2(a) - The biology of the animal model must be as comparable as possible to human biology. Although cell cultures or invertebrates such as Drosophila or C. elegans may be useful for gerontological studies and may be informative about what life-extension experiments might be attempted, they are inappropriate for a definitive life-extension study, because their biologies are not sufficiently homologous to that of humans. A mammal must be used.

2(b) - The appropriate mammal should be practical in terms of husbandry, costs, and length of life. For example, although monkeys are the animal model which is most similar to human biology, they are difficult to acquire, and entail a great deal of expertise and money to maintain. They live some 40 years, and thus, it would take decades to obtain solid information. Further, they are phenotypically heterogeneous and not inbred, which makes it impossible to have comparable cohorts in different experiments. Consequently, except under very exceptional circumstances and for modalities that are already well developed, monkeys would not be feasible. The mammal should not hibernate nor have other metabolic characteristics which are different from humans and which might be influenced inadvertently by therapeutic agents and thereby cause a life-extension effect. Rodents (either mice or rats) are the most appropriate model.

2(c) - The life-span trajectory of growth, maturation, and senescence must be analogous to that of humans. Many rats grow throughout most of their life-span; and consequently, those which do would be inappropriate because any agent which retarded growth would likely have a life-extension effect. Indeed, that is a common artifact in many life-extension experiments. The mouse does not have this characteristic; and that is a more appropriate model.

2(d) - For homogeneity and consistency, the strain of mouse must be a well studied inbred strain or a well studied hybrid, first generation (F1) cross between two well studied inbred strains. This insures experimental consistency among different experiments, by different investigators, asynchronously.

2(e) - The strain must be genetically well characterized, so that it is positioned for more fundamental research into molecular mechanisms of disease and biological vitality.

2(f) - The strain must be a long lived strain, so that it ages.

2(g) - Optimal, ad libitum, survival curves must be well established by multiple investigators.

2(h) - Optimal, calorically restricted, survival curves should be well established by multiple investigators.

2(i) - The causes of mortality and pathologies of the strain should be well known and as analogous as possible to the pathologies of humans, and multiple, so that no one pathology is the dominant cause of death. The following table contains the major causes of morbidity and mortality in humans and can be used as a schematic reference.

2(j) - The particular strain, must be readily available from a certified breeding colony that has accreditation from an organization such as the "Association for Assessment and Accreditation of Laboratory Animal Care." {http://www.aaalac.org}

2(k) - In evidence that the strain ages, there must be a decline in lean body mass in the last quartile of the life-span. Also, to insure homogeneity of the experimental cohort, there should not be a marked difference in body mass within the population.

2(l) - The life-span trajectory should have three stages (growth, maturation, and senescing) and be proportionately comparable to that of humans. See the table below.

2(m) - The following table of criteria serves as a check list of the major relevant factors, as defined above, for selecting a mouse strain for a life-extension study. The answer to each should be "yes".

3(a) - It is the prevailing opinion that no single strain of mouse sufficiently fits all of the above criteria in such a manner as to constitute one definitive model for life-extension experiments. This is the opinion of the investigators surveyed here and is the published opinion for gerontological studies [3a]. Consequently, depending on the resources and interests of a particular investigator, either a sequential or parallel research strategy is required in which different strains of long-lived inbred or their F1 hybrids are used. In terms of such a sequential strategy, an appropriate animal model must be selected for the initial experiment. Depending on the preliminary experimental results, the protocol can be subsequently expanded to other strains. Perhaps, at about the age of 100 weeks, an analysis of the Gompertzian characteristics of the survival curve might indicate that a life-extension effect is being achieved; and therefore, the initiation of parallel experiments is justified by the original investigator or by collaborators. If an investigator is well funded and enthusiastic about a particular line of investigation, an experiment can be initiated in multiple strains simultaneously. Finally, where positive results are gained or discernible within the course of an experiment, an experiment might be conducted with the caloric restriction paradigm; however, there are some important qualification that must be considered in the usage of that paradigm; and they are discussed briefly, below, at 3e.

[3a] Hazzard DG, Bronson RT, McClearn GE, Strong R. Selection of an appropriate animal model to study aging processes with special emphasis on the use of rat strains. J Gerontol 1992 May;47(3):B63-4. MEDLINE: 1573179.

