Final Report Summary - MAGNANOVES (Magnetically responsive nanoparticle-vesicle hydrogels as "smart" biomaterials for the spatiotemporal control of cellular responses)
Summary description of the project objectives: The fundamental aim of this project is to develop new “smart” biomaterials able to release active compounds upon the application of an external magnetic field, compounds that trigger biochemical responses in cells. After these “smart” biomaterials have delivered their payload, a subsequent magnetic signal will trigger the “self-destruction” of the biomaterial. Our strategy uses magnetic nanoparticle-vesicle aggregates (MNPVs) as the active elements, which in this project contain reversible covalent linkages (such as disulfide bonds) between nanoparticles and vesicles in the place of non-covalent interactions used previously. With this covalent crosslinking design we hoped for efficient control over the release of enzymes and other bioactive compounds from MNPVs.
Description of the work performed since the beginning of the project and the main results achieved: Initial work focused both on the development of new magnetically responsive MNPV assemblies and the magnetically triggered release of macromolecules from the MNPVs. The former is a key step of the project since these new MNPVs will be the active units and hence the key component of the biomaterial that will be developed, whilst the latter will allow the dual release of a prodrug and enzyme.
[1.] New magnetic nanoparticle-vesicle (MNPV) assemblies from reversible covalent chemistry. Two different methods for the reversible covalent crosslinking of MNPVs were explored.
(1.1.) Boronate esters. Our new approach to obtain covalently linked MNPVs started with the diol/boronic acid link. The ability of boronic acids to bind diols and give cyclic boronate esters is well known and may open a new path for nanoparticle-vesicle interactions. Both dopamine-gluconate coated magnetic nanoparticles (MNPs) and gluconate-functionalized amino-coated magnetic nanoparticles were created and mixed with vesicles doped with an azoboronolipid. However MNPV formation was not observed, which suggested the positive effect of multivalency is not enough to overcome low association constants at vesicle surfaces, an observation verified through the study of simple polygluconamides. Nonetheless these studies with boronic acid capped lipids has been published in part (see Org. Biomol. Chem. 2014, 12, 2576), with further observations on diol-boronic acid interactions at the bilayer interface to be published.
(1.2.) Disulfides. The next method used to achieve this objective was disulfide exchange. A series of new thiol-containing lipids have been synthesized by two methods: by adding sulfhydryl functionality directly onto a lipidic core as well as through opening of gamma-thiobutyrolactone and Traut’s reagent by amino- terminated lipids. To prepare MNPVs, activated mixed disulfide lipids were created by reaction of these thiol lipids with Ellman’s reagent. These “Ellmans-capped” lipids were doped into phospholipid vesicles, to undergo a thiol/disulfide exchange with free thiol groups present in the surface of coated MNPs.
Different coating procedures have been used to prepare suitable thiolated MNPs, which had a significant density of free surface –SH groups, good dispersability under the conditions to create MNPVs, and a high efficiency for converting magnetic energy into heat (a good specific activity ratio (SAR) value), which will ensure effective magnetic release of the MNPV contents. The best results have been obtained with double shell silica-coated MNPs: an initial silica coating followed by a 3-mercaptopropyltrimethoxysilane coating providing the desired thiol functionality on the MNPs. With this coating, aggregates with different MNP/vesicle ratios could be observed upon mixing vesicles and nanoparticles, although significant leakage of the vesicle contents points at the need for further improvements. Measurements of the zeta potential of analogous double coated MNPs suggested cationic surface charge in the MNPs may have contributed to this leakage. Initial studies revealed that these MNPVs formed from thiolated MNPs and vesicles doped with an Ellmans capped lipid (with a short di(ethylene glycol) spacer) gave ~26% level of release after application of an alternating magnetic field pulse, as well as significant leakage. To diminish the amount of leakage, longer PEG spacers (n = 44) were to be used in a series of new lipids capped with free thiols or activated disulfides; work in this area was ongoing at the end of the Fellowship. Both the thiol coated MNPs and Ellmans coated MNPs (see below), as well as MNPVs resulting from them, could have exciting applications in regenerative medicine, particularly in conjunction with HyStem hydrogel which was shown to have higher biocompatibility and lower long-term induced leakage of MNPV contents in the absence of a magnetic signal.
