gene delivery applications

European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb

Review Article

Polydioxanone-based bio-materials for tissue engineering and drug/gene delivery applications Nowsheen Goonoo, Roubeena Jeetah, Archana Bhaw-Luximon, Dhanjay Jhurry ⇑ ANDI Centre of Excellence for Biomedical and Biomaterials Research, MSIRI Building, University of Mauritius, Réduit, Mauritius

a r t i c l e

i n f o

Article history: Received 13 December 2014 Revised 27 May 2015 Accepted in revised form 28 May 2015

Keywords: Polydioxanone Poly(ester-ether)s Biomaterials Tissue engineering Biodegradable scaffolds Drug delivery systems

a b s t r a c t Since the commercialization of polydioxanone (PDX) as a biodegradable monofilament suture by Ethicon in 1981, the polymer has received only limited interest until recently. The limitations of polylactide-co-glycolide (PLGA) coupled with the growing need for materials with enhanced features and the advent of new fabrication techniques such as electrospinning have revived interest for PDX in medical devices, tissue engineering and drug delivery applications. Electrospun PDX mats show comparable mechanical properties as the major structural components of native vascular extracellular matrix (ECM) i.e. collagen and elastin. In addition, PDX’s unique shape memory property provides rebound and kink resistance when fabricated into vascular conduits. The synthesis of methyl dioxanone (MeDX) monomer and copolymers of dioxanone (DX) and MeDX have opened up new perspectives for poly(ester-ether)s, enabling the design of the next generation of tissue engineering scaffolds for application in regenerating such tissues as arteries, peripheral nerve and bone. Tailoring of polymer properties and their formulation as nanoparticles, nanomicelles or nanofibers have brought along important developments in the area of controlled drug or gene delivery. This paper reviews the synthesis of PDX and its copolymers and provides for the first time an exhaustive account of its applications in the (bio)medical field with focus on tissue engineering and drug/gene delivery. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Biodegradable aliphatic polyesters such as poly(lactide) (PLA), poly(lactide-co-glycolide) (PLGA) and polycaprolactone (PCL) have attracted much interest for applications ranging from medical implants, bone fixation parts, scaffold fabrication, controlled drug release devices to sustained release systems for pesticides and fertilizers [1,2]. The major advantage with their use is that their degradation products can be removed by natural metabolic pathways. Generally, the copolymer PLGA is preferred compared to its constituent homopolymers for the fabrication of bone substitute constructs mainly because PLGA offers superior control of degradation properties by varying the ratio of LA and GA monomers. PLGA has a wide range of degradation rates, governed by the composition of chains, both hydrophobic/hydrophilic balance and crystallinity [3]. The possibility of controlling polymeric degradation rates allows matching with tissue regeneration rate for tissue engineering applications and control of drug release kinetics for drug delivery. ⇑ Corresponding author. Tel.: +230 4651347. E-mail address: [email protected] (D. Jhurry). http://dx.doi.org/10.1016/j.ejpb.2015.05.024 0939-6411/Ó 2015 Elsevier B.V. All rights reserved.

Unlike poly(ester)s, the biodegradable poly(ester-ether) PDX has not been much investigated since its introduction on the market in 1981 by Ethicon and is best known for its clinical use as a monofilament suture [4]. More recently, it has been used in the fabrication of rings for pediatric mitral and tricuspid heart valve repair [5,6] and as plates for orbital floor reconstruction [7]. Several studies have investigated the use of PDX stents in both vascular and non-vascular organs such as esophagus, trachea and intestine [8–12]. PDX shows several interesting unique properties compared to polyesters. For instance, as shown by Boland et al. [13], bulk material properties of electrospun PDX are of the same order of magnitude as the major structural components of native vascular ECM, in particular collagen and elastin. In addition, PDX has shape memory, and hence its fabrication into vascular conduits provides rebound and kink resistance [13]. However, the shape memory property of PDX makes knot retention difficult and hence undesirable when used as PDSÒ suture. Copolymerization with new dioxanone (DX) analogues has made it possible to modulate degradation rate compared to PDX and has opened up new perspectives for drug and gene delivery applications.

372

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Sn(Oct) 2

II

Al(OiPr) 3 DX Zn

Lipase

I

(A)

(B)

Fig. 1. (A) Different catalysts used for DX homopolymerization and (B) 1H NMR spectra (CDCl3) of DX (I) and PDX (II).

This paper reviews PDX-based biomaterials and their applications in tissue engineering and drug/gene delivery. An introductory overview of the synthesis of DX monomer and its methyl derivative (MeDX) as well as their homopolymers and copolymers is given while the focus will be on their use as scaffolds for tissue engineering applications as well as for the elaboration of drug delivery devices. 2. Synthesis of DX and its derivatives The synthesis of DX was first reported by Doddi et al. [14] who reacted sodium glycolate with chloroacetic acid. The resulting hydroxy acid undergoes cyclization at 200 °C in the presence of MgCO3, to yield 1,4-dioxan-2-one (DX). The crude product was purified to 99% by multiple crystallization and distillation with 50–70% yield. In an attempt to improve on yield, Nakatami et al. [15] purified the sodium salt of the hydroxy acid to eliminate ethylene glycol and any residual chloroacetic acid or chloroacetate. However, the yield of the product (67%) did not improve significantly. A one-step synthesis was described by Forschner et al. [16] whereby oxidative dehydrogenation of diethylene glycol is performed in the presence of copper oxide catalysts supported on silica particles at 275 °C under hydrogen atmosphere. However, the exact reaction conditions and preparation of the catalyst still remain buried in patents. More recently, the synthesis of a dioxanone analogue namely D,L-3-methyl-1,4-dioxanone (MeDX) [17,18] was reported. The monomer was synthesized using a modified version of Doddi’s method (Scheme 1). Briefly, sodium glycolate was made to react with D,L-2-chloropropionic acid. Intramolecular esterification of the hydroxyl acid was performed in the presence of dicyclohexylcarbodiimide (DCC) as a catalyst. Purified MeDX monomer was obtained by cryodistillation in approximately 50% yield.

2.1. Homopolymerization of DX and MeDX Polydioxanone (PDX) is synthesized by ring-opening polymerization (ROP) of DX. The catalyst plays an important role in the polymerization process, influencing not only the polymeric parameters such as reaction rate, conversion and yield, but also the properties of the polymer such as molecular weight and polydispersity

[19]. Several catalysts such as organic tin (e.g. Sn(Oct)2) [20–29], organic aluminum (e.g. Al(OiPr)3) [30] as well as organic zinc compounds [31] have been found effective for the ROP of DX. As reported by Nishida et al. [25], PDX has a ceiling temperature of 265 °C. Special precautions should be taken to process the material at the lowest temperature possible to avoid depolymerization to DX. In fact, the monomer–polymer equilibrium was described by the microreversibility model according to which all growing chains (Pn and Pn+1) are capable of depolymerizing until a constant monomer concentration is reached (Eq. (1)) [25]: kp

Pn þ M Pnþ1 kdp

ð1Þ

where Pn and Pn+1 are propagating polymer chains, M is the monomer DX, and kp and kdp are the rate constants for polymerization and depolymerization respectively. It is crucial to remove all traces of metallic catalysts in PDX before use in biomedical and pharmaceutical fields to avoid any adverse biological reactions. Metallic catalyst residues may be removed using solvent extraction [30]. This step is costly and has prompted researchers to try enzymatic polymerization of DX. Enzymatic activity is influenced by temperature, origin of enzyme and the reaction media [32]. Nishida et al. [33] reported on the successful enzymatic polymerization of DX using 5 wt% immobilized lipase at 100 °C for 15 h. However, the rate of monomer conversion is very low and hence cannot be used for large scale application. The different catalysts used for DX homopolymerization and 1H NMR spectra of DX and PDX are given in Fig. 1. Homopolymerization of racemic MeDX has been carried out using various initiator systems for instance Sn(Oct)2/n-BuOH, Sn(Oct)2 and Al(OiPr)3 [18]. However DX is found to polymerize to higher conversions than MeDX irrespective of the nature of the initiator. Highest molar mass polymers with low polydispersity indices were reported with Al(OiPr)3 in the temperature range 40–60 °C. Prolonged polymerization times led to a decrease in molar masses and a broadening of molar mass distribution, due probably to increased depolymerization and side reactions. The different catalysts used for MeDX homopolymerization and 1H NMR spectra of MeDX and poly(methyl dioxanone) (PMeDX) are given in Fig. 2. PDX is a semi-crystalline polymer with the glass transition temperature of about 10 °C and melting temperature of around 110 °C. PMeDX is amorphous and is soluble in tetrahydrofuran

373

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

O OH

+

HO

R

+

Na (s)

OH Cl

R: H, CH3

Reflux at 100 oC under vacuum Neutralisation

O HO

O

OH R

+

OH

HO

1. Cyclisation using DCC 2. Cryodistillation

O

O 1,4-dioxan-2-one (DX) R = H D,L-3-methyl-1,4-dioxanone (MeDX) R = CH3

O

R

Scheme 1. DX and MeDX synthetic pathway [17,18].

II

Sn(Oct)2

MeDX

I Al(OiPr)3

(A)

(B) 1

Fig. 2. (A) Different catalysts used for MeDX homopolymerization and (B) H NMR spectra (CDCl3) of MeDX (I) and PMeDX (II).

and dichloromethane. Compared to PDX, PMeDX has a lower ceiling temperature (104 °C) which explains the lower polymer conversion and lower molar masses obtained at higher polymerization temperatures [18]. Thermal degradation of PMeDX showed that degradation occurs in a single step with onset degradation temperature (Tonset) of 78 °C. Degradation continues until 300 °C. 2.2. DX and MeDX copolymer microstructure Lochee et al. reported on the synthesis of a range of random P(DX-co-MeDX) copolymers with varying co-monomer ratios by the non-sequential polymerization of DX with MeDX using initiators such as tin(II) octanoate, tin(II) octanoate/n-butyl alcohol, and aluminum tris(isopropoxide) [17,34]. The five carbonyl signals in 13C NMR spectrum were assigned to the triads MeDX–MeDX–MeDX + DX–MeDX–MeDX (173.0 ppm), MeDX–MeDX–DX + DX–MeDX–DX (172.9 ppm), MeDX–DX–MeDX

(171.1 ppm), DX–DX–MeDX + MeDX–DX–DX (170.8 ppm) and DX–DX–DX (170.3 ppm) indicating a pseudo-periodic pattern with a random distribution of MeDX units in the copolymer. In general, the copolymers were slightly enriched in DX. Thermal properties of the copolymer changed significantly with increasing percentage of MeDX units, in particular a decrease in the melting temperature and crystallinity of the copolymer was noted with an increase in MeDX units. Furthermore, analysis of the copolymers by MALDI-TOF MS showed the presence of a small amount of macro-cycles.

2.3. Biodegradation of PDX homopolymers and copolymers 2.3.1. Mechanism of hydrolytic degradation The hydrolytic degradation of PDX is postulated as a ‘‘cleavage-induced crystallization process’’ which occurs in two stages whereby amorphous regions are more easily attacked than

374

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 3. SEM images of outer layer of EG/PEG/PG scaffold after aging in hydrolytic medium for (a) 5 days, (b) 10 days, (c) 20 days, and (d) 30 days. Arrows show fiber breaking or cracks formed due to aging (reproduced with permission from Thomas et al. [38]).

crystalline ones [35]. According to Sabino et al. [36], chain scission proceeds in two steps with the first occurring in the amorphous regions of microfibrils and intermicrofibrillar space. The second one then occurs in the crystalline regions. Scission of PDX chains in the amorphous regions leads to higher mobility of chains which results in an increase in the crystallinity. Similar to the biodegradable polyesters, PDX is also assumed to degrade via a bulk erosion mechanism (Scheme 2). Resulting products of in vivo degradation of PDX are in accordance with the normal metabolites of the body. Most of them are excreted via the respiratory tract, and the rest are excreted via the alimentary tract [37]. PDX was approved by the FDA for use as a biodegradable monofilament suture (PDSÒ) in 1981. O H2C

H2 C

H2 C

O

C

2.3.2. Controlling degradation 2.3.2.1. Via blending. Blending synthetic polymers with natural ones or copolymerizing hydrophobic monomers with hydrophilic ones in varying ratios can help tailor the degradation rate to either match the growth rate of the regenerating tissue or to control drug release kinetics. Blends of PDX and elastin or gelatin prepared for arterial scaffolds exhibited increased hydrophilicity [38]. The phosphate buffer solution (PBS) uptake of PDX/elastin and PDX/gelatin was 395% and 363% compared to 206% for PDX. However, no mention about polymer miscibility in the blends was made in the paper. Moreover, the inner layer (gelatin/elastin) of a trilayered vascular graft was found to degrade faster than the outer layer which OH

H2O O

O

H2 C

O

C OH

OHO

O

OH

HO

C

H2C

HO

P

P

O-

O

O

P

H

H

O

O C

C H2

OH

Scheme 2. Proposed scheme for the hydrolysis of PDX in phosphate buffer solution [36].