3(b) - In addition to the above criteria for the animal model, an experiment must be initiated at an age after which growth has ceased, so that any life-extension due to growth retardation is not a potential artifact.

3(c) - The number of animals to be used for statistical significance should be no less than 50 and need not exceed 100 per cohort. A good protocol would be 56 males and 56 females. This would mean: a minimum of 200 experimental animals (50 male Experimental Controls, 50 female Experimental Controls, 50 male Experimental Treated, 50 female Experimental Treated)

3(d) - Standard scientific criteria of reproducibility by independent investigators is an obvious component of this or any epistemology.

3(e) - The two experimental paradigms are: 1) ad libitum feeding; and 2) caloric restriction. Here, only the ad libitum paradigm is considered in detail. Although the caloric restriction paradigm is considered the "gold standard" for gerontology and for elucidating ageing parameters, it is problematic for testing life-extension agents. First, life-extension epistemology requires that the experimental animal model be similar to humans; and humans will not voluntarily conform to a caloric restriction regimen. Thus, without an appetite suppressant that is non-toxic and that induces life-extension by caloric restriction in a way that people will use, the caloric restriction paradigm does not have good correspondence to real life. In addition, in animal studies, when caloric restriction is initiated after growth and development has ceased, there is not much of a life-extension effect [3e1].

[3e1] Goodrick CL, Ingram DK, Reynolds MA, Freeman JR, Cider N. Effects of intermittent feeding upon body weight and lifespan in inbred mice: interaction of genotype and age. Mech Ageing Dev 1990 Jul;55(1):69-87. Beginning at either 1.5, 6 or 10 months of age, male mice from the A/J and C57BL/6J strains and their F1 hybrid, B6AF1/J were fed a diet (4.2 kcal/g) either ad libitum every day or in a restricted fashion by ad libitum feeding every other day. Relative to estimates for ad libitum controls, the body weights of the intermittently-fed restricted C57BL/6J and hybrid mice were reduced and median and maximum life span were incremented when the every-other-day regimen was initiated at 1.5 or 6 months of age. When every-other-day feeding was introduced at 10 months of age, again both these genotypes lost body weight relative to controls; however, median life span was not significantly affected although maximum life span was increased. Among A/J mice, intermittent feeding did not reduce body weight relative to ad libitum controls when introduced at 1.5 or 10 months of age; however, this treatment did increase median and maximum life span when begun at 1.5 months, while it decreased median and maximum life span when begun at 10 months. When restricted feeding was introduced to this genotype at 6 months of age, body weight reduction compared to control values was apparent at some ages, but the treatment had no significant effects on median or maximum life span. These results illustrate that the effects of particular regimens of dietary restriction on body weight and life span are greatly dependent upon the genotype and age of initiation. Moreover, when examining the relationship of body weight to life span both between and within the various groups, it was clear that the complexity of this relationship made it difficult to predict that lower body weight would induce life span increment. MEDLINE: 2402168

Consequently, if the experiment must be initiated after development (as it must to avoid the artifact of the retardation of growth, as stipulated in 1(f) above), then caloric restriction would have to be initiated after the cessation of growth, which would have only a minimal life-extension effect. Therefore, in terms of life-extension science in animal models, it would seem that the caloric restriction paradigm would have little, if any, greater scientific merit than the ad libitum paradigm, which is much better studied and easier to perform.

In general, a fairly wide range of husbandry conditions yield similar life-span results. Different frequency of cage changes (7, 14, and 21 days) and different cage ventilation rates (30, 60 and 100 air changes per hour) have not shown any differences in health status [4-1].

[4-1] Reeb-Whitaker CK, Paigen B, Beamer WG, Bronson RT, Churchill GA, Schweitzer IB, Myers DD. The impact of reduced frequency of cage changes on the health of mice housed in ventilated cages. Lab Anim 2001 Jan;35(1):58-73. MEDLINE ID: 11201289.

And fairly dirty, unsanitary conditions had no significant effect on the life-span - (4-2)

[4-2] Chino F, Makinodan T, Lever WE, Peterson WJ. The immune systems of mice reared in clean and in dirty conventional laboratory farms. I. Life expectancy and pathology of mice with long life-spans. J Gerontol 1971 Oct;26(4):497-507. PubMedID: 4329048.

