E. coli's control systems

CSGnet Archive CSG2007
1

[From Bill Powers (2007.05.01.1650 MDT)]
Here is a quote from the abstract of a report in Science for 27 April
2007 by Ishii et. al., Science 316, p. 593,
(2007).

Analysis of metabolic enzyme gene disruptants revealed
unexpectedly small changes in messenger RNA and proteins for most
disruptants. Overall, metabolite levels were also stable, reflecting the
rerouting of fluxes inm the metabolic network. In contrast, E. coli
actively regulated enzyme levels to maintain a stable metabolic state in
response to changes in growth rate. E. coli thus seems to use
complementary strategies that result in a metabolic network rubust
against perturbations.

Perhaps they should be told by someone that they are looking at a
hierarchy of control systems.

Best,

Bill P.

2

[From Rick Marken (2007.05.01.1825)]

Bill Powers (2007.05.01.1650 MDT)

Here is a quote from the abstract of a report in Science for 27 April 2007
by Ishii et. al., Science 316, p. 593, (2007).

Analysis of metabolic enzyme gene disruptants revealed unexpectedly small
changes in messenger RNA and proteins for most disruptants. Overall,
metabolite levels were also stable, reflecting the rerouting of fluxes inm
the metabolic network. In contrast, E. coli actively regulated enzyme levels
to maintain a stable metabolic state in response to changes in growth rate.
E. coli thus seems to use complementary strategies that result in a
metabolic network rubust against perturbations.

Perhaps they should be told by someone that they are looking at a hierarchy
of control systems.

Not being a biologist I find this difficult to follow. I'd appreciate
it if you could explain what they found.

Best

Rick

···

--
Richard S. Marken
Adjunct Professor of Psychology UCLA
Statistical Analyst VHA
rsmarken@gmail.com

3

Bill Powers (2007.05.01.1650
MDT)

Here is a quote from the abstract of a report in Science for 27
April 2007

by Ishii et. al., Science 316, p. 593, (2007).

Analysis of metabolic enzyme gene disruptants revealed unexpectedly
small

changes in messenger RNA and proteins for most disruptants. Overall,

metabolite levels were also stable, reflecting the rerouting of fluxes
inm

the metabolic network. In contrast, E. coli actively regulated enzyme
levels

to maintain a stable metabolic state in response to changes in growth
rate.

E. coli thus seems to use complementary strategies that result in a

metabolic network rubust against perturbations.

Perhaps they should be told by someone that they are looking at a
hierarchy

of control systems.

Not being a biologist I find this difficult to follow. I’d
appreciate

it if you could explain what they found.
From Bill Powers (2007.05.02.0745 MDT)]
Rick Marken (2007.05.01.1825) –
I find it hard, too. I’ll copy this to Bob Franza and see if he can give
us any help.
As near as I can read between the esoteric chemicalspeaks, the authors
have been growing colonies of E. coli (K-12 strain) and subjecting them
to various kinds of disturbances, including disturbances of genetic
expressions and of enzymes. They sample the colonies at frequent
intervals, homogenize the samples, and do a detailed chemical analysis.
Here’s part of what they say:
"In the present study, we measured the global response of E. coli
K-12 cells to genetic and environmental perturbations and compared this
with a detailed quantitative ananysis of specific metabolic
pathways.
“To probe the global cellular response, we determined the relative
quantities of gene and protein expression using DNA microarray analysis
and two-dimensional gel electrophoresis.”
They also used several other quantitative methods to add detail, ending
up measuring dozens of substances over time.
They did the Test for the Controlled Variable on all these substances,
applying disturbances known to have effects on each component, and
measuring the resulting variations in the rate of production of each
component in comparison to the rate observed in an unperturbed sample
from a control group. The observed result, which apparently was
surprising, was that the perturbations had very little effect. I can even
interpret some of their findings to suggest that E. coli has a
reorganizing system,
“E. coli can respond to changes in the concentration of
growth-limiting substrate by regulating the level of enzyme expression to
maximize growth rate, which is reflected in the observed stability of
metabolite levels.” [“maximizing” is the wrong term; they
just meant “restoring” – nothing indicates that the bacteria
tried to grow at the maximum conceivable rate.] [Also,
“regulating” just means “varying”. They do not mean
the level of enzyme expression was held constant, but that it was varied]
and
“E. coli may survive by accumulating additional mutations as we
observed for pfkA and rpiA disruptants.”

