as silence is different from a voice”. Isabelle Alfandary explores this theme in Cummings’ poetry in her essay, “Voice and Silence in E. E. Cummings’ Poetry“, giving some context to how the poet explored the meanings and consequences of voice and silence. Take for example, his poem “silence”
silence
.is
a
looking
bird:the
turn
ing;edge, of
life
(inquiry before snow
e.e. cummings
Lately, it seems that the brain imaging community is similarly beginning to explore the meanings and consequences of the brain when it speaks (activations whilst performing certain tasks) and when it rests quietly. As Cummings beautifully intuits the profoundness of silence and rest, I suppose he might have been intrigued by just how very much the human brain is doing when we are not speaking, reading, or engaged in a task. Indeed, a community of brain imagers seem to be finding that the brain at rest has quite a lot to say – moreso in people who carry certain forms of genetic variation (related posts here & here).
A paper by Perrson and colleagues “Altered deactivation in individuals with genetic risk for Alzheimer’s disease” [doi:10.1016/j.neuropsychologia.2008.01.026] asked individuals to do something rather ordinary – to pay attention to words – and later to then respond to the meaning of these words (a semantic categorization task). This simple endeavor, which, in many ways uses the very same thought processes as used when reading poetry, turns out to activate regions of the temporal lobe such as the hippocampus and other connected structures such as the posterior cingulate cortex. These brain regions are known to lose function over the course of life in some individuals and underlie their age-related difficulties in remembering names and recalling words, etc. Indeed, some have described Alzheimer’s disease as a tragic descent into a world of silence.
In their recordings of brain activity of subjects (60 healthy participants aged 49-79), the team noticed something extraordinary. They found that there were differences not in how much the brain activates during the task – but rather in how much the brain de-activates – when participants simply stare into a blank screen at a small point of visual fixation. The team reports that individuals who carry at least one copy of epsilon-4 alleles of the APOE gene showed less de-activation of their their brain (in at least 6 regions of the so-called default mode network) compared to individuals who do not carry genetic risk for Alzheimer’s disease. Thus the ability of the brain to rest – or transition in and out of the so-called default mode network – seems impaired in individuals who carry higher genetic risk.
So, I shall embrace the poetic wisdom of E. E. Cummings and focus on the gaps, empty spaces, the vastness around me, the silences, and learn to bring my brain to rest. And in so doing, perhaps avoid an elderly descent into silence.
The human brain is renown for its complexity. Indeed, while we often marvel at the mature brain in its splendid form and capability, its even more staggering to consider how to build such a powerful computing machine. Admittedly, mother nature has been working on this for a long time – perhaps since the first neuronal cells and cell networks appeared on the scene hundreds of millions of years ago. In that case, shouldn’t things be pretty well figured out by now? Consider the example of Down syndrome, a developmental disability that affects about 1 in 800 children. In this disability, a mere 50% increase in a relative handful of genes is enough to alter the development of the human brain. To me, its somehow surprising that the development of such a complex organ can be so sensitive to minor disruptions – but perhaps that’s the main attribute of the design – to factor-in aspects of the early environment whilst building. Perhaps?
So what are these genes that, in the case of Down syndrome, can alter the course of brain development? Well, it is widely known that individuals with Down syndrome have an extra copy of chromosome 21. However, the disorder does not necessarily depend on having an extra copy of each and every gene on chromosome 21. Rare partial trisomies of only 5.4 million base-pairs on 21q22 can produce the same developmental outcomes as the full chromosome trisomy. Also, it turns out that mice have a large chunk of mouse chromosome 16 that has the very same linear array of genes (synteny) found on human chromosome 21 (see the figure here). In mice that have an extra copy of about 104 genes, (the Ts65Dn segment above) many of the developmental traits related to brain structure and physiology are observed. In mice that have an extra copy of about 81 genes, this is also the case (the Ts1Cje segment).
To focus this line of research even further, the recent paper by Belichenko et al., “The “Down Syndrome Critical Region” Is Sufficient in the Mouse Model to Confer Behavioral, Neurophysiological, and Synaptic Phenotypes Characteristic of Down Syndrome” [DOI:10.1523/JNEUROSCI.1547-09.2009] examine brain structure, physiology and behavior in a line of mice that carry an extra copy of just 33 genes (this is the Ts1Rhr segment seen in the figure above). Interestingly, these mice display many of the various traits (admittedly mouse versions) that have been associated with Down syndrome – thus greatly narrowing the search from a whole chromosome to a small number of genes. 20 out of 48 Down syndrome-related traits such as enlargement of dendritic spines, reductions of dendritic spines, brain morphology and various behaviors were observed. The authors suggest that 2 genes in this Ts1Rhr segment, in particular, look like intriguing candidates. DYRK1A a gene, that when over-expressed can lead to hippocampal-dependent learning deficits, and KCNJ6, a potassium channel which could readily drive neurons to hyperpolarize if over-expressed.
In Robert Sapolsky’s book, “Why Zebras Don’t Get Ulcers“, he details a biological feedback system wherein psychological stress leads to the release of glucocorticoids that have beneficial effects in the near-term but negative effects (e.g. ulcers, depression, etc.) in the long-term. The key to getting the near-term benefits and avoiding the long-term costs – is to be able to turn OFF the flow of glucocorticoids. This is normally dependent on circuitry involving the frontal cortex and hippocampus, that allow individuals to reset their expectations and acknowledge that everything is OK again. Here’s the catch (i.e. mother nature’s ironic sense of humor). These very glucocorticoids can initiate a kind of reorganization or ‘shrinkage’ to the hippocampus – and this can disable, or undermine the ability of the hippocampus to turn OFF the flow of glucocorticoids. Yes, that’s right, the very switch that turns OFF glucocorticoid flow is disabled by exposure to glucocorticoids! Can you imagine what happens when that switch (hippocampus) get progressively more disabled? Your ability to turn OFF glucocorticoids gets progressively worse and the negative effects of stress become more and more difficult to cope with.
