Poly(oxazoline) Block Copolymers for Biomedical Applications - ACS

Jul 8, 2013 - One approach is to convert a commercially available α ... 0 and 2.3 ppm, and the protons connected to the nitrogen appear at 3.3-3.5 ...
2 downloads 0 Views 2MB Size
Chapter 5

Poly(oxazoline) Block Copolymers for Biomedical Applications Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

Michael J. Isaacman1 and Luke Theogarajan*,2 1Department

of Chemistry, University of California, Santa Barbara, Santa Barbara, California 93106 2Department of Electrical and Computer Engineering, University of California, Santa Barbara, Santa Barbara, California 93106 *E-mail: [email protected].

Poly(oxazolines) have emerged as a viable alternative to poly(ethyleneglycol), as a polymer coating for therapeutic applications. In this chapter, we will outline some of the key features of poly(oxazoline), its synthesis and its use as a hydrophilic segment of an amphiphilic block copolymer. Both macrointiation based block copolymer synthesis and polymer-polymer conjugation by click chemistry will be discussed in the context of synthesizing these block copolymers. The self-assembling properties of these block copolymers are then discussed. We show that nanoscale vesicular topologies are readily formed by these polymers.

Introduction and Motivation Translation from the lab-bench to the bedside of effective therapies, especially against cancer, has proved enormously challenging (1). A key reason for this gap between the advances in our knowledge of cancer biology and effective outcome in the clinic can be mainly attributed to the ineffectiveness of the drug delivery techniques employed. For example, only 1-10 parts in a 100,000 of intravenously delivered monoclonal anti-bodies reach the intended target (2). The main challenges to effective targeted delivery are the biological and physiological barriers that need to be overcome (3). Accumulation of the drug in the desired tissue is achieved by most targeted therapies via antigen binding, © 2013 American Chemical Society In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

since molecular targets that are different between cancerous and normal tissue are hard to find. Liposomal delivery has been the main approach to overcoming some of the barriers. Liposomes that utilize antibody based targeting, termed immunoliposomes, have been moderately successful in pre-clinical (4). The main challenge in using liposomes, apart from their poor scale-up, high-cost and short shelf-life, is that they are rapidly cleared by the reticulo-endothelial system (RES). High circulation times are essential for liposomes to effectively deliver their payload to targets such as tumors. Stealth properties that prevent rapid clearance by the RES can be imparted to liposomes by incorporating a small fraction (3-7 mol%) of lipids that have been conjugated to hydrophilic polymers. The increased circulation times are believed to be due to the increase in the hydrodynamic radius, thus avoiding renal clearance. For many decades, PEG has been the hydrophilic polymer of choice for biomedical conjugation (or PEGylation). One school of thought is that PEGylation of liposomes reduces surface opsonization by steric repulsion thus enabling macrophage resistance, though it is increasingly becoming clear that this is not the case (5). It seems a combination of effects such as limited concentration of opsonins in the blood; weak interaction, hydrophobic effects and favorable dysopsonin interaction with the polymer surface are responsible for the prolonged circulation times (5). Despite some commercial success PEGylation has its drawbacks. One major cause for concern is the so-called accelerated blood clearance (ABC) effect (6), where a second dose of PEGylated liposomes, administered within a few days, are rapidly cleared by the RES. It is believed that the complement activation by the serum proteins by the first dose is responsible for this effect (7), though an antibody activated pathway may also be at work (8, 9). Irrespective of the exact mechanism, it is widely suspected that the type of surface presented to the cell by the liposome is largely responsible for the fate of the liposome in-vivo (10). Unfortunately, liposome surfaces are not easily amenable to manipulation since a delicate balance exists between the amounts of PEGylated (3-10 mol%) to non-PEGylated lipids (11). Another polymer that has been investigated for drug delivery applications is Hydroxyethyl starch (HES), a polymer derived from the natural polymer amylopectin. HES found its main use as plasma volume substitute. Recently, it has been shown that conjugation of Hydroxyethyl starch (Hesylation) to proteins and drugs have been found to be more effective than the unconjugated versions of the drug (31). The main attribute that makes Hydroxyethyl starch useful is that it can degraded extracellularly by the plasmatic α–amylase (primary pathway) or intracellularly by α-glucosidase enzyme (32), enabling clearance through the renal pathway. The rate of clearance can be tailored via the amount of hyroxyethylation of the natural polymer. Hesylated drugs and proteins exhibit low immunogenicity, less interference with bindig sites in-vitro and exhibit low viscocity enabling high concentration formulations (31). Recently, it has been shown that by modification of HES, via esterification, with fatty acid chains such as lauric, stearic and palmitic acids, amphiphillic polymers that assemble into micelles and vesicles can be synthesized (30). Despite these advantages due to its highly branched nature, it is more difficult to synthesize polymers that exhibit a well-defined structure-function relationship. 54 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

