Logo unite MMSB


Sur ce site

Sur le Web du CNRS

Accueil du site > Équipes > ATPase/GTPase bactériennes : résistance aux antibiotiques et nouvelles enzymes. (JM Jault) > Bacterial nucleotide-binding proteins : resistance to antibiotics and new targets (JM Jault)

Bacterial nucleotide-binding proteins : resistance to antibiotics and new targets (JM Jault)

We work primarily on bacterial proteins with two main goals :
- To understand the functioning mechanism of transporters involved in multiple antibiotic resistance (known as multidrug transporters),
- To unravel the role of new enzymes, essential for bacterial growth and absent in eukaryotes, that might be targeted for the search of new antibiotics.

  • Multidrug ABC transporters

Resistance of microbes to conventional treatments poses a huge threat to public health and one of the major mechanisms exploited by bacteria to escape antibiotics is to use membrane pumps capable to expel continuously the molecules outside the cell. Some of these pumps recognize a wide variety of structurally unrelated compounds and are therefore capable to confer a multidrug resistance (MDR) phenotype. Interestingly, similar pumps are found in eukaryotic pathogens such as parasites or yeasts and are responsible for resistance towards antiparasitic or antifungal compounds. Moreover, in some cancers, related human transporters recognize and expel anticancerous molecules outside the cell and are thus responsible for chimiotherapy failures. All these related MDR transporters belong to the largest transporter superfamily, found in all kingdoms of life, the ABC (‘ATP-Binding Cassette’) transporters. These transporters are able to translocate, either as importers or exporters, a huge array of compounds including sugars, amino acids, ions, hydrophobic organic molecules of unrelated structure and even large proteins (> 100 kDa). In human, the aberrant functioning of several ABC transporters leads to severe genetic diseases such as cystic fibrosis, adrenoleukodystrophy or congenital hyperinsulinism.
In contrast to most ABC transporters that are rather specific for one (or a class of) substrate(s), multidrug transporters expel many structurally unrelated compounds and thus show a polyspecificity. All ABC transporters share a similar topology with four domains, two transmembrane and two cytosolic, either borne by the same polypeptide or found on separate polypeptides (up to four ; Fig. 1). ATP hydrolysis, by a mechanism presumably common to all ABC transporters, drives the opening of the transmembrane domain thereby leading to substrate translocation.

Figure 1. Modular structure of ABC transporters . From left to right, oligopeptide permease, histidine permease, the vitamin B12 importer, MsbA : the lipid A exporter, glycoprotein-P which expels many drugs used in chimiotherapies and CFTR, whose mutations cause cystic fibrosis. Importers have a supplementary subunit involved in substrate capture outside of the cell (subunit A, J and F).

We study the functioning mechanism of bacterial MDR ABC transporters. The first one, BmrA (for ‘Bacillus multidrug resistance ATP’) is constitutively expressed in Bacillus subtilis and was overexpressed in E. coli. After purification, its 3D structure was determined at low resolution and in a membrane environment (23Å) by cryo-electron microscopy (coll. with Daniel Lévy, Institut Curie, Paris). BmrA associates into oligomers to form two types of supramolecular structures as a double precision ring containing a total of 24 or 39 homodimers (Fig. 2). Within these dimers, BmrA adopts an open conformation in the resting state with its two NBDs physically disconnected. This shows that multidrug ABC transporters are indeed capable to adopt an open conformation in a lipidic bilayer. Although this supramolecular structure is purely artificial, it allowed us to monitor how effectors can affect its stability. Hence, binding of ATP-Mg which induces the dimerisation of the two NBDs, produces a conformational change large enough to destroy the rings. This result is consistent with the hypothesis that the ATP-induced NBDs dimerization generate enough energy to allow drug transport.

Figure 2. 3D model of a BmrA ring in a lipidic environment obtained by cryo-electron microscopy, and manual fitting with 3D structures of domains from ABC transporters. Top left panel, one ring is actually made of two stacked rings, each containing 12 BmrA homodimers ; two homodimers are indicated by 1,2 and 1’,2’. The three domains of BmrA are colored in white (NBD), in green (intracellular domain) and in bleu (TMD). The black bar represents 5 nm. Bottom left panel, the envelope of a BmrA homodimer was fitted with the 3D structure of the mouse MDR transporter (P-gp). Right panels, a ring seen from above with two NBD from a BmrA homodimer fitted with the NBD of the P-gp with the same orientation as in the left panel (top) or with tow NBD from Tap 1 (bottom) in a different orientation (coll. D. Lévy, Institut Curie, Paris).

