bell-peter at yale.edu
Sun Apr 11 11:54:06 EST 1993
In article <9304110351.AA33827 at itsa.ucsf.EDU> jharper at ITSA.UCSF.EDU (M. Jane Harper) writes:
>I'm not sure this question fits in *any* BIONET section, but it came up in
>my neuro course, so here goes ...
>Can anybody explain what evolutionary advantage may lie in manufacture of
>neuropeptides my subtraction? That is, why make a B I G L O N G
>peptide and then whack off little pieces of it until Mama Nature gets
>what she needs? Is there an evolutionary advantage? If so, what?
In a few cases, such as the instance of proopiomelanocorticotropin, a long
precursor is manufactured and several alternate pathways of cleavage are
available, possibly being useful in coordinating the production of several
peptides which might (I don't know if this applies to this example) be
usefully regulated in synchrony.
Another important question, though, is whether there is a selective disad-
vantage to this strategy-- what is the marginal energetic cost of producing
large precursors and then cleaving them up? It is possible that some of
what is going on is targeting, and that some of it simply reflects the
history of the peptide's evolution. If there is no selection against
the apparently "wasteful" production path, there is no reason for it to
change. The null hypothesis for any characteristic has to be that it
is selectively neutral-- if the null hypothesis is that it is the result
of natural selection, then the hypothesis of natural selection has been
advanced to the status of "proven in all cases". Sewall Wright did some
great work on clarifying the point that if evolution is taken to be
change in the relative frequency of a charactertic over time, then evolution
by natural selection is only one hypothesis that might explain that change.
At least one other explanation is that stochastic processes may explain
the change, especially given that populations are finite and that change can
result from random selection of individuals to be represented in subsequent
generations. A random sample of a finite set can easily be biased, enough
to overcome a selective pressure if the selection is weak on the character
I have no doubt that i have done some violence to Wright's work-- you might
try asking the question in bionet.molbio.evolution to get a better answer
from the folks who do this for a living....
>Here endeth my wise-*ss question of the week. Thanks in advance.
>Doctoral Student, Neuroscience/Nursing
>University of California at San Francisco
>jharper at itsa.ucsf.edu
bell at minerva.cis.yale.edu
More information about the Neur-sci