Sunday, September 15, 2019

The paradoxical fat mice

This paper is very interesting. I think I picked it up via Raphi on twitter. It comes from Jim Johnson's lab.

Caloric Restriction Paradoxically Increases Adiposity in Mice With Genetically Reduced Insulin

The background is in these two papers:

Phenotypic alterations in insulin-deficient mutant mice

Compensatory Responses in Mice Carrying a Null Mutation for Ins1 or Ins2

The paradoxical mice all had the Ins2 gene fully knocked out and in addition to this some mice also had one allele for the Ins1 gene knocked out (Ins1+/-). So the mice in the study had either a half or a quarter of the normal mouse insulin gene complement. Some mice were fed ad-lib, some were 40% calorie restricted (CR).

The CR, lowest insulin gene group (Ins1+/-) had significantly elevated total fat mass and a significantly elevated percentage of bodyweight as fat. That's a paradox to the insulin hypothesis of obesity and so really interesting. The Ins1+/+ group also had a (ns) increase in percentage body fat but not in absolute fat mass, so the trend is there too, but only a trend.

Metabolically, the split is between ad-lib vs CR groups.

All mice had the same maximal insulin response to a 2g/kg intraperitoneal glucose tolerance test but the CR groups had a very significantly reduced peak and AUC for glucose, ie they were much more insulin sensitive. The intra-peritoneal insulin tolerance test result might be worth a post in its own right, it's paradoxical too but there won't be space to cover it today.

So let's have a look at energy expenditure (EE) from Fig3 C.

To make things a bit clearer I've copy pasted the light period from the left half of the graph on to the end of the dark period to give more of an idea of the EE curves are really like during dark to light transition:

The red line starts horizontally with no significant difference in EE between ad-lib fed mice or CR mice. There is a modest increase during the dark (active) period when the CR mice get their three very small meals, as indicated. After the last meal a precipitous and highly significant fall in EE occurs. The mice enter torpor, a state of extreme lassitude and hypothermia. At around two hours in to the next light period the mice wake up and EE returns to just below that of the ad-lib mice and the cycle repeats. The ad-lib mice behave like normal mice.

The CR mice have a profound hypometabolic period every day. You could argue, if you are a cico-tard, that this is why they store excess fat. They eat all the food they can get but expend relatively little energy so they become fat: CICO. But I would disagree.

Here's my guess as to what is happening. Speculation warning.

We know that the CR mice are exquisitely insulin sensitive. They are that way because they have a low number of insulin genes and they never get enough food to trigger a major insulin spike anyway. The CR is the dominant factor but it needs the genetic background to get the paradox to occur. Insulin-induced insulin resistance, acute or chronic, does not occur due to lifetime low insulin exposure. The fact that all mice are capable of producing the same maximal insulin response to an IPGTT does not mean that the CR group experience an equivalent insulin exposure to the ad-lib group during their routine lives. They never get enough food to trigger a maximal insulin response.

The CR mice spent the bulk of the light period with a slightly low EE. Dark period arrives and with it food. As the food is eaten there is an upward trend in EE followed by a drop. The second small meal arrives, again an upswing followed by a drop. The third and final meal gives the same upswing but the drop in EE which follows just goes on downward. The mice enter torpor, a state of profound lassitude and hypothermia.

I think torpor happens because the mice simply have no accessible calories.

This is despite the fact that it occurs immediately after the third of their calorie restricted meals. Their problem is that the meals generate an insulin response. The mice are so insulin sensitive that calories are lost in to adipocytes (and probably hepatocytes) under the over-effective action of insulin.

They lose calories in to adipocytes. These are calories out. The adipocytes get bigger with the lost fat.

Torpor occurs BECAUSE the mice have become fatter.

This is the equivalent of the hunger which follows for a human under a euglycaemic (or even hyperglycaemic) hyperinsulinaemic clamp. There is no hypoglycaemia but fatty acids become locked in to adipocytes by the hyperisulinaemia and hunger follows due a lack of available calories. I posted about it here.

At two hours in to the light period insulin drops low enough to allow lipolysis. The mice wake up.

That's all.

Except: Why do the CR mice have paradoxically (although ns) elevated fasting insulin cf the ad lib mice? There are two reasons. Here are the blood sample times added to the EE graph. The arrows are not quite in the correct clock times, as detailed in the methods, but the times related to feeding/metabolism are approximately correct.

The green arrow is the sampling time for the ad-lib fed mice. It is about six hours in to the light period and the mice would normally have been asleep during the hours leading up to it. Light-period snacking, from the respiratory exchange ratio (RER) graph in Fig3 D, would not normally have started by this time so it's a very simple physiological fasting sample.

The blue arrow for the fasting CR mice just hits the end of torpor. I'm not sure these mice ever have a time when they wouldn't eat, given the chance, but here they are in their hypometabolic phase and have minimal access to calories. At this time insulin is actually a little (ns) higher than for the ad-lib groups (Fig1 D). Higher insulin means fat stays in adipocytes. Why is insulin high?

Calorie restriction does many things in addition to dropping metabolic rate. If you fast a hard working group of humans for 5 days they develop a post prandial increase in GIP (glucose-dependent insulinotropic hormone). This was found in the CR mice in both the fasting and fed state (Fig4 C). GIP facilitates insulin release, hence insulin is a little higher the CR mice and loss of calories in to adipocytes more severe, necessitating torpor.

It's interesting as to why GIP might be elevated under hunger conditions. Possibly generating and saving fat becomes a priority when calories are low. The Ins1+/- CR mice certainly have the highest RER (>1.05) after their third meal, suggesting that they prioritise the conversion of glucose to fat. This DNL, should it occur in the liver, might go some way to explaining the elevated triglycerides in both of the CR groups. Maybe. Accentuated DNL in people who have undergone massive weight loss via gastric bypass surgery is routine during an OGTT. Like these people.

Anyway, I'll stop now. This post is about 1/4 the length it started out as, so if corners seem a bit cut then mea culpa.


Thursday, September 05, 2019

Hyperlipid Protons ambassadors

Many people may have noticed that the blog Hyperlipid is not exactly the most user friendly of blogs. The prose clearly appears to make sense but some of the concepts are not always particularly simple unless you have the Protons idea well understood.

Last year (2018) Mike Eades made a sterling presentation which summarised the concept in a talk at Low Carb Down Under in terms that were much more accessible

and Brad Marshall now has a blog on which, throughout 2019, he has been writing around ideas which derive partly from the Protons thread on Hyperlipid. But in significantly more user friendly language, while still being on the spot.

Fire in a Bottle

is his website, a very neat name. It's good. He farms low PUFA pigs too.


*Very few things in life are quite so disappointing as finding that insulin interacts directly with the NOX4 (NADH oxidase 4) complex to generate the bulk of the initiating low levels of superoxide/H2O2 which trigger insulin signalling. Some ROS do come from the electron transport chain but NOX4 seems rather important. Sigh. Bulk ROS to terminate/blunt insulin signalling do appear to be ETC derived...

Sunday, September 01, 2019

The sweet taste of DMEM

I mentioned in the last post that few papers ever specify the glucose concentration used for cell culture. This came up in comments from Alex and I think it deserves a mention in a post of its own.

The standard methods description for almost all cell culture usually specifies that DMEM was used, along with assorted ancillary chemicals and a source of growth factors, usually foetal bovine serum, but rarely the glucose concentration.