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

375

Fig. 4. SEM micrographs of 4 samples (magnification: 500) during the hydrolysis in phosphate buffer solution (pH 7.4) at 37 °C after 8, 10 and 14 weeks (reproduced with permission from Ding et al. [42]).

Fig. 5. P(BDX-co-DX) and P(HDX-co-DX) copolymers.

consisted of PDX/gelatin, as evidenced by loss in the inner fibrous morphology by 5 days followed by complete delamination by 10 days of in vitro degradation. In contrast, major fiber breaking was observed in the outer layer after 30 days (Fig. 3). An increase in crystallinity was noted with increasing degradation time due to a decrease in chain entanglements in the amorphous region, and the successive incorporation of these chains into the crystalline regions, forming new crystal lamellas. Additionally, increasing poly(vinyl alcohol) (PVA) content in porous PDX/PVA scaffolds improved the resulting hydrophilicity [39]. In another study, the authors showed that the rate of hydrolysis in PDX/PVA-g-PDX blends increased with increasing content of the graft copolymer [34]. This was explained by the fact that the addition of PVA-g-PDX to PDX results in less regularity in the PDX crystalline phase due to dispersion of the graft copolymer in PDX matrix. 2.3.2.2. Via copolymerization. To improve the degradation rate of PDX, two strategies have been adopted. First, copolymerization allows the adjustment of the length of either or both blocks in

order to control the degradation rate of the copolymer. For instance, the introduction of phosphate or starch segments in the backbone led to an enhancement of the degradation of the copolymer in the case of P(DX-co-ethyl ethylene phosphate) [40] and starch-g-PDX [41]. On the other hand, increase in the weight fraction of butylene succinate (BS) in PDX-b-PBS [42] led to a decrease in the degradation rate of the copolymer (Fig. 4). The second strategy involves the synthesis of derivatives of DX monomer and its copolymerization with DX. For instance, Li et al. [43] synthesized a novel dioxanone monomer, 5-benzyloxymethyl-1,4-dioxan-2-one (BDX) and carried out bulk ROP. The protecting benzyl groups of poly(5-benzyloxyme thyl-1,4-dioxan-2-one) (PBX) were subsequently removed by catalytic hydrogenation to give poly(5-hydroxymethyl-1,4-dioxa n-2-one) (PHDX) with pendant hydroxyl groups. The authors also prepared random copolymers of P(BDX-co-DX)s. Deprotection of the benzyl groups in the copolymers afforded the corresponding hydroxyl-containing P(HDX-co-DX)s. In comparison with PDX and PBDX, the degradation rate of PHDX was greatly enhanced

376

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

VicrylÒ (PLGA) [46]. No difference was found in complication rates or cellular reaction to the suture material between VicrylÒ and PDSÒ [47]. However, there are several negative aspects to the use of PDSÒ as a suturing material such as loosening of fine surgical knots. This occurs for two reasons. PDX has shape memory which causes the suture to retain its spooled shape and thus, can be difficult to use. In addition, PDSÒ has low surface friction allowing it to glide through tissues easily, but the lack of friction coupled with shape memory makes knot retention difficult [48]. However, the shape memory effect which is undesirable for PDSÒ suture is actually an interesting property of PDX when used as vascular conduits. Another downside of PDSÒ suture concerns the degradation rates which may be too rapid to provide a durable closure for wounds with abnormal healing and the suture ends may be sharp when cut. 3.2. Commercially available implants

due to its higher hydrophilicity which can be accounted for by the presence of the pendant hydroxyl groups (Fig. 5).

Over the past decades, surgical strategies for mitral regurgitation have shifted from mitral valve replacement to mitral valve repair [49]. In 2006, EU-approved Kalangos Biodegradable Ring based on PDX was developed for pediatric mitral and tricuspid heart valve repair [5,6]. The ring is presently commercialized by Parvulus Suisse (Lonay, Switzerland). The biodegradable ring has a curved C-shape segment consisting of PDX colored with a blue dye and attached at both ends with a needle holding extension monofilament polyvinyl suture. The ring is implanted ‘‘into’’ the annulus which decreases the risk of postoperative thromboembolic complications [50]. Moreover, there is no need for anticoagulant therapy given that the ring is not in direct contact with the circulating blood. The polymer undergoes slow degradation by hydrolysis resulting in an inflammatory reaction (Fig. 6) [51]. The latter induces fibrosis which enables remodeling of the posterior mitral annulus. Histological biocompatibility of the biodegradable ring in a porcine model showed that the biodegradable ring was gradually replaced by fibrous tissue, with complete hydrolytic degradation within 6 months [5]. Echocardiography showed no signs of tricuspid valve dysfunction, a preserved ventricular contractility, and physiological growth of the tricuspid valve orifice. Polydioxanone plates (PDS™ flexible plates), commercialized by Ethicon Inc., Johnson & Johnson is available as perforated and unperforated plates in varying shapes and sizes [7]. Boenisch et al. showed that PDS™ plates were histoconductive and stimulated bone regeneration in orbital floor reconstruction [52]. Flexibility of PDX together with its biocompatibility rationalize its use as scaffolds (PDS™ plates) for extracorporeal septorhinoplasty, columellar struts, septal extension grafts, endonasal septoplasty and septal perforation repair [53].

3. Applications of PDX

3.3. Electrospun PDX scaffolds for tissue engineering

3.1. Commercial sutures

Polyesters have been investigated for use as scaffolds in tissue engineering by several research groups. However, a major concern with the use of PLAs is that their degradation products decrease local pH, which in turn, induces an inflammatory reaction and affects cells at the implant site [54]. As a result, the search for new materials is still ongoing and PDX was found to perform significantly better than PLAs. As shown by Madurantakam et al. [55], better scaffold mineralization was observed for PDX containing 50% HA compared to the corresponding PLGA scaffolds in ionic simulated body fluid (i-SBF) and revised simulated body fluid (r-SBF). The acidic degradation products of PLGA were shown to inhibit mineral growth on the scaffolds. Recent in vitro studies carried out using human blood clearly pointed out that electrospun PDX nanofibers did not perturb the phagocytic function of human monocytes and neutrophils [56]. Cytokine secretion by monocytes

Fig. 6. Assessment of the degradation and fibrotic characteristics of the intraannular ring in the tricuspid position in a growing pig model. (A) Macroscopic observation of the ring and fibrotic reaction (bold arrows) at 6 months after implantation into the tricuspid annulus. Dotted arrows indicate the fixation points of the ring on the antero-septal and postero-septal commissures (AL = anterior leaflet of the tricuspid valve, PL = posterior leaflet of the tricuspid valve). (B) Histological analysis of tricuspid valve annulus at 12 months post-ring implantation showing complete degradation of the ring associated with intense fibrotic reaction (Masson’s trichrome, 40) (reproduced with permission from Cikirikcioglu et al. [51]).

PDX was first introduced by Ethicon in 1981 and marketed as an FDA approved biodegradable monofilament suture PDSÒ [4]. PDSÒ suture had greater pliability than polypropylene suture and also had greater strength than that of other monofilament sutures. The flexibility of PDSÒ suture can be explained by the presence of an ether oxygen in the backbone structure. PDSÒ retained 25% of its tensile strength at 42 days and was absorbed within 130–180 days [44]. This time scale is advantageous for slower healing wounds (compared to DexonÒ). However, the monofilament PDSÒ is not retained as long as poly(lactic-acid)-based sutures [45]. PDSÒ elicits a low order of tissue response and is absorbed by simple hydrolysis. PDSÒ retains its tensile strength longer compared to the two multifilament sutures, DexonÒ (PGA) and

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

377

Fig. 7. Between 1 and 4 months, glycosaminoglycan concentration decreased whereas elastin concentration increased to become close to that of the ECM of the native tissue, scale bar: 0.5 mm (reproduced with permission from Kalfa et al. [61]).

and lymphocytes as well as NK cell cytotoxic effector cell functions was undisturbed. In vivo studies conducted in mice with rheumatoid arthritis suggested that PDX nanofibers had anti-inflammatory effects as evidenced by an increase in IL-10 [56]. Moreover, the mechanical properties of PDX are more attractive than those of PLAs. PDX shows greater softness and flexibility compared to polylactides/glycolides due to the presence of the ether bond and the additional methylene group. Electrospinning of PDX was first reported by Boland et al. in 2005 using 1,1,1,3,3,3-hexafluro-2-propanol (HFIP) as a solvent [13]. Both aligned and random fibrous structures could be obtained by varying the mandrel target speed. Pore area, modulus of elasticity, peak stress and strain at failure could be tailored by careful variation in solution concentration. In general, the aligned fibers tested in the longitudinal direction had higher moduli, peak stresses and lower strain at break than the corresponding fibers tested in the orthogonal direction. Importantly, the mechanical properties of electrospun PDX were comparable to the two major structural components of soft tissues i.e. collagen and elastin as can be noted from Table 1 [57]. Peak stress exhibited by the electrospun PDX is within the range of elastin and near the lower limit of collagen. The percent strain at a break range of electrospun PDX spans both the ranges of elastin and collagen, and exceeds the upper limit of collagen. These data show that from a mechanical standpoint the bulk

Table 1 Comparison of mechanical properties of electrospun PDX with soft tissue components. Polymers

Modulus (MPa)

Peak stress (MPa)

Strain (%)

Reference

PDX PLLA PCL Collagen Elastin

2–46 Approx. 56 3.2 100–2900 0.3–0.6

1.7–12.1 – 4.3 5–500 0.36–4.4

31–240 – 103.2 5–50 100–200

[13] [59] [60] [58] [13]

material properties of electrospun PDX are of the same order of magnitude as the major structural components of the native vascular ECM. In addition, an electrospun PDX patch was investigated for potential use in rotator cuff repair [58]. Tendon derived cells adhered and proliferated very well on the electrospun PDX mat for a period of up to 21 days, with no visible clumps of cells. Cells appeared elongated along the electrospun nanofibers while forming numerous cell–cell contacts. 3.4. Blends of PDX for tissue engineering applications Blends of PDX with natural and synthetic polymers have been electrospun or processed by other techniques for applications such as vascular tissue engineering, bone tissue engineering, urologic tissue engineering and cartilage tissue engineering. An overview of these different tissue engineering applications will now be presented. 3.4.1. Vascular tissue engineering Electrospun biodegradable scaffolds such as poly(lactic acid) (PLA), poly(glycolic acid) (PGA), poly(caprolactone) (PCL), in addition to natural polymers such as collagen and elastin have been extensively investigated for vascular tissue engineering. However, most of these materials by themselves have had limited success. PGA ultimately leads to failure of the graft material under physiological pressures due to rapid degradation. Additionally, PGA has the potential to produce inflammation at the site of the graft or downstream due to its rapid degradation and the subsequent decrease in local pH levels. Although PLA degradation is slower than PGA, its increased stiffness makes it to be less desirable when designing a vascular graft. On the other hand, shape memory of PDX allows it to become a promising material for the creation of tissue engineered scaffolds. Fabrication of PDX into vascular conduits provides rebound and kink resistance. This is particularly useful when designing small diameter vascular grafts where the

