Biomacromolecules 2004, 5, 306-311
306
Water-Soluble and Nonionic Polyphosphoester: Synthesis, Degradation, Biocompatibility and Enhancement of Gene Expression in Mouse Muscle Shi-Wen Huang,†,‡ Jun Wang,†,§ Peng-Chi Zhang,† Hai-Quan Mao,†,| Ren-Xi Zhuo,*,‡ and Kam W. Leong*,†,§ Tissue & Therapeutic Engineering Lab, Johns Hopkins Singapore, Singapore 117597, Department of Biomedical Engineering, Johns Hopkins University, School of Medicine, Baltimore, Maryland 21205, Department of Chemistry, Wuhan University, Wuhan 430072, P. R. China, and Department of Materials Science and Engineering, Johns Hopkins University, Baltimore, Maryland 21218 Received July 16, 2003; Revised Manuscript Received November 28, 2003
A nonionic and water-soluble polyphosphoester, poly(2-hydroxyethyl propylene phosphate) (PPE3), was synthesized by chlorination of poly(4-methyl-2-oxo-2-hydro-1,3,2-dioxaphospholane), followed by esterification with 2-benzyloxyethanol and deprotection of the hydroxyl group by catalytic hydrogenation in the presence of Pd-C. PPE3 degraded rapidly in PBS 7.4 at 37 °C. The cytotoxicity and tissue compatibility assays suggested good biocompatibility of PPE3 in vitro and in vivo. The expression of pVR1255 Luc plasmid in mouse muscle after intramuscular (i.m.) injection of DNA formulated with PPE3 solution in saline was enhanced up to 4-fold compared with that of naked DNA. These results suggest the potential of this polyphosphoester for naked DNA-based gene therapy. The advantages of this polymer design include the biodegradability of the polyphosphoester and its structural versatility, which allows the fine-tuning of the physicochemical properties to optimize the enhancement of gene expression in muscle. Introduction Strategies for naked DNA-based gene therapy have been widely investigated and applied in clinical trials since Wolff et al. first reported the successful expression of plasmid DNA in muscle after intramuscular injection of a foreign gene.1-3 Naked DNA, the simplest gene medicine, can be taken up by muscle cells at the DNA injection site and yields longterm expression in muscle, in some cases the expression lasting for 2 years.4,5 DNA taken up by myoblasts remains unintegrated and extrachromosomal. Transgene expression in the muscle may be used as gene medicines or as DNA vaccines.6-8 Compared with the application of recombinant viral vectors, intramuscular injection of naked DNA has the advantages of simplicity, safety, and cost-effectiveness. However, only relatively low transgene expression can be achieved, particularly at low DNA doses. This is most likely an issue of low bioavailability. More than 98% of injected DNA is rapidly degraded by extracellular nucleases or eliminated from the muscle, and only a small portion of the injected DNA can enter into the muscle cells.9-11 Efficient gene delivery systems that can improve the DNA availability would enhance the therapeutic potential of naked DNA administration. * To whom correspondence should be addressed. (K.W.L.) 726 Ross Building, 720 Rutland Avenue, Baltimore MD21218. Phone: +1-410-6143741. Fax: +1-410-955-0075. E-mail:
[email protected]. (R.-X.Z.) Department of Chemistry, Wuhan University, Wuhan 430072, P. R. China. Phone & Fax: +86-27-8764-8509. E-mail:
[email protected]. † Johns Hopkins Singapore. ‡ Wuhan University. § Johns Hopkins University, School of Medicine. | Johns Hopkins University.
Although condensation of plasmid DNA with cationic polymer can form polyelectrolyte complexes or nanoparticles to protect DNA from degradation by nucleases and deliver the foreign gene into targeting cells or tissues more efficiently,12,13 cationic polymeric carriers have rarely been shown to improve gene expression in muscle, with the exception of a biodegradable polyphosphoester (PPE) containing primary amino side chains.14 One possible explanation for low transfection efficiency mediated by cationic polymerDNA complexes in muscle is that the connective tissue surrounding the myofibers and muscle fasciculus significantly limits the diffusion of the condensed DNA particles through the muscle.15,16 Mumper et al. has reported a series of protective and interactive noncondensation (PINC) polymer systems. For example, poly(vinylpyrrolidone) and its copolymers protect plasmids from nuclease degradation by forming loose PNIC complexes with DNA. The complexes purportedly could diffuse throughout the muscle for reasonable gene expression.2,15,17 Other PINC gene delivery systems include poloxamers (pluronic L61 and F127)18 and maltodextrin.19 We have previously reported a new family of cationic gene carriers based on PPEs. The molecular design offers a good degree of versatility to allow tailored design of the polymeric structures.14,20-23 PPEs, particularly the noncharged PPEs, have good tissue compatibility and low cytotoxicity.24 With an aim to explore if electrostatically neutral PPE with polar side chains could enhance gene expression when delivered together with plasmid DNA, we have designed a water soluble, nonionic PPE with hydroxyethyl side chains (Scheme
10.1021/bm034241l CCC: $27.50 © 2004 American Chemical Society Published on Web 02/06/2004
Water-Soluble and Nonionic Polyphosphoester
Biomacromolecules, Vol. 5, No. 2, 2004 307
Figure 2. In vitro degradation of PPE3 in 0.1 M PBS (pH7.4) at 37 °C.