A brief trip to any commercial supplier's website gives you approximately 50 DMEM formulations. Three or four have zero glucose and are intended to allow you to specify your own glucose concentration. Another five or six use the original 1g/l of glucose giving 5mmol/l, ie a physiological glucose concentration. This is described as "low glucose".

The other forty-odd specify 4.5g/l, ie 25mmol/l.

I've never done cell culture but I gather the normal technique is to use 25mmol/l medium, let the cells use the glucose to grow and when the glucose drops toward 10mmol/l then they are re-fed with new medium at 25mmol/l of glucose. This works. For decades.

It also provides you with enormous information about the behaviour of cells under hyperglycaemic conditions.

Obviously, 25mmol/l of glucose in an intact human is pure pathology but it is well tolerated by the cells in culture because there is no other caloric source which inputs as FADH2. No free fatty acids.

At the most basic level a glucose concentration of 25mmol/l should a) never occur at all and b) if it does occur it should promptly trigger, via insulin acting on adipocytes, a fall in FFAs to 100 micromol/l or less.

This graph is from a paper on pancreatic beta cell death. From the open circles it is clear that, with glucose held at physiological concentration of 5mmol/l (G5), palmitate is harmless at up to at least 400micromol/l.

This should surprise no one because a 60 hour fast in a healthy human will provide 2000micromol/l of mixed free fatty acids with a glucose of just under 5mmol/l. And no multiple organ failure.

Compare that with the closed circles where glucose is pathologically high at 20mmol/l. Physiological fasting levels of FFAs then unmask the gross pathology of glucose at 20mmol/l. Typical of ordinary cell culture concentration.

Given the idiotic stand of the cardiologist community against saturated fat and given the apparent safety of pathologically elevated glucose in fat-free or PUFA supplemented cell culture medium it should come as no surprise that very few labs worry about the consequences of their grossly non-physiological DMEM.

Demonstrating the pathology of hyperglycaemia using (and blaming) saturated fatty acids or the converse lack of acute toxicity from PUFA (low FADH2 input) is a route to funding which would surely discourage any questioning of the techniques of cell culture which have been successfully used for decades. As folks say:

"We've always done it that way".

That's all I really wanted to say but here are a couple of odds and ends:

There are groups for whom glucose matters. Even if stuck in the incorrect complex I inhibition paradigm for metformin's action, the concentration of glucose is recognised as important by some.

This group:

Metabolic determinants of cancer cell sensitivity to glucose limitation and biguanides

have developed a system for glucose homeostasis in cell culture

Their approach is pretty well unique. It is NOT how cell culture is normally done.

The other brief aside while I'm on cell culture from an outsider viewpoint is FBS, foetal bovine serum. This contains essential growth factors. You can pretty well translate it as "insulin" or "IGF-1".

When cell culture shows that metformin works to reduce cancer cell growth, it is working in the presence of insulin signalling (usually combined with hyperglycaemia). Which metformin blocks at the glycerophosphate shuttle level, without the need for a lethal blockade of complex I.


Wednesday, August 28, 2019

Who gave you pancreatitis?

Background: Propofol is a mainstay anaesthetic induction agent. Its use is associated with occasional pancreatitis episodes. I won't wander aside about the poor dog which was referred for cataract surgery and which did eventually recovered from its perioperative fulminating pancreatitis.

Propofol is dissolved in what is essentially Intralipid, a soybean emulsion used for parenteral nutrition. The lipid emulsion gives a transient hypertriglyceridaemia. Hypertriglyceridaemia from any cause is associated with pancreatitis.

Me, being me, would automatically blame the PUFA in the Intralipid. But then I would.

I came across this paper by accident this morning:

Distinctive roles of unsaturated and saturated fatty acids in hyperlipidemic pancreatitis

It's good. The group even give you the glucose concentration used in their cell culture, 0.2%, ie 200mg/dl or around 10mmol/l. Not normal but hardly seriously pathological.

Aside: Less innocent groups keep quiet about glucose concentration and can reliably show endoplasmic reticulum stress and any other nasty attributable to palmitic acid. With how much glucose??? End aside.

This is what they found:

"Unsaturated fatty acids at high concentrations but not saturated fatty acids induced intra-acinar cell trypsin activation and cell damage and increased PKC expression"

So. If you have a genetic hypertriglyceridaemia, say lipoprotein lipase deficiency, and you get acute necrotising pancreatitis (not fun) there is every possibility that it was induced by the high content of polyunsaturated fatty acids in those triglycerides and the FFAs derived from them.

Which means that your cardiologist put you in the ITU. Avoid saturated fats, replace them with polyunsaturated oils. Thank you Public Health England and your equivalents world wide.

The converse might well be that loss of the gene for lipoprotein lipase, or similar loss, might not have been a big deal when humans lived by eating elephants. Or even until corn oil took off as a cholesterol lowering scam.


Protons (50) The video

Dave Speijer has a great video up on YouTube based largely around his recent paper in BioEssays.

Can All Major ROS Forming Sites of the Respiratory Chain Be Activated By High FADH2/NADH Ratios?

It's nice to hear the man himself on a subject about which he has spent a great deal of time thinking. Here it is:

Many thanks to Andrew Moore, editor-in-chief at BioEssays and publisher of many of Dr Speijer's papers, for the heads up that this video had been produced.


Monday, August 19, 2019

Protons (49) Complex III

Dave Speijer, an extremely insightful person if ever there was one, has a new paper out:

Can All Major ROS Forming Sites of the Respiratory Chain Be Activated By High FADH2/NADH Ratios?

the link to which I am extremely grateful to Bob for forwarding to me. This concept is purely from Dr Speijer. But I like it. A lot.

I'll start with an old doodle I produced about a decade ago depicting the front end of the electron transport chain. Matrix is at the top, cytoplasm at the bottom:

Electrons travel from NADH to Coenzyme Q, reducing it to QH2 which donates them to complex III being oxidised back to Q in the process.

The unlabelled blobs in the diagram are complex II (succinate dehydrogenase), electron transporting flavoprotein dehydrogenase and mitochondrial glycerophosphate dehydrogenase, all of which compete with complex I for CoQ as an electron acceptor. Given a high membrane voltage, a deeply reduced CoQ couple with most of the CoQ as QH2, then reverse electron transport back through complex I gives superoxide/H2O2 generation (provided the mitochondrial NAD pool is also highly reduced so unable to accept electrons (ie little NAD+, lots of NADH).

This is pretty straight forward and is the gist of the Protons thread. The extension of this is that, under high substrate availability, H2O2 from this process stops insulin facilitated caloric ingress to the cell.

Another major site of ROS generation in the ETC is complex III. Dave Speijer would hold that this is also triggered, like RET, by a deeply reduced CoQ couple. This is why.

So here is a stripped out version of the above doodle:

The problem here is that it's not that simple. Complex III does rather odd things with its two electrons. Electron bifurcation is a standard enzymic technique perfected very early on by proto-biology and it is exactly what happens here. One electron travels to cytochrome C (which only ever carries one electron at a time) and then on to complex IV and O2. The other does not:

The second electron is transferred backwards to another  CoQ molecule and so partially reduces it to QH*, the radical semiquinone.

So (oxidised) Q is a necessary electron acceptor for complex III. Under high substrate availability and with most of the CoQ couple present as QH2 there will be very little Q available.