Besides, Hydrogen/Deuterium exchange coupled to mass spectrometry have shown that in the resting state (or open conformation), BmrA is highly flexible and can sample multiple conformations whereas in the ATP-bound state (or closed conformation), BmrA is quite rigid. This intrinsic flexibility in the resting state might allow the transporter to recognize a wide variety of molecules to efflux (Fig. 3).

Figure 3. H/D exchange of peptides using BmrA 3-D models. 3-D models of BmrA in the open conformation (left side) or in the closed conformation (right side). One subunit was drawn in wheat colour and the other in light grey. Identified peptides are drawn with rainbow colours (in one monomer and in all ICDs) according to their percentage of deuterium exchange after 3600 s (scale of exchange shown on the right in %).

More recently, we started the characterization of new putative multidrug transporters. In contrast to BmrA, these transporters work as heterodimers leading to a structural and thus functional asymmetry (the two ATP-binding sites are non equivalent). In Bacillus subtilis, the BmrC/BmrD transporter is capable to expel multiple drugs and its synthesis is induced by many antibiotics. A homologous transporter found in Streptococcus pneumoniae is responsible for the resistance towards fluoroquinolones in vitro (coll. Thierry Vernet, IBS, Grenoble) but also in humans who are infected by this pathogen. 

Finally, we initiated the study of a new ABC transporter of Streptococcus pneumoniae which works together with a two-component system (histidine kinase plus response regulator) to confer a resistance towards several antimicrobial peptides (coll. with Thierry Vernet and Franck Fieschi teams, IBS, Grenoble). We seek to identify the physiological role of this transporter and its putative link to virulence during infection. 

Recent or ongoing collaborations on ABC transporters :
- Thierry Vernet and Franck Fieschi, IBS, Grenoble
- Pierre Falson and Attilio Di Pietro, BMSSI, Lyon
- Carole Gardiennet and Anja Böckmann, BMSSI, Lyon
- Daniel Lévy, Institut Curie, Paris,
- David Cafiso, Univ. of Virginia, Charlottesville, USA

  • New bacterial ATPases/GTPases

Resistance towards conventional antibiotics reaches nowadays an alarming level. To strike back, one solution is to identify new enzymes which are essential to bacterial growth that might be targeted by new inhibitors.
We initiated a study of three of these new families of bacterial enzymes, which are prevalent in most, if not all, pathogenic species. Two are GTPases (called YphC and YsxC in Bacillus subtilis), and we have shown (coll. Robert Britton, Michigan State University, USA), that they are very likely involved in the biogenesis of the large ribosomal subunit (50S ; Fig. 4).

Figure 4. Role of YsxC and YphC in ribosome biogenesis. Both GTPases, YsxC and YphC, are very likely involved in bacterial ribosome biogenesis because a depletion of either enzyme leads to a ‘crippled’ ribosome with an abnormal size of the large subunit (about 45S instead of 50S). This crippled ribosome tends to fall apart into its two constitutive subunits, 45S and 30S.

In the case of YsxC, we identified some of its protein partners that belong to the 50S ribosomal subunit. Regarding YphC, it is a very special GTPase because it bears two GTP-binding domains in tandem. We recently obtained two high resolution 3D structures of YphC allowing us to hypothesize the role of each GTP-binding domain in the molecular mechanism of this peculiar enzyme (coll. with Dominique Housset, IBS, Grenoble). Finally, the third enzyme family (YdiB in B. subtilis) is a new class of ATPases, of essential yet unknown function, and which has the ability to assemble into homo-oligomers both in vitro and in vivo (coll E. Brown, McMaster Univ., Canada). We are currently trying to identify partners of this new family of enzymes that might give us some clues about the physiological role of these enzymes in vivo (coll. Christophe Grangeasse, BMSSI, Lyon). 

Recent or ongoing collaborations on new bacterial enzymes :
- Guy Schoehn, IBS, Grenoble
- Christophe Grangeasse, IBCP, Lyon
- Anne Galinier, LCB, Marseille
- Robert Britton, Michigan State University, USA
- Eric Brown, McMaster University, Canada