Biomedical Engineering Reference
In-Depth Information
replication error during cell division, which are consequences of metabolism.
The predictions of this theory are supported by molecular divergence data from
mitochondrial and nuclear genomes (55). By accounting for the effects of body
size and temperature on metabolic rate, a single molecular clock explains het-
erogeneity in rates of nucleotide substitution across genes, taxa, and thermal
environments. This clock, however, "ticks" at a constant substitution rate per
mass-specific metabolic rate rather than per unit time. For taxonomic groups
with DNA repair rates and genome lengths that are roughly constant, this pre-
dicts that the number of substitutions per lifetime is invariant because lifetime
scales inversely with metabolic rate (Figure 2).
In collaboration with our colleagues, we are currently analyzing other cellu-
lar level properties, including scaling laws for cell size, coding and non-coding
genome lengths, and RNA abundances. Unicellular eukaryotes vary in mass and
volume by several orders of magnitude (10
-10
to 10
-6
g), and their metabolic
rates scale as cell mass to the 3/4-power (Figure 1). This raises interesting ques-
tions about how external exchange surfaces and internal structures and rate
processes within cells scale (13). Moreover, recent data for multicellular organ-
isms suggest that the mass of certain cell types scale with body size whereas
other types remain roughly constant. The scaling of cell size is therefore not just
a question of the size of an organism or of endothermy versus ectothermy, but is
also related to questions of functionality. Recent progress on the scaling of cell
size and genome length suggests that these problems may well be interrelated.
By understanding how much variation in genome length is due just to
changes in the size and temperature of the organism and identifying the residuals
of these more dominant patterns, it may be possible to provide a better definition
of biological or genomic complexity and possibly help resolve the C-value para-
dox (56-60), i.e., why is genome length seemingly uncorrelated with organismal
complexity. A few groups have begun to study connections between the theory
of biological scaling and genome length (14,15). Furthermore, it is well known
that in eukaryotes total genome length scales linearly with cellular mass. Our
colleagues and we believe that we have begun to understand this relationship,
and we have discovered similar relationships for coding genome length. Further
progress on the scaling of genome length may lead to very simple methods for
estimating the gene number of an organism based on the cellular mass, and this
could potentially facilitate gene searches.
3.3. Drug Dosing: Scaling from Rats to Humans
Tests to determine the appropriate level of drug dosage are often done on
mice or rats and then extrapolated to humans (61,62). Guiot et al. (9) have dis-
cussed the use of the growth model described in ยง3.1, as applied to tumor
growth, for determining levels of therapeutic tumor dosages. In mammals the
extrapolation of drug dosage for the treatment of a given condition is, for sim-
plicity, sometimes calculated assuming a linear dependence with body mass.
