Friday, October 11, 2019

Ketogenic diets are unhelpful and dangerous for managing mitochondrial diseases. Maybe.

I think this is probably an abstract from a short communication at a conference. I picked it up from Miki Ben-Dor on twitter

I think we can say that ketogenic diets are pretty rubbish for managing mitochondrial illnesses.

By chance there is also a twitter discussion on-going relating to omega 6 based ketogenic diets.

First, nutritionists LOVE omega 6 PUFA and HATE saturated fats. In case anyone hadn't noticed. This tweet came from laura cooper, this time picked up via Raphi and Tucker.

Supported by this

It seems very likely that we can combine these two concepts and come up with some sort of an explanation.

I think we can accept that the ketone induced metabolic changes noted by Veech in isolated working rat hearts, resulting in increased energy yield per ATP molecule, still apply even with poorly functional mitochondria, because there is comparable improvement in the control of abnormal mitochondria induced intractable epilepsy vs ordinary intractable epilepsy. The ketogenic diet is clearly doing something...

That's good.

Everything else is bad.

If there is a significant problem with the structure/function of complex I ketones will not be directly helpful. They deliver acetyl-CoA to the TCA and essentially nothing else. There will be a small FADH2 input from succinate dehydrogenase but all other electrons will be presented as NADH, which needs a functional complex I to do anything much.

To bypass a poorly functional complex I we really need input as FADH2 directly to the CoQ couple without having to turn the TCA. That means beta oxidation of fatty acids, in particular it needs those fatty acids to be fully saturated because electron transporting flavoprotein only receives electrons to form FADH2 from the first desaturation step at the start of beta oxidation. Any double bonds skip this step.

Using PUFA immediately reduces energy sourced via this route.

The next thing we need to realise that modern nutritionist derived ketogenic diets cause, amongst other things, pancreatitis. I posted about pancreatitis, Intralipid and propofol here. It should come as no surprise that the side effects (from here) of PUFA based ketogenic diets in children can be severe, they're probably a lot higher in PUFA than even F3666 rodent chow...

"Other early-onset complications, in order of frequency, were hypertriglyceridemia, transient hyperuricemia, hypercholesterolemia, various infectious diseases, symptomatic hypoglycemia, hypoproteinemia, hypomagnesemia, repetitive hyponatremia, low concentrations of high-density lipoprotein, lipoid pneumonia due to aspiration, hepatitis, acute pancreatitis, and persistent metabolic acidosis. Late-onset complications also included osteopenia, renal stones, cardiomyopathy, secondary hypocarnitinemia, and iron-deficiency anemia".

Then there are cardiolipins. Each cytochrome C molecule is anchored to the outer surface of the inner mitochondrial membrane by four lipid anchors. Their nature is largely controlled by the dietary lipid supply. Modern PUFA based ketogenic diets will result in highly unsaturated cardiolipin anchors. Damaged mitochondria produce an excess of ROS. ROS break PUFA based cardiolipins giving apoptosis or, if ATP levels are too low for this, necrosis. Not going to do your ragged red muscle fibres any good. Or you cardiomyopathic cardiomyocytes.

I could go on, but you get the flavour.

Is there any end to the damage done by the lipid hypothesis?

Probably not.



Vern said...

I avoid PUFAs, eat a low carb diet and practice time restricted eating. I guess that it is even more important to avoid PUFAs on a ketogenic diet. Very interesting.

Wout Mertens said...

I'm currently lazily getting most of my nutrition from keto meal shakes, where I get to add my own choice of fat. I use coconut oil, sometimes butter, and my body fat.

I also eat hard cheese and nuts, mostly cashews (I don't want to go too low carb, I like bowel movements). The latter is my only significant source of PUFA, O6.

I've been eating this way for almost 5 months, lost a bunch of body fat and generally feel great.

I noticed that if I don't eat fatty fish (salmon, lazy again), I get itchy after a week or so. Eating salmon makes the itch stop within a day.

