Diseño y aplicaciones biomédicas de biomateriales funcionalizados basados en poliésteres de origen bacteriano

Abstract

Polyhydroxyalkanoates (PHA), commonly known as “bioplastic", are biodegradable polymers accumulated by certain bacteria in the form of intracellular inclusions. These biopolymers are synthesized from renewable sources, can be biodegraded under controlled conditions and have similar physico-chemical characteristics to plastic materials derived from the petrochemical industry (Rehm, 2010). Considering the fact that they are biodegradable, biocompatible and non toxic, these compounds have great potential for multiple applications, such as biomaterials. Synthesis of new PHA is of great interest from both the industrial and biomedical standpoint, as the novel biopolymers and monomers that constitute them present physico-chemical and mechanical properties that differ from conventional PHA. Implant-associated infections and the alarming rise in antibiotic resistance highlight the need to generate new antimicrobial surfaces to develop biomaterials and tissueengineered products. Bacterial polyesters, PHA, have become one of the leading biomaterials under investigation due to their biocompatible and absorbable features. Among other properties, one of the characteristics required for clinical or alimentary use is their capacity to act as bacteriostatic or bactericidal polymers, so designed packaging materials or devices stop bacterial proliferation (bacteriostatic property) or even cause bacterial death (bactericidal property). Antimicrobial polymers represent a class of biocides of increasing importance as an alternative to existing biocides, and in some cases even antibiotics (Siedenbiedel and Tiller, 2012). Bearing all this in mind, in this PhD Thesis two different polymer functionalization strategies were followed to obtain added-value PHA for industrial and biomedical applications. The first strategy involved in vivo bacterial production of tailor-made functionalized nano-beads, where proteins attached to the natural PHA granule were engineered to display fusion proteins of interest. Such PHA granules can be utilized as nano-beads to immobilize recombinant proteins using the BioF system previously designed in our laboratory. This system is based on the use of the N-terminal domain of Pseudomonas putida KT2440 PhaF phasin as affinity tag, which enables the recombinant protein to anchor to the PHA granule (Moldes et al., 2004). One of the drawbacks of the BioF system is its low yield, which is due to the fact that not all fusion proteins are attached to the granule and that an important part of the granule surface is occupied by natural phasins. This system has been improved by studying the process of tag binding to the PHA granule and identifying the minimal necessary factors that drive PHA granule formation and segregation between daughter cells during cell division. The second approach is based on the use of metabolic engineering to design bacterial strains able to produce new non-natural polyesters (Escapa et al., 2011). We studied the properties of the new PHA family obtained, named PHACOS (poly(3- hydroxyacilthioalkanoate-co-[R]-3-hydroxyalkanoate)), carrying functionalized groups in the side chain.

Results and discussion To optimize the BioF system for in vivo immobilization of proteins to PHA granules, we studied physiological function of PhaF phasin, determining the factors that influence the granule segregation process during cell division, as well as granule intracellular localization. We identified two different functions of the C- and N-terminal domain of this protein. It was demonstrated that C-terminal contains AAKP-like tandem repeats, characteristic of the histone H1-family, and binds DNA in a nonspecific manner, whereas N-terminal domain is responsible for granule binding and shares high sequence similarity with PhaI phasin. Transmission electronic microscopy (TEM), confocal microscopy and multidimensional microscopy revealed that, in the cells producing granules at early growth stage, PhaF directs the PHA granules to the center of the P. putida KT2440 cells, forming a characteristic needle array. Moreover, by means of flow cytometry studies, we demonstrated the presence of two markedly different cell populations in the strain lacking the PhaF protein, i.e., cells with and without PHA. Complementation studies definitively demonstrated a key role of PhaF in granule segregation during cell division, ensuring the equal distribution of granules between daughter cells. Confocal microscopy in vivo monitoring of green fluorescent protein (GFP) and C-terminal domain of PhaF fusion protein showed that C-terminal binds nucleoid. All these findings suggested PhaF plays a major role in the PHA apparatus through interactions with the segregating chromosome. In this PhD Thesis, we have also studied the involvement of the other P. putida KT2440 phasin, PhaI protein, in the PHA machinery, as PhaI phasin and BioF domain of PhaF share a similar primary structure. We demonstrated the cooperative work of phasin domains (PhaI and N-terminal of PhaF) to obtain optimal PHA granule formation and distribution. On the one hand, we studied the production of recombinant protein under different growth conditions, both in P. putida wild type and phasin mutant strains. The said mutant strains were constructed by gene deletion using the pK18mobsacB plasmid. This system represents an environmentally friendly approach as directed genetic deletions are produced without incorporating antibiotic markers. On the other hand, we studied the localization of phasin proteins fused with GFP in the presence and absence of PHA granules. These constructions were performed using the pCNB5 plasmid, which carries a mini-transposon facilitating stable insertion of genetic construction, such as monocopy in the P. putida chromosome. These experiments enabled us to determine the optimal conditions and host strain for in vivo production of bioactive nano-beads. Studies of phasin-mutant P. putida strains producing BioF-GFP recombinant protein performed by epifluorescent microscopy and flow cytometry demonstrated that balanced distribution after cell division is recovered when PhaF, and BioF or PhaI are produced concurrently, thereby showing the swappable nature of phasin domains. By means of Western blot analysis, we confirmed that low concentrations of natural PhaF phasin are sufficient for proper granule segregation between daughter cells during cell division. All these findings enabled the construction and design of an optimal host strain for BioF system expression, derived from the model P. putida KT2440 strain with minimal genetic modifications, making it ideal for bioactive nano-bead production. Quantification of the recombinant protein was performed by fluorimetry and flow cytometry.

