Sunday, April 28, 2013

Hyperglycaemia is bad

Hyperglycaemia does whatever you want it to. Want to show it increases glycolysis and/or oxidative phosphorylation? No problem. Want to show it decreases both? Equally no problem. Choose your tissue, choose your duration, choose your insulin level, choose your glucose level, choose your tissue culture medium before test, choose... With the correct combination you can show anything.

But certain patterns emerge from lots of papers. In the short term hyperglycaemia increases both glycolysis and oxidative phosphorylation. Acute hyperglycaemia in neurons induces an equally acute hyperpolarisation of the inner mitochondrial membrane (a pre requisite for reverse electron flow through complex I), followed by a burst of free radicals (from reverse electron transport in the face of a low NAD+/NADH ratio?), followed by a collapse of the inner mitochondrial membrane potential (from free radical induced loss of cytochrome c?), soon to be followed by apoptosis, as you might expect

These guys set out the events nicely but suggest the mechanism is unclear. I would be willing to bet on G-3-P dehydrogenase as driving reverse electron flow using the high membrane potential from glycolysis. It seems that, under "mitochondrial preparation" conditions, ignoring reverse electron flow, G-3-P dehydrogenase also spills a reasonable dose of free radicals not only inwards towards the matrix but also outwards to the inter membrane space, in roughly equal amounts. As does complex III of course, but complex III is not specifically driven by a short side branch of hyperglycaemia-induced hyperactive glycolysis. Goodness only knows if this happens in-vivo, but let's accept that it does. Cytochrome c is on the outer surface of the inner mitochondrial membrane and spilling free radicals outwards seems a good way to oxidise the cardiolipin anchors and release one of the most important pro apoptotic proteins we have, cytochrome c.

So acute hyperglycaemic injury, in a cell type where glucose entry is essentially concentration driven, is potentially apoptotic if the injury is severe enough. Lesser but sill significant injury may come from spills of superoxide from complex I on to the mitochondrial DNA, another potentially interesting effect. Research on G-3-P dehydrogenase is still in its infancy and there are no clear cut answer as to how important this scenario might be, but I rather like it. Is it true? Who knows. It's hard to tell.

Exactly how difficult it is to transfer information from "preparations" to any semblance of "in vivo" is reviewed by Martin Brand. I like this chap, he really looks at the limitations of how much we currently know (not much, it appears) plus he came up through Naked Mole Rat research, another positive. Here's his summary of where free radicals might be produced:

Outwards spillage, directly on to cytochrome c, from G-3-P dehydrogenase and complex III...

It's quite clear that hyperglycaemia is not invariably acutely fatal to all neurons on first exposure. It takes years of following the advice of the ADA and AHA to develop diabetic neuropathy or to kill off enough central neurons (around 70%) to get the clinical label of Alzheimers and, while recurrent hyperglycaemia might get us there directly, the indirect effects are much more interesting to a mitochondriac like myself.

Chronic hyperglycaemia is where we have a depressed inner mitochondrial membrane potential, reduced glycolysis and electron transport with subsequent failure to generate superoxide.

Badness too.



Sabine said...

Thank you for this post.
I read about this first about ten years ago, how sugar and also cortisol lead to apoptosis (via mitochondria) of the cell, and I always wanted to know more.

Puddleg said...

For reference purposes, what would be the F:N ratio of glucose?

Mitochondriac - I like it.

Just to complicate things:
"Mitochondria are recognized as an important intracellular Ca2+ store and actively participate in Ca2+ signaling. Increased mitochondrial Ca2+ has been noted to have a number of effects including increasing electron transport, increasing ROS production, and, under some conditions, causing the opening of the permeability transition pore (13)⇓ . The mitochondrial Ca2+ content is regulated by both Ca2+ uptake and efflux. Mitochondria have a strong negative matrix electrochemical potential, and under normal conditions they take up Ca2+ specifically via a Ca2+-selective channel in the inner membrane called the Ca2+ uniporter. Efflux of Ca2+ from mitochondria occurs as well, but needs to be coupled to another ion gradient through Na+/Ca2+ exchange or H+/Ca2+ exchange (14)⇓."

Peter said...


Agree, Ca is v important. I have a large review paper summarising what we knew 4 years or so ago, incl Ca, which I have yet to sit down and take on board in detail... I had promised myself that at the end of the protons thread I would take a bit of a break and not only read this review but also Nick Lane's books, The Hallucinogens and a few other odds and ends IRL. One day...


Purposelessness said...


david said...

off topic, but any views on jaminet's post?

Peter said...


Seen too many redmeatwillkillyou studies to worry about any of them. Re Paul J's post, a pound of meat a day is well above what I would eat unless I was having a protein overdose day, which is my main occasional dietary sin. Adequate protein is my aim as much as is practical...

Sad to see Barry Groves has died, only info I have is via Jimmy Moore's website.


karl said...

Wondered about thoughts on this:

BigWhiskey said...

What is your range of "adequate protein", Peter?

karl said...

Speaking of inflammation - a new paper:

Puddleg said...

