DINITROPHENYLATION OF RIBONUCLEASE A 47 



Had there been alkylation at the thioether sulphur of this amino acid, it is 

 hkely that the acid decomposition products of the alkyl derivative (cf. 

 Gundlach et al. [7]), particularly homoserine lactone, would have been 

 formed and detected. The analyses for the hydroxyamino acids are also 

 satisfactorily constant in the series, though it is clear that for an amino 

 acid like serine, of which there are fifteen residues in ribonuclease A, the 

 precision attainable by difference analysis would make it impossible to 

 detect a change of half a residue. 



Comparison of the values for lysine, histidine, and tyrosine in Table I 

 shows that substitution on lysine is clearly the most rapid reaction. Sub- 

 stitution on histidine does not become detectable before the 60-minute 

 point, while reaction at tyrosine is not evident until the 240-minute 

 point (analyses of the products obtained at longer time intervals are not 

 included in Table I). A further breakdown of the results of the lysine 

 analyses is possible. The values for e-dinitrophenyl-lysine show that this 

 derivative appears to the extent of 1-04 equivalents in 40 minutes, at 

 which time the sum of the equivalents of unsubstituted lysine and e- 

 dinitrophenyl-lysine is equal (within the combined errors involved) to the 

 total available lysine. This sum becomes significantly less than 10 at 60 

 minutes, with e-dinitrophenyl-lysine at 1-4 equivalents, and falls to a 

 value of 9 at 180 minutes, at which time the value for e-dinitrophenvl-lvsine 

 has reached 2 • i equivalents. A value of 9 equivalents for the sum of free 

 lysine and e-dinitrophenyl-lysine is the value to be expected of a protein 

 in which the terminal a-amino group has been completely substituted. 

 Since determinations of a,e-bis-dinitrophenyl-lysine were not carried out, 

 an estimate of double substitution at the terminal lysine residue in ribonu- 

 clease A could not be made. Thus the values for e-dinitrophenvl-lvsine 

 represent minimum estimates of the extent of substitution of the e-amino 

 groups. Nonetheless, the results in Table I demonstrate that at least two 

 e-amino groups in ribonuclease A may react faster with fluorodinitro- 

 benzene than the a-amino group under these conditions. 



Further progress in interpreting these observations depends on the 

 isolation and characterization of the partly substituted protein derivatives. 

 Moreover, without further information about the initial steps, it would be 

 impossible to investigate the kinetics of subsequent substitutions in the 

 molecule. As will be seen some success in the direction of isolating the 

 initial reaction products at pH 8 has been attained. Before considering 

 the isolation work, the effect of dinitrophenylation on the activity of the 

 enzyme will be described. 



Loss of enzymic activity on dinitrophenylation 



Ribonuclease A (acetate) is rapidly inactivated by dinitrophenvlation 

 at pH 8. After an initial lag phase (also obvious from the results in Table I), 