I suppose none of this really pertains to the article, I just felt like sharing ;)

Alex said...

(Don't know anything about the KD-ready to use formulas...)

Children's KD is high in PUFAs? Really? Why THE HELL would they do such a thing??

John said...

according to the USDA, mackerel has as much saturated fat as polyunsaturated:

and twice the saturated fat as a steak, if the steak has been trimmed:

but of course fish=good, red meat=bad, we all know that!

raphi said...

It's predictable, ketogenic diets were to become become bothersome pigs we couldn't ignore, so we put lipstick on them 8 (i.e. made them "plant-based" & with PUFA oils). We made them more palatable for mainstream expectations. Anyways...

I enjoyed the ragged-red fibers reference, I wrote about that damn mitochondrial disease back in 2014 as a topic for one of my molecular biology courses

"The elimination of dysfunctional mitochondria by modulating mitophagy, apoptosis and other cellular quality control dynamics has been proposed. This approach is endorsed by DiMauro et Schon who argue that the “possibility of mitochondrial dysfunction needs to be taken into account by every medical subspecialty […] progress in this field has been striking enough to amply justify the term ‘mitochondrial medicine’”(55). Current pharmaceutical drugs targeting mitochondria are still imprecise, weak and replete with side-effects that are often times worse than the disorder itself. Fortunately, a lot of the scientific literature from disparate fields has uncovered epigenetic influences — also known as lifestyle factors — affecting numerous quality control mechanisms of the organelle, inside and outside of it. Despite these encouraging avenues, mainstream research remains committed to distilling these hypercomplex effects into drug form before exploring non-drug approaches. Thankfully, there are roads to explore in the mean time. Examples include — but are not limited to — lifestyle factors guided by an evolutionarily concordant framework revolving around: nutrition, circadian rhythms entrainment, appropriate movement and activity and electrical stimulation of muscles and the brain(56,67). These attempts are both old and novel in many respects. They are often times significantly more reliable, reproducible, impactful and safe compared to conventional pharmaceutical drug treatments. Labelling such approaches as ‘holistic’ seems appropriate as the influences on mitochondria arise from many yet undiscovered pathways and feedback systems. What is being affected is not a single organ system, but organelles that allow the fundamentally life-sustaining process of respiration, carried out in nearly every cell in our body. Mitochondrial medicine is poised to modernize the medical paradigm."

Hap said...

I did not see you weigh in on the plant based keto thread started by dr Weiss who had a friend on standard keto but suggested PB. drastically lowered LDL. ALL IS WELL NOW. Anyway , criticizing avoided mostly a matter of civility. Isn’t it reasonabl to expect that most PB KD to be very high in omega 6 and subject to the issues you wrote about here? People are lauding this PB effort.

Also you may have written about metformin and complex I partial inhibition. Are we still to infer not a good idea on KD...but better on LC?

Passthecream said...


"and electrical stimulation of muscles and the brain(56,67)."

I have one of these in a cupboard somewhere, thanks for the suggestion, I'll dust it off, polish the electrodes and put a new dry cell in it --- could be so much more convenient than giving up Tim Tams and the olive oil liver flushes.


cavenewt said...

"Other early-onset complications...repetitive hyponatremia..."

No doubt those official nutritionist-formulated keto diets incluse low salt? Those of us doing keto know that it requires extra salt.

Passthecream said...

Polyunsaturatophilia is much more like a religion than anything else, cultural rather than medical. ( ... is there really a difference I wonder ...) It's easy to get good saturated fats on a plant based diet, the stuff is abundant in the vegetable kingdom - palm, coconut, cacao, some other detoxified ones like eg tallow tree, etc Coconut- mct alone travelling the direct path through the portal vein should be a good choice for medical ketosis, although, insulin???? In a mix perhaps.

