I think there's a very strong case to be made that almost all humans currently practicing caloric restriction are doing so in a way that fundamentally and significantly differs from how caloric restriction is implemented in and affects rodents. Here is my argument:
The overwhelming majority of caloric restriction(CR) experiments in rodents, whether studying lifespan, longevity or health effects, involve rodents on CR being served a single ration of food per day(with a few exceptions, which are addressed in detail below.)
This delivery of a single ration of food per day is usually used in CR'ed animals simply because it is the easiest way to implement a calorie-restricted diet. All the food for the day is measured out and then delivered to the feeding area of the animal all at once, which is much easier than delivering multiple precisely measured batches of food throughout the day.
In comparison, the "control" ad-libitum(free-feeding) rodents in these experiments usually have a constant supply of food available to them.
It is sparsely mentioned or dealt with in the CR literature, but this difference in feeding regimes is known to result in dramatic differences in feeding behaviours between CR'ed and ad-libitum fed rodents.
Ad-libitum rodents engage in a natural "nibbling" feeding pattern, eating many meals throughout the day. Ad-libitum fed rats rarely go for longer than 3 or 4 hours at at time without feeding(1) and ad-libitum mice rarely go for longer than 1 hour without feeding(11.)
On the other hand, rodents subjected to CR gorge on their single food ration and finish it all in a very short period of time(2). This feeding pattern, unlike in the ad-libitum animals, thus results in long periods of complete food deprivation(fasting) in the CR'ed rodents until the next day's meal is supplied to them.
It has been shown that 50% of rats on a conventional CR feeding regime finish their entire daily food ration within 3.5 hours, and 90% of rats finish it all in under 5 hours(2). This means that 50% of CR'ed rats are going for more than 20.5 hours per day without any food at all, and 90% of the CR'ed rats are going for more than 19 hours without any food per day.
CR'ed mice eat their entire food ration within just an hour(11) and are thus subjected to a full 23 hours without food per day with conventional CR feeding regimes.
This major difference in fasting duration between ad-libitum and CR rodents results in a raft of very significant physiological differences(3.) When subjected to only 16 hours without food(considerably less than that experienced by the average CR'ed rodent), rats lose an average of 6.9% body weight, their serum glucose concentrations drop by an average of 34.9%, triglycerides drop by 48.4% and cholesterol drops by 21.7%(3). CR'ed rats deplete their glycogen stores during the period of fasting between meals and transition into a daily state of ketosis, as well as experiencing hypothermia during this period(11). Mice rapidly switch to fatty acid oxidation(and ketosis) for fuel after just 6 hours without food(12).
The time scale for many aspects of rodent metabolism is radically different to that of humans, so what seems like a short period of time for a human may actually represent a very long period by rodent standards.
How might this translate and scale into human terms? Here are some comparisons: when subjected to complete food deprivation(fasting) it takes humans about 10 days to lose 10% body weigh(6), it takes rats about 24 hours to lose 10% body weight(3) and it takes mice less than 24 hours to lose 10% body weight(7).
Non-obese humans can survive about 60 days of complete starvation(8), rats can survive about 10 days of starvation(9), mice can only survive about 6 days of starvation(10).
By these measures, which arguably seem very relevant to the matters being discussed, there is from a 6:1 to 10:1 ratio between humans and rodents in terms of speed and duration of these particular metabolic processes.
Taking into consideration these ratios, along with the known feeding patterns of rodents on CR, it could be suggested that the approximately 20 hours of fasting that regular CR'ed rodents experience on a daily basis, is equivalent to 120 to 200 hours of fasting for a human(5 to 8 days.) The approximately 4 hours of food gorging in CR'ed rodents might be the equivalent of a human eating about a normal week's worth of food within a period of one to two days. For a human male of 70kg, the daily weight-cycling of about 7% that rodents experience on CR would be equivalent to continuously rapidly losing and then gaining 5kg of body weight.
Could these long periods of fasting, short periods of food gorging and the associated major impacts on physiology be the source of some or all of the beneficial effects seen in CR rodent experiments?
So far, there have been a very small number of studies that have attempted to answer these questions. These studies reported that the life-extending properties of CR still occur when more frequent feeding regimes are used in CR'ed rodents. But I think there are some real problems with these studies.
The two studies(that I am aware of) that increased feeding frequency and reduced fasting duration the most, still left rats with a daily 16 hour gap between food rations(4), and left mice for a 12 hour daily gap between food rations(5). As noted above, ad libitum rats normally experience a maximum of just 3 to 4 hours between meals, and ad libitum mice just 1 hour gap.