[2.] Release of macromolecules from MNPVs and dual release. To broaden the number of areas to which MNPVs could be applied, three goals were identified: release of macromolecules from MNPVs after a magnetic pulse, dual release of substrates and enzymes from MNPVs after a magnetic pulse, and magnetic release from MNPVs in suspension. Using a modified biotin-avidin cross-linked MNPV system it was shown for the first time that proteins could be released from MNPVs in suspension (not fixed within gels). These assays also successfully showed magnetic release of the protease trypsin from vesicle compartments, which digested a polypeptide in solution. However the corresponding release of a soluble trypsin substrate from another vesicle compartment did not give a response, which suggested that direct vesicle-vesicle transfer of substrates/enzymes using MNPVs is not currently feasible. In our 2015 publication of these results (Phys. Chem. Chem. Phys. 2015, 17, 15579) we outlined three key limitations in this area, namely: (a) conditions for encapsulation of large biomolecules in vesicles; (b) not all small hydrophilic compounds are suitable for release, especially if they have low solubility, significant hydrophobic character or cationic groups; (c) cost of current encapsulation technologies. Nonetheless these results indicate that MNPVs could be a new type of drug delivery system, where the drug delivered is an enzyme rather than a small molecule, with the catalytic activity of the enzyme potentially producing a greater response.
[3.] Combining synthetic biology with magnetic nanoparticles. The development of routes to thiolated MNPs then allowed these to be activated by reaction with Ellman’s reagent to form mixed disulfides on the surface (“Ellmans MNPs”). A reliable method of producing these Ellmans MNPs, which could react with thiol groups on vesicle surfaces, was found and was extended to the reaction with surface thiols on biomolecules. Of particular interest was using this technology to ligate proteins and enzymes onto the surface of magnetic nanoparticles; the enzymes reversibly attached to the surface of the nanoparticles in the MNPVs could act on substrate prodrugs that are magnetically released from the MNPVs by a magnetic pulse. Rather than using the limited number of naturally occurring proteins with surface thiols, we used the latest synthetic biology techniques to link horseradish peroxidase (HRP) modified with a ybbR tag to the surface of Ellmans-coated MNPs. The ybbr-modified HRP was reacted with soluble CoA under phosphopantetheinyl transferase Sfp catalysis, followed by ligation to the surface of Ellmans-MNPs through a thiol-disulphide exchange reaction. After washing, the activity of attached peroxidase was confirmed. We believe this methodology will encourage further work in this field, e.g. other ybbR tagged enzymes could be attached to the MNPs giving applications as sensors and industrial biocatalysts.
Potential impact and use of the results: There are two key results from this research. The release of biomolecules like cytochrome c and trypsin from MNPVs in suspension, and the demonstration that catalytic activity is retained by the latter, shows that these MNPV nano-constructs should have an exciting future for the remote magnetically triggered delivery of biopharmaceuticals in vivo. The development of magnetic nanoparticles with an activated disulfide coating (Ellmans MNPs) and showing techniques from synthetic biology can be used to modify these Ellmans MNPs with a range of modified proteins and peptides will be of wide interest. We illustrated this by enzymatically attaching a ybbr-tagged enzyme and showing catalytic activity was retained. Such enzyme-MNP conjugates linked through reversible covalent disulfide bonds have possible applications as sensors and in industrial biocatalysis.
Description of the work performed since the beginning of the project and the main results achieved: Initial work focused both on the development of new magnetically responsive MNPV assemblies and the magnetically triggered release of macromolecules from the MNPVs. The former is a key step of the project since these new MNPVs will be the active units and hence the key component of the biomaterial that will be developed, whilst the latter will allow the dual release of a prodrug and enzyme.
[1.] New magnetic nanoparticle-vesicle (MNPV) assemblies from reversible covalent chemistry. Two different methods for the reversible covalent crosslinking of MNPVs were explored.
(1.1.) Boronate esters. Our new approach to obtain covalently linked MNPVs started with the diol/boronic acid link. The ability of boronic acids to bind diols and give cyclic boronate esters is well known and may open a new path for nanoparticle-vesicle interactions. Both dopamine-gluconate coated magnetic nanoparticles (MNPs) and gluconate-functionalized amino-coated magnetic nanoparticles were created and mixed with vesicles doped with an azoboronolipid. However MNPV formation was not observed, which suggested the positive effect of multivalency is not enough to overcome low association constants at vesicle surfaces, an observation verified through the study of simple polygluconamides. Nonetheless these studies with boronic acid capped lipids has been published in part (see Org. Biomol. Chem. 2014, 12, 2576), with further observations on diol-boronic acid interactions at the bilayer interface to be published.
(1.2.) Disulfides. The next method used to achieve this objective was disulfide exchange. A series of new thiol-containing lipids have been synthesized by two methods: by adding sulfhydryl functionality directly onto a lipidic core as well as through opening of gamma-thiobutyrolactone and Traut’s reagent by amino- terminated lipids. To prepare MNPVs, activated mixed disulfide lipids were created by reaction of these thiol lipids with Ellman’s reagent. These “Ellmans-capped” lipids were doped into phospholipid vesicles, to undergo a thiol/disulfide exchange with free thiol groups present in the surface of coated MNPs.
