- Email: [email protected]
Dextrin–rhEGF conjugates as bioresponsive nanomedicines for wound repair
Contents lists available at ScienceDirect
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
Dextrin–rhEGF conjugates as bioresponsive nanomedicines for wound repair Joseph Hardwicke a,b,⁎, Elaine L. Ferguson b, Ryan Moseley a, Phil Stephens a, David W. Thomas a,⁎, Ruth Duncan b,⁎ a
Wound Biology Group, Cardiff Institute of Tissue Engineering and Repair (CITER), Tissue Engineering and Reparative Dentistry, School of Dentistry, Heath Park Campus, Cardiff University, Cardiff CF14 4XY, UK Centre for Polymer Therapeutics, Welsh School of Pharmacy, Redwood Building, King Edward VII Avenue, Cardiff CF10 3XF, UK
b
a r t i c l e
i n f o
Article history: Received 9 April 2008 Accepted 11 July 2008 Available online 22 July 2008 Keywords: Dextrin–rhEGF conjugate Wound repair PUMPT Polymer therapeutics Nanomedicine
a b s t r a c t Growth factors are known to act in concert to promote wound repair, but their topical application rarely leads to a significant clinical improvement of chronic wounds due to premature inactivation in wound environment. The aim of this study was to synthesise a polymer–growth factor conjugate and investigate whether the novel concept called Polymer-masking-UnMasking-Protein Therapy (PUMPT) might be used to generate bioresponsive polymer therapeutics as nanomedicines able to promote tissue repair. Succinoylated dextrin (∼ 85,000 g/mol; ∼ 19 mol% succinoylation), and rhEGF were chosen as a first model combination. The conjugate synthesised contained ∼ 16%wt rhEGF and b 1% free protein. It exhibited increased stability towards proteolytic degradation by trypsin and the clinically relevant enzyme neutrophil elastase. The dextrin component was degraded on addition of α-amylase leading to sustained release of free rhEGF over time (52.7% release after 168 h). When biological activity was assessed (±α-amylase) in proliferation assays using epidermoid carcinoma (HEp2) cells and HaCaT keratinocytes, as anticipated, polymer conjugation reduced rhEGF bioactivity (p = 0.0035). However, exposure to physiological concentrations of α-amylase triggered dextrin degradation and this led to protein unmasking with restoration of bioactivity to the level seen for unmodified rhEGF. Indeed, prolongation of HEp2 proliferation was observed over 8 days. The inability of dextrin, succinoylated dextrin or α-amylase alone to induce proliferative effects, and the ability of αamylase-exposed dextrin–rhEGF to induce phosphorylation of the epidermal growth factor receptor (EGFR) in HEp2 cells confirmed a mechanism of action by stimulation of classical signal transduction pathways. These observations suggest that this dextrin–rhEGF, and other dextrin-growth factor conjugates have potential for further development as bioresponsive nanomedicines for tissue repair. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Although the complex array of cellular and molecular mechanisms of wound healing are becoming increasingly well understood [1,2] there has been very limited success in harnessing this knowledge to promote effective wound healing in the clinical setting. Impaired dermal wound healing (e.g. venous leg ulcers, diabetic foot ulcers and pressure sores) results in an extremely poor quality of life [3]. It is characterised by the chronic persistence of inflammatory cells, the disordered synthesis and remodelling of the extracellular matrix (ECM), and a lack of re-epithelialisation [4]. Growth factors, including platelet-derived growth factor (PDGF), epidermal growth factor (EGF), vascular endothelial growth factor (VEGF), and transforming growth factor-β (TGF-β), normally act in concert to promote wound repair
⁎ Corresponding authors. Wound Biology Group, Cardiff Institute of Tissue Engineering and Repair (CITER), Tissue Engineering and Reparative Dentistry, School of Dentistry, Heath Park Campus, Cardiff University, Cardiff CF14 4XY, UK. Tel.: +44 2920 745454. E-mail address: [email protected] (J. Hardwicke). 0168-3659/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2008.07.023
(reviewed in [1,5,6]). However, chronic wound fluids have shown that the wound environment has the capacity to induce premature growth factor degradation and inactivation [7,8]. Polymer therapeutics are finding increasing clinical application as anticancer and antiviral agents (reviewed by Duncan [9,10]) and poly (ethyleneglycol) (PEG)-protein conjugation has been widely used to enhance the protein stability, reduce immunogenicity and optimise pharmacokinetics (reviewed in [11,12]). However, polymer therapeutics have rarely been designed to promote tissue repair [13,14]. The aim of this study was to examine whether Polymer-maskingUnMasking-Protein Therapy (PUMPT) could be used to develop a bioresponsive polymer therapeutic able to promote wound healing. The novel concept PUMPT uses a biodegradable polymer to transiently mask a protein during transit (thus stabilising/inactivating the protein), whilst subsequently allowing triggered polymer degradation, protein unmasking and the restoration of protein bioactivity [15]. The biodegradable polysaccharide dextrin, and a recombinant human epidermal growth factor (rhEGF) were chosen here as models. The hypothesis is illustrated schematically in Fig. 1. Theoretically, the dextrin–rhEGF conjugate has the potential to treat acute or chronic
NANOMEDICINE
Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
276
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 1. Schematic showing the proposed mechanism of action of dextrin–rhEGF. Panel (A) shows the potential routes of administration; topical or intravenous for targeting by the enhanced vascular permeability effect. Panel (B) shows the putative mechanism of action of dextrin–rhEGF conjugates in the wound.
wounds, most simply by topical administration or alternatively, via intravenous (i.v.) administration, to localise the rhEGF using enhanced vascular permeability at sites of inflammation. The bioactivity of rhEGF is well documented in animal and human studies [16–18]. It is a small protein (molecular weight 6230 g/mol), containing three internal disulfide bridges, that binds with high affinity to EGF receptors (EGFR) [19]. This causes their dimerisation and receptor phosphorylation, initiating the signal transduction pathways leading to increased cell proliferation [20,21] and migration [22]. Dextrin, α-1,4 poly(glucose), is a biocompatible polymer used clinically in end-stage renal failure patients as a peritoneal dialysis solution [23] and as a carrier for the anticancer agent, 5-fluorouracil [24]. Dextrin is particularly useful for PUMPT, as it is readily degraded by α-amylase to maltose, maltotriose or maltotetraose [23]. Following succinoylation, it can easily be conjugated to proteins and moreover, the rate of dextrin degradation can be tailored by controlling the degree of polymer succinoylation [25]. rhEGF has two lysine residues (Lys28, Lys48), with one being conveniently located peripherally, allowing easy access for conjugation. Here, dextrin–rhEGF conjugates were first synthesised and characterised. The ability of conjugation to protect against proteolytic
degradation by trypsin (as a model) and human neutrophil elastase (the most prevalent chronic wound protease [7]) was investigated. Intracellular mechanistic studies confirmed EGFR phosphorylation by dextrin–rhEGF and conjugate activity (±α-amylase) was assessed using an epidermoid carcinoma (HEp2) cell model and HaCaT keratinocyte proliferation assay. 2. Materials and methods 2.1. Materials 2.1.1. General chemicals, polymers and reagents rhEGF was obtained from Prospec-Tany Technogene Ltd, (Rehovot, Israel) and dextrin (Mw ∼ 42,000 g/mol) from ML Laboratories (Liverpool, UK). 4-Dimethylaminopyridine (DMAP), anhydrous dimethylformamide (DMF), diethyl ether, succinic anhydride, sodium dodecyl sulphate (SDS), 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyltetrazolium bromide (MTT), human neutrophil elastase and human salivary αamylase, were from Sigma-Aldrich (Poole, UK). Dialysis membrane (10,000 g/mol cut-off) was obtained from BDH (Poole, UK). 1-Ethyl-3[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC), N-
hydroxysulfosuccinimide (sulfo-NHS), and bicinchoninic acid (BCA) reagents were obtained from Pierce (Cramlington, UK). All other chemicals were of analytical reagent grade. 2.1.2. Cell culture Human epidermoid carcinoma (HEp2) (ATCC No: CCL-23) and Eagle's minimum essential media (EMEM) with L-glutamine and Earle's balanced salt solution adjusted to contain 1.5 g/L sodium bicarbonate, non-essential amino acids, and sodium pyruvate was purchased from LGC Promochem (Teddington, UK). Epilife medium and HKGS supplement kit (bovine pituitary extract, bovine insulin, hydrocortisone, bovine transferrin) were obtained from Cascade Biologics (Mansfield, UK). Trypsin-EDTA, penicillin G, streptomycin sulphate, amphotericin B, hydrocortisone, adenine, cholera toxin, insulin and fetal calf serum (FCS) were purchased from Invitrogen (Paisley, UK). 2.1.3. Antibodies and ELISA The polyclonal antibody anti-EGFR (phospho Y1173) (rabbit) was obtained from Abcam (Cambridge, UK) and the monoclonal antibody anti-actin (C-2) (mouse) from Santa Cruz Biotechnology (Santa Cruz, USA). A human EGF-ELISA kit was purchased from R&D Systems (Abingdon, UK). 2.2. Methods 2.2.1. Synthesis and characterisation of succinoylated dextrin Dextrin was succinoylated as described previously [26]. Briefly, dextrin (1 g) was added to DMAP (80 mg), purged with nitrogen and DMF added (7.5 mL) to the stirred mixture. Succinic anhydride (182.5 mg) was dissolved in DMF (2.5 mL) and added to the reaction mixture. The solution was stirred under nitrogen at 50 °C for 14 h and the DMF removed by evaporation under high vacuum to leave a few drops of reaction mixture. This was added drop-wise to vigorously stirred diethyl ether (250 mL) and left for 10 h, under continual stirring (Fig. 2). This solution was filtered under vacuum, and the resultant product was purified by dialysis in double-distilled water. Functionalisation of succinoylation dextrin was confirmed qualitatively by FTIR (Avatar 360 E.S.P. Spectrometer and EZ OMNIC E.S.P. 5.2 software (Thermo Nicolet, Loughborough, UK)) using 64-300 scans over 400–4000 cm− 1 with background subtracted. The weight average (Mw), number average (Mn) molecular weight and polydispersity of the polymers were determined with gel permeation chromatography (GPC) using Ultrahydrogel 1000 (200 × 8 mm) and Ultrahydrogel 250 columns (Waters Ltd, Elstree, UK) in series by RI (Gilson, Middleton, USA) and Severn Analytical UV absorbance (Jaytee Biosciences, Whitstable, UK) detectors. Succinoylated dextrin in PBS (3 mg/mL) was injected (60 μL) onto the column and elution profiles were monitored over 30 min. Polysaccharide (pullulan) standards were used to estimate Mw and Mn. 2.2.2. Synthesis and characterisation of dextrin–rhEGF conjugates The succinoylated dextrin intermediate was conjugated to rhEGF. Briefly, succinoylated dextrin (8.2 mg in 500 μL ddH2O) was added to EDC (1.12 mg) and stirred for 10 min, at room temperature. To this, sulfo-NHS (1.2 mg) was added and the solution was stirred for a further 40 min. rhEGF (2 mg in 300 μL ddH2O) was added and the reaction was allowed to proceed for 18 h, pH 8 (Fig. 2). The product was purified using FPLC (AKTA FPLC, UV detector, Unicorn 3.20 software, Amersham Biosciences, Little Chalfont, UK). The conjugate fractions (6–9 mL) were collected and concentrated using Vivaspin 6 centrifugal concentrators (Mw cut-off 30 kDa, Sigma-Aldrich; 3500 g, 10 min), prior to lyophilisation. The BCA protein assay was used, as per the manufacturer's instructions, to determine the total rhEGF content of the conjugates. FPLC and SDS-PAGE were used to determine the levels of free rhEGF (non-conjugated rhEGF) and/or degradation