Were these studies sufficient to rule out the effects of long periods of fasting? As shown above, even just 16 hours of food deprivation in rats results in very significant physiological effects. Mice likely experience even quicker alteration in physiology due to food deprivation, such that 12 hours without food may still be a significant period for them to go without food. These rodents were quite likely still undergoing ketosis, for instance.
To give a comparison: using the best-case scenario, based on the feeding regimes used in the small number of studies that attempted to limit the periods between feedings, and the human:rodent metabolic ratios described above, a human-scaled equivalent feeding regime would still lead to somewhere between about 72 and 160 hours between meals. That's a minimum of three days without eating in human terms, and that's given the best case scenario currently available from studies into rodents.
So where does this leave human caloric restriction? Should humans be engaging in multi-day fasts, interspersed with short periods of food gorging, if they want to mimic the rodent experiments? Others have suggested that a regime like this may in fact be necessary to reproduce the effects seen in rodents(13). Several tantalizing hypotheses and pieces of evidence lend support for this view. Some of the most popular and promising theories explaining the effects of CR, including modulation of autophagy, apoptosis, lipolysis and protein turnover, seem likely to require longer-term fasting in humans to be effectively activated(14,15.) Changes in gene expression have also shown that fasting seems to regulate the genes that respond to CR(16.)
In short, I believe that there is a very strong argument to be made, that the conventional caloric restriction regimes that are currently used by most human CR practitioners(ie eating at least one or more meals every day or two), are not suitably replicating the regimes or many of the major physiological effects of CR seen and used in any CR rodent studies to date.
(1) "Meal patterns in the genetically obese Zucker rat: a reexamination"
Castonguay TW, Upton DE, Leung PM, Stern JS. PMID: 7100292
(2) "Methods for Inducing and Monitoring Liver Autophagy Relative to Aging and Antiaging Caloric Restriction in Rats"
Donati A, Cavallini G, Bergamini E. PMID: 19200897
(3) "Effect of fasting duration on clinical pathology results in Wistar rats"
Kale VP, Joshi GS, Gohil PB, Jain MR. PMID: 19351329
(4) "Temporal Pattern of Food Intake Not a Factor in the Retardation of Aging Processes by Dietary Restriction"
Masoro EJ, Shimokawa I, Higami Y, McMahan CA, Yu BP. PMID: 7814779
(5) "Meal-timing, circadian rhythms and life span of mice"
Nelson W, Halberg F. PMID: 3794831
(6) "Metabolic Aspects of Acute Starvation in Normal Humans (10 Days)"
Consolazio CF, Matoush LO, Johnson HL, Nelson RA, Krzywicki HJ - PMID: 6036255
(7) "Acute Starvation Protects Mice Against Listeria monocytogenes"
Wing EJ, Young JB. - PMID: 6772566
(8) "Fasting - the ultimate diet?"
Johnstone AM. PMID: 17444963
(9) "Factors influencing survival of rats in fasting; metabolic rate and body weight loss"
RIXON RH, STEVENSON JA. PMID: 13411211
(10) "Response of germfree, conventional, conventionalized and E. coli monocontaminated mice to starvation"
Tennant B, Malm OJ, Horowitz RE, Levenson SM. PMID: 4866341
(11) "Effect of chronic caloric restriction on physiological variables related to energy metabolism in the male Fischer 344 rat"
Duffy PH, Feuers RJ, Leakey JA, Nakamura K, Turturro A, Hart RW. PMID: 2661930
(12) "Calorie restriction increases fatty acid synthesis and whole body fat oxidation rates"
Bruss MD, Khambatta CF, Ruby MA, Aggarwal I, Hellerstein MK. PMID: 19887594
(13) "Caloric restriction in C57BL/6J mice mimics therapeutic fasting in humans"
Mahoney LB, Denny CA, Seyfried TN. PMID: 16709251
(14) "Proteolytic and lipolytic responses to starvation"
Finn PF, Dice JF. PMID: 16815497
(15) "The role of autophagy in aging: its essential part in the anti-aging mechanism of caloric restriction"
Bergamini E, Cavallini G, Donati A, Gori Z. PMID: 16709251
(16) "Starvation response in mouse liver shows strong correlation with life-span-prolonging processes"
Bauer M, Hamm AC, Bonaus M, Jacob A, Jaekel J, Schorle H, Pankratz MJ, Katzenberger JD. PMID: 14762175