N-Bromosuccinimide modification of Lac repressor protein.

The modification of lnc repressor protein with N-bromosuccinimide has been studied. Treatment with S-bromosuccinimide resulted in the oxidation of one of the two tryptophan residues in the repressor monomer; in the presence of inducer, isopropyl-1-thio-/3-o-galactoside, the tryptophan was protected from oxidation. Anti-inducer, o-nitrophenylP-n-fucoside, did not protect the reactive residue. The tryptophan involved was identified as residue 209. Some loss of ability to bind inducer was observed upon treatment with libromosuccinimide; however, this loss was not correlated with the oxidation of any amino acid and appeared to be attributable to a general destabilization of the protein. The facts that (n) inducer protected the oxidation of tryptophan while anti-inducer did not, and (h) the oxidation of tryptophan did not alter the affinity of repressor for inducer imply that tryptophan 209 is affected by the conformational change upon inducer binding but is not in the inducerlantiinducer binding site. This is consistent with conclusions drawn from previous spectral studies implicating tryptophan as participating in a conformational isomerization rather than interacting directly with ligand. Loss of operator binding activity was observed with low levels of iii-bromosuccinimide. The presence of inducer or anti-inducer did not protect against this loss of operator binding activity. Methionine or cysteine (or both), which were each oxidized by IV-bromosuccinimide, may be involved in the binding of operator DNA, and their oxidation may have directly affected I)NA binding; more likely, their oxidation may have produced a secondary effect which elicited a change in the conformation of the protein rendering it incapable of binding operator DNA. At levels of S-bromosuccinimide where loss of ability to bind operator DNA occurred, ability to bind nonspecifically to DNA was maintained. This provides chemical evidence that the nonspecific DNA and the operator DNA binding sites are functionally separable and that operator binding can be altered without affecting the binding to nonspecific DNA.

From the Department of Biochemistr,y, Rice University, Houston, Texas 77001 The modification of lnc repressor protein with N-bromosuccinimide has been studied. Treatment with S-bromosuccinimide resulted in the oxidation of one of the two tryptophan residues in the repressor monomer; in the presence of inducer, isopropyl-1-thio-/3-o-galactoside, the tryptophan was protected from oxidation. Anti-inducer, o-nitrophenyl-P-n-fucoside, did not protect the reactive residue.    Since no change was found in the K,, for IPTG (K,, = 3 2 1 x lo-';, in 0.5 M Tris/Cl, pH 7.0), one portion of the repressor population retained unaltered IPTG binding activity while another portion did not bind IPTG to an extent detectable by equilibrium dialysis. Measurement of operator binding activity reflected a loss of 80% of the ability to bind operator in repressor modified with less than a 5-fold excess of N-bromosuccinimide per monomer (Fig. 4) IPI'G, but the decreased ability to bind inducer could not be correlated to the oxidation of a specific amino acid. This loss in activity was confined to only a fraction of the protein, since equilibrium dialysis yielded a dissociation constant equivalent to that obtained for repressor in the same buffer system. The observation that exposure to light in the presence of N-bromosuccinimide resulted in the loss of IPTG binding activity implies that inactivation was probably due to secondary reactions. With a 16-fold excess of N-bromosuccinimide, the loss in IPTG binding activity was also accompanied by a change in the circular dichroism spectrum of the protein; this circular dichroism change reflected partial unfolding of the repressor protein. The fact that even with high excesses of N-bromosuccinimide, at least one-half of the population of repressor retained full inducer binding activity implies that tryptophan 209 is not part of the inducer binding site. This conclusion is in agreement with spectral studies which have indicated that tryptophan is perturbed by the conformational change upon inducer binding rather than by the ligand itself (2)(3)(4)(5).
In contrast to inducer binding activity, the ability of the protein to bind operator DNA was markedly affected by Nbromosuccinimide modification.
With less than a 5-fold excess of reagent, the protein lost 80% operator binding activity while retaining 95% of its ability to bind IPTG. The loss in operator binding activity could not be directly correlated to the oxidation of any single amino acid. It is possible that the loss in activity was due to the oxidation of cysteine or methionine (or both) with a subsequent alteration of the secondary or tertiary structure of the surrounding region of the protein. The oxidation of tryptophan 209 was probably not responsible for the loss in operator binding activity since its reaction was protected by the presence of inducer and operator activity was not. Chromatography on Sephadex G-100 indicated that the decreased binding to operator DNA was not due to an alteration of the oligomeric structure of the protein. Likewise, circular dichroism spectroscopy indicated no detectable change in the helical backbone structure of the protein. However, given the size of the operator binding site and its lability, it is not inconsistent that changes undetectable by these methods could still result in alterations in the active structure of the protein.
Previous chemical studies have indicated a concomitant loss of both operator and nonoperator DNA binding activities of the lac repressor protein. Treatment of the protein with heat or trypsin, or the presence of actinomycin, affected the two DNA binding activities identically (23). However, Schlotmann et al. (24) and Beyreuther (25) have isolated a small class of mutants which will bind to DNA cellulose but have lost the ability to bind operator DNA. Modification of repressor with N-bromosuccinimide provides the first chemical evidence that the operator and nonoperator binding sites are functionally separable. With low excesses of N-bromosuccinimide, 80% of the operator binding activity was lost, while no alteration in the ability to bind nonoperator DNA was observed.
This study of the reaction of lac repressor with N-bromosuccinimide yields two conclusions regarding the function of the protein molecule. (a) Tryptophan 209 is affected by the conformational change which accompanies inducer binding but does not appear to be located directly in the inducer binding site. (6) The operator DNA and nonspecific DNA binding sites are functionally separable; alteration of the operator DNA site does not necessarily affect alteration of the nonspecific DNA site.