Finally, animals raised in completely germ free conditions do not show a significant improvement in survival curves. Early reports did show an improved life-span from germ free husbandry [4-3], [4-4], [4-5], [4-6].

[4-3] Gordon HA, Bruckner-Kardoss E, Wostmann BS. Aging in germ-free mice: life tables and lesions observed at natural death. J Gerontol 1966 Jul;21(3):380-7. MEDLINE 5944800.

[4-4] Pollard M. Senescence in germfree rats. Gerontologia 1971;17(5):333-8. MEDLINE 4948858.

[4-5] Pollard M. Spontaneous prostate adenocarcinomas in aged germfree Wistar rats. J Natl Cancer Inst 1973 Oct;51(4):1235-41. MEDLINE 4745857.

However, those findings were contravened by later studies showing that germ-free animals had a lower caloric intake with that being the likely cause of extended life-span. "The reduced early food intake and smaller body weight of adult (germ-free) GF rats may be the reason ad libitum fed GF rats live slightly longer than their (conventional) CV counterparts, but GF life was without additional effect on life span when food intake was restricted." [4-6]. This is a prime example of the artifact of caloric restriction even from an unusual source such as germ free environment.

[4-6] Snyder DL, Pollard M, Wostmann BS, Luckert P. Life span, morphology, and pathology of diet-restricted germ-free and conventional Lobund-Wistar rats. J Gerontol 1990 Mar;45(2):B52-8. MEDLINE 2313040

yes no no-response.

Comments:

4(a) - Animals should be housed 1 per cage.

This will avoid various confounding factors, such as: territorial fights that cause lesions and, sometimes, death; and over-grooming dominance which can cause lesions. Infection transmission is mitigated. Animal identification is certain and makes unnecessary tatooing (which can scarified or be nibbled off) or ear notching (which can be obliterated by fighting when animals are housed together) or electronic identification implants (which are expensive). Finally, in cages with multiple animals and as each animal dies, the space per animal changes and that can cause a stress variable that may impact on the relatively mortality rates within the cohort. In single caging, minimum cage floor size is 180 cm2 with cage height of 12 cm.

4(b) - Water should be changed weekly.

4(c) - Change bedding at least every 14 days and, weekly if housing two to four mice per cage.

4(d) - Temperature. If housed singly, 25 degrees Centigrade or 77 degrees Fahrenheit. If housed multiply, 20-24 degrees Centigrade or 68-75 Fahrenheit.

4(e) - Humidity - 50-60%

4(f) - Ventilation or air changes - 15-20 per hour.

4(g) - Light/dark cycle 12/12 hours.

4(h) - Protect personnel from allergens and animals from infection by using masks and gloves.

4(i) - Clean and sterilize cages, bottles, drinking tubes, and feeders. Autoclaving is preferrable; however, this can be done without elaborate equipment by having a dual set of cages, bottles, tubes, and feeders, which are cleaned with soap and left to soak in highly chlorinated solution between cage changes.

4(j) - If lack of capital and smaller facilities require a greater density than 1 animal per cage, then house males and females separately and no more than four per cage. Change bedding, cages, and accouterments weekly.

4(k) - Watch for leaking water tubes which can be caused either by a mis-fit with the ball baring insert or by animals stuffing bedding in the tube.

4(l) - As a general reference, use the guidelines in Principles of Laboratory Animal Science. Van Zutphen LF, Baumans V, & Beynen AC [Eds.] Elsevier 2001.

5(a) - Report the survival curves as represented in section 2, above, plotting Controls and Treated against a range of optimal Reference Controls. Also, report body weights and food consumption.

5(b) - Chronology. Unless otherwise necessary, use weeks for the chronology of percent surviving, the recording of body weights, food consumption, pathologies, and incidence of other events.

5(c) - Units of weight. Use grams for units of weight; and, if recording energy, use calories rather than joules.