I don't know if you've seen this.
The SFA rats got fatter, not surprisingly as they ate a lot more; but their livers stayed slim.

Long term highly saturated fat diet does not induce NASH in Wistar rats

MCDD fed rats rapidly lost weight and showed NASH features. Rats fed coconut (86% of saturated fatty acid) or butter (51% of saturated fatty acid) had an increased caloric intake (+143% and +30%). At the end of the study period, total lipid ingestion in term of percentage of energy intake was higher in both coconut (45%) and butter (42%) groups than in the standard (7%) diet group. No change in body mass was observed as compared with standard rats at the end of the experiment. However, high fat fed rats were fattier with enlarged white and brown adipose tissue (BAT) depots, but they showed no liver steatosis and no difference in triglyceride content in hepatocytes, as compared with standard rats. Absence of hepatic lipid accumulation with high fat diets was not related to a higher lipid oxidation by isolated hepatocytes (unchanged ketogenesis and oxygen consumption) or hepatic mitochondrial respiration but was rather associated with a rise in BAT uncoupling protein UCP1 (+25–28% vs standard).

Unknown said...

Great stuff Peter.

Peter said...

Hi BigWhisky, 40-70g/d is my target.

karl, nice, but the study really needs to be retitled along the lines of "Our super dooper interventions obtund the NF-kB mediated effects of hyperglycaemia in a mouse strain which has been known to have defective insulin secretion since late in the last century". Interpret in normal humans with great caution. In T2 diabetics, perhaps not eating sugar might have the same effect?

Response to medium chain FFAs is normal. It's impossible to over emphasise how important the C57BL/6 mouse is to everything we ever find out about C57BL/6 mice!

George, It's on the agend for when I ever get to pufa and the liver and iron etc!


Peter said...

George, 0.2

Lactate is the same at 0.2 but there is no glycolysis when lactate goes to pyruvate. ie G-3-P dehydrogenase doesn't get a look in to the CoQ couple. You CAN generate glycerol from lacatate but I can't see neurons doing this when they normally run on a controlled lactate supply from glial cells.

Neurons should never be forced to develop insulin resistance. Getting superoxide at complex I from the CoQ couple does them no good in terms of energy control as they never evolved to deal with a BG of 30mmol/l. Insulin signalling is needed for correct ETC function (probably through phosphorylation of ETC proteins) in neurons just as much as elsewhere. So failing to respond to insulin is bad for nerve cells, but does not protect them from caloric overload.

That's the idea at the moment.....


karl, OMG, halibut is going to kill us all!!!!! And did I see peas as a major source of nasties too? Glad Chis M did the donkey work, but really, red meat kills? I get past caring about this sort garbage.

Reminds me of an anecdote about EPIC Norfolk. They trawled their data to show nitrates in the drinking water were carcinogenic in the stomach (we have tons her in agriland) via nitrosamines etc etc. They don't. They are protective. Never publised

Puddleg said...

"Insulin signalling is needed for correct ETC function" I wonder if this has anything to do with insulin upregulating methylation?
Remember Jane on the methylation required for complex 1?
Glucose favours complex 1, insulin promotes methylation. Fat favours ROS and insulin resistance, IR promotes trans-sulfuration, which supplies cysteine for glutathione.
Which will not only buffer the mito ROS but also deal with the H2O2 from the peroxisomal oxidation. Maybe.

Jane said...

George, it looks like insulin is needed for correct ETC function because of its effects on your friend FOXO1 and HMOX1 (haem oxygenase-1), which I always thought was a good guy but which apparently disrupts complex III and IV of the respiratory chain. OMG I've just realised why: these complexes use haem, right? HMOX1 degrades haem. So insulin prevents haem degradation!

'...In line with dysregulated OXPHOS gene expression in the insulin-resistant liver [59], insulin signaling was recently shown to underpin ETC integrity and activity by suppressing FOXO1/HMOX1 and maintaining the NAD+/NADH ratio, the mediator of SIRT1/PGC1α pathway for mitochondrial biogenesis and function 41 and 49. These findings strongly suggest that insulin signaling is required for normal mitochondrial function in metabolism 41, 49, 60, 61 and 62...'
'Insulin signaling meets mitochondria in metabolism'

Puddleg said...

Haem is also needed for catalase, which, for all the hype about glutathione peroxidase, is still the busiest antioxidant enzyme.
Sulfite oxidase also uses heme and molybdenum and transfers electrons to cytochrome C when sulfate is produced.

Paleo Phil said...

Interesting post, Peter. I share your fascination with mitochondria.

Do I understand correctly that the evidence so far suggests:
chronic hyperglycemia -> chronic down-regulation of UCP3 (which is mainly in muscle tissue) -> glucose-induced injury and apoptosis of DRG neurons (and possibly other cells) -> diabetic neuropathy and Alzheimer's (and heart disease, ...)

