Figure 1.
Isolation and in vivo analysis of the peptide K79.
(A) Scheme of the Yeast Two Hybrid selection used to isolate rtTA-binding peptides. The yeast strain harbors a lacZ reporter gene and a HIS3 selection marker controlled by an upstream activating sequence (UAS). The TetR moiety of rtTA was used as a single-chain variant (sc TetR) fused to the C-terminus of the Gal4 DNA binding domain (Gal4-BD). A peptide library was fused to the C-terminus of the Gal4 activation domain (Gal4-AD). (B) E. coli WH207(λtet50) containing lacZ under Tet control was used to analyze K79-mediated effects in E. coli. The TetR expressing plasmid constitutively expresses LacI and the respective TetR variant. The TetR dimer, depicted by grey ovals indicating the DNA- and inducer-binding domains, controls transcription of lacZ. The TrxA-K79 expressing plasmids encode either wildtype TrxA or the C-terminal K79 fusion to TrxA under control of the Ptac promoter. Antagonistic activity of the TrxA-K79 peptide fusion leads to an increase of β-galactosidase expression by reducing dox-mediated repression of the reverse TetR variant. Dox is shown as black pentagon.
Figure 2.
Antagonistic activity of K79 in E. coli.
(A) K79-mediated effects were analyzed using the in vivo system depicted in Figure 1 in which the reverse TetR variant rtTA controls expression of a lacZ reporter gene. In the absence of dox (white bar), β-galactosidase activity is maximal, while it is reduced by rtTA-mediated repression of lacZ expression in the presence of dox (black bar). Expression of TrxA has no effect on β-galactosidae activity, while expression of the C-terminal peptide fusion TrxA-K79 leads to an increase in β-galactosidae activity. Basal expression of TrxA and TrxA-K79 is shown as light grey bar, IPTG-induced overexpression as dark grey bar. Expression of TrxA in the presence of dox (right-hand side of the diagram) has no effect on β-galactosidae activity. Expression of TrxA-K79 in the presence of dox leads to a complete reversal of rtTA-mediated repression of lacZ. The data are shown as mean values ± standard deviations. (B) Western blot analysis of TrxA and TrxA-K79 expression levels in E. coli using an anti-TrxA antibody (bottom row). Protein lysates were obtained from the log-phase cultures used in (A). IPTG was used at 60 µM when indicated. DnaK, which served as loading control, was detected with an anti-DnaK antibody (top row). Molecular weights are indicated on the right-hand side. TrxA, 11.8 kDa; TrxA-K79, 14 kDa; DnaK, 70 kDa.
Figure 3.
Sequence requirements necessary for the antagonistic activity of K79.
(A) Alanine scanning was done in order to identify residues within the K79 peptide sequence contributing to its antagonistic activity. Each position was mutated to alanine and the respective TrxA-K79 fusion was analyzed in E. coli as depicted in Figure 1. The K79 sequence is shown with one-letter abbreviations and the effect of each alanine exchange on rtTA-controlled lacZ expression is depicted underneath the sequence in % antagonistic activity normalized to K79 wildtype which was set to 100%. (B) K79 variants, in which the N-terminal residues 1 to 9 were truncated successively, were analyzed. The antagonistic activity of the respective K79 deletion variant is shown on the right-hand side of the sequence. The antagonistic activity of the K79 wildtype was set to 100%.
Figure 4.
Analysis of K79 “loss of function” mutants in E. coli.
K79 mutants were expressed as C-terminal TrxA fusions and analyzed using the in vivo system shown in Figure 1. Mutations M10A, W11A and G13A were combined to obtain the double mutants M10A/W11A, M10A/G13A and W11A/G13A. Their ability to abolish rtTA-mediated repression of lacZ is compared to K79 wildtype. β-galactosidae activity in the absence of dox is depicted with black bars. Basal expression of TrxA or a specific TrxA-K79 fusion in the presence of dox is depicted with white bars, while IPTG-induced overepression is depicted with grey bars. The sequences of the K79 mutants are shown with one-letter abbreviation on the left-hand side. The data are shown as mean values ± standard deviations.
Figure 5.
K79-mediated effects on gene expression controlled by wildtype TetR.