277
products present in the conjugates. SDS-PAGE was conducted using 12.5% linear gels, with rhEGF and protein molecular weight markers as a reference (Mini-Protean 3 Gel Electrophoresis System, Bio-Rad Laboratories Ltd., Hemel Hempstead, UK). All subsequent concentrations of dextrin–rhEGF are derived from a theoretical 100% degradation of the conjugate. 2.2.3. Degradation of dextrin, succinoylated dextrin and dextrin–rhEGF by α-amylase 2.2.3.1. GPC assay. Dextrin, succinoylated dextrin, and dextrin–rhEGF (all 1 mg/mL) were dissolved in PBS, human salivary α-amylase (93 i.u./L) added, thoroughly mixed, and incubated at 37 °C for 168 h. GPC was used to estimate Mw and Mn as before. 2.2.3.2. FPLC assay. Samples (200 μL) of dextrin–rhEGF (with or without exposure to human salivary α-amylase) were collected, as above. All samples were defrosted and analysed by FPLC. The area of the peaks corresponding to the dextrin–rhEGF conjugate and free rhEGF were determined, and the data shown expressed as the ratio of rhEGF:dextrin–rhEGF with time. 2.2.4. Determination of EGF release by ELISA assay Samples of dextrin–rhEGF were diluted to 250 pg/mL rhEGFequivalent in EGF-ELISA dilution buffer (RD5E; R&D Systems, Abingdon, UK). Samples were then incubated at 37 °C for 168 h in the presence and the absence of α-amylase (93 i.u./L). Samples (200 μL × 3) were taken at times up to 168 h and snap-frozen in liquid nitrogen. The EGF-ELISA assay was then used to quantify the rhEGF content in each sample, according to the manufacturer's instructions and by using a rhEGF standard curve. Data was expressed as rhEGF released as % theoretical total of conjugated rhEGF. 2.2.5. Dextrin–rhEGF conjugate stability in the presence of trypsin and neutrophil elastase The dextrin–rhEGF conjugate and rhEGF (both 1 mg/mL) were dissolved separately in PBS. To assess the stability to proteolytic degradation, trypsin-EDTA (0.025%) or human neutrophil elastase (100 μg/mL) were added to these samples and incubated at 37 °C for 6 h. Samples (200 μL) were taken at times from 0 to 24 h and snapfrozen in liquid nitrogen. At the end of the experiment, all samples were defrosted and analysed by FPLC as described above. The area of the dextrin–rhEGF conjugate and free rhEGF peaks was calculated. 2.2.6. Cell culture HEp2 cells were seeded at 1 × 105 cells in a 75 cm2 tissue culture flasks (Barloworld Scientific, Stone, UK) and incubated in 10 mL of HEp2-serum containing medium (H-SCM: EMEM with penicillin G (100 U/mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL) and 10% v/v FCS), at 37 °C/5% CO2. Upon reaching 80–90% confluence, the cells were sub-cultured, as follows. The cells were washed (×2) with PBS and detached from the plastic by incubation with 1 mL of trypsin/EDTA (0.05% trypsin, 0.53 mM EDTA), for approximately 5 min, at 37 °C. Each flask was washed with H-SCM (9 mL) to collect trypsinised cells. The solution was then transferred to a 15 mL centrifuge tube and centrifuged at 1500 g for 5 min. The supernatant was aspirated and the resulting pellet re-suspended in 1 mL of H-SCM and viable cells (assessed by trypan blue staining) counted, prior to sub-culture. The cell culture medium was changed twice weekly. Immortalised HaCaT keratinocytes (Boukamp [27]) were seeded at 1 × 105 cells in 75 cm2 tissue culture flasks and incubated in 10 mL of keratinocyte-serum containing medium (K-SCM: 67.5% DMEM/22.5% Ham's F12 nutrient media, supplemented with penicillin G (100 U/ mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL), hydrocortisone (400 ng/mL), adenine (0.089 mM), cholera toxin
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
278
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 2. Dextrin–rhEGF synthesis. Dextrin (A) is functionalised with succinic anhydride (B) to produce succinoylated dextrin (C). This reacts with EDC (D) to form an amine-reactive O-acylisourea intermediate (E) and sulfo-NHS (F) then forms an amine-reactive sulfo-NHS ester (G). Lysine residues within the rhEGF (H) molecule form stable amide bonds resulting in dextrin–rhEGF (I).
(0.08 μg/mL), insulin (5 μg/mL), and 10% v/v FCS). Upon reaching 80– 90% confluence, keratinocytes were passaged, as described above. The cell culture medium was changed every 2 days. 2.2.7. Measurement of EGFR phosphorylation by Western blotting In order to investigate whether dextrin–rhEGF retained the ability to induce EGFR-phosphorylation HEp2 cells were grown on 100 mm tissue culture dishes (Cellstar, Greiner Bio-One, UK) and EGFR phosphorylation measured with time. Cells were serum starved for 24 h (in H-SFM:EMEM with penicillin G (100 U/mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL)) prior to the addition of rhEGF (400 ng/mL in H-SFM), dextrin–rhEGF (400 ng/mL equiv. in H-SFM) or α-amylase-exposed dextrin–rhEGF (400 ng/mL equiv. in H-SFM, containing 93 i.u./L α-amylase, incubated at 37 °C for 24 h) for 2.5 min. At times between 0 and 40 min post-exposure, cells
were washed twice with cold PBS and protein extracted by scraping into 1.0 mL of 50 mM Tris-HCl buffer, pH 7.5, containing 5 mM EDTA and 1 mM dithiothreitol, and sonicated (5 × 20 s, Model W-220F, Ultrasonic Inc., USA). Cell suspensions were harvested, centrifuged at 15,000 g (4 °C) for 20 min and protein content was determined using the Bio-Rad DC Protein Assay (Bio-Rad Laboratories, Hemel Hempstead, UK). Samples (70 μg) were loaded onto SDS-PAGE pre-formed 4–15% gradient gels (Bio-Rad Laboratories, Hemel Hempstead, UK) and transferred onto PVDF membranes (G.E. Healthcare, Chalfont St. Giles, UK). Immunodetection was performed using antibodies against phosphorylated EGFR and actin. 2.2.8. Cell proliferation assay HEp2 cells and HaCaT keratinocytes were seeded into 96-well microtitre plates (BD Biosciences, Oxford, UK), at a cell density of
2.5 × 103 cells/well in 10% SCM and incubated for 24 h. The cells were then incubated in serum-free media (SFM) for a further 24 h. For the proliferation assays stock solutions were prepared as follows: For rhEGF, serial dilutions of a stock solution 6.4 μg/mL in K-SFM (Epilife media + HKGS) and a stock solution of 400 ng/mL in H-SFM, were prepared for addition to HaCaT and HEp2 cells, respectively. In each case, the concentration range chosen was based upon preliminary proliferation assays undertaken using these cells (data not shown). For dextrin–rhEGF, serial dilutions of 6.4 μg/mL rhEGF-equivalent stock solution in K-SFM and 50 ng/ mL rhEGF-equivalent in H-SFM were prepared. The rhEGF and dextrin–rhEGF samples were incubated at 37 °C, with or without, α-amylase (20–521 i.u./L), or with trypsin-EDTA (0.025%) for 24 h. The samples exposed to trypsin-EDTA (0.025%) were purified by FPLC and subsequently exposed to α-amylase (93 i.u./L) for a further 24 h. At the beginning of the proliferation assays, either free rhEGF, dextrin–rhEGF (both without or without prior α-amylase exposure), or rhEGF and dextrin–rhEGF exposed to trypsin-EDTA (0.025%) and α-amylase, were added to cells and incubated at 37 °C. After 67 h, cell proliferation was assayed using MTT. MTT (25 μL; 5 mg/mL in PBS) was added to the cell media and incubated for 5 h at 37 °C. The media was then removed. Lysis buffer was added (200 μL; DMF (50%) in H2O containing SDS (20%), glacial acetic acid (2.5%) and HCl (1 M, 2.5%), pH 4.7) and incubated at
279
37 °C for a further 3 h. The absorbance of each sample was then assayed using a Dynex MRX Spectrophotometer (Dynex Technologies, Worthing, UK) with a 540 nm filter. To examine effects over a longer time period, similar cell proliferation assays were performed with HEp2 cells, but proliferation was assessed daily over 8 days. 2.2.9. Statistical analysis Statistical analyses were undertaken using GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA). Data were compared using a Student's t-test and Mann–Whitney U-test, and a one-way ANOVA with a Bonferroni post-test was used for group analysis. Results are expressed as a mean and standard deviation (S.D.), or standard error of the mean (S.E.M.). All p values are two-tailed. Statistical significance was considered at a probability of p b 0.05. 3. Results 3.1. Synthesis and characterisation of the dextrin–rhEGF conjugates The succinoylated dextrin intermediate was characterised by FTIR spectroscopy to confirm the incorporation of carboxylic acid, indicated by a peak at 1720 cm− 1. By titration, the degree of polymer modification was calculated as ∼ 19 mol%. Reaction efficiency was
Fig. 3. Characterisation of the dextrin–rhEGF conjugate and degradation of dextrin, succinoylated dextrin and the conjugate in the presence of α-amylase. Panels (A) and (B) show FPLC analysis of rhEGF (P) and dextrin–rhEGF (- - - -) and SDS-PAGE analysis, respectively. Panel (C) shows GPC chromatograms of dextrin, succinoylated dextrin and dextrin– rhEGF. Panel (D) shows the change in relative molecular weight (GPC analysis) of dextrin (- - ■ - -), succinoylated dextrin (—▴—), and dextrin–rhEGF (- - ♦ - -), in the presence of αamylase (93 i.u./L), over 24 h (n = 3; mean ± S.D., ⁎p = 0.0734, ⁎⁎ p = 0.0012, ANOVA and Bonferroni post hoc test).
