cron-web.org

[A1CR]

Do Older Humans Live Longer on CR Diet? Yes!

In this 2004-04-06 post, MR provides his usual erudite analysis -- this time on the recent Spindler paper which found relatively late onset CR can work very well at extending mouse lifespan. On the surface Spindler's findings seems to contradict earlier studies showing only very modest lifespan gains from relatively late onset CR. But in this post, MR shows that Spindler's findings aren't really that far out of line with previous studies.

According to MR, the explanation comes down to his favorite mantra "Calories, Calories, Calories." Significant restriction in terms of absolute calorie intake can result in substantial life extension, even when begun late in life -- at least in rodents. This is an important post well worth reading, particularly for people who have started (or are considering starting) CR in middle age or later.

Dean wrote:
>
> MR wrote:
> > I'll confess that I skipped over the calculations here, but these are
> > the actual data [from the new Spindler adult onset CR paper [1]]:
> > ...
> > Again, if we peg this down to max LS 100 [for humans], we get [via
> > extrapolation] more believable numbers on a population basis: 77 [AL mean
> > lifespan], 89 [CR mean LS], 94 [AL max LS], and 109 years [CR max LS].
>
> MR, thank you for the pointer to the abstract and your comments on this
> new Spindler paper on relatively late onset CR. But your comments beg the
> big question: Why did Spindler observe such a large and apparently robust
> increase in mean and max lifespan in mice who started CR relatively late in
> life, when other researchers before him have seen much more modest gains
> from adult onset CR?

It would SEEM that this can in large part be explained by (wait for it!)
Calories, Calories, Calories.

Let's compare the results of (4), a classic get-'em-while-they're-young
study, with 3 adult-onset studies: the classic Weindruch & Walford
study, which first CONVINCINGLY showed that adult-onset CR works (2); a
later paper in animals initiated at about the same age but with
different Caloric intake (3); & Spindler's new breakthru' report (1).
(Below, "Max LS" means "mean LS of longest-lived decile;" life extension
"vs remaining LS" means taking the LS remaining to the CR cohort AFTER
the time of initiation, vs the LS remaining after the SAME time point in
the AL cohort. Apologies if the table comes thru' as a mess; it looks
clean as I send it out).

Citation (4) (2) (3) (1)
Strain C3B10F1 C3B10F1 C57BL6 C3B6F1
Age Initiated 1 12 12 19
Calories/week
AL 85 160 84 93
CR 50 90 62 52.2
%CR (vs AL) 41 44 26 44
Lifespan (months):
Mean 43 36.9 33.2 35.4
Max 51 45.1 41.8 43.6
% Lifespan Extension
Vs Total LS
Mean 30 19 11 15
Max 28 11 11 16
Vs. Remaining LS
Mean 31 18 16 40
Max 28 16 16 32

Now IMO this is a pretty remarkable table. If you look at the STATED %
CR for (4) & (2), it sure LOOKS as if the LS extension gained is
substantially reduced in adulthood, even when you consider it in terms
of REMAINING LS (the reduction in ABSOLUTE LS gains is of course
predictable in any intervention that merely slows (as vs. arresting or
reversing) aging). But then look at the ABSOLUTE Caloric intake. What
the fuck were Weindruch & Walford doing feeding these critters so much
food?? 90 Cal/week?? NONE of the other studies report the "AL" (ie,
10-20% restricted -- I threw out the literal "AL" cohort from (4) for
this purpose) anywhere NEAR this high, even in the same strain. (2) says
that "Control mice ... were fed enough of diet 1 (~160 kcal/week) to
maintain their initial body weights throughout much of their subsequent
lifespans", but why were they allowed to get this big in the first
place? If they'd imported them from some mouse-fattening operation, I'd
understand (tho' I'd consider it a flaw none the less), but the paper is
quite clear that the mice were born, weaned, bred & raised in W&W's colony.

So what we see here is that the mice were allowed to develop an
excessively high setpoint, whereupon a relatively large "%CR" actually
involved only NORMALIZING of the animals' Caloric intake, as vs all of
the other colonies -- including W&W's own previous experiment with the
same starain of mouse in (4)!

Now compare the resultst in (3) with those in (2). the animals wer
initiated at the same age, & while (3) used a much smaller "%CR", it was
relative to a much more normative AL intake. & while you don't get QUITE
as long a LS out of these guys, you do get a much longer LS than you
might expect.

This is esp remarkable since the C57BL6 mouse strain used in (3), while
longevous, is known to naturally be about 10% more *short-lived* than
the C3B10F1 strain from (2) & (4), per the Biomarkers of Aging Program
(5) in which a whole whack of different strains & species of rodents
were maintained under relatively consistent conditions in a large,
long-term LS study. ISTM that this fact makes us justified in paying at
least as much attention to the *%* LS gains as the absolute LS data -- &
here, remarkably, the effects are very similar indeed.

