Limits...
Prevention of cross-talk in conserved regulatory systems: identification of specificity determinants in RNA-binding anti-termination proteins of the BglG family.

Hübner S, Declerck N, Diethmaier C, Le Coq D, Aymerich S, Stülke J - Nucleic Acids Res. (2011)

Bottom Line: This analysis revealed the key role of an arginine side-chain for both the high affinity and specificity of LicT for its cognate RAT.Introduction of this Arg at the equivalent position of SacY (A26) increased the RNA binding in vitro but also resulted in a relaxed specificity.Altogether our results suggest that this family of anti-termination proteins has evolved to reach a compromise between RNA binding efficacy and specific interaction with individual target sequences.

View Article: PubMed Central - PubMed

Affiliation: Department of General Microbiology, Institute of Microbiology and Genetics, Georg-August University Göttingen, Grisebach strasse 8, D-37077 Göttingen, Germany.

ABSTRACT
Each family of signal transduction systems requires specificity determinants that link individual signals to the correct regulatory output. In Bacillus subtilis, a family of four anti-terminator proteins controls the expression of genes for the utilisation of alternative sugars. These regulatory systems contain the anti-terminator proteins and a RNA structure, the RNA anti-terminator (RAT) that is bound by the anti-terminator proteins. We have studied three of these proteins (SacT, SacY, and LicT) to understand how they can transmit a specific signal in spite of their strong structural homology. A screen for random mutations that render SacT capable to bind a RNA structure recognized by LicT only revealed a substitution (P26S) at one of the few non-conserved residues that are in contact with the RNA. We have randomly modified this position in SacT together with another non-conserved RNA-contacting residue (Q31). Surprisingly, the mutant proteins could bind all RAT structures that are present in B. subtilis. In a complementary approach, reciprocal amino acid exchanges have been introduced in LicT and SacY at non-conserved positions of the RNA-binding site. This analysis revealed the key role of an arginine side-chain for both the high affinity and specificity of LicT for its cognate RAT. Introduction of this Arg at the equivalent position of SacY (A26) increased the RNA binding in vitro but also resulted in a relaxed specificity. Altogether our results suggest that this family of anti-termination proteins has evolved to reach a compromise between RNA binding efficacy and specific interaction with individual target sequences.

Show MeSH

Related in: MedlinePlus

Effect of reciprocal point mutations in SacY CAT and LicT CAT on RNA recognition in vivo and in vitro. (A and B) Relative anti-termination activity of the wild-type RNA binding domains (SacY CAT and LicT CAT) and their variants carrying the indicated amino acid substitution, in B. subtilis reporter strains SA501 and SA504 expressing the lacZ gene under the control of sacB RAT (grey bars) or licS RAT (black bars). β-Galactosidase activities are expressed in units/mg of proteins, above the background level (about 20 U/mg) measured for transformants of strain SA501 or SA504 harbouring the empty pRL23 cloning vector (with no CAT gene). Note that in this in vivo assay, the anti-termination activity of SacY CAT at the sacB RAT locus appears about 5-fold higher than that of LicT CAT at the licS RAT locus. All measurements were performed on two different transformants and two different extracts from the same bacterial culture. (C and D) Relative RNA binding activity of the wild-type and mutant CATs measured by SPR. The amount of sacB RAT (grey bars) or licS RAT (black bars) RNA bound at equilibrium is expressed as the ΔRU measured using GST alone in the reference flow-cell. The mean value and standard deviation are shown for each GST-CAT fusion, obtained from two independent experiments on different sensor chips (only one experiment for the H9N variant). All measurements were performed on two different transformants and two different extracts from the same bacterial culture. Standard deviations were <10%, except for weak activities (below 50 U/mg) where they were up to 30%.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3105407&req=5

Figure 4: Effect of reciprocal point mutations in SacY CAT and LicT CAT on RNA recognition in vivo and in vitro. (A and B) Relative anti-termination activity of the wild-type RNA binding domains (SacY CAT and LicT CAT) and their variants carrying the indicated amino acid substitution, in B. subtilis reporter strains SA501 and SA504 expressing the lacZ gene under the control of sacB RAT (grey bars) or licS RAT (black bars). β-Galactosidase activities are expressed in units/mg of proteins, above the background level (about 20 U/mg) measured for transformants of strain SA501 or SA504 harbouring the empty pRL23 cloning vector (with no CAT gene). Note that in this in vivo assay, the anti-termination activity of SacY CAT at the sacB RAT locus appears about 5-fold higher than that of LicT CAT at the licS RAT locus. All measurements were performed on two different transformants and two different extracts from the same bacterial culture. (C and D) Relative RNA binding activity of the wild-type and mutant CATs measured by SPR. The amount of sacB RAT (grey bars) or licS RAT (black bars) RNA bound at equilibrium is expressed as the ΔRU measured using GST alone in the reference flow-cell. The mean value and standard deviation are shown for each GST-CAT fusion, obtained from two independent experiments on different sensor chips (only one experiment for the H9N variant). All measurements were performed on two different transformants and two different extracts from the same bacterial culture. Standard deviations were <10%, except for weak activities (below 50 U/mg) where they were up to 30%.

