74 replies to this topic

[niner]

I'm getting less excited about PUFA the more I read:
http://pmid.us/16620...976221 17702671
PUFAs are sensitive to oxidative damage and therefore species with long life spans tends to minimize their use.

Thanks for the abstracts. I'd run across this before in the context of different species but some of these papers show that you can see correlations between degree of membrane unsaturation and longevity within species as well. A question, however: Is the degree of membrane unsaturation determined by diet or by levels of endogenous desaturation and acylation/deacylation pathways? Or is it both? It seems plausible that it could at least be both, but I'll bet someone has looked at it. Has anyone done the experiment where animals were fed either a high or low PUFA diet to see which lived the longest? Someone MUST have done that as part of a CVD study somewhere.

BTW, cool use of pmid.us! I never knew about the "+" operator.

[tunt01]

Thanks for the abstracts. I'd run across this before in the context of different species but some of these papers show that you can see correlations between degree of membrane unsaturation and longevity within species as well. A question, however: Is the degree of membrane unsaturation determined by diet or by levels of endogenous desaturation and acylation/deacylation pathways? Or is it both? It seems plausible that it could at least be both, but I'll bet someone has looked at it. Has anyone done the experiment where animals were fed either a high or low PUFA diet to see which lived the longest? Someone MUST have done that as part of a CVD study somewhere.

BTW, cool use of pmid.us! I never knew about the "+" operator.

i think this will partially answer your question:

http://www.imminst.o...showtopic=40307

he discusses membrane composition in the context of dietary PUFA intake and the outcomes (heart disease).

[Skötkonung]

Dr Eades discusses membrane saturation:
http://www.proteinpo...ane-saturation/

[tunt01]

Dr Eades discusses membrane saturation:
http://www.proteinpo...ane-saturation/

thx for this link. along the lines of my thinking. low PUFA, moderate, but balanced SFA/MUFA. but the other question is then, how much total fat as a % of caloric intake (assuming constant ratios)?

last 3 days for me is 19g SFA, 26g MUFA, 8.2g PUFA (3.1 n-3, 4.8 n-6). not perfect, but maybe going in the right direction. intake of coconut oil was low last 3 days (waiting for shipment).

Fat is ~30% of total caloric intake.
9.5% is SFA
13.1% is MUFA
4.1% is PUFA

I may target something like 10% SFA, 10% MUFA, 5% (max) PUFA until this issue is more clear. think there are going to end up being other factors at work here, besides just MUFA/SFA/PUFA intake. like Vitamin E, CoQ10, etc. it will be a multi-factor situation, where just looking at XYZ diet that +/- SFAs might not be entirely revealing of the matter (note the comments in the Eades link saying that Vit. E. is needed to protect PUFA against oxidation).

Edited by prophets, 14 April 2010 - 11:32 AM.

[Michael]

I'm getting less excited about PUFA the more I read:
http://pmid.us/16620...976221 17702671
PUFAs are sensitive to oxidative damage and therefore species with long life spans tends to minimize their use.

To be clear, the issue is the degree of unsaturation (number of double bonds) per molecule of each constituent membrane phospholipid's PUFA molecules, not (as Dr. Eades and Skotkonung appear to misunderstand) the number of phospholipid fatty acids that are saturated vs PUFA, which latter is hardwired by the structures of the phospholipids themselves: most membrane phospholipids are by chemical nature composed of one saturated and one unsaturated fatty acyl tail, and a given species (in health) has a given balance of these in a given cell type.

Ie, you're "stuck" with a given number of unsaturated fatty acids in your cell membranes, but species and individuals vary in the degree of unsaturation of the the unsaturated fats, depending (to answer Niner's question) on both dietary intake (more intake of longer-chain PUFA means more longer- vs. shorter-chain PUFA in the cell membranes) and on active, species-specific metabolic regulation by desaturases, acylation/deacylation, remodeling by phospholipases, selective peroxisomal beta-oxidation directed by PPARs, etc. Longer-lived species are more rigorous in actively keeping the peroxidizabiity ratio (number of double bonds per UFA molecule) low despite dietary availability. Similarly, CR animals are more aggressive about this than AL animals (including, based on some very preliminary data from the CR Society cohort taken from plasma membrane PUFA, CR humans); this appears to be one of the mechanisms whereby CR retards the aging process.

The issue is that peroxidizability rises exponentially with each double bond beyond the first; if you look at the various studies, in fact, you'll find that the only membrane FA that is really consistently negatively associated with lifespan is DHA, with EPA and AA making an occasional appearance. The link appears to be via the mitochondrial inner membrane, which makes sense as they are very close to 'ground zero' for ROS generation in vivo (the electron transport chain) and are physically linked to the mitochondrial DNA, to which they can pass on free radical damage through propagation of lipid peroxidation.

While badly out of date, I have discussed this extensively here; for a semi-update , see references (1-15) here. These also partially address Niner's important diet-and-lifespan question. This is why I specifically recommend that people on CR favor ALA (eg flax) over long-chain omega-3s (EPA/DHA/fatty fish), which has cardiovascular outcomes at least as favorable and doesn't potentially undermine CR's effects on mitochondrial inner membrane composition.

[Skötkonung]

Michael,
I take it you have seen the findings (here, here, here) that strongly link the level of saturation of cell-membrane omega-3 phospholipids (ie short-chain ALA) to maximum lifespan in various vertebrates?

First, (and I’m not going to quote from specific papers right now because I don’t have them in front of me–this response is based on memory, which is always dangerous) it’s apparent that the unsaturation index can be lowered by consuming more saturated fats (by that I don’t necessarily mean saturated fats, but fats that are less unsaturated). A lowered unsaturation index means that the fat acids (primarily in the cell and mitochondrial membranes) are less prone to peroxidation. Less prone to peroxidation means longer lived. Evidence shows that dietary restriction reduces the unsaturation index, making the membranes less prone to peroxidation, meaning that (probably) a reduced unsaturation index does indeed lead to greater longevity.

