Mechanism of Tet repressor induction by tetracyclines: length compensates for sequence in the alpha8-alpha9 loop.

Natural Tet repressor (TetR) variants are alpha-helical proteins bearing a large loop between helices 8 and 9, which is variable in sequence and length. We have deleted this loop consisting of 14 amino acid residues in TetR(D) and rebuilt it stepwise with up to 42 alanine residues. All except the mutant with the longest alanine loop show wild-type repression, but none is inducible with tetracycline. This demonstrates the importance of the alpha8-alpha9 loop and its amino acid sequence for induction. The induction efficiencies increase with loop length, when the more tightly binding inducer anhydrotetracycline is used. The largest increase of inducibility was observed for TetR mutants with loop lengths between eight and 17 alanine residues. Since loop residues Asp/Glu157 and Arg158 are conserved in the natural TetR sequence variants, we constructed a mutant in which all other residues of the loop were replaced by alanine. This mutant exhibits increased anhydrotetracycline induction compared to the corresponding alanine variant. Thus, these residues are important for induction. Binding constants for the anhydrotetracycline-TetR interaction are below the detection level of 10(5) M(-1) for the mutant with a loop of two alanine residues and increase sharply until a loop size of ten residues is reached. TetR variants with longer loops have similar anhydrotetracycline-binding constants, ranging between 2.6 x 10(9) M(-1) and 8.0 x 10(9) M(-1), about 500-fold lower than wild-type TetR. The increase of the affinity occurs at shorter loop lengths than that of inducibility. We conclude that the induction defect of the polyalanine variants arises from two increments: (i) the loop must have a minimal length-to allow efficient inducer binding; (ii) the loop must structurally participate in the conformational change associated with induction.

Tet repressor induction without Mg2+.

We have examined anhydrotetracycline (atc) binding to Tet repressor (TetR) in dependence of the Mg(2+) concentration. Of all tc compounds tested so far, atc has the highest affinity for TetR, with a K(A) of 9.8 x 10(11) M(-1) in the presence of Mg(2+) and 6.5 x 10(7) M(-1) without Mg(2+). Thus, it binds TetR with 500-fold higher affinity than tc under both conditions. The Mg(2+)-free binding of atc to TetR leads to induction in vitro, demonstrating that the metal is not necessary to trigger the associated conformational change. To obtain more detailed information about Mg(2+)-free induction, we constructed and prepared to homogeneity four single-alanine substitution mutants of TetR. Three of them affect residues involved in contacting Mg(2+) (TetR H100A, E147A, and T103A), and one altered residue contacts tc TetR N82A. TetR H100A and E147A are induced by atc, with and without Mg(2+), showing 110-fold and 1000-fold decreased Mg(2+)-dependent and unchanged Mg(2+)-independent atc binding, respectively. Thus, the contacts of these residues to Mg(2+) are not necessary for induction. TetR N82A is not inducible under any of the conditions employed and shows an about 4000-fold decreased atc binding constant. The Mg(2+)-dependent affinity of TetR T103A for atc is only 400-fold reduced, but no induction with atc was observed. Thus, Thr103 must be essential for the conformational change associated with induction in the absence of Mg(2+).

Conformational changes induced in the Tet repressor protein TetR(B) upon operator or anhydrotetracycline binding as revealed by time-resolved fluorescence spectroscopy on single tryptophan mutants.

We have analyzed the tryptophan (trp) fluorescence-decay kinetics of single trp mutants of the Tet repressor protein in the free, the tet operator and anhydrotetracycline (atc)-bound states. The position of the single trp varies between residues 164 and 171, in close proximity to one entrance of the tetracycline-binding pocket. A good fit of the trp fluorescence decay needed generally three exponentials. The decay times vary with detection wavelength, the extent of this variation being correlated to the variation of the emission maximum. Quenching experiments with neutral (acrylamide), cationic (N-methylpyridinium chloride) and anionic quencher (KI) support the interpretation of the three fluorescence components within a conformer model. Operator and atc binding change the ratio of the relative amplitudes of the medium- and long-lived component, thus pointing to structural changes as indicated also by the changes in decay time. Since the fluorescence decay is different between the free, atc- and operator-bound states we conclude that the protein structure is different in each of these three states. The fluorescence quenching constants reflect not only the variation in solvent exposure with position, but also the fact that the net surface charge in this region is negative, because the quenching constants by the cationic quencher are up to 10-fold higher. The atc fluorescence appears to decay monoexponentially with about the same decay time for all mutants, except W170, in which the trp residue sterically interferes with atc.

Backbone dynamics of Tet repressor alpha8intersectionalpha9 loop.

A set of single Trp mutants of class B Tet repressor (TetR), in which Trp residues are located from positions 159 to 167, has been engineered to investigate the dynamics of the loop joining the alpha-helices 8 and 9. The fluorescence anisotropy decay of most mutants can be described by the sum of three exponential components. The longest rotational correlation time, 30 ns at 10 degrees C, corresponds to the overall rotation of the protein. The shortest two components, on the subnanosecond and nanosecond time scale, are related to internal motions of the protein. The initial anisotropy, in the 0.16-0.22 range, indicates the existence of an additional ultrafast motion on the picosecond time scale. Examination of physical models for underlying motions indicates that librational motions of the Trp side chain within the rotameric chi(1) x chi(2) potential wells contribute to the picosecond depolarization process, whereas the subnanosecond and nanosecond depolarization processes are related to backbone dynamics. In the absence of inducer, the order parameters of these motions, about 0.90 and 0.80 for most positions, indicate limited flexibility of the loop backbone. Anhydrotetracycline binding to TetR induces an increased mobility of the loop on the nanosecond time scale. This suggests that entropic factors might play a role in the mechanism of allosteric transition.