Mentions: In a complementary approach, we have undertaken site-directed mutagenesis of the CAT domains of SacY and LicT based on the structural information available for these proteins. The structure of the LicT CAT-RAT anti-termination complex was solved by NMR, and the residues making direct interaction with the RAT hairpin were identified (17). For SacY CAT the protein–RNA contact region has been mapped by NMR foot printing (5) and it overlaps very well with that of LicT CAT at the dimer interface (Figure 1B). At the RNA level, the sacB RAT and licS RAT sequences differ by only 2 nt (Figure 1A), therefore the 3D structures of the anti-terminator stem–loop are expected to be very similar. In spite of these very strong structural similarities, SacY CAT and LicT CAT display very different affinity and specificity towards their cognate RAT targets (19). The origin of these differences was investigated by introducing point mutations into the RNA-binding domains of SacY and LicT. Four non-conserved residues within the RNA-contacting region of the SacY CAT were targeted (Lys-4, His-9, Ala-26 and Asn-31) and replaced with the amino acid side-chain found at the corresponding positions in the LicT CAT (Ala-4, Asn-9, Arg-27 and Gln-32, respectively). The genes encoding the resulting variants of SacY CAT (K4A, H9N, A26R and N31Q, respectively), or the reciprocal variants of LicT CAT (A4K, N9H, R27A and Q32N, respectively) were introduced into B. subtilis or E. coli expression vectors, and the effect of the mutations on the recognition of sacB RAT and licS RAT was tested both in vivo and in vitro (Figure 4).Figure 4.


Prevention of cross-talk in conserved regulatory systems: identification of specificity determinants in RNA-binding anti-termination proteins of the BglG family.

Hübner S, Declerck N, Diethmaier C, Le Coq D, Aymerich S, Stülke J - Nucleic Acids Res. (2011)

Effect of reciprocal point mutations in SacY CAT and LicT CAT on RNA recognition in vivo and in vitro. (A and B) Relative anti-termination activity of the wild-type RNA binding domains (SacY CAT and LicT CAT) and their variants carrying the indicated amino acid substitution, in B. subtilis reporter strains SA501 and SA504 expressing the lacZ gene under the control of sacB RAT (grey bars) or licS RAT (black bars). β-Galactosidase activities are expressed in units/mg of proteins, above the background level (about 20 U/mg) measured for transformants of strain SA501 or SA504 harbouring the empty pRL23 cloning vector (with no CAT gene). Note that in this in vivo assay, the anti-termination activity of SacY CAT at the sacB RAT locus appears about 5-fold higher than that of LicT CAT at the licS RAT locus. All measurements were performed on two different transformants and two different extracts from the same bacterial culture. (C and D) Relative RNA binding activity of the wild-type and mutant CATs measured by SPR. The amount of sacB RAT (grey bars) or licS RAT (black bars) RNA bound at equilibrium is expressed as the ΔRU measured using GST alone in the reference flow-cell. The mean value and standard deviation are shown for each GST-CAT fusion, obtained from two independent experiments on different sensor chips (only one experiment for the H9N variant). All measurements were performed on two different transformants and two different extracts from the same bacterial culture. Standard deviations were <10%, except for weak activities (below 50 U/mg) where they were up to 30%.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3105407&req=5

