Limits...
Interaction sites of DivIVA and RodA from Corynebacterium glutamicum.

Sieger B, Bramkamp M - Front Microbiol (2015)

Bottom Line: Deletion of rodA had drastic effects on morphology and growth, likely a result from misregulation of penicillin-binding proteins and cell wall precursor delivery.Deletion of these 10 AAs significantly decreased the interaction efficiency with DivIVA.Our results corroborate the interaction of DivIVA and RodA, underscoring the important role of DivIVA as a spatial organizer of the elongation machinery in Corynebacterineae.

View Article: PubMed Central - PubMed

Affiliation: Biocenter - Ludwig-Maximilians-University Munich Munich, Germany.

ABSTRACT
Elongation growth in actinobacteria is localized at the cell poles. This is in contrast to many classical model organisms where insertion of new cell wall material is localized around the lateral site. We previously described a role of RodA from Corynebacterium glutamicum in apical cell growth and morphogenesis. Deletion of rodA had drastic effects on morphology and growth, likely a result from misregulation of penicillin-binding proteins and cell wall precursor delivery. We identified the interaction of RodA with the polar scaffold protein DivIVA, thus explaining subcellular localization of RodA to the cell poles. In this study, we describe this interaction in detail and map the interaction sites of DivIVA and RodA. A single amino acid residue in the N-terminal domain of DivIVA was found to be crucial for the interaction with RodA. The interaction site of RodA was mapped to its cytoplasmic, C-terminal domain, in a region encompassing the last 10 amino acids (AAs). Deletion of these 10 AAs significantly decreased the interaction efficiency with DivIVA. Our results corroborate the interaction of DivIVA and RodA, underscoring the important role of DivIVA as a spatial organizer of the elongation machinery in Corynebacterineae.

No MeSH data available.


(A) Topology model of RodA according to topology prediction (TMHMM). Scissors indicate truncation sites. The protein possesses 12 transmembrane domains and both ends are at the cytoplasmic site. (B) Fluorescence microscopy images of full length DivIVA-YFP and the two truncation mutants RodA1/2-CFP and RodA2/2-CFP. While individually expressed RodA1/2-CFP localizes to the membrane (first column), co-localization with DivIVA-YFP seems to be abolished (second column). RodA2/2-CFP lost completely its membrane localization and appears cytoplasmic (third column); however, when co-expressed with DivIVA, it is co-localized to DivIVA foci (forth column). (C) Localization of RodA-CFP, RodAΔC10-CFP, and RodAΔC80-CFP (-DY) and co-localization with DivIVA-YFP (+DY). Full length RodA co-localizes to 95% of all DivIVA foci, RodAΔC10-CFP co-localizes to 20%, and RodAΔC80-CFP does not co-localize with DivIVA-YFP. FRET measurements confirm these observations. RCY values are 1.15 for full length RodA, 1.03 for RodAΔC10-CFP, and 0.99 for RodAΔC80-CFP.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4285798&req=5

Figure 4: (A) Topology model of RodA according to topology prediction (TMHMM). Scissors indicate truncation sites. The protein possesses 12 transmembrane domains and both ends are at the cytoplasmic site. (B) Fluorescence microscopy images of full length DivIVA-YFP and the two truncation mutants RodA1/2-CFP and RodA2/2-CFP. While individually expressed RodA1/2-CFP localizes to the membrane (first column), co-localization with DivIVA-YFP seems to be abolished (second column). RodA2/2-CFP lost completely its membrane localization and appears cytoplasmic (third column); however, when co-expressed with DivIVA, it is co-localized to DivIVA foci (forth column). (C) Localization of RodA-CFP, RodAΔC10-CFP, and RodAΔC80-CFP (-DY) and co-localization with DivIVA-YFP (+DY). Full length RodA co-localizes to 95% of all DivIVA foci, RodAΔC10-CFP co-localizes to 20%, and RodAΔC80-CFP does not co-localize with DivIVA-YFP. FRET measurements confirm these observations. RCY values are 1.15 for full length RodA, 1.03 for RodAΔC10-CFP, and 0.99 for RodAΔC80-CFP.

