However, there is an important difference between the cellular homeostatic mechanisms we reviewed in chapter 6 (compartmentalisation and enzyme control) and the control of gene expression. The mechanisms outlined in chapter 6 are usually very rapid. Typically, they have time-courses in the order of milliseconds. But a change in gene expression takes effect much more slowly: the new protein appears in minutes, not milliseconds. This is why we regard the homeostatic mechanisms of chapter 6 as aspects of internal state and the control of gene expression as a way of adjusting or altering the internal state.
Imagine a series of times: ti , t2 , t3 and so on. At each of these times the cell has a particular internal state, S, and a particular pattern of gene expression, G. "G" represents the set of genes that are being transcribed and their transcription rates. Let us use "S1" to mean the internal state at time t1 and "G1" to mean the pattern of gene expression at time t1. In the nutrient-storage case outlined above, S1 includes the suddenly increased amount of nutrient, and G1 includes the rapid expression of the "storage enzyme" gene. It will be several minutes before this enzyme is available to the cell; we have now reached time t2. Once the enzyme is available, the internal state is changed: the nutrient is converted to storage form and the amount of free nutrient in the cell falls. S1 becomes S2.
In so far as S2 is different from S1, it will alter the pattern of gene expression, which now becomes G2. But it will be time t3 before G2 affects the internal state, changing it from S2 to S3. The new internal state S3 will then change G2 to G3. And so on.
The cell does not really make a series of sudden jumps (from G2 and S2 at t2 to G3 and S3 at t3 etc.). There is a succession of smooth changes in S (internal state) and G (pattern of gene expression) over a continuous time-course. Genes differ in their expression rates and their responses to transcription factors and repressors. They are not all switched on and off,
accelerated or repressed at the same time. Nor are their transcription rates all identical. Nevertheless the message of Fig. 8-2 is valid. The internal state affects the pattern of gene expression more or less immediately; but changes in gene expression alter the internal state after a delay.
Suppose the cell needs to maintain its internal state, not to change it. (This is the situation we have been considering up to now.) Events such as a sudden uptake of nutrient perturb the internal state (say at time ti). This perturbation lasts until the consequent changes in gene expression take effect (time t2). But once the gene expression pattern has been changed, the change is likely to persist for a while. A gene switched on at time t1 might be switched off again at t2, but the enzyme or other protein made while the gene is active might not be removed or inactivated immediately. Indeed, the messenger RNA for this protein might be stable. This can perturb the internal state in a different direction; the cell "overcorrects" Gene expression then changes once more to correct the overcorrection. This oscillating behaviour can go on more or less indefinitely. In some respects, a cell's internal state tends to behave like a car fishtailing along an icy road. Every time the rear of the car goes out of line the steering is adjusted; the car overcorrects; the steering is adjusted again; the car overcorrects again; and so on. A cell's internal state tends to oscillate over time.
The effects of this are apparent in some hormone-secreting cells. When they are active, these cells do not usually secrete the hormone smoothly and continuously, but in a succession of short pulses15. The blood stream irons out these pulses so that the target tissue experiences a steadily increased concentration of the hormone, but the behaviour of the secreting cell itself is oscillatory.
Was this article helpful?