378

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

latter is more likely to collapse after clamping. This unique property combined with the polymer‘s excellent biocompatibility, strength and elasticity may be highly beneficial in the creation of vascular grafts [13]. Kalfa et al. [61] reported on the use of a PDX electrospun valved transannular patch to replace the right ventricular outflow in lambs. The patch was seeded with mesenchymal stem cells and was implanted in the lambs at the age of 3 months. Polarized light microscopy confirmed the complete disappearance of the PDX scaffolds from 4 months onward the time of implantation. A three layered tissue, comprising of an inner endothelium like structure, a central structure rich in collagen, glycosaminoglycans, and elastic fibers resembling an ECM and also a vascularized loose connective tissue layer was formed as demonstrated by echography and MRI. The composition of this neo-matrix evolved between 1 and 4 months to become closer to the extracellular matrix of the native tissue with a decreasing glycosaminoglycan concentration (Fig. 7). An increase in the diameter of the reconstructed pulmonary artery from 16.5 to 17.7, 18.9 and 20 mm at 1 month, 4 months and 8 months respectively was observed. Moreover, no aneurismal evolution was detected. Uncross-linked and cross-linked PDX/elastin scaffolds were fabricated by McClure et al. [62]. PDX/elastin scaffolds exhibited mechanical properties resembling that of both pig artery and native human artery. Following this study, Sell et al. [63] designed an electrospun cardiovascular graft composed of PDX and elastin. PDX was chosen to provide mechanical integrity to the prosthetic, while elastin provides elasticity and bioactivity. Addition of elastin significantly reduced peak stress and strain at break of the scaffolds. Results showed that the 50:50 PDX/elastin closely mimic the compliance of native femoral artery, while grafts that containing less elastin exceeded the suture retention strength of native vessel. Preliminary cell culture studies using human dermal fibroblasts (HDFs) showed that cells cultured on PDX/elastin scaffolds migrated through their full thickness within 7 days, but failed to migrate into pure PDX scaffolds. Electrospun tissue engineering scaffolds composed of PDX, elastin and collagen were investigated for use as a potential vascular graft [64]. PDX was chosen to provide tensile support and to prevent vessel rupture while elastin and collagen would provide elasticity and bioactivity respectively. Uniaxial tensile test results demonstrated an overall decrease in peak stress, tangential modulus, and strain to break with increasing elastin and collagen contents. Results showed that the ECM of pig femoral artery stress relaxed significantly more than any of the electrospun polymer blends tested, suggesting a clear disparity in viscoelastic properties. However, both pig artery and the blended scaffolds reached a constant tension after 400 cycles. McClure et al. [65] fabricated artificial grafts composed of PCL/silk and PDX/silk in blend ratios 100/0 and 50/50. Increasing mandrel speeds had no effect on fiber diameter but increased the fiber packing density resulting in a more compact matrix. Better fiber alignments were obtained with increasing rotational speeds. Higher rotational rates were found to result in larger average peak stress and modulus in all samples except for PDX which showed an opposite trend. Most of the blends electrospun at high rotational rates showed an average burst pressure superior to that of the human saphenous vein (HSV), while the majority of the low rotational ones were below the HSV average. Smith et al. [66] designed a dual-layered 1.5 mm ID electrospun PDX/elastin tube with either one or two 6-0 PDS™II bioresorbable PDX sutures wound around the tubes between the layers. Cross-linked suture-reinforced PDX/elastin conduits had burst pressures more than 10 times greater than normal systolic pressures and showed a range of compliance values, within those of the native artery.

Moreover, a trilayered tubular conduit (4 mm) consisting of blends of PDX, elastin and gelatin was fabricated [38]. The first, second and third layers comprised of blends of elastin/gelatin (EG), PDX/elastin/gelatin (PEG) and PDX/gelatin (PG) respectively. SEM showed a randomly oriented nanofibrous structure, with a well inter-connected network of pores with porosity of around 80%. Uniaxial tensile testing of the hydrated scaffold revealed a tensile strength of 1.77 MPa and a modulus of 5.74 ± 0.30 MPa with a failure strain value of 75 ± 10%, comparable to that of native arteries. 3.4.2. Bone tissue engineering Hydrophobicity of PLAs in addition to their acidic degradation products is a major limitation. Due to their inherent hydrophobic nature, PLAs reduce expression of phenotypic markers by adherent osteoblasts, and further influence cell attachment and proliferation due to the denaturation of adhesive ligands of cells, which are undesirable [67]. Moreover, despite being biocompatible, the clinical application of pure PLGA for bone regeneration is hampered by poor osteoconductivity and exhibits suboptimal mechanical properties for use as load-bearing applications [3]. This led researchers to consider the use of PDX as scaffolds for bone tissue engineering. As reported by Madurantakam et al. [55], one of the major limitations in scaffold-based bone tissue engineering concerns the inability to increase the loading of biologically active inorganic mineral. PDX/Hydroxyapatite (HA) was blended and electrospun in view of overcoming this problem [55]. The amount of mineral deposited was found to vary directly with HA concentration, thereby confirming that HA acted as nucleation sites for further crystal growth. In addition, better scaffold mineralization was observed for PDX/50% HA compared to the corresponding PLGA scaffolds in ionic simulated body fluid (i-SBF) and revised simulated body fluid (r-SBF), both having ion concentrations equal to those of blood plasma in dissociated and total amounts, respectively [55]. The mineralization potential of electrospun PDX/HA/fibrinogen (PDX/HA/Fg) blended scaffolds in different simulated body fluids (SBF) was investigated by Rodriguez et al. [68]. They found that mineralized electrospun Fg scaffolds without PDX were mechanically stable only for the first 5 days, but they had superior mineralization capabilities which produced a thick bone-like mineral (BLM) layer throughout the scaffolds. 50/50/0 scaffolds incubated in r-SBF and c-SBF for 5 and 14 days respectively produced scaffolds with high mineral content and individual-mineralized fibers. These mineralized scaffolds were still porous and were considered as promising candidate in optimization studies as bone substitute. BMPs immobilized-PDX/Pluronic F127 porous particles were developed as a bone graft using a melt-molding particulateleaching method. Single BMP-2 or dual BMP-2/BMP-7 were easily immobilized onto the porous surface of the particles via heparin binding and were released in a sustained manner regardless of BMP type [69]. In vitro and in vivo experiments demonstrated that compared to porous particles without BMP, the continuously releasing BMP particles provided a better environment for bone marrow stem cells (BMSC) osteogenesis and new bone formation respectively. Greater bone regeneration was noted for the BMP-2 and dual BMP groups compared to the BMP-7 group. Overall, the authors concluded that the system may be considered as a promising bone graft candidate for delayed and insufficient bone healing. Lee et al. [70] tried to achieve tissue-engineered bone formation using periosteal-derived cells and a PDX/pluronic F127 scaffold with pre-seeded adipose tissue-derived CD146 positive endothelial-like cells. The scaffold was prepared using a modified melt-molding particulate-leaching method. The periosteal-derived cells seeded on the PDX/pluronic F127 scaffold grew into the scaffold as a thick and dense cell layer. Radiographic experiments carried out to examine in vivo bone

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

379

Fig. 8. Optical micrographs of hematoxylin and eosin (H and E) stained electrospun PDX/fibrinogen composite scaffolds at 20 times magnification (reproduced with permission from McManus et al. [71]).

formation in a pig model revealed that newly formed bone was seen more clearly in the defect with periosteal-derived cells and PDX/pluronic F127 scaffold with pre-seeded adipose tissue-derived CD146 positive endothelial-like cells compared to other defects with periosteal-derived cells and PDX/pluronic F127 scaffold without adipose tissue-derived CD146 positive endothelial-like cells.

3.4.3. Urologic tissue engineering PDX/fibrinogen composite scaffolds have been investigated by McManus et al. for urologic TE applications [71]. Fibrinogen was chosen due to its innate ability to induce cellular interaction and subsequent scaffold remodeling. Increasing fibrinogen content led to increased cell migration and collagen deposition while increasing PDX concentration improved the scaffold mechanical

380

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

properties. The fibrinogen to PDX ratio could be adjusted to achieve the required properties for specific tissue engineering applications. In general, modulus of elasticity and peak stress data varied linearly with increasing PDX concentration (0–50% PDX). In vitro cell culture studies using human BSM (bladder smooth muscle cells) showed that the cells migrated through the full thickness of the scaffolds by day 7. Increased cell spreading and cell layering were noted at higher PDX concentrations, showing that the scaffold supported cells and directed cell growth and migration (Fig. 8). Overall, scaffolds with higher PDX content allowed for slower scaffold remodeling and stimulated cells to adopt morphologies closer to their native state [71].

thereby showing increasing miscibility of PCL and P(DX-co-MeDX) segments. The activation energies of thermal degradation of the P(DX-co-MeDX) segment in the copolymer were found to decrease linearly with increasing MeDX content of the copolymer. In vitro hydrolytic degradation studies showed that the copolymers degraded via bulk erosion, with copolymers containing a higher mol% of MeDX degrading at a faster rate. Overall, increasing MeDX content influenced both thermal properties and degradation kinetics through phase mixing of segments in the copolymers. The possibility of controlling miscibility and degradation of copolymers through incorporation of MeDX units allow designing tissue engineering scaffolds for several applications.

3.4.4. Cartilage tissue engineering Cartilage is a specialized tissue consisting of chondrocytes and an abundant amount of ECM. Cartilage tissue engineering emerged as a potential solution to the limited regeneration capacity of cartilaginous tissue after injury. Scaffold consisting of a 90/10 PCL/PDX blend was fabricated using compression molding and salt-leaching techniques [72]. Chondrocytes were found to adhere well to the scaffold and the cells showed a continuous monolayer and flattened-spindle-like appearance. Moreover, the cells expressed collagen type II after 3 weeks in culture. In addition, Fontana et al. [73] evaluated the primary stability of a PLGA/PDX membrane for autologous chondrocyte implantation inserted into defects in the femoral head and acetabulum. The implanted membranes showed stability in 83.3% and 33.3% of the acetabular defects and femoral defects respectively after 50 cycles. Results indicated that this polymeric membrane has potential clinical applications in autologous chondrocyte implantation. Scheme 3 summarizes the different tissue engineering applications that have been under study for PDX-based blends.

3.5.2. Blend films Blend films of semi-crystalline PDX and amorphous PMeDX have been prepared by solvent casting method and fully characterized in terms of their mechanical performance, thermal and degradation behaviors [75]. Viscosity analysis coupled with SEM and AFM images indicated immiscibility of the blends over the whole range of compositions. Incorporation of PMeDX in the blend films caused a significant decrease in the melting temperatures (Tm) compared to that of PDX homopolymer. As demonstrated by thermal analysis, mechanical properties and surface morphology images, low PMeDX contents in the blends acted as plasticizer. The plasticizing effect on blends increased with decreasing reduced viscosity of PMeDX (molar mass). Lower molecular weight-PMeDX had a greater plasticizing effect on PDX. In vitro hydrolytic degradation studies in PBS showed that blend films with higher PMeDX contents degraded at faster rates with profiles differing from PDX film. Although a decrease in mechanical properties is noted upon PMeDX addition to PDX, blend films can be a means to tune degradation rates.

3.5. New poly(ester-ether)-based materials for tissue engineering

3.5.3. Nanofibers Copolymerization of monomers or blending of polymers allows adjustment of biological, chemical and mechanical properties for tissue engineering applications. Electrospun diblock PCL-b-PDX nanofibrous scaffold with average fiber diameter of 3 lm was fabricated by Kratz et al. [76]. The nanofibers had excellent shape memory properties with high recovery rate (Rf) values in the range of Rf = 92–98% and a recovery stress (rmax) values of 4.6–5.0 MPa. Hydrolytic degradation studies showed a linear mass loss profile

3.5.1. Diblock polyester-poly(ester-ether) copolymers PCL-b-poly(DX-co-MeDX) were synthesized in a range of compositions of the two segments and with varying MeDX units [74]. The thermal properties, miscibility and hydrolytic degradation of these copolymers were studied and compared with PDX. Only one crystallization exotherm was noted for all copolymers, decreasing with increasing MeDX content in the copolymer,

Urologic • Appropriate mechanical properes • Slow remodeling with cell infiltraon

Bone • •

Carlage

TE applicaons

Good structural integrity Good mineralizaon potenal

• Appropriate mechanical properes • Cell migraon

Vascular • • •

Shape memory Elascity, strength Cell infiltraon

Material/ Blend

Modulus/MPa

Peak stress/MPa

Compliance/mmHg

PDX/elastin 50/50 PDX/elastin/collagen 50/25/25 Pig artery PDX/silk 8000 rpm PDX/elastin/gelatin

9.64 ± 0.66 7 6 3 5.74 ±3

3.25 ± 0.24 2 4 2 NA

3.8 NA NA 1 NA

Burst pressure/mm Hg NA NA NA 3336 NA

Scheme 3. Summary of TE applications investigated for PDX based blends [38,63–65].