Figure 1.
1H
NMR spectra of PPE2 (CDCl3) and PPE3 (DMSO-d6).
1, PPE3). In this report, we describe the preparation, in vitro degradation and biocompatibility of PPE3, and the intramuscular transfection efficiency of naked DNA co-delivered with PPE3. Results and Discussion Synthesis of Poly(2-hydroxyethyl propylene phosphate) (PPE3). PPE3 was synthesized in a method similar to poly(2-aminoethyl propylene phosphate) (PPE-EA) we described previously14 (Scheme 1). A precursor polymer PPE1, obtained by ring opening polymerization of 4-methyl-2-oxo2-hydro-1,3,2-dioxaphospholane was chlorinated according to the method described by Penczek et al.,26,27 resulting in highly nucleophile-reactive P-Cl bonds. Reacting those P-Cl bonds to an excess of 2-benzyloxyethanol in chloroform using 4-(dimethylamino)-pyridine as a catalyst yielded intermediate PPE2. PPE2 is soluble in chloroform, dimethyl sulfoxide, and methanol. The 1H NMR spectrum of PPE2 in Figure 1 indicated a complete side chains transformation from P-Cl to P-OCH2CH2OCH2Ph. PPE3 was prepared by removal of benzyloxy protecting groups by Pd/C catalytic hydrogenation in methanol.28 Complete removal of the benzyl
protecting group required a relatively longer reaction time (16 h). The absence of peaks assigned to the benzyloxy group in the 1H NMR spectrum (δ 7.22 ppm and δ 4.56 ppm) of PPE3 indicated a complete deprotection (Figure 1). The polymer was partially degraded during the catalytic hydrogenation process, and as a result, terminal P(O)-OH groups were formed. This was reflected in the 1H NMR spectrum of PPE3 by the appearance of a side peak at δ 1.05 ppm, which was assigned to methyl groups adjacent to P(O)-OH groups. PPE3 is soluble in water, dimethyl sulfoxide, and methanol. Its weight average molecular weight was 9100 with a polydispersity index of 1.44 as determined by the GPC/LS/RI method, which corresponded to a number average degree of polymerization (DPn) of 34.6. In Vitro Degradation. An obvious advantage of this polymer design is that the backbone of PPE3 is hydrolytically cleavable. The hydrolytic degradation kinetics of PPE3 was studied in phosphate buffer (0.1 M, pH 7.4) at 37 °C according to the method reported previously.14 Figure 2 shows the degradation profile of PPE3 as a function of time. Backbone cleavage is evident in the corresponding decrease of both Mw and Mn, almost in a linear fashion. The Mw decreased by 9% in the first day. By day 7, the Mw decreased to 67% of its original value, whereas Mn dropped by 80%. This rapid degradation is attributed to the aliphatic nature of PPE3. This degradation rate is in the same order of magnitude as that observed previously with a polyphosphate PPE-EA that contains the same propylene backbone and an ethylamine side chain instead of an ethanol side chain in
Scheme 1. Synthesis of Water-Soluble and Nonionic Poly(2-hydroxyethyl propylene phosphate) (PPE3)
308
Biomacromolecules, Vol. 5, No. 2, 2004
Figure 3. In vitro cytotoxicity of PPE3 in COS-7 cells and HEK293 cells (n ) 6).
PPE3.14 The Mw of the PPE-EA dropped 15% in 24 h, and 83% by day 7. The less polar side chain in PPE3 than in PPE-EA is attributed to the slightly lower degradation rate observed. In Vitro Cytotoxicity. Biocompatibility is an important factor in gene carrier designs. We have opted for a polymeric structure that is hypothesized to breakdown into innocuous compounds. The final degradation products of PPE3 should include phosphate salt, ethylene glycol, and 1,2-propanediol, all of which have good safety profiles. Quantitatively, complete hydrolysis of 1 mg of PPE3 will give 0.42 mg of 1,2-propandiol and 0.34 mg of ethylene glycol. To confirm this hypothesis, we first analyzed the in vitro cytotoxicity of PPE3 by the MTT assay in COS-7 cells and HEK 293 cells,29 and compared with PEI (Mw ) 25 KDa), a wellknown gene carrier. The results shown in Figure 3 indicated
Huang et al.
that PPE3 has minimal toxicity to both COS-7 cells and HEK 293 cells at a polymer concentration as high as 12.5 mg/ mL. The LC50 of PPE3 in these two cell lines are in the range of 43-50 mg/mL, in contrast to
