Limits...
Escherichia coli RNA polymerase-associated SWI/SNF protein RapA: evidence for RNA-directed binding and remodeling activity.

McKinley BA, Sukhodolets MV - Nucleic Acids Res. (2007)

Bottom Line: Specifically, the formation of stable RapA-RNA intermediates in transcription and other, independent lines of evidence presented herein indicate that RapA binds and remodels RNA during transcription.Our results are consistent with RapA promoting RNA release from DNA-RNA polymerase-RNA ternary complexes; this process may be accompanied by the destabilization of non-canonical DNA-RNA complexes (putative DNA-RNA triplexes).Taken together, our data indicate a novel RNA remodeling activity for RapA, a representative of the SWI/SNF protein superfamily.

View Article: PubMed Central - PubMed

Affiliation: Laboratory of Biochemistry, Department of Chemistry, Lamar University, Beaumont, TX 77710, USA.

ABSTRACT
Helicase-like SWI/SNF proteins are present in organisms belonging to distant kingdoms from bacteria to humans, indicating that they perform a very basic and ubiquitous form of nucleic acid management; current studies associate the activity of SWI/SNF proteins with remodeling of DNA and DNA-protein complexes. The bacterial SWI/SNF homolog RapA-an integral part of the Escherichia coli RNA polymerase complex-has been implicated in remodeling post-termination DNA-RNA polymerase-RNA ternary complexes (PTC), however its explicit nucleic acid substrates and mechanism remain elusive. Our work presents evidence indicating that RNA is a key substrate of RapA. Specifically, the formation of stable RapA-RNA intermediates in transcription and other, independent lines of evidence presented herein indicate that RapA binds and remodels RNA during transcription. Our results are consistent with RapA promoting RNA release from DNA-RNA polymerase-RNA ternary complexes; this process may be accompanied by the destabilization of non-canonical DNA-RNA complexes (putative DNA-RNA triplexes). Taken together, our data indicate a novel RNA remodeling activity for RapA, a representative of the SWI/SNF protein superfamily.

Show MeSH

Related in: MedlinePlus

Predicted partial structure of the RapA NTPase/(putative) DNA-binding module. RapAs amino acid sequence was threaded into the SsoRad54–DNA structure via the SWISS-MODEL Protein Modeling Server (32–34). (A) Homology between SWI/SNF subdomain ‘I’ of RapA and that of the Sulfolobus solfataricus Rad54 homolog (SsoRad54). Identical and homologous amino acids are marked by, respectively, red and blue boxes. A limited sequence homology between SWI/SNF subdomain ‘I’ of RapA and an RNA-binding, or ‘S1’-module of the ribosomal protein S1 is also indicated (dashes, colons). The RapA SWI/SNF subdomains are shown schematically at the top; the aligned segments represent the section that yielded the homology model shown below. (B and C) Predicted structure of the RapA NTPase/(putative) DNA-binding module (sticks and mesh shown in color) superimposed with the homologous domain of SsoRad54 (gray mesh) bound to DNA (shown as CPK spheres), as reported by the Hopfner group (24). (D–G) Predicted partial structure of the RapA NTPase module (shown in color) superimposed with the complete SsoRad54 ATPase core–DNA complex (shown in gray scale) (24). Key amino acid changes in the RapA NTPase module relative to that of SsoRad54 are highlighted as CPK spheres. Lys183 (a highly conserved residue present in both proteins), alanine substitution of which results in significantly reduced ATPase activity in purified RapA (23), is shown in salmon pink. Several amino acids present in RapA but not found in SsoRad54, such as Arg222, Arg221, Tyr235 and Glu233, nearly co-align at certain viewpoints along the axis pointing into the DNAs major groove. The figures were prepared using PyMOL (DeLano Scientific LLC, San Carlos, CA).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2175355&req=5

Figure 8: Predicted partial structure of the RapA NTPase/(putative) DNA-binding module. RapAs amino acid sequence was threaded into the SsoRad54–DNA structure via the SWISS-MODEL Protein Modeling Server (32–34). (A) Homology between SWI/SNF subdomain ‘I’ of RapA and that of the Sulfolobus solfataricus Rad54 homolog (SsoRad54). Identical and homologous amino acids are marked by, respectively, red and blue boxes. A limited sequence homology between SWI/SNF subdomain ‘I’ of RapA and an RNA-binding, or ‘S1’-module of the ribosomal protein S1 is also indicated (dashes, colons). The RapA SWI/SNF subdomains are shown schematically at the top; the aligned segments represent the section that yielded the homology model shown below. (B and C) Predicted structure of the RapA NTPase/(putative) DNA-binding module (sticks and mesh shown in color) superimposed with the homologous domain of SsoRad54 (gray mesh) bound to DNA (shown as CPK spheres), as reported by the Hopfner group (24). (D–G) Predicted partial structure of the RapA NTPase module (shown in color) superimposed with the complete SsoRad54 ATPase core–DNA complex (shown in gray scale) (24). Key amino acid changes in the RapA NTPase module relative to that of SsoRad54 are highlighted as CPK spheres. Lys183 (a highly conserved residue present in both proteins), alanine substitution of which results in significantly reduced ATPase activity in purified RapA (23), is shown in salmon pink. Several amino acids present in RapA but not found in SsoRad54, such as Arg222, Arg221, Tyr235 and Glu233, nearly co-align at certain viewpoints along the axis pointing into the DNAs major groove. The figures were prepared using PyMOL (DeLano Scientific LLC, San Carlos, CA).

