English   español  
Por favor, use este identificador para citar o enlazar a este item: http://hdl.handle.net/10261/141891
Compartir / Impacto:
Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL

Inhibidores sintéticos de FtsZ con actividad antibacteriana

AutorRuiz-Avila, Laura B.
DirectorAndreu, José Manuel ; Huecas, Sonia
Palabras claveDivisión celular bacteriana
Ensamblaje de proteínas
Bacterial cell division
Protein assembly
Fecha de publicación2014
EditorCSIC - Centro de Investigaciones Biológicas (CIB)
Universidad Complutense de Madrid
ResumenCell division is one of the most fundamental biological processes, essential for the propagation of living species. Cell division is carried out with the participation of cytoskeleton or cytomotive proteins. The eukaryotic cytoskeleton is composed of three types of cytoskeletal fibers formed by actin, intermediate filaments and tubulin. Prokaryotic cells have cytomotive homologues of each of these proteins and lack motor proteins. Thus, MreB, ParM and FtsA (van den Ent et al., 2001a; van den Ent et al., 2001b; van den Ent and Lowe, 2000) were described as bacterial proteins with a fold similar to actin. Crescentin (Ausmees et al., 2003) was found in Caulobacter crescentus bacteria and is homologue of intermediate filaments. The essential bacterial cell division protein FtsZ, bacterial tubulin BtubA/B and TubZ are the three prokaryotic tubulinhomologues. The aim in this Thesis is the assembly of FtsZ as a target for the discovery of new antibiotics. This protein was identified by the isolation of mutations that causes the formation of filaments at a nonpermisive temperature andit isfor that reason that is called FtsZ (Filamentous temperature sensitive Z). FtsZ is the most conserved bacterial cell division protein and it is also present in some plastids and mitochondria of several groups of the Eukarya (Margolin, 2005). FtsZ is the first known protein to localize to the midcell, where it polymerizes into a ring-like structure known as Z-ring (Bi et al., 1991) that marks the position of the future division site. It is still unclear how the Z-ring is built, but it has been proposedthat it is formed by the association of short FtsZ protofilaments. The Z-ring persists during the division guiding the synthesis and location of the division septum and it serves as a scaffold for the recruitment of other proteins involves in the bacterial cell division process, which all together form the divisome.
The function of many of the proteins that are involved in bacterial cell division is not fully understood; however, their roles involve i) assisting in the formation and stabilization of the Zring, ii) permitting the correct DNA replication and avoiding the formation of Z-ring over the nucleoid, iii) recruiting and stabilizing the divisome at the division site and iv) directing the synthesis of peptidoglycan in order to form the division septum (Harry et al., 2006). The position of Z-ring isimportant to prevent its assembly at inappropriate locations and to permit the correct chromosome replication and segregation. There are three systems that influence the position of Z-ring: the MINCDE/DivIVA system in Escherichia coli and Bacillus subtilis, the nucleoid occlusion guiding by SlmA and Noc in E.coli and B.subtilis respectively,and the regulation by the protein MipZ in C. crescentus. FtsZ and tubulin are thought to share an ancestral homologue. Both proteins have the same fold (Lowe and Amos, 1998; Nogales et al., 1998) despite of the low overall amino acid identity. FtsZ and tubulin are GTPases and the residues required for GTP binding and hydrolysis are those which show the highest degree of conservation. FtsZ is composed of two globular subdomains which are independently folded. The two subdomains are connected by a long helix, H7, which is sandwiched between them (Figure A). FtsZ is a self-activating GTPase and the hydrolysis of GTP depends on the polymerization of the protein.The GTPase catalytic site is formed during filament assembly. For polymerization, the T7 loop in the “upper” subunit is inserted into the nucleotide binding pocket of the “lower” subunit where the catalytic residues activate GTP hydrolysis (Lowe and Amos, 1999). For these reasons the assembly and dynamic behavior of FtsZ polymers is driven by guanosine triphosphate (GTP) binding and hydrolysis. Figure A. Schematic representation of FtsZ-ring in a bacterial cell (on the left) and cristal structure of the complex FtsZ from Staphylococcus aureus with GDP and PC190723 (on the rigth).Is shown N-terminal domain (cyan), Cterminal domain (blue) and H7 helix (grey). Ligands GDP and PC190723 are shown in green.
