Induction of single chain tetracycline repressor requires the binding of two inducers

Abstract

In this article we report the in vivo and in vitro characterization of single chain tetracycline repressor (scTetR) variants in Escherichia coli. ScTetR is genetically and proteolytically stable and exhibits the same regulatory properties as dimeric TetR in E.coli. Urea-dependent denaturation of scTetR is independent of the protein concentration and follows the two-state model with a monophasic transition. Contrary to dimeric TetR, scTetR allows the construction of scTetR mutants, in which one subunit contains a defective inducer binding site while the other is functional. We have used this approach to establish that scTetR needs occupation of both inducer binding sites for in vivo and in vitro induction. Single mutations causing loss of induction in dimeric TetR lead to non-inducible scTetR when inserted into one half-side. The construction of scTetR H64K S135L S138I (scTetR(i2)) in which one half-side is specific for 4-dedimethylamino-anhydrotetracycline (4-ddma-atc) and the other for tetracycline (tc) leads to a protein which is only inducible by the mixture of tc and 4-ddma-atc. Fluorescence titration of scTetR(i2) with both inducers revealed distinct occupancy with each of these inducers yielding roughly a 1:1 stoichiometry of each inducer per scTetR(i2). The properties of this gain of function mutant clearly demonstrate that scTetR requires the binding of two inducers for induction of transcription.

Figure 1

Schematic presentation of dimeric and scTetR mutants. The protein arrangements possible with different genetic situations are shown. (A) The TetR dimers arising from two different tetR alleles expressed in the same cell. The filled star indicates a mutation. The three possible dimers are two homodimers and one heterodimer. (B) The scTetR gene (sctetR) in which one half-side carries the mutation (filled star). The first tetR sequence (white arrow) contains the desired mutation (filled star), followed by the linker encoding sequence designated (SG₄)₅ and the second synthetic tetR sequence (grey arrow). The resulting monomeric protein contains the mutation (filled star) in one half-side.

Figure 2

Western blots of TetR and scTetR. The influence of different cell disruption methods on the integrity of TetR proteins is shown. Soluble proteins from crude cell extracts (35 μg) were loaded in lanes 1–6. Proteins from E.coli WH207_λtet50 cells carrying pWH1925ΔtetR were loaded in lanes 1 and 4; proteins from cells carrying pWH1925 (tetR) were loaded in lanes 2 and 5, and cells carrying pWH1925sc (sctetR) were loaded in lanes 3 and 6. Lane 7 contains 60 ng of purified TetR, and lane 8 contains 20 ng of purified scTetR. The lysis methods employed are indicated below the respective lanes.

Figure 3

Induction efficiencies of sctetR with mutations in one inducer binding pockets. (A) β−gal activities of E.coli WH207_λtet50 transformed with plasmids expressing tetR (black columns) or sctetR (white columns) with mutations causing induction deficiency in tetR (designated N82A, T103A or E147A) in one half-side are shown. The β−gal activity in the absence of tetR was ∼8000 Miller Units and was set to 100% (data not shown). β−gal activities in the absence of inducer were <1% for all variants (data not shown). β−gal expression in the presence of inducers is shown for 0.4 μM tc and 0.4 μM atc as indicated in the figure. (B) EMSA with purified TetR and scTetR N82A carrying a mutation in one inducer binding pocket are shown in the insert. The compounds and the amounts present in the respective reaction mixtures are listed in the table above the lanes. The positions of bands corresponding to free (f) and complexed (c) DNA are indicated on the right side.

Figure 4

Induction efficiencies of scTetR harbouring different inducer specificities. (A) β−gal activities of E.coli WH207_λtet50 transformed with plasmids expressing sctetR or sctetRⁱ² (H64K, S135L, S138I in one half-side) are depicted. The β−gal activity in the absence of tetR is ∼8000 Miller Units and was set to 100% (not shown). Measurements were performed in the absence and presence of 0.4 μM of each inducer as indicated in the figure. (B) EMSA were performed with purified scTetRⁱ² and the compounds and amounts indicated in the table above the lanes. ‘Random’ refers to a non-tetO containing DNA. The positions of bands corresponding to free (f) and complexed (c) DNA are indicated on the right side.

Figure 5

Titration of scTetRⁱ² with tc and 4-ddma-atc. (A) The fluorescence intensity of 2 nmol of purified scTetRⁱ² titrated with up to 8.5 nmol of tc (corresponding to a concentration of 8.5 μM) followed by titration with up to 8.5 nmol of 4-ddma-atc (corresponding to a concentration of 8.5 μM) is shown. Since fluorescence was excited at 370 nm and emission was measured at 515 nm only the binding of tc to the protein is detected. (B) The same experiment except that the excitation and emission wavelengths were 420 and 540 nm, repectively, where only 4-ddma-atc binding is observed. The equivalence points are indicated by the dotted lines.

Figure 6

Urea-dependent denaturation of scTetR. (A) Fluorescence spectra of native (solid line), denatured (long-dashed line) and renatured (dashed–dotted line) scTetR are shown. The difference spectrum between native and denatured forms is shown by the dotted line. (B) The figure shows the change of fluorescence in dependence of the urea concentration. The fluorescence in the absence of urea was set to 100% of folded protein. The denaturation curves were determined at different concentrations of scTetR (filled triangle, 5 μM; circle, 1 μM; open triangle, 0.4 μM, scTetR).