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 9. Effects of DX:MeDX ratio copolymer ratio and fiber diameter on crystallinity (adapted with permission from Wolfe et al. [77]).

with a continuous decrease in elongation at break from 800% to 15% observed during a time period of 92 days. Random P(DX-co-MeDX) copolymers were electrospun from a solution of HFIP and the thermal and mechanical properties of the resulting electrospun nanofibers compared with those of PDX [77]. Uniaxial tensile testing revealed that the addition of MeDX alters the mechanical properties of the electrospun mats. A decrease in melting and crystallization temperatures was also noted due to increasing amorphous character of the copolymer with increasing MeDX content. For copolymers with comparable molar masses, a linear relationship between the crystallinity of the electrospun fibers and their fiber diameter was noted (Fig. 9). The greater the % crystallinity, the larger the fiber diameter. The wide range of fiber diameters, as well as mechanical and thermal properties should allow for the fabrication of scaffolds to meet a variety of biomedical engineering applications.

381

The fabrication of electrospun PDX/PMeDX nanofibrous mats was also reported in an in-depth study of their thermal, mechanical and degradation behaviors and a comparison with their blend films [78]. Increasing degree of morphological heterogeneity was noted with increasing PMeDX content. Electrospun PDX/PMeDX blend fibers exhibited a melting transition at 109 °C independently of the PMeDX content, which corresponds to the melting of PDX nanofibers. The Young’s moduli decreased as the PMeDX content of the fibers increased and an increase in strain at break and peak stress was noted as a result of a decrease in the fiber diameter. Electrospun mats were thermally more stable than blend films. In the same study, hydrolytic degradation of the mats followed a surface erosion mechanism as opposed to bulk degradation observed for blend films. Degradation of fibers was found to be mainly dependent on their diameter (Fig. 10). On the other hand, hydrolytic degradation of blend films depended on the overall crystallinity of the blends. In addition, electrospun PDX/PMeDX scaffolds supported HDF cell attachment and proliferation. In fact, a greater density of viable cells was observed on the PDX/PMeDX scaffolds compared to electrospun PDX. Cell migration analysis revealed that on average, the cells migrated up to a maximum of 45.1% (±11.8%) through the 98/2 scaffold after 7 days while almost no cell infiltration was observed in electrospun PDX mat. This study shows the potential of these electrospun poly(ester-ether) blend nanofibrous scaffold for tissue engineering applications. The physico-chemical properties (thermal and mechanical properties), degradation behavior and cellular response of electrospun mats of semi-crystalline PCL and PLLA blended with amorphous poly(methyl dioxanone) (PMeDX) were recently investigated [79]. Almost no change in crystallization temperature of PCL was noted with increasing PMeDX content, suggesting immiscibility of the two homopolymers. The appearance of two crystallization peaks for PLLA/PMeDX blends suggested stereocomplex formation. Initial degradation of electrospun mats over a period

Fig. 10. SEM (500 magnification, scale bar = 50 lm) of electrospun mats of 98/2 and 85/15 at weeks 0 and 5 (reproduced with permission from Goonoo et al. [78]).

382

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 11. SEM image showing bipolar spindle (A), cobble-stone morphology (B) and fluorescence microscopy image showing 100% HDF infiltration in 85/15 PLLA/PMeDX (C) (adapted with permission from Goonoo et al. [79]).

Table 2 PDX-based microspheres. Co/polymer

Diameter (lm)

Reference

Starch-g-PDX CDA-g-PDX

20 5

[41] [81]

of 5 weeks appears to occur via a surface erosion mechanism. A greater density of viable HDF cells was noted on electrospun PCL/PMeDX and PLLA/PMeDX scaffolds compared to PCL and PLLA mats respectively. A change in cell morphology was noted with increasing PMeDX content from bipolar spindle (Fig. 11A) to a cobble-stone morphology (Fig. 11B) which was more pronounced with PLLA/PMeDX than PCL/PMeDX. Indeed, the presence of cobble-stone HDF morphology was detected as from 15 wt% PMeDX for PCL/PMeDX mat and as early as 2 wt% PMeDX for PLLA/PMeDX mat. No cell infiltration was noted for PCL/PMeDX mats independently of PMeDX composition and varying extent of HDF migration was observed for PLLA/PMeDX depending on blend composition. Noteworthy is the fact that HDFs infiltrated through the entire thickness of electrospun 85/15 PLLA/PMeDX scaffold (Fig. 11C). 3.6. Drug delivery applications Applications of PDX have been mainly restricted to the medical field and tissue engineering. Studies toward drug delivery applications have been attracting interest for the past decade or so prompted by the search for more efficient polymeric drug delivery systems and the advantages conferred by nanoformulations. This section will review the types of PDX-based drug and gene delivery systems developed so far, more specifically microspheres/microparticles, nanoparticles, nanomicelles and electrospun nanofibers. 3.6.1. PDX-based drug/gene carriers Microparticles and nanoparticles are usually made of biodegradable polymers that can encapsulate a variety of drugs. Microparticles/microspheres are solid, nearly spherical particles in the size of 1–1000 lm whereas nanoparticles are mostly in the range of 200–800 nm. The larger size of microparticles allows their use to deliver not only drugs but also peptides and hormones [80]. Nanomicelles consist of self-assembled amphiphilic polymer chains, having diameters ranging from 10 to 100 nm, characterized by a core–shell morphology in which the hydrophobic blocks make up the inner core entrapping hydrophobic drug molecules and the hydrophilic blocks make up the surrounding corona.

3.6.1.1. Micron-sized delivery systems. 3.6.1.1.1. Microparticles and microspheres. Bovine serum albumin, a model protein drug, was encapsulated within PDX, PBDX [poly(5benzyloxymethyl-1,4-dioxan-2-one] or P(HDX-co-DX) [poly(5-hy droxymethyl-1,4-dioxan-2-one-co-DX)] to form microparticles. Results indicated that the drug loading and encapsulation efficiency increased with increasing hydrophobicity (higher DX content) as confirmed with a P(HDX-co-DX) 12:88 microparticle compared with a 57:43 microparticle. P(HDX-co-DX) also showed a faster release rate than PBDX and PDX [43]. Well-defined starch-g-PDX copolymers were synthesized via ROP of DX with an acetylated starch initiator. Microspheres with an average diameter of 20 lm with potential use in controlled release of drugs were successfully prepared (Table 2) [41]. Another graft copolymer was synthesized by Zhu et al. [81] based on synthetic cellulose diacetate, CDA-g-PDX. Porous microspheres with a diameter of about 5 lm could be obtained through the solvent evaporation of water-in-oil emulsion, with potential applications in drug delivery (Table 2). 3.6.1.2. Nano-sized DDS. 3.6.1.2.1. Nanoparticles. Chitosan-g-PDX copolymers were synthesized via two different pathways. Liu et al. [82] used bulk ROP of DX initiated by the hydroxyl group or amino group of chitosan in the presence of Sn(Oct)2 as catalyst. The copolymers were loaded with ibuprofen and in vitro release in PBS (pH = 7.4) showed an initial burst release followed by sustained release. The release of ibuprofen from the copolymers was slower (58.7–72.9%) compared with the release from pure chitosan carrier (86.9%). The release rate decreased with increasing degree of grafted PDX chains. Wang et al. [83] applied a protection-graft-deprotection technique using N-phthaloyl-chitosan to access the same chitosan-g-PDX. The copolymers were loaded with the analgesic drug sinomenine and release behavior was recorded in both artificial gastric juice and phosphate buffer solution. The graft copolymer showed sustained release in both media compared to chitosan. However, the release rate was dependent on the hydrophobicity, that is, the number of PDX chains. For instance over 5 h, cumulative release decreased from 100% to 42.5% in PBS and 85.2% to 36.4% in artificial gastric juice as the copolymer hydrophobicity increased, slowing down water penetration and thus drug release. Generally, a higher drug release is observed in gastric juice than in PBS as rapid degradation of the polymer, demicellization of the drug-loaded micelles or hydrolysis of linkages conjugating drugs to polymer backbones or antibodies occur in an acidic medium [84]. Poly(x-pentadecalactone-co-dioxanone) [poly(PDL-co-DX)] copolymers, prepared by Liu et al. [85], gave tissue responses comparable to PDX in in vivo experiments using mice model.

383

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 12. TEM micrographs of mbPDX8-b-PEG12.5 (A), mbPDX8-b-PEG45.5 (B), mbPDX11-b-PEG12.5 (C) and linear PDX8-b-PEG45.5 block copolymer (D) (reproduced with permission from Chen et al. [87]).

Poly(PDL-co-DX) with up to 69 mol% DX units were successfully loaded with either doxorubicin or a polynucleotide, siRNA as reported in Table 3. Biphasic release was noted for both DOX and siRNA consisting of an initial burst followed by sustained release. About 30–40% of the total encapsulated DOX was released within the first day, with an additional 30–40% of the drug being released over the next 20 days. For siRNA-loaded nanoparticles, an initial burst extended over the first 3 days (20%) followed by sustained release over the following 60 days (additional 25%). The authors also entrapped siLUC into the same nanoparticles with an encapsulation efficiency of 27%. siLUC-loaded nanoparticles were tested for their effectiveness in inhibiting luciferase reporter gene expression in LUC-labeled RKO human cancer colon cells. At 48 h posttransfection and 50 nM siLUC dosage, there was a 76% and 50% reduction in luciferase gene expression by the combination of siLUC and LF2K

(commercial transfection agent, Lipofectamine 2000) and siLUC-loaded nanoparticles, respectively. After 72 h, there was no further decrease in the case of siLUC/LF2K while an additional decrease of 14% was noted with siLUC-loaded nanoparticles. This offers the possibility of using the poly(PDL-co-DX) nanoparticle for controlled and continuous release of siLUC to control the expression of target genes in cells over time [85]. 3.6.1.2.2. Nanomicelles. Copolymerization allows tailoring of polymer properties with either the resulting copolymer having properties ranging between those of the homopolymers or having new intrinsic properties. Copolymers of PDX with poly(ethylene oxide) (PEO), PCL, PGA, PLA and trimethylenecarbonate (TMC) possessed new physico-chemical properties [19]. Amphiphilic block or graft copolymers, characterized by a hydrophilic segment chemically tethered to a hydrophobic segment, have been used extensively

Table 3 PDX-based nanoparticles. Co/polymer

Drug

DDS characteristics

Reference

CS-g-PDX

IBU SIN

Controlled release of drug in PBS Controlled drug release in artificial gastric juice and PBS

[82] [83]

Poly(PDL-co-DX)

DOX or siRNA

Diameter = 190–250 nm DOX: 30–40% released on day 1 followed by an additional 30–40% release over the next 20 days siRNA: 20% release over the first 3 days followed by an additional 25% over the next 60 days

[85]

384

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 13. TEM images of (a) blank microspheres and (b, b1) DOX-loaded micelles (reproduced with permission from Zhang et al. [93]).