With one electron securely on cytochrome C, with any delay in the availability of Q, the second electron can be left sitting on one of the two haem groups along its route, easily available for donation to O2 to give superoxide, so adding its ROS signal to that of complex I. Both indicate that enough substrate is present and it is time to limit insulin signalling, to limit caloric ingress.

Nothing, absolutely nothing, about the construction of the ETC is random.

The general principle that a highly reduced CoQ couple is a signal to halt caloric ingress in to the cell applies to complex III just as much as to complex I.

When you want to resist insulin, you really want to resist insulin. Superoxide is then your friend.

If you fail to limit caloric ingress then eventually ROS from complex III are ideally placed to break the cardiolipins which anchor the water soluble cytochrome C to the inner mitochondrial membrane. Destroy these anchors and the ROS signal changes from"resist insulin" (good) to "perform apoptosis" (possibly not quite so good)...

And of course ROS in excess of physiological signalling are going to activate all sorts of inflammatory pathways.


Aside: The process of ROS generation is probably limited by pairing of complex IIIs (complex III is always a dimer in-vivo) for cooperation to use one Q to accept an electron from each of the pair. This should limit ROS generation as one Q will be available to two complex IIIs, twice that avaiable if they were each working alone. Nothing is random. End aside.

Tuesday, July 23, 2019

On belief structures in lipidology (2) KODA-CP

Altavista put up an excellent link to an article describing an AI algorithm capable of mining pubmed abstracts and coming up with gold. The exact opposite of modern meta-analysis (repeat after me: the meta-analysis of dross is dross). And the AI algorithm could use old data to predict recent discoveries! Sadly, the university I graduated from is currently teaching final year vet students that any publication over five years old is unimportant/can be ignored. The end of science. Alt was prompted to share his gem following this link put up by Hap:

C‐reactive protein promotes atherosclerosis by increasing LDL transcytosis across endothelial cells

Which is worth thinking about (and ignoring). Unless anyone thinks that LDL transcytosis across human umbilical vein derived endothelial cells (HUVECs) in culture resembles the process of arteriosclerosis, I think we can safely ignore this aspect of the modelling in the paper. Just ask yourself how severe is venous arteriosclerosis, with or without C-reactive protein. And the cells in the model die if you expose them to anything greater than 35mg/ld LDLc! But it brings up the apoE-/- mouse, a truly fascinating subject.

The apoE-/- mouse is also a model. They develop hyperlipidaemia and rapidly progressive "arteriosclerosis". Obviously, the apoE-/- lipid particles, which lack their apoE attachment protein, work their way (using active transcytosis of course) through the endothelium of blood vessels and cause cholesterol to accumulate in the subendothelial space and... Well. It's a pretty neat evolutionary dead end, if you believe it.

The apoE-/- model tells us more than anyone could ever want to know about apoE-/- mice. I've just spent some considerable time on Pubmed and SciHub trying to find out if there is any sort of full blown apoE-/- syndrome in humans. Not apoE2/E2 etc, more like no apoE at all. Zero. Zilch. Like the mice.

No luck finding it so far.

EDIT Yay, Adam found it

Effects of the absence of apolipoprotein e on lipoproteins, neurocognitive function, and retinal function

No suggestion of CVD. Just shows how important adding over 1% by weight of dried (oxidised) egg yolk to the diet to generate the "model" might be!


But apoE-/- mice are interesting in their own right. They really do accumulate lipids on their arterial linings in a manner exuberant enough to make a lipidologist wet their knickers. None of this messing about with the non-lipid intimal thickening so characteristic of real human arteriosclerosis. Lipids, lots of them, just "invade" the arterial walls and stick. Right there, pretty well on the surface.


Every now and then you trip over an interesting paper, in this case about why apoE-/- mice really have vascular problems. Here's one:

TLR2 Plays a Key Role in Platelet Hyperreactivity and Accelerated Thrombosis Associated with Hyperlipidemia

The paper is long and complex and very, very clever. As per usual. And less than five years old.

Here is the scene-setter from the discussion

"Patients with enhanced platelet reactivity are at increased risk for cardiovascular events.4, 37–39 Enhanced platelet reactivity is associated with chronic and acute inflammation, infections, diabetes, and a number of pathophysiological states related to dyslipidemia, including atherosclerosis, diabetes, and metabolic syndrome". My italics.

The mechanism appears to be through CD36, a multifunctional scavenger-type receptor present on most cells but here they are looking at platelets:

"Previously we have linked platelet hyperreactivity in dyslipidemia to accumulation in circulation of specific oxidized phospholipids, oxPC-CD36, which activate platelets via the scavenger receptor CD36"

The "previously" citation is to:

Phosphoproteomic Analysis of Platelets Activated by Pro-Thrombotic Oxidized Phospholipids and Thrombin

which introduces us to KODA-CP, or to put it more elegantly 9-keto-12-oxo-10-dodecenoic-phosphatidyl choline. This was just one of the more effective CD36 activators of the many lipid products present in oxidised lipoproteins.

So. Hyperlipidaemia facilitates the generation of KODA-CP which activates CD36, which activates TLR2, which makes platelets super sticky.

KODA-CP must have linoleic acid or arachidonic acid as part of its parent molecule.

The platelets stick. In apoE-/- mice enough of them stick to form massive aggregates on the arterial surface that look a bit like late stage lipid infiltrated arteriosclerosis plaques. It's a model.

BTW, platelets carry apoB labeled lipoproteins, among the many physiologically appropriate contents of their cytoplasmic granules (link below). Under the more normal generation of arterial intimal hyperplasia which precedes pathology I consider this lipid will simply be used for normal repair/hyperplasia processes. But there is nothing physiological about apoE-/- mice. They look like they should stick a ton of platelets to any damaged vascular wall, with more apoB labels than any-(mouse)-body knows what to do with. Given enough omega 6 PUFA to generate the KODA-CP.

Apart from the cardiologist derived omega 6 PUFA (another link below), did you notice the core involvement of the CD36 receptor? CD36 also facilitates free fatty acid uptake in to many cells. It is stored within cells and translocates to the cell surface, a bit like GLUT4 proteins, when needed. Stored in the cell, translocated when needed.

What controls CD36 translocation to the cell surface?

Insulin, of course (another post there).

Just thought you might like to know.


Let's just summarise. Lacking the apoE protein limits the utilisation of lipoproteins, much as having a fully non functional LDL receptor does in familial hypercholesterolaemia. This increases the concentration of lipoprotein particles in the circulation.

Applying the Dunning-Kruger effect to lipidology: Lots of LDLc particles = lots of invasion. QED. That's been it for the last 50 years. I don't thing many lipidologists every get past this obvious, unarguable, simple fallacy. Oh, also core to lipid "therapy" has been, and still is, giving corn oil to lower the LDLc count.

In reality the elevated lipoproteins are a marker of reduced utilisation and are associated with an increased residency time in the circulation. Given lipids based on palmitic, stearic or oleic acids I don't think that would matter.

Given lipids filled with linoleic acid, the essential precursor of KODA-CP, you will get a progressive rise in KODA-CP associated with increasing persistence of the lipoproteins. The more KODA-CP the more activation of platelets via CD36/TLR2 (and undoubtedly other pathways) and the stickier the platelets become.