7 - BASED ON THE CRITERIA ESTABLISHED IN SECTION 2, THE C57BL/6j MOUSE IS A VALID ANIMAL MODEL

The lineage of this mouse dates as follows. From Chinese through Japanese and English "fanciers", this and almost all inbred strains of mice used in experimental science today were developed between 1903-1915 at Abbie Lathrop's mouse farm in Granby, Maryland. One was labeled "C57". Through Clarence Little, a "Black" sub-line was developed between 1921 and 1937. From that, another sub-line called "6" was developed at Jackson Laboratory in 1947 - thus becoming named as C57BL/6j (http://www.jax.org/). The strain was incorporated into the breeding colonies of various institutions. In 1951, the National Institutes of Health (HIH) bred this mouse as C57Bl/6N. It was bred by the National Institute of Aging via the NIH as C57Bl/6NNia. In 1974, NIH contracted breeding with the commercial vendor, Charles River Laboratories as C57BL/6Cr (http://www.criver.com/) and subsequently the NIH line became maintained by Simonsen Laboratories, Inc. as C57BL/6N Sim (http://www.simlab.com/). It is assumed that all lines have virtually identical survival and pathological criteria. [7-1] [7-2] [7-3] [7-4]

[7-1] Heston WE. Development of inbred strains in the mouse and their use in cancer research, p.9-13. In: Lectures on genetics, cancer, growth and social behavior. Roscoe B. Jackson Memorial Laboratory, Bar Harbor, Maine, 1949.

[7-2] Biology of the Laboratory Mouse. EL Green (Ed.) McGraw-Hill Book Co. 1966, pg.1-8.

[7-3] The Mouse in Biomedical Research. Foster HL, Small JD, Fox JG (Editors). Academic Press, 1981, pg. 7.

[7-4] Malakoff D. The Rise of the Mouse, Biomedicine's Model Mammal. Science Apr 14 2000: 248-253. MEDLINE 10777401

yes no no-response.

Comments:

7(a) - Given the criteria in Section 2(m), the C57BL/6j mouse fits the essential criteria for a life-extension study

yes no no-response. Comments:

7(b) - As stipulated in Section 2(l), the C57BL/6j mouse has a life-span trajectory sufficiently similar to humans.

7(c) - As stipulated in Section 2(i), the C57BL/6j mouse has pathologies that are analogous to those of humans. (Note. This needs a lot more work.)

(See also: Bronson RT. Rate of Occurrence of Lesions in 20 Inbred and Hybrid Genotypes of Rats and Mice Sacrificed at 6 Month Intervals During the First Years of Life. Ch. 17. In: Genetic Effects on Aging II, by David Harrison. Telford Press, 1990)

7(d) - The C57BL/6j mouse is well characterized and studied. In the U.S., 61% of the mice used for ageing studies were C57BL/6 [7d1]. See also the MEDLINE searches below.

[7d1] Sprott RL & Ramirez I. Current Inbred and Hybrid Rat and Mouse Models. ILAR Journal, 1997 vol.38 no.3, pg 104-108.

7(e) - Male & Female Reference Survival Curves, Body weights, and Food Consumption.

Between 1948 to 1956, the median life-span of this strain was 522 days for males and 581 days for females. After 1960, it increased to 878 days for males and 794 days for females. [7e1] [7e2] The major shift occurred around 1960. While a "genetic drift" may have been a factor, most-likely improved husbandry and diet more suitable to the genotype (possibly a higher fat content) was the major reason.

[7e1] Russel ES. Lifespan and Aging Patterns. pg. 511-519. In: Biology of the Laboratory Mouse. Green EL (Ed.), McGraw-Hill Book Company. 1966.

[7e2] Kunstyr I, Leuenberger HG. Gerontological data of C57BL/6J mice. I. Sex differences in survival curves. J Gerontol 1975 Mar;30(2):157-62. A group of 1,052 C57BL/6J mice (296 males and 756 females) was kept under well-defined, clean laboratory conditions from the age of 6 weeks until natural death. The survival curves of males and females (computer-produced 3, 4, and 5 parameter curves, Gompertz curve histogram) were established and shown to follow a logistic function. The average life-span amounted to 878 plus or minus 10 days for males and 794 plus or minus 6 days for females. These values distinctly exceed comparable values given in the literature. They are attributed to favorable conditions of animal care and to supposed alterations in genetic background. A genetic drift in sex-dependent median survival time occurred in the genetically unstable C57BL/6J strain between 1966 and 1970. Before this drift, the males died sooner; after it, they lived longer. MEDLINE: 1123533

 

Note: the charts which follow have been drawn from the studies which are referenced.