O

O

O

HN

HN

HN O

O

O

O

O

3(S)-methyl-morpholine2,5-dione (MMD)

3(S)-isobutyl-morpholine-2,5dione (BMD)

O

3(S)-sec-butyl-morpholine2,5-dione (SBMD)

Fig. 14. Monomers of depsipeptides copolymerized with dioxanone.

in pharmaceutical applications for drug and gene delivery. In aqueous solution, polymeric micelles are formed at or above the critical micelle concentration. Upon micellization, the hydrophobic core can be used to encapsulate hydrophobic drugs or genes. PDX-based amphiphilic block or graft copolymers have been synthesized in recent years using the classical hydrophilic polymer, PEG [86,87]. Other hydrophilic polymers such as cellulose diacetate [81], poly(vinyl alcohol) [88], dextran [89] and chitosan [90] have also been reported. A series of multiblock PDX–PEG copolymers, synthesized via coupling the prepolymers of dihydroxyl-terminated PDX and dicarboxylated PEG were found to self-assemble in aqueous solution to give micelles of average size of 90 nm at 40 °C, with an increase in the micellar size (98–100 nm) as the temperature decreased below 40 °C [77]. Chen et al. [87] also synthesized

multi-branched PEG-b-PDX copolymers and observed that the molecular characteristics of the multi-branched copolymers, such as chain length of the blocks and branch density, are easily tunable. The multi-branched copolymers formed ‘‘star anise’’-like micelles in aqueous solution due to the presence of the branched and crystallizable PDX blocks. In comparison, linear PEG-b-PDX copolymers formed more spherical micelles (Fig. 12). Bhattarai et al. [91] synthesized P(DX-co-LLA)-bPEG-b-P(DX-co-LLA) triblock copolymers giving micelles in the size range 60–165 nm. The particle size increased with increasing copolymer concentration in acetonitrile. For instance, a 3 mg/mL copolymer solution afforded micelle of size 60 nm whereas micelles of 152 nm were obtained when the concentration was increased to 100 mg/mL. Bahadur et al. [92] also reported the synthesis of triblock copolymers P(DX-co-CL)-b-PEG-b-P(DX-co-CL).

385

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391 Table 4 PDX-based micelles. Co/polymer

Drug

DDS characteristics

Reference

CS-b-PDX

CPT

Diameter = 77–115 nm Encapsulation efficiency: 82% Drug release: after 100 h, 71–85% Cytotoxicity: drug-free micelles showed good compatibility with L929 fibroblasts and HeLa cells

[90]

P(MMD-co-DX)-b-PEG-b-P(MMD-co-DX)

DOX

Diameter = 50–100 nm CMC = 0.41–0.66 lg/mL Encapsulation efficiency >95% Drug release: sustained release of DOX up to 65% over 72 h

[93]

PDMAEMA-b-P(IBMD-co-DX)-b-PEG-b-P(IBMD-coDX)-b-PDMAEMA

IBU and DOX

Diameter = 80–120 nm Combined encapsulation of IBU and DOX Encapsulation efficiency: up to 61.29% Drug release: sustained release of IBU–DOX in PBS

[94]

P(SBMD-co-DX)-b-PEG-b-P(SBMD-co-DX)

DOX

Encapsulation efficiency: 62.5% Drug release: Sustained release over 72 h

[95]

MPEG-b-P(DX-co-MeDX) and P(DX-co-MeDX)-bPEG-b-P(DX-co-MeDX)

KET

Drug loading: 36–86% Sustained in vitro drug release over 7 days

[96]

MPEG-b-P(DX-co-MeDX) and P(DX-co-MeDX)-bPEG-b-P(DX-co-MeDX)

RIF, PZA and INH

Encapsulation efficiency Single-loaded: RIF = 45–74%, PZA = 26–28% Dual-loaded RIF and INH = 42–79% and 17–25% Sustained drug release: RIF: 9 days, PZA: 5 days, INH: 4 days Hemolysis: No erythrocyte agglutination between 0.125 and 1 mg/mL for drugfree micelles Brine shrimp lethality assay: no toxicity profile at 200 lg/mL for drug-free micelles (<50% mortality)

[98]

MPEG-b-P(DX-co-MeDX) and P(DX-co-MeDX)-bPEG-b-P(DX-co-MeDX)

DOX, PTX and CPT

Encapsulation efficiency DOX: 16–46%, PTX: 52–60%, CPT: 49–64% Sustained drug release: 5 days for all three drugs

[99]

MPEG-P(CL-co-DX)

CUR

Diameter = 30 nm Encapsulation efficiency >95% Drug release: Max release of 75% after 308 h Cytotoxicity: micelles inhibited the growth of PC-3 human prostate cancer cells in a dose-dependent manner

[100]

CS-g-PDX

NIM

CMC = 4.8  104–1.26  103 mg/mL Diameter = 40–130 nm Encapsulation efficiency: 30% Drug release: 71–85% over 100 h

[102]

Scheme 4. Self-assembly mechanism of PDMAEMA-b-P(IBMD-co-DX)-b-PEG-b-P(IBMD-co-DX)-b-PDMAEMA (reproduced with permission from Zhang et al. [94]).

The latter were found to self-assemble in aqueous solution, giving rise to micelles with average diameters ranging from 90 to 113 nm. Amphiphilic block copolymer micelles of chitosan and PDX were synthesized for the first time by Tang et al. [90] wherein antitumor drug camptothecin (CPT) was encapsulated. The release of CPT from the micelles was faster at pH 5.0 than at pH 7.4. Drug-free micelles were nontoxic in preliminary in vitro cytotoxicity assays.

Zhang et al. [93] prepared triblock P(MMD-co-DX)-bPEG-b-P(MMD-co-DX) copolymers (Fig. 13) by ROP of 3(S)-methyl-morpholine-2,5-dione (MMD) (Fig. 14) and DX in the presence of PEG(6000). DOX was encapsulated by hydrogen-bond interaction in the stable micelles and was released at a sustained rate of 72 h, with a maximum of around 65%. The release rate of DOX from the micelles was faster in pH 4.0 than pH 7.4 PBS (Fig. 15). The faster release of DOX in acidic condition could be

386

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391 O

O

H3CO

O O

O

H

O

x

O

m

n

MPEG-b-P(DX-co-MeDX) Sn(Oct)2 O

O

O

H3CO

O

H O

MPEG

+ O DX

x

O MeDX H2N

O O

y

x

Sn(Oct)2

NH2

O

z

Jeffamine® ED-2003

O H

O

O

O Om

O

H N On

HN

O O

O

m

O

O

H

n

P(DX-co-MeDX)-b-PEG-b-P(DX-co-MeDX) Scheme 5. Synthesis of diblock MPEG-b-P(DX-co-MeDX) and triblock P(DX-co-MeDX)-b-PEG-b-P(DX-co-MeDX) copolymers (reproduced with permission from BhawLuximon et al. [97]).

due to cleavage of hydrogen bonds formed between the copolymer and DOX at pH 4.0 and the degradation of hydrophobic P(MMD-co-DX) micellar core being faster in pH 4.0. The same authors reported on a multiblock copolymer system, PDMAEMA-b-P(IBMD-co-DX)-b-PEG-b-P(IBMD-co-DX)-b-PDMAEMA, synthesized from DX and depsipeptides monomers (Fig. 13) coupled with PDMAEMA and PEG using a combination of ROP and atom transfer radical polymerization [94]. The copolymers readily self-assembled into nanomicelles in aqueous solution, with average diameter 70–110 nm. The amphiphilic nanoparticles were composed of hydrophobic P(IBMD-co-PDX) blocks forming the core while the hydrophilic PEG and PDMAEMA blocks formed the corona shell (Scheme 4). Ibuprofen and doxorubicin as a combined model drug was encapsulated within these particles with diameter 80–120 nm, high encapsulation efficiencies and sustained release of IBU–DOX in phosphate buffered solution. Lv et al. [95] recently synthesized triblock copolymer P(SBMD-co-PDX)-b-PEG-b-P(SBMD-co-PDX) by the ROP of 3(S)-se c-butyl-morpholine-2,5-dione (SBMD) (Fig. 13) and DX with PEG as the initiator, which self-assembled into micelles in PBS solution. Doxorubicin was successfully encapsulated into the micelles with encapsulation efficiency of around 60%. The release of doxorubicin was faster at pH 4 and also when triblock copolymers of lower molecular weight/lower PDX composition were used. Jeetah et al. [96] synthesized diblock MPEG-b-P(DX-co-MeDX) and triblock P(DX-co-MeDX)-b-PEG-b-P(DX-co-MeDX) copolymers using MPEG and di-amino-terminated PEG-based JeffamineÒ ED-2003 as a macroinitiator in the presence of Sn(Oct)2 (Scheme 5). The copolymers shifted from semi-crystalline to amorphous as the ratio of MeDX increased in the copolymers. These amphiphilic block copolymers gave rise to micelles in aqueous solution which were loaded with anti-inflammatory drug ketoprofen, anti-tuberculosis drugs rifampicin, pyrazinamide and isoniazid and anti-cancer drugs doxorubicin, paclitaxel and camptothecin [96,98,99]. Rifampicin was found to have the highest encapsulation in both single and dual drug-loading (Table 4). Interestingly, the in vitro drug release could be controlled by the MeDX content of the copolymers. Copolymer micelles with higher MeDX content underwent faster degradation and therefore faster diffusion of drug from the micellar core was observed. The

Fig. 15. In vitro release profiles of free DOX and DOX-loaded microspheres in PBS solutions (pH 4.0 and 7.4) at 37 °C (reproduced with permission from Zhang et al. [93]).

chemical structure of the encapsulated drug was also found to have a bearing on the release, i.e. the stronger the polymer–drug interaction/binding, the slower is its release from the micelles. For instance, the release of drug was fastest for doxorubicin, followed by camptothecin and paclitaxel (Fig. 16). These results are in line with the determined binding constants of drugs to the micellar core. Hemolysis studies and brine shrimp lethality assays confirmed the absence of toxicity of copolymers with LD50 at 200 lg/mL (<50% mortality of shrimps). Song et al. [100] prepared polymeric micelles of MPEG-P(CL-co-DX) copolymers and loaded curcumin by a solid dispersion method with a high encapsulation efficiency (>95%). The 30-nm sized curcumin-loaded micelles were stable for at least 24 h at room temperature when lyophilized and reconstituted in water. Curcumin was slowly released from the micelles without any burst effect with a maximum release of around 75% occurring after 308 h. The cytotoxicity assay indicated that curcumin-loaded MPEG-P(CL-co-DX) micelles inhibited the growth of PC-3 human

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

Fig. 16. Release profiles of doxorubicin, paclitaxel and camptothecin-loaded P(DX-co-MeDX)-PEG-P(DX-co-MeDX) copolymer micelles, %DX:%MeDX = 92:8 in PBS (pH = 7.4) as a sink solution at 37 °C. Each value represents the mean ± S.D. (n = 3). P < 0.05 (reproduced with permission from Jeetah et al. [99]).

prostate cancer cells in a dose-dependent manner. There was no significant difference in cell viability at 2.5 and 5 lg/mL between free curcumin and curcumin-loaded micelles while the curcumin-loaded micelles showed lower cytotoxicity than free curcumin at higher curcumin concentrations. Graft copolymers of PDX with hydrophilic counterparts also form micellar structure in aqueous solution. PDX-g-PVA copolymers were synthesized via a three-step reaction consisting of partial silylation of the hydroxyl groups of PVA, followed by ring opening polymerization (ROP) of PDX initiated from the unprotected hydroxyl groups of silylated PVA and finally, deprotection of the silylated hydroxyl groups gave graft copolymers with well-defined architectures [88] with sizes varying from 83 to 105 nm. Similarly, dextran-g-PDX copolymers were obtained by silylation of dextran followed by ROP of DX [82]. AFM studies

387

revealed the formation of uniform and spherical micelles in the range of 30–58 nm. Hydroethylcellulose-g-PDX [HEC-g-PDX] was synthesized by coupling PDX onto the backbone of HEC in the presence of 1,6-hexamethylene diisocyanate [101]. These amphiphilic HEC-g-PDX copolymers can self-assemble into micelles in the range of 200–800 nm depending on the degree of substitution on the HEC backbone. Larger sized micelles were obtained as the degree of substitution increased. The latter increases the hydrophobic chain length/density and hence an enlargement of the micellar core due to the strong hydrophobicity and crystallization of the PDX branches [101]. Amphiphilic CS-g-PDX copolymers were synthesized by Wang et al. [102] and the self-assembled micelles were encapsulated with model drug nimodipine. The latter was released in a sustained manner from the micelles without any burst effect. MTT assay showed that CS-g-PDX copolymers exhibited low cytotoxicity and good biocompatibility with L929 mouse cells, highlighting the potential of these copolymers as drug carriers. Few reports exist for PDX-based micelles tested for gene delivery. Bhattarai et al. [103] have shown that a combination of micelles and Lipofectin could be used as a non-invasive gene delivery system for lung cancer. They combined commercial cationic lipid Lipofectin and amphiphilic triblock copolymer PEG-b-P(DX-co-LLA) into an aerosol system to deliver tumor suppressor gene PTEN to the lungs of C57BL/6 mice bearing the B16-F10 melanoma in vivo. The polymeric micelles with the combination of Lipofectin/PTEN significantly improved gene expression of PTEN with no cell toxicity or acute inflammation and lung metastasis was reduced with extension in the survival time of the mice (Fig. 17). Bhattarai et al. [104] tested their PEG-b-P(DX-co-LLA) as an enhancer for the cationic lipid mediated uptake of plasmid DNA by cells in monolayer culture. PEG-b-P(DX-co-LLA) enhanced complex formation between DNA and cationic lipid (Lipofectin). No recovery of the plasmid encoded b-galactosidase into MCF-7 breast

Fig. 17. Inhibition of B16-F10 lung metastasis by PTEN delivered to mouse lung by aerosol with Lipofectin/PTEN/polymeric micelles. (a) Tumor burden as tumor index (lung weight  number of foci  size of foci); (b) representative lungs; and (c) lung weights (reproduced with permission from Bhattarai et al. [103]).

