Christmas food for thought: Feed me, all 100 trillion of me

The morning before Christmas eve, I’m sitting here in the dining room munching happily on the bits and pieces of what’s left of our gingerbread house that was only erected to its full glory the night before. I have not consumed this amount of carbohydrates in over a year.

Inside, a few species of my extensive gut microbe community are screaming bloody murder.

E. coli bacteria

FEEEEED ME!!!! Source:

When you eat, you’re not only feeding your own fleshy vessel, but also the 100 trillion of microbugs that thrive in your intestines. Hardly “along for the ride”, these bugs not only help us digest foodstuff, ferment carbohydrates and proteins but also heavily impact our metabolism and general health. Depending on their composition, they tweak our risk of cardiovascular diseases, Type II diabetes and may even cause obesity in humans. There’s tantalizing evidence that their reach extends to the brain, influencing mood, anxiety and cognition in mice.

However, the gut microbiota* is a fluid, ever-changing beast. In one previous study, researchers transplanted gut-free mice with fresh or frozen human poop to inoculate them with a microbiome of known composition. When researchers switched these mice’s plant-based diet to a high-fat, high-sugar one, the structure of the established microbiome changed within a single day: some species dwindled in number, while others exploded onto the intestinal stage, bringing with them their particular metabolic tricks. (*The word “microbiome” refers to the set of genes in the gut bugs).

Similar diet-induced changes have been found in humans. When babies are weaned from their mothers’ milk and switch to solid food, their gut bug community simultaneously go through tumultuous changes. The gut bugs of African hunter-gatherers vastly differ from those in people grown on a Western diet. But these changes take weeks, even lifetimes. Just how fast can the microbiome adapt and change to a new diet?

In a new study, researchers recruited ten volunteers and put them on two drastically different extreme diets for 5 days – as you can see below, the plant-based diet was rich in grains, fruits and vegetables (high-carb and high-fibre), while the animal-based diet consisted of meats, eggs and cheeses (high-fat, high-protein and low/no-fibre). Each day, the volunteers handed in a poop sample for the researchers to monitor.

Screen Shot 2013-12-23 at 10.24.18 AM

In general, the animal-based diet had a greater impact on gut flora than the plant-based one. It significantly increased the diversity of gut flora, enriching 22 species whilst decreasing the fibre-intake associated Prevotella in a life-long vegetarian on this meaty diet. The plant-based diet, on the other hand, only increased the abundance of 3 species, mostly those associated with carbohydrate fermentation.

Many of the changes made sense. An animal-based diet enriched putrefactive microbes, shifting carbohydrate fermentation into amino acid digestion, thus helping the body break down the onslaught of heaps animal protein. Several strains of immigrant bacteria – particularly those used for cheese- and sausage-making –settled down and made themselves comfortable in the native gut flora community. The meat-heavy diet also triggered microbes to activate pathways that degrade cancer-causing compounds found in charred meats, and enhanced the synthesis of vitamins.

On the other hand, several strains of potentially health-negative bacteria also multiplied in the meat-eaters. On a high-fat diet, we excrete more bile – a bitter fluid that may ruin a good fish dish – to deal with the digestion of fat. Bile is toxic to many gutbugs, but not to the mighty Bilophila (“bile-loving”) wadsworthia – a bile-resistant bacterium stimulated by saturated fats in milk that may cause intestinal inflammation, at least in mice. The high-fat content in the animal-based diet also triggered increased levels of microbe-produced DCA, which is previously linked to liver cancer in mice. However, as of now there’s no evidence that these risks also apply to people, and researchers caution against making health-related judgments (although some can’t resist the temptation).

On the whole, plant- and animal-based diets induced changes in host microbiome gene structure that resembled those of herbivorous and carnivorous mammals within a few days. Furthermore, the volunteer’s microbiome reversed back to their previous composition only 2 days after the end of the experiment. Researchers believe we might be looking at a fast-forwarded movie of millions of years of co-evolution between humans and their microbugs: when animal food sources fell scarce, our ancestors were forced to switch to a plant-heavy diet; a flexible gut-bug community could quickly and appropriately shift their repertoire and function to help digestion, thus increasing the flexibility of human diets and chances of survival.