Apparently these disruptants altered the genome and the effect was
counteracted by a reorganization of the genes – “accumulated
mutations”. That implies a reorganizing system as opposed to an
ordinary control system.

I wish I could read their diagrams. Or else I wish they had a glimmer of
understanding about hierarchical control systems.

Best,

Bill P.

4

[from Gary Cziko 2007.05.02 16:33 GMT]

I sent the “E. coli’s control systems” conversation to my daughter, Anne-Marie Cziko, who is finishing her doctorate in neuroscience at the University of Arizona. She and her lab deals with micro aspect of nervous systems including genes, synaptogenesis and such.

She responded:

These kinds of things have been known about in biological systems for
years! All signaling networks have robustness built in that can handle
a huge deal of disturbance. We talked about this for several weeks in
one of my graduate student classes.

She will send me some references on “synaptic homeostasis” as an example (those impatient can Google “synaptic homeostasis” and get 881 hits.

–Gary

P.S. It would be interesting for someone to write a paper or a book about all these biological control systems so nonspecialists could understand them, and then show how similar systems might even be involved in animal and human behavior!

5

[from Gary Cziko 2007.05.02 17:10 GMT]

Anne-Marie tells me that Gina Turrigiano is doing research on neuronal homeostasis. There are lots of references on her website:

http://www.bio.brandeis.edu/faculty01/turrigiano.html

–Gary

6

[From Rick Marken (2007.05.02.1310)]

Gary Cziko (2007.05.02 17:10 GMT) --

Anne-Marie tells me that Gina Turrigiano is doing research on neuronal
homeostasis. There are lots of references on her website:

Gina Turrigiano | Faculty | Department of Biology | Brandeis University

I guess I don't quite understand how this relates to the E. coli
stuff. But it is interesting as physiology. I think the research is
about how neural firing properties are controlled but I don't quite
understand the findings. For example, at the site Turrigiano says:

how do neurons and circuits maintain stability in their firing properties
in the face of such dramatic synaptic reconfiguration?

But she doesn't say what firing properties (variables) are thought to
be kept stable (controlled)? Has there been research to determine what
firing properties are controlled?

One possibility is that neuronal activity levels can
homeostatically regulate the properties of neural
circuits to maintain firing rates within certain boundaries.

OK, so firing rate is thought to be a controlled variable. But then
the diagram at the site is confusing; it shows quite different firing
amplitudes following what I think would be called different
experiential disturbances; so firing amplitude doesn't seem to be
controlled. Is the point of the graph that the firing rate remains the
same despite changes in the firing amplitude? If so, then why does the
caption talk only about amplitude?

There are several possible targets for such activity-dependent
regulation of firing rates. First, activity could modify intrinsic
neuronal excitability by modifying the balance of conductances
expressed by a neuron.

This sounds like speculation about _how_ firing rate might be
controlled, which is fine if they have already determined _that_
firing rate is controlled.

My lab is concerned with asking whether such homeostatic mechanisms
operate in mammalian neocortical circuits

It looks like she is looking to see whether a means of controlling
firing rate exists. I hope she has already determined that firing rate
is a controlled variable.