Given the pathological intake of linoleic acid promoted by cardiologists and lipidologists working under their cholesterophobic hypothesis it seems perfectly possible that seed oils (and insulin) may well be drivers of the platelet adhesion which is core to the vascular damage in apoE-/- mice. Platelets even carry apoB100 labeled lipoproteins in their cytoplasmic granules which allows us to immuno-stain lipid accumulations with this LDLc implicating flag.

Given lipoproteins which lack apoE on their surface, accumulation of KODA-CP, hyperreactive platelets and a surfeit of insulin we are in a position to understand how the apoE-/- mouse works. Which is cool for those of us who like to understand things.

How much of this applies to actual human arteriosclerosis? Increasing platelet stickiness will amplify the normal response to arterial injury. I think this may be real. The rest is just a very extreme, rather bad model.

Most models, like this one, are usually useless.

Is it conceivable that cardiological dietary advice represents the exact opposite of the correct approach? That it would actively worsen the problem it is aiming to ameliorate?

Yep. But we knew that anyway.

Increasing linoleic acid in the diet is undoubtedly a facilitator of the generation of KODA-CP and the activation of the subsequent cascade goes a long way to explain the Sydney Diet Heart Study and the Minnesota Coronary Experiment. People died. From corn oil.

I'll stop now.

Some helpful links that didn't integrate neatly in to the text.

Effects of saturated and polyunsaturated fat diets on the chemical composition and metabolism of low density lipoproteins in man (1980, written on papyrus)

Apolipoprotein B release from activated human platelets (1986, probably on parchment, safe to ignore).

Monday, June 24, 2019

On belief structures in lipidology

Dr Thomas "Just-take-the-statin" Dayspring writes on twitter:

"Any apoB-lipoprotein less than 70 nm in diameter can pass be pass thru endothelium - The LDLs are 20.5-25 nm. Remnants and IDLs are less than 70 nm and greater than 30 nm. The term small, dense LDL is way too simplistic - Big LDLs, like small LDs (less than 20.5 nm) if present in excess can invade artery"

This statement makes a prediction. It predicts that the accumulation of lipid in arteriosclerosis will be, initially, sub-endothelial.

As in this review of transcytosis (because passive "leakage" of LDLc down a concentration gradient across an endothelial cell layer is laughably impossible for particles over 6 nm across. No, that 6 nm is not a typo, according to the review). Not that "impossible" means anything to a lipidologist.

"During the initial stages of atherosclerosis, LDL particles are transported [transcytosed] across the EC [endothelial cell] barrier and accumulate in the subendothelial space".

So. All we need to do is a few post mortem examinations, find some poor people who had early arteriosclerosis at the time they died, and look for that lipid which will be sitting neatly under that single layer of endothelial cells lining their arteries. That is the prediction embedded in Dayspring's tweet.

If we go to this paper:

Early Human AtherosclerosisAccumulation of Lipid and Proteoglycans in Intimal Thickenings Followed by Macrophage Infiltration

EDIT: from a link in Subbotin's excellent "Excessive intimal hyperplasia in human coronary arteries before intimal lipid depositions is the initiation of coronary atherosclerosis and constitutes a therapeutic target" END EDIT

we can find real images of real arteries from real people who died of non related causes while carrying different levels of arteriosclerosis:

Left side images are van Giesen stained for histology, central images are with Sudan IV for lipid and right hand are immunostained with anti-CD68 antibodies to show macrophages. The pairs of small black arrowheads indicate the level where the intima stops and the outer media layer begins. The vascular endothelium is a single cell layer at the top of each image.

Let's look at the circled image in detail. This is an example of early atherosclerosis from a real human with real early changes who died of non related causes. It's just the sort of place you might hope to catch an LDLc particle creeping between the cells of a single endothelial layer or freshly spat out after transcytosis by an endothelial cell. Lipid stains bright red:

Well, there's the lipid, deep, deep down at the junction of the intima and the media, right between the arrow heads...

There is none anywhere near the endothelial cell layer. If you believe that LDLc, as a result of a concentration gradient between artery and sub endothelial layers, "moves" or "invades" across that endothelial cell layer you have to explain how there is none at all in the sub-endothelial area and there is a progressive accumulation at the intima-muscularis junction. How does the lipid get from the top of the image to the deep spaces without any of it showing up in the lipid-free zone between the two?

"Beam me down, Scotty" is undoubtedly the most plausible explanation.

It is very, very hard to explain how utterly disreputable the lipid hypothesis is. All of this angst about increased LDLc and/or apoB counts on LC diets is based on the assumption that somewhere, somehow, cholesterol is the cause of heart disease. How LDLc "invades" (by active and controlled transcytosis!) the sub-endothelial space, disappears from there and then suddenly appears at over 200micrometres deeper, with none showing in the intervening zone requires a belief tenet which bears no resemblance to reality...

This was bollocks in the 1950s. My question is, as always, at what time did it stop being bollocks?

No one would reasonably doubt that the lipid deep down at the intima/media junction level comes from lipoproteins (though there are other plausible explanations). No one would doubt that loading the lipoproteins with with linoleic acid is likely to be a Bad Thing. No one would doubt that generating oxidative derivatives of the lipids in those lipoproteins might be a Bad Thing.

But trans-endothelial "invasion" is beyond belief.

This would suggest that all lipidologists are talking crap.

Nothing new there then.


Tuesday, May 28, 2019

Metform (11) metformin vs mtG3Pdh knockout

This paper is interesting (and badly written):

A high carbohydrate diet does not induce hyperglycaemia in a mitochondrial glycerol-3-phosphate dehydrogenase-deficient mouse

It uses a mtG3Pdh knockout mouse, which is essentially a mouse which behaves as if it was on an enormous dose of metformin without all of that toxic blockade of complex I which gives a potentially lethal lactic acidosis at high dose rates. If you feed these mice standard crapinabag they are phenotypically normal. If you feed them a diet consisting some casein, a little PUFA to avoid EFA deficiency and the rest of the calories from pure sucrose they become rather interesting.

Eating pure sucrose does not make normal mice fat. It does make them insulin resistant and hyperinsulinaemic and, of course, insulin resistant adipocytes refuse to retain fat unless insulin action is facilitated by the oxidation of PUFA. Hence the normal body weight.

But the knockout mice actually become slim on sucrose. Here are the data, we can ignore the heterozygous (HET) groups:

They are slim because insulin levels are low. From the Protons perspective the function of the glycerophosphate shuttle in the pancreas is to drive enough reverse electron transport through complex I to trigger insulin release. Less RET, less insulin release, less fat storage, less hunger. Here is the isolated response of the perfused pancreas model to hyperglycaemia:

First phase insulin release is about a third of that in the normal mice. The mice are not diabetic because, in the absence of the glycerophosphate shuttle, RET to allow insulin signalling is generated by beta oxidation supplying electron transfering flavoprotein the CoQ couple via mtETFdh. Insulin signalling still happens but at the "cost" of increased lipid oxidation in the peripheral tissues.

What doesn't happen is sucrose induced insulin resistance. Again I consider this is triggered via the glycerophosphate shuttle causing RET at a level to shut down insulin signalling, which simply doesn't happen in the knockout mice. Lack of glycerophosphate shuttle also stops the generation of insulin-induced insulin resistance under conditions of high insulin concentrations coupled with energy replete cells.