Thus, when you gobble down the vast selection of Christmas dishes this year, remember to thank the flexibility of your gut flora for your diverse digestive powers. And remember that we can’t say one diet is better than the other for our microbiota; the take-home message is that they are incredible flexible, more so than we previously thought. In the end, it still comes down to the age-old wisdom: you are what you eat.
David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, Biddinger SB, Dutton RJ, & Turnbaugh PJ (2013). Diet rapidly and reproducibly alters the human gut microbiome. Nature PMID: 24336217

Scientific American Best of the Blogs Video

I’m over at #SciAm videos, talking about the (potential) neuroprotective effects of a ketogenic diet. The short clip is based on a blog post The Fat-Fueled Brain I wrote a while back for their Guest Blogs (the article’s older sibling, Brain, living on Ketones can be found here).

The talented Dr. Carin Bondar, host of the series, suffered through the process of piecing my camera-dodging, hair-pulling, face-making video clips together (I’m EXTREMELY camera shy). She has my deepest respect.


SciAm also has a great video series called Instant Egghead that’s worth checking out. Fun – and super random- science info in ~2min clips with great graphics. Here’s an example: why toothpaste makes OJ taste bad.

The fat-fueled brain: unnatural or advantageous?

This post was originally published in Scientific American MIND Blogs.


It’s not bacon; it’s therapy! Source:

The ketogenic diet is a nutritionist’s nightmare. High in saturated fat and VERY low in carbohydrates, “keto” is adopted by a growing population to paradoxically promote weight loss and mental well-being. Drinking coffee with butter? Eating a block of cream cheese? Little to no fruit? To the uninitiated, keto defies all common sense, inviting skeptics to wave it off as an unnatural “bacon-and-steak” fad diet.

Yet versions of the ketogenic diet has been used to successfully treat drug-resistant epilepsy in children since the 1920s – potentially even back in the biblical ages. Emerging evidence from animal models and clinical trials suggest keto may be therapeutically used in many other neurological disorders, including head ache, neurodegenerative diseases, sleep disorders, bipolar disorder, autism and brain cancer. With no apparent side effects.

Sound too good to be true? I feel ya! Where are these neuroprotective effects coming from? What’s going on in the brain on a ketogenic diet?

Ketosis in a nutshell

In essence, a ketogenic diet mimics starvation, allowing the body to go into a metabolic state called ketosis (key-tow-sis). Normally, human bodies are sugar-driven machines: ingested carbohydrates are broken down into glucose, which is mainly transported and used as energy or stored as glycogen in liver and muscle tissue. When deprived of dietary carbohydrates (usually below 50g/day), the liver becomes the sole provider of glucose to feed your hungry organs – especially the brain, a particularly greedy entity accounting for ~20% of total energy expenditure. The brain cannot DIRECTLY use fat for energy. Once liver glycogen is depleted, without a backup energy source, humanity would’ve long disappeared in the eons of evolution.

The backup is ketone bodies that the liver derives primarily from fatty acids in your diet or body fat. These ketones – β-hydroxybutyrate (BHB)*, acetoacetate and acetone – are released into the bloodstream, taken up by the brain and other organs, shuttled into the “energy factory” mitochondria and used up as fuel. Excess BHB and acetoacetate are excreted from urine, while acetone, due to its volatile nature, is breathed out (hence the characteristically sweet “keto breath”). Meanwhile, blood glucose remains physiologically normal due to glucose derived from certain amino acids and the breakdown of fatty acids – voila, low blood sugar avoided!

*Chemically speaking, BHB is not a ketone – it’s a carboxylic acid)


Brain on ketones: Energetics, Oxidation and Inflammation

So the brain is happily deriving energy from ketones – sure, but why would this be protective against such a variety of brain diseases?

One answer may be energy. Despite their superficial differences, many neurological diseases share one major problem – deficient energy production. During metabolic stress, ketones serve as an alternative energy source to maintain normal brain cell metabolism. In fact, BHB (a major ketone) may be an even more efficient fuel than glucose, providing more energy per unit oxygen used. A ketogenic diet also increases the number of mitochondria, so called “energy factories” in brain cells. A recent study found enhanced expression of genes encoding for mitochondrial enzymes and energy metabolism in the hippocampus, a part of the brain important for learning and memory. Hippocampal cells often degenerate in age-related brain diseases, leading to cognitive dysfunction and memory loss. With increased energy reserve, neurons may be able to ward off disease stressors that would usually exhaust and kill the cell.

