8-Anilion-1-naphthalenesulfonate (Ans), recrystallized from water as the magnesium salt, contains a fluorescent impurity representing 0.3% of the absorbance at 351 nm. This impurity can be removed by Sephadex LH-20 chromatography. The chromatographic and spectral properties of this impurity suggest that it is bis(Ans), a dimer of Ans. This bis(Ans) impurity makes a significant contribution to the fluorescence increment observed when lac repressor is added to recrystallized Ans. This occurs because bis(Ans) binds much more tightly to this protein than does Ans. The dissociation constant divided by the number of binding sites per subunit is 3.1 X 10(-6) M for bis(Ans); the corresponding value for Ans is greater than 1 X 10(-4) M. Because of their differing absorption spectra, the impact of this bis(Ans) impurity is especially large with excitation wavelengths above 400 nm. Users of recrystallized Ans should consider the potential consequences of this impurity whenever working with a protein to which Ans binds weakly.

The association of lac repressor with poly[d(A-T)] was monitored with the fluorescent prob 8-anilino-1-naphthalenesulfonate (Ans). Excess poly[d(A-T)] decreased the emission intensity of the repressor--Ans complex by 30%. Fluorescence titrations indicated that 33 +/- 4 base pairs were required to bind all of the repressor. Sedimentation studies indicated, however, that all of the repressor sedimented as a protein--DNA complex with as few as 10 to 15 base pairs per tetramer, even in the presence of Ans. These data are interpreted with two models: one where repressors bind to both sides of the DNA (Butler, A. P., et al. (1977) Biochemistry 16, 4757: Zingsheim, H.P., et al. (1977) J. Mol. Biol. 115, 565), the other where a double layer of repressors bind to a single side of the DNA. Removal of the amino-terminal regions from the repressor decreased the fluorescence from bound Ans by 77%. The amino-terminal fragments alone did not enhance Ans fluorescence.

We describe a homogeneous substrate-labeled fluorescent immunoassay for human IgM. In this competitive-binding method we use a fluorogenic substrate for Escherichia coli beta-galactosidase, N-(6-aminohexyl)-7-beta-galactosyl-coumarin-3-carboxamide, which is covalently coupled to IgM. The fluorescence emission intensity of the labeled IgM at 450 nm (with excitation at 400 nm) is negligible, but when beta-galactosidase is added, the acetal linkage of the galactosyl moiety is hydrolyzed and the product has a greatly enhanced fluorescence. Formation of this fluorescent product is inhibited when antibody specific for IgM is bound to the labeled protein. In competitive-binding reactions, IgM from the serum competes with the labeled IgM for antibody binding sites and consequently the fluorescence produced by the enzymic reaction is related to the IgM concentration. The working range of the assay is between 0.5 and 5.0 g of IgM per liter when a 50-fold predilution of the sera is used. The assay does not cross react significantly with immunoglobulins G or A.