(A) K79 variants were expressed in an E. coli strain in which lacZ expression is controlled by TetR(B) wildtype. In the absence of dox (white bar), lacZ is repressed by binding of TetR to tetO (compare Figure 1), while in the presence of dox (black bar), lacZ expression is induced. Basal expression of TrxA or a specific TrxA-K79 fusion is depicted with light grey bars, and IPTG-induced overepression is depicted with dark grey bars. The data are shown as mean values ± standard deviations. (B) Same as in (A), but lacZ expression is controlled by the wildtype TetR variant TetR(BD) in which the inducer binding-/dimerization domain is replaced against the class D sequence.
Figure 6.
Structure-function analysis of K79 activity and inducer binding pocket of TetR.
K79 activity is monitored in E. coli in which lacZ expression is controlled by TetR(B) wildtype and various TetR(B) mutants with single-residue exchanges located within the inducer-binding pocket. Mutated amino acids are depicted with respect to their orientation towards tetracycline (tc). The magnesium ion within the tc⋅Mg2+ complex is illustrated as light grey ball, while water molecules are shown as dark grey balls. TetR mutants are grouped with regard to their tc contact (A-ring, Mg2+-complex and D-ring contacting residues). K79 activity with TetR(B) wildtype was set to 100% β-galactosidae activity. Basal expression of TrxA-K79 is depicted with light grey bars, and IPTG-induced overepression is depicted with dark grey bars. The data are shown as mean values ± standard deviations. Values of the β-galactosidae activities monitored are also listed in the Table 1.
Table 1.
Activity of TetR-inducing peptides K79, TIP and TIP2 to induce TetR variants compared to TetR wildtype.
Figure 7.
Methodology to characterize K79 activity in HeLa cells.
(A) The HeLa cell line harbors a stably integrated luciferase gene (luc) under control of a Tet-responsive promoter (TRE). In the presence of dox (white pentagon), the reverse transregulator rtTA binds to the TRE and leads to activation of gene expression. The peptide K79 is fused to the C-terminus of a glucocorticoid receptor variant (GRΔNLS1), and, in the absence of dexamethasone (yellow triangle), resides within the cytoplasm where it cannot interact with nuclear rtTA. (B) After addition of dexamethasone (Dex), GRΔNLS1-K79 is transported to the nucleoplasm via the nuclear pore complex (NPC) where it can interact with rtTA via the K79 moiety. Due to the antagonistic activity of K79, this leads to a decrease in luc expression.
Figure 8.
In vivo characterization of K79 activity in HeLa cells.
Analysis of K79-mediated regulation of luc expression in a HeLa cell line as described in Figure 7. Luciferase expression is controlled by rtTA and, in the presence of doxycycline (dox), leads to activation of gene expression (white bar). K79 is expressed as C-terminal fusion to a glucocorticoid receptor variant (GRΔNLS1). In the absence of dexamethasone, GRΔNLS1 and GRΔNLS1-K79 reside within the cytoplasm (grey bars), while in the presence of dexamethasone (Dex), GRΔNLS1 and GRΔNLS1-K79 are transported to the nucleus (black bars). When dox is used in the latter samples, GRΔNLS1-K79 can interact with rtTA via the K79 peptide which leads to reversal of rtTA-mediated activation of luc expression. The data are shown as mean values ± standard deviations.
Figure 9.
Switching off gene expression: efficiency of K79 compared to dox removal.
(A) The HeLa cell line illustrated in Figure 7 was transiently transfected with an rtTA-encoding plasmid – either alone, or double-transfected with a GRΔNLS1-K79-, or a GRΔNLS1-encoding plasmid, respectively. Before measuring luciferase activity, dox-containing medium was exchanged with dox-free medium at the indicated time points for cells transfected with rtTA only (sample 1; black circles, solid line). For cells additionally transfected with GRΔNLS1 (sample 2; open circles, broken line) or GRΔNLS1-K79 (sample 3; grey triangles, solid line), respectively, 2 µM dexamethasone was added at these time points. The data are shown as mean values ± standard deviations. (B) Experimental procedure as in (A). 10 h time points are shown as x-fold decrease in luciferase expression. The diagram illustrates the effect of dox-removal by medium exchange (left bar), GRΔNLS1-K79 expression (middle bar) and GRΔNLS1-K79 expression together with dox-removal by medium exchange (right bar).