Search the Community

All: Pinged by a diligent contributor to and reader of the Forums, I'm reposting some material I've previously posted in the Archives, since it's clear they won't be resurrected any time soon and since the issue is hot at the moment on the more practially-oriented side in a thread initiated by Tasbin (welcome Tasbin! And welcome back Sirtuin! I'd thought you'd left us — not for the grave : just for other discussion groups). Reduced Circulating IGF-1 and the "CR Triad": Impaired Beta-Cell Maintenance/Renewal? So as most folks will know, many models of reduced IGF-1 signaling slow the degenerative aging process (albeit to varying magnitudes and degrees of convincingness). This evidence includes the existence numerous rodent "longevity mutants" with low IGF-1 signaling, amongst rodents and recently apparently amongst humans with exceptional familial longevity. Part of this data is the mostly compelling evidence of a role for reduced circulating IGF-1 in mediating at least some of the anti-aging effect of CR — particularly its protection against cancer. So the lack of such a reduction in most people on CR in the WUSTL CR Society subjects — indeed, the finding of HIGH IGF-1 in most of us -- was a disconcerting piece of evidence that seemed to point against the human translatability of the anti-aging action of CR into humans. Experimental dietary manipulation proved that what was going on was that in humans, consuming substantially more than the RDA of protein (which many of us were at that time) blocks the CR-induced inhibition of IGF-1 — and that the IGT largely coincided with the most OTHERWISE "CR-like" pattern of low fasting glucose, T3, and testosterone (1,2). However, a very puzzling dichotomy emerged within our group: everyone had good fasting glycemia, but several folks had impaired glucose tolerance. Surprisingly, the IGT folks were also found to have the same insulin sensitivity and initial insulin output on OGTT as those with normal or better glucose tolerance. And when they looked into the factors separating the 2 groups, one pattern that emerged pretty clearly was that folks who had good OGTT (most of us, myself included) ) had normal or high IGF-1, while most folks with good or excellent glucose has the full triad of low IGF-1, T3, and testosterone — exactly as you'd "want" to see on CR. Unfortunately, of course, this implies that lowering your protein to bring down your IGF-1 will lead to this impaired glucose tolerance problem :( . A few years ago now, evidence emerged to suggest that the CR-associated IGT might be the result of the reduction in GH/IGF-1 itself, rather than being an effect directly caused by the macronutrient shift, possibly due to the need for GH/IGF-1 in supporting beta-cell survival or function: Now, of course, CR folk do not eat high-Calorie diets. But even the low-fat-fed GH-deficient mice had abnormally low fasting and postprandial insulin levels (see Fig 4B, above). OTOH, while FASTING insulin is low in low-protein, low-IGF-1, impaired glucose tolerance folks just as it is in the rest of us, the insulin response to OGTT in the IGT folks was actually normal, not low, during most of the glucose tolerance curve -- and in fact, when it starts to fall in those with normal glucose tolerance, it keeps rising even higher in the IGT folks, presumably because the body is still pumping out more insulin to try to get the glucose down [Edit, 2016-08-09: this is a debatable interpretation of the data as Dean notes in a later post, which I should have emphasized and of which I'm actually now inclined to be a bit suspicious (see projected response to Dean): here are the relevant graphs, made postable by Dean]: So these animals do not offer a very compelling model for explaining the paradoxical glycemic response in low-protein, lower-IGF-1 CR folk. [Again, I now acknowledge, debatable]. Still, it's worrisome. Impaired beta-cell function or numbers might be irreversible, short beta-cell transplantation. Reduced IGF-1 Signaling: Loss of Adiponectin-Releasing