388

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

cancer cells, CT-26 colon cancer cells and NIH 3T3 mouse embryo cells at 37 °C was observed when cells were incubated with free DNA or DNA with block copolymer. Moreover, administration of the DNA/Lipofectin complex in the presence of PEG-bP(DX-co-LLA) further increased plasmid uptake and enhancement of transfection was also noted. Increased uptake of plasmid DNA and transfection activity may be explained by the hydrophilic PEG segment helping to reduce the precipitation of DNA/polycation (DNA/Lipofectin) complex and bring about greater dispersion in the aqueous phase with less negative surface charge on the complex [104]. 3.6.1.2.3. Electrospun drug-loaded nanofibers. The potential of electrospun nanofibers for drug delivery applications has been extensively reviewed [105]. Drug-eluting nanofibrous mats using a variety of polymers including PDX have been investigated for several applications ranging from cancer therapy to wound dressings. The anti-microbial effects of vancomycin (VANC) and

rifampicin-loaded (RIF) electrospun PDX fibers were assessed by Waeiss et al. [106]. VANC and RIF were loaded in the electrospun fibers in varying wt% by solution electrospinning The effect of drug-loaded fibers on the inhibition of biofilm growth containing osteomyelitis (OM)-associated pathogens was investigated. The authors concluded that the 10% VANC + RIF-loaded PDX fibrous mat was the best combination to control bacterial growth (Fig. 18). Triclosan was successfully loaded into PDX fibers by molecular diffusion with a swelling solvent and loading by means of a coating based on polycaprolactone or polycaprolactone/magnesium stearate mixtures [107]. In another study by Bottino et al. [108], anti-microbial properties of metronidazole (MET) or ciprofloxacin (CIP)-loaded electrospun PDX fibers were investigated using Porphyromonas gingivalis/Pg and Enterococcus faecalis/Ef. The amount of drug released was dependent on the drug type and wt% of drug. CIP-containing scaffolds inhibited biofilm growth of both bacteria significantly. Conversely, MET-containing

Fig. 18. Representative SEM images of adherent Staphylococcus on electrospun matrices after in vitro biofilm formation from the following groups: (A) polydioxanone only; (B) 10% rifampicin (RIF); (C) 5% VANC and (D) 10% VANC + RIF. Images acquired using 20 kV, SS10, and WD08 at 1000 magnification, with 5000 and 10,000 magnification for insets A and B, respectively denoted as a and b (reproduced with permission from Waeiss et al. [106]).

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

scaffolds inhibited only Pg growth. Furthermore, cytotoxicity studies using human dental pulp stem cells (hDPSCs) revealed that only the 25 wt% CIP-containing scaffolds were cytotoxic. Recently, Li et al. [109] fabricated a 5-fluorouracil-loaded PDX weft-knitted stent for the treatment of colorectal cancer. Biodegradable PDX monofilament was weft-knitted into a tubular stent. 5-FU-loaded poly-L-lactide fibrous membranes were then electrospun onto the surface of the PDX stent. Increasing % of 5-FU in PLA fibrous mat resulted in faster release of 5-FU. Cell cycle and apoptosis evaluation showed that the 12.8% 5-FU-loaded PLLA membranes demonstrated better anti-tumor effect compared to the 5-FU solution at IC50 concentration. The implanted 12.8% 5-FU-loaded PLLA membranes showed better in vivo anti-tumor capability than an intraperitoneal injection of 5-FU at LD50 concentration. 4. Conclusion Restricted for a long time as the only commercial synthetic biodegradable monofilament suture, PDX and analogues are now attracting increasing interest not only for medical applications but also for tissue engineering and drug/gene delivery applications. The uniqueness of PDX by virtue of its shape memory property makes it an interesting material for use as scaffolds for TE. PDX nanofibers now find applications not only in conventional medicine but also in alternative medicine for instance in chronic inflammatory diseases. In addition to PDX, novel DX analogues and a range of new poly(ester-ether)s have been developed with tunable physico-chemical and biological properties to meet the stringent criteria of biodegradability, biocompatibility and mechanical performance of scaffolds and drug/gene delivery devices as defined by regulatory bodies. The elaboration of new PDX-based materials through copolymers, blend films, nanofibers, nanoparticles and nanomicelles opens new perspectives for various tissue engineering and drug/gene delivery applications. Much remains to be accomplished in terms of biological/compatibility testing of the newly developed materials on a range of human cells or tissues. More effort has thus to be put to drive these materials through preclinical testing and overcome the barriers of EU or FDA approval before eventual commercialization. The fabrication of nanofibers from blends of PDX and copolymers with natural polymers could also provide interesting perspectives for the next generation of smarter materials. The limited commercial supply of PDX remains a limitation at present but increasing demand for PDX and copolymers offers new opportunities for commercial ventures. In summary, as highlighted throughout this review, future developments of PDX-based materials will depend on a sound combination of macromolecular engineering, pharmaceutical design, scaffold fabrication technique and biological testing. Acknowledgments We are most thankful to the Editor for inviting us to contribute to this special edition. The collaboration we have had with Prof. Robert Gurny and his colleagues at the University of Geneva, which led to the teaching of a Masters module in Pharmaceutical Technology at the University of Mauritius and to training of students was a most enriching one. We are most indebted to him for this initiative which helped consolidate drug delivery research. We thank the Tertiary Education Commission (Mauritius) for awarding PhD scholarships to N. Goonoo and R. Jeetah. We are grateful to the Mauritius Research Council (Mauritius) for supporting biomaterials and drug delivery research at the ANDI Centre of Excellence for Biomedical and Biomaterials Research (CBBR). We are also most indebted to our international collaborators, Prof.

389

Gary Bowlin, Prof. Holger Schoenherr, Dr. Daniel Wesner, Prof. Gary Hulse, Prof. Viness Pillay and Prof. David Kaplan for their continuous support.

References [1] K.M. Nampoothiri, N.R. Nair, R.P. John, An overview of the recent developments in polylactide (PLA) research, Bioresour. Technol. 101 (22) (2010) 8493–8501. [2] J.F. Mano, R.A. Sousa, L.F. Boesel, N.M. Neves, R.L. Reis, Bioinert, biodegradable and injectable polymeric matrix composites for hard tissue replacement: state of the art and recent developments, Compos. Sci. Technol. 64 (2004) 789–817. [3] R.P. Felix Lanao, A.M. Jonker, J.G. Wolker, J.A. Jansen, J.C. Van Hest, S.C. Leeuwenburg, Physicochemical properties and applications of poly(lactic-coglycolic acid) for use in bone regeneration, Tissue Eng. B Rev. 19 (2013) 380– 390. [4] J.A. Ray, N. Doddi, D. Regula, J.A. Williams, A. Melveger, Polydioxanone (PDS), a novel monofilament synthetic absorbable suture, Surg. Gynecol. Obstet. 153 (4) (1981) 497–507. [5] A. Kalangos, J. Sierra, D. Vala, M. Cikirikcioglu, B. Walpoth, X. Orrit, J. Pomar, C. Mestres, S. Albanese, D. Jhurry, Annuloplasty for valve repair with a new biodegradable ring: an experimental study, J. Heart Valve Dis. 15 (6) (2006) 783–790. [6] J.T. Christenson, A. Kalangos, Use of a biodegradable annuloplasty ring for mitral valve repair in children, Asian Cardiovasc. Thorac. Ann. 17 (2009) 11– 12. [7] PDS Flexible Plates. . [8] A. Repici, F.P. Vleggaar, C. Hassan, P.G. Van Boeckel, F. Romeo, N. Pagano, A. Malesci, P.D. Siersema, Efficacy and safety of biodegradable stents for refractory benign esophageal strictures: the BEST (Biodegradable Esophageal Stent) study, Gastrointest. Endosc. 72 (5) (2010) 927–934. [9] R. Lischke, J. Pozniak, D. Vondrys, M.J. Elliott, Novel biodegradable stents in the treatment of bronchial stenosis after lung transplantation, Eur. J. Cardiothorac. Surg. 40 (2011) 619–624. [10] D. Vondrys, M.J. Elliott, C.A. McLaren, C. Noctor, D.J. Roebuck, First experience with biodegradable airway stents in children, Ann. Thorac. Surg. 92 (2011) 1870–1874. [11] G. Li, Y. Li, P. Lan, J. Li, Z. Zhao, X. He, J. Zhang, H. Hu, Biodegradable weftknitted intestinal stents: fabrication and physical changes investigation in vitro degradation, J. Biomed. Mater. Res. A 102 (4) (2014) 982–990. [12] P. Zamiri, Y. Kuang, U. Sharma, T.F. Ng, R.H. Busold, A.P. Rago, L.A. Core, M. Palasis, The biocompatibility of rapidly degrading polymeric stents in porcine carotid arteries, Biomaterials 31 (2010) 7847–7855. [13] E.D. Boland, B.D. Coleman, C.P. Barnes, D.G. Simpson, G.E. Wnek, G.L. Bowlin, Electrospinning polydioxanone for biomedical applications, Acta Biomater. 1 (2005) 115–123. [14] N. Doddi, C.C. Versfelt, D. Wasserman, Synthetic Absorbable Surgical Devices of Poly-dioxanone, United States Patent No. 4,052,988, 1977. [15] S. Nakatami, T. Matsumoto, Y. Nakahara, H. Akieda, T. Ishioyu, Purified salt of b-hydroxyethoxy acetic acid, purified 2-p-dioxanone, and manufacturing method therefore, United States Patent No. 6,384,241, 2002. [16] F.C. Forschner, Process for Preparing Para-dioxanones, United States Patent No. 5,310,945, 1994. [17] Y. Lochee, A. Bhaw-Luximon, D. Jhurry, A. Kalangos, Novel biodegradable poly(ester-ether)s: copolymers from 1,4-dioxan-2-one and D, L-3-methyl-1,4dioxan-2-one, Macromolecules 42 (19) (2009) 7285–7291. [18] Y. Lochee, D. Jhurry, A. Bhaw-Luximon, A. Kalangos, Biodegradable poly(esterether)s: ring-opening polymerization of D, L-3-methyl-1,4-dioxan-2-one using various initiator systems, Polym. Int. 59 (9) (2010) 1310–1318. [19] K.K. Yang, X.L. Wang, Y.Z. Wang, Poly(p-dioxanone) and its copolymers, J. Macromol. Sci. C – Polym. Rev. C 42 (3) (2002) 373–398. [20] T.C. Forschner, D.E. Gwyn, C.A. Veith, Method for preparing Poly-dioxanone Polymer, United States Patent No. 5,652,331, 1997. [21] T.C. Forschner, Method for preparing Poly-p-dioxanone Polymer, U.S. Patent 5,717,059, 1998. [22] T.C. Forschner, Method for preparing poly-p-dioxanone polymer, WO 97 (1997) 21753. [23] D.D. Jamiolkowski, H.D. Newman Jr., Medical Devices containing High Inherent Viscosity Poly(p-dioxanone), EP 0 691 359 A2, 1996. [24] H.D. Newman Jr., D.D. Jamiolkowski, Medical Devices containing High Inherent Viscosity Poly(p-dioxanone). U.S. Patent 5,869,597, 1999. [25] H. Nishida, M. Yamoshita, T. Endo, Y. Tokiwa, Equilibrium polymerization behavior of 1,4-dioxan-2-one in bulk, Macromolecules 33 (2000) 6982–6986. [26] H. Kricheldorf, R. Kreiser-Saunders, C. Boettcher, Polylactones: 31. Sn(II)octoate-initiated polymerization of L-lactide: a mechanistic study, Polymer 36 (1995) 1253–1259. [27] G. Schwach, J. Coudane, R. Engel, M. Vert, More about the polymerization of lactides in the presence of stannous octoate, J. Polym. Sci. A: Polym. Chem. 35 (1997) 3431–3440. [28] A.J. Nijenhuis, D.W. Grijpma, A.J. Penjings, Lewis acid catalyzed polymerization of L-lactide. Kinetics and mechanism of the bulk polymerization, Macromolecules 25 (1992) 6419–6424.