A ketogenic diet may also DIRECTLY inhibit a major source of neuronal stress, by –well- acting like a blueberry. Reactive oxygen species are unfortunate byproducts of cellular metabolism. Unlike the gas Oxygen, these “oxidants” have a single electron that makes them highly reactive, bombarding into proteins and membranes and wrecking their structure. Increased oxidants are a hallmark of aging, stroke and neurodegeneration.

Ketones directly inhibit the production of these violent molecules, and enhance their breakdown through increasing the activity of glutathione peroxidase, a part of our innate anti-oxidant system. The low intake of carbohydrates also directly reduces glucose oxidation (something called “glycolysis”). Using a glucose-like non-metabolized analogue, one study found that neurons activate stress proteins to lower oxidant levels and stabilize mitochondria.

Due to its high fat nature, keto increases poly-unsaturated fatty acids (PUFAs, such as DHA and EPA, both sold over-the-counter as “brain healthy” supplements), which in turn reduces oxidant production and inflammation. Inflammatory stress is another “root of all evil”, which PUFAs target by inhibiting the expression of genes encoding for pro-inflammatory factors.

Neurons on Ketones: Dampen that enthusiasm!

Excited neurons transmit signals, process information and form the basis of a functioning brain. OVER-excited neurons tend to die.

The brain teeters on a balance between excitation and inhibition through two main neurotransmitters, the excitatory glutamate and the inhibitory GABA. Tilt the scale towards glutamate, which occurs in stroke, seizures and neurodegeneration, and you get excitotoxicity. In other words, hyper-activity is toxic.

Back in the 1930s, researchers found that direct injection of various ketone bodies into rabbits prevented chemically-induced seizures through inhibiting glutamate release, but the precise mechanism was unclear. A recent study in hippocampal neurons showed that ketones directly inhibited the neuron’s ability to “load up” on glutamate – that is, the transmitter can’t be packaged into vesicles and released – and thus decreased excitatory transmission. In a model of epilepsy that used a chemical similar to glutamate to induce damage, the diet protected mice against cell death in the hippocampus by inhibiting pro-death signaling molecules. On the other end of the excitation-inhibition balance, ketones increase GABA in the synapses (where neurotransmitters are released) of rats and in the brains of some (but not all) epileptic humans subjects. This increase in inhibition may confer both anti-seizure effects and neuroprotection, though data is still scant.

Then there are some fringe hypotheses. The acidity of ketones may decrease the pH of certain brain microdomains, which might to be the mechanism of keto’s positive effect on Type II Bipolar disorder (lot’s of mays and mights, I know).  As keto affects the whole body, global changes due to calorie restriction and regulation of the satiety hormone Leptin are bound to alter brain function, and play a circumstantial role.

Neuroprotection? Show me the evidence!

All these molecular changes suggest that a ketogenic diet is protective against brain injury. But is there any REAL evidence?

A study with 23 elderly with mild cognitive impairment showed that a ketogenic diet improved verbal memory performance after 6 weeks compared to a standard high carbohydrate diet. In a double-blind, placebo-controlled study, 152 patients with mild- to moderate Alzheimer’s disease were given either a ketogenic agent or a placebo, while maintaining a normal diet. 90 days later, those receiving the drug showed marked cognitive improvement compared to placebo, which was correlated with the level of ketones in the blood.

In a pilot study in 7 patients with Parkinson’s disease, 5 were able to stick to the diet for 28 days and showed marked reduction in their physical symptoms. In an animal model of Amytrophic Lateral Sclerosis (ALS), a ketogenic diet also led to delayed motor neuron death and histological and functional improvements, although it did not increase life span; clinical trials are on the way.

Remarkably, a long-term ketogenic diet does not seem to be associated with significant side effects, although constipation, dehydration and electrolyte and micronutrient deficiencies are common complaints. More serious complications include increased chance of kidney stones, gallbladder problems and bone fractures, especially in children. Menstrual irregularities often occur in women, with potential impact on fertility. Although ketoacidosis – acidification of the blood due to pathological levels of ketones – was historically proposed as a side effect, nutritional ketosis simply cannot achieve the level of ketones required to induce this life-threatening state. Nevertheless, there are no studies directly monitoring the side effects of ketosis yet, hence it’s too early to conclude that the diet is completely safe for everyone.

