Jim Anderson wrote:
>Stephen Oliver proposes that determining functional hierarchies rests on th=
e
>principles of "top-down" metabolic control analysis. Is there an real, or
>experimental relationship?
Although I've not heard the full exposition of Stephen Oliver's ideas on
this, when they first came to my attention last year I gave it some
thought. I presume that he means a real experimental program, not just an
analogy. The following is my ideas about what Oliver might be saying,
although I don't know if this is what he really has in mind. In principle
the use of top down control analysis to provide a rational programme for
hunting the function of ORFs is an attractive idea, but I'm not sure if and
how it would actually work.
Top down control analysis (or more appropriately in this context, what we
have called top down elasticity analysis) starts with a complex system (in
this case it would be a yeast cell culture) and identifies which are the
primary and which the secondary effects of some change (in this case
presumably over- or under-expression of an ORF).
Consider a simple example; a pathway that converts a fixed pool of A to a
variable pool of B to a fixed pool of C, and an effector that alters the
steady state rate of A conversion to C. Both step 1 (A to B) and step 2 (B
to C) change rate in the new steady state with effector present, but which
effect is primary, and which is the secondary response to the resulting
steady state change in [B]? To answer this you first manipulate step 1 and
measure the dependence of step 2 on [B] in new steady states. Separately
you manipulate step 2 and measure the dependence of step 1 on [B]. The
primary site(s) of action will have a changed dependence on [B]; the
secondary site will not.
One of the strengths of the top down approach is that you can group
reactions together, so each step could include large chunks of a cell's
activities. To oversimplify slightly, step 1 could be all catabolism, B
could be ATP and step 2 could be all anabolism. So in principle you could
find out if your primary change was in catabolism or in anabolism. Say it
was in catabolism - you would then regroup catabolism around a different
'B', eg pyruvate, and do another set of experiments to find out if the
primary change was in transport and glycolysis (upstream of pyruvate) or in
the TCA cycle (downstream of pyruvate). And so on, dividing from the top
level down into the system until you identify the unique site or sites of
primary action.
So how would you apply this method to a yeast cell culture (or mammalian
cell line) with a known change in gene expression? Seems to me there are
two approaches.
The first is the steamroller approach. You draw up some sort of global
block reaction scheme of everything you think the cell does, connected by
intermediates. So you might have twenty blocks of functions connected by
thirty intermediates. Then you measure the response of each block to
changes in each intermediate. Then you make your gene expression change,
and remeasure all the responses. Clearly, the simpler your block diagram,
the less work you will have to do, but the less discriminating is your
initial screen. Blocks containing primary sites of action will have
changed rates at the original concentration of one or more of the
intermediates. Once you find a primary effect you divide the relevant
block up into a new set of blocks, and so on until you find the culprit
protein. You could set up a committee to decide how to divide up cell
function, then allocate different blocks to different laboratories, and get
each laboratory to screen each ORF until the whole lot have been
identified. The problem, of course, is to define a sufficiently small
number of meaningful blocks that are experimentally accessible and yet
cover all the complex functions of the cell culture.
In the second approach you have to notice some change caused by altering
expression of your ORF. Then you use that change as your assay, and divide
the system up into blocks that seem relevant to the change you saw, i.e a
different subdivision for every ORF. Then proceed as above, but maybe
doing a lot less work on the way. I suspect that this is pretty much what
people do anyway, but here there is more emphasis on changed responses of
whole blocks of reactions to specified intermediates, rather than
speculative tests of specific targets that seem likely to give the observed
phenotype.
What to other people think about this? Is it feasible? Is it useful? Is it
better than the current strategy of inspired guesswork?
____________________________________________________________________________=
____
Some background for anyone who wants to read more:
Pedro Mendes' Metabolic Control Analysis pages at Aberystwyth -
http://144.124.112.51/metab/mca_home.htm
Athel Cornish-Bowden's pages at Marseilles -
http://ir2lcb.cnrs-mrs.fr/lcbpage/athel/home_page.html
Our own homepages -
http://www.bio.cam.ac.uk/dept/biochem/brand/
____________________________________________________________________________=
____
Some of our relevant publications (several of these are on-line on our
homepages):
Brand, M.D. (1996) Top down metabolic control analysis. J. Theor. Biol.
In press
Brand, M.D. and Kesseler, A. (1995) Control analysis of energy metabolism
in mitochondria. Biochem. Soc. Trans. 23, 371-376
Harper, M.-E. and Brand, M.D. (1995) Use of top-down elasticity analysis
to identify sites of thyroid hormone-induced thermogenesis. Proc. Soc.
Exp. Biol. Med. 208, 228-237
Kesseler, A. and Brand, M.D. (1994) Localisation of the sites of action of
cadmium on oxidative phosphorylation in potato tuber mitochondria using
top-down elasticity analysis. Eur. J. Biochem. 225, 897-906
Buttgereit, F., Grant, A., M=FCller, M. and Brand, M.D. (1994) The effects
of methylprednisolone on oxidative phosphorylation in
Concanavalin-A-stimulated thymocytes. Top down elasticity analysis and
control analysis. Eur. J. Biochem. 223, 513-519
Brand, M.D. and Brown, G.C. (1994) The experimental application of control
analysis to metabolic systems. In "Biothermokinetics" (ed H.V.
Westerhoff), Intercept, Andover, U.K. pp 27-35
Harper, M.-E., Ballantyne, J.S., Leach, M. and Brand, M.D. (1993) Effects
of thyroid hormones on oxidative phosphorylation. Biochem. Soc. Trans. 21,
785-792
Brand, M.D., Chien, L-F. and Rolfe, D.F.S. (1993) Control of oxidative
phosphorylation in liver mitochondria and hepatocytes. Biochem. Soc.
Trans. 21, 757-762
Brand, M.D. (1993) Control of oxidative phosphorylation in liver
mitochondria and cells: top-down control analysis and top-down elasticity
analysis. In "Surviving Hypoxia: mechanisms of control and adaptation"
(eds. P.W. Hochachka, P.L. Lutz, T. Sick, M. Rosenthal and G. van den
Thillart), Chapter 20, pp 295-309, CRC Press, Boca Raton
Martin D. Brand
Department of Biochemistry
University of Cambridge
Tennis Court Road
Cambridge CB2 1QW UK
phone: +44 1223 333648
fax: +44 1223 333345
Internet: http://www.bio.cam.ac.uk/dept/biochem/brand/