Mentions: This model was considered in our earlier study (23), and it was based on the speculation that RNA polymerase ‘trapped’ on DNA cannot disengage from it in order to reinitiate new cycles of transcription efficiently. If RapA were indeed to promote transcriptional cycling by displacing RNA polymerase from DNA, under most circumstances [the exception being a nearly instantaneous transfer of the polymerase from the terminator to the promoter section in the DNA template, which, at least in theory, cannot be entirely ruled out due to possible co-alignment of these two sections in supercoiled DNA (see Ref. 23; Figure 8 therein)] this should be accompanied by a measurable increase in the fraction of free RNA polymerase in the system. We tested this possibility using PAGE- and ultracentrifugation-based techniques. In the first, PAGE-based assay, in vitro transcription reactions carried out to stationary phase (with or without RapA present) were fractionated on non-denaturing 5% polyacrylamide gels in the presence of magnesium, and the amounts and subunit composition of the DNA-bound polymerase were determined (Figure 6). These experiments showed no detectable reduction in the amount of DNA-associated RNA polymerase in the presence of RapA (Figure 6, densitogram; compare the levels of the large RNA polymerase subunits in reactions with or without RapA). Also, this set of experiments showed that >85% of RapA dissociated from the DNA-bound RNA polymerase, while >50% of the sigma70 subunit was retained by the DNA-bound enzyme (Figure 6, graph). In the second approach, in vitro transcription reactions were subjected to ultracentrifugation in order to determine the fractions of DNA-bound and free RNA polymerase in reactions with or without RapA. Similarly, these experiments showed no effect of excess RapA on the ratio of free and DNA-bound RNA polymerase in the system (data not shown). Furthermore, a number of other independent experiments, which assessed the amount of DNA-associated RNA polymerase in reactions with or without RapA consistently showed no effect of RapA on the amount of DNA-associated polymerase (for example, see Figure 4 above).Figure 6.


Escherichia coli RNA polymerase-associated SWI/SNF protein RapA: evidence for RNA-directed binding and remodeling activity.

McKinley BA, Sukhodolets MV - Nucleic Acids Res. (2007)

Predicted partial structure of the RapA NTPase/(putative) DNA-binding module. RapAs amino acid sequence was threaded into the SsoRad54–DNA structure via the SWISS-MODEL Protein Modeling Server (32–34). (A) Homology between SWI/SNF subdomain ‘I’ of RapA and that of the Sulfolobus solfataricus Rad54 homolog (SsoRad54). Identical and homologous amino acids are marked by, respectively, red and blue boxes. A limited sequence homology between SWI/SNF subdomain ‘I’ of RapA and an RNA-binding, or ‘S1’-module of the ribosomal protein S1 is also indicated (dashes, colons). The RapA SWI/SNF subdomains are shown schematically at the top; the aligned segments represent the section that yielded the homology model shown below. (B and C) Predicted structure of the RapA NTPase/(putative) DNA-binding module (sticks and mesh shown in color) superimposed with the homologous domain of SsoRad54 (gray mesh) bound to DNA (shown as CPK spheres), as reported by the Hopfner group (24). (D–G) Predicted partial structure of the RapA NTPase module (shown in color) superimposed with the complete SsoRad54 ATPase core–DNA complex (shown in gray scale) (24). Key amino acid changes in the RapA NTPase module relative to that of SsoRad54 are highlighted as CPK spheres. Lys183 (a highly conserved residue present in both proteins), alanine substitution of which results in significantly reduced ATPase activity in purified RapA (23), is shown in salmon pink. Several amino acids present in RapA but not found in SsoRad54, such as Arg222, Arg221, Tyr235 and Glu233, nearly co-align at certain viewpoints along the axis pointing into the DNAs major groove. The figures were prepared using PyMOL (DeLano Scientific LLC, San Carlos, CA).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2175355&req=5