Brain ❤ Bacon?

While promising, large-scale placebo-controlled clinical trials in patients with neurological disorders are still lacking. The existing data needs to be interpreted carefully to avoid generating false hope or encourage patients to “ditch drugs for diet”. Nevertheless, the possibility that we can reduce symptoms of untreatable neurological disorders through modifying dietary composition is quite incredible; that a ketogenic diet may benefit physical and cognitive performance in healthy individuals is an even more tantalizing idea.

As the science behind this age-old dietary therapy gradually comes to light, social issues such as low adherence and public prejudice will need to be resolved. In the meantime, to those neuroscientists interested in studying keto: pass the bacon and I VOLUNTEER!

Brain, livin’ on ketones – a molecular neuroscience look at the ketogenic diet

Edited October 3, 2013: A 2.0 version of this post can be found at Scientific American MIND Guest blogs, here. And here’s me talking about it. Feel free to check it out!


WARNING: Wall of text on the yummy neuroprotective effect of ketosis from a molecular neuroscience point of view. Proceed with caution.


Delicious neuroprotection on a plate (Don’t forget veggies!) source:

Remember when your high school biology teacher said that the brain absolutely NEEDS glucose to function? Well, that’s not entirely true. Under severe carbohydrate restriction, the brain can adapt and start burning ketones as fuel.

Originally devised as a therapy for drug-resistant epilepsy in children, the ketogenic diet (keto) has been gaining popularity lately. It’s a high fat, moderate protein and low carbohydrate diet (LCHF) designed to force the body to go into a state called metabolic ketosis. With the advent of books like “Good Calories, Bad Calories” and “Why we get fat”, LCHF diets are increasingly touted as the magic bullet to weight loss. While there is considerable interest in the medical community in using the ketogenic diet to manage metabolic syndrome or prevent cardiovascular disease, more attention has focused on its role in drug-resistant seizure management and (potentially) neuroprotective effects in brain damage. In the last decade, keto has been shown to improve memory in patients at risk for Alzheimer’s disease, stabilize mood in type II bipolar disorder, reduce symptoms in Parkinson’s disease and even ameliorate some behavioral and social deficits in autism. Keto also seems to decrease brain cancer progression. ALL without observable side effects. Although most of these studies were unblinded (hence placebo can’t be ruled out), the effect is still amazing.

What is going on in the brain? And why aren’t pharmaceutical companies racing to package keto into a convenient treat-all 3-a-day pill?

How does the body go into ketosis?

Simple speaking, strict carbohydrate restriction depletes liver glycogen and forces the body to turn to other macronutrients for energy. Proteins are metabolically costly to utilize (not to mention dangerous – heart is also a muscle), and is often used as a last resort. By providing adequate amounts of fat, the liver uses dietary and body fat as fuel and produces ketones. “Ketones”, or “ketone bodies” is actually an umbrella term for 3 different molecules, β-hydroxybutyrate (BHB), acetoacetate (ACA) and acetone. All three can be delivered into the brain and metabolically converted into ATP in both neurons and glia. The three are interrelated: BHB and ACA can convert into each other, while ACA can turn into acetone. Extra ketones are eliminated through urine, or in the case of acetone, breath.

What are ketones doing in the brain? Answer: it’s complicated

There’s a reason neuroscientists and neurologists still haven’t figured out why keto is so effective in treating epilepsy. Holistically, the body is now running in a different metabolic state with changes in hormone levels (to say the least) which influences the nervous system. Locally, the brain is running on 3 types of semi interchangeable ketone bodies, the effects of which often can’t be teased apart. Hence it’s hard to say whether specific molecular and cellular alterations observed clinically or experimentally in animal models are a direct effect of ketosis, or simply a secondary phenomenon. Nevertheless, several hypotheses have been put forward to explain keto’s neuroprotective effects.