Now, this is all a little iffy ... but ISTM that thinkgs become very
clear when we compare all of this with Spindler's remarkable new study.
FIRST, let's compare (1) with (2) -- the weaning-onset study, performed
in a very closely-related strain of mouse. In this comparison, we see,
that at the same absolute Caloric intake, while (of course)
weaning-onset still gave a longer TOTAL LS, still, remarkably, THE
EXTENSION OF THE REMAINING LIFESPAN IS NEARLY THE SAME. The effect is
esp close for MAX LS -- the proxy for aging per se, the sine qua non of
the CR effect.

This is remarkable, and quite counterintuitive if one assumes that the
irreversible accumulation of damage "ought" to lead to a kind of
"momentum" which, while it can still be slowed, still prevents the full
RELATIVE impact of CR from manifesting. Instead, we apparently see that
while you can't undo what's already been done with CR (which is why
everyone should throw money at the Reversal Prize at <
http://www.methuselahfoundation.org > ), still you can slow down the
RATE of aging to an equal degree at this late date.

This very finding was reported in fruit flies a few months ago (6), &
while it was misinterpreted by some as suggesting that CR started late
in life would lead to the same TOTAL life extension as that started in
youth, still I was skeptical of the relevance to mammals even of the
ACTUAL finding: that the full retardation of aging of CR could be
accessed at any time.

Now, this isn't quite the whole picture, because (aside from the absence
of Gompertz analysis (mortality-rate doubling times)), Spindler actually
found that "Linear regression and breakpoint analysis showed that
between 21 and 31 months of age [ie, AFTER they'd been on CR for 2 mo --
the effect is not INSTANTANEOUS], newly initiated CR decreased the rate
of age-associated mortality 3.1-fold (P < 0.001). Beginning at the
first breakpoint ... the mean time to death increased from 11.8 +/- 0.7
(SE) to 16.8 +/- 1.2 months (SE; P 0.004), a 42% increase. LT-CR
initiated after weaning extends the lifespan of this mouse genotype by
40% (13). [BUT] After the second breakpoint in the CR survival curve (31
months of age), *the mortality rate approximated that of the control
mice,* but the survival curve was shifted to the right by 4.7 months."
Ie, at this point, the LS extension is entirely a "carryover" from
previous CR: "normal" aging hs taken over, & they're just starting from
a biologically younger starting point. WHY this is is even more puzzling
to me; it also makes me wonder if returning to AL at this late point
would have any negative effect on LS.

This is a very optimistic finding, in any case -- and doubtless one that
warms CRON4healthyfuture's heart .

The SECOND important point about the above comparison is to compare the
effects of (2) (Weindruch & Walford's breakthru' -- but in retrospect,
overly pessimism-inducing -- study) and (1). The RELATIVE extension of
LS was actually not that far off -- but only because both the controls &
the CR group in (2) were apparently overfed, overstating the degree of
CR therein. When we then turn to the ABSOLUTE LSs, we see that they are
nearly IDENTICAL. And the reason appears to be dead freaking' simple. To
answer Dean's entirely sensible quesition:

> What gives? Has Spindler found the magic formula to get adult onset CR to
> work very well in rodents, or was his data just a fluke?

Indeed he has. The magic formula is: Calories, Calories, Calories.

-MR

>
> > [1] Joseph M. Dhahbi, Hyon-Jeen Kim, Patricia L. Mote, Robert J. Beaver,
> > and Stephen R. Spindler
> > Temporal linkage between the phenotypic and genomic responses to
> > caloric restriction
> > http://www.pnas.org/cgi/content/abstract/0305300101v1

2. Weindruch R, Walford RL.
Dietary restriction in mice beginning at 1 year of age: effect on
life-span and
spontaneous cancer incidence.
Science. 1982 Mar 12;215(4538):1415-8.
PMID: 7063854 [PubMed - indexed for MEDLINE]

3. Pugh TD, Oberley TD, Weindruch R.
Dietary intervention at middle age: caloric restriction but not
dehydroepiandrosterone sulfate increases lifespan and lifetime cancer
incidence in mice.
Cancer Res. 1999 Apr 1;59(7):1642-8.
PMID: 10197641 [PubMed - indexed for MEDLINE]

4. Weindruch R, Walford RL, Fligiel S, Guthrie D.
The retardation of aging in mice by dietary restriction: longevity, cancer,
immunity and lifetime energy intake.
J Nutr. 1986 Apr;116(4):641-54.
PMID: 3958810 [PubMed - indexed for MEDLINE]

5. Turturro A, Witt WW, Lewis S, Hass BS, Lipman RD, Hart RW.
Growth curves and survival characteristics of the animals used in the
Biomarkers of Aging Program.
J Gerontol A Biol Sci Med Sci. 1999 Nov;54(11):B492-501.
PMID: 10619312 [PubMed - indexed for MEDLINE]

6. Mair W, Goymer P, Pletcher SD, Partridge L.
Demography of dietary restriction and death in Drosophila.
Science. 2003 Sep 19;301(5640):1731-3.
PMID: 14500985 [PubMed - indexed for MEDLINE]
Pop press:
http://www.mail-archive.com/eristocracy@merrymeet.com/msg00255.html