A similar profile of essential criteria needs to be done for all commonly used inbred and F1 hybrid long-lived strains of mice and also for F344 and other rats that are used in life-extension studies.

9(a) - The term "epistemology", as used here, means a set of definitions and criteria which circumscribe valid knowledge within a particular science. ("Epistemic" would perhaps be a better term but it is less familiar.) The consideration here is with life-extension science as it pertains to animal research. The central issue can be stated as follows: if a modality is tested in experimental animals for its effects on life-extension and control of ageing, then what criteria make such an experiment valid? Again, the focus here is only on the formulation of an epistemology for life-extension experiments in an animal model. There is a different epistemology for life-extension experiments in humans; and that is only briefly explained below.

9(b) - In general terms, "life-extension science" is engaged in the development of modalities which extend the human life-span. In more objective terms, life-extension science is about the enhancement of biological vitality, the effect of which would be to extend life-expectancy. Refining this further, although life-extension science may include modalities which prevent, manage, or cure disease or support health and thereby increase average life-expectancy, the primary target of life-extension science is that cluster of degenerative processes which come under the category of biological ageing, the successful treatment of which would enable survival beyond the natural, maximum life-span. The definition of life-extension science can entail many nuances because it is not a thing or a condition. Rather it is a potential that is inherent in a condition. It is stochastic and based on objective conjecture - i.e., given certain intrinsic conditions and baring certain extrinsic factors, life-expectancy will be prolonged.

9(c) - The prevailing public health and medical paradigms focus on the common causes of death (i.e., heart attack, stroke, cancer, diabetes, infections, accidents, etc.). However, even if measures were invented that either cured or prevented all such causes of mortality, that would only increase the median life-expectancy by about 12 years and would do nothing to increase the maximum life-span, which would remain unchanged at about 100. With such "perfect" medicine, virtually everyone would live out to the natural, maximum potential of human genetics. But, ironically, that would result in both a personal and social catastrophe because, then, everyone would survive into advanced senility and end-up spending a decade or more in the geriatric ward. Indeed, that is the direction things are presently headed. However, if ageing were cured, then health would remain optimal; the probability of disease would be minimal; if disease did occur, then treatment would be effective and recovery, rapid; and the potential, maximum life-span would be open-ended. Thus, in every respect (i.e., life-extension, disease prevention, and clinical medicine), the control of ageing is the conditio sine qua non, upon which all things depend and without which only marginal benefits can be derived. That is a dramatic, but simple fact, which goes completely ignored by the general consciousness and professional attention. Because the existing paradigms are inadequate, life-extension science via the control of ageing is, a fortiori, the future of health and medicine.

9(d) - "Gerontology" is about the study of the ageing processes; and "life-extension science" is about those modalities which increase life-expectancy. Gerontology is mostly analytic and descriptive; and life-extension science is mostly prescriptive and practical. The two disciplines are distinct but not separate; and they inform each other. Gerontological studies of the ageing process suggest therapeutic approaches to be tested in life-extension studies; and the results of such life-extension studies inform gerontology about mechanisms which may or may not warrant aggressive research. Indeed, empirical life-extension would be the ultimate proof-of-principle for the validity of gerontological theories; and this is evidenced by the fact that it is the technique of caloric restriction, which was discovered empirically to extend maximum life-span in animals, has set the dominant paradigm for modern gerontological studies. The mechanisms by which caloric restriction works are still not understood even though the life-extension effects are certain.