FtsZ assembles into polar tubulin-like one-stranded protofilaments (Huecas and Andreu, 2003; Huecas et al., 2008; Romberg et al., 2001). Monomers of protein can exist in two states: the inactive state with low affinity to form polymers and active state with high affinity for polymerization (Chen and Erickson, 2011; Huecas et al., 2008; Martin-Galiano et al., 2010). The inactive form of FtsZ monomers has been observed in most crystal structures of FtsZ from different species (Oliva et al., 2007). The crystal structure of an FtsZ protofilament (Elsen et al., 2012; Matsui et al., 2012; Tan et al., 2012) has recently shown what has been proposed to be the active conformation, and possibly the conformational changes between both FtsZ states. The important role of FtsZ in bacterial cell division, and the fact that currently no available antibiotics specifically inhibit the bacterial division, have made FtsZ an attractive target for antibacterial drug discovery (Sass and Brotz-Oesterhelt, 2013; Vollmer, 2006). The emergence of bacterial resistance to antibiotics is a major health problem and, therefore, it is critical to develop new antibiotics with novel modes of action for effective treatments against new multidrug resistant pathogens such as methicillin-resistant Staphylococcus aureus (MRSA). Several small-molecule inhibitors have been reported to modulate the assembly/disassembly dynamics of FtsZ, some of which showed important antibacterial activity against important pathogens and were efficacious in in vivo models of infection (Schaffner-Barbero et al., 2012). The described compounds that target FtsZ have different origins: high-throughput screening (HTS) of chemical libraries, natural product discovery, previously known antibacterial compounds, compounds from tubulin field and synthetic compounds specifically developed to inhibit FtsZ. Some of these compounds are: difluoro-benzamide derivatives such as PC190723 (Haydon et al., 2008) and 8J (Adams et al., 2011), viriditoxin (Wang et al., 2003a), totarol (Jaiswal et al., 2007), zantrins (Margalit et al., 2004), PC170942 and PC58538 (Stokes et al., 2005), chrysophaentin A (Plaza et al., 2010)and its fragment hemi-chrysophaentin (Keffer et al., 2013), cinnamaldehyde (Domadia et al., 2008), sanguinarine (Beuria et al., 2005), C8-GTP analogs (Lappchen et al., 2008) and amikacin (Possoz et al., 2007). However, is still unknown where these compounds bind to FtsZ except the GTP analogs, chrysophaentin A and its fragment, which bind to the nucleotide site of FtsZ (Keffer et al., 2013) and PC190723, which binds between the N- and C-terminal domains of FtsZ.