A couple of papers I’ve come across seem to indicate that insulin increases the activity of desaturation enzymes, meaning that elevated insulin levels would lead to more desaturation, i.e. fats that are more unsaturated, and, consequently, a higher unsaturation index. If a lower unsaturation index leads to a longer life, one would assume that a higher unsaturation index would imply membranes more susceptible to peroxidation and a shorter life.

ALA (again, according to a lot of reading I’ve done) is burned preferentially as a fuel source instead of being incorporated into membranes or shunted down the elongase-desaturase pathway. ALA is also converted to ketone bodies easily because of its propensity for being rapidly oxidized.

The sum total of these sort of random thoughts is that (to me, at least) the amounts of EPA/DHA consumed as fish oil are minuscule in terms of overall fat consumption and are (in my opinion) not major players in establishing the unsaturation index. Consuming more saturated fats while keeping insulin levels low will maximally decrease the unsaturation index (but not to the point where membrane rigidity becomes a problem), providing membranes much less prone to free radical attack, and, consequently, increase longevity. And since ALA doesn’t really enter the pipeline much to speak of, I’ve never considered it as a replacement for fish oil.

[Michael]

I take it you have seen the findings (here, here, here) that strongly link the level of saturation of cell-membrane omega-3 phospholipids (ie short-chain ALA) to maximum lifespan in various vertebrates?

You're somehow misreading these abstracts, Skot: none of those studies find that, except "negatively." The Hurlburt review (see the full text here) rightly links the association to DHA, and his theory (contrary to most researchers' view that it's due to peroxidizability per se) is that "both the amount of membranes and their acyl composition, notably the relative balance between monounsaturated and polyunsaturated acyl chains, especially docosahexaenoic acid (DHA), are a pacemaker for metabolic activity (Hulbert and Else, 1999, 2000). Space limitations restrict our discussion here to the compositional aspects of membranes." The bird-vs-rat study finds that "the fatty acid double bond content of total lipids and phosphatidylcholine, phosphatidylethanolamine and cardiolipin fractions of heart mitochondria is intrinsically lower in pigeons (MLSP = 35 years) than in rats (MLSP = 4 years). This is mainly due to a lower content of the most highly unsaturated docosahexaenoic acid (22:6n-3) and in some fractions arachidonic acid (20:4n-6)." And the study of liver mitochondria from eight mammals abstract is quite explicit on this: " the total number of double bonds and the peroxidizability index are negatively correlated with maximum life span (r = -0. 88, P < 0.003; r = -0.87, P < 0.004, respectively). This is not due to a low content of unsaturated fatty acids in longevous animals, but mainly to a redistribution between kinds of the polyunsaturated n-3 fatty acids series, shifting from the highly unsaturated docosahexaenoic acid (r = -0.89, P < 0.003) to the less unsaturated linolenic acid (r = 0.97, P < 0.0001)."

The sum total of these sort of random thoughts is that (to me, at least) the amounts of EPA/DHA consumed as fish oil are minuscule in terms of overall fat consumption and are (in my opinion) not major players in establishing the unsaturation index.

It's less important than species, but if you look at the studies I cite in my posts, you'll see that dietary intake does shift cell and (more importantly) mitochondrial membrane PUFA content.

And since ALA doesn’t really enter the pipeline much to speak of, I’ve never considered it as a replacement for fish oil.

Not sure what you're getting at, here.

[Jay]

Michael,
As I understand it, balance of EPA, DHA, and arachidonic acid is important to maintain proper eicosanoid signaling. I think Skot is wondering whether, on account of ALA's limited conversion to EPA and DHA, eicosanoid concerns might trump membrane concerns?

Also, if we didn't care about eicosanoid biology (e.g., balanced LC n-6 and n-3), would linoleic acid be just a good as ALA for your membrane peroxidation concerns? And, if so, wouldn't the average western dieter, consuming lots of LA and little DHA, have minimal membrane peroxdiation? Are you aware of any such evidence?
Thanks, Jay

Edited by Jay, 20 April 2010 - 10:06 PM.

[Skötkonung]

Michael,
As I understand it, balance of EPA, DHA, and arachidonic acid is important to maintain proper eicosanoid signaling. I think Skot is wondering whether, on account of ALA's limited conversion to EPA and DHA, eicosanoid concerns might trump membrane concerns? Also, if we didn't care about eicosanoid biology (e.g., balanced LC n-6 and n-3), would linoleic acid be just a good as ALA for your membrane peroxidation concerns? And, if so, wouldn't the average western dieter, consuming lots of LA and little DHA, have little membrane peroxdiation? Are you aware of any such evidence?
Thanks, Jay

Yes, I was wondering about the relevance of his ALA suggestion given how easily it is converted to ketones. It seems more pertinent to suggest consuming longer chain fats (SFA, in particular) to lower the unsaturation index, given the research I have seen.

The papers were ones I had read supporting Michael's hypothesis. I'll try locating the papers from which I drew my conclusions in the above post when I get home.

[HaloTeK]

I agree with Michael on limiting poly sources, but want to know his full reasoning on favoring monos vs saturated fats when there have been some studies linking monos to heart disease.

I'm just playing devil's advocate here because I always feel better when I consume olive oil vs things like butter.

Edited by HaloTeK, 20 April 2010 - 11:19 PM.

[Sillewater]

Thanks MR for that post. I was reading through some studies and your old posts and was wondering what the main factors were. Originally I thought that if you ate more PUFA's you would have on an absolute basis more PUFA's in you membranes. Instead its just that each phospholipid will have a unsaturated fatty acid with more double bonds.