Figure 8: Predicted partial structure of the RapA NTPase/(putative) DNA-binding module. RapAs amino acid sequence was threaded into the SsoRad54–DNA structure via the SWISS-MODEL Protein Modeling Server (32–34). (A) Homology between SWI/SNF subdomain ‘I’ of RapA and that of the Sulfolobus solfataricus Rad54 homolog (SsoRad54). Identical and homologous amino acids are marked by, respectively, red and blue boxes. A limited sequence homology between SWI/SNF subdomain ‘I’ of RapA and an RNA-binding, or ‘S1’-module of the ribosomal protein S1 is also indicated (dashes, colons). The RapA SWI/SNF subdomains are shown schematically at the top; the aligned segments represent the section that yielded the homology model shown below. (B and C) Predicted structure of the RapA NTPase/(putative) DNA-binding module (sticks and mesh shown in color) superimposed with the homologous domain of SsoRad54 (gray mesh) bound to DNA (shown as CPK spheres), as reported by the Hopfner group (24). (D–G) Predicted partial structure of the RapA NTPase module (shown in color) superimposed with the complete SsoRad54 ATPase core–DNA complex (shown in gray scale) (24). Key amino acid changes in the RapA NTPase module relative to that of SsoRad54 are highlighted as CPK spheres. Lys183 (a highly conserved residue present in both proteins), alanine substitution of which results in significantly reduced ATPase activity in purified RapA (23), is shown in salmon pink. Several amino acids present in RapA but not found in SsoRad54, such as Arg222, Arg221, Tyr235 and Glu233, nearly co-align at certain viewpoints along the axis pointing into the DNAs major groove. The figures were prepared using PyMOL (DeLano Scientific LLC, San Carlos, CA).
Mentions: This model was considered in our earlier study (23), and it was based on the speculation that RNA polymerase ‘trapped’ on DNA cannot disengage from it in order to reinitiate new cycles of transcription efficiently. If RapA were indeed to promote transcriptional cycling by displacing RNA polymerase from DNA, under most circumstances [the exception being a nearly instantaneous transfer of the polymerase from the terminator to the promoter section in the DNA template, which, at least in theory, cannot be entirely ruled out due to possible co-alignment of these two sections in supercoiled DNA (see Ref. 23; Figure 8 therein)] this should be accompanied by a measurable increase in the fraction of free RNA polymerase in the system. We tested this possibility using PAGE- and ultracentrifugation-based techniques. In the first, PAGE-based assay, in vitro transcription reactions carried out to stationary phase (with or without RapA present) were fractionated on non-denaturing 5% polyacrylamide gels in the presence of magnesium, and the amounts and subunit composition of the DNA-bound polymerase were determined (Figure 6). These experiments showed no detectable reduction in the amount of DNA-associated RNA polymerase in the presence of RapA (Figure 6, densitogram; compare the levels of the large RNA polymerase subunits in reactions with or without RapA). Also, this set of experiments showed that >85% of RapA dissociated from the DNA-bound RNA polymerase, while >50% of the sigma70 subunit was retained by the DNA-bound enzyme (Figure 6, graph). In the second approach, in vitro transcription reactions were subjected to ultracentrifugation in order to determine the fractions of DNA-bound and free RNA polymerase in reactions with or without RapA. Similarly, these experiments showed no effect of excess RapA on the ratio of free and DNA-bound RNA polymerase in the system (data not shown). Furthermore, a number of other independent experiments, which assessed the amount of DNA-associated RNA polymerase in reactions with or without RapA consistently showed no effect of RapA on the amount of DNA-associated polymerase (for example, see Figure 4 above).Figure 6.

Bottom Line: Specifically, the formation of stable RapA-RNA intermediates in transcription and other, independent lines of evidence presented herein indicate that RapA binds and remodels RNA during transcription.Our results are consistent with RapA promoting RNA release from DNA-RNA polymerase-RNA ternary complexes; this process may be accompanied by the destabilization of non-canonical DNA-RNA complexes (putative DNA-RNA triplexes).Taken together, our data indicate a novel RNA remodeling activity for RapA, a representative of the SWI/SNF protein superfamily.

View Article: PubMed Central - PubMed

Affiliation: Laboratory of Biochemistry, Department of Chemistry, Lamar University, Beaumont, TX 77710, USA.

ABSTRACT
Helicase-like SWI/SNF proteins are present in organisms belonging to distant kingdoms from bacteria to humans, indicating that they perform a very basic and ubiquitous form of nucleic acid management; current studies associate the activity of SWI/SNF proteins with remodeling of DNA and DNA-protein complexes. The bacterial SWI/SNF homolog RapA-an integral part of the Escherichia coli RNA polymerase complex-has been implicated in remodeling post-termination DNA-RNA polymerase-RNA ternary complexes (PTC), however its explicit nucleic acid substrates and mechanism remain elusive. Our work presents evidence indicating that RNA is a key substrate of RapA. Specifically, the formation of stable RapA-RNA intermediates in transcription and other, independent lines of evidence presented herein indicate that RapA binds and remodels RNA during transcription. Our results are consistent with RapA promoting RNA release from DNA-RNA polymerase-RNA ternary complexes; this process may be accompanied by the destabilization of non-canonical DNA-RNA complexes (putative DNA-RNA triplexes). Taken together, our data indicate a novel RNA remodeling activity for RapA, a representative of the SWI/SNF protein superfamily.

Show MeSH
Related in: MedlinePlus