390

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391

[29] A. Kowalski, A. Duda, S. Penczek, Mechanism of cyclic ester polymerization initiated with tin(II)octoate. 2. Macromolecules fitted with tin(II)alkoxide species observed directly in MALDI-TOF spectra, Macromolecules 33 (2000) 689–695. [30] J.M. Raquez, Ph. Degee, R. Narayan, Ph. Dubois, ‘‘Coordination–Insertion’’ ring-opening polymerization of 1,4-dioxan-2-one and controlled synthesis of diblock copolymers with e-caprolactone, Macromol. Rapid Commun. 21 (2000) 1063–1071. [31] H.R. Kricheldorf, D.O. Damrau, Polylactones, 42a. Zn L-lactate-catalyzed polymerization of 1,4-dioxan-2-one, Macromol. Chem. Phys. 199 (1998) 1089–1097. [32] S. Matsumura, K. Tsukada, K. Toshima, Enzyme-catalyzed ring-opening polymerization of 1,3-dioxano-2-one to poly(trimethylene carbonate), Macromolecules 30 (1997) 3122–3124. [33] H. Nishida, M. Yamashita, M. Nagashima, T. Endo, Y. Tokiwa, Synthesis of metal-free poly(1,4-dioxan-2-one) by enzyme-catalyzed ring-opening polymerization, J. Polym. Sci. A: Polym. Chem. 38 (2000) 1560–1567. [34] Y. Lochee, Into the Polymerization Studies of 1,4 dioxa-2-one and Analogues, Synthesis of Copolymers and Electrospinning of Fibres, PhD thesis, UoM, 2010. [35] S.C. Chen, X.L. Wang, Y.Z. Wang, K.K. Yang, Z.X. Zhou, G. Wu, In vitro degradation of biodegradable blending materials based on poly(p-dioxanone) and poly(vinyl alcohol)-graft- poly(p-dioxanone) with high molecular weights, J. Biomed. Mater. Res. A 80 (2) (2007) 453–465. [36] M.A. Sabino, S. GonzaÂlez, L. MaÂrquez, J.L. Feijoo, Study of the hydrolytic degradation of polydioxanone PPDX, Polym. Degrad. Stab. 69 (2000) 209– 216. [37] Y.M. Cao, D.M. Yu, Bioabsorbable Poly(p-dioxanone) and application, Novel Mater. Chem. Eng. 1996 (1921) 3 (in Chinese). [38] V. Thomas, X. Zhang, Y.K. Vohra, A biomimetic tubular scaffold with spatially designed nanofibers of protein/PDS1 bio-blends, Biotechnol. Bioeng. 104 (5) (2009) 1025–1033. [39] K.J. Cho, D.K. Song, S.H. Oh, Y.J. Koh, S.H. Lee, M.C. Lee, J.H. Lee, Fabrication and characterization of hydrophilized polydioxanone scaffolds for tissue engineering applications, Key Eng. Mater. 342–343 (2007) 289–292. [40] F. Li, J. Feng, R. Zhuo, Synthesis and characterization of novel biodegradable poly (p-dioxanone-co-ethyl ethylene phosphate)s, J. Appl. Polym. Sci. 102 (6) (2006) 5507–5511. [41] F. Lu, X.L. Wang, S.C. Chen, K.K. Yang, Y.Z. Wang, An efficient approach to synthesize polysaccharides-graft-poly (p-dioxanone) copolymers as potential drug carriers, J. Polym. Sci. A: Polym. Chem. 47 (20) (2009) 5344–5353. [42] S.D. Ding, G.C. Zheng, J.B. Zeng, L. Zhang, Y.D. Li, Y.Z. Wang, Preparation, characterization and hydrolytic degradation of poly [p-dioxanone-(butylene succinate)] multiblock copolymer, Eur. Polym. J. 45 (11) (2009) 3043–3057. [43] M.X. Li, R.X. Zhuo, F.Q. Qu, Study on the preparation of novel functional poly(dioxanone) and for the controlled release of protein, React. Funct. Polym. 55 (2003) 185–195. [44] R.S. Bartholomew, PDS (polydioxanone suture): a new synthetic absorbable suture in cataract surgery. A preliminary study, Ophthalmologica 183 (2) (1981) 81–85. [45] J.C. Middleton, A.J. Tipton, Synthetic biodegradable polymers as medical devices, Med. Plast. Biomater. 3 (1998) 30. [46] J.M. Desbordes, H. Hamard, Y. Pouliquen, Absorbable polydioxanone monofilament in the surgery of the anterior segment of the eye, J. Fr. Ophtalmol. 6 (12) (1983) 1007–1010. [47] E. Andersen, K. Sondenaa, J. Holter, A comparative study of polydioxanone (PDSÒ) and polyglactin 910 (VicrylÒ) in colonic anastomoses in rats, Int. J. Colorectal Dis. 4 (4) (1989) 251–254. [48] PDS II (Polydioxanone), The Veterinarian’s Suture Guide. . [49] R.H.L. Wong, A.P.W. Lee, C.S.H. NG, I.Y.P. Wan, S. Wan, M.J. Underwood, Mitral valve repair: past, present, and future, Asian Cardiovasc. Thorac. Ann. 18 (6) (2010) 586–595. [50] P.O. Myers, A. Kalangos, Valve repair using biodegradable ring annuloplasty: from bench to long-term clinical results, Heart Lung Vessels 5 (4) (2013) 213– 218. [51] M. Cikirikcioglu, S. Cherian, A. Kalangos, Mitral annuloplasty using an intraannular ring, Multimedia Manual Cardio-Thorac. Surg. (2012), http:// dx.doi.org/10.1093/mmcts/mms019. [52] M. Boenisch, H. Tama´s, G.J. Nolst Trenite´, Influence of polydioxanone foil on growing septal cartilage after surgery in an animal model: new aspects of cartilage healing and regeneration (preliminary results), Arch. Facial Plast. Surg. 5 (4) (2003) 316–319. [53] J. Rimmer, L.M. Ferguson, H.A. Saleh, Versatile applications of the polydioxanone plate in rhinoplasty and septal surgery, Arch. Facial Plast. Surg. 14 (5) (2012) 323–330. [54] H. Liu, E.B. Slamovich, T.J. Webster, Less harmful acidic degradation of poly(lactic-co-glycolic acid) bone tissue engineering scaffolds through titania nanoparticle addition, Int. J. Nanomed. 1 (4) (2006) 541–545. [55] P.A. Madurantakam, I.A. Rodriguez, C.P. Cost, R. Viswanathan, D.G. Simpson, M.J. Beckman, P.C. Moon, G.L. Bowlin, Multiple factor interactions in biomimetic mineralization of electrospun scaffolds, Biomaterials 30 (2009) 5456–5464. [56] S. Gollapudi, C.S. So, M. Formica, S. Agrawal, A. Agrawal, Safety and efficacy of polydioxanone nanofibers as anti-inflammatory agents, J. Nanomed. Biotherapeut. Discov. 4 (2) (2014), http://dx.doi.org/10.4172/2155983X.1000127.