Btw to add to your evolution/ adaptation point from recently, 'medicine' in a broad sense is also adaptation. The human brain is possibly capable of solving many diet and environmental related adaptive issues by exhibiting flexible behaviours and innovation although given the above pufa- mania, we can see how evolution takes some amongst us down dead end pathways, even so.

cavenewt said...

Peter, would you consider cancer a metabolic disease in this context? Almost every cancer has crippled mitochondria, according to Warburg and others. While a ketogenic diet might be bad for the tumor, it's good for the rest of the body.

Wonder if there's a way of delivering PUFAs directly to a tumor, while shielding the rest of the body from them

Gyan said...

Is there any empirical support for the assertion that human cardiolipin depends upon dietary lipid supply?
In Oxygen Nick Lane gives examples from animals that unsaturation of the cell membranes does not depend upon unsaturation of the dietary lipids.

Peter said...

Hi Gyan,

I missed that in Oxygen! There are lots of rodent examples of changing cardiolipin saturation through diet. This was my third hit, one and two were similar but not quite so good. I think humans are big mice, though not as crippled genetically as BL6 strains!

cave, yes. I consider that PUFA allow continued ingress of calories when they should stop and I consider them poorly able to generate ROS when ROS are needed for apoptosis. The converse is stearic acid and re PUFA So I think I would target the who body with stearic acid rather than try to weaken mitochondria selectively with targeted PUFA...


Justin said...

I've been following your blog for years and really appreciate you getting into the nuts and bolts of things even if it makes my head hurt! Anyway here is an interesting paper on the incorporation of supplemented unsaturated fatty acids into cardiolipin. Would love to here your take on it. Few cited references that look interesting too, especially the one palmitic acid incorporation and apoptosis. Anyway, thanks again for maintaining such an interesting blog!

Justin said...

Gyan said...

Peter, Justin

From the paper referred by Justin
"in the mammalian heart and skeletal tissues, the most common CLs are 18-carbon fatty acids, especially linoleic acid (LA), followed by the minor species stearic acid (18:0) and oleic acid (OA). The abundance of LA in CL in these tissues is maintained delicately [8], and its dysregulation is associated with aging and diseases. An animal study reported that compared with the heart of a young rat, the heart of an aged rat contained lower levels of LA-containing CL and higher levels of highly unsaturated fatty acids. "

1) LA is abundant in mammalian CL and abundance is regulated.
2) Dysregulation of LA is associated with aging and leads to lower LA but higher levels of more unsaturated lipids.

Peter said...


Always start with the methods. You can assess any paper in about 30 seconds.

Justin's paper:

"Cultures of GL15 glioblastoma multiforme cell line were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum, 100 IU/mL penicillin G, 100 lg/mL streptomycin, 2 mM glutamine, and 1 mM sodium pyruvate. After trypsinization, the flasks were incubated 3 days at 37°C in a 5% CO2 humidified atmosphere to obtain semi-confluent cells. Medium was removed and cells were incubated for 12–36 h in the presence of serum-free DMEM containing fatty acid/bovine serum albumin (BSA) solutions (molar ratio, 2 : 1; 0.6 mM final fatty acid concentration). The fatty acids used were palmitate (16:0), oleate (18:1), and linoleate (18:2). Control cells were incubated with media containing fatty acid-free BSA at the same concentration as the fatty acid-exposed cells. Where indicated, cells were incubated with palmitate in the presence of 1.5 mM cycloserine (Bla ́zquez et al. 2000), or 20 lM GW4869 (Delgado et al. 2006), or 5 lM desipramine hydrochloride (Delgado et al. 2006), or 5 lM AY-9944 (Yoshikawa and Sakuragawa 1992). In some experiments, cells were pre-incubated with CL liposomes (0.2 mM) for 10 min in phosphate-buffered saline (PBS) acidified to pH 5.8 before treatment with palmitate"

What was the glucose concentration used for cell culture with the palmitate?


Also how many in-vivo cells are ever exposed to 100% palmitate w/o oleate, especially with with glucose maintained continuously at 20mmol/l?