ABA Self-Assembling Triblock Copolymers Clearly, a targeted, tailored, cost-effective and versatile method for drug delivery is needed. Rather than use the problematic lipid-centric approach, a modular polymer-centric approach would be preferable. The membrane of the vesicle will then be composed of a triblock copolymer that mimics the lipid-bilayer, transforming the liposome into a polymersome. The triblock comprises of a hydrophobic siloxane block flanked by two hydrophilic polymethyloxazoline blocks, an approach originally pioneered by Meier and co-workers (15). Polyoxazolines display the same “stealth” behavior like PEG without its adverse biological side-effects (12, 13). Since the packing density of the hydrophilic polymer is extremely high, the polymersome has an intrinsic advantage over PEGylated-lipid containing liposomes. The ease of tailoring the properties of the polymer makes it more attractive than HESylation. Additionally, we have extended the above concept to enable functionalization of the hydrophobic block (22). This allows for either tuning of the self-assembling properties or enabling a prodrug strategy. In the next few sections we will briefly review the polymerization of the individual blocks and then synthesis of the ABA triblock copolymer, followed by its self-assembly properties.

Figure 1. Mechanism of cationic ring-opening polymerization of 2-oxazolines. R can either be a small molecule such as propargyl or a macromolecule. R1 is an alkyl substituent, such as methyl, ethyl or phenyl, on the oxazoline monomer and OTs is the Tosylate leaving group. 55 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

Cationic Ring Opening Polymerization of Polyoxazolines Oxazolines belong to the class of endocyclic imino ethers susceptible to ring-opening isomerization polymerization. The polymerization of oxazolines are initiated by electrophilic species like protic acids, alkyl sulfonates like triflates and tosylates and aklyl halides and hence it is also called electrophilic polymerization. The mechanism of polymerization is shown in Figure 1. Depending on the nature of the initiator the propagation can proceed through an ionic species or a covalent species or in some cases both. The stability of the propagating species makes the nature of the polymerization living under appropriate conditions, 2-oxazolines can be unsubstituted or substituted and the nature of the substituent determines much of the properties of the polymeric block. If the substituent is methyl or ethyl the block is hydrophillic and bulkier substituents make it hydrophobic. The living nature of the polymerization makes it possible to terminate the polymerization with functional groups yielding macroinitiators and telechelics. One very useful end group is the acrylate group, which can be conveniently introduced by terminating the reaction with acrylic acid in an appropriate base like triethylamine or 2,6-lutidine. The carboxylate being nucleophilic enough to terminate the polymerization yielding post-polymerizable acrylate end-groups. Recently, Hoogenboom et. al. (14) have shown that under microwave irradiation conditions oxazolines bearing pendant alkyl groups can be readily polymerized.

Polysiloxanes as Hydrophobic Cores Polysiloxanes are generally prepared from cyclic monomers by ring opening polymerization using an acid catalyst (cationic) or base (anionic) initiator. Though the Si-O bond is highly stable under neutral conditions it is readily cleaved in highly acidic or basic conditions. Siloxane bonds are constantly broken and reformed in both anionic and cationic polymerizations, leading to both linear and cyclic species being formed until the reaction reaches a thermodynamic equilibrium. Hence, this polymerization is often termed an equilibration or redistribution polymerization. However in anionic ring-opening, the beginning of the reaction is most probably a kinetically controlled process and only in the later stages does the equilibration process take over. To obtain high molecular weights with low polydispersity careful time-controlled quenching of the reaction mixture is necessary. The main advantage of anionic polymerization is that it can be used to polymerize cycles that contain bulky side groups like phenyls that cannot proceed via equilibrium polymerization. However, anionic polymerization requires strained cycosiloxane rings such as trisiloxanes. Cationic polymerization, is poorly understood and the current state of understanding is that the acid catalyzes the ring-opening in the initiation step and continues to catlyze the reaction via the silyl ester that is formed via the condensation of the silanol group and the counterion. The other mechanism that is thought to be active in the propagation step is the conversion of the sianol end-group to a leaving group (water) by the acid catalyst. Unlike the Si-O bond the Si-C bond is stable under these reaction conditions and if molecules containing Si-C bonds are present, they will terminate the growing chain and serve as the end-blocker (17). 56 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

