840 LIGHT AND LIFE 



water) , is nearly as efficient as DPNH in both reactions. In both 

 reactions its fluorescence emission is shifted toward bhie and is en- 

 hanced upon being bound. It is clear that each enzyme binds and 

 polarizes the coenzyme in its own particular way. 



The transfer of excitation energy from the protein to the coenzyme 

 is well illustrated in the case of lactic dehydrogenase. Its emission 

 seems predominantly to relate to its 28 tryptophan residues. The 

 quantum yield of LDH is nearly 40 per cent, in contrast to the 10 

 per cent of GPD, and in terms of the light absorbed by tryptophan 

 it actually amounts to 70 per cent. The fluorescence emission of LDH 

 is strongly quenched by denaturation in 6 M urea, whereas the emis- 

 sion of free tryptophan is enhanced under the same conditions; hence 

 the high yield must be related to the tertiary structure of the pro- 

 tein. Polarization of the emitted light is about 25 per cent, or about 

 the same as that of free tryptophan in a highly viscous medium or at 

 a low temperature. The partial depolarization occuring in this and 

 other large proteins therefore arises probably from vibration or oscil- 

 lation of the indole side-chains. When the excitation energy passes 

 from the protein to the bound coenzyme a new coenzyme excitation 

 band appears and at the same time the protein emission is quenched 

 by two-thirds, if all the binding sites for DPNH on the enzyme are 

 occupied. In short, the data suggest that the protein emission is 

 from tryptophan, that tryptophan residues are the donors of the 

 energy transferred to the bound coenzyme, and that tyrosine trans- 

 fers little or none of its absorbed light energy to tryptophan. The 

 titration curves of enzyme with DPNH, using either the quenching 

 of protein fluorescence or the emission by the excited dinucleotide 

 as measures, are strictly stoichiometric. It follows that each mole- 

 cule of nucleotide bound quenches only the tryptophan residues 

 within its own domain, with negligible overlap. There is no indica- 

 tion that these proteins are aggregates which may dissociate ( except 

 for glutamic dehydrogenase, the subunits of which are still of large 

 molecular weight and each carry 3 or 4 binding sites for DPNH) . 

 The demarcation of "domains" strictly corresponding to individual 

 bound coenzyme molecules therefore becomes quite a problem, less 

 difficult in the case of an enzyme with low quenching, such as GPD, 

 than in such a case as the LDH-APNH complex, where the quench- 

 ing is 80 per cent — for in the latter instance each molecule of the 

 bound coenzyme analogue must either capture all of the energy from 