Metabolic Fat? Ames dwarves are one of those mouse models of slow aging with reduced IGF-1 signaling: their Prop-1 mutation leads to failure of pituitary gonadotropes, somatotropes, & other pituitary cells to develop, resulting in deficient LH, FSH, GH, prolactin, and TSH. They hhave similar insulin sensitivity to CR animals on the same background, and CRing them increases it even further -- and CR robustly increases their LS. OTOH, GHR-KO mice (similar to human Laron syndrome) enjoy only very modest further increases in LS or insulin sensitivity when on CR, and what effect there is is more pronounced on BOTH fronts in males than felmales. Phosphorylation of the insulin receptor tyrosine in response to insulin is enhanced by CR in WT but not GHR-KO, with similar differences are seen in multiple insulin signal transducers. But, insulin and glucose levels are already as low or lower in AL-KO as in WT-CR, and are not FURTHER lowered by CR. Suprisingly, GHR-KO mice are fat. In recent studies,(1) intact GHR-KO mice have similar insulin sensitivity and NS lower insulin than AL mice after visceral fat removal (VFR, which increases insulin sensitivity in AL WT animals, and partially normalized the LS in a study with short-lived controls); VFR leads to still-lower insulin levels in GHR-KO mice, but while glucose levels remain normal in AL, they climb in VFR mutants, associated with a reduction in both insulin and glucose tolerance (ie, when you shoot VFR GHR-KO mice up with glucose, they become hyperglycemic, and they FAIL to become hypoglycemic when shot up with insulin). The reason appears to be that GHR-KO mice have high levels of the insulin-sensitizing adipokine adiponectin, and VFR robs them of this advantage. (In humans, adiponectin is mostly produced in subcutaneous fat; in obesity, its release from subcu adipocytes is normal, but release from VF is low "and better predicts obesity-associated metabolic abnormalities" (pubmed/19219061), but/and people with more VF have disproportionately low adiponectin (pubmed/14747242); these animals seem to have lots of VF adiponectin). Similarly, GHR-KO mice have similar (low) levels of ectopic fat in skeletal muscle to VFR WT; AL WT *and* VFR GHR-KO mice have HIGHER levels than the aforementioned. Interestingly, adiponectin-overexpressing TG mice have what IGF-1 investigator Bartke characterizes as a similar metaboic profile to that of Ames and GHR-KO mice (pubmed/17204560), altho' that seems to be only very fuzzily so to me; similarly, he notes several recent and older studies (PMID: 12543271, 21070591, 20398121, 20157552, 17228087, 16891987, 8967479, 21191145, 15582274) showing that centenarians (mostly drawing from cohorts with familial longevity) and offspring in longevous families show resistance to the "metabolic syndrome" (notably, lower fasting insulin levels & higher sensitivity and/or lower fasting glucose vs normals, and lower incidence of diabetes; PMID 15582274 was actually a genotype rather than phenotype study). I do note that PMID: 16891987 does specifically report that "In [general-population] centenarians we found that adiponectin concentrations were significantly increased, compared with young, early elderly and obese women. Insulin concentrations were lower than those in young and obese subjects. HOMA-IR [insulin resistance] was significantly lower than in obese women. Positive correlations were found between adiponectin and HDL, and negative correlations between adiponectin and HOMA-IR, total cholesterol, LDL, triglycerides, blood pressure and BMI." And, the Leiden Longevity Study and a couple of others found no effect on IGF-1 levels, tho' in other cohorts there are eg. changes in receptor polymorphism in and similar transduction pathways in some others. Based on the discordant effects of CR, Ames mutation, and GHR-KO on insulin sensitivity & longevity, Bartke had already proposed (pubmed/19304940 ) that "Insulin sensitivity %5Bis%5D a key mediator of growth hormone actions on longevity"; I don't know if they already had the results of