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
280
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
limit N 0.15 μg) (Fig. 3A, B). This value was estimated as FPLC analysis showed a clear distinction between rhEGF and dextrin– rhEGF, and rhEGF was undetectable at concentrations b1.6 μg/mL. For a typical conjugate batch containing 16 wt.% rhEGF, this equated to b1% free rhEGF. 3.2. Studies of the rate of α-amylase triggered degradation, rhEGF release and stability Using pullulan standards, the dextrin–rhEGF conjugate had a Mn 70,000; Mw 106,600; PDI 1.5, however these figures should be viewed as an approximation (Fig. 3C). GPC analysis of the molecular weight change in the presence of α-amylase showed that succinoylated dextrin and the dextrin–rhEGF conjugate were degraded much more slowly (t1/2 N 24 h) than parent dextrin (t1/2 b 6 min) (p = 0.0012) (Fig. 3D). Addition of α-amylase led to sustained release of free rhEGF over time, while ELISA confirmed that 52.7% of the rhEGF was released after 168 h (Fig. 4A). Moreover, the dextrin–rhEGF conjugate showed negligible degradation in α-amylase-free conditions (Fig. 4B), at both 4 °C and 37 °C in PBS for 168 h, or when lyophilised and stored at −20 °C (data not shown). In addition, whereas native rhEGF was rapidly degraded by neutrophil elastase (t1/2 b1 h) (Fig. 4C), the dextrin–rhEGF was much more stable, with only 50% degradation at 24 h. 3.3. Investigation of the ability of dextrin–rhEGF to induce EGFR phosphorylation Western blot analysis showed that addition of rhEGF to HEp2 cells induces EGFR phosphorylation (Fig. 5A). This effect was impeded by dextrin conjugation (Fig. 5B), but following exposure
Fig. 4. Effect of addition of α-amylase on rhEGF release and stability of the conjugate in the presence of neutrophil elastase. Panel (A) shows the percentage free rhEGF quantified using EGF-ELISA on addition of human salivary α-amylase (93 i.u./L) (n = 9; mean ± S.E.M., ⁎p = 0.0012 (Student's t- and Mann–Whitney U-test)). Panel (B) shows the release of free rhEGF (conditions as for panel (A), measured by FPLC) expressed as ratio of free:conjugated rhEGF. Panel (C) shows the stability (FPLC) of rhEGF and dextrin–rhEGF in the presence of neutrophil elastase from human leukocytes (100 μg/ mL) (n = 2; mean ± S.D., ⁎p = 0.0401 (Student's t-test)).
typically 50–60% [26]. GPC showed an increased hydrodynamic volume for the succinoylated dextrin (estimated Mw 85,000 g/ mol), compared to the parent dextrin (estimated Mw 42,000 g/ mol), probably due to coil expansion of the negatively charged polymer. The optimum conditions for synthesis of the dextrin– rhEGF conjugates were found to be a 1:2 molar ratio of dextrin: rhEGF with an 18 h reaction time. Synthesis was reproducible, the mean rhEGF content was 16.3 ± 4.4 wt.% protein by BCA assay (n = 5, mean ± S.D.; detection limit N 1.6 μg/mL) and the free rhEGF content of the conjugate was b1% by FPLC and SDS-PAGE (detection
Fig. 5. Effect of rhEGF and dextrin–rhEGF on EGFR phosphorylation. Western blotting was used to determine EGFR phosphorylation after addition of rhEGF (A), dextrin– rhEGF (B), or dextrin–rhEGF exposed to α-amylase (93 i.u./L) (C).
281
rhEGF was between 0.19 and 6.25 ng/mL compared to native rhEGF, which displayed its highest activity between 3.13 and 100 ng/mL. When α-amylase was added to the incubation medium HEp2 cell proliferation was restored to levels observed after the addition of native rhEGF at 72 h (p = 0.0035, Fig. 6B). The extent of proliferation induced by the conjugate was clearly related to the α-amylase concentration present (Fig. 6C). It is important to note that addition αamylase at a physiological concentration (93 i.u./L), dextrin or succinoylated dextrin did not affect cell proliferation (Fig. 6B). When cells were exposed to rhEGF or the dextrin–rhEGF conjugate (without α-amylase) for 8 days, the native rhEGF caused the greatest increase in proliferation, apparent over the first 5 days of exposure (Fig. 7A). Subsequently proliferation returned the control levels. In contrast, the α-amylase-activated dextrin–rhEGF conjugate initially showed lower activity. At day 5, maximum proliferation was observed and this was equivalent to that for rhEGF at day 2. However, the α-amylaseactivated dextrin–rhEGF conjugate retained a significant stimulation of proliferation, compared to the controls, until day 6. Moreover, the conjugate was significantly more active than rhEGF on days 5 and 6 (p b 0.001, Fig. 7B). Utilising HaCaT keratinocytes to study the effect of rhEGF and conjugate, the biphasic proliferative response was less pronounced then seen for HEp2 cells (Fig. 8A). However, these more clinically relevant cells were sensitive to rhEGF growth stimulation over a wider range of concentrations (0.1–100 ng/mL). Using HEp2 cells, it was also shown that exposure of rhEGF to trypsin, before addition to cells led to a reduction in activity, compared to the controls. In contrast, dextrin– rhEGF exposed to trypsin and then to α-amylase retained bioactivity,
Fig. 6. HEp2 proliferation (assessed using the MTT assay) in the presence of rhEGF and α-amylase-activated dextrin–rhEGF. Panel (A) shows proliferation of HEp2 cells after 72 h, in the presence of rhEGF (—w—) and α-amylase-activated dextrin–rhEGF (- -■- -), at the rhEGF concentrations shown (mean ± S.E.M., n = 18; ⁎p b 0.001, ANOVA and Bonferroni post hoc test). Panel (B) compares the proliferation of HEp2 cells after 72 h, in the presence of rhEGF and the dextrin–rhEGF conjugate in the presence and absence of α-amylase (93 i.u./L) (mean ± S.E.M.; n = 18; ⁎p = 0.0035 (ANOVA and Bonferroni post hoc test)). Controls (amylase 93 i.u./L, dextrin 6.25 mg/mL, 19 mol% succinoylated dextrin 6.25 mg/mL) show no significant effect (p N 0.05; ANOVA and Bonferroni post hoc test). Panel (C) shows that restoration of rhEGF activity was related to the concentration of αamylase added (mean ± S.D., n = 6, ⁎p b 0.01, ⁎⁎p b 0.001, ANOVA and Bonferroni post hoc test).
to α-amylase, the dextrin–rhEGF conjugate was again able to induce EGFR phosphorylation (Fig. 5C). 3.4. Stimulation of HEp2 and HaCaT proliferation HEp2 cells demonstrated and displayed a biphasic pattern of growth in the presence of rhEGF and dextrin–rhEGF. Addition of both the growth factor and conjugate stimulated proliferation at lower concentrations, but inhibited proliferation at higher concentrations (N400 ng/mL) (Fig. 6A). The peak concentration for activity of dextrin–
Fig. 7. Panel (A) shows HEp2 growth curves over 8 days, when cells were grown in serum-free conditions (control), or in the presence of rhEGF or dextrin–rhEGF, following exposure to α-amylase (521 i.u./L). Panel (B) shows growth of HEp2 compared to controls (mean ± S.E.M., n = 12; ⁎p b 0.001, ANOVA and Bonferroni post hoc test).
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
282
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 8. Effect of rhEGF and dextrin–rhEGF on the proliferation of HaCaT keratinocytes, and the effect of trypsin exposure on rhEGF and dextrin–rhEGF. Panel (A) shows proliferation of HaCaT keratinocytes cells in response to rhEGF and unmasked dextrin– rhEGF after 72 h (mean ± S.E.M.; n = 18; ⁎p b 0.05, ANOVA and Bonferroni post hoc test). Panel (B) shows proliferation of HEp2 cells in response to rhEGF after protease exposure is reduced to that of rhEGF-free controls (p N 0.05), whereas dextrin–rhEGF retains bioactivity compared to rhEGF-free controls (mean ± S.D., n = 6, ⁎p b 0.001, ⁎⁎p b 0.05, ANOVA and Bonferroni post hoc test).