Other evidence suggests that there are ways to up-regulate UCP3, like fasting (probably especially restriction of glucose and protein), exercise, muscle building, a high fat diet (especially high intake of long chain fatty acids), ATF-1 from hypoxia, heat shock therapy ( and, and sufficient thyroid hormone (especially T3), much of which can be found by Googling "upregulates UCP3". My guess is that most of these measures would be most beneficial when intermittent and acute/vigorous (within safe tolerances), rather than chronic and constant.

There are even suggestions that cold-induced nonshivering thermogenesis may contribute to UCP3 upregulation, though some evidence points to that being a brief, acute effect and that longer term cold actually down-regulates UCP3 (and upregulates UCP1 in brown adipose tissue). It's apparently "a subject of an ongoing debate" ( Any thoughts on this?

It's interesting that scientific research is echoing nature where it suggests that "Heat Treatment Improves Glucose Tolerance" via upregulation of heat shock protein 72 ( and cold exposure increases UCP3, which increases fatty acid oxidation (, This matches well with seasonal food availability.

Apparently, it may be possible to up-regulate UCP3 too far, as one report suggested "a possible relation" between excessive overexpression of UCP3 and "muscle wasting during cancer" (, though it seems speculative and I didn't see corroborating reports.

There is a suggestion that Sprague-Dawley rats may be "may be less than appropriate" for neurological investigations (, though the Vincent et al paper you linked to also references reseach that includes human evidence, albeit also in vitro (such as here:

blogblog said...

the problem was the high level of Provetella bacteria.

The only humans who have high Provetella levels are people who eat a predominantly grain based diet.

Peter said...

Hi Phil,

UCP3 does appear to be a double edged sword... It may help thermogenesis but equally may interfere with elite level athletic performance. But no one would suggest that being an elite level athlete is a recipe for health of course.

I'm really looking at AMP kinase and the PGC-1 alpha, SIRT1 axis in the chronic case.


Paleo Phil said...

Cool, thanks Peter. The double-edged sword is a good analogy for hormesis in general and it coincidentally came to my mind yesterday as I read about hormetic influences on UCP3 and mitochondria (ie, mitohormesis). I look forward to your future posts on mitochondria, as I'm sure other mitochondriacs do.

Puddleg said...

@ Paleo Phil,
I have a hormesis hypothesis;
Toxin X is cytotoxic through ROS, so a good adaptation to exposures is an increase in cellular antioxidant defenses, and increased stem cell differentiation.
But - a better adaptation is to detoxify Toxin X altogether via CYP450 and conjugation. Being a more specific adaptation, and one not required before, this takes longer to evolve. At some point the organism is left, for a time, in the position of being able to both detoxify the toxin and be strengthened by it.

Quote for mitochondriacs:

Hold somebody's hand and feel its warmth. Gram per gram, it converts 10 000 times more energy per second that the sun. You find this hard to believe? Here are the numbers: an average human weighs 70 kilograms and consumes about 12 600 kilojoules / day; that makes about 2 millijoules / gram.second, or 2 milliwatts / gram. For the sun it's a miserable 0.2 microjoules / gram.second. Some bacteria, such as the soil bacterium "Azotobacter" convert as much as 10 joules / gram.second, outperforming the sun by a factor 50 million. I am warm because inside each of my body cells there are dozens, hundreds or even thousands of mitochondria that burn the food I eat.

- Gottfried Schatz, from Jeff's View on Science and Scientists.

Puddleg said...

@ Jane, Schatz on Manganese

You can find all the Jeff's View essays online here:

I found this by following Alexander Arsov's excellent reviews on Amazon.

karl said...


Well what do expect from a Nature paper?

But ignoring the rats - and just looking at the inflammation's effects on aging is interesting.

I'm also taken by this:

Anecdotal - but it reminds me of my grandfather that lived to 101 - ate bacon everyday. While it has been popular to vilify nitrites in the popular press, one should realize that they are potent antioxidants .

The risk with nitrites are they can form nitrosamine's which are carcinogens - but perhaps the risk reward trade off is better than statins?

It would be pretty funny if good research finds that nitrite soaked bacon is a health food...

I really like good bacon..

Unknown said...

Hi there, your blog is really informative. Thanks for sharing.

Jane said...

George, THANKS
I can't wait for Peter to get interested in manganese. Please Peter, read what George linked. You will love it.

One thing Schatz doesn't mention is that divalent manganese is extremely magnetic. 5 unpaired electrons! I've thought for years that the reason mitotic spindles look just like drawings of dipole magnetic fields is because the microtubules have bound manganese. In vitro, manganese will bind tubulin and polymerise it.

John said...


You sure do love manganese--and copper. Out of curiosity, what do you then eat? Have you restricted iron for a long time?

Jane said...

Hi John
I eat a Hunza diet. You can read about the Hunza and their extraordinary health here.

Conventional thinking says deficiencies of iron and zinc are common, and deficiencies of their partners manganese and copper are rare. Evidence is coming out all the time that this is not correct.

John said...

Thanks for the link Jane. The diet seems pretty similar to some others'. I sure do like sour milk products. Regarding Mn, I would guess coconut cultures, so especially Tokelauns, would have a high intake, but of course many variables are different there.