Different coating procedures have been used to prepare suitable thiolated MNPs, which had a significant density of free surface –SH groups, good dispersability under the conditions to create MNPVs, and a high efficiency for converting magnetic energy into heat (a good specific activity ratio (SAR) value), which will ensure effective magnetic release of the MNPV contents. The best results have been obtained with double shell silica-coated MNPs: an initial silica coating followed by a 3-mercaptopropyltrimethoxysilane coating providing the desired thiol functionality on the MNPs. With this coating, aggregates with different MNP/vesicle ratios could be observed upon mixing vesicles and nanoparticles, although significant leakage of the vesicle contents points at the need for further improvements. Measurements of the zeta potential of analogous double coated MNPs suggested cationic surface charge in the MNPs may have contributed to this leakage. Initial studies revealed that these MNPVs formed from thiolated MNPs and vesicles doped with an Ellmans capped lipid (with a short di(ethylene glycol) spacer) gave ~26% level of release after application of an alternating magnetic field pulse, as well as significant leakage. To diminish the amount of leakage, longer PEG spacers (n = 44) were to be used in a series of new lipids capped with free thiols or activated disulfides; work in this area was ongoing at the end of the Fellowship. Both the thiol coated MNPs and Ellmans coated MNPs (see below), as well as MNPVs resulting from them, could have exciting applications in regenerative medicine, particularly in conjunction with HyStem hydrogel which was shown to have higher biocompatibility and lower long-term induced leakage of MNPV contents in the absence of a magnetic signal.
[2.] Release of macromolecules from MNPVs and dual release. To broaden the number of areas to which MNPVs could be applied, three goals were identified: release of macromolecules from MNPVs after a magnetic pulse, dual release of substrates and enzymes from MNPVs after a magnetic pulse, and magnetic release from MNPVs in suspension. Using a modified biotin-avidin cross-linked MNPV system it was shown for the first time that proteins could be released from MNPVs in suspension (not fixed within gels). These assays also successfully showed magnetic release of the protease trypsin from vesicle compartments, which digested a polypeptide in solution. However the corresponding release of a soluble trypsin substrate from another vesicle compartment did not give a response, which suggested that direct vesicle-vesicle transfer of substrates/enzymes using MNPVs is not currently feasible. In our 2015 publication of these results (Phys. Chem. Chem. Phys. 2015, 17, 15579) we outlined three key limitations in this area, namely: (a) conditions for encapsulation of large biomolecules in vesicles; (b) not all small hydrophilic compounds are suitable for release, especially if they have low solubility, significant hydrophobic character or cationic groups; (c) cost of current encapsulation technologies. Nonetheless these results indicate that MNPVs could be a new type of drug delivery system, where the drug delivered is an enzyme rather than a small molecule, with the catalytic activity of the enzyme potentially producing a greater response.
[3.] Combining synthetic biology with magnetic nanoparticles. The development of routes to thiolated MNPs then allowed these to be activated by reaction with Ellman’s reagent to form mixed disulfides on the surface (“Ellmans MNPs”). A reliable method of producing these Ellmans MNPs, which could react with thiol groups on vesicle surfaces, was found and was extended to the reaction with surface thiols on biomolecules. Of particular interest was using this technology to ligate proteins and enzymes onto the surface of magnetic nanoparticles; the enzymes reversibly attached to the surface of the nanoparticles in the MNPVs could act on substrate prodrugs that are magnetically released from the MNPVs by a magnetic pulse. Rather than using the limited number of naturally occurring proteins with surface thiols, we used the latest synthetic biology techniques to link horseradish peroxidase (HRP) modified with a ybbR tag to the surface of Ellmans-coated MNPs. The ybbr-modified HRP was reacted with soluble CoA under phosphopantetheinyl transferase Sfp catalysis, followed by ligation to the surface of Ellmans-MNPs through a thiol-disulphide exchange reaction. After washing, the activity of attached peroxidase was confirmed. We believe this methodology will encourage further work in this field, e.g. other ybbR tagged enzymes could be attached to the MNPs giving applications as sensors and industrial biocatalysts.
Potential impact and use of the results: There are two key results from this research. The release of biomolecules like cytochrome c and trypsin from MNPVs in suspension, and the demonstration that catalytic activity is retained by the latter, shows that these MNPV nano-constructs should have an exciting future for the remote magnetically triggered delivery of biopharmaceuticals in vivo. The development of magnetic nanoparticles with an activated disulfide coating (Ellmans MNPs) and showing techniques from synthetic biology can be used to modify these Ellmans MNPs with a range of modified proteins and peptides will be of wide interest. We illustrated this by enzymatically attaching a ybbr-tagged enzyme and showing catalytic activity was retained. Such enzyme-MNP conjugates linked through reversible covalent disulfide bonds have possible applications as sensors and in industrial biocatalysis.