The remaining cyclic non-functional side products can be removed by vacuum distillation or precipitation. If the end-blocker is a siloxane dimer it will yield bifunctional siloxane telechelics and simultaneously provide a method for the control of the molecular weight.

Polyoxazoline Containing Triblock Block Copolymers

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

Triblock Synthesis by Macroinitation Previously reported syntheses of non-functionalized poly(siloxane)-bpoly(oxazoline) triblock copolymers relied on one of two major strategies. One approach is to convert a commercially available α,ω-bishydroxypoly(dimethylsiloxane) into a macroinitator by converting the hydroxyl group into a trifluoromethanesulfonic acid ester (triflate) (15), that initiate the ring-opening polymerization of 2-methyl-2-oxazoline. The second approach is to synthesize the poly(siloxane) with reactive end-groups, namely benzyl chloride, and use those as a macroinitiator for the ring-opening polymerization of 2-ethyl-2-oxazoline with Sodium iodide as catalyst (16). Another approach to the synthesis of a diblock copolymer that merits attention is based upon the hydrosilylation of a SiH-terminated poly(dimethylsiloxane) with allyl alcohol. The hydroxyl end group thus obtained is converted into a tosylate (17) and used as the macroinitiator as in the syntheses described above. The central theme of all the reported syntheses has been to generate a macrointiator capable of initiating the ring-opening polymerization of oxazolines. A wide variety of intiators have been reported of which the triflates are the most effective since they are highly electrophilic (18). However, triflates are generally less air and moisture stable, requiring very careful handling. The approach of using chlorides was also not very attractive since it needed fairly high temperatures (130°C) and the use of iodide as a catalyst can cause unwanted side reactions in the functional groups. Tosylates offer a good balance between the required leaving group ability, tolerance to other functional groups and stability. Functionalization of the hydrophobic core requires a copolymer of dimethylsiloxane and methylhydrosiloxane, P(DMS-co-MHS. Therefore reported schemes like the one mentioned earlier (17) cannot be used since it involves a hydrosilylation step. Also, post derivatization of a hydroxyl end-group is not attractive since the termination may neither be quantitative or bi-functional, and may lead to very tedious work-up strategies to isolate the bifunctional tosylates. Additionally, hydroxybutyl and hydroxypropyl terminated polysiloxanes degrade upon heating, through the loss of the end-groups (20). Recently, we reported (22) a novel and facile route to synthesizing quantitatively terminated bifunctional P(DMS-co-MHS) tosylated telechelics, which is a modification of a reaction reported by Yilgor et al. (16, 19). Triblock copolymers were synthesized by cationic ring opening polymerization of 2-methyl-2-oxazoline using the bistosylate terminated siloxane telechelics as a macroinitiator. As in the telechelic synthesis, reaction conditions were first determined using the PDMS homopolymer. The reaction scheme (Generation I) is shown in Figure 2 along with the 1H NMR spectrum of the 57 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

triblock copolymer. The spectrum exhibits the classical polyoxazoline segment peaks of the side chain methyl protons between 2. 0 and 2.3 ppm, and the protons connected to the nitrogen appear at 3.3-3.5 ppm. Similarly, triblock copolymers were formed when P(DMS-co-HMS) was used as macroinitiator, Figure 2, Generation II. The 1H NMR spectrum verifies the structure of the triblock copolymer, clearly indicating the presence of the poly(oxazoline), Si-CH3, and Si-H moieties. The presence of the fairly reactive Si-H groups does not seem to hinder the block polymerization under the reaction conditions used.