Does anyone recall this figure from Metformin (01) post back in 2017?

Insulin was given at 90 minutes. At 150 minutes in the upper (non metformin-ed) rats insulin action starts to fail, at about the correct time for insulin-induced insulin resistance. By 180 minutes that upper trace, the non-metformin group, shows an upward trend in glucose as exogenous insulin levels are no longer high enough to overcome insulin-induced insulin resistance (the rats are DM T1 under insulin withdrawal).

At 180 minutes in the lower line showing metformin treated rats we can see the continued action of insulin being facilitated by the metformin because it blocks insulin-induced insulin resistance. It was mention in the comments to the post that, clinically, this effect of metformin might worsen the possibility of hypos in humans if combined with exogenous insulin usage. Potentially fatal hypos.

So what happens if you inject a sucrose treated mtG3Pdh knockout mouse with exogenous insulin to check their insulin sensitivity? Insulin sensitivity is preserved, to fatal effect:

All of the mice with the mtG3Pdh knockout died under exogenous insulin. This is exactly how I would expect metformin to behave in humans using insulin. A functional glycerophosphate shuttle allowed a sucrose diet to block this fatal sensitivity to exogenous insulin.

Obviously the mtG3Pdh mice have a normal complex I. Might they still develop lactic acidosis? Sadly the group didn't look at this (they had no idea back in 2003 that they had developed a meformin mimic model mouse). I do think there might be some elevation of systemic lactate despite a normal complex I.

In the absence of the glycerophosphate shuttle glycolysis is going to run directly to lactate to maintain redox balance. If glycolysis proceeds at a rate in excess of oxidation of the lactate within mitochondria (recall oxphos is slow compared to glycolysis) then some glycolytic lactate will spill outwards, though this is never likely to reach ICU-needing levels. No need to have a complex I blockade to generate mild lactic acidosis.

Does this metformin-ed like mouse have the exercise gains seen in human cyclists after popping 500mg of metformin pre-race?

That requires that we look at a different model.


Wednesday, April 17, 2019

Life (31) Chinese whispers

Back in 2008 Noha Mesbha published her excellent PhD thesis


which introduced the world, via this paper

The halophilic alkalithermophile Natranaerobius thermophilus adapts to multiple environmental extremes using a large repertoire of Na+(K+)/H+ antiporters

to Natranaerobius thermophilus and its antiporter nt-Nha. Which gives every impression of being a stand alone Na/H+ antiporter very closely related to the invariably operon controlled MrpA protein (named shaA) of Clostridium tetani. As she says

"Gene nt-Nha had 35% identity to the shaA (mrpA) gene of Clostridium tetani. The Mrp proteins belong to the monovalent cation/proton antiporter-3 protein family. ...Sequence analysis of the regions surrounding gene nt-Nha, however, did not show that it was part of an operon. This indicates that gene nt-Nha does not encode a subunit of an Mrp system, but rather a mono-subunit antiporter".

All well and good.

Then in 2010 Morino et al published

Single Site Mutations in the Hetero-oligomeric Mrp Antiporter from Alkaliphilic Bacillus pseudofirmus OF4 That Affect Na+/H+ Antiport Activity, Sodium Exclusion, Individual Mrp Protein Levels, or Mrp Complex Formation

Although the whole paper was about B subtilis (and how none of its Mrp subunits worked in any incomplete combination to antiport anything) they did make this throw-away comment:

"A MrpA/MrpD homologue encoded by a “stand alone” gene from polyextremophilic Natranaerobius thermophilus was recently reported to exhibit Na+/H+ and K+/H+antiport activity in anaerobically grown E. coli KNabc (24)"

where (24) is Mesbha's PhD paper. Notice that at this stage Mesbha's

nt-Nha ~ shaA, very closely related at 35% conserved gene sequence,

has been changed by Morino in to

nt-Nha ~ An "MrpA/MrpD homologue".

This is a just about acceptable per se because MrpA and MrpD are homologous to each other and nt-Nha is closely related to MrpA (shaA) of C tetani. But Mesbha herself never mentions MrpD in her 2009 paper or in her PhD thesis. And "MrpA/MrpD" is open to mis-interpretation. So we have "definition-creep" here, where nt-Nha could be accidentally seen as some sort of composite of MrpA in combination with MrpD. Ouch.

So next we have the 2017 offering by Jasso-Chávez et al

Functional Role of MrpA in the MrpABCDEFG Na+/H+ Antiporter Complex from the Archaeon Methanosarcina acetivorans

where we have this bizarre statement

"On the other hand, a fused MrpA-MrpD homolog in the alkaliphilic Natranaerobius thermophilus displayed Na+/H+ antiport activity when produced in E. coli strain KNabc (5, 28)"

Ref (5) is Morino's paper on B subtilis Mrp, in which one rather misleading citation suggests that nt-Nha is an "MrpA/MrpD homologue". This has developed to the extent that nt-Nha has now "become" a fusion of two genes to give a rather mythical monster.

Ref (28) is just Mesbha's PhD paper in which nothing of the sort was suggested.

So the Jasso-Chávez paper is utterly flawed due to misinterpreting a poorly phrased statement and adding an erroneous modification so as to grossly mis-represent an initial very solid finding by Mesbha. The Jasso-Chávez discussion of nt-Nha can be distilled as:

"Send three and fourpence, we're going to a dance".

The chance of their understanding how nt-Nha or their very own archaeal MrpA subunit work as a stand-alone antiporter appears to be approximately zero.

Very sad.


BTW The folks who worked from the actual gene to model the nt-Nha protein structure suggest that

"The final model presents 13 transmembrane α-helices organized in a similar arrangement as the NuoL subunit".

You know the picture but here it is again 'cos I think it's lovely

Sunday, April 14, 2019

To the gym?

This press release in advance of a poster presentation is doing the rounds on social media at the moment:

Ability to lift weights quickly can mean a longer life

A factual association is probably true between being above the median in ability to do work on a specifically selected gym machine and longevity.

What does this mean? It's observational. Hypothesis generation only.

Perhaps non-athletes with higher power output might live longer because they have excellent mitochondria and our mitochondria are essentially what determine our longevity. Having good mitochondria might well mean that you exercise spontaneously without describing yourself as an athlete. As in I might carry a couple of sacks of chicken food myself rather than wait for Paul the yard-man and his barrow. But this is an effect, not a cause.

Non-athletes with low power output may be the converse. They have mitochondria in which the cardiolipins anchoring their cytochrome-C to the mitochondrial inner membrane are as flimsy as a PUFA in a deep fat fryer, which are willing to trigger apoptosis at the drop of a superoxide molecule. Poor mitochondria, less muscle fibres giving sarcopaenia, shorter longevity. Probably already giving diabetes in-situ. An AHA poster-child.

Taking someone from the second category and making them exercise might convert them to being healthy with a long life span. Maybe.

Or, there again, it might make the gym sessions so unbearable that they quit.

Or, if you don't let them quit, it might simply make no difference.

Or, if you don't let them quit, it might kill them sooner!

As Eeyore said "think of all the possibilities, Piglet, before you settle down to enjoy yourselves".

The study (as per press release) tests nothing. Researchers are giving very specific advice based on an untested observation which might be going to make them look very, very stupid when the results of an intervention to test their hypothesis actually refutes it. But that will take decades and hopefully the researches will be retired by then. More likely it will be ignored or termed a paradox. All the possibilities Piglet.......