The main cavities available for ligand binding in FtsZ monomer are the nucleotide-binding cup in the N-terminal domain and the long cleft between N- and C-terminal domains, known as PC190723 binding site (Figure A). The nucleotide binding site in FtsZ is conserved among FtsZs from different organisms (Oliva et al., 2007). Some compounds targeting the nucleotide site have similar chemical structure to the nucleotide, such as the C8-GTP derivatives that selectively inhibit FtsZ but promote tubulin assembly (Lappchen et al., 2008), while others have different chemical structures such as PC170942 (Stokes et al., 2005). In this Thesis, the effects on the functional activity of FtsZ of both types of ligands have been examined. The discovery of the PC190723 binding site in FtsZ is recent(Haydon et al., 2008) andthe crystal structure of FtsZ with bound PC190723 was made available last year (Elsen et al., 2012; Tan et al., 2012 and Matsui et al., 2012). Few compounds that bind to this site have been reported, all of them derivatives of PC190723. In this work we have characterized the effect of PC190723, its fragments and an analog (8J) on the assembly and GTPase activity of FtsZ. The objectives in this Thesis were i) to find new compounds targeting the nucleotide binding site in the protein and ii) to explore the PC190723 binding site in FtsZ. To our first purpose, an in-depth study of the mechanism of action of the C8 substituted GTP analogs on the assembly of FtsZ from several species were carried out. We also were interested in searching for small molecules structurally different to GTP that could replace the nucleotide in FtsZ and with antimicrobial activity. Once this goal was achieved, we attempted to optimize the compound’s chemical structure to obtain ligands with higher affinity for our target and a better antimicrobial activity. Regarding the second objective, the exploration of the PC190723 binding site in FtsZ, we had this synthetic ligand, its fragments difluorometoxibenzamide (DFMBA) and cloro-tihazol-pyridine (CTPM), and an analog (8J) available. The goal was the biochemical characterization of these compounds on the assembly and GTPase activity of FtsZ. On the other hand, we also were interested in the development of a fluorescence probe to detect and characterize novel compounds that bind to the PC190723 site in FtsZ. The approaches used in this worktowards these objectives are summarized inFigure B.Once we had the compounds to be tested, we proceeded with two main controls: determination of its solubility in our experimental buffer and a register of fluorescence to discard those compounds which could interfere with mant-GTP signal. In the case of the small-molecules supposed to bind to the nucleotide site of FtsZ we first detected indirectly their binding using a competition assay with mant-GTP, which allows us to determine the competitor binding affinity (Kb). For those compounds we were more interested in, we confirmed their binding affinities with another competition assay using the radioactive ligand 3H-GTP. For direct detection and characterization of ligands binding, we performed different assays such as analytic ultracentrifugation, difference absorption spectroscopy, nuclear magnetic resonance and computational experiments (docking and molecular dynamics).
We use diverse methods to study the effect of compounds on the assembly of FtsZ and to check for possible cross-effects on tubulin. By measuringthe right angle light scattering and small sample ultracentrifugation we observed the effect of compounds in the polymerization of the protein, and electron microscopy gave us qualitative information of polymer structure. Knowing that GTPase activity depends on polymerization it was also important to determinate the rate of GTP hydrolysis in the presence of the ligands.Finally, we used microbiological methods to study the effect of inhibitors on FtsZ in vivo such as the antibacterial spectrum and potency of a compound, determining its minimal inhibitory concentration (MIC) and observing the presence of cell filamentation or FtsZ delocalization. We also tested FtsZ inhibitors for possible cross-effect on microtubule assembly on human cell lines. For those compounds targeting PC190723 binding site, we studied their effect on the assembly of FtsZ as above. Since we were interested in the development of a fluorescence assay to detect and characterize novel compounds that bind to the PC190723 site in FtsZ we measured fluorescense intensity and anisotropy of fluorescent PC derivatives to detect any changes between their free and FtsZ-bound forms.