the unpublished & unmentioned VFR experiments to hand at the time, but in later papers he uncovered evidence that the lipokine adiponectin is a key mediator of all of this. This might be an elegant, if somewhat disturbing, explanation for the about the curious intersection of impaired glucose tolerance with low IGF-1 and other putative mediators of CR in a subgroup of our human cohort (see discussioin in my post linked above, and the post linked inside of it in turn), except that BOTH subgroups of human practitioners seem to have high adiponectin levels and good insulin sensitivity! In response to a separate question from me, re: the aforementioned (3) study showing impaired insulin release and associated glucose intolerance in inducible, adult-onset GH deficiency in mice, and about how it might intersect with the findings in our human cohort (recall that this doesn't quite match what SEEMS to be going on in the human subgroup, as first-phase insulin release SEEMS normal but just doesn't keep glucose down at the peak, leading to ongoing insulin release when it's returning to baseline in the other subgroup, despite ALL CR folk apparently having excellent insulin sensitivity per se: PMID 19904628) -- in response to my inquiry on this, and its relationship to the mutants he studies, Bartke referred to an earlier (1995) finding of his that "Islet volume ... decreased 2- to 5-fold in [Ames] dwarf mice. Analysis of the distributions of islet sizes revealed that almost all of the volume ... decreases in dwarf mice were accounted for by alterations in the numbers and sizes of large (diameter, > 150 microns) islets" - PMID: 7720649. He also SAID, at a previous CR Society Conference, that he had found this for insulin secretory reserve. Moreover, digging, I see from an independent group that "Adult GHR(-/-) mice exhibited significant reductions in the levels of blood glucose and insulin, as well as insulin mRNA accumulation. Immunohistochemical analysis of pancreatic sections revealed normal distribution of the islets despite a significantly smaller size ... [ie,] only one-third of that in wild-type littermates. Total beta-cell mass was reduced 4.5-fold in GHR(-/-) mice, significantly more than their body size reduction. This reduction in pancreatic islet mass appears to be related to decreases in proliferation and cell growth. GHR(-/-) mice were different from the human Laron syndrome in serum insulin level, insulin responsiveness, and obesity"(5). This last statement (that THESE guys' GHR-KO mice are NOT obese) appears to be rather misleading, when you dig into the full text: Reviewing the possible human translatability of impaired GH/IGF-1 signaling: again, several centenarian populations DO seem to have reduced signaling thru' the pathway (PMID 19489743, 18316725, 15771611), tho' it must be said that in the general population low IGF-1 levels put the elderly at incraesed risk of mortality (I think it's reasonable to suggest that this is reverse causation, or failure to compensate, rather than a lifelong effect on aging: CR animals notably have low IGF-1 in youth but higher IGF-1 than age-matched AL in late life). Untreated patients with lifelong isolated GH deficiency have no early athero deespite central obesity, elevated cholesterol, and BP (PMID 16522693); not cited by Bartke, I find that the same group "conducted a cross-sectional study of 20 IGHD individuals (seven males; age, 50.8 ± 14.6 yr) and 22 control subjects (eight males; age, 49.9 ± 11.5 yr) ... Adiponectin was higher [12.8 (7.1) vs. 9.7 (5) ng/ml; P = 0.041] ... whereas no difference was observed in leptin [7.3 (6.3) vs. 9.3 (18.7 ng/ml] and UAE [urinary albumin excretion, which they say is "a marker of endothelial disease"] [8.6 (13.8) vs. 8.5 (11.1) μg/min]. ... [H]igh adiponectin and normal leptin levels may delay vascular damage and lesions of the renal (and in this context, vascular? -MR] endothelium" - pubmed/20016047] Similarly, "Obese adults with ... Classic Laron Syndrome ... a recessive disease of insulin-like growth factor