According to this paper:
Am J Physiol. 1999 Jan;276(1 Pt 2):H149-58.
PUFA and aging modulate cardiac mitochondrial membrane lipid composition and Ca2+ activation of PDH.
Pepe S, Tsuchiya N, Lakatta EG, Hansford RG.
PMID: 9887028 [PubMed - indexed for MEDLINE

A diet higher in omega-6 in rats caused a shift in lipid composition. Specifically resulting in more phosphatidylcholine. Which lipids cause the lowest levels of peroxidizability?

This one is definitely worth a look. Compares different muscles whose cardiolipin contains different amounts of SAFA and PUFA:
J Membr Biol. 2010 Apr;234(3):207-15. Epub 2010 Mar 25.
Skeletal muscle type comparison of subsarcolemmal mitochondrial membrane phospholipid fatty acid composition in rat.
Stefanyk LE, Coverdale N, Roy BD, Peters SJ, LeBlanc PJ.

This article mentions the decrease in fluidity of mitochondrial membranes in aged rates. (Don't PUFA's keep them more fluid, I'm reminded of Art De Vany's reason for consuming Fish Oil).
Eur J Biochem. 1978 Oct;90(2):385-90.
Influence of mitochondrial radical formation on energy-linked respiration.
Nohl H, Breuninger V, Hegner D.
PMID: 710436 [PubMed - indexed for MEDLINE]

Kinda scary, the metabolites can diffuse away:
Chem Phys Lipids. 2009 Jan;157(1):1-11. Epub 2008 Oct 14.
Lipid peroxidation of membrane phospholipids generates hydroxy-alkenals and oxidized phospholipids active in physiological and/or pathological conditions.
Catalá A.

Just a good read on mitochondrial dysfunction:
Crit Care. 2008;12(1):206. Epub 2008 Feb 18.
Bench-to-bedside review: Mitochondrial injury, oxidative stress and apoptosis--there is nothing more practical than a good theory.
Bayir H, Kagan VE.

Ageing Res Rev. 2010 Apr;9(2):200-10. Epub 2009 Oct 1.
Lipid peroxidation in relation to ageing and the role of endogenous aldehydes in diabetes and other age-related diseases.
Dmitriev LF, Titov VN.

J Neurochem. 2010 Apr 1;113(2):465-76. Epub 2010 Jan 22.
Membrane lipid modification by polyunsaturated fatty acids sensitizes oligodendroglial OLN-93 cells against oxidative stress and promotes up-regulation of heme oxygenase-1 (HSP32).
Brand A, Bauer NG, Hallott A, Goldbaum O, Ghebremeskel K, Reifen R, Richter-Landsberg C.

[Jay]

It is interested, though, that DHA seems to protect LDL from oxidation... (cite)

Edited by Jay, 21 April 2010 - 03:17 AM.

[kismet]

 

Yes, I was wondering about the relevance of his ALA suggestion given how easily it is converted to ketones. It seems more pertinent to suggest consuming longer chain fats (SFA, in particular) to lower the unsaturation index, given the research I have seen.

What evidence in particular? I am looking for papers linking diet and membrane composition in humans (esp. mtMembranes or those most representative for max. life span in other species). The data from rats I have seen in Hulbert's reviews suggests that the slope between diet and a. membrane unsaturation index and b. the likely more important peroxidation index is very, very close to zero.

(Unsaturation index may be quite misleading anyway, although I am not sure as of now - but Hulbert et al. explain the difference: "For example, a membrane bilayer consisting solely of MUFA will have an unsaturation index of 100 and a peroxidation index of 2.5, while a membrane bilayer consisting of 95% SFA and 5% DHA will have an unsaturation index of 30 and a peroxidation index of 40. This means that although the 5% DHA-containing membrane has only 30% the density of double bonds of the monounsaturated bilayer, it is 16 times more susceptible to peroxidative damage.")

In fact the review MR links to contains one of these graphs.

One human study found that if anything a diet higher in total PUFA & total fat decreased or maintained phospholipid membrane peroxidation index due to increased PUFA but decreased HUFA content. But I haven't figured out which membranes are representative in this context nor how to search medline for useful studies on this topic.

I agree with Michael on limiting poly sources, but want to know his full reasoning on favoring monos vs saturated fats when there have been some studies linking monos to heart disease.

I would re-think that statement. MUFA has been linked to CHD, but SaFa hasn't? The evidence for the latter seems considerably stronger, which should be a sufficient answer.

EDIT:
IIIRC several (or even most of) the pertinent links in your cr-society post are dead.

Edited by kismet, 21 April 2010 - 04:35 PM.

[kismet]

Does anyone haven an idea whether I am wrong or why I'm wrong? To reiterate what I said in the above post: I haven't seen but would appreciate evidence showing that (regular) PUFA meaningfully influences membrane unsaturation index or more importantly membrane peroxidation index.

Edited by kismet, 07 May 2010 - 02:52 PM.

[Jay]

Does anyone haven an idea whether I am wrong or why I'm wrong? To reiterate what I said in the above post: I haven't seen but would appreciate evidence showing that (regular) PUFA meaningfully influences membrane unsaturation index or more importantly membrane peroxidation index.

Don't know if you are right or wrong, but it's entirely plausible that LA does decrease the membrane peroxidation index. Longer chain PUFAs oxidize much more readily than short chain PUFAs. Membranes are composed, in part, of PUFAs so eating more short chain PUFAs might yield membranes with more shorter chain PUFA. MR stated (but I don't think he referenced any study) that ALA increases the membrane unsaturation index vis a vis EPA/DHA. I assumed this was why.

One would expect trans fats to do the same thing, right? Trans fats, including both industrial and natural trans fats, inhibit elongation of fatty acids by D6D... Does that make LA, ALA, and i-trans fats good? Maybe good for something but probably not good, IMO.

Lastly, I haven't looked into this whole fatty acid deficiency thing. If one doesn't consume enough LA (say only 1 or 2% of calories) and consumes only 1 gram or so of LC omega 3, will such person's membranes be composed, in part, of something else? Mead acid?