again demonstrating the resistance of the conjugate to proteolytic degradation (Fig. 8B) afforded by this PUMPT conjugate. 4. Discussion The dextrin–rhEGF conjugate described here was synthesised with the aim of protecting the growth factor from the harsh chronic wound fluid environment whilst allowing regeneration of activity with time when exposed to physiological concentrations of α-amylase using the PUMPT concept [15]. Normal serum α-amylase occurs at a range of 30–110 i.u./L [28], whilst levels in chronic wounds occur in a similar range (unpublished data). In this way it was hoped to generate novel therapeutics (other growth factors may also be used) for improved wound repair. Traditional wound therapy involves the use of topical agents to clean, disinfect and stimulate healing. This can be a protracted, painful process that is often associated with high levels of morbidity. Chronic wounds fail to show the controlled inflammatory, proliferative and remodelling phases of normal wound healing. Numerous factors can be responsible for this malfunction, including wound infection, tissue hypoxia, failed angiogenesis and the underlying uncoupling of the normal growth factor cascade. Many attempts have been made to find more effective therapies for chronic wounds, but there has been little success [29]. Natural and synthetic polymers
have been explored as biodegradable matrices or hydrogels for the controlled release of growth factors (reviewed in [30]), but although their topical application in conjugation with EGF can increase the rate of re-epithelialisation [31], little or no significant clinical improvement in chronic wound healing has ever been demonstrated [32,33]. Overall, the limited success of growth factor therapy has been attributed to (i) an inappropriate release profile in the clinical setting, (ii) the lack of stability of the protein in the delivery system or postdelivery, (iii) rapid protein degradation of rhEGF by chronic wound proteases or denaturation by reactive oxygen species (ROS) [34,35], (iv) the lack of stability of the protein in the delivery vehicle used for application, and/or the toxicity of the polymeric carrier used (e.g. a polyethyleinimine-nitric oxide adduct [36]). It should be noted, however, that one approach, Regranex® (Becaplermin), which is a topically applied gel containing recombinant human platelet-derived growth factor (rhPDGF), recently became the first recombinant growth factor to be approved by the US Food and Drug Administration for the acceleration of wound healing and it is currently used in the treatment of diabetic foot ulcers. However, even in this case, Regranex® has been shown to be only effective in selected wounds, in conjunction with good wound care, with no reduction in ulcer recurrence, when compared to good wound care alone [37]. The dextrin–rhEGF conjugates synthesised here contained, on average, 16 wt.% protein and this corresponded to ∼ 1.5 rhEGF molecules per polymer chain. Heterogeneity of the product was unavoidable (evident by SDS-PAGE electrophoresis; Fig. 3B) and this is largely attributable to the polydisperse dextrin fraction used for conjugation. A narrower molecular weight dextrin fraction could be used to improve conjugate homogeneity if needed for clinical evaluation. A dextrin of average molecular weight of 42,000 g/mol with ∼19 mol% succinoylation was chosen based on earlier studies [15] that suggested this would maximise steric ‘masking’ of rhEGF (lower molecular weight dextrin chains cannot) whilst allowing a sufficiently slow rate of dextrin degradation by α-amylase (Fig. 3D) to liberate rhEGF at an optimal rate for sustained bioactivity. Also, whilst αamylase will degrade the glycosidic bonds within the dextrin polymer, the degradation is slowed by the presence of carboxylic groups and amide bonds [26]. Full degradation of the conjugate would require protease activity. It was confirmed that conjugation of dextrin to rhEGF significantly increased its stability towards proteolysis by the clinically most important protease in wound fluid, neutrophil elastase (Fig. 4C). As anticipated polymer conjugation also reduced biological activity of rhEGF. However, consistent with the PUMPT concept, it was demonstrated that α-amylase, when added with concentrations known to be present in human wound fluid was able to unmask/ release rhEGF. α-Amylase-triggered rhEGF liberation was confirmed by both FPLC (Fig. 3B) and ELISA (Fig. 4A). Polymer degradation would, at least initially, be expected to leave oligosaccharides and/or maltose linked to the rhEGF surface. However, the fact that the liberated rhEGF was clearly identifiable by EGF-immunoassay, and moreover, able to stimulate the clinically relevant parameter of EGF-mediated cell proliferation, underlined the integrity of the product. HEp2 cells were chosen for the cell proliferation assays as they have high levels of EGFR, and their relatively rapid doubling time made it possible to explore the effect of dextrin–rhEGF conjugates over the more clinically relevant time frame of 8 days. The importance of α-amylase as a trigger for conjugate activation was clear throughout all the cell proliferation studies, and it was clearly demonstrated that growth stimulation of HEp2 cells by dextrin– rhEGF was directly related to the units of α-amylase present (Fig. 6B, C). In the extended HEp2 cell proliferation assay (Fig. 7A), it was apparent the initial delayed release of rhEGF release from the conjugate led to a lag phase of approximately 24 h in the cellular response compared to that seen for free rhEGF. However, the sustained rhEGF release from the conjugate led to enhanced activity far beyond that achieved with rhEGF alone. This is even
more impressive given that the EGF-ELISA assay showed release of only 52.7% rhEGF after 168 h. The value of a slow, sustained introduction of bioactive rhEGF is underlined by the biphasic, concentration-dependence of rhEGFinduced proliferation (Fig. 6A). Clearly, high local concentrations of rhEGF must be prevented if inhibition of cell growth is to be avoided. Although HaCaT keratinocytes contain lower levels of EGFR, and grow more slowly than HEp2 cells, they can be considered a more clinically relevant model in relation to wound re-epithelialisation. HaCaT cells also showed enhanced growth in the presence of dextrin–rhEGF activated by α-amylase. Furthermore, it was demonstrated that conjugates exposed to trypsin (as a model protease) before α-amylase exposure still retained the ability to promote HaCaT proliferation. Preliminary experiments to investigate the cellular mechanism of action of dextrin–rhEGF in HEp2 cells found that conjugate exposed to α-amylase was able to, like free rhEGF, induce EGFR phosphorylation (Fig. 5). This suggests a direct effect on the EGFR and promotion on signal transduction. In addition, experiments in progress have shown that α-amylase-activated dextrin–rhEGF conjugates can also induce the migration of HaCaT cells and fibroblasts isolated from human chronic dermal wounds (Hardwicke et al., 2008; unpublished). Such enhanced migration and wound repopulation is key to the repair of chronic wounds in vivo. In conclusion, these studies demonstrate for the first time the potential of a bioresponsive polymer therapeutic as a nanomedicine for wound healing. The conjugate uses a clinically welltolerated polymer, dextrin, and rhEGF as a potential growth factor. The positive effects seen for dextrin–rhEGF in the in vitro studies, suggest that further evaluation of the conjugate towards in vivo and clinical trials is warranted. First clinical studies would certainly involve local, topical application of the conjugate, probably to diabetic or venous leg ulcers. These pathologies are causative of the majority of non-healing wounds and, as such, there is real clinical need for a new treatment. However, this conjugate also has potential for i.v. administration as it would be expected to exhibit targeted delivery to wounds developing new blood vessels by angiogenesis and thus, displaying localised vascular hyperpermeability. Given that recent studies have shown that polymer conjugates can be tailored to deliver anticancer combination therapy [38], the approach reported here has also has the versatility to deliver other growth factors (and or their combinations) to enhance wound healing. This study also verifies the broader therapeutic potential of the PUMPT concept, especially using the combination of dextrin and α-amylase as the enzyme is widely distributed in serum and extracellular fluids [39]. Acknowledgements J.H. was funded by a Fellowship from The Healing Foundation (Registered Charity number 1078666) and the Welsh Office for Research and Development (WORD) and the authors would like to acknowledge support from EPSRC platform grant No. EP/C 013220/1. References [1] J. Li, J. Chen, R. Kirsner, Pathophysiology of acute wound healing, Clin. Dermatol. 25 (2007) 9–18. [2] N.B. Menke, K.R. Ward, T.M. Witten, D.G. Bonchev, R.F. Diegelmann, Impaired wound healing, Clin. Dermatol. 25 (2007) 19–25. [3] P. Price, K. Harding, Cardiff Wound Impact Schedule: the development of a condition- specific questionnaire to assess health-related quality of life in patients with chronic wounds of the lower limb, Int. Wound J. 1 (2004) 10–17. [4] S.E. Herrick, P. Slo, an, M. McGurk, L. Freak, C.N. McCollum, M.W. Ferguson, Sequential changes in the histologic pattern and extra-cellular matrix deposition during the healing of chronic venous leg ulcers, Am. J. Pathol. 141 (1992) 1085–1095. [5] R. Blakytny, E. Jude, The molecular biology of chronic wounds and delayed healing in diabetes, Diabet. Med. 23 (2006) 594–608. [6] R. Goldman, Growth factors and chronic wound healing: past, present and future, Adv. Skin Wound Care 17 (2004) 24–35.