I (IGF-I) deficiency and primary growth hormone insensitivity, clinically characterized by dwarfism and marked obesity ... have normal endothelial function" - pubmed/17320443 . "Patients with congenital deficiency of IGF-I seem protected from the development of malignancies: a preliminary report" (PMID: 17166755) -- NB, in a young cohort. And see also the recent PMID 21325617. Observed CR-Associated IGT in Mice, Monkeys, and Men Some have suggested that this might be a question of some people tested in Luigi's studies having higher BMI and/or adipose tissue and/or energy intake because of varying practices (Dean, tho' he was "CR proper" at the time; Paul McG) and/or the sheer statistical noise of the small size of our cohort. There are aspects of the data that wold seem consistent with that idea, but overall I think the evidence tends to rule out the former kind of explanation: "BMI was significantly lower in the CR-IGT subgroup than in the CR-NGT subgroup (Table 4). Total energy intake was not different between the two CR subgroups [tho' it should be noted that in fact it was nonsignificantly higher in the IGT group, which if real would support your hypothesis], but fiber intake was significantly higher in the CR-IGT group."(1) The CR-IGT subgroup includes a lot of people with quite low Calorie intake; my own apparent transition into CR-IGT (see below) happened while I was lowering Calorie intake (I had been >1900 Cal/d then, and was <1800 Cal when I first observed it ); and perhaps most tellingly the CR-IGT subgroup has the MOST "CR-like" endocrinology: low IGF-1, testosterone, and active T3, suggesting that they are MORE CRed from a physiological/metabolic standpoint. (As a reminder, when I was originally tested by Luigi on a relatively high-protein diet and with very high IGF-1 in 2006, my OGTT was the best he had ever seen (and I was tested after Saul, whose OGTT was the best out of the rest of the cohort by a significant margin). I haven't been in for an OGTT since my protein reduction, but my HbA1c has gone from ≤5 (normal) in all tests prior to lowering my protein intake to ~5.6 (high-normal to at-risk) afterward in a series of tests. Since it's not fasting glucose that's the problem, it's evidently postprandial). (I also note, without intending to alarm him and with apologies for Fermat's Theorem risk, that Dean's recent bloodwork (Posted 01 July 2016) is no longer quite in line with this profile). [MR, 2016-08-09: Somehow miscalculated this. Dean's bloodwork is fully in line therewith]. Another counterargument to this being truly a CR phenomenon: It is DEFINITELY not the right metric, since (first) the CR-IGT subgroup appears, based on HOMA-IR, to have *excellent* insulin sensitivity. You're right that the CR animals don't develop diabetes as measured by fasting glycemia, but our fasting glycemia is also excellent -- whereas there IS evidence, in mice(6) and nonhuman primates (7), of impaired glucose tolerance and in particular apparent impairment in insulin output. Now, the abstract of (6) might actually seem to support a (very odd) too-many-Calories-on-CR hypothesis: "Glucose tolerance curves were unchanged by age in ad libitum fed or 50% restricted animals, but in 80% ad libitum [ie, 20% CR] groups, older animals showed evidence of decreased glucose tolerance with respect to young animals." In fact, however, it finds the opposite of what my interlocutor is proposing. As is plain from their Figure 2 (excerpt below), the young and old AL animals have similar "OGTT" to one another, rising from 100 mg/dL fasting to <150 under glucose challenge. The young 20%-CR mice have a similar curve to the young and old AL animals (perhaps even slightly better), but the "OGTT" worsens in the old 20% CR animals, going up to ~175 mg/dL, unlike the AL aged mice. And the kicker is the 50% CR animals, who have far worse OGTT at any age than any other group, spiking to ~210 mg/dL. Ie, more severe CR --> worse OGTT, and earlier in life. And, while it's hard to know for sure, it sure LOOKS like the reason