Edited by Jay, 07 May 2010 - 03:16 PM.

[Michael]

All:

To reiterate what I said in the above post: I haven't seen but would appreciate evidence showing that (regular) PUFA meaningfully influences membrane unsaturation index or more importantly membrane peroxidation index.

... and more importantly mitochondrial membrane peroxidizability index. As I said in a related context to Kismet, my apologies for leaving him (and the rest of you) you waiting ... I have a serious backlog of folks to whom I should be getting back. Part of the reason I am always falling behind is because I have an all-or-nothing perfectionist streak, which leads to a little bit of 'all' and far too much 'nothing' ... To get out SOMETHING, I'm going to largely just do a big link list, starting above all with asking folks to review the line of argumentation and the LIVE links in this post and supporting studies (1-15) here. The following studies then reconstruct those that haven't apparently been permanently lost (one of which, unfortunately, was the most important of all) and a couple of new ones. And, I also apologize because at least sme are also redundant to embedded links above. So please see:

MiRFAA: MiFR: Flax, not Fish? - sci.life-extension | Google Groups
MiFR: Flax, not Fish
(Yes, those are 2 different posts)

crsociety : Message: [CR] DHA Causes mt Peroxidation (and related stuff on fats)
"RE: Michael Rae's mitochondrian theory" post. Note the link takes you to a mishmash of 2 entirely different posts; to see the relevant material, SKIP DOWN TO the post by Dean P beginning with "Greg Watson wrote:".
Effects of Dietary Fatty Acids on Mitochondrial and Tissue Membrane Fatty Acid Composition & Peroxidizability Index.

Happily, albeit entailing additional work for me, 2 extremely useful studies, which I hadn't seen since last revisiting the issue, have provided some of the most comprehensive results yet. First is a study on the "Effects of alpha-linolenic acid vs. docosahexaenoic acid supply on the distribution of fatty acids among the rat cardiac subcellular membranes after a short- or long-term dietary exposure. (1) Table 3 shows total "Cardiac phospholipids fatty acid profiles (% of total fatty acids) from rats fed experimental diets for 2 or 6 months", and more importantly Table 6 shows Mitochondrial fatty acid profiles. Rearranging that table to fit here:

Cardiac mitochondrial membrane phospholipid DHA (22:6 n-3) after 2 months
Control diet 4.4 ± 0.28c
DHA diet 24.5 ± 0.97a
ALA diet 9.4 ± 0.25b

Cardiac mitochondrial membrane phospholipid DHA (22:6 n-3) after 6 months
Control diet 3.4 ± 0.10c ***
DHA diet 20.4 ± 2.73a ***
ALA diet 8.2 ± 0.35b

(If it looks as if the control diet is better than the ALA diet, well, it is: the control diet provides only the minimum essential fatty acids (LA and ALA) for the rodent RDA, whereas the ALA and DHA diets are necessarily enriched. We want as little n3 and n6 as lowers disease risk, and can make up the rest of our dietary fat (and it's pretty clear that one should be getting ≥30% fat in the diet) as MUFA, which is even less desaturated (1 double bond) than ALA (3 double bonds) or LA (2)).

The results for total cardiac membrane phospholipids are parallel, albeit less important.

(2) covers the effects of a wider range of diets ("olive oil, sunflower oil, fish oil, soybean oil, palmitic acid, or 82% palmitic acid plus 18% soybean oil, supplying the essential fatty acid") on the tissue membranes of several different tissues ("thymocytes, pancreatic exocrine, muscle and adipose tissues "), tho' we're mostly concerned with mt membranes and with postmitotic tissues like muscle; the results are harder to post, but give this a whirl:





Hey! That works pretty well, so long as your vision is good ...

Again, I expect that the relatively small benefits of this will acrue to those who are getting up against the limits of human life expectancy, not the average Jo(e) on the street or even a very healthy non-CR person; it's likely important to CR so that you aren't overriding a known effect and likely mechanism if indeed CR works in humans, and it might also be important to people who either get lucky statistically, or have 'longevity genes.'

It is interested, though, that DHA seems to protect LDL from oxidation... (cite)

First, that's in vitro oxidizability of LDL, which is totally irrelevant to the way that LDL is oxidized in vivo: what you want to look at is the level of endogenously-oxidized LDL derived in situ from blood samples. And second, even if you had that, it would be LDL oxidation -- our issue is tissue (and, again, especially mitochondrial) membrane peroxidizability.

One would expect trans fats to do the same thing, right? Trans fats, including both industrial and natural trans fats, inhibit elongation of fatty acids by D6D... Does that make LA, ALA, and i-trans fats good? Maybe good for something but probably not good, IMO.

First, you can't take advantage of a slightly lower rate of biological aging if you're already dead of heart disease, so this would be a bad idea anyway . But the main thing is direct displacement, not inhibitioin of desaturation (not, NB, elongation) by D6D -- although mutual competition for each step in the desaturation and elongation pathways separating the EFA per se from the long-chain derivative of the same series is an additional benefit to taking in the short-chain, essential PUFAs (ALA and LA) instead of the long-chain products (EPA, DHA, AA), which overrride metabolism and 'dump' the long-chain products directly into the membranes.

References
1. Brochot A, Guinot M, Auchere D, Macaire JP, Weill P, Grynberg A, Rousseau-Ralliard D.
Effects of alpha-linolenic acid vs. docosahexaenoic acid supply on the distribution of fatty acids among the rat cardiac subcellular membranes after a short- or long-term dietary exposure.
Nutr Metab (Lond). 2009 Mar 25;6(1):14. PubMed PMID: 19320987; PubMed Central PMCID: PMC2670308.