283
[7] N.J. Trengove, M.C. Stacey, S. MacAuley, N. Bennett, J. Gibson, F. Burslem, G. Murphy, G. Schultz, Analysis of the acute and chronic wound environments: the role of proteases and their inhibitors, Wound Rep. Regen. 7 (1999) 442–452. [8] S.M. Chen, S.I. Ward, O.O. Olutoye, R.F. Diegelmann, I. Kelman Cohen, Ability of chronic wound fluids to degrade peptide growth factors is associated with increased levels of elastase activity and diminished levels of proteinase inhibitors, Wound Rep. Regen. 5 (1997) 23–32. [9] R. Duncan, The dawning era of polymer therapeutics, Nat. Rev. Drug. Discov. 2 (2003) 347–360. [10] R. Duncan, Polymer conjugates as anticancer nanomedicines, Nat. Rev. Cancer 6 (2006) 688–701. [11] J.M. Harris, R.B. Chess, Effect of PEGylation on pharmaceuticals, Nature Rev. Drug Discov. 2 (2003) 214–221. [12] M. Francesco, F. Veronese, J. Milton Harris (Eds.), Peptide and protein PEGylation III: advances in chemistry and clinical applications, Adv Drug Delivery Rev., vol. 60, 2008, pp. 1–88. [13] S. Shaunak, S. Thomas, E. Gianasi, A. Godwin, E. Jones, I. Teo, K. Mireskandari, P. Luthert, R. Duncan, S. Patterson, P. Khaw, S. Brocchini, Polyvalent dendrimer glucosamine conjugates prevent scar tissue formation, Nat. Biotechnol. 22 (2004) 977–984. [14] M.J. Vicent, E. Perez-Paya, Poly-L-glutamic acid (PGA) aided inhibitors of apoptotic protease activating factor 1 (Apaf-1): an antiapoptotic polymeric nanomedicine, J. Med. Chem. 49 (2006) 3763–3765. [15] R. Duncan, H.R.P. Gilbert, R.J. Carbajo, M.J. Vicent, Polymer Masked-Unmasked Protein Therapy (PUMPT) 1. Bioresponsive dextrin-trypsin and -MSH conjugates designed for a-amylase activation, Biomacromolecules (2008) Web release date: 19th March, DOI 10.1021/bm701073n. [16] Y.B. Kwon, H.W. Kim, D.H. Roh, S.Y. Yoon, R.M. Baek, J.Y. Kim, H. Kweon, K.G. Lee, Y.H. Park, J.H. Lee, Topical application of epidermal growth factor accelerates wound healing by myofibroblast proliferation and collagen synthesis in rat, J. Vet. Sci. 7 (2006) 105–109. [17] J.P. Hong, Y.W. Kim, H.D. Jung, K.I. Jung, The effect of various concentrations of human recombinant epidermal growth factor on split-thickness skin wounds, Int. Wound J. 3 (2006) 123–130. [18] J.P. Hong, H.D. Jung, Y.W. Kim, Recombinant human epidermal growth factor (EGF) to enhance healing for diabetic foot ulcers, Ann. Plast. Surg. 56 (2006) 394–398. [19] H. Lu, J. Chai, M. Li, B. Huang, C. He, R. Bi, Crystal structure of human epidermal growth factor and its dimerization, J. Biol. Chem. 276 (2001) 34913–34917. [20] G. Carpenter, S. Cohen, Human epidermal growth factor and the proliferation of human fibroblasts, J. Cell Physiol. 88 (1976) 227–237. [21] K. Hashimoto, Regulation of keratinocyte function by growth factors, J. Derm. Sci. 24 (2000) S46–S50. [22] R.C. Kurten, et al., Coordinating epidermal growth factor-induced motility promotes efficient wound closure, Am. J. Physiol. Cell Physiol. 288 (2005) C109–C121. [23] J.E. Frampton, G.L. Plosker, Icodextrin: a review of its use in peritoneal dialysis, Drugs 63 (2003) 2079–2105. [24] K.B. Hosie, D.J. Kerr, J.A. Gilbert, M. Downes, G. Lakin, G. Pemberton, K. Timms, A. Young, A. Stanley, A pilot study of adjuvant intraperitoneal 5-fluorouracil using 4% icodextrin as a novel carrier solution, Eur. J. Surg. Oncol. 29 (2003) 254–260. [25] D. Hreczuk-Hirst, D. Chicco, L. German, R. Duncan, Dextrins as potential carriers for drug targeting: tailored rates of dextrin degradation by introduction of pendant groups, Int. J. Pharm. 230 (2001) 57–66. [26] D. Hreczuk-Hirst, L. German, R. Duncan, Dextrins as carriers for drug targeting: reproducible succinoylation as a means to introduce pendant groups, J. Bioact. Compat. Polym. 16 (2001) 353–365. [27] P. Boukamp, R.T. Petrussevska, D. Breitkreutz, J. Hornung, A. Markham, N.E. Fusenig, Normal keratinisation in a spontaneously immortalised aneuploid human keratinocyte cell line, J. Cell Biol. 106 (1988) 761–771. [28] P. Branca, R.M. Rodriguez, J.T. Rogers, D.S. Ayo, J.P. Moyers, R.W. Light, Routine measurement of pleural fluid amylase is not indicated, Arch. Intern. Med. 161 (2001) 228–232. [29] J.R. Mekkes, M.A.M. Loots, A.C. Vanderwal, J.D. Bos, Causes, investigation and treatment of leg ulceration, Br. J. Dermatol. 148 (2003) 388–401. [30] W.M. Saltzman, W.L. Olbricht, Building drug delivery into tissue engineering, Nat. Rev. Drug Discov. 1 (2002) 177–186. [31] A. Tanaka, T. Nagate, H. Matsuda, Acceleration of wound healing by gelatin film dressings with epidermal growth factor, J. Vet. Med. Sci. 67 (2005) 909–913. [32] V. Falanga, W.H. Eaglstein, B. Bucalo, M.H. Katz, B. Harris, P. Carson, Topical use of human recombinant epidermal growth factor (h-EGF) in venous ulcers, J. Dermatol. Surg. Oncol, 18 (1992) 604–606. [33] D.M. Danilenko, et al., Growth factors in porcine full and partial thickness burn repair. Differing targets and effects of keratinocyte growth factor, platelet-derived growth factor-BB, epidermal growth factor, and neu differentiation factor, Am. J. Pathol. 147 (1995) 1261–1277. [34] T. James, M.A. Hughes, G.W. Cherry, R.P. Taylor, Evidence of oxidative stress in chronic venous ulcers, Wound Rep. Regen. 11 (2003) 172–176. [35] R. Moseley, J.E. Stewart, P. Stephens, R.J. Waddington, D.W. Thomas, Extracellular matrix metabolites as potential biomarkers of disease activity in wound fluid: lessons learned from other inflammatory diseases? Br. J. Dermatol. 150 (2004) 401–413. [36] J.A. Bauer, W. Rao, D.J. Smith, Evaluation of linear polyethyleneimine/nitric oxide adduct on wound repair: therapy versus toxicity, Wound Rep. Regen. 6 (1998) 569–577. [37] D.L. Steed, Clinical evaluation of recombinant human platelet-derived growth factor for the treatment of lower extremity ulcers, Plast. Reconstr. Surg. 117 (2006) S143–S149. [38] M.J. Vicent, F. Greco, R.I. Nicholson, R. Duncan, Polymer–drug conjugates as a novel combination therapy for the treatment of hormone-dependent cancers, Angew. Chem. Int. Ed. 44 (2005) 4061–4066. [39] K. Miyakasa, S.S. Rothman, Redistribution of amylase activity accompanying its secretion by the pancreas, Proc. Natl. Acad. Sci. 79 (1982) 5438–5442.
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
Dextrin–rhEGF conjugates as bioresponsive nanomedicines for wound repair Joseph Hardwicke a,b,⁎, Elaine L. Ferguson b, Ryan Moseley a, Phil Stephens a, David W. Thomas a,⁎, Ruth Duncan b,⁎ a
Wound Biology Group, Cardiff Institute of Tissue Engineering and Repair (CITER), Tissue Engineering and Reparative Dentistry, School of Dentistry, Heath Park Campus, Cardiff University, Cardiff CF14 4XY, UK Centre for Polymer Therapeutics, Welsh School of Pharmacy, Redwood Building, King Edward VII Avenue, Cardiff CF10 3XF, UK
b
a r t i c l e
i n f o
Article history: Received 9 April 2008 Accepted 11 July 2008 Available online 22 July 2008 Keywords: Dextrin–rhEGF conjugate Wound repair PUMPT Polymer therapeutics Nanomedicine
a b s t r a c t Growth factors are known to act in concert to promote wound repair, but their topical application rarely leads to a significant clinical improvement of chronic wounds due to premature inactivation in wound environment. The aim of this study was to synthesise a polymer–growth factor conjugate and investigate whether the novel concept called Polymer-masking-UnMasking-Protein Therapy (PUMPT) might be used to generate bioresponsive polymer therapeutics as nanomedicines able to promote tissue repair. Succinoylated dextrin (∼ 85,000 g/mol; ∼ 19 mol% succinoylation), and rhEGF were chosen as a first model combination. The conjugate synthesised contained ∼ 16%wt rhEGF and b 1% free protein. It exhibited increased stability towards proteolytic degradation by trypsin and the clinically relevant enzyme neutrophil elastase. The dextrin component was degraded on addition of α-amylase leading to sustained release of free rhEGF over time (52.7% release after 168 h). When biological activity was assessed (±α-amylase) in proliferation assays using epidermoid carcinoma (HEp2) cells and HaCaT keratinocytes, as anticipated, polymer conjugation reduced rhEGF bioactivity (p = 0.0035). However, exposure to physiological concentrations of α-amylase triggered dextrin degradation and this led to protein unmasking with restoration of bioactivity to the level seen for unmodified rhEGF. Indeed, prolongation of HEp2 proliferation was observed over 8 days. The inability of dextrin, succinoylated dextrin or α-amylase alone to induce proliferative effects, and the ability of αamylase-exposed dextrin–rhEGF to induce phosphorylation of the epidermal growth factor receptor (EGFR) in HEp2 cells confirmed a mechanism of action by stimulation of classical signal transduction pathways. These observations suggest that this dextrin–rhEGF, and other dextrin-growth factor conjugates have potential for further development as bioresponsive nanomedicines for tissue repair. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Although the complex array of cellular and molecular mechanisms of wound healing are becoming increasingly well understood [1,2] there has been very limited success in harnessing this knowledge to promote effective wound healing in the clinical setting. Impaired dermal wound healing (e.g. venous leg ulcers, diabetic foot ulcers and pressure sores) results in an extremely poor quality of life [3]. It is characterised by the chronic persistence of inflammatory cells, the disordered synthesis and remodelling of the extracellular matrix (ECM), and a lack of re-epithelialisation [4]. Growth factors, including platelet-derived growth factor (PDGF), epidermal growth factor (EGF), vascular endothelial growth factor (VEGF), and transforming growth factor-β (TGF-β), normally act in concert to promote wound repair
⁎ Corresponding authors. Wound Biology Group, Cardiff Institute of Tissue Engineering and Repair (CITER), Tissue Engineering and Reparative Dentistry, School of Dentistry, Heath Park Campus, Cardiff University, Cardiff CF14 4XY, UK. Tel.: +44 2920 745454. E-mail address: [email protected] (J. Hardwicke). 0168-3659/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2008.07.023
(reviewed in [1,5,6]). However, chronic wound fluids have shown that the wound environment has the capacity to induce premature growth factor degradation and inactivation [7,8]. Polymer therapeutics are finding increasing clinical application as anticancer and antiviral agents (reviewed by Duncan [9,10]) and poly (ethyleneglycol) (PEG)-protein conjugation has been widely used to enhance the protein stability, reduce immunogenicity and optimise pharmacokinetics (reviewed in [11,12]). However, polymer therapeutics have rarely been designed to promote tissue repair [13,14]. The aim of this study was to examine whether Polymer-maskingUnMasking-Protein Therapy (PUMPT) could be used to develop a bioresponsive polymer therapeutic able to promote wound healing. The novel concept PUMPT uses a biodegradable polymer to transiently mask a protein during transit (thus stabilising/inactivating the protein), whilst subsequently allowing triggered polymer degradation, protein unmasking and the restoration of protein bioactivity [15]. The biodegradable polysaccharide dextrin, and a recombinant human epidermal growth factor (rhEGF) were chosen here as models. The hypothesis is illustrated schematically in Fig. 1. Theoretically, the dextrin–rhEGF conjugate has the potential to treat acute or chronic
NANOMEDICINE
Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
276
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 1. Schematic showing the proposed mechanism of action of dextrin–rhEGF. Panel (A) shows the potential routes of administration; topical or intravenous for targeting by the enhanced vascular permeability effect. Panel (B) shows the putative mechanism of action of dextrin–rhEGF conjugates in the wound.