The results of this Thesis are divided in two main sections. In the first section we analyze the effects of compounds targeting the nucleotide binding site in FtsZ and in the second section the PC190723 binding site in FtsZ is explored. A. Inhibitors that bind to the nucleotide site of FtsZ  C8 substituted GTP analogs: We found that C8 substituted GTP analogs bind to the nucleotide site of FtsZ from a bacteria, Bacillus subtilis (Bs-FtsZ) and an archaea, Methanococcus jannaschii (Mj-FtsZ) inhibiting the assembly of both proteins. The ligands bind to nucleotide binding site with similar affinities according to the structural similarity of nucleotide pocket in FtsZ. We observed that the assembly inhibitory potencies of the C8-GTP analogs on Bs-FtsZ correlate with their binding affinities to this protein. Unexpectedly, we found a different relationship of the assembly inhibitory capacity with the binding affinity of these compounds in Mj-FtsZ. Moreover, the analog of GTP, MorphGTP does not inhibit the assembly of FtsZ from the arqueobacteria but induces an abnormal polymerization of the protein. We have analyzed by Nuclear Magnetic Resonance (NMR) STD and trNOESY methods the molecular recognition of the C8-GTP analogs by bacterial and archaeal FtsZ. The results show significant differences in the binding mode of the nucleotide and its C8-modified analog to the proteins and they also exhibit a distinct geometric conformation (anti and syn) between its forms free or bound to FtsZ from both species. By computational experiments (docking and molecular dynamic simulations) we propose models of how the C8-analogs inhibit FtsZ. C8-analogs generate significant changes in size, shape and electrostatic surface at the interface between FtsZ monomers, which probably lead to the observed inhibition of FtsZ assembly.  Non-nucleotide synthetic inhibitors: For our purpose of finding small molecules structurally different to GTP that could replace the nucleotide in FtsZ, we tested several compounds which came from: i) virtual screening of chemical libraries, ii) previously described FtsZ-interacting molecules, or iii) synthetic compounds selected from an in-house library after docking into the Bs-FtsZ GTP site. In order to detect the binding of these compounds to the GTP site and measure their affinities we used a fluorescence assay, which measure the anisotropy change of mant-GTP upon binding to nucleotide-free FtsZ monomers.
Compounds from a virtual screening campaign and several which has been reported to interact with FtsZ; amikacin, PC170942 and the polyphenolic compounds curcumin, chlorogeinc acid and caffeic acid, gave negative results in mant-GTP competition assays. Nevertheless, we observed that PC170942 (Stokes et al., 2005) inhibits FtsZ polymerization and GTPase activity. We propose this functional inhibitor of FtsZ either binds weakly to the nucleotide site of FtsZ or binds to an allosteric site. In collaboration with the Medicinal Chemistry Lab, Universidad Complutense de Madrid (UCM), we found several compounds targeting the nucleotide binding site FtsZ from their inhouse synthetic library. Most of them are polyphenolic derivatives and we identified the hit compound UCM05 (Figure C), which binds to the nucleotide site of Bs-FtsZ,Kb = 4.30 ± 0.04 · 105 M-1. Analogs of UCM05 were synthesized and two compounds that increase the binding affinity for the GTP site of Bs-FtsZ were easily found: the simplified tetra-hydroxy analog UCM44, Kb = 1.5 ± 0.3 · 106 M-1 and the chlorinated analog UCM53, Kb = 1.3 ± 0.2 · 106 M-1. We analyzed the effects of UCM05 and UCM44 on the assembly of FtsZ from three different species: Bs-FtsZ, Ec-FtsZ (FtsZ from Eschericia coli) and Mj-FtsZ. In Bs-FtsZ, the compounds inhibit the assembly of the protein in the presence of nucleotide and, without adding GTP, UCM05 y UCM44 induce the formation of disordered polymers and aggregates. By docking and molecular dynamic simulation experiments we observed UCM05 and UCM44 bind to Bs-FtsZ in two possible modes. In addition, we performed analytic ultracentrifugation assays in which we observed that ligands enhanced the formation of FtsZ dimers. Taken these results together, we propose UCM compounds are probably distorting FtsZ self-association interface leading the formation of disorder polymers but we cannot discard an allosteric binding of the ligands. In contrast to the results obtained with Bs-FtsZ, UCM05 and UCM44 did not reduce the polymerization of Ec-FtsZ, so we conclude they are weak inhibitors in FtsZ from this species. UCM05 and UCM44 induce the polymerization of Mj-FtsZ in the absence of nucleotide, forming well-ordered protofilaments of this protein different to the Mj-FtsZ filaments forming with GTP. The analysis of images of the UCM05 induced polymer is compatible with two possibilities: a pair of FtsZ protofilaments or hollow tubes formed by 5-7 monomers per turn. These results suggested that UCM05 and UCM44 replace GTP in Mj-FtsZ, inducing a structural rearrangement different to GTP, which leads the formation of tubule-like polymers.