Trp scanning analysis of Tet repressor reveals conformational changes associated with operator and anhydrotetracycline binding.

We analysed the conformational states of free, tet operator-bound and anhydrotetracycline-bound Tet repressor employing a Trp-scanning approach. The two wild-type Trp residues in Tet repressor were replaced by Tyr or Phe and single Trp residues were introduced at each of the positions 162-173, representing part of an unstructured loop and the N-terminal six residues of alpha-helix 9. All mutants retained in vivo inducibility, but anhydrotetracycline-binding constants were decreased up to 7.5-fold when Trp was in positions 169, 170 and 173. Helical positions (168-173) differed from those in the loop (162-167) in terms of their fluorescence emission maxima, quenching rate constants with acrylamide and anisotropies in the free and tet operator-complexed proteins. Trp fluorescence emission decreased drastically upon atc binding, mainly due to energy transfer. For all proteins, either free, tet operator bound or anhydrtetracycline-bound, mean fluorescence lifetimes were determined to derive quenching rate constants. Solvent-accessible surfaces of the respective Trp side chains were calculated and compared with the quenching rate constants in the anhydrotetracycline-bound complexes. The results support a model, in which residues in the loop become more exposed, whereas residues in alpha-helix 9 become more buried upon the induction of TetR by anhydrotetracycline.

Structural investigation of Tet repressor loop 154-167: a time-resolved fluorescence study of three single Trp mutants.

We have studied the time-resolved fluorescence of three engineered Tet repressor (TetR) mutants bearing a single Trp residue at positions 162, 163, and 165 in the C-terminal part of the loop joining helices 8 and 9. Detailed analysis indicates that, at 20 degrees, the fluorescence decay of each Trp can be described as the sum of three exponential components with lifetimes in the 1-, 3-, and 6-ns range. Emission wavelength and temperature dependence studies are consistent with a model in which these components are due to the existence of three classes of Trp residues non-interconverting on the nanosecond timescale. Within the framework of the rotamer model, the weak temperature dependence of the lifetimes strongly suggests that the secondary structure of the loop, at least in the 162-165 range, is not altered with temperature. The equilibrium between the rotamers is characterized by an enthalpy-entropy compensation effect which strongly suggests the involvement of background structural regions of TetR in the thermodynamics of the process. The very high deltaH degrees and TdeltaS degrees observed (up to 18 kcal/ mol) should reflect the temperature-dependent conformational change of a large part of the protein which would alter the rotamer distribution of the Trp residues. Taken together, our results are consistent with the existence of (at least) two conformations of the loop and suggest a model for loop motion.

Regulation of expression, genetic organization and substrate specificity of xylose uptake in Bacillus megaterium.

Xylose uptake in Bacillus megaterium depends on expression of a putative H+/xylose symporter encoded by xylT, the last gene in the xyl operon. Insertional inactivation of xylT leads to an apparent uptake deficiency determined with whole cells and severely slower growth on xylose as sole carbon source. Expression of xylT is xylose inducible and subject to carbon catabolite repression mediated by CcpA and cre. Northern analysis of the xyl mRNA reveals that a potential stem-loop structure located in the non-translated region between xylA and xylB presumably acts as a transcriptional terminator, as it leads to different amounts of the respective mRNA sections: the 5'-xylA portion is very abundant, while the 3'-xylBT portion constitutes only a fraction of it. XylT has an apparent Michaelis constant (KM) of approx. 100 microM and is competitively inhibited by glucose with an inhibitor constant KI of 16 mM.

Tetracycline analogs affecting binding to Tn10-Encoded Tet repressor trigger the same mechanism of induction.

We examined the influence of substituents in tetracycline (tc) analogs modified at positions 2 and 4-9 and anhydrotetracycline (atc) on induction of the Tn10-encoded Tet repressor (TetR) by a quantitative in vitro induction assay. The equilibrium association constants of the modified tc to TetR were independently determined to distinguish effects on binding from those on induction. We found a correlation between the binding affinity and induction of TetR for most tc analogs. While a substitution at position 5 revealed only minor effects, changes at position 6 increased binding and induction efficiencies up to 20-fold. A chlorine at position 7 or 8 enhanced binding and induction about 4- and 9-fold, respectively. Substituents at position 9 decreased binding up to 5-fold. Epimerization of the dimethylamino function at position 4 in 4-epi-tc resulted in about 300-fold-reduced binding and 80-fold-reduced induction. Substitution of this grouping by hydrogen in 4-de(dimethylamino)-tc resulted in no binding and no induction. The respective atc analog failed to induce as well, although binding was still observed. The dimethylamino function may, thus, play a role in triggering the conformational change of TetR necessary for induction. Substitution of the 2-carboxamido by a nitrilo function did not influence binding and induction efficiencies. Atc showed about 30-fold increased binding and induction, being the most effective inducer tested in this study. The equilibrium association constants of most TetR-[Mg-tc]+ and TetR-([Mg-tc]+)2 analog complexes with tet operator are decreased about 10(2)- and 10(8)-fold, respectively, as compared to those of free TetR. This suggests that these tc analogs share the same molecular mechanism of TetR induction.