We have shown that activity can scale the strength of synaptic
connections between pyramidal neurons in such a way as to
maintain stability in firing rates;

But is firing rate what is controlled? Or is it firing rate as a
function of input level at the cell body? Or something else? It looks
like they have determined that "firing rate" is controlled; it would
just be nice to get a clearer definition of what the "firing rate"
variable is that is controlled. What is the controlled variable? It
obviously can't be firing rate per se since neural impulses carry
information in terms of time variations in spike rate, since spikes
tend to be the same amplitude. So it would be really a bad idea to
have neural control systems that keep a neuron firing at a
predetermined rate.

Best

Rick

···

--
Richard S. Marken
Adjunct Professor of Psychology UCLA
Statistical Analyst VHA
rsmarken@gmail.com

7

Anne-Marie tells me that Gina
Turrigiano is doing research on neuronal homeostasis. There are lots of
references on her website:


http://www.bio.brandeis.edu/faculty01/turrigiano.html
[From Bill Powers (2007.05.02.1930 MDT)]
Gary Cziko 2007.05.02 17:10 GMT –
I’m not clear about some details. She says:
“We have shown that activity can scale the strength of synaptic
connections between pyramidal neurons in such a way as to maintain
stability in firing rates; increased activity decreases synaptic
strengths, and vice versa.”
What determines firing rates, and firing rates of what? One diagram
suggests that it is the amplitude of individual EPSP (Excitatory
Post-Synaptic Potential) impulses that is regulated, which of course has
no effect on the rate at which such impulses occur. It’s not explained
why there are any impulses in the first place, or where they’re coming
from. Is electrical stimulation of the in vitro dendrites being
used?
Given a constant rate of EPSPs at the input, the output firing rate of a
neuron should increase and decrease with the amplitude of the EPSPs, so
changing that amplitude would affect the output firing rate, but not, of
course, the input impulse rate. This would hold for cell bodies with
relative large capacitances, so the potential at the axon hillock would
be an integrated average over many EPSPs. That mean potential determines
the recovery time after an output impulse, and so determines the
output impulse rate.

What this looks like is an automatic gain control system. Also, it has
potential applications to the circuitry of color vision, since apparently
the color signals received by midbrain neurons are somehow normalized to
“gray”. I’ve speculated about a global negative feedback loop
in the perceptual input function that adjusts the gain of all inputs to
maintain an average color-signal of gray (in the manner of Land’s theory
of color vision). This work seems to suggest that a single neuron could
accomplish something similar.

Is any of this being related to a model of the neuron?

Last thought. Many neurons have local feedback loops that branch off the
output axon and feed back either to the cell body or (maybe) the
dendrites. Could this be part of the homeostatic mechanism, or would that
feedback be too fast?

Best.

Bill P.

8

[From Bill Powers (2007.05.04.0726 MDT)]

Bo Wang(2007.05.03) –

Referring
neuroscience, I remember when I was looking for some info on “biased
random walk” (the behavior of E. coli) an article titled
“Biased Random-Walk Learning: A Neurobiological Correlate to
Trial-and-Error” came across me. It talked about the possibility
that “random variation in synaptic strength” plays a role in
the learning progress of neural networks.

The fulltext could be found by scholar.google.com with that
title.

Most interesting, Bo. The mathematics is far beyond me but the
principles seem closely related to my E. coli method of reorganization. I
notice that several papers seem to be using this method to reorganize
perceptual input functions. Also, one paper applies it to the grazing of
animals, which is appealing since it is like the E. coli way of finding
food! Perhaps Richard Kennaway, who has nothing else to do with his time
(sorry, Richard) will look at this material.

In the chapter on Learning (Chapter 14) in B:CP the principles of biased
random walk learning is laid out – everything except the biased random
walk. When I wrote that book I didn’t see how a purely random
reorganization process could be efficient enough. I say on p. 188 “I
do not assert, however, that this model is detailed enough to account for
the way in which reorganization actually takes place.” What was
missing, of course, was the E. coli principle of moving through
(hyper)space and reorganizing by choosing different directions of
movement at random. The way to do this was not clear to me until Daniel
Koshland published his 1980 book, “Bacterial chemotaxis as a model
behvioral system” (New York: Raven Press, 1980). I think I saw a
review of that book in Science magazine, bought it, and saw immediately
that it solved my problem. Koshland’s book was published long before the
papers in the Google list – but I didn’t see any references to
it.