Sounds depressing. Indeed it is, and there are many findings of reduced hippocampal volume in various depressive illnesses. The complex problem at hand, then, is how to reverse the runaway-train-like (depression leads to glucocorticoids which leads to smaller hippocampus which leads to more depression) effects of stress and depression?
One new avenue of research has been focused on the ability of the hippocampus to normally produce new cells – neurogenesis – throughout life. Might such cells be useful in reversing hippocampal remodeling (shrinkage)? If so, what molecules or genes might be targeted to drive this process in a treatment setting?
The recent paper by Joffe and colleagues, “Brain derived neurotrophic factorVal66Met polymorphism, the five factor model of personality and hippocampal volume: Implications for depressive illness” [doi: 10.1002/hbm.20592] offers some key insights. They examined 467 healthy participants of the Brain Resource International Database (a personalized medicine company with a focus on brain health) using personality tests, structural brain imaging and genotyping for an A-to-G variation (valine-to-methionine) polymorphism in the BDNF gene. They report that lower volume of the hippocampus was associated with higher scores of neuroticism (worriers) – but, this negative relationship was not found in all people – just those who carry the A- or methionine-allele. Thus, those individuals who carry the G/G (valine/valine) genotype of BDNF may be somewhat more protected from the negative (hippocampal remodeling) effects of psychological stress. Interestingly, the BDNF gene seems to play a role in brain repair! So perhaps this neuro-biochemical pathway can be explored to further therapeutic benefit. Exciting!!
By the way, the reason zebras don’t get ulcers, is because their life revolves around a lot of short term stressors (mainly hungry lions) where the glucocorticoid-stress system works wonderfully to keep them alive. Its only homo sapiens who has enough long-term memory to sit around in front of the TV and incessantly fret about the mortgage, the neighbors, the 401K etc., who have the capacity to bring down all the negative, toxic effects of chronic glucocorticoids exposure upon themselves. My 23andMe profile shows that I am a G/G valine/valine … does this mean I’m free to worry more? Now I’m worried. More on BDNF here.
With this premise in mind, it is fascinating to ponder some recent findings reported by Huffaker et al. in their research article, “A primate-specific, brain isoform of KCNH2 affects cortical physiology, cognition, neuronal repolarization and risk of schizophrenia” [doi: 10.1038/nm.1962]. Here, the research team has identified a gene, KCNH2, that is both differentially expressed in brains of schizophrenia patients vs. healthy controls and that contains several SNP genetic variants (rs3800779, rs748693, rs1036145) that are associated with multiple different patient populations. Furthermore, the team finds that the risk-associated SNPs are associated with greater expression of an isoform of KCNH2 – a kind of special isoform – one that is expressed in humans and other primates, but not in rodents (they show a frame-shift nucleotide change that renders their ATG start codon out of frame and their copy non-expressed). Last I checked, primates and rodents shared a common ancestor many millenia ago. Very neat – since some have suggested that newer evolutionary innovations might still have some kinks that need to be worked out.
In any case, the research team shows that the 3 SNPs are associated with a variety of brain parameters such as hippocampal volume, hippocampal activity (declarative memory task) and activity in the dorsolateral prefrontal cortex (DLPFC). The main suggestion of how these variants in KCNH2 might lead to these brain changes and risk for schizophrenia comes from previous findings that mutations in this gene screw up the efflux of K+ ions during the repolarization phase of an action potential. In the heart (where KCNH2 is also expressed) this has been shown to lead to a form of “long QT syndrome“. Thus, the team explores this idea using primary neuronal cell cultures and confirms that greater expression of the primate isoform leads to non-adaptive, quickly deactivating, faster firing patterns, presumably due to the extra K+ channels.
The authors hint that fast & extended spiking is – in the context of human cognition – is thought to be a good thing since its needed to allow the binding of multiple input streams. However, in this case, the variants seem to have pushed the process to a non-adaptive extreme. Perhaps there is a seed of an interesting evolutionary story here, since the innovation (longer, extended firing in the DLPFC) that allows humans to ponder so many ideas at the same time, may have some legacy non-adaptive genetic variation still floating around in the human lineage. Just a speculative muse – but fun to consider in a blog post.
In any case, the team has substantiated a very plausible mechanism for how the genetic variants may give rise to the disorder. A scientific tour-de-force if there ever was one.
On a personal note, I checked my 23andMe profile and found that while rs3800779 and rs748693 were not assayed, rs1036145 was, and I – boringly – am a middling G/A heterozygote. In this article, the researchers find that the A/As showed smaller right-hippocampal grey matter volume, but the G/A were not different that the G/Gs. During a declarative memory task, the GGs showed little or no change in hippocampal activity while the AA and GA group showed changes – but only in the left hippocampus. In the N-back task (a working memory task), the AA’s showed more changes in brain activation in the right DLPFC compared to the GGs and GAs.
For further edification, here is a video showing the structure of the KCNH2-type K+ channel. Marvel at the tiny pore that allows red K+ ions to leak through during the repolarization phase of an action potential. **PODCAST accompanies this post**
![Reblog this post [with Zemanta]](https://i0.wp.com/img.zemanta.com/reblog_c.png)
![Reblog this post [with Zemanta]](https://i0.wp.com/img.zemanta.com/reblog_e.png)
Genes 2 Brains 2 Mind 2 Me 