is failure to secrete enough insulin in response to the glucose load. Here is what Walford & colleagues say in their Discussion, further complicating the matter and reducing any comfort from mouse glycemia studies: [Edit: added 2016-08-23]: Somewhat similarly, though with a more optimistic ultimate finding, the "CR" primates appear to have lower beta-cell response to glucose (although in both cases there is ambiguity about their "CR" status). At NIA, "Several measures of the insulin response (baseline, maximum, and integrated areas under curve) increased with age and were lower in DR monkeys. ... {I]ntegrated insulin response was lowered in DR monkeys compared with controls .... [similar to a previous report on] DR in rats. Acute insulin response (first-phase) and second-phase insulin response, represented by areas under the curve for O-10 and 10-60 min, respectively, were also lowered in DR monkeys."(10) However, in the NIA primates, this was accompanied by a lower peak glucose concentration during OGTT, suggesting that beta-cell function was adequate, and was lower simply because it didn't need to be any higher, due to the animals' enhanced insulin sensitivity. "These findings confirm earlier reports (5) that age-related increases in insulin levels, which could develop into hyperinsulinemia and diabetes, are ameliorated by calorie restriction."(10) There is a parallel at WUSTL, too, though the result is not actually the same, and we have to be particularly skeptical of the relevance of the result, as it is pretty clear that the WUSTL "CR" study was not a proper CR study, but a study of avoidance of obesity by controlled energy intake. Still, it's still notable (and in some senses even more striking) that even following mere obesity-avoidance amongst the nonhuman primates at WUSTL, (10) "[CR] monkeys had less body fat, lower basal β-cell sensitivity to glucose (Ø(b)), greater insulin sensitivity, and lower first-phase plasma insulin response. DR did not significantly affect first-phase and second-phase β-cell sensitivity to glucose."(11) This is not the contradiction it seems: basal sensitivity is the unchallenged sensitivity of the beta-cell to fasting glucose, and first-phase insulin sensitivity is the degree to which beta-cells are activated in response to rising glucose; first-phase insulin response reflects the level of plasma insulin that appears in circulation in response to incoming glucose from IVGTT. What they find, then, is that their "CR" animals' beta-cells do respond to incoming glucose by producing more insulin, but it is more rapidly cleared from the plasma: All well and good, so long as insulin secretion and plasma levels are adequate to match postprandial glucose. The problem for CR-IGT is that whatever the underlying mechanism, the latter is not happening without additional steps to control it.[Here endeth edit of 2016-08/23] mTOR and "Starvation Diabetes" CR does inhibit the autophagy-repressing, protein-synthesis-enabling mTOR, which is the mechanism of action of rapamycin, the first (but, happily, no longer the sole) true anti-aging drug in rodents. And so a parallel phenomenon in rapamycin-treated mice, and related matters, as well as a potential mechanism for it involving the differing metabolic effects of inhibiting the two mTOR complexes (mTORC1 and mTORC2), are suggestive — and way around the effect in drug design is promising, if not clearly actionable (tho' Dean has hypothesized that it may be via cold exposure): On the one hand, the recent finding of the predicted additive or synergistic effect of adding metformin to rapamycin — despite the lack of effect of metformin alone on lifespan in normal, healthy mice — is supportive of the models in (8,9). On the oher hand, a lot of papers have now been published showing divergent effects of CR and rapamycin, and it's not clear to me how this impacts the translatability (tho' I've not looked into this more than superficially). Apologies for lack of elaboration ... More may possibly come later, but no promises ... Health