Does Caloric Restriction Induce Hormesis?

Harold Boxenbaum, Ph.D.

Arishel Inc., 14621 Settlers Landing Way, North Potomac, MD 20878-4305

Tel: 301-424-2806, Fax: 301-424-8563


The contribution by Turturro et al provides useful and cogent insights into the phenomenon termed hormesis. Conceptualizing hormesis from two posited definitions (approaches), the authors' arguments are compelling and persuasive.

Hormetic agents are virtually anything ­ chemicals, stress, food, diet, radiation, etc. The older definition of hormesis is growth stimulation at low exposures to an agent and growth retardation at higher exposures. This phenomenon has commonly been referred to as "growth hormesis." The herbicide phosfon, for example, stimulates peppermint plant growth at low soil concentrations and inhibits it at normal higher concentrations (Calabrese and Howe, 1976). The second definition of hormesis used here is a non-monotonic dose-response function following exposure to an agent. A good example is exposure to high and low doses of ionizing radiation in mice; at low exposures, longevity is enhanced, but is reduced at high exposures. A plot of ratios of mean survival time (control mice/exposed mice) vs. chronic g-radiation dose-rate is U- or J-shaped, equal to unity at no exposure, less than unity at low doses, but exceeding unity at higher doses (Boxenbaum et al, 1988). The reason is g-radiation enhances longevity through one mechanism, but reduces it via another. The net result is dose-dependent and depends on which of the two actions dominates. Using definition 2 in the context of longevity, selenium, magnesium, boron, chromium and iron are all hormetic or co-hormetic.

In the approach of Turturro et al, enhanced rodent longevity observed with caloric restriction in rats is viewed as hormetic (2nd definition), whereas Neafsey et al (1989) view it as non-hormetic, viz., a simple change (slowing) in the aging rate constant. Within the context of each groups' definitions and/or criteria, both are correct. But this is an attribute of perception and ascribed meaning, not objective, well-defined and specific criteria. And herein lies the problem. There is simply no generally accepted agreement of what constitutes hormesis. Turturro et al (1st definition) view hormesis as the existence of dose-dependent proliferation vs. proliferation inhibition. But what is proliferation. Caloric restriction proliferates molecular elements (growth/hormesis), promoting apoptosis (killing/anti-hormetic), but inhibiting oxidative cell damage through reduction of free radical formation (retardation, hormetic and/or anti-hormetic). Where do the respective processes fit into the hormetic scheme of things, and doesn't the overall result depend on descriptive criteria for "proliferation?" Are we proliferating enzymes ­ enzyme inhibitors, cells ­ cell toxins, health ­ disease, weight gain ­ weight-loss, mortality ­ longevity, function ­ dysfunction, apoptosis ­ reproduction, etc? Is a cancer chemotherapeutic agent hormetic, non-hormetic, or anti-hormetic when it kills cancer cells vs. destroying normal cells. If an agent proliferates toxic metabolites, is this hormesis? Is hormesis the proliferation of good, bad or neutral system elements, and who decides the criteria of what is good or bad?. Turturro et al tacitly recognize this dilemma, but like most of us, fail to come to grips with the ambiguities. My own view of longevity hormesis (Boxenbaum, 1992) is analogously arbitrary and muddled.

For most or all of us, hormesis characterization is more a functional property of definition and process than of cohesiveness and consistency. The hormesis critics are justified in their harsh criticisms. Although we hormesis-believers may be mavericks at a frontier, we are also Don Quixote-ish, culpable and blameworthy. Any present hormetic designation is arguably gratuitous, predominantly or exclusively dependent on subjective perspectives. Look at some processes designated as hormetic: growth, development, hatching success, length of reproductive life, behavior parameters, enzyme induction, fecundity, cancer reduction, disease resistance, viability, respiration, radiation resistance, wound healing, resistance to infection, and longevity enhancement (Boxenbaum et al, 1988).

What are the real or imagined substrates of hormesis? ­ cells, molecules, tissues, organs, intact organisms, actions, behaviors, etc. And does hormesis or lack of it exist along a time line (e.g., age-specific mortality), and/or is it a time-dependent hybridized outcome (e.g., life span)? Given present thinking, virtually any inhomogeneous phenomenon or system may be characterized as possessing hormetic elements. Hormesis is a hodgepodge of biological phenomena/events, lacking focus, uniformity and objective significance. Unless and until hormesis is better defined, it will continue to be ridiculed and ignored. Confidence in any scientific principle is roughly proportional to its conceptual precision. Viewed thusly, hormesis is promiscuous, credulous and inane. Our challenge is to survey, condense, simplify, infer, define, dissect, categorize, focus, restrict, etc. ­ to order the relevant information from the sea of data in which we are adrift. But does anyone think this can be done without knowledge of mechanism? We don't even concur if hormesis is an event, an outcome, both or neither.


Boxenbaum H. Hypotheses on mammalian aging, toxicity, and longevity hormesis: Explication by a generalized Gompertz function. In: "Biological Effects of Low Level Exposures to Chemicals and Radiation," E Calabrese (ed), Lewis Publishers, Boca Raton, 1992, Chapter 1, pp. 1-39.

Boxenbaum H, Neafsey PJ, Fournier DJ. Hormesis, Gompertz functions, and risk assessment. Drug Met Rev 19:195-229(1988).

Calabrese EJ & Howe KJ. Stimulation of growth of peppermint (Mentha piperita) by phosfon, a growth retardant. Physiol Plant 37:163-165(1976).

Neafsey PJ, Boxenbaum H, Ciraulo DA, Fournier DJ. A Gompertz age-specific mortality rate model of aging: Modification by dietary restriction in rats. Drug Met Rev 21:351-365(1989).