Mentions: Next we aimed to identify the RodA interaction site with DivIVA. Figure 4A shows a topology model of RodA according to a topology prediction simulation (TMHMM; Arnold et al., 2006). The protein harbors 12 transmembrane domains and both termini are facing the cytoplasm. To identify the interaction site with DivIVA we first divided the protein into two CFP-tagged halves and expressed them individually and together with DivIVA in E. coli (Figure 4B). It turned out that the N-terminal part (RodA1/2) localized to the membrane, however, it did not co-localize with DivIVA. The C-terminal part (RodA2/2) appeared cytoplasmic, but co-localized completely with DivIVA, implicating that the interaction site must be in the C-terminal half of the protein, although the truncated protein is apparently not inserted correctly into the membrane. We then made CFP-tagged truncations of 10 and 80 AAs from the C-terminus, ensuring cytoplasmic localization of the fluorophore. Whereas >90% of full length RodA-CFP co-localized with DivIVA foci (RCY = 1.15), co-localization of RodAΔC10-CFP was reduced to approximately 20% (RCY = 1.03) and completely abolished for RodAΔC80-CFP (RCY = 0.99; Figure 4C). Apparently, the C-terminal 10 AAs contribute to the RodA–DivIVA interaction. We finally tested several point mutations in the C-terminal domain of RodA. We reasoned that maybe a negatively charged residue might interact with K20 that we identified within DivIVA to be responsible for interaction. In spite of this, the variant RodAE438G did not abolish interaction (RCY = 1.10). RodAmut10C-CFP, a strain where all C-terminal 10 AAs of RodA were mutated into 10 AAs with similar residues (K → R, Q → N, A → G, RCY = 1.17, Figure 5) preserved the interaction. However, point mutants K434G, Q435G, and the double mutant S433G–S437G decreased interaction with DivIVA (RCY = 1.04, 1.04, and 1.05), implicating an essential role of these four AAs in DivIVA–RodA interaction (Figure 5).


Interaction sites of DivIVA and RodA from Corynebacterium glutamicum.

Sieger B, Bramkamp M - Front Microbiol (2015)

(A) Topology model of RodA according to topology prediction (TMHMM). Scissors indicate truncation sites. The protein possesses 12 transmembrane domains and both ends are at the cytoplasmic site. (B) Fluorescence microscopy images of full length DivIVA-YFP and the two truncation mutants RodA1/2-CFP and RodA2/2-CFP. While individually expressed RodA1/2-CFP localizes to the membrane (first column), co-localization with DivIVA-YFP seems to be abolished (second column). RodA2/2-CFP lost completely its membrane localization and appears cytoplasmic (third column); however, when co-expressed with DivIVA, it is co-localized to DivIVA foci (forth column). (C) Localization of RodA-CFP, RodAΔC10-CFP, and RodAΔC80-CFP (-DY) and co-localization with DivIVA-YFP (+DY). Full length RodA co-localizes to 95% of all DivIVA foci, RodAΔC10-CFP co-localizes to 20%, and RodAΔC80-CFP does not co-localize with DivIVA-YFP. FRET measurements confirm these observations. RCY values are 1.15 for full length RodA, 1.03 for RodAΔC10-CFP, and 0.99 for RodAΔC80-CFP.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4285798&req=5