2. Soriguer FJ, Tinahones FJ, Monzón A, Pareja A, Rojo-Martínez G, Moreno F, Esteva I, Gómez-Zumaquero JM.
Varying incorporation of fatty acids into phospholipids from muscle, adipose and pancreatic exocrine tissues and thymocytes in adult rats fed with diets rich in different fatty acids.
Eur J Epidemiol. 2000 Jun;16(6):585-94. PubMed PMID: 11049103.

Edited by Michael, 28 May 2010 - 05:39 PM.

[Sillewater]

That was a great and very convincing read. Thanks MR.

[Blue]

Michael, it would be easier if the arguments were in one place instead of spread over a myriad forums.

Anyway, lets assume your argument that on theoretical grounds one should minimze membrane DHA. Obviously we cannot eliminate DHA since it is essential. But is that not an argument for taking a low amount of DHA instead of ALA? The conversion of ALA to DHA is uncertain and and varied. So in order to get the essential minimum of DHA one must take an amount of ALA that will most likely produce more DHA than the essential minimum just to be sure that if the conversion do happens to be poor one still get the minimum.

The two rodent study you cite above only shows that taking a high amount of DHA produces a high amount of membrane DHA. Does not prove that taking a low amount of DHA is worse than taking ALA.

Of course, there are numerous other theoretical possibilites. Like that dietary ALA must take part of numerous conversions that dietary DHA avoid. Do these conversions involving polyunsaturated fats where only a small fraction end up as DHA produce bad effects like free radical damage or other toxic (or good) products that dietary DHA avoid?

Looking instead at epidemiology or clinical studies, I do not think that the evidence clearly favors ALA. If the low but still essential DHA theory is right the optimum intake would likely be quite narrow for both ALA and DHA which may confuse the studies.
http://lpi.oregonsta...rnuts/omega3fa/

Edited by Blue, 02 June 2010 - 06:06 AM.

[JLL]

Ie, you're "stuck" with a given number of unsaturated fatty acids in your cell membranes, but species and individuals vary in the degree of unsaturation of the the unsaturated fats, depending (to answer Niner's question) on both dietary intake (more intake of longer-chain PUFA means more longer- vs. shorter-chain PUFA in the cell membranes) and on active, species-specific metabolic regulation by desaturases, acylation/deacylation, remodeling by phospholipases, selective peroxisomal beta-oxidation directed by PPARs, etc. Longer-lived species are more rigorous in actively keeping the peroxidizabiity ratio (number of double bonds per UFA molecule) low despite dietary availability.

Does this apply to fat cells also? It looks like in this rabbit study (which I've written about here) even the percentages of SAs, MUFAs and PUFAs change with different diets. If the number of unsaturated fatty acids is fixed, wouldn't you expect the SA percentage to stay the same and only the MUFA/PUFA ratio to change?

[kismet]

Looking instead at epidemiology or clinical studies, I do not think that the evidence clearly favors ALA. If the low but still essential DHA theory is right the optimum intake would likely be quite narrow for both ALA and DHA which may confuse the studies.
http://lpi.oregonsta...rnuts/omega3fa/

BTW you are mixing epidemiology on disease outcomes with longevity aspects in your last sentence. MR's *hypothesis* bears (basically) no relevance to the epidemiology assessing CVD and all-cause mortality in at risk populations. Small benefits on the rate of aging would never be reflected in short term studies, or at least I suppose they'd all be underpowered to detect any such effect over one or two decades. And, yeah, a few authors indeed suggest that ALA is "clearly inferior", FWIW.

JLL, did they actually test membranes? I am not sure *storage fat* has any correlation with membrane FAs. The mode of storage is completely different AFAIK, to quote wik!:
"Adipocytes, also known as lipocytes and fat cells, are the cells that primarily compose adipose tissue...White fat cells or monovacuolar cells contain a large lipid droplet surrounded by a layer of cytoplasm...fat stored is in a semi-liquid state, and is composed primarily of triglycerides and cholesteryl ester."

Edited by kismet, 02 June 2010 - 12:33 PM.

[JLL]

Looking instead at epidemiology or clinical studies, I do not think that the evidence clearly favors ALA. If the low but still essential DHA theory is right the optimum intake would likely be quite narrow for both ALA and DHA which may confuse the studies.
http://lpi.oregonsta...rnuts/omega3fa/

BTW you are mixing epidemiology on disease outcomes with longevity aspects in your last sentence. MR's *hypothesis* bears (basically) no relevance to the epidemiology assessing CVD and all-cause mortality in at risk populations. Small benefits on the rate of aging would never be reflected in short term studies, or at least I suppose they'd all be underpowered to detect any such effect over one or two decades. And, yeah, a few authors indeed suggest that ALA is "clearly inferior", FWIW.

JLL, did they actually test membranes? I am not sure *storage fat* has any correlation with membrane FAs. The mode of storage is completely different AFAIK, to quote wik!:
"Adipocytes, also known as lipocytes and fat cells, are the cells that primarily compose adipose tissue...White fat cells or monovacuolar cells contain a large lipid droplet surrounded by a layer of cytoplasm...fat stored is in a semi-liquid state, and is composed primarily of triglycerides and cholesteryl ester."

You're right, I don't think they tested membrane saturation. Though, looking at the charts Michael posted, the two do seem to correlate.

[Michael]

Here is a totally bizarre update to the literature on this (hey, kids: do not try this at home!):

Biochemistry (Mosc). 2010 Dec;75(12):1491-7.
Dietary supplementation of old rats with hydrogenated peanut oil restores activities of mitochondrial respiratory complexes in skeletal muscles.

Bronnikov GE, Kulagina TP, Aripovsky AV.