wounds, most simply by topical administration or alternatively, via intravenous (i.v.) administration, to localise the rhEGF using enhanced vascular permeability at sites of inflammation. The bioactivity of rhEGF is well documented in animal and human studies [16–18]. It is a small protein (molecular weight 6230 g/mol), containing three internal disulfide bridges, that binds with high affinity to EGF receptors (EGFR) [19]. This causes their dimerisation and receptor phosphorylation, initiating the signal transduction pathways leading to increased cell proliferation [20,21] and migration [22]. Dextrin, α-1,4 poly(glucose), is a biocompatible polymer used clinically in end-stage renal failure patients as a peritoneal dialysis solution [23] and as a carrier for the anticancer agent, 5-fluorouracil [24]. Dextrin is particularly useful for PUMPT, as it is readily degraded by α-amylase to maltose, maltotriose or maltotetraose [23]. Following succinoylation, it can easily be conjugated to proteins and moreover, the rate of dextrin degradation can be tailored by controlling the degree of polymer succinoylation [25]. rhEGF has two lysine residues (Lys28, Lys48), with one being conveniently located peripherally, allowing easy access for conjugation. Here, dextrin–rhEGF conjugates were first synthesised and characterised. The ability of conjugation to protect against proteolytic
degradation by trypsin (as a model) and human neutrophil elastase (the most prevalent chronic wound protease [7]) was investigated. Intracellular mechanistic studies confirmed EGFR phosphorylation by dextrin–rhEGF and conjugate activity (±α-amylase) was assessed using an epidermoid carcinoma (HEp2) cell model and HaCaT keratinocyte proliferation assay. 2. Materials and methods 2.1. Materials 2.1.1. General chemicals, polymers and reagents rhEGF was obtained from Prospec-Tany Technogene Ltd, (Rehovot, Israel) and dextrin (Mw ∼ 42,000 g/mol) from ML Laboratories (Liverpool, UK). 4-Dimethylaminopyridine (DMAP), anhydrous dimethylformamide (DMF), diethyl ether, succinic anhydride, sodium dodecyl sulphate (SDS), 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyltetrazolium bromide (MTT), human neutrophil elastase and human salivary αamylase, were from Sigma-Aldrich (Poole, UK). Dialysis membrane (10,000 g/mol cut-off) was obtained from BDH (Poole, UK). 1-Ethyl-3[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC), N-
hydroxysulfosuccinimide (sulfo-NHS), and bicinchoninic acid (BCA) reagents were obtained from Pierce (Cramlington, UK). All other chemicals were of analytical reagent grade. 2.1.2. Cell culture Human epidermoid carcinoma (HEp2) (ATCC No: CCL-23) and Eagle's minimum essential media (EMEM) with L-glutamine and Earle's balanced salt solution adjusted to contain 1.5 g/L sodium bicarbonate, non-essential amino acids, and sodium pyruvate was purchased from LGC Promochem (Teddington, UK). Epilife medium and HKGS supplement kit (bovine pituitary extract, bovine insulin, hydrocortisone, bovine transferrin) were obtained from Cascade Biologics (Mansfield, UK). Trypsin-EDTA, penicillin G, streptomycin sulphate, amphotericin B, hydrocortisone, adenine, cholera toxin, insulin and fetal calf serum (FCS) were purchased from Invitrogen (Paisley, UK). 2.1.3. Antibodies and ELISA The polyclonal antibody anti-EGFR (phospho Y1173) (rabbit) was obtained from Abcam (Cambridge, UK) and the monoclonal antibody anti-actin (C-2) (mouse) from Santa Cruz Biotechnology (Santa Cruz, USA). A human EGF-ELISA kit was purchased from R&D Systems (Abingdon, UK). 2.2. Methods 2.2.1. Synthesis and characterisation of succinoylated dextrin Dextrin was succinoylated as described previously [26]. Briefly, dextrin (1 g) was added to DMAP (80 mg), purged with nitrogen and DMF added (7.5 mL) to the stirred mixture. Succinic anhydride (182.5 mg) was dissolved in DMF (2.5 mL) and added to the reaction mixture. The solution was stirred under nitrogen at 50 °C for 14 h and the DMF removed by evaporation under high vacuum to leave a few drops of reaction mixture. This was added drop-wise to vigorously stirred diethyl ether (250 mL) and left for 10 h, under continual stirring (Fig. 2). This solution was filtered under vacuum, and the resultant product was purified by dialysis in double-distilled water. Functionalisation of succinoylation dextrin was confirmed qualitatively by FTIR (Avatar 360 E.S.P. Spectrometer and EZ OMNIC E.S.P. 5.2 software (Thermo Nicolet, Loughborough, UK)) using 64-300 scans over 400–4000 cm− 1 with background subtracted. The weight average (Mw), number average (Mn) molecular weight and polydispersity of the polymers were determined with gel permeation chromatography (GPC) using Ultrahydrogel 1000 (200 × 8 mm) and Ultrahydrogel 250 columns (Waters Ltd, Elstree, UK) in series by RI (Gilson, Middleton, USA) and Severn Analytical UV absorbance (Jaytee Biosciences, Whitstable, UK) detectors. Succinoylated dextrin in PBS (3 mg/mL) was injected (60 μL) onto the column and elution profiles were monitored over 30 min. Polysaccharide (pullulan) standards were used to estimate Mw and Mn. 2.2.2. Synthesis and characterisation of dextrin–rhEGF conjugates The succinoylated dextrin intermediate was conjugated to rhEGF. Briefly, succinoylated dextrin (8.2 mg in 500 μL ddH2O) was added to EDC (1.12 mg) and stirred for 10 min, at room temperature. To this, sulfo-NHS (1.2 mg) was added and the solution was stirred for a further 40 min. rhEGF (2 mg in 300 μL ddH2O) was added and the reaction was allowed to proceed for 18 h, pH 8 (Fig. 2). The product was purified using FPLC (AKTA FPLC, UV detector, Unicorn 3.20 software, Amersham Biosciences, Little Chalfont, UK). The conjugate fractions (6–9 mL) were collected and concentrated using Vivaspin 6 centrifugal concentrators (Mw cut-off 30 kDa, Sigma-Aldrich; 3500 g, 10 min), prior to lyophilisation. The BCA protein assay was used, as per the manufacturer's instructions, to determine the total rhEGF content of the conjugates. FPLC and SDS-PAGE were used to determine the levels of free rhEGF (non-conjugated rhEGF) and/or degradation
277
products present in the conjugates. SDS-PAGE was conducted using 12.5% linear gels, with rhEGF and protein molecular weight markers as a reference (Mini-Protean 3 Gel Electrophoresis System, Bio-Rad Laboratories Ltd., Hemel Hempstead, UK). All subsequent concentrations of dextrin–rhEGF are derived from a theoretical 100% degradation of the conjugate. 2.2.3. Degradation of dextrin, succinoylated dextrin and dextrin–rhEGF by α-amylase 2.2.3.1. GPC assay. Dextrin, succinoylated dextrin, and dextrin–rhEGF (all 1 mg/mL) were dissolved in PBS, human salivary α-amylase (93 i.u./L) added, thoroughly mixed, and incubated at 37 °C for 168 h. GPC was used to estimate Mw and Mn as before. 2.2.3.2. FPLC assay. Samples (200 μL) of dextrin–rhEGF (with or without exposure to human salivary α-amylase) were collected, as above. All samples were defrosted and analysed by FPLC. The area of the peaks corresponding to the dextrin–rhEGF conjugate and free rhEGF were determined, and the data shown expressed as the ratio of rhEGF:dextrin–rhEGF with time. 2.2.4. Determination of EGF release by ELISA assay Samples of dextrin–rhEGF were diluted to 250 pg/mL rhEGFequivalent in EGF-ELISA dilution buffer (RD5E; R&D Systems, Abingdon, UK). Samples were then incubated at 37 °C for 168 h in the presence and the absence of α-amylase (93 i.u./L). Samples (200 μL × 3) were taken at times up to 168 h and snap-frozen in liquid nitrogen. The EGF-ELISA assay was then used to quantify the rhEGF content in each sample, according to the manufacturer's instructions and by using a rhEGF standard curve. Data was expressed as rhEGF released as % theoretical total of conjugated rhEGF. 2.2.5. Dextrin–rhEGF conjugate stability in the presence of trypsin and neutrophil elastase The dextrin–rhEGF conjugate and rhEGF (both 1 mg/mL) were dissolved separately in PBS. To assess the stability to proteolytic degradation, trypsin-EDTA (0.025%) or human neutrophil elastase (100 μg/mL) were added to these samples and incubated at 37 °C for 6 h. Samples (200 μL) were taken at times from 0 to 24 h and snapfrozen in liquid nitrogen. At the end of the experiment, all samples were defrosted and analysed by FPLC as described above. The area of the dextrin–rhEGF conjugate and free rhEGF peaks was calculated. 2.2.6. Cell culture HEp2 cells were seeded at 1 × 105 cells in a 75 cm2 tissue culture flasks (Barloworld Scientific, Stone, UK) and incubated in 10 mL of HEp2-serum containing medium (H-SCM: EMEM with penicillin G (100 U/mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL) and 10% v/v FCS), at 37 °C/5% CO2. Upon reaching 80–90% confluence, the cells were sub-cultured, as follows. The cells were washed (×2) with PBS and detached from the plastic by incubation with 1 mL of trypsin/EDTA (0.05% trypsin, 0.53 mM EDTA), for approximately 5 min, at 37 °C. Each flask was washed with H-SCM (9 mL) to collect trypsinised cells. The solution was then transferred to a 15 mL centrifuge tube and centrifuged at 1500 g for 5 min. The supernatant was aspirated and the resulting pellet re-suspended in 1 mL of H-SCM and viable cells (assessed by trypan blue staining) counted, prior to sub-culture. The cell culture medium was changed twice weekly. Immortalised HaCaT keratinocytes (Boukamp [27]) were seeded at 1 × 105 cells in 75 cm2 tissue culture flasks and incubated in 10 mL of keratinocyte-serum containing medium (K-SCM: 67.5% DMEM/22.5% Ham's F12 nutrient media, supplemented with penicillin G (100 U/ mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL), hydrocortisone (400 ng/mL), adenine (0.089 mM), cholera toxin
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
278
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 2. Dextrin–rhEGF synthesis. Dextrin (A) is functionalised with succinic anhydride (B) to produce succinoylated dextrin (C). This reacts with EDC (D) to form an amine-reactive O-acylisourea intermediate (E) and sulfo-NHS (F) then forms an amine-reactive sulfo-NHS ester (G). Lysine residues within the rhEGF (H) molecule form stable amide bonds resulting in dextrin–rhEGF (I).