Figure 4: (A) Topology model of RodA according to topology prediction (TMHMM). Scissors indicate truncation sites. The protein possesses 12 transmembrane domains and both ends are at the cytoplasmic site. (B) Fluorescence microscopy images of full length DivIVA-YFP and the two truncation mutants RodA1/2-CFP and RodA2/2-CFP. While individually expressed RodA1/2-CFP localizes to the membrane (first column), co-localization with DivIVA-YFP seems to be abolished (second column). RodA2/2-CFP lost completely its membrane localization and appears cytoplasmic (third column); however, when co-expressed with DivIVA, it is co-localized to DivIVA foci (forth column). (C) Localization of RodA-CFP, RodAΔC10-CFP, and RodAΔC80-CFP (-DY) and co-localization with DivIVA-YFP (+DY). Full length RodA co-localizes to 95% of all DivIVA foci, RodAΔC10-CFP co-localizes to 20%, and RodAΔC80-CFP does not co-localize with DivIVA-YFP. FRET measurements confirm these observations. RCY values are 1.15 for full length RodA, 1.03 for RodAΔC10-CFP, and 0.99 for RodAΔC80-CFP.
Mentions: Next we aimed to identify the RodA interaction site with DivIVA. Figure 4A shows a topology model of RodA according to a topology prediction simulation (TMHMM; Arnold et al., 2006). The protein harbors 12 transmembrane domains and both termini are facing the cytoplasm. To identify the interaction site with DivIVA we first divided the protein into two CFP-tagged halves and expressed them individually and together with DivIVA in E. coli (Figure 4B). It turned out that the N-terminal part (RodA1/2) localized to the membrane, however, it did not co-localize with DivIVA. The C-terminal part (RodA2/2) appeared cytoplasmic, but co-localized completely with DivIVA, implicating that the interaction site must be in the C-terminal half of the protein, although the truncated protein is apparently not inserted correctly into the membrane. We then made CFP-tagged truncations of 10 and 80 AAs from the C-terminus, ensuring cytoplasmic localization of the fluorophore. Whereas >90% of full length RodA-CFP co-localized with DivIVA foci (RCY = 1.15), co-localization of RodAΔC10-CFP was reduced to approximately 20% (RCY = 1.03) and completely abolished for RodAΔC80-CFP (RCY = 0.99; Figure 4C). Apparently, the C-terminal 10 AAs contribute to the RodA–DivIVA interaction. We finally tested several point mutations in the C-terminal domain of RodA. We reasoned that maybe a negatively charged residue might interact with K20 that we identified within DivIVA to be responsible for interaction. In spite of this, the variant RodAE438G did not abolish interaction (RCY = 1.10). RodAmut10C-CFP, a strain where all C-terminal 10 AAs of RodA were mutated into 10 AAs with similar residues (K → R, Q → N, A → G, RCY = 1.17, Figure 5) preserved the interaction. However, point mutants K434G, Q435G, and the double mutant S433G–S437G decreased interaction with DivIVA (RCY = 1.04, 1.04, and 1.05), implicating an essential role of these four AAs in DivIVA–RodA interaction (Figure 5).

Bottom Line: Deletion of rodA had drastic effects on morphology and growth, likely a result from misregulation of penicillin-binding proteins and cell wall precursor delivery.Deletion of these 10 AAs significantly decreased the interaction efficiency with DivIVA.Our results corroborate the interaction of DivIVA and RodA, underscoring the important role of DivIVA as a spatial organizer of the elongation machinery in Corynebacterineae.

View Article: PubMed Central - PubMed

Affiliation: Biocenter - Ludwig-Maximilians-University Munich Munich, Germany.

ABSTRACT
Elongation growth in actinobacteria is localized at the cell poles. This is in contrast to many classical model organisms where insertion of new cell wall material is localized around the lateral site. We previously described a role of RodA from Corynebacterium glutamicum in apical cell growth and morphogenesis. Deletion of rodA had drastic effects on morphology and growth, likely a result from misregulation of penicillin-binding proteins and cell wall precursor delivery. We identified the interaction of RodA with the polar scaffold protein DivIVA, thus explaining subcellular localization of RodA to the cell poles. In this study, we describe this interaction in detail and map the interaction sites of DivIVA and RodA. A single amino acid residue in the N-terminal domain of DivIVA was found to be crucial for the interaction with RodA. The interaction site of RodA was mapped to its cytoplasmic, C-terminal domain, in a region encompassing the last 10 amino acids (AAs). Deletion of these 10 AAs significantly decreased the interaction efficiency with DivIVA. Our results corroborate the interaction of DivIVA and RodA, underscoring the important role of DivIVA as a spatial organizer of the elongation machinery in Corynebacterineae.

No MeSH data available.