Source: Melo TM et al. Neuronal–glial interactions in rats fed a ketogenic diet. Neurochemistry InternationalVolume 48, Issues 6–7, May–June 2006, Pages 498–507

Source: Melo TM et al. Neuronal–glial interactions in rats fed a ketogenic diet. Neurochemistry International
Volume 48, Issues 6–7, May–June 2006, Pages 498–507

An oldie turned newbie: local changes in pH

Ketone metabolism generate pH-lowering metabolites, hence a change in pH was proposed early on as a way keto influences brain function. However there’s no evidence that keto significantly lowers brain pH, although mild decreases in pH may be possible in local microdomains. This hypothesis is attractive as many receptors are modulated by pH, such as acid-sensing ion channel (involved in stroke) and NMDA receptors (involved in learning, memory and excitotoxicity in stroke/neurodegenerative diseases), which may explain keto’s possible effect in stroke protection or cognitive improvement. Discarded a while back, this hypothesis recently resurfaced as the mechanism behind keto’s positive effect on Type II bipolar disorder management, which relies on blood acidification.

A favorite: Bioenergetics?

Ketones can be turned into energy effectively by the brain. In fact, BHB may provide a more efficient source of energy for brain per unit oxygen than glucose. A microarray study showed that keto induced a coordinated upregulation of genes encoding energy metabolism and mitochondrial enzymes, increasing the number of mitochondria in the hippocampus, a brain area associated with learning and memory. This increased energy capacity has been shown to enable hippocampal neurons to better withstand low glucose exposure, which happens in stroke. Better bioenergetics is proposed to limit seizure activity by stabilizing neuron resting membrane potential (so they’re not as excitable) or activate KATP channels through adenosine release. There are no studies directly addressing if energy efficiency is the reason for cognitive improvement in neurodegenerative diseases under keto, or if it contributes to enhanced cognitive performance in healthy individuals.

Cute as it is, there's not a hell lot of evidence.

Cute as it is, there’s not a hell lot of evidence. Source:

Another favorite: Antioxidative & anti-inflammatory effects?

Mitochondrial respiration, while generating ATP, also produces many reactive oxygen species (ROS). An acute increase in ROS is associated with stroke damage, while accumulation of ROS is one of the major hallmarks of aging and age-related neurodegenerative diseases. Keto can induce upregulation of mitochondrial uncoupling proteins (UCPs) in rat, which correlates with decreased ROS generation and increased resistance towards chemically induced seizure. Keto also increases the body’s own antioxidant defense system, namely glutathione levels in the hippocampus and protects mitochondrial DNA from ROS damage. If, and how, these antioxidative effect relate to neuroprotection is not yet clear.

Poly unsaturated fatty acids (PUFA) in the keto diet, such as DHA and EPA have garnered a lot of attention. Some evidence links them to decreasing neuronal excitability in hippocampus, which may contribute to decreasing seizure generation. PUFAs can also directly act on receptors called peroxisome proliferator-activated receptors (PPARs), the latter of which translocates to the nucleus and shuts down expression of pro-inflammatory factors. As inflammation increasingly recognized as a contributor to seizures, Alzheimer’s and metabolic syndrome, this mechanism may be a crucial one in keto’s favorable effects.

Maybe: direct drug-like actions of ketone bodies

Direct injection of ACA and acetone into animal models of epilepsy prevented seizures, hinting that ketone bodies may directly suppress seizure activity. However, other studies show that ketone levels may not correlate with seizure control. BHB and ACA have also been proposed to directly influence excitatory/inhibitor neural transmission. However, direct application of the two ketones had no effect on (1) excitatory responses in hippocampal neurons after stimulation (2) spontaneous epilepsy-like activity in a brain slice model of epilepsy (3) whole-cell currents evoked by glutamate, kainate, and GABA in cultured hippocampal neurons. Looks like a nail in the coffin for that theory. As far as I know, direct actions of ketone has not been linked to neuroprotection.

 Directly inhibiting cell death?

Keto seems to suppress the expression of pro-cell death proteins such as caspase-3 and clusterin (both of which mediates cell death in Huntington’s disease, among others), which correlates with enhanced recovery from seizure episode or stroke in patients. How much of a role this mechanism plays in disease states is still unknown.

Putting it all together: no pill yet!

The effects of keto are multifactorial and complicated. At the moment it’s impossible to tease apart which mechanisms are the driving forces behind keto’s powers, and which ones are secondary manifestations. Or they may operate equally, who knows? Hence packing everything up into a neat little keto pill is going to require a lot of effort (… …don’t even mention raspberry ketones!).

