500 SESSION V. DISCUSSION 



The results obtained, shown in Fig. i, show that when chymotrypsin is combined with 

 DNA it is inaaivated, but that the degree of inactivation of the proteolytic activity of 

 the chymotr>'psin in complexes of varying protein content is different. 



The specific proteolytic activity of the enzyme was very low in complexes containing a 

 minimal quantity of protein; as the amount of protein in the complex increased, the specific 

 activity of the enzyme increased sharply, almost reaching the original level of activity of 

 the chymotrypsin (when the protein content was about 68-70%); as the protein content 

 of the complex was further increased the activity of the enzyme again decreased markedly. 



The phenomenon may be explained in many ways; it is, however, beyond doubt that 

 in these explanations the presence of structural alterations in protein when it combines 

 with the DNA molecule plays a definite part, accompanied, perhaps, by partial blocking 

 of the active centres and alterations in the arrangement of the reacting groups in the 

 molecule of the enzyme. Thus, it is possible that the nature of the structural changes in 

 the protein on the DNA is not uniform and the relative amounts of protein unfolded in 

 various ways on the molecule of DNA determine the overall effect of the activity of the 

 enzyme in complexes having different protein contents. 



We are at present studying the mechanism of these phenomena. 



The estabhshment of the fact of the dependence of the proteolytic activity of the enzyme 

 in the complex on the amount of protein entering into combination with the DNA and 

 on the structure of the protein in the molecule of nucleoprotein is clearly of decisive 

 importance for an understanding of the metabolic processes which characterize the 

 biological synthesis of protein. 



J. D. Bernal (Great Britain): 



I should like to make a few small observations on Deborin's paper and on the contri- 

 butions of two other speakers on the question of the association between proteins and 

 other molecules in the cell, namely lipids and nucleic acids. This interaction is determined 

 to some extent, by the shape of the molecules themselves and does not depend on special 

 biological factors. 



It has been shown that in lipoproteins there is a layer of lipid which also determines the 

 structure. This is well seen in the case of plastids and the double membranes within cells. 

 We have also shown that in the proteins of gluten which have this structure even in the 

 grain, the lipid is so closely bound to the protein that it has not been possible to observe 

 it. The other case occurs when the fundamental molecule is in the form of an extended 

 polymer. Sometimes molecules of this or a similar type twist themselves round one 

 another and this does not necessarily occur at a time of growth. As Mitchell has shown, 

 two such molecules, when they twist up, may become entangled with one another in vitro 

 automatically and, from the physical point of view, it may be shown that we are dealing 

 with two types of nucleoprotein. In one case the fundamental partner is the nucleic acid 

 as, for example, in the case of the nucleoprotein of sperm; in the other case the pre- 

 dominant partner is the protein, as in viruses. 



Thus, most of the structures which we find in the cell are seen to be the result of the 

 purely colloidal and crystallographic properties of matter. 



I. Prigogine (Belgium): 



I should like to answer, very briefly, the interesting communication of Prof. Pasynskiï. 

 I agree with most of his remarks but I should like to say that in our work with Wiame 

 one must distinguish clearly between that which has been demonstrated conclusively and 

 that which is more or less hypothetical. It has been shown that, near to the position of 

 equilibrium, the rate of production of entropy by the system has a characteristic magni- 

 tude, characterizing the stationary state. Can one, as we suggest, identify the rate of pro- 

 duction of entropy with the rate of the metabolic process ? This is a question which, in 

 my view, the biologists should solve and it is beyond my power. Now, about the non- 

 linear and linear nature of the equations of entropy. In the first place one must not set 

 the kinetic and thermodynamic characteristics of the process in opposition to one another. 

 In other words, the kinetic characteristics must satisfy the requirements of thermodynamics. 