The effect of dietary supplementation of old rats (26-33 months) with hydrogenated peanut oil on the activity of mitochondrial enzymes in skeletal muscles has been studied. ... The activities of respiratory complexes I and IV were shown to significantly decrease with the age compared to the activity of the same enzymes in young animals, while the activity of citrate synthase was virtually unchanged. The fatty acid composition of muscle homogenates of old rats differed from that of young animals by a reduced content of myristic, oleic, linoleic, and α-linolenic acids and enhanced content of dihomo-γ-linolenic, arachidonic, and docosahexaenoic acids. [So far, in line with multiple previous studies , tho' I don't remember anyone mentioning an effect on myristic acid-MR]

Per oral supplementation of the old rats with hydrogenated peanut oil completely restored the activity of complex IV and increased the activity of complex I to 80% of the value observed in muscles of young animals, reducing the content of stearic, dihomo-γ-linolenic, arachidonic, eicosapentaenoic, docosapentaenoic, and docosahexaenoic acids relative to that in the groups of old and young rats. The content of oleic and linoleic acids increased relatively to that in the group of the old rats, as well as young animals. The possible mechanisms of the restoration of the activity of the respiratory enzymes under the administration of hydrogenated peanut oil are discussed. PMID: 21314620

I find all of this very puzzling; and tho' I've not yet had access to the full text, I suspect I'm going to continue to find it puzzling once I get a copy. IAC, I don't recommend this particular approach to the problem ...

[brunotto]

The most clear-cut findings of this study are that postmenopausal women who developed breast cancer had erythrocyte membranes characterized by higher levels of MUFAs (especially oleic acid) and a lower SI than those who did not develop breast cancer. Case women were also characterized by lower levels of the n-6 PUFA 20 : 2,n-6c and n-3 docosahexaenoic acid than women who remained disease free.

Linoleic acid (18 : 2,n-6c).

We found a small inverse association between erythrocyte membrane linoleic acid levels and the risk of breast cancer. After absorption, the essential fatty acid linoleic acid undergoes desaturation/saturation and chain elongation/shortening reactions, producing other n-6 fatty acids. A correlation study between diet and erythrocyte membrane content (Fuhrman B: unpublished data) suggest that our finding of reduced breast cancer risk in women with high levels of linoleic acid could be partially due to protection by a diet high in this fatty acid residue and partially due to its possibly low transformation rate into other n-6 PUFAs.
Other n-6 PUFAs.

While high proportions of the fatty acids on the pathway from linoleic to arachidonic acid were not associated with breast cancer, the 20 : 2,n-6c fatty acid was statistically significantly protective and was the only long-chain residue to be so. 20 : 2,n-6c is derived from linoleic acid by a different elongation pathway from that leading to arachidonic acid. Hence, the possibility arises that diversion of linoleic acid metabolism away from arachidonic acid (and hence from arachidonic acid-derived prostaglandins) might be protective against breast cancer (27); this possibility might be worth investigating further. It has been found (9) that, even when long-chain n-6 PUFAs are a minor constituent of dietary triglycerides, erythrocyte phospholipids are constitutionally rich in them; therefore, they cannot be considered to be a reliable marker of dietary intake.
α-Linolenic acid (18 : 3,n-3c).

α-Linolenic acid, the other essential PUFA in humans, was not associated with breast cancer risk. The α-linolenic acid content of erythrocyte membranes is very low and is largely unrelated to dietary intake [(9); Fuhrman B: unpublished data].
Other n-3 PUFAs.

Docosahexaenoic acid, the most abundant n-3 fatty acid in erythrocyte membranes, was inversely associated with breast cancer risk. Docosahexaenoic acid was the only fatty acid associated with fish consumption (r = .48; P<.000) in the ORDET cohort (Fuhrman B: unpublished data) in spite of the low intake of fish (average intake, 15 g daily).
MUFAs and SFAs.

We found membrane MUFAs to be positively associated and the SI to be inversely associated with breast cancer risk, whereas there was no association with SFA levels. The idea that low-erythrocyte-membrane SI is associated with increased risk of breast and other cancers has been tested in several case–control studies (10–13,28) with conflicting results, although all of the studies were small and had cross-sectional designs. Our finding of an inverse association between red blood cell membrane SI and postmenopausal breast cancer risk is consistent with the results of a Swedish cohort study (14) in which high plasma SI was found to be inversely associated with breast cancer risk. In that study, however, the apparent protective effect of high membrane SI was due to high levels of stearic acid (SI numerator), whereas the apparent protective effect was due to low levels of oleic acid (SI denominator) in the present study. These findings suggest that, rather than the individual components, the ratio between the two fatty acids may be more directly related to breast cancer risk.
A diet high in MUFAs is probably one determinant—but not the major one—of erythrocyte membrane MUFAs (8,29,30), which are extensively synthesized in the body (7,31). Therefore, the association of high levels of membrane oleic acid and low SI with increased breast cancer risk might be related to factors other than diet. Most oleic acid in mammalian tissue is derived from the saturated stearic acid residue (7,32). This key conversion is controlled by the enzyme Δ9-d (Fig. 118), hepatoma cells (33), and human leukemia and lymphoma cells (34). In all of these cases, the required MUFA levels are assured by overexpression of the genes encoding Δ9-d. Inhibition of Δ9-d and, consequently, of oleic acid biosynthesis blocks the growth of transplanted mammary tumor in rats (19). Furthermore, in vitro inhibition of mammary carcinogenesis has been achieved by the addition of an inhibitor of Δ9-d (20).
The fat content of the diet has an important effect on Δ9-d activity. High levels of SFAs increase Δ9-d activity by twofold to threefold, whereas PUFAs decrease it (35–40); as a consequence, high levels of oleic acid and low SI in tissues may be the result of a diet poor in PUFAs.


http://jnci.oxfordjo...93/14/1088.full

Edited by brunotto, 19 February 2012 - 09:32 PM.