In the end, what do all these data tell us? A ketogenic diet with calorie restriction is most likely beneficial to weight loss. It is effective in controlling seizures in children and adults. It may improve cognition in patients with neurodegeneration or enhance mood stability in patients with Type II bipolar disorder. And that’s a BIG “may”. Without randomized controlled trials, it’s really difficult to say.

What about keto as a potential therapy (or adjunct therapy) for neurodegeneration? As increased ROS accumulation and mitochondrial dysfunction are common threads in age-related neurodegenerative diseases, it is conceivable that keto could be beneficial with its antioxidative and anti-inflammatory actions. To what degree is anyone’s guess.

Pass the bacon, maybe?

The ketogenic diet (along with other low-carb diets) is gaining popularity as a weight-loss measure. Some go on the keto diet because they experience “fewer sugar crashes, enhanced energy levels, mental clarity and decreased hunger”. A quick browse through progress photos on is probably enough to convince most people with a few extra pounds to give the diet a shot. Based on the above studies, keto is touted as a safe, effective tool for weight loss, with the added bonus of improved cognition. A few studies in epileptic children and healthy volunteers demonstrate that keto exterts a biphasic effect on cognition, with initial lethargy and subsequent heightened vitality, physical functioning, and alertness. Whether this translates into overall cognitive enhancement remains to be seen.

What are the potential side effects? A small study involving 21 obese women showed impaired higher cognitive function after 28-days on the keto diet. On the contrary, another study involving 83 obese patients showed improved blood lipid profile after being on a 24-week ketogenic diet without significant side effects. A caveat of most of these studies is the lack of an adequate control group, making results hard to interpret.

Some have also raised concerned regarding metabolic and long-term complications. Children following ketogenic diets have higher rates of dehydration, constipation, and kidney stones. Other reported adverse effects include hyperlipidemia, impaired neutrophil function, optic neuropathy, osteoporosis, and protein deficiency. However, ketogenic diet for the management of seizure is different than that for weight-loss. The former generally requires 80-90% fat calories (although this has decreased slightly in the Modified Atkins Diet) while the latter proposes ~60% fat calories with sufficient protein for muscle maintenance.

Whether keto promotes cognitive improvement and neuroprotection in the general public remains to be seen. While waiting for science to catch up though, I’m going for that bacon, spinach & cheese omelet. For science!

Edited Sep 2,2013: For those interested in exploring more, here is an accessible journal review that looks at potential therapeutic uses of nutritional ketosis in many other diseases. Note I am not promoting using keto as a sole treatment option – if there are efficient pharmaceuticals available please do not forgo them in favour of ketosis.


Masino and Jong. 2012. Mechanisms of Ketogenic Diet Action. Jasper’s basic mechanisms of the epilepsies. 4th ed.

Hallböök T, Ji S, Maudsley S, & Martin B (2012). The effects of the ketogenic diet on behavior and cognition. Epilepsy research, 100 (3), 304-9 PMID: 21872440

Dashti HM et al. 2004. Long-term effects of a ketogenic diet in obeses patients. Exp Clin Cardiol. 9(3): 200-205.

Rho, J., & Sankar, R. (2008). The ketogenic diet in a pill: Is this possible? Epilepsia, 49, 127-133 DOI: 10.1111/j.1528-1167.2008.01857.x

Phelps, J., Siemers, S., & El-Mallakh, R. (2012). The ketogenic diet for type II bipolar disorder Neurocase, 1-4 DOI: 10.1080/13554794.2012.690421

Ruskin DN, Ross JL, Kawamura M Jr, Ruiz TL, Geiger JD, & Masino SA (2011). A ketogenic diet delays weight loss and does not impair working memory or motor function in the R6/2 1J mouse model of Huntington’s disease. Physiology & behavior, 103 (5), 501-7 PMID: 21501628

Krikorian R, Shidler MD, Dangelo K, Couch SC, Benoit SC, & Clegg DJ (2012). Dietary ketosis enhances memory in mild cognitive impairment. Neurobiology of aging, 33 (2), 2147483647-27 PMID: 21130529

Denke, M. (2001). Metabolic effects of high-protein, low-carbohydrate diets The American Journal of Cardiology, 88 (1), 59-61 DOI: 10.1016/S0002-9149(01)01586-7