[57] W.M. Abbott, R.P. Cambria, Control of physical characteristics (elasticity and compliance) of vascular grafts, in: J.C. Stanley (Ed.), Biological and Synthetic Vascular Prostheses, Grune and Stratton, New York, NY, 1982, pp. 189–220. [58] O. Hakimi, R. Murphy, U. Stachewicz, S. Hislop, A.J. Carr, An electrospun polydioxanone patch for the localization of biological therapies during tendon repair, Eur. Cells Mater. 24 (2012) 344–357. [59] G. Liao, K. Jiang, S. Jiang, H. Xia, Synthesis and characterization of biodegradable poly(e-caprolactone)-b-poly(L-lactide) and study on their electrospun scaffolds, J. Macromol. Sci. A 47 (11) (2010) 1116–1122. [60] N. Abbasi, S. Soudi, N. Hayati-Roodbari, M. Dodel, M. Soleimani, The effects of plasma treated electrospun nanofibrous poly (e-caprolactone) scaffolds with different orientations on mouse embryonic stem cell proliferation, Cell J. 16 (3) (2014) 245–254. [61] D. Kalfa, A. Bel, A. Chen-Tournoux, A. Della-Martina, P. Rochereau, C. Coz, V. Bellamy, M. Bensalah, V. Vanneaux, S. Lecourt, E. Mousseaux, P. Bruneval, J. Larghero, P. Menasche, A polydioxanone electrospun valved patch to replace the right ventricular outflow tract in a growing lamb model, Biomaterials 31 (2010) 4056–4063. [62] M.J. McClure, S.A. Sell, C.P. Barnes, W.C. Bowen, G.L. Bowlin, Cross-linking electrospun polydioxanone-soluble elastin blends: material characterization, J. Eng. Fibers Fabr. 3 (1) (2008) 1–10. [63] S.A. Sell, M.J. McClure, C.P. Barnes, D.C. Knapp, B.H. Walpoth, D.G. Simpson, G.L. Bowlin, Electrospun polydioxanone–elastin blends: potential for bioresorbable vascular grafts, Biomed. Mater. 1 (2006) 72–80. [64] M.J. McClure, S.A. Sell, D.G. Simpson, G.L. Bowlin, Electrospun polydioxanone, elastin, and collagen vascular scaffolds: uniaxial cyclic distension, J. Eng. Fibers Fabr. 4 (2) (2009) 18–25. [65] M.J. McClure, S.A. Sell, C.E. Ayres, D.G. Simpson, G.L. Bowlin, Electrospinning aligned and random polydioxanone-polycaprolactone-silk fibroin blended scaffolds: geometry for a vascular matrix, Biomed. Mater. 4 (2009) 055010. [66] M.J. Smith, M.J. McClure, S.A. Sell, C.P. Barnes, B.H. Walpoth, D.G. Simpson, G.L. Bowlin, Suture-reinforced electrospun polydioxanone–elastin smalldiameter tubes for use in vascular tissue engineering: a feasibility study, Acta Biomater. 4 (2008) 58–66. [67] S.N. Stephansson, B.A. Byers, A.J. Garcia, Enhanced expression of the osteoblastic phenotype on substrates that modulate fibronectin conformation and integrin receptor binding, Biomaterials 23 (12) (2002) 2527–2534. [68] I.A. Rodriguez, M.A. Maduramtakam, J.M. McCool, S.A. Sell, H. Yang, P.C. Moon, G.L. Bowlin, Mineralization potential of electrospun PDOhydroxyapatite-fibrinogen blended scaffolds, Int. J. Biomater. (2012) 12, http://dx.doi.org/10.1155/2012/159484. Article ID: 159484. [69] T.H. Kim, S.H. Oh, S.Y. Chun, J.H. Lee, Bone morphogenetic proteinsimmobilized polydioxanone porous particles as an artificial bone graft, J. Biomed. Mater. Res. A 102A (5) (2014) 1264–1274. [70] J.H. Lee, J.H. Kim, S.H. Oh, S.J. Kim, Y.S. Hah, B.W. Park, et al., Tissueengineered bone formation using periosteal-derived cells and polydioxanone/pluronic F127 scaffold with pre-seeded adipose tissuederived CD146 positive endothelial-like cells, Biomaterials 32 (2011) 5033– 5045. [71] M.C. McManus, S.A. Sell, W.C. Bowen, H.P. Koo, D.G. Simpson, G.L. Bowlin, Electrospun fibrinogen-polydioxanone composite matrix: potential for in situ urologic tissue engineering, J. Eng. Fibers Fabr. 3 (2) (2008) 12–21. [72] I.A. Chaim, M.A. Sabino, M. Mendt, A.J. Muller, D. Ajami, Evaluation of the potential of novel PCL/PPDX biodegradable scaffolds as support materials for cartilage tissue engineering, J Tissue Eng. Regen. Med. 6 (2012) 272–279. [73] A. Fontana, A. Bistolfi, M. Crova, G. Massazza, Primary stability of a PGLA/ Polydioxanone membrane for potential autologous chondrocyte implantation in the hip joint. A cadaveric study, Hip Int. 23 (3) (2013) 337. [74] N. Goonoo, A. Bhaw-Luximon, G.L. Bowlin, D. Jhurry, Diblock poly(ester)poly(ester-ether) copolymers: I. Synthesis, thermal properties, and degradation kinetics, Ind. Eng. Chem. Res. 51 (2012) 12031–12040. [75] N. Goonoo, A. Bhaw-Luximon, I.A. Rodriguez, G.L. Bowlin, D. Jhurry, Poly(ester-ether)s: I. Investigation of the properties of blend films of polydioxanone and poly(methyl dioxanone), Int. J. Polym. Mater. Polym. Biomater. 63 (10) (2014) 527–537. [76] K. Kratz, R. Habermann, T. Becker, K. Richau, A. Lendlein, Shape memory properties and degradation behavior of multifunctional electrospun scaffolds, Int. J. Artif. Organs 34 (2) (2011) 225–230. [77] P.S. Wolfe, Y. Lochee, D. Jhurry, A. Bhaw-Luximon, G.L. Bowlin, Characterization of electrospun novel poly(ester-ether) copolymers: 1,4dioxan-2-one and D, L-3-methyl-1,4-dioxan-2-one, J. Eng. Fibres Fabr. 6 (4) (2011) 60–69. [78] N. Goonoo, A. Bhaw-Luximon, I.A. Rodriguez, D. Wesner, H. Schoenherr, G.L. Bowlin, D. Jhurry, Poly(ester-ether)s: II. Properties of electrospun nanofibres from polydioxanone and poly(methyl dioxanone) blends and human fibroblast cellular proliferation, Biomater. Sci. 2 (2014) 339–351. [79] N. Goonoo, A. Bhaw-Luximon, I.A. Rodriguez, D. Wesner, H. Schoenherr, G.L. Bowlin, D. Jhurry, J. Mater. Chem. B (2015), http://dx.doi.org/10.1039/ c4tb01350f (in press). [80] V. Lassalle, M.L. Ferreira, PLA nano-and microparticles for drug delivery: an overview of the methods of preparation, Macromol. Biosci. 7 (6) (2007) 767– 783. [81] J. Zhu, H.C. Dang, W.T. Wang, X.L. Wang, Y.Z. Wang, Cellulose diacetate-g-poly (p-dioxanone) co-polymer: synthesis, properties and microsphere preparation, J. Biomater. Sci. Polym. Ed. 22 (8) (2011) 981–999.

N. Goonoo et al. / European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 371–391 [82] G.Y. Liu, Y.L. Zhai, X.L. Wang, W.T. Wang, Y.B. Pan, X.T. Dong, Y.Z. Wang, Preparation, characterization, and in vitro drug release behavior of biodegradable chitosan-graft-poly (1, 4-dioxan-2-one) copolymer, Carbohydr. Polym. 74 (4) (2008) 862–867. [83] X.L. Wang, Y. Huang, J. Zhu, Y.B. Pan, R. He, Y.Z. Wang, Chitosan-graft poly (pdioxanone) copolymers: preparation, characterization, and properties, Carbohydr. Polym. 344 (6) (2009) 801–807. [84] T. Yoshida, T.C. Lai, G.S. Kwon, K. Sako, PH- and ion-sensitive polymers for drug delivery, Expert Opin. Drug Deliv. 10 (11) (2013) 1497–1513. [85] J. Liu, Z. Jiang, S. Zhang, C. Liu, R.A. Gross, T.R. Kyriakides, W.M. Saltzman, Biodegradation, biocompatibility, and drug delivery in poly (xpentadecalactone-co-p-dioxanone) copolyesters, Biomaterials 32 (27) (2011) 6646–6654. [86] X.L. Wang, Y.R. Mou, S.C. Chen, J. Shi, Y.Z. Wang, A water-soluble PPDO/PEG alternating multiblock copolymer: synthesis, characterization, and its gel–sol transition behaviour, Eur. Polym. J. 45 (4) (2009) 1190–1197. [87] S.C. Chen, L.L. Li, H. Wang, G. Wu, Y.Z. Wang, Synthesis and micellization of amphiphilic multi-branched poly (p-dioxanone)-block-poly (ethylene glycol), Polym. Chem. 3 (5) (2012) 1231–1238. [88] G. Wu, S.C. Chen, Q. Zhan, Y.Z. Wang, Well-defined amphiphilic poly (pdioxanone)-grafted poly (vinyl alcohol) copolymers: synthesis and micellization, J. Polym. Sci. A: Polym. Chem. 48 (21) (2012) 4811–4822. [89] M.P. Bajgai, D.C. Parajuli, J.A. Ko, H.K. Kang, M.S. Khil, H.Y. Kim, Synthesis, characterization and aqueous dispersion of dextran-g-poly (1,4-dioxan-2one) copolymers, Carbohydr. Polym. 78 (4) (2009) 833–840. [90] D.L. Tang, F. Song, C. Chen, X.L. Wang, Y.Z. Wang, A pH-responsive chitosan-bpoly (p-dioxanone) nanocarrier: formation and efficient antitumor drug delivery, Nanotechnology 24 (14) (2013) 145101. [91] N. Bhattarai, S.R. Bhattarai, H.K. Yi, J.C. Lee, M.S. Khil, P.H. Hwang, H.Y. Kim, Novel polymeric micelles of amphiphilic triblock copolymer poly (pdioxanone-co-L-lactide)-block-poly(ethylene glycol), Pharm. Res. 20 (12) (2003) 2021–2027. [92] R.K.C. Bahadur, S.R. Bhattarai, S. Aryal, M.S. Khil, N. Dharmaraj, H.Y. Kim, Novel amphiphilic triblock copolymer based on PPDO, PCL, and PEG: synthesis, characterization, and aqueous dispersion, Colloids Surf. A 292 (2007) 69–78. [93] L. Zhang, Y. Feng, H. Tian, C. Shi, M. Zhao, J. Guo, Controlled release of doxorubicin from amphiphilic depsipeptide–PDO–PEG-based copolymer nanosized microspheres, React. Funct. Polym. 73 (9) (2013) 1281–1289. [94] L. Zhang, Y. Feng, H. Tian, M. Zhao, M. Khan, J. Guo, Amphiphilic depsipeptidebased block copolymers as nanocarriers for controlled release of ibuprofen with doxorubicin, J. Polym. Sci. A: Polym. Chem. 51 (15) (2013) 3213–3226. [95] J. Lv, L. Zhang, M. Khan, X. Ren, J. Guo, Y. Feng, Biodegradable depsipeptide– PDO–PEG-based block copolymer micelles as nanocarriers for controlled release of doxorubicin, React. Funct. Polym. 82 (2014) 89–97.

391

[96] R. Jeetah, A. Bhaw-Luximon, D. Jhurry, New amphiphilic PEG-b-P(ester–ether) micelles as potential drug nanocarriers, J. Nanopart. Res. 14 (10) (2012) 1–13. [97] A. Bhaw-Luximon, R. Jeetah, N. Goonoo, A. Veeren, Y. Jugdawa, D. Jhurry, A review of polymeric biomaterials research for tissue engineering and drug delivery applications at the Centre for Biomedical and Biomaterials Research, Mauritius, Afr. J. Sci. Technol. Innov. Dev. 6 (2) (2014) 145–157. [98] R. Jeetah, A. Bhaw-Luximon, D. Jhurry, Dual encapsulation and controlled delivery of anti-TB drugs from PEG-block-poly (ester-ether) nanomicelles, J. Nanopharm. Drug Deliv. 1 (3) (2013) 240–257. [99] R. Jeetah, A. Bhaw-Luximon, D. Jhurry, Polymeric nanomicelles for sustained delivery of anti-cancer drugs, Mutat. Res. – Fund. Mol. M. 768 (2014) 47–59. [100] L. Song, Y. Shen, J. Hou, L. Lei, S. Guo, C. Qian, Polymeric micelles for parenteral delivery of curcumin: preparation, characterization and in vitro evaluation, Colloids Surf. A 390 (1) (2011) 25–32. [101] X.T. Dong, W.T. Shi, H.C. Dang, W.Y. Bao, X.L. Wang, Y.Z. Wang, Thermal, crystallization properties, and micellization behavior of HEC-g-PPDO copolymer: microstructure parameters effect, Ind. Eng. Chem. Res. 51 (43) (2012) 14037–14046. [102] X.L. Wang, Y.L. Zhai, D.L. Tang, G.Y. Liu, Y.Z. Wang, Self-assembly, drugdelivery behavior, and cytotoxicity evaluation of amphiphilic chitosan-graftpoly (1, 4-dioxan-2-one) copolymers, J. Polym. Res. 19 (9) (2012) 1–9. [103] S.R. Bhattarai, S.Y. Kim, K.Y. Jang, H.K. Yi, Y.H. Lee, N. Bhattarai, S.Y. Nam, D.Y. Lee, H.Y. Kim, P.H. Hwang, Amphiphilic triblock copolymer poly (pdioxanone-co-L-lactide)-block-poly (ethylene glycol), enhancement of gene expression and inhibition of lung metastasis by aerosol delivery, Gene Ther. 14 (6) (2006) 476–483. [104] S.R. Bhattarai, H.K. Yi, N. Bhattarai, P.H. Hwang, H.Y. Kim, Novel block copolymer (PPDO/PLLA-b-PEG): enhancement of DNA uptake and cell transfection, Acta Biomater. 2 (2) (2006) 207–212. [105] N. Goonoo, A. Bhaw-Luximon, D. Jhurry, Drug loading and release from electrospun biodegradable nanofibers, J. Biomed. Nanotechnol. 10 (2014) 2173–2199. [106] R.A. Waeiss, T.C. Negrini, R.A. Arthur, M.C. Bottino, Antimicrobial effects of drug-containing electrospun matrices on osteomyelitis-associated pathogens, J. Oral Maxillofac. Surg. 72 (7) (2014) 1310–1319. [107] M.G. Blanco, L. Franco, J. Puiggali, A. Rodriguez-Galan, Incorporation of triclosan into polydioxanone monofilaments and evaluation of the corresponding release, J. Appl. Polym. Sci. 114 (2009) 3440–3451. [108] M.C. Bottino, K. Kamocki, G.H. Yassen, J.A. Platt, M.M. Vail, Y. Ehrlich, K.J. Spolnik, R.L. Gregory, Bioactive nanofibrous scaffolds for regenerative endodontics, J. Dent. Res. 92 (11) (2013) 963–969. [109] G. Li, Y. Chen, J. Hu, X. Wu, J. Hu, X. He, et al., A 5-fluorouracil-loaded polydioxanone weft-knitted stent for the treatment of colorectal cancer, Biomaterials 34 (2013) 9451–9461.