[brunotto]

A high-cholesterol diet also seems to increase Δ9-d activity, resulting in increased monounsaturation of the membrane fatty acids in rat liver to partially compensate for the “rigidizing” effect of cholesterol incorporation in the membrane (31).
Δ9-d is activated by carbohydrate administration (37), and it is well established that insulin enhances Δ9-d activity (39–44). The possible relationships among dietary carbohydrate, insulin resistance, and breast cancer risk is receiving increasing attention (45–51).
Δ9-d activity might be also enhanced by estrogens (52,53) and testosterone (31). Estradiol has been shown to markedly lower the SI of erythrocyte membranes in premenopausal and postmenopausal women (54), suggesting a role of this hormone in the regulation of membrane fluidity. There is strong evidence that high levels of these hormones increase the risk of breast cancer (21).

Edited by brunotto, 19 February 2012 - 09:38 PM.

[kismet]

And since ALA doesn't really enter the pipeline much to speak of, I've never considered it as a replacement for fish oil.

Not sure what you're getting at, here.

I think Skot meant something like this (1):

Several investigators have performed detailed studies of the in vivo conversion of ALA to its long-chain n−3 derivatives EPA, docosapentaenoic acid (DPA), and DHA in humans by using uniformly labeled [13C]- or [2H]ALA as a tracer (reviewed in references 35 and 36). These studies have consistently shown that ≥15–35% of dietary ALA is rapidly catabolized to carbon dioxide for energy (37-40), and that only a small proportion, estimated by using compartment models to be <1%, is converted to DHA (41, 42). In fact, ALA has the highest rate of oxidation among all unsaturated fatty acids (43). The fractional conversion of ALA to EPA, estimated by measuring peak or area under the curve plasma contents of the labeled fatty acids, varies between 0.3% and 8% in men, and the conversion of ALA to DHA is <4% and often undetectable in males (39, 40, 44, 45). Conversion of ALA to long-chain n−3 fatty acids appears to be more efficient in women: up to 21% is converted to EPA and up to 9% is converted to DHA (38), with a concomitant reduction in the rate of ALA oxidation (≈22% compared with ≈33% in men) (38-40). The conversion of DPA to DHA is the rate-limiting step in the conversion of ALA to DHA, and dietary DHA and EPA down-regulate this step by 70% (41). Others have also shown that dietary EPA and DHA reduce the conversion of ALA to long-chain n−3 fatty acids (37, 39). Diets high in LA may influence, through substrate competition and inhibition of Δ6 desaturase enzyme, the metabolism of the n−3 fatty acids. Emken et al (44) showed that a diet high in LA reduces the conversion of ALA to its long-chain derivatives by 40%, with a net reduction in long-chain n−3 fatty acid accumulation of 70%. Diets high in ALA appear to increase the rate of ALA oxidation, limiting its accumulation in plasma and reducing its conversion rate to EPA and DHA (37). Considerable variability in the conversion rates among individuals has been reported, even when the subjects have similar background diets (44).This interindividual variability, along with modest ALA intakes[!] and high amounts of LA[!] in the American diet, suggests that ALA cannot reliably replace EPA and particularly DHA in the diet.

But the authors go on to point out why this is a minor issue in the healthy:

Several case studies involving n−3-deficient patients reported that intervention with ALA results in marked increases in plasma concentrations of both EPA and DHA (50, 51). In addition, vegans who consume ALA but not EPA and DHA in their diets have low but stable concentrations of DHA in plasma (52, 53). .

and veg*ans nonetheless suffer from fewer chronic diseases than omnivores - arguing against clinically relevant deficiency, and ALA is as (in)effective as fish oil for the prevention of heart disease - arguing against clinically relevant deficiency. The evidence ain't perfect, but taken together it all fits.

Clearly, this excludes deficiency, but not modest insufficiency. Given the down-sides of fish oil, I would still recommend isolated ALA from flax seed oil or ALA+low dose fish oil over moderate-to-high-dose fish oil (i.e. >300mg/d total EPA+DHA). This rule of thumb applies to both ad libitum(normal) and CR diets.

Edited by kismet, 22 July 2013 - 02:19 PM.

[misterE]

Insulin increases saturated-fat conversion into monounsaturated-fat (via delta-9-desaturase up-regulation). Insulin-resistance impairs this conversion.

[Logic]

Docosahexaenoic acid suppresses the expression of FoxO and its target genes.
http://www.ncbi.nlm....pubmed/22444500

"...The scientists were able to show that animals without FoxO possess significantly fewer stem cells. Interestingly, the immune system in animals with inactive FoxO also changes drastically. „Drastic changes of the immune system similar to those observed in Hydra are also known from elderly humans“..."
http://www.uni-kiel....foxogen-e.shtml

FoxO is a critical regulator of stem cell maintenance in immortal Hydra
http://www.pnas.org/...nt/109/48/19697

[niner]

Wow, very interesting. Thanks for that find, Logic. Here is the first link:

J Nutr Biochem. 2012 Dec;23(12):1609-16. doi: 10.1016/j.jnutbio.2011.11.003. Epub 2012 Mar 22.
Docosahexaenoic acid suppresses the expression of FoxO and its target genes.
Chen YJ, Chen CC, Li TK, Wang PH, Liu LR, Chang FY, Wang YC, Yu YH, Lin SP, Mersmann HJ, Ding ST.

Center for Biotechnology / Department of Animal Science and Technology, National Taiwan University, Taipei 106, Taiwan.