(0.08 μg/mL), insulin (5 μg/mL), and 10% v/v FCS). Upon reaching 80– 90% confluence, keratinocytes were passaged, as described above. The cell culture medium was changed every 2 days. 2.2.7. Measurement of EGFR phosphorylation by Western blotting In order to investigate whether dextrin–rhEGF retained the ability to induce EGFR-phosphorylation HEp2 cells were grown on 100 mm tissue culture dishes (Cellstar, Greiner Bio-One, UK) and EGFR phosphorylation measured with time. Cells were serum starved for 24 h (in H-SFM:EMEM with penicillin G (100 U/mL), streptomycin sulphate (100 μg/mL), amphotericin B (0.25 μg/mL)) prior to the addition of rhEGF (400 ng/mL in H-SFM), dextrin–rhEGF (400 ng/mL equiv. in H-SFM) or α-amylase-exposed dextrin–rhEGF (400 ng/mL equiv. in H-SFM, containing 93 i.u./L α-amylase, incubated at 37 °C for 24 h) for 2.5 min. At times between 0 and 40 min post-exposure, cells
were washed twice with cold PBS and protein extracted by scraping into 1.0 mL of 50 mM Tris-HCl buffer, pH 7.5, containing 5 mM EDTA and 1 mM dithiothreitol, and sonicated (5 × 20 s, Model W-220F, Ultrasonic Inc., USA). Cell suspensions were harvested, centrifuged at 15,000 g (4 °C) for 20 min and protein content was determined using the Bio-Rad DC Protein Assay (Bio-Rad Laboratories, Hemel Hempstead, UK). Samples (70 μg) were loaded onto SDS-PAGE pre-formed 4–15% gradient gels (Bio-Rad Laboratories, Hemel Hempstead, UK) and transferred onto PVDF membranes (G.E. Healthcare, Chalfont St. Giles, UK). Immunodetection was performed using antibodies against phosphorylated EGFR and actin. 2.2.8. Cell proliferation assay HEp2 cells and HaCaT keratinocytes were seeded into 96-well microtitre plates (BD Biosciences, Oxford, UK), at a cell density of
2.5 × 103 cells/well in 10% SCM and incubated for 24 h. The cells were then incubated in serum-free media (SFM) for a further 24 h. For the proliferation assays stock solutions were prepared as follows: For rhEGF, serial dilutions of a stock solution 6.4 μg/mL in K-SFM (Epilife media + HKGS) and a stock solution of 400 ng/mL in H-SFM, were prepared for addition to HaCaT and HEp2 cells, respectively. In each case, the concentration range chosen was based upon preliminary proliferation assays undertaken using these cells (data not shown). For dextrin–rhEGF, serial dilutions of 6.4 μg/mL rhEGF-equivalent stock solution in K-SFM and 50 ng/ mL rhEGF-equivalent in H-SFM were prepared. The rhEGF and dextrin–rhEGF samples were incubated at 37 °C, with or without, α-amylase (20–521 i.u./L), or with trypsin-EDTA (0.025%) for 24 h. The samples exposed to trypsin-EDTA (0.025%) were purified by FPLC and subsequently exposed to α-amylase (93 i.u./L) for a further 24 h. At the beginning of the proliferation assays, either free rhEGF, dextrin–rhEGF (both without or without prior α-amylase exposure), or rhEGF and dextrin–rhEGF exposed to trypsin-EDTA (0.025%) and α-amylase, were added to cells and incubated at 37 °C. After 67 h, cell proliferation was assayed using MTT. MTT (25 μL; 5 mg/mL in PBS) was added to the cell media and incubated for 5 h at 37 °C. The media was then removed. Lysis buffer was added (200 μL; DMF (50%) in H2O containing SDS (20%), glacial acetic acid (2.5%) and HCl (1 M, 2.5%), pH 4.7) and incubated at
279
37 °C for a further 3 h. The absorbance of each sample was then assayed using a Dynex MRX Spectrophotometer (Dynex Technologies, Worthing, UK) with a 540 nm filter. To examine effects over a longer time period, similar cell proliferation assays were performed with HEp2 cells, but proliferation was assessed daily over 8 days. 2.2.9. Statistical analysis Statistical analyses were undertaken using GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA). Data were compared using a Student's t-test and Mann–Whitney U-test, and a one-way ANOVA with a Bonferroni post-test was used for group analysis. Results are expressed as a mean and standard deviation (S.D.), or standard error of the mean (S.E.M.). All p values are two-tailed. Statistical significance was considered at a probability of p b 0.05. 3. Results 3.1. Synthesis and characterisation of the dextrin–rhEGF conjugates The succinoylated dextrin intermediate was characterised by FTIR spectroscopy to confirm the incorporation of carboxylic acid, indicated by a peak at 1720 cm− 1. By titration, the degree of polymer modification was calculated as ∼ 19 mol%. Reaction efficiency was
Fig. 3. Characterisation of the dextrin–rhEGF conjugate and degradation of dextrin, succinoylated dextrin and the conjugate in the presence of α-amylase. Panels (A) and (B) show FPLC analysis of rhEGF (P) and dextrin–rhEGF (- - - -) and SDS-PAGE analysis, respectively. Panel (C) shows GPC chromatograms of dextrin, succinoylated dextrin and dextrin– rhEGF. Panel (D) shows the change in relative molecular weight (GPC analysis) of dextrin (- - ■ - -), succinoylated dextrin (—▴—), and dextrin–rhEGF (- - ♦ - -), in the presence of αamylase (93 i.u./L), over 24 h (n = 3; mean ± S.D., ⁎p = 0.0734, ⁎⁎ p = 0.0012, ANOVA and Bonferroni post hoc test).
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
280
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
limit N 0.15 μg) (Fig. 3A, B). This value was estimated as FPLC analysis showed a clear distinction between rhEGF and dextrin– rhEGF, and rhEGF was undetectable at concentrations b1.6 μg/mL. For a typical conjugate batch containing 16 wt.% rhEGF, this equated to b1% free rhEGF. 3.2. Studies of the rate of α-amylase triggered degradation, rhEGF release and stability Using pullulan standards, the dextrin–rhEGF conjugate had a Mn 70,000; Mw 106,600; PDI 1.5, however these figures should be viewed as an approximation (Fig. 3C). GPC analysis of the molecular weight change in the presence of α-amylase showed that succinoylated dextrin and the dextrin–rhEGF conjugate were degraded much more slowly (t1/2 N 24 h) than parent dextrin (t1/2 b 6 min) (p = 0.0012) (Fig. 3D). Addition of α-amylase led to sustained release of free rhEGF over time, while ELISA confirmed that 52.7% of the rhEGF was released after 168 h (Fig. 4A). Moreover, the dextrin–rhEGF conjugate showed negligible degradation in α-amylase-free conditions (Fig. 4B), at both 4 °C and 37 °C in PBS for 168 h, or when lyophilised and stored at −20 °C (data not shown). In addition, whereas native rhEGF was rapidly degraded by neutrophil elastase (t1/2 b1 h) (Fig. 4C), the dextrin–rhEGF was much more stable, with only 50% degradation at 24 h. 3.3. Investigation of the ability of dextrin–rhEGF to induce EGFR phosphorylation Western blot analysis showed that addition of rhEGF to HEp2 cells induces EGFR phosphorylation (Fig. 5A). This effect was impeded by dextrin conjugation (Fig. 5B), but following exposure
Fig. 4. Effect of addition of α-amylase on rhEGF release and stability of the conjugate in the presence of neutrophil elastase. Panel (A) shows the percentage free rhEGF quantified using EGF-ELISA on addition of human salivary α-amylase (93 i.u./L) (n = 9; mean ± S.E.M., ⁎p = 0.0012 (Student's t- and Mann–Whitney U-test)). Panel (B) shows the release of free rhEGF (conditions as for panel (A), measured by FPLC) expressed as ratio of free:conjugated rhEGF. Panel (C) shows the stability (FPLC) of rhEGF and dextrin–rhEGF in the presence of neutrophil elastase from human leukocytes (100 μg/ mL) (n = 2; mean ± S.D., ⁎p = 0.0401 (Student's t-test)).
typically 50–60% [26]. GPC showed an increased hydrodynamic volume for the succinoylated dextrin (estimated Mw 85,000 g/ mol), compared to the parent dextrin (estimated Mw 42,000 g/ mol), probably due to coil expansion of the negatively charged polymer. The optimum conditions for synthesis of the dextrin– rhEGF conjugates were found to be a 1:2 molar ratio of dextrin: rhEGF with an 18 h reaction time. Synthesis was reproducible, the mean rhEGF content was 16.3 ± 4.4 wt.% protein by BCA assay (n = 5, mean ± S.D.; detection limit N 1.6 μg/mL) and the free rhEGF content of the conjugate was b1% by FPLC and SDS-PAGE (detection
Fig. 5. Effect of rhEGF and dextrin–rhEGF on EGFR phosphorylation. Western blotting was used to determine EGFR phosphorylation after addition of rhEGF (A), dextrin– rhEGF (B), or dextrin–rhEGF exposed to α-amylase (93 i.u./L) (C).