EDIT or the intervention might show benefit. That's not impossible! END EDIT


Declaration: I have nothing against exercise. Nowadays I mostly boulder because it is three dimensional problem solving using muscle groups to failure on a regular basis. Plus I also try to keep my cardiolipins as saturated as practical!

Life (30) Guesses about Na+ channels

Complex I exists in various states, the two main ones being activated and deactivated. The deactivated form is convincingly a Na+/H+ antiporter. There is a pretty good case made in this paper, which has a number of flaws but is generally probably correct:

The deactive form of respiratory complex I from mammalian mitochondria is a Na+/H+ antiporter

As always, the suspect antiporter is thought to be NuoL, the distal "antiporter-like" subunit. You can see the logic for this which is supported by the stand-alone antiporting homologues of NuoL seen in nt-Nha and at least one archaeal MrpA subunit. Personally I'm not sure this is the case. The phenomenal difficulty in trying to interpret exactly what is happening in a structure as intricate and as minute as complex I allows many views of the available data.

It also appears to be the case that in bacteria which use menaquinone as their electron acceptor (rather than CoQ) some degree of Na+/H+ antiporting occurs under normal active NADH oxidation/proton pumping. That's in here

Respiratory complex I: A dual relation with H(+) and Na(+)?

and here's the energetics doodle from the same Fig 2 as I pinched the NADH:ubiquinone doodle from in the last post:

Because transferring two electrons from NADH to menaquinone only provides 480mV/2e- the complex uses Na+ moving down its concentration gradient to "top up" energy availability and so get the extra energy needed to pump the full four protons.

Aside: The paper has lots of good ideas but they are very wedded to the concept that NuoL, M and N are still antiporters and that loss of "control" by the redox cytoplasmic arm allows this antiporting to re establish. It's a very reasonable idea but I think it can be improved upon, especially now we have more detailed information about the Na+ pumping of the P furiosus MBH, where this is not what has happened. End aside.

Why on earth should it matter whether complex I pumps the full complement of four protons? If there is only enough Gibbs free energy for two or three protons, why not just pump two or three protons?

What occurred to me is that for complex I to pump the four protons it might be necessary to have a full "priming" of the membrane arm with enough Gibbs free energy for a full "push to the left", as in this doodle, discussed in a previous post:

What if you only have 480mV/2e- available, giving a half hearted "nudge" to the left when you need the full shove from 840mV/2e-? Is it possible that, under these circumstances, nothing at all happens? There is no flip of the glutamate/lysine pairs from together to apart, which triggers all of the opening/closing of water channels that allows proton translocation? Zero proton translocation?

If you wanted to restore the full "kick to the left" it might be a reasonable energetic top-up to supply the extra energy from Na+ ingress up near the Q binding site to allow the menaquinone plus Na+ ingress to generate the full Gibbs free energy for activation of the membrane arm. That needs a trans-membrane channel, so we are looking at the really complicated region around NuoH and NuoA/J/K. Not easy.

The best characterised Na+ channel in the whole related set of the complex I, MRP and MBH systems is the one in the MBH of P furiosus, described by Yu et al in

Structure of an Ancient Respiratory System

which gives the Na+ channel looking like

with the red blobs being the modelled Na+ binding sites. This is made up of proteins from four separate genes and is thought to be homologous to the Na+ channel of the MRP antiporter, also a multigene structure. They show them as identical in their discussion doodles, like this for the MRP multigene Na+ channel, simplified to the subunit shown as MrpG:

The equally multigene Na+ channel of the MBH of P furiosus (in this doodle the multiple Mbh genes are simplified to MbhC) is shown in exactly the same location with the same critical broken helix, in the same shade of green:

The final doodle in this figure is complex I. Here is their image:

which keeps things nice and simple between NuoH and the Nuo A/J/K region, where Nqo10 is shown but nothing else. So I wanted to know if there could be a Na+ channel which could be used to either top up the energy of the NADH:menaquinone couple or to allow the antiporting function to occur in deactivated standard CoQ based complex I. Of course no one is looking for Na+ channels in complex I in quite the same way as Yu et al were looking for one in MBH, where they knew it was crucial. Anyway, I went looking. In here

Structure and function of mitochondrial complex I

is where I found find this image (which is unfortunately left to right transposed in its view) that includes a rather nice broken helix shown in pink which I have circled in red:

If we take this broken helix, colour it green and transplant it in to Yu et al's complex I we get this:

Might it be the Na+ channel I need?  No one knows (yet). It is known that there is a conformational change in the region of the CoQ binding pocket when complex I changes from the active to the deactivated state, close to the region of the broken helix head. I would suggest this conformational change might allow the Na+ antiporting function to occur when pumping in complex I is deactivated. Protons would be allowed to enter the cell/mitochondrion in exchange for Na+ expulsion (theoretically reversible but that seems unlikely physiologically unless you are a bacterium using menaquinone where some Na+ ingress is worth it to trigger the membrane arm). Logic says that this trade off would be preserved if antiporting provided a net benefit to the organism under which ever conditions might have favoured deactivation of complex I.


Addendum. Here are the three complexes lined up. Because PowerPoint lets you do it. No other reason.

Wednesday, April 03, 2019

Life (29) Applied billiards to MRP and MBH

Here is a nice doodle of complex I taken from this interesting but off topic paper

It shows that taking two electrons from NADH and dropping them to CoQ provides a Gibbs free energy of 840mV for the pair. That's all well and good and is why cells love oxygen based electron transport chains. But I'd like to look at this in reverse. We know that complex I works perfectly well in reverse. So we can say that dropping four protons from extracellular to intracellular down a membrane voltage of 150mV across a proton tight membrane will allow the performance of 840mV of work.

For complex I running in reverse this 840mV of work is always used for NADH generation from NAD+ (or superoxide generation) because that's the unit bolted on to the membrane arm. Other metabolic units can be used in place of NADH dehydrogenase and many of these can be found in various bacteria using various substrates with various numbers of antiporter-like subunits in the membrane arm. All will work in reverse and provide/utilise differing Gibbs free energies.

So I would like to view the membrane arm of all of these complexes as a machine in its own right which, when running in reverse, can harvest the energy available from a proton gradient for whatever function the cell might like. Obviously this is not the current arrangement, but is possible.

So here is the intact membrane arm of the complex I of T thermophilus, once again taken from Luca et al

running in reverse to give 840mV/2e- of usable energy. As in the last post we can imagine each proton passing through each channel "kicks" to the right ('cos we're running in reverse here) until the cumulation of four "kicks" is enough to cause the conformation change in Nqo 8 which drives two electrons back up the FeS chain to an NAD+. Think of the membrane arm as a machine to accumulate multiple "kicks" to give a big Gibbs free energy.

Now let's remove some of the sub units and see what we get. How about this:

Two proton translocations will provide much less Gibbs free energy than four proton translocations. I've faintly allowed a third proton channel but there is no suggestion this is a functional feature of complex I at all. It's Nqo 10, never mentioned in the paper. Note that all of the subunits are kicking in the same direction.