UCM05, UCM44 and UCM53 inhibit the bacterial cell division and induce filamentation of B.subtilis cells. All of them are active against Gram-positive pathogens but havemuch less antimicrobial activity against Gram-negative bacteria. UCM53 is also active on meticillinresistat Staphylococcus aureus (MRSA), ampicillin- and levofloxacin-resistant Enterococcus faecalis. Furthermore, these inhibitors induces FtsZ delocalization with formation of numerous punctuate foci and distort the Z-ring structure in B.subtilis. Finally, we have confirmed these ligands acts specifically on FtsZ and they do not affect the formation of tubulin microtubules.  Optimization of new FtsZ inhibitors: After we had found the hit compound, UCM05, we conducted a systematic chemical structure modification of the compound to improve affinity to nucleotide binding site of FtsZ (in vitro potency) and antibacterial activity (in vivo efficacy). For this purpose we measured the binding affinities of the ligands for the nucleotide binding site of Bs-FtsZ and we determined the Minimal Inhibitory Concentration (MIC) on B.subtilis cells. In the search of new derivatives of UCM05 compounds were synthesized with modification in: i) the central core of the molecule, replacing 1,3-naphthalene by different chemical groups such as phenyl and cyclohexane rings, ii) the ester bonds used as a spacer in the molecule by other spacers such as amides, sulfonamides and double bonds and iii) In the hydroxy groups of the phenyl rings by reduction of the number, modifying of position of these groups or replacing them by other substituents such as methoxy or chlorine.
Based on the determination of binding affinity and antimicrobial activity, we have confirmed that naphthalene scaffold and ester bonds are the most favorable central core and linker in the molecule. Also position R1 and R3 present in UCM05 are the best locations for the phenyl rings on the naphthalene core. The reduction in the number of hydroxy groups on both phenyl rings improves the binding affinity comparing with UCM05. These last series of modifications have permitted us to identify the hydroxy groups in positions R1, R4 and R5 as the most important in the molecule since compounds where they are present have significantly improved Kb and MIC values. Finally, we observed that the combination of hydroxy and chlorine substituents in the same phenyl ring improves affinity (from 105 to 106 M-1) and antibacterial activity (20-fold) when comparing with UCM05.The modified ligands inhibit FtsZ assembly, leading to functional inhibition of FtsZ and blocking the bacterial cell division. They are more active against Gram-positive resistant pathogens including different strains of MRSA, Listeria monocytogenes and E.faecalis comparing with UCM05. B. Study of the PC190723 binding site in FtsZ We first determined the effect of PC190723 (PC) and its fragments on the assembly and GTPase activity of FtsZ from B.subtilis and E.coli. We confirmed that PC190723 (PC) is a stabilizer of FtsZ polymers from susceptible species (Bs-FtsZ) (Andreu et al., 2010; Haydon et al., 2008) and reduces the GTPase activity of the protein. We have also identified the benzamide fragment (DFMBA) as the principal of the molecule since retain the stabilizer effects on polymerization and reduces GTPase activity than PC in Bs-FtsZ. From several derivatives of PC190723 or DFMBA attached to different fluorophores, we found a compound NBD-DFMBA (Figure D) that specifically bind to PC190723 site in FtsZ from susceptible species. This probeincreases its anisotropy upon binding to FtsZ polymers rather than to FtsZ monomers. We have observed that the anisotropy change in the compound follows the time course polymerization of the protein. These results support the proposed structural mechanism of the FtsZ assembly switch and should facilitate screening for new antibacterial compounds binding to the PC190723 site of FtsZ.
Descripción217 p.-53 fig.-14 tab.-3 Anexos
Aparece en las colecciones: (CIB) Tesis
Ficheros en este ítem:
Fichero Descripción Tamaño Formato  
Tesis_Ruiz_Avila_Laura_2014.pdf9,4 MBAdobe PDFVista previa
Mostrar el registro completo

NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.