281
rhEGF was between 0.19 and 6.25 ng/mL compared to native rhEGF, which displayed its highest activity between 3.13 and 100 ng/mL. When α-amylase was added to the incubation medium HEp2 cell proliferation was restored to levels observed after the addition of native rhEGF at 72 h (p = 0.0035, Fig. 6B). The extent of proliferation induced by the conjugate was clearly related to the α-amylase concentration present (Fig. 6C). It is important to note that addition αamylase at a physiological concentration (93 i.u./L), dextrin or succinoylated dextrin did not affect cell proliferation (Fig. 6B). When cells were exposed to rhEGF or the dextrin–rhEGF conjugate (without α-amylase) for 8 days, the native rhEGF caused the greatest increase in proliferation, apparent over the first 5 days of exposure (Fig. 7A). Subsequently proliferation returned the control levels. In contrast, the α-amylase-activated dextrin–rhEGF conjugate initially showed lower activity. At day 5, maximum proliferation was observed and this was equivalent to that for rhEGF at day 2. However, the α-amylaseactivated dextrin–rhEGF conjugate retained a significant stimulation of proliferation, compared to the controls, until day 6. Moreover, the conjugate was significantly more active than rhEGF on days 5 and 6 (p b 0.001, Fig. 7B). Utilising HaCaT keratinocytes to study the effect of rhEGF and conjugate, the biphasic proliferative response was less pronounced then seen for HEp2 cells (Fig. 8A). However, these more clinically relevant cells were sensitive to rhEGF growth stimulation over a wider range of concentrations (0.1–100 ng/mL). Using HEp2 cells, it was also shown that exposure of rhEGF to trypsin, before addition to cells led to a reduction in activity, compared to the controls. In contrast, dextrin– rhEGF exposed to trypsin and then to α-amylase retained bioactivity,
Fig. 6. HEp2 proliferation (assessed using the MTT assay) in the presence of rhEGF and α-amylase-activated dextrin–rhEGF. Panel (A) shows proliferation of HEp2 cells after 72 h, in the presence of rhEGF (—w—) and α-amylase-activated dextrin–rhEGF (- -■- -), at the rhEGF concentrations shown (mean ± S.E.M., n = 18; ⁎p b 0.001, ANOVA and Bonferroni post hoc test). Panel (B) compares the proliferation of HEp2 cells after 72 h, in the presence of rhEGF and the dextrin–rhEGF conjugate in the presence and absence of α-amylase (93 i.u./L) (mean ± S.E.M.; n = 18; ⁎p = 0.0035 (ANOVA and Bonferroni post hoc test)). Controls (amylase 93 i.u./L, dextrin 6.25 mg/mL, 19 mol% succinoylated dextrin 6.25 mg/mL) show no significant effect (p N 0.05; ANOVA and Bonferroni post hoc test). Panel (C) shows that restoration of rhEGF activity was related to the concentration of αamylase added (mean ± S.D., n = 6, ⁎p b 0.01, ⁎⁎p b 0.001, ANOVA and Bonferroni post hoc test).
to α-amylase, the dextrin–rhEGF conjugate was again able to induce EGFR phosphorylation (Fig. 5C). 3.4. Stimulation of HEp2 and HaCaT proliferation HEp2 cells demonstrated and displayed a biphasic pattern of growth in the presence of rhEGF and dextrin–rhEGF. Addition of both the growth factor and conjugate stimulated proliferation at lower concentrations, but inhibited proliferation at higher concentrations (N400 ng/mL) (Fig. 6A). The peak concentration for activity of dextrin–
Fig. 7. Panel (A) shows HEp2 growth curves over 8 days, when cells were grown in serum-free conditions (control), or in the presence of rhEGF or dextrin–rhEGF, following exposure to α-amylase (521 i.u./L). Panel (B) shows growth of HEp2 compared to controls (mean ± S.E.M., n = 12; ⁎p b 0.001, ANOVA and Bonferroni post hoc test).
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
NANOMEDICINE
282
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283
Fig. 8. Effect of rhEGF and dextrin–rhEGF on the proliferation of HaCaT keratinocytes, and the effect of trypsin exposure on rhEGF and dextrin–rhEGF. Panel (A) shows proliferation of HaCaT keratinocytes cells in response to rhEGF and unmasked dextrin– rhEGF after 72 h (mean ± S.E.M.; n = 18; ⁎p b 0.05, ANOVA and Bonferroni post hoc test). Panel (B) shows proliferation of HEp2 cells in response to rhEGF after protease exposure is reduced to that of rhEGF-free controls (p N 0.05), whereas dextrin–rhEGF retains bioactivity compared to rhEGF-free controls (mean ± S.D., n = 6, ⁎p b 0.001, ⁎⁎p b 0.05, ANOVA and Bonferroni post hoc test).
again demonstrating the resistance of the conjugate to proteolytic degradation (Fig. 8B) afforded by this PUMPT conjugate. 4. Discussion The dextrin–rhEGF conjugate described here was synthesised with the aim of protecting the growth factor from the harsh chronic wound fluid environment whilst allowing regeneration of activity with time when exposed to physiological concentrations of α-amylase using the PUMPT concept [15]. Normal serum α-amylase occurs at a range of 30–110 i.u./L [28], whilst levels in chronic wounds occur in a similar range (unpublished data). In this way it was hoped to generate novel therapeutics (other growth factors may also be used) for improved wound repair. Traditional wound therapy involves the use of topical agents to clean, disinfect and stimulate healing. This can be a protracted, painful process that is often associated with high levels of morbidity. Chronic wounds fail to show the controlled inflammatory, proliferative and remodelling phases of normal wound healing. Numerous factors can be responsible for this malfunction, including wound infection, tissue hypoxia, failed angiogenesis and the underlying uncoupling of the normal growth factor cascade. Many attempts have been made to find more effective therapies for chronic wounds, but there has been little success [29]. Natural and synthetic polymers
have been explored as biodegradable matrices or hydrogels for the controlled release of growth factors (reviewed in [30]), but although their topical application in conjugation with EGF can increase the rate of re-epithelialisation [31], little or no significant clinical improvement in chronic wound healing has ever been demonstrated [32,33]. Overall, the limited success of growth factor therapy has been attributed to (i) an inappropriate release profile in the clinical setting, (ii) the lack of stability of the protein in the delivery system or postdelivery, (iii) rapid protein degradation of rhEGF by chronic wound proteases or denaturation by reactive oxygen species (ROS) [34,35], (iv) the lack of stability of the protein in the delivery vehicle used for application, and/or the toxicity of the polymeric carrier used (e.g. a polyethyleinimine-nitric oxide adduct [36]). It should be noted, however, that one approach, Regranex® (Becaplermin), which is a topically applied gel containing recombinant human platelet-derived growth factor (rhPDGF), recently became the first recombinant growth factor to be approved by the US Food and Drug Administration for the acceleration of wound healing and it is currently used in the treatment of diabetic foot ulcers. However, even in this case, Regranex® has been shown to be only effective in selected wounds, in conjunction with good wound care, with no reduction in ulcer recurrence, when compared to good wound care alone [37]. The dextrin–rhEGF conjugates synthesised here contained, on average, 16 wt.% protein and this corresponded to ∼ 1.5 rhEGF molecules per polymer chain. Heterogeneity of the product was unavoidable (evident by SDS-PAGE electrophoresis; Fig. 3B) and this is largely attributable to the polydisperse dextrin fraction used for conjugation. A narrower molecular weight dextrin fraction could be used to improve conjugate homogeneity if needed for clinical evaluation. A dextrin of average molecular weight of 42,000 g/mol with ∼19 mol% succinoylation was chosen based on earlier studies [15] that suggested this would maximise steric ‘masking’ of rhEGF (lower molecular weight dextrin chains cannot) whilst allowing a sufficiently slow rate of dextrin degradation by α-amylase (Fig. 3D) to liberate rhEGF at an optimal rate for sustained bioactivity. Also, whilst αamylase will degrade the glycosidic bonds within the dextrin polymer, the degradation is slowed by the presence of carboxylic groups and amide bonds [26]. Full degradation of the conjugate would require protease activity. It was confirmed that conjugation of dextrin to rhEGF significantly increased its stability towards proteolysis by the clinically most important protease in wound fluid, neutrophil elastase (Fig. 4C). As anticipated polymer conjugation also reduced biological activity of rhEGF. However, consistent with the PUMPT concept, it was demonstrated that α-amylase, when added with concentrations known to be present in human wound fluid was able to unmask/ release rhEGF. α-Amylase-triggered rhEGF liberation was confirmed by both FPLC (Fig. 3B) and ELISA (Fig. 4A). Polymer degradation would, at least initially, be expected to leave oligosaccharides and/or maltose linked to the rhEGF surface. However, the fact that the liberated rhEGF was clearly identifiable by EGF-immunoassay, and moreover, able to stimulate the clinically relevant parameter of EGF-mediated cell proliferation, underlined the integrity of the product. HEp2 cells were chosen for the cell proliferation assays as they have high levels of EGFR, and their relatively rapid doubling time made it possible to explore the effect of dextrin–rhEGF conjugates over the more clinically relevant time frame of 8 days. The importance of α-amylase as a trigger for conjugate activation was clear throughout all the cell proliferation studies, and it was clearly demonstrated that growth stimulation of HEp2 cells by dextrin– rhEGF was directly related to the units of α-amylase present (Fig. 6B, C). In the extended HEp2 cell proliferation assay (Fig. 7A), it was apparent the initial delayed release of rhEGF release from the conjugate led to a lag phase of approximately 24 h in the cellular response compared to that seen for free rhEGF. However, the sustained rhEGF release from the conjugate led to enhanced activity far beyond that achieved with rhEGF alone. This is even
more impressive given that the EGF-ELISA assay showed release of only 52.7% rhEGF after 168 h. The value of a slow, sustained introduction of bioactive rhEGF is underlined by the biphasic, concentration-dependence of rhEGFinduced proliferation (Fig. 6A). Clearly, high local concentrations of rhEGF must be prevented if inhibition of cell growth is to be avoided. Although HaCaT keratinocytes contain lower levels of EGFR, and grow more slowly than HEp2 cells, they can be considered a more clinically relevant model in relation to wound re-epithelialisation. HaCaT cells also showed enhanced growth in the presence of dextrin–rhEGF activated by α-amylase. Furthermore, it was demonstrated that conjugates exposed to trypsin (as a model protease) before α-amylase exposure still retained the ability to promote HaCaT proliferation. Preliminary experiments to investigate the cellular mