Jump to content

  • Log in with Facebook      Sign In   
  • Create Account       

Photo

Human CR Greatly Suppresses Senescent Cell Accumulation in Colon

senescent cells human cr

  • Please log in to reply
7 replies to this topic

#1 Michael R

Michael R
  • Moderator
  • 572 posts
  • Website URL:http://www.sens.org/

Posted 28 March 2018 - 10:33 PM

All:
 

Aging Cell. 2018 Mar 25:e12746. doi: 10.1111/acel.12746. [Epub ahead of print]
The effects of graded caloric restriction: XII. Comparison of mouse to human impact on cellular senescence in the colon.
Fontana L1,2, Mitchell SE3, Wang B4, Tosti V1, van Vliet T4, Veronese N1, Bertozzi B1, Early DS1, Maissan P4, Speakman JR3,5, Demaria M

... Here, we show that a number of classical transcriptomic markers of senescent cells are reduced in adult but relatively young mice under CR. [FULL: "mice were aged 20 weeks when they entered four levels of CR for 12 weeks: 10%, 20%, 30%, and 40% restriction from baseline food intake"]. Moreover, we demonstrate that such senescence markers are not induced in the colon of middle-age human volunteers under CR in comparison with age-matched volunteers consuming normal Western diets.

[Specifically: "we recruited and studied 12 middle‐aged (61.7 ± 8.4 years), weight‐stable very lean (BMI = 19.1 ± 1.3 kg/m2) members of the Calorie Restriction Society who have been practicing ~30% CR with adequate nutrition (at least 100% of RDI for each nutrient) for an average of 10.1 years ... and a control group of 12 nonobese (BMI = 27.4 ± 2.5 kg/m2) age‐matched sedentary controls eating a typical Western diet (WD‐o; Figure 2a) ... [whom] we compared ... with younger (24.3 ± 2.0 years, range 21–27 years) nonobese (BMI = 25.7 ± 0.9 kg/m2) humans (WD‐y). All the genes measured were expressed at higher level in WD‐o than in WD‐y volunteers."

"Levels of p16 were significantly lower in the CR compared to WD‐o volunteers (Figure 2b). Levels of p21 followed the trend observed in p16, but did not reach statistical significance (Figure 2c). In accordance with a previous study, we observed significantly lower level of the pro‐inflammatory cytokine IL‐6 in the CR colon mucosa (Figure 2d; You, Sonntag, Leng & Carter, 2007). The other SASP factors analyzed Cxcl1, Il8, Il1a, and Mmp9 followed similar trends, but only the latter two reached statistical significance (Figure 2e)." [I can't post the Figure here due to Forum software rules, but see

 

https://wol-prod-cdn.literatumonline.com/cms/attachment/ea5d511e-f43a-42f1-9361-9284e344a7fa/acel12746-fig-0002-m.jpg

[MR: the gist of it, tho' the numbers are not all statistically significant, is pretty clearly that 61-y.o. CR practitioners' senescent cell burden is clearly much closer to that of 24-y.o. AL than to age-matched AL controls, with p16, IL-6, and IL-1α (which last enforces the SASP through a positive feedback loop) being the most dramatic.]

PMID: 29575469

https://onlinelibrar...1111/acel.12746

 

Diet quality may play into this, and the fact that the 27.4 ± 2.5 kg/m2 average BMI in the controls includes some significantly overweight people. And it's n-12, and only in the colon, where narrow diet effects might be most strongly observed (note that dietary influences on colorectal and GI cancers are more consistent than other tissues). But it seems pretty good evidence that human CR suppresses SC accumulation rather strongly, consistent with CR's tranlatability.



#2 Mechanism

Mechanism
  • Supporting
  • 473 posts

Posted 29 March 2018 - 06:08 PM

Interesting study Michael, promising but also agree with the limitations you mentioned. Thank you for posting it.

Additionally, exercise does the same thing: http://diabetes.diab...6.full-text.pdf

This poses a couple of closely related questions:

1) Could this have been achieved with exercise as an alternative to CR ( with a low normal range BMI)

2) Since the CR group is self-selected, how much of the difference observed in the CR group was mediated by potential differences in exercise itself? — CR tends to attract the health-conscious to orthorexic crowd and while some sedentary CR individuals were no doubt included in the CR group, these can easily be outnumbered by higher intensity exercise within the CR group even if the younger control group was not sedentary per se. Is it unreasonsble to entertain that confounding may mediate anywhere from some to all of the purported effects of CR in this study?

Excercise and CR are of course not mutually exclusive and it is possible they have synergistic effect.... but It may be equally conceivable that either is sufficient for full or near-maximal effect.

While hedging bets by doing both is one strategy, if the cost/effort for one is substantially lower/higher than for the other ( eg- exercise over CR), clarity whether exercise alone is adequate is helpful

N.b., yes, at least one study I know of suggested for one breed that CR+exercise>CR alone though the outcome variable was not senescence per se and moreover in contrast with this study, it was not in carried out in human subjects which was a key strength of this study. Of course there is a vast CR literature I am not familiar with and I am interested to evaluate any examples and counterpoints.

Edited by Mechanism, 29 March 2018 - 08:35 PM.


#3 Michael R

Michael R
  • Moderator
  • 572 posts
  • Website URL:http://www.sens.org/

Posted 30 March 2018 - 07:42 PM

Interesting study Michael, promising but also agree with the limitations you mentioned. Thank you for posting it.

Additionally, exercise does the same thing: http://diabetes.diab...6.full-text.pdf

This poses a couple of closely related questions:

1) Could this have been achieved with exercise as an alternative to CR ( with a low normal range BMI)


Well, the paper you link clearly rather strongly suggests that it can't. Note that the human CR paper looked at people aged on av'g 61 at time of sampling, who had been on CR an average of 10 years — but they had senescent cell burdens similar to those of lean controls who were 37 years younger. By contrast, in the paper you link, there were no significant differences in SC burden between lean, Normal Diet-fed sedentary mice and even leaner Normal Diet-fed exercised mice, despite the latter having less visceral fat— and if anything, in the visceral fat tissue of greatest interest (as well as in pancreas), it looked as if these animals had more senescent cells:
 

F3.medium.gif


Exercise-induced weight loss blunted the excess accumulation of senescent cells inflicted on animals by a high-fat, high-HFCS, obesogenic diet, but it did nothing for the "normal" age-related burden of SC.
 

2) Since the CR group is self-selected, how much of the difference observed in the CR group was mediated by potential differences in exercise itself? — CR tends to attract the health-conscious to orthorexic crowd and while some sedentary CR individuals were no doubt included in the CR group, these can easily be outnumbered by higher intensity exercise within the CR group even if the younger control group was not sedentary per se. Is it unreasonsble to entertain that confounding may mediate anywhere from some to all of the purported effects of CR in this study?


I have no idea what you're talking about here.
 

Excercise and CR are of course not mutually exclusive and it is possible they have synergistic effect.... but It may be equally conceivable that either is sufficient for full or near-maximal effect.


Again, there is evidently none from exercise in already-lean mice.
 

While hedging bets by doing both is one strategy, if the cost/effort for one is substantially lower/higher than for the other ( eg- exercise over CR), clarity whether exercise alone is adequate is helpful

N.b., yes, at least one study I know of suggested for one breed that CR+exercise>CR alone though the outcome variable was not senescence per se and moreover in contrast with this study, it was not in carried out in human subjects which was a key strength of this study. Of course there is a vast CR literature I am not familiar with and I am interested to evaluate any examples and counterpoints.


There is a substantial body of research showing that CR retards aging and exercise doesn't except possibly as an add-on to more moderate CR, most importantly the lifespan studies summarized in a post that is MIA in the Archives, but includes notably PMIDs 9177588 and 9049716.

#4 Mechanism

Mechanism
  • Supporting
  • 473 posts

Posted 30 March 2018 - 08:39 PM

Thanks Michael, I appreciate your thoughts on the study-

I agree with your statement: “Exercise-induced weight loss blunted the excess accumulation of senescent cells inflicted on animals by a high-fat, high-HFCS, obesogenic diet, but it did nothing for the "normal" age-related burden of SC.”

Might the reduced rate of SC accumulation in the context of an obesogenic diet understate the benefits of exercise had the mice been given a normal diet? ie, reducing SC relative to sedentary age-matched controls rather than merely slowing down their accelerated accumulation in the setting of an obesogenic diet. If this is an open question, then confounding by differences in exercise status relative to controls may over longer stretches of time mediate some of the difference in SC burden in the CR group relative to the sedentary controls.

Edited by Mechanism, 31 March 2018 - 05:43 AM.


#5 Michael R

Michael R
  • Moderator
  • 572 posts
  • Website URL:http://www.sens.org/

Posted 31 March 2018 - 10:38 PM

Thanks Michael, I appreciate your thoughts on the study-

I agree with your statement: “Exercise-induced weight loss blunted the excess accumulation of senescent cells inflicted on animals by a high-fat, high-HFCS, obesogenic diet, but it did nothing for the "normal" age-related burden of SC.”

Might the reduced rate of SC accumulation in the context of an obesogenic diet understate the benefits of exercise had the mice been given a normal diet? ie, reducing SC relative to sedentary age-matched controls rather than merely slowing down their accelerated accumulation in the setting of an obesogenic diet. If this is an open question, then confounding by differences in exercise status relative to controls may over longer stretches of time mediate some of the difference in SC burden in the CR group relative to the sedentary controls.


But the point is that we don't have to guess about this: they had both sedentary and exercised Normal-Diet-fed, lean mice, and in these animals, and there was no effect (or, if anything, it made the burden of SC in the visceral fat worse). It didn't slow it down; it didn't reduce it; it did diddley squat. So this strongly suggests that exercise isn't some kind of cryptic confounder in comparing lean and CRed humans, either.



#6 Mechanism

Mechanism
  • Supporting
  • 473 posts

Posted 01 April 2018 - 07:39 PM

Please point out if I am mistaken, but on my read that does not appear to be quite right:

The mouse treatment study I cited found, in the words of the author ( p. 1611) "To explore whether exercise mitigates diet-induced cellular senescence, we conducted gene expression profiling on visceral adipose extracts. Similar to the prevention study, exposure to the FFD for 30 weeks resulted in a significant fourfold increase in p16 expression in visceral fat of sedentary mice (Fig. 5E). Despite continued consumption of the FFD, the treatment of obese mice with exercise for 14 weeks reverted this effect, as evidenced by reduced expression of p16 to levels that were not different from ND-fed control mice."

They went on there fill in some additional nuances on the other biomarkers with some mixed results but on the whole felt and concluded "Cumulatively, these results point to diminution of a diet-induced senescent phenotype in visceral adipose tissue as a means by which exercise provides restorative benefit to abate chronic disease after long term sedentary behavior and nutrient excess."

Now I completely appreciate the distinction you emphasized between

(1) exercise preventing and even reversing [ at least in visceral fat] a senescent phenotype in the setting of an obesogenic sedentary lifestyle ( such as we saw in the mouse study)

versus on the other hand

(2) long term CR practitioners having expression levels of senescence and SASP markers in the colon more closely resembling the youthful controls than those for the age-matched controls.

You emphasize in this distinction that preventing or even reversing diet-induced senescence down to control group levels of senescence is not the same thing as having senescence revert to or more closely match levels found in more youthful individuals than in age-matched controls.  

However here lies the problem with the study you cited - Whereas the mouse study controls were similar to the intervention groups ( and had the added benefit of randomization of unknown as well as known potential as it was an experimental design), in contrast, the CR society study you cited has a self-selected intervention group of CR practitioners with middle-aged controls that are not suitable for the inferences drawn.

 

Not only do we have reason to believe that individuals selecting CR are likely to be more health conscious than average ( lest of all orthorexic), and hence more likely to exercise and consume a less obesogenic diet than average, practices and in all likelihood a variety of other pro-longevity lifestyle practices, but also the authors additionally and explicitly indicated that the age-matched controls were weak indeed ( here we go NIA monkey trial again!) characterizing them as e" a control group of 12 nonobese (BMI = 27.4 ± 2.5 kg/m2) age‐matched sedentary controls eating a typical Western diet"

So these controls (1) had higher BMIs ( BMI of 27.4 is not obese, but a far cry from lowest morbidity end of the BMI curve in the 18-24 range [ I explicitly factor in that they are older with a wider range of BMIs as some data suggests a little extra weight may help]  and (2) they had a "typical Western diet" . 

 

A true CR study would compare CR with a healthy if not truly optimal at the very least "Dietary Guidelines for Americans" diet) and (3) they were described by the authors as "sedentary."  The CR hypothesis pertains to an impact of calories controlling for other variables and I would be hard pressed to find more impactful potential confounders requiring control that were not accounted for..

The relevance here is that we don't know whether it is CR per se that mediated the observed differences, or whether a better diet, exercise, other healthiful lifestyle factors, or some potentially synergistic combination of health behaviors differentiate the very health-conscience CR practitioner group from a handful of age-matched sedentary ~27 BMI adults consuming the standard American diet. There is no reason to believe it is the CR that made them more like the lean young controls than due to any other health behavior.

Now recall in the mouse study that SC burden was lower in visceral fat -- so we have:

(1) The observations in your study does not exclude that any number of health seeking behavior related variables besides CR ( yet associated with health-conscious individuals who practice CR relative to sedentary individuals on the SAD) is responsible for the reduced colon tissue SC in the human study.

 

and 

(2) In the CR Society study the only tissue sampled was from the colon, so I see no evidence presented that this population had a global reduction in SC ( outside the GI tract) either ( we saw no reduction in SC in the pancreas, liver, heart in the mouse study).  Granted, there is no evidence of an absence of such effect in the CRSociety study since it was examined to my knowledge ( please correct me if I am mistaken), but the burden of proof is to demonstrate an effect in other organ systems rather than for it to the default assumption.

Now playing devil's advocate on my poin (2) above - I do recognize that the mouse study did fail to show reversal of SC in the GI tract which you point out was seen in the CR Society study --- a key observation that deserves explicit mention here.  I absolutely agree with you here that this is a "promising" finding and deserves further investigation.  However, the fact remains that given the major differences in health behaviors ( not limited to exercise) between CR group and the age-matched controls we have a major attribution problem: We cannot attribute it to CR per se,

 

This situation is exacerbated by the long-term nature of CR practice: it may be explained in part or whole by the other health behaviors in the CR group over 10+ years ( I include the plus(+) because pro-healthful behaviors often start well before establishing a CR practice, lest such behaviors indeed by confounding the observed differences in colon SC burden).  Already simply maintaining a long-term CR practice says something very special about that individual who perhaps represents the top 0.1%-1% of health conscious individuals.

I would suggest that the next study incorporate an age-matched "healthy but not CR diet" and "physically active lifestyle" control, more akin to the exercise ( non-CR) group from Fontana's classic RCT on the 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. doi: 10.1007/s11357-009-9118-z
PMID 19904628
https://www.ncbi.nlm...es/PMC2829643/

It did not elude me that Dr Fontana is one of the authors of the very same CR Society study under discussion here.  I have much respect for him and his work ( thank you Dr. Fontana!), and he indeed generally produces very high quality work.  It is not the quality of the methodology but inferences drawn that I object to. Namely, that is CR that is responsible versus merely that exercise and avoiding the SAD and embracing other prohealth behaviors may be responsible for some to to all of the differences noted.   

 

I am very pleased the research done, but much more work needs to be done to determine whether this effect was from CR or something else, as well as to confirm and quantify the hard health status outcome(s) clinical impact of the finding.

 

I sincerely hope that some facts have evaded me in this regard and I would be quick to update my position if I missed something.   As a student of science and in the spirit of learning and mutual exchange I pose these issues for discussion in my exchange with you Michael on this forum.


Edited by Mechanism, 02 April 2018 - 06:28 AM.


#7 Michael R

Michael R
  • Moderator
  • 572 posts
  • Website URL:http://www.sens.org/

Posted 08 April 2018 - 02:04 PM

Please point out if I am mistaken


OK: you're mistaken ;)xyz .
 

The mouse treatment study I cited found, in the words of the author ( p. 1611) "To explore whether exercise mitigates diet-induced cellular senescence, we conducted gene expression profiling on visceral adipose extracts. Similar to the prevention study, exposure to the FFD for 30 weeks resulted in a significant fourfold increase in p16 expression in visceral fat of sedentary mice (Fig. 5E). Despite continued consumption of the FFD, the treatment of obese mice with exercise for 14 weeks reverted this effect, as evidenced by reduced expression of p16 to levels that were not different from ND-fed control mice."


Right. You're looking at the wrong group, Mechanism. Again: yes, the data shows that if you're obese, eating a ridiculously high-fat/high-sugar diabetes-inducing diet (FFD), you'll accumulate supernumerary senescent preadipocytes, and doing exercise (and accompanying weight loss) will bend the curve on its progression to similar to sedentary animals eating a Normal Diet (ND). But the same study also shows that exercise has no effect on those merely eating a Normal Diet, and thus not becoming overweight. There is no excess here to slow.
 
In the mouse study accompanying the human study, even 10 or 20% CR was inadequate to reduce the burden of senescent cells purely attributable to aging (normal but ad libitum diet) (tho' arguably the "10% CR" group here were the real controls); it's at 30 (or maybe we should say 20) or 40% (30%) CR that you see a genuine effect on obligate aging-induced SC.
 
In the human study itself, all the CR subjects are lean and eating healthy diets, whether they exercise or not, and even the young controls are only slightly overweight; an effect against obesity/crap-diet-induced excess just falls out of it.



#8 Mechanism

Mechanism
  • Supporting
  • 473 posts

Posted 08 April 2018 - 07:22 PM

Right Michael, I appreciate that point, which I could have more clearly acknowledged in my post above. Thank you for the exchange and as always your open mind and good sense of humor.

My post above was lengthy and could have been better organized -- buried in it, and unaddressed as far as I can tell, is my fundamental concern: It is the poor choice of controls in the CR study that I object to.

I would characterize both sets of CR study controls as follows: The differences between the controls and the CR participants were highly clinically significant and troubling as a potential source of confounding -- Per the study authors, these were BMI = 27.4 ± 2.5 kg/m2 age‐matched "sedentary" controls consuming a "standard American diet."

Regarding my residual references to exercise following your initial post I want to clarify and emphasize here that I am not so much at this point concerned about exercise as an isolated variable per se, which you addressed ( though in all fairness we can't rule out exercise synergistically and even conditionally interacting with differences in other health related behaviors; i.e., through "effect modification”), so much as much as the collective impact of all the practices that go with individuals who tend to exercise or be health conscious ( CR participants) versus those of individuals described as "sedentary" and consuming the "standard american diet.”

Not only are the sedentary, SAD-consuming controls less health conscience (based on these descriptions taken directly from the manuscript) than healthy non-CR potential controls (not used in this study) who would at least try to follow the basics of the "Dietary Guidelines for Americans," but we additionally know well from our forum here that CR practitioners tend to border on orthorexia, optimizing diet with nutrient dense whole foods (the "ON" in CRON), being mindful of evidence-based health practices including regular visits to the doctor, lest of all cold exposure, other biohacking, etc.

I would suggest that this study failed to incorporate ( and the next study should incorporate) age-matched "healthy but not CR diet" controls more akin to the exercise ( non-CR) group from Fontana's classic RCT on the 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. doi: 10.1007/s11357-009-9118-z
PMID 19904628
https://www.ncbi.nlm...es/PMC2829643/

I do not mean that the controls need to run or ride long distances per se ( again exercise not a variable we are looking at in isolation but a surrogate variable for other health conscious behavior including dietary choices), but rather that at the very least they should have a BMI in the 18.5 - 22 range ( as opposed to 27.4), and have a whole foods based diet ina non-sedentary population. The latter can be identified based on food frequency survey or other survey data inclusion criteria; likewise they can be assembled based on membership in a nutrition club, gym members who frequent the gym a certain # of times a week minimum, etc. Identifying such a group takes more groundwork, but is very realistic to deploy for such a small study (and has been done successfully in other studies), where the similarity of controls for all but the independent variable of interest ( CR) is critical for drawing reliable conclusions CR is the intervention responsible for the effect.

In essence, what this study DID show us that CR lifestyle+optimal diet / lifestyle ( plus all the other healthy practices in this population) collectively slows SC cell development in the colon relative to a standard American diet in a sedentary population ( plus all the comparatively less healthy practices associated with a control population choosing such practices).

I appreciate that assessment is by its nature in the eye of the beholder; it’s just that in my mind this is not quite the same thing as "pretty good evidence that human CR suppresses SC accumulation rather strongly, consistent with CR's tranlatability. [sic] "

With better controls we could hopefully have more confidence that it was CR that was responsible for the observed differences between the groups rather than the impact of the rather disparate non-CR health-related practices that was truly responsible for the SC differences observed - would you agree this is a fair assessment ?

Edited by Mechanism, 09 April 2018 - 10:34 AM.






Also tagged with one or more of these keywords: senescent cells, human cr