Clearly rat cardiolipins change with time. By dysfunction? Or because evolution chooses saturates with age because that's how you stave of apoptosis when fed rodent chow with linoleate as the main fatty acid? Who knows?


Justin said...

Good stuff Peter. Like I said, nuts and bolts. I remember that post and need to familiarize myself with normo-physiological glucose concentrations so I can spot the BS. It makes sense that the cells would try to stabilize by incorporating palmitate into the CL's. Good point about the probability of cells being exposed to pure palmitate without the oleate. I can understand the need to bookend responses to changes variables by making gross changes, but that is definitely outside of the normal physiological realm. Lol!

Quick question for you. I currently raise hair sheep, chickens and quail for food. I am having a hard time locating data on manipulating pufa content in chicken fat and eggs. I know you used to keep chickens and I was wondering if you ever stumbled upon any useful data? If so, I would greatly appreciate it if you could share any info! Thanks again for checking out that paper.


Hap said...

Those posts of yours very good. Enjoyed the course description as well as the focused sarcasm regarding mainstreaming of KD

Hap said...

“It's predictable, ketogenic diets were to become become bothersome pigs we couldn't ignore, so we put lipstick on them 8 (i.e. made them "plant-based" & with PUFA oils). We made them more palatable for mainstream expectations. Anyways...”

A Procrustean Bed

Justin said...

I don't mean to muddy up this post, but I couldn't help but this here. I do motor oil degradation studies at work and stumbled across this statement in an article I was reading a while back. Reading all of these posts on PUFA made me think of it. Made me chuckle a little.

"Unsaturated molecules are by their very nature more reactive than saturated molecules. Anyone who has visited a doctor or dietitian recently will know that switching from saturated fat to unsaturated fat in one’s diet has some health benefits. In this instance, unsaturated molecules are more desirable, because they are more readily broken down by the body - they react faster."


Passthecream said...

Justin, high pufa oils make good cooked oil varnishes eg lso or walnut or low oleic sunflower. Olive oil not so much, it is too saturated on average to make a strong film.

Justin said...
This comment has been removed by the author.
Passthecream said...

Justin I have a vague memory that chicken meat is innately higher in pufa even on a wild diet, one of the highest %. They descend from small wet area birds, no?, perhaps that calls for lower melting point fats, lower core temperatures perhaps? Most of the good sat fats seem to come from larger land animals.

raphi said...

Hi Hap,

Thanks for saying so :) It was a fun assignment, but I do cringe now reading a few of those things I wrote. For instance, I invoked epigenetics when what I should've said is "gene regulatory" - something I'm now quite critical of when people misuse it. Oh well, live and learn

Unknown said...

So, don't drink the Kool Aid, nor the Crisco, Nutritionists are dolling out...

Passthecream said...

Yet another plant source of useful fats is cupuassu. It's a relative of cacao and very popular in Brasil.

" Main fatty acid components of cupuaçu butter are stearic acid (38%), oleic acid (38%), palmitic acid (11%) and arachidic acid (7%).[20][21]"

Spinds like a promising blend appart from that high oleic content. Arachidic acid is fully saturated 20 carbon fat, the saturated equivalent of arachidonic acid. How does that rate in the lipid pantheon?

Peter said...

Arachidic acid is peroxisome bound. It will generate heat plus caprylic acid, which it will ship to the mitochondria. Too much FADH2 (yes, it can happen) in C20, stearate is where physiology stops for the mitochondria...


Passthecream said...

It is miraculously well engineered and leads me to wonder how many millions of years our ancestor species spent as small mouse-like creatures while evolution sorted all this out.

When oil soluble inputs are mixing with lipids and bile, and being re-esterified via enterocytes for shipping out on chylomicrons eg retinol, tocopherol and cholesterol, humans seem to use any handy fatty acid so you get retinoyl linoleate/ palmitate etc. but some mustelids only use stearic acid for that purpose --- ferrets? or weasels? Maybe others.