the effect of the genetic control on differentia- 

 tion. 



TS'O: Of course. 



B. WRIGHT: I don't see any way to sim- 

 plify this picture. 



LOVETT: Isn't the important thing the 

 relationship between how much of it is simple 

 interaction between pre-existing systems and 

 how much is neiy? A cell turns out a certain 

 amount of enzyme as it grows at a certain rate 

 for a certain amount of time. How much is just 

 interaction between products in that kind of 

 pathway and another pathway to which it con- 

 nects? For instance, perhaps some of these 

 products are going somewhere else and affect- 

 ing enzymes because of their actual concentra- 

 tion level and thus preventing a pathway from 

 functioning. Even though you might measure it 

 in vitro and get a certain level, in fact, the 

 enzyme isn't doing it nearly that fast because 

 it's been shut off by some kind of a feedback 

 mechanism. This could occur in the cytoplasm 

 or by a circuitous path. 



B. WRIGHT: This is pure speculation be- 

 cause there are practically no systems in which 

 we have knowledge enough to do anything except 

 point to one little effect somewhere. 



LOVETT: Well, specifically, yes. However, 

 there's certainly good enough precedent that 

 small molecules having nothing to do with the 

 function of a specific enzyme can cause things 

 like inhibition. Why does one have to restrict 

 it to a previous enzyme in the same pathway? 

 Why couldn't it be used in a coordinating sense 

 to regulate more than one pathway? 



TS'O: Well, actinomycin is actually a good 

 demonstration of this. Presumably, actino- 

 mysin-D doesn't affect enzyme activity, 



LOVETT: Yes, but I think the best answer 

 for that is some other experiments I have done 

 with another organism where I am really worried 

 about actinomycin effects: that is, it causes a 

 lot of turnover of RNA when the cells are not 

 making any protein. 



STROTHER: You're talking about a single 

 cell situation here, but even so it seems to me 

 that the geometry must be taken into account. 

 For instance, the nucleus is surrounded by a 

 cytoplasm, in general, and the cytoplasm is 

 surrounded by the external environment. Now, 

 it would seem to me that what you're really 

 missing in your overall schematic here is the 

 interaction that you observe in a multicellular 

 organism in the very close interaction, even 

 in the single cell, with the environment sur- 

 rounding it. I think you're missing a very 



important part with regard to your signals. 

 Also, I don't know of any part of a consistent 

 theory that doesn't involve statistical analysis 

 of some sort, and I don't see where that appears 

 here. Are they buried in the noise? 



TS'O: Well, I'd like to take that in two 

 parts. First, the membrane is of the greatest 

 importance. Dr. Kahn, Dr. Gregg and myself 

 have certainly talked a lot on that. We' re very 

 conscious of the membrane, but at the present 

 I didn't have enough time to bring in all the 

 elaborations. 



The second point is that, as far as the 

 noise level is concerned, you really couldn't 

 argue this question unless you knew more about 

 the details of this system. You'd have to know 

 more about the hardware and the mechanics 

 of doing it, before you could start posing 

 questions of this kind. You need to know, for 

 instance, when you make a protein, how much 

 error you made in the process. Questions like 

 that are beginning to be approached experi- 

 mentally by Bob Loftfield and others, questions 

 as to how often you make a wrong transcrip- 

 tion. However, I don't think it's germane at this 

 moment to put that into it. The point is not to 

 make the system as complicated as possible 

 but to keep it simple with enough essential 

 parts to help our own decision-making in con- 

 ducting our research. 



GROSS: I'd like to add to both your re- 

 sponses. First of all, it's clear what we're 

 really trying to do is decide what we mean 

 when we say something is differentiated. Em- 

 bryologists have really not been able to agree 

 on that. Eventually this might lead to agree- 

 ment; it might not. Now there are, in fact, 

 systems such as this one cell which, in the 

 absence of other cells, will do what we agree 

 is differentiation. A single sea urchin egg, 

 isolated from all other eggs, can be fertilized 

 and, presumably, it will develop. I suspect that 

 a single spore ot Blastocladiella will germinate. 

 So, to this extent, it's reasonable as a first 

 approximation to talk about this rather simply. 



The statistical point is a very good one and 

 it's precisely there that I would locate all the 

 considerations that motivate Barbara Wright's 

 remarks because the noise level in this system 

 represents the degree of deviation of the over- 

 all result of metabolism from the norm of the 

 population as a whole. There's no doubt that in 

 the cells there are momentary but significant 

 fluctuations. It's the delicate interplay of all 

 these separate steps in the pathway that pre- 

 vents those fluctuations from becoming large 



200 