Figure 2. Generation I was used as a proof of concept to verify the synthetic scheme of using a tosylate terminated polysiloxane as a macroinitiator for the ring opening polymerization of 2-methyl-2-oxazoline. The 1H NMR spectrum shows the classical polyoxazoline peaks and are indicated as peaks b and c. Generation II was designed so that the polymeric backbone could be derivatized with functional molecules via the hydrosilylation reaction. The Si-H peak d at 4.7 ppm clearly indicate that under the reactions conditions the Si-H protons do no hinder the polymerization. (Reprinted with permission from Ref. (22), Copyright 2008 John Wiley & Sons.)

The triblock copolymers described above with the methylhydrosiloxane moieties containing B-block were further derivatized via a hydrosilylation reaction. The effect of the reaction order was investigated by either (i) first forming the triblock copolymers followed by the hydrosilylation reaction, and (ii) the hydrosilylation was conducted first, followed by the ring-opening polymerization 58 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

of 2-methyl-2-oxazoline. It was discovered that the hydrosilylation reaction, that is the attachment of the tethered supramolecules to the P(DMS-co-HMS) block, needs to be conducted prior to the block copolymer formation with 2-methyl-2-oxazoline (i.e. option (ii)). Reactions in which the triblock copolymer was formed first, option (i), and the side-chain moieties were tethered later met with failure. We believe this is due to the interaction of the platinum metal catalyst with the poly(oxazoline) block. We have not investigated the use of platinum catalyst’s other than Kardsedt’s catalyst since it is the most widely used and gentle method of hydrosilylation. Hence, the reaction scheme was modified so that after the copolymerization of D4 and D4H, a hydrosilylation reaction was performed yielding a P(DMS-co-HMS) copolymer derivatized with an appropriate side-chain, see Figure 3, Generation III.

Figure 3. Generation III shows a modification to the earlier scheme to yield a functionalized triblock copolymer. Here the backbone was first derivatized and then used as a macroinitiator for the ring opening polymerization of 2-methyl-2-oxazoline. As evidenced by the 1H NMR spectrum this route is successful and is the preferred route for the synthesis of functionalized triblock co-polymers. (Reprinted with permission from Ref. (22), Copyright 2008 John Wiley & Sons.)

Triblock Formation via Click Chemistry Recently, click chemistry has found considerable use in polymer-polymer conjugation via the copper-catalyzed azide-alkyne cycloaddtion (CuAAC) reaction (33–35). Click-based conjugation allows for the modular synthesis of block copolymer architectures with well-defined block lengths and end-groups. 59 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

The basic idea of the approach is shown in Figure 4. This allows for systematic investigation of the effect of a single parameter variation on the self-assembling properties of the macromolecular system. For example, by using individually well-characterized hydrophobic and hydrophilic blocks, the effect of the hydrophilic block length on the self-assembling properties could be studied. This exquisite level of control allows for sophisticated scientific exploration and will enable the understanding of the underlying structure-function relationships that influence the physics of self-assembly in these macromolecules.

Figure 4. Schematic view of our click-based triblock copolymer synthesis. (Reprinted with permission from Ref. (21), Copyright 2012 John Wiley & Sons.)

Poly(siloxane)s were converted into clickable partners for the alkynefunctionalized poly(oxazoline) A-blocks by conversion of the iodide or tosylate end groups into azides, using sodium azide. We found that while tosylate end-groups can be converted into azides under mild heating, this was not suitable for Si-H bearing copolymers. To circumvent this we used an iodide end-blocker that could be converted to an azide at room-temperature and was tolerant to the Si-H groups in the copolymer. The hydrophilic PMOXA A-block was synthesized via CROP using a propargyl tosylate initiator, which ensured terminal alkyne functionality on one end of the polymer. 60 In Tailored Polymer Architectures for Pharmaceutical and Biomedical Applications; Scholz, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2013.

61

Downloaded by UNIV OF PITTSBURGH on July 20, 2013 | http://pubs.acs.org Publication Date (Web): July 8, 2013 | doi: 10.1021/bk-2013-1135.ch005

Table 1. Copper-Catalyzed Alkyne-Azide Cycloaddition Methodology No.

Poly(siloxane) B-block

Copper Source

Solvent

Time (hours)

Result

Product

1

N3-PDMS-N3

Cu/Asc

H2O/THF

12