cultured fibroblasts of cell lines from animals of different size have similar activities for a range of different enzymes and metabolic rates that converge on the similar values despite the metabolic rates of the animals of different sizes having quite distinctly different basal metabolic rates (BMRs). In discussing this “normalization,” the data support experimentally a previous hypothesis based on circumstantial evidence. However, the experimental plan and the cell type chosen for analysis do not provide the best evidence to confirm this hypothesis.
In mammals (homeotherms), it is thought that a general allometric relationship exists between metabolic rate and body mass. While this is not usually disputed as a rule of thumb, some groups have argued persistently that the relationship is more precise than others would maintain. The formula proposed is that BMR is proportional to M3/4 (14). It would be surprising to find these formulae applied across the board in mammals, and equally surprising of West and Brown (13) to consider that many deviations would not undermine this relationship. Where exceptions are numerous, any hypothesis is inevitably weakened. Unless one can explain exceptions to the alleged “norm,” the general relationship cannot be regarded as a “mathematical law” of which, incidentally, there are none in biology. The problem of the range within any species also needs to be mentioned (8).
Proposed Basis of the Relationship Between Body Mass and Metabolic Rate
Small mammals have high mass-specific BMRs, whereas large mammals have much lower BMRs. Although measurement of BMR is not straightforward, this general rule has been known for decades. Small mammals lose heat rapidly because their surface area-to-body mass ratio is high, whereas it decreases in large animals as mass increases. This gives a 2/3 rather than a 3/4 power law, and does not take into account that some animals might be better insulated than others (e.g., the arctic fox). Thus, the BMR of a man will vary depending on whether he is clothed or naked. Since we are talking about whole organisms (typically mammals), it is reasonable to assume that the heat produced by metabolic activity will be the normalized sum of that produced by the cells from every organ and tissue of the body (10, 12), e.g., liver cells have much higher metabolic throughputs than connective tissue. West et al. (14), however, attribute the allegedly universal 3/4 power scaling not to the different contributions of different organs nor to the surface-to-volume ratios in animals of different body mass, but to the fractal geometry of resource supply networks, which they consider to be a biologically universal.
What Sets the Metabolic Rates in Different Tissue and Different Species?
In this issue of American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, Brown et al. (2) proclaim in the title that “Metabolic rate does not scale with body mass in cultured mammalian cells,” which will in literal terms not be self-evident to general readers. But the underlying question in it is: What sets the metabolic rates in different tissues and animals? The premise underlying much of the discussion of this problem is that some intrinsic regulatory mechanism is involved. Any such regulatory mechanism would have to operate down to the cellular and molecular levels; but what sort of control could possibly regulate general metabolic activity? No one has a compelling answer (1).
Role of the Mitochondrion
The organelle that clearly demands attention is the mitochondrion. In general, heat production by a homeotherm maintains the body temperature at ∼37°C, although a lot may be insensibly lost. When extra heat is required, mitochondria can become increasingly uncoupled. Porter and Brand (11) showed that freshly-isolated mitochondria from horse liver metabolize 10 times slower than from mice liver, and on three accounts they explained why this was so. Their findings corroborate the work of Krebs (9), who found significant differences in the metabolic activity of freshly isolated liver slices from animals of considerably different masses, supporting the notion of intrinsic differences among cells (tissues) from different species. Different enzyme affinities, different membrane transporter densities, levels of cofactors, etc., will all contribute to these “intrinsic differences.”
Metabolic Activity in Cultured Cells
Returning to cells placed in culture, one might expect that these differences would level out in time in a standardized medium, and therefore cells from different species would converge toward similar metabolic activities. This argument was promulgated by Wheatley and Clegg (16), backed up by evidence presented in Clegg and Wheatley (3), and supporting Coulson's (5) beautifully clear exposition on the importance of delivery rates of metabolites in animals of different sizes. Coulson (4) made another simple point, that once a substrate has entered the cell, the intracellular enzymes have little choice but to act on it. If we knew more about the delivery rates of substrates to cells in situ in different-sized animals, we may well find a scaling formula that is much more relevant and reasonable. In principle, this affords some justification for the “resource supply network” approach taken by West et al. (15).
An Experimental Attempt to Prove the Hypothesis
In Brown et al. (2), the crucial issue was to carry out experiments that establish whether or not a scaling factor is retained in recently isolated homologous cells (cell “lines”) from small and large animals. Unfortunately, Brown's group did not use primary cultures. Even if they had, it would have taken quite a number of generations to obtain sufficient cells for their analyses. The choice of fibroblasts was also unfortunate because dermal fibroblasts in vivo are quite inactive, with very low metabolic activity. Their activity will rise in culture, so cells from different species might all have risen to similar levels simply by placing them in the same rich medium (as expected on Coulson's principle). Relevant details are also missing on the American Type Culture Collection (Manassas, VA)-acquired “cell lines”, and we are not told much about how long the cultures were incubated before being analyzed. However, the fact that the enzymes chosen for study all had similar activities suggests the normalization of activity would not be unexpected.
In conclusion, the outcome of the study is nevertheless incontrovertible, that any scaling that might have existed in vivo from different species is lost when the cells are cultured, as already predicted (2, 16). Examination of metabolic changes during culturing of cells that become normalized is now a crucial requirement, and the obverse experiment (setting out cells of high activity in vivo in culture) would be more relevant.
An Alternative Hypothesis
Whereas the West et al. (14) model predicts that (metabolically active) cells will lose their in situ scaling characteristics when they are grown in culture, the quantum metabolism theory of Demetrius (6, 7) predicts the opposite. Quantum metabolism attempts to infer whole organism metabolic rates from the metabolic activities of component cells. Based on chemiosmotic energy transduction, a concept of energy storage in membrane biomolecules and the perception that these energy storage levels are in principle quantized, the argument for the theory is developed using the same mathematical formalism as the quantum theory of solids. This theory makes a number of predictions that are critically testable by experiment, of which the retention of scaling by appropriately cultured cells is one. It is therefore particularly unfortunate that the experiments of Brown et al. (2) were conducted in a manner that failed to provide either compelling corroboration or refutation of either the West et al. or the Demetrius model. The fact that (at least) two alternative accounts of allometric scaling give opposite predictions in respect of such experiments makes the need for an examination of cultured cells from metabolically active tissues even more crucial.
I thank my colleague Dr. Paul Agutter for his comments on this review.
- Copyright © 2007 the American Physiological Society