The fluorescence emission spectrum for reduced diphosphopyridine nucleotide (DPNH) in Escherichia coli uridine diphosphate galactose 4-epimerase-DPNH complexes has a maximum at 435 nm, which is about twice as intense when the excitation is at 280 nm as at 340 nm. The fluorescence excitation spectrum monitored at 460 nm has two maxima, one at 340-345 nm and another about twice as intense at 280 nm. The polarization of DPNH fluorescence by these complexes is 0.43-0.44 compared with 0.46 for DPNH immobilized in propylene glycol at -20 degrees C. The small degree of fluorescence depolarization is due to rotational relaxation of the protein, relaxation time 205 ns. The excited-state lifetimes in epimerase-DPNH-nucleotide complexes are 3.5-4.2 ns. The fluorescence data show that the dihydropyridine ring in these complexes is highly immobilized and exhibits no detectable independent motion relative to rotational motions of the protein. The inhibition constants for uridine monophosphate (UMP) and 2,2,6,6-tetramethyl-4-piperidinyl-1-oxyl uridyl pyrophosphate acting as competitive reversible inhibitors of epimerase-DPN+ are 1.2 and 0.2 mM, respectively, at 27 degrees C in 0.1 M sodium bicinate buffer at pH 8.5. A collection of Ki and Km values for uridine nucleotide inhibitors and substrates indicates that the principle substrate binding interactions involve the nucleotide moieties of substrates. Dissociation constants for uridine nucleotides dissociating from epimerase-DPNH-nucleotide complexes, measured by ultraviolet absorption and fluorescence techniques, are 12 muM for UMP, 14 muM for UDP-hexopyranoses, 4 muM for UDP-pentopyranoses, 27 muM for p-bromoacetamidophenyl uridyl pyrophosphate, 0.14 muM for UDP-4-ketohexopyranose intermediate, and 0.36 muM for UDP-4-ketopentopyranose intermediate at 27 degrees C in 0.1 M sodium bicinate buffer at pH 8.5. Analysis of these data shows conclusively that the major part of the binding free energy for UDP-4-ketopyranose intermediates binding to epimerase-DPNH is attributable to the uridylpyrophosphoryl components and that the glycosyl-binding free energies are much smaller. The data show that the action of this enzyme does not require tight binding between the active site and glycosyl groups of either substrates or intermediates, although there is favorable binding of the uridylpyrophosphoryl components, particularly by epimerase-DPNH. It is postulated that nonstereospecific action results from and depends upon relatively weak, nonspecific active site binding of glycosyl groups in substrates and intermediates and that the uridylpyrophosphoryl groups serve as binding anchors in the epimerization process.