The above is obviously the basis of the MRP antiporter, which looks like this:

where the energy of two protons is cumulated to allow a single Na+ ion to be extruded. How much of a Gibbs free energy will be developed by the two protons will obviously be influenced by the membrane voltage through which the protons fall. It will also be influenced by the permeability to protons of the membrane in which MRP is embedded. If the membrane was 100% permeable to protons there is no point trying to harvest a gradient which isn't there. The more permeable the membrane the less energy can be harvested per proton. But we do know that in E coli under CCCP with a membrane voltage of as little as 15mV that MRP can cumulate enough Gibbs free energy to antiport Na+ when monogenetic antiporters fail completely.

In the ancestral hydrothermal vent scenario a membrane voltage from the pH gradient between vent and oceanic fluids might be, ideally, around 150-200mV. I still consider MRP is most likely an adaptation to allow colonisation of vent environments where apposition of fluids is not perfect or some mixing has already occurred and the available membrane voltage might be very low. Given MRP LUCA might survive on the edges of vents in addition to thriving in the luxurious conditions of perfect apposition near the centre.

A step on from MRP is, hypothetically, to add a power unit which will allow Na+ pumping when membrane voltage is too low even for MRP. You can think of this as an adaptation to the tail end of the vent system. Some membrane voltage is still present as are complex molecules (from dying prokaryotes in the better vent environments) out of which ferredoxin can be synthesised. This is not quite the membrane bound hydrogenase of P furiosus as we will see. It's an augmented "kick-to-the-right" through MRP to assist the generation of enough Gibbs free energy to allow Na+ extrusion:

Notice that all of the "kick" arrows run in the same direction as for all of the illustrations in this post, looking to generate a significant Gibbs free energy at the right hand end.

The final stage of generating the MBH of P furiosus (and freedom from the vent gradient completely) is like this: We have to reverse the direction of the proton travel in MBH subunit H. The "kick" from proton ingress is to the right. We want to impose a kick towards the left to reverse the proton flow to give egress, using ferredoxin power. This requires a barrier between MbhM and MbhH to stop that kick to the right, completely different to complex I and any doodle so far. As Yu at al comment in the legend to Figure 3

"A hydrophilic axis across MbhM membrane interior is also identified but it is separated from that in MbhH due to a gap between the two subunits"

and in the legend to Figure S4:

"Note the large gap between subunits M and H in MBH (A). There are four elongated densities located to the lower region of the gap (A) inset; marked by blue dashed lines), which stack against several hydrophobic regions of subunits M and H. These densities are likely from two phospholipid molecules that may stabilize the structure and prevent ion leakage across the membrane bilayer. The dashed curves in (C) and (D) highlight the fact that the chain of hydrophilic residues found in complex I is continuous (D), but is discontinuous in MBH (C)".

This is the genuine article:

and my version, more crudely:

As well as stopping MbhM "kicking" the central water channel protons of MbhH to the right, the conformation change from the NiFe hydrogenase has to be imposed on MbhH to force a proton outwards.

So the white arrow of proton "kick" in the central water channel of MbhH is reversed by the green cranks using the power from the hydrogenase (green arrow) to enforce this. The bright red proton channel is present as specified by Yu et al and is both essential and functional. Quite what the remnant of this is doing complex I I don't know and quite what the proton channel within MbhM is preserved for I also can't imagine (yet). But the general principles of proton movement as set down here are much more satisfying than my initial thoughts on MRP and MBH.


Tuesday, April 02, 2019

Life (28) Proton Billiards

Here we are looking at complex I from from Thermus thermophilus as featured in this paper:

Symmetry-related proton transfer pathways in respiratory complex I

Because it's bacterial it is all labelled up in Nqo terminology. In mammalian complex I we use Nuo terms, where NuoH is Nqo8, NuoL is Nqo 12, NuoM is Nqo 13 and NqoN is Nqo 14. We can ignore all of the other subunits.

Here are the water channels used to allow protons to move through complex I, red beads being water molecules:

The CoQ binding site (and the NADH dehydrogenase unit) are at the right hand end. What is most important is that the water channels are not all open (hydrated) at the same time. In the resting state the N-side channels are open. A conformational change in Nqo8 is induced by CoQ reduction which opens its water channel and allows a proton to enter from the cytoplasm. This triggers a chain of conformational changes horizontally along the central water channel, moving a proton from right to left within each antiporter-like subunit and which also closes the N-side water channels and opens the P-side water channels, to allow protons to move outwards in to the periplasm. This is their doodle from the discussion:

If anyone goes through the diagram in the sort of detail I did they can see that Nqo13 doesn't make sense because glutamic acid E377 is not a lysine (K abbreviation), which it is in the other two subunits. That messes up all of the charge movements and the inter-subunit electrostatic binding. From Fig 1 section B elsewhere in the paper you can see there is an arginine (R163) just "north" of E377 which might be doing this job by binding to the aspartate (D166) of Nqo12 but I can't see that this is addressed anywhere in the paper. So it's just my guess. Still. The basic concept is pretty convincing.

TLDR: The reduction of CoQ to CoQH2 clunks protons horizontally within the central hydrated channel of each antiporter-like subcomplex from their input zone to their output zone.

We have to bear several things in mind. First is that the system is completely reversible today. As in reverse electron transport using a high membrane potential and reduced CoQ couple to reduce NAD+ to NADH, and generate ROS when NAD+ is all used up... Protons will move inwards from periplasm to cytoplasm as this happens.

Also this is complex I, it is a relatively late addition to modern bacterial metabolism dependent on proton tight membranes and the availability of molecular oxygen.

Third is that our best remnant of LUCA is the Na+ pumping membrane bound hydrogenase of P furiosus and this drives from left to right through an Nqo14/NuoN related subunit (and will almost certainly be equally reversible) to its Na+ channel.

I have some Powerpoint doodles to take this a little further.


Wednesday, March 27, 2019

Life (27) Alphabet soup

So this is the predicted structure of  the modern MrpA-like functional antiporter from N thermophilus, nt-Nha:

courtesy of

A missing link between complex I and group 4 membrane-bound [NiFe] hydrogenases

As you can see it is a double channel and it is homologous to NouL from mammalian (and others) complex I. The two marked amino acids are a glutamic acid and a lysine which are conserved in NuoL. I would guess that the left hand channel is the proton channel and the right hand one is for the antiported Na+ ion. In the modern bacterium N thermophilus this antiporter is electrogenic, ie it moves more than one H+ inwards for each Na+ outwards. This may or may not have been the case in LUCA.

So nt-Nha looks a lot like NuoL and so quite like NuoM and NuoN (and MrpA, MrpD and MbhH) but it is the only one to antiport. The antiport mechanism really hasn't been worked out yet.

I guess the next thing to look at is from the same paper. This time we're looking at NuoH, the anchor point at the base of all complex I family of proton pumps. Here it is as a model with NuoH in gold superimposed over the "left hand" channel of NuoL in grey:

Half of NuoL is structurally pretty well identical to NuoH.

From the proto-Ech posts I consider NuoH to be derived from the primordial channel designed to allow oceanic pH penetration in towards the NiFe catalytic site (at alkaline hydrothermal vent pH) to generate reduced ferredoxin as the core power molecule of early LUCA, able to reduce CO2 to CO at the CODH/ACS complex.