Implications Of course, this is all on its face scary shit: both possible beta-cell loss, and postprandial hyperglycemia, and as discussed in recent threads like this and this, people exhibiting CR-IGT are prudently attempting to manage the hyperglycemia in particular. It must be said, tho', that however intuitively bad this all sounds, CR mice none the less retard the aging process and live dramatically longer than AL mice, despite exhibiting these same phenomena. This would seem to stand as a counterargument even in the face of uncertainty about the human translatability of CR, since this is exactly a translatable phenomenon — both the glucoregulatory pattern itself, and the associated endocrinology. And, as discussed here, the evidence on actual implications of all of this on health and aging are often surprising in their absence, in mouse and man. I certainly still think it's prudent to work to bring down postpradial hyperglycemia, but there are clearly some things going on that my own little brain has not managed to capture and digest. References 1: Luigi Fontana, Samuel Klein, John O. Holloszy. Effects of long-term calorie restriction and endurance exercise on glucose tolerance, insulin action, and adipokine production. Age (Dordr). 2010 Mar;32(1):97-108. Epub 2009 Nov 11. PMID: 19904628 [And see, when it's up: http://www.crsociety.org/archive/read.php?2,197226,197244#msg-197244 2: Fontana L, Weiss EP, Villareal DT, Klein S, Holloszy JO. Long-term effects of calorie or protein restriction on serum IGF-1 and IGFBP-3 concentration in humans. Aging Cell. 2008 Oct;7(5):681-7. PubMed PMID: 18843793; PubMed Central PMCID: PMC2673798. 3. PLoS One. 2011 Jan 19;6(1):e15767. Metabolic impact of adult-onset, isolated, growth hormone deficiency (AOiGHD) due to destruction of pituitary somatotropes. Luque RM, Lin Q, Córdoba-Chacón J, Subbaiah PV, Buch T, Waisman A, Vankelecom H, Kineman RD. PMID: 21283519 http://dx.plos.org/10.1371/journal.pone.0015767 4. Masternak MM, Bartke A, Wang F, Spong A, Gesing A, Fang Y, Salmon AB, Hughes LF, Liberati T, Boparai R, Kopchick JJ, Westbrook R. Metabolic effects of intra-abdominal fat in GHRKO mice. Aging Cell. 2011 Oct 31. doi: 10.1111/j.1474-9726.2011.00763.x. [Epub ahead of print] PubMed PMID: 22040032. http://dx.doi.org/10.1111/j.1474-9726.2011.00763.x 5. Liu JL, Coschigano KT, Robertson K, Lipsett M, Guo Y, Kopchick JJ, Kumar U, Liu YL. Disruption of growth hormone receptor gene causes diminished pancreatic islet size and increased insulin sensitivity in mice. Am J Physiol Endocrinol Metab. 2004 Sep;287(3):E405-13. Epub 2004 May 11. PubMed PMID: 15138153. 6. Harris SB, Gunion MW, Rosenthal MJ, Walford RL. Serum glucose, glucose tolerance, corticosterone and free fatty acids during aging in energy restricted mice. Mech Ageing Dev. 1994 Mar;73(3):209-21. PubMed PMID: 8057691. 7. Gresl TA, Colman RJ, Havighurst TC, Allison DB, Schoeller DA, Kemnitz JW. Dietary restriction and beta-cell sensitivity to glucose in adult male rhesus monkeys. J Gerontol A Biol Sci Med Sci. 2003 Jul;58(7):598-610. PubMed PMID: 12865475. 8. Blagosklonny MV. Once again on rapamycin-induced insulin resistance and longevity: despite of or owing to. Aging (Albany NY). 2012 May;4(5):350-8. Review. PubMed PMID: 22683661; PubMed Central PMCID: PMC3384435. 9. Lamming DW, Ye L, Katajisto P, Goncalves MD, Saitoh M, Stevens DM, Davis JG, Salmon AB, Richardson A, Ahima RS, Guertin DA, Sabatini DM, Baur JA. Rapamycin-induced insulin resistance is mediated by mTORC2 loss and uncoupled from longevity. Science. 2012; 335: 1638-1643. 11: Lane MA, Ball SS, Ingram DK, Cutler RG, Engel J, Read V, Roth GS. Diet restriction in rhesus monkeys lowers fasting and glucose-stimulated glucoregulatory end points. Am J Physiol. 1995 May;268(5 Pt 1):E941-8. PubMed PMID: 7762649. 11: Gresl TA, Colman RJ, Havighurst TC, Allison DB, Schoeller DA, Kemnitz JW. Dietary restriction and beta-cell sensitivity to glucose in adult male rhesus monkeys. J Gerontol A Biol Sci Med Sci. 2003 Jul;58(7):598-610. PubMed PMID: 12865475.