I recommend a re-reading of Chapter 14 because near its end there appears
the ghost of that chapter on “An experiment with awareness”
that the editors decided to leave out of B:CP. It’s all about how
awareness directs reorganization, so that a probable principle of therapy
would be “change demands consciousness from the point of view that
needs changing.” I think that’s the earliest allusion to the method
of levels in print.

That was a good find, Bo. Also, I did’t know about Scholar.Google.com –
very useful. Thanks. See you in July?

Best,

Bill P.

9

[From Dag Forssell (2007.05.04.1805 PST)]

Bill Powers (2007.05.04.0726 MDT)]

Bill,

I recommend a re-reading of Chapter 14 because near its end there appears the ghost of that chapter on “An experiment with awareness” that the editors decided to leave out of B:CP.

Make that not “An experiment with awareness” but “An experiment with levels”. PCTers will find it in LCSII, pages 41-53.

It’s all about how awareness directs reorganization, so that a probable principle of therapy would be “change demands consciousness from the point of view that needs changing.” I think that’s the earliest allusion to the method of levels in print.

While working with Tim Carey on MOL, we reviewed this. Tim made the point that he got started with MOL because of this chapter in LCSII.

Best, Dag

10

[From Dag Forssell (2007.05.04.1805 PST)]

Bill Powers (2007.05.04.0726
MDT)]

Bill,

I recommend a re-reading of
Chapter 14 because near its end there appears the ghost of that chapter
on “An experiment with awareness” that the editors decided to
leave out of B:CP.

Make that not “An experiment with awareness” but “An
experiment with levels”. PCTers will find it in LCSII, pages
41-53.

It’s all about how awareness
directs reorganization, so that a probable principle of therapy would be
“change demands consciousness from the point of view that needs
changing.” I think that’s the earliest allusion to the method of
levels in print.

While working with Tim Carey on MOL, we reviewed this. Tim made the point
that he got started with MOL because of this chapter in LCSII.

Best, Dag

11

Please give this newbie the references for
LCSII and MOL.

Ted

···

From: Control Systems
Group Network (CSGnet) [mailto:CSGNET@LISTSERV.UIUC.EDU] On Behalf Of Dag Forssell
Sent: Friday, May 04, 2007 7:11 PM
To: CSGNET@LISTSERV.UIUC.EDU
Subject: Re: E. coli’s control
systems

[From Dag Forssell (2007.05.04.1805 PST)]

Bill Powers (2007.05.04.0726 MDT)]

Bill,

I recommend a re-reading of Chapter 14 because near its end there
appears the ghost of that chapter on “An experiment with awareness”
that the editors decided to leave out of B:CP.

Make that not “An experiment with awareness” but “An experiment
with levels”. PCTers will find it in LCSII, pages 41-53.

It’s all about how awareness directs reorganization, so that a probable
principle of therapy would be “change demands consciousness from the point
of view that needs changing.” I think that’s the earliest allusion to the
method of levels in print.

While working with Tim Carey on MOL, we reviewed this. Tim made the point that
he got started with MOL because of this chapter in LCSII.

Best, Dag

12

[From Dag Forssell (2007.05.05.1330 PST)]

···

At 08:47 PM 5/4/2007, you wrote:

Please give this newbie the references for LCSII and MOL.

Ted

Check out www.livingcontrolsystems.com. Check it again in a month or three. I am updating and upgrading.

Best, Dag

13

[From Dag Forssell (2007.05.05.1330 PST)]

···

At 08:47 PM 5/4/2007, you wrote:

Please give this
newbie the references for LCSII and MOL.

Ted

Check out

www.livingcontrolsystems.com. Check it again in a month or three. I
am updating and upgrading.

Best, Dag