Figure 4: Effect of reciprocal point mutations in SacY CAT and LicT CAT on RNA recognition in vivo and in vitro. (A and B) Relative anti-termination activity of the wild-type RNA binding domains (SacY CAT and LicT CAT) and their variants carrying the indicated amino acid substitution, in B. subtilis reporter strains SA501 and SA504 expressing the lacZ gene under the control of sacB RAT (grey bars) or licS RAT (black bars). β-Galactosidase activities are expressed in units/mg of proteins, above the background level (about 20 U/mg) measured for transformants of strain SA501 or SA504 harbouring the empty pRL23 cloning vector (with no CAT gene). Note that in this in vivo assay, the anti-termination activity of SacY CAT at the sacB RAT locus appears about 5-fold higher than that of LicT CAT at the licS RAT locus. All measurements were performed on two different transformants and two different extracts from the same bacterial culture. (C and D) Relative RNA binding activity of the wild-type and mutant CATs measured by SPR. The amount of sacB RAT (grey bars) or licS RAT (black bars) RNA bound at equilibrium is expressed as the ΔRU measured using GST alone in the reference flow-cell. The mean value and standard deviation are shown for each GST-CAT fusion, obtained from two independent experiments on different sensor chips (only one experiment for the H9N variant). All measurements were performed on two different transformants and two different extracts from the same bacterial culture. Standard deviations were <10%, except for weak activities (below 50 U/mg) where they were up to 30%.
Mentions: In a complementary approach, we have undertaken site-directed mutagenesis of the CAT domains of SacY and LicT based on the structural information available for these proteins. The structure of the LicT CAT-RAT anti-termination complex was solved by NMR, and the residues making direct interaction with the RAT hairpin were identified (17). For SacY CAT the protein–RNA contact region has been mapped by NMR foot printing (5) and it overlaps very well with that of LicT CAT at the dimer interface (Figure 1B). At the RNA level, the sacB RAT and licS RAT sequences differ by only 2 nt (Figure 1A), therefore the 3D structures of the anti-terminator stem–loop are expected to be very similar. In spite of these very strong structural similarities, SacY CAT and LicT CAT display very different affinity and specificity towards their cognate RAT targets (19). The origin of these differences was investigated by introducing point mutations into the RNA-binding domains of SacY and LicT. Four non-conserved residues within the RNA-contacting region of the SacY CAT were targeted (Lys-4, His-9, Ala-26 and Asn-31) and replaced with the amino acid side-chain found at the corresponding positions in the LicT CAT (Ala-4, Asn-9, Arg-27 and Gln-32, respectively). The genes encoding the resulting variants of SacY CAT (K4A, H9N, A26R and N31Q, respectively), or the reciprocal variants of LicT CAT (A4K, N9H, R27A and Q32N, respectively) were introduced into B. subtilis or E. coli expression vectors, and the effect of the mutations on the recognition of sacB RAT and licS RAT was tested both in vivo and in vitro (Figure 4).Figure 4.

Bottom Line: This analysis revealed the key role of an arginine side-chain for both the high affinity and specificity of LicT for its cognate RAT.Introduction of this Arg at the equivalent position of SacY (A26) increased the RNA binding in vitro but also resulted in a relaxed specificity.Altogether our results suggest that this family of anti-termination proteins has evolved to reach a compromise between RNA binding efficacy and specific interaction with individual target sequences.

View Article: PubMed Central - PubMed

Affiliation: Department of General Microbiology, Institute of Microbiology and Genetics, Georg-August University Göttingen, Grisebach strasse 8, D-37077 Göttingen, Germany.

ABSTRACT
Each family of signal transduction systems requires specificity determinants that link individual signals to the correct regulatory output. In Bacillus subtilis, a family of four anti-terminator proteins controls the expression of genes for the utilisation of alternative sugars. These regulatory systems contain the anti-terminator proteins and a RNA structure, the RNA anti-terminator (RAT) that is bound by the anti-terminator proteins. We have studied three of these proteins (SacT, SacY, and LicT) to understand how they can transmit a specific signal in spite of their strong structural homology. A screen for random mutations that render SacT capable to bind a RNA structure recognized by LicT only revealed a substitution (P26S) at one of the few non-conserved residues that are in contact with the RNA. We have randomly modified this position in SacT together with another non-conserved RNA-contacting residue (Q31). Surprisingly, the mutant proteins could bind all RAT structures that are present in B. subtilis. In a complementary approach, reciprocal amino acid exchanges have been introduced in LicT and SacY at non-conserved positions of the RNA-binding site. This analysis revealed the key role of an arginine side-chain for both the high affinity and specificity of LicT for its cognate RAT. Introduction of this Arg at the equivalent position of SacY (A26) increased the RNA binding in vitro but also resulted in a relaxed specificity. Altogether our results suggest that this family of anti-termination proteins has evolved to reach a compromise between RNA binding efficacy and specific interaction with individual target sequences.

Show MeSH
Related in: MedlinePlus