Docosahexaenoic acid (DHA), an n-3 polyunsaturated fatty acid, has previously been shown to ameliorate obesity-associated metabolic syndrome. To decipher the mechanism responsible for the beneficial effects of DHA on energy/glucose homeostasis and the metabolic syndrome, 30 weaned cross-bred pigs were randomly assigned to three groups and fed ad libitum with a standard diet supplemented with 2% of beef tallow, soybean oil or DHA oil for 30 days, and the gene expression profile of various tissues was evaluated by quantitative real-time polymerase chain reaction. The DHA-supplemented diets reduced the expression of forkhead box O transcription factor (FoxO) 1 and FoxO3 in the liver and adipose tissue. DHA treatments also decreased the expression of FoxO1 and FoxO3 in human hepatoma cells, SK-HEP-1 and human and porcine primary adipocytes. In addition, DHA also down-regulated FoxO target genes, such as microsomal triacylglycerol transfer protein (MTP), glucose-6-phosphatase, apolipoprotein C-III (apoC-III) and insulin-like growth factor binding-protein 1 in the liver, as well as reduced total plasma levels of cholesterol and triacylglycerol in the pig. Transcriptional suppression of FoxO1, FoxO3, apoC-III and MTP by DHA was further confirmed by reporter assays with each promoter construct. Taken together, our study indicates that DHA modulates lipid and glucose homeostasis in part by down-regulating FoxO function. The down-regulation of genes associated with triacylglycerol metabolism and very low density lipoprotein assembly is likely to contribute to the beneficial effects of DHA on the metabolic syndrome.

PMID: 22444500

My first thought when I read this was that 2% of calories was a lot of DHA, and it is, but not as much as you might think. For a 2500 Cal diet, it's 5.5 grams. There are people here who are megadosing fish oil who may be getting that much. It helps if you have metabolic syndrome, though you can certainly see those lipid-lowering effects with much lower doses. If FoxO downregulation is the mechanism behind the lipid lowering, then much lower doses of DHA are causing FoxO downregulation in humans. The big question is what's the effect of this FoxO downregulation on stem cells? I've never been a fan of fish oil megadosing, and this isn't helping.

[Michael]

I am puzzled as to why Mister E and Logic think their posts are relevant to this discussion.

In any case, there are now two additional points in favor of this hypothesis. First, the regulated reduction in membrane long-chain PUFA (and especially DHA) in CR animals fed the same amount of essential fatty acids (LA and ALA) as AL animals is now reported also in Ames dwarf mice, another mouse model with greatly retarded aging:(1)

Figures 1, 2, 3 and 4 illustrate the proportion of phospholipid DHA (C 22:6 n-3) and LA (C 18:2 n-6) of freshly weaned (1 month old), young adult (2 months old) and adult (6 months old) Ames dwarf mice compared to the heterozygous controls from the same strain. These two FAs were the most abundant PUFA in all tissues (except for LA in the brain) (Table 3). In heart, skeletal muscle and liver, we found increasing differences between the phenotypes as age increased, with lower amounts of DHA and higher proportions of LA in Ames dwarf mice.




Fig. 1: Heart phospholipid docosahexaenoic acid (DHA) content (a) and linoleic acid content (b) in 1-, 2- and 6-month-old Ames dwarf mice and normal-sized littermates. Total n _{Ames dwarf mice} = 18, total n _{normal littermates} = 21; means ± SEM




Fig. 2: Skeletal muscle phospholipid docosahexanenoic acid (DHA) content (a) and linoleic acid content (b) in 1-, 2- and 6-month-old Ames dwarf mice and normal-sized littermates. Total n _{Ames dwarf mice} = 18, total n _{normal littermates} = 21; means ± SEM




Fig. 3: Liver phospholipid docosahexanenoic acid (DHA) content (a) and linoleic acid content (b) in 1-, 2- and 6-month-old Ames dwarf mice and normal-sized littermates. Total n _{Ames dwarf mice} = 18, total n _{normal littermates} = 21; means ± SEM




Fig. 4: Brain phospholipid docosahexanenoic acid (DHA) content (a) and linoleic acid content (b) in 1-, 2- and 6-month-old Ames dwarf mice and normal-sized littermates. Total n _{Ames dwarf mice} = 18, total n _{normal littermates} = 21; means ± SEM

[Note that "short-lived control" here actually just means normal, healthy mice, which are "short-lived" relative to Ames dwarves — not genuinely short-lived animals -MR] ... The proportions of seven out of 13 FAs were affected by phenotype (C 18:0, C 18:1 n-9, C 18: 2 n-6, C 18:3 n-3, C 20:5 n-3, C 22:5 n-3 and C 22:6 n-3), even after the influence of body weight was accounted for (p < 0.05 each time). Note that the amount of C 18:2 n-6 in brain phospholipids was below 1 % in both phenotypes, so brain differed in tissue composition from all the others. DHA content significantly differed between tissues (p  < 0.0001) in both phenotypes, except for brain and muscle amongst the Ames dwarf mice, where the proportions were similar. ... The results from the liver mitochondria revealed again the same pattern with more n-3 PUFAs, namely DHA in the control animals compared with the Ames dwarf mice.(1)

The relative lack of effect on brain PL is also seen in CR animals. It would have been more useful if they had measured mitochondrial membranes instead of total cellular PL, but this clearly supports the Hypothesis.

Second, the same study also reports a (relatively modest) negative correlation between membrane n3 PUFA (no breakdown by EPA/DHA vs. ALA, unfortunately) and lifespan, even in ad libitum animals(1):





Again, it would have been more useful if they had measured mitochondrial membranes instead of total cellular PL.

Reference
1: Valencak TG, Ruf T. Phospholipid composition and longevity: lessons from Ames dwarf mice. Age (Dordr). 2013 May 3. [Epub ahead of print] PubMed PMID: 23640425.

Edited by Michael, 02 August 2013 - 01:02 PM.

[Logic]

I am puzzled as to why Mister E and Logic think their posts are relevant to this discussion.

The title of the thread is 'DHA-Accelerated Aging Hypothesis' Michael.

If I understand all this correctly the hypothesis that a lower amount of DHA than was previously thought may increase lifespan in longer lived mammals, and the evidence seems to support it.

I feel that an increase in FOXO expression by the same means adds weight to the argument, albeit from a different direction to the means currently dominating the discussion.