9(e) - Like any medical therapy, a life-extension modality will ultimately count in and be validated by its beneficial effects in a particular, individual, human being. However, the role of the animal experiment is much more central in life-extension science than in conventional medical science. Usually, in medical therapies, human pathology suggests a therapeutic approach, and animal models are used in phase I studies to work out surgical techniques or elucidate the pharmacological kinetics and to evaluate adverse effects before going into application on small human cohorts where the proof-of-principal is mostly studied. From there, an effective modality (measured in statistical significance) is advanced to trials on larger cohorts of human and finally into application on individual humans in clinical practice where it either works or does not work. In contrast, in the invention of life-extension therapies, mechanisms of degeneration in gerontological studies in animals are more likely to suggest degenerative processes in humans; and proof-of-principle for the therapeutic modalities will be demonstrated first in animals. Further, if life-extension were to be obtained in the animal experiments, it is likely that the intermediate human cohort studies would be by-passed, and administration would begin in individuals, validating efficacy in the process of clinical application. This is justified on the basis of both practicality and safety. In terms of practicality, intermediate human cohort studies would be completely unfeasible because the amount of time which it would take to test a life-extension agent in a human cohort would be decades and there would be a requirement of thousands of participants in order to demonstrate statistical significance. In terms of ethics, something which caused life-extension in animals is almost certainly non-toxic and without significant adverse effects. Several examples elucidate this point. Caloric restriction which clearly causes life-extension in animal models is being recommended for individual human application without doing first any clinical trails in humans. The drug, selegiline, which is licensed for certain medical usage, was investigated for its life-extension effects in animals. Life-extension was reported as well as enhanced reproductive activity, as a bio-marker; and based on that physicians began prescribing it and individuals began taking it for life-extension and vitality. Unfortunately, the original design of the animal experiments for selegiline were flawed (something which would have been avoided if the epistemology herein had been employed); and other investigators had to redo the animal studies, demonstrating the drug's lack of efficacy. Another case would be human growth hormone (HGH), which, in a small study in humans over short period of time, was reported to improve certain parameters of ageing (i.e., bone density, lean body mass, and skin thickness); and from that little evidence, physicians have begun prescribing and individuals taking HGH for an anti-ageing effect. Human growth hormone has not been subjected to proper a life-span study in animals, so it is unclear what, if any, effect it might have. Proper animal studies would have saved a lot of time, effort, money, and credibility.

9(f) - If we take McCay's animal experiment in 1935 as the beginning of scientific life-extension, then this discipline has a history of some 70 years. During this time, proper methods have been developed to yield valid experiments and various artifacts have been well identified. These artifacts include the following: a) an improper animal model is used; b) the survival curve of the control animals is sub-optimal due to deficient husbandry such that, if a therapeutic agent mitigates some of those deficiencies, then it appears that life-extension is achieved, but in fact the survival curve has just been normalized; and c) food consumption and body weights of the animals are not measured, such that treated animals may be calorically restricted, with that being the cause of the life-extension rather than the experimental agent. These experimental conditions have been identified for decades by many investigators; however, they have been ignored in most of the animals studies in the literature (except those involving caloric restriction) and continue to be disregarded even in the design of contemporary studies, with the result being that the literature is and will continue to be replete with mis-information.

9(g) - There is sufficient knowledge about what constitutes a valid life-extension experiment in animals such that a solid (albeit provisional) epistemology can be defined, the framework for which is the purpose of this present effort. It needs to be stated that this epistemology will evolve as the science improves and becomes more certain. A time can be foreseen when the bio-markers of vitality and ageing are so well know in humans and bio-informatic databases for simulation so well developed for toxicology and mechanisms of action that animal experiments may become sub-ordinate or irrelevant; but that will be some time in the more distant future. By adopting and promulgating a basic epistemology for life-extension science, the design of initial research by investigators and the evaluation of that research can be facilitated in both life-extension science and in gerontology. Further, with the completion and success of the Human Genome Project, the initiation of the next big program in molecular biology (Proteomics), and the continued surge in bio-technology enterprise, virtually every scientific discipline is getting into biology. And because, as previously noted, the disease-cure approach to medicine is, in effect, a cul-de-sac, there will be nowhere to go other than life-extension and control of ageing. Many of these new investigators will not be familiar with the well established artifacts in this science; and hopefully by using the guideline of a proper epistemology for animal experiments this "land-rush" into the field can be channeled into productive areas rather than further complicate the subject by more mis-information.

The epistemology for life-extension experiments in animals is different from the epistemology for life-extension experiments in humans. Human life-extension science would entail: 1) the determination of individual, human, biological parameters which functionally correspond to vitality, longevity, and disease and ageing; 2) the ability to test those parameters in a clinical environment in a particular individual and to show modification toward optimum values over short durations of time; and 3) therapies which restored those parameters back to optimum values. Beyond this general sketch there will not be here an exposition of an epistemology for life-extension science in humans.