Taken together, this system provides a useful tool for in vivo immobilization of active proteins to a biodegradable support, with many potential applications. Applying a second strategy, previously described by Escapa et al., (2011), functionalized biomaterials called PHACOS were synthesized. In this PhD Thesis, we focused on studying the new properties of this non-natural polyester carrying functionalized groups at the side chain. There have been no previous reports of the antibacterial activity of PHA polymers that have not been modified or mixed with other compounds for that purpose. However, it is known that PHA-derived monomers, [R]-3- hydroxyalkanoates (3HA), show antibacterial activity against Stapylococcus aureus (Ruth et al., 2007) with antibacterial efficiency of 3HA showing very high MIC values (Minimal Inhibitory Concentration), 1-5 mM. The antibacterial activity of PHACOS and non-functionalized control polymer PHO against several Gram-positive and Gram-negative bacterial strains was determined following ISO 22196:2011 “Measurement of Antibacterial Activity on Plastics Surfaces” standard protocol, with certain modifications. The results demonstrated that PHACOS inhibit growth of S. aureus, including three methicilline resistant (MRSA) clinical isolates, showing less than 10% bacterial cell survival when compared to the control. Moreover, we studied the ability of PHACOS to prevent bacterial biofilm formation on its surface in comparison to other materials. To that end, S. aureusT and Pseudomonas aeruginosa CECT 4122 were used as model strains able to form biofilm on different materials. The ability of bacteria to form biofilm in vitro on polymer surfaces was examined by: (i) environmental scanning electron microscopy (ESEM); (ii) crystal violet assay and (iii) colony forming unit (CFU) counting. Results show that biofilm formation of S. aureus decreased 2-fold in the presence of PHACOS as compared to biofilm formation on the control polymer; however, this was not the case for P. aeruginosa. Furthermore, we studied the ability of PHACOS to kill bacteria adhered to its surface, in comparison with other materials. The number of viable bacterial cells adhering to the surface of biopolymers was determined by fluorescence microscopy using the Bacterial Viability test LIVE/DEAD BacLightTM (Invitrogen L13152). Interestingly, results demonstrated that the number of viable S. aureus cells incubated on PHACOS was much lower (80%) in comparison to those on PHO. However, just 17% of cells incubated on the control polymer were non viable. The total number of bacterial cells (viable and non viable) adhering to disks coated with PHACOS and the total number of bacteria adhering to disks coated with control polymer were similar. This demonstrated that there was no decrease in bacterial adhesion but that bacterial survival was affected.

Moreover, biocompatibility and antibacterial activity of PHO and PHACOS were studied and demonstrated in vivo using BALB/c male mice and fluorescent dye (H-ICG). Therefore, we concluded that PHACOS have intrinsic antibacterial activity, as they are obtained from bacterial fermentation, and do not require chemical modification. In general, three types of antimicrobial polymers are known: biocidal polymers, polymeric biocides and biocide-releasing polymers. Biocidal polymers are usually positively charged macromolecules that interact with microbial cells, which generally carry a negative net charge at the surface due to their membrane proteins, teichoic acids of Gram-positive bacteria, and negatively charged phospholipids on the outer membrane of Gram-negative bacteria. Polymeric biocides consist of bioactive repeating units, i.e., the polymers are just multiple interconnected biocides, which act similarly to monomers. Often, polymerization of biocidal monomers does not lead to active antimicrobial polymers, either, because the polymers are water-insoluble or the biocidal functions do not reach their target. The last experiments that form part of this PhD Thesis demonstrated the antibacterial activity of PHACOS-derived dimers/trimers and therefore we propose that PHACOS act as a polymeric biocide.