SIDXEY I. VELICK Vi\ 



is distributed closer to the binding site than it would be in a random 

 or iniiforni distribution. The interacting tryptophan residues and 

 the bound coenzyme are in fixed orientation in the protein molecule 

 and a gain in transfer efficiency may occiu' through a iavorable value 

 of .v-. The angle 9 is related to the polarization as 



_ 3 cos-g - 1 , . 



^^ ~ cos20 + 3 ^^ 



In Fig. 8 it is seen that the DPNH emission excited at 340 m/x with 

 unpolarized light exhibits nearly the maximal theoretical polarization, 

 whereas the polarization of the fluorescence excited by transfer from 

 the tryptophan is quite low. This was to be expected since the energy 

 reaches the coenzyme from seven different directions. The average 

 value of 6 of DPNH and tryptophan in an LDH domain is about 

 45° but the a and b angles are unknown, x^ is a sufficiently sensitive 

 function of a and b to account for the demarcation of domains in 

 cases where the transfer efficiency is high within a domain. It is to 

 be expected that cases will arise, depending upon the structures of 

 protein and ligand, in which there will be a strong overlap between 

 domains. Because of this possibility, the quenching of protein fluores- 

 cence as a quantitative indicator of ligand attachment must be used 

 with discretion. 1 



' The tryptophan emission of hemoglobin and myoglobin is tjuenched more than 

 911 per cent bv the heme prosthetic group (41). In these cases each heme is as- 

 sociated with a protein siibimit of about 18,000 g molecular weight, which is 

 about half the size of a domain of interaction in the pyridine-nucleotide-enzyme 

 c )mplexes. Extremely efficient quenching is also a consequence of the high oscilla- 

 tor strength of the heme. Titration of the chemically resolved apoprotein with 

 heme (Fiufine and Velick, unpublished), with quenching of protein fluorescence as 

 indicator, shows normal stoichiometry. The quenching is about 85 per cent in 

 the reconstituted system. The absence of strong overlap between domains in this 

 case is associated with the fact that the apoprotein, in the absence of heme, is a 

 mixture of monomer and dimer subunits, chiefly dimer. One prosthetic group per 

 dimer promotes aggregation to the tetramer, corresponding in molecular weight 

 to the normal hemoglobin. Failure to obtain complete quenching and the ap- 

 parent absence of quenching overlap between subimits is a sign that the recon- 

 stituted complex is not identical with the starting material although it is quite 

 similar in its other properties. 



An unusual example of the quenching of protein fluorescence is ])ro\ided by a 

 pure antibody directed against 2,4-dinitrophcnyl haptene groups. Tlie antibody, 

 of molecular weight approximately 180,000, combines with two molecules of the 

 haptene, 2,4-dinitrophenyl-L-lysine, with a dissociation constant of 10" M, re- 

 markably low for an immunochemical system. Ihere is a strong o\erlap between 

 the einissicm band of the protein and the al)sorption band of the haptene, and 

 the quenching is about 7.5 per cent (X'elick, Parker, and Risen, Proc. Xat. Acad. 

 Sri. U. S., in press). 