So a derivative of the ancestral proton channel formed half of NuoL. I think it is very likely that this is the case for all of the "antiporter-like" subcomplexes, though clearly some changes in function have occurred. What made up the other channel of the ancestral antiporter? For a clue as to this one we have to change subunit and change nomenclature. The work has not been done for NuoL but it has been done for the membrane bound hydrogenase of P furiosus (subunit MbhH) which is structurally more closely related to NuoN rather than NuoL.

Let's switch papers and go back to

Structure of an Ancient Respiratory System

Here we are viewing subunit MbhH, remember that we think that the left hand channel is probably derived from the ancestor of NuoH:

Now, here's the clever bit. If we draw a horizontal line across the left hand channel, say just below the "9", we can rotate the channel around this line so all of the labels 9, 10, 11, 12 and 13 end up on the bottom far side of the model. With this done the two channels are superimposable, rather neatly:

So the right hand side of MbhH is simply the inverse of the left hand side.

We only need one protein, the original ancestral proton channel, to make both sides of the original ancestral antiporter. Minor modification would allow one side for H+ and the other for Na+. Then some sort of coupling to convert H+ in to drive Na+ out. But basically we can make all of the modern "antiporter-like" subunits from an ancestral derivative of two conjoined NuoH-like channels.

That is so neat it has to be correct. So it's probably wrong!

That coupling process might be extractable from work being done with modern complex I but I have only skim read the paper so far, so I'm not sure how much further I can progress the current fairy tale.


Tuesday, March 26, 2019

Life (26) MrpA MrpD NuoL NuoM and NuoN. Plus nt-Nha.

A few years ago I mentioned this paper

Homologous protein subunits from Escherichia coli NADH:quinone oxidoreductase can functionally replace MrpA and MrpD in Bacillus subtilis

In brief they had Bacillus subtilis strains with either an MrpA knockout or an MrpD knockout. The E coli complex I equivalent of NuoL can replace MrpA and the NuoN equivalent can replace MrpD in B subtilis. NouM doesn't seem do either. But all five subunits look very similar to each other and all are clearly related. NuoL, NuoM and NuoN are always described as "antiporter-like". MrpA and MrpD are thought to be antiporters but none ever work alone, the whole complex is needed, so they are probably "antiporter-like" too. They all appear to have been derived from an antiporter but any intrinsic antiporting has been lost.

Which makes me sad because it seems very probable that all of the above subunits are derived from the primordial antiporter at the origin of life which initiated Na+ bioenergetics and all that followed on from that.

Then came Natranaerobius thermophilus. N thermophilus is not really in the league of P furiosus, it's okay growing at up to 57 degC (which will still scald you) but no higher and it has adapted its membrane to remain proton tight at this temperature. BTW it's strictly anaerobic, is an halophile and an alkaliphile. It has (among several) one family of modern antiporters which are clearly genetically related to the MrpA of modern Clostridium tetani (and probably all other MrpAs). Modern nt-Nha is a fully functional antiporter as a stand-alone single gene protein. As these folks say:

The halophilic alkalithermophile Natranaerobius thermophilus adapts to multiple environmental extremes using a large repertoire of Na+(K+)/H+ antiporters

"Gene nt-Nha had 35% identity to the shaA (mrPA) gene of Clostridium tetani. The Mrp proteins belong to the monovalent cation/proton antiporter-3 protein family. This family is composed of multi-component Na+/H+ and K+/H+ antiporters encoded by operons of six or seven genes, and all genes are required for full function in Na+ and alkali resistance (Ito et al., 2000). Sequence analysis of the
regions surrounding gene nt-Nha, however, did not show that it was part of an operon. This indicates that gene nt-Nha does not encode a subunit of an Mrp system, but rather a mono-subunit antiporter".

EDIT number 2: These people have isolated an MrpA from an archaeal species which will antiport on its own, which makes it very similar to nt-Nha. There is also some evidence that complex I can function as a partial Na+/H+ antiporter as in this paper. NuoL is the main suspect. END EDIT.

Neither the MrpA-like nt-Nha nor the modern MrpA subunit of C tetani is in any way primordial. Both are used to extrude Na+ in exchange for H+ but this is not to drive Na+ energetics, they are much more associated with resistance to high Na+ concentrations and to alkaline pH environments. So it is possible that the N Thermophilus nt-Nha is a relatively modern derivative (it does use a proton tight membrane) of a relatively modern MrpA.

Or, more excitingly, it's possible that an ancestral Na+/H+ antiporter gave rise to both nt-Nha and MrpA. This would be the interesting option as it is possible that the Na+ binding sites, the route across the antiporter for Na+ and the mechanism for activation might just give us the technique used by the original ancestral antiporter. Genetics and structure-function modeling look to be the way to go but I can't see that it's been done yet.

Edit: Found the structure homology studies in here, lying around on my hard drive for years too. End edit.


Monday, March 25, 2019

Life (25) Left or right hand?

Here is the basic Mrp antiporter structure as suggested in

Structure of an Ancient Respiratory System

Quite how many protons are exchanged for how many Na+ is uncertain but there are papers using modern Mrp set in proton tight membranes that suggest it might be more than one H+ in per one Na+ out. ie Mrp is electrogenic, or rather it consumes pH gradient to extrude Na+. This would be no problem in the hydrothermal vent scenario, protons being freely available there. Personally I'd like electroneutrality but that's just my biases as to how P furiosus works. Anyway, Mrp is much like this, with uncertainty about the numbers of ions:

Here is exactly the same antiporter but broken down in to the main channels:

If we ignore the arrows for the H+ in the diagram all we have to do is remove the bulk of MrpA (the N-terminal) and replace it with a power source "pushing" in from the left and we have the membrane bound hydrogenase of P furiosus, still retaining the MrpG Na+ channel and working as a Na+ pump:

The paper then goes on to talk about Complex I and how that, in the Mrp nomenclature, the combined MrpD plus the fused-on C-terminal of MrpA are flipped around. I spent a long time mentally lining up various channels in my head until I twigged the simplest way to look at it was to keep Mrp channels unchanged but look at the NADH dehydrogenase of complex I as simply pumping in to a completely un-flipped Mrp but being bolted on to the opposite end, in the place of the MrpG Na+ channel. Leaving the N-terminal of MrpA still in place, like this:

I've squeezed in an extra MrpD because that's what complex I appears to have done as a modified duplication of either MrpA or MrpD. In mammalian mitochondrial nomenclature the MrpA N-terminal derivative is NuoL, the narrowed (only in this image, not really) MrpD gives NuoM and the full sized MrpD is NuoN. Yu et al only use the bacterial  complex I terminology based around the Nqo numbers. I've avoided these numbers (just used the Mrp terminology throughout the doodles) as the switches from terminology to terminology did my head in (as we say here in Norfolk) for weeks. MBH, Mrp and Nqo. Alphabet soup for the subunits!

But the core insight for me was that if you supply power from the left you pump Na+. Supply it from the right and you pump protons. This looks very much like motorising Mrp from one end makes it work in the Na+ extrusion antiport mode. Adding the power source to the opposite end, coincidentally removing the Na+ channel at the same time, drives the antiporter in the H+ expulsion direction, reversing the primordial function of Mrp and so forming the origin of the complex I family.

Obviously there is nothing primordial about complex I. It is reliant on a proton tight membrane and the ability to extract large amounts of energy from NADH, which usually means the presence of molecular oxygen. The least altered representative of antiquity is undoubtedly the MBH of P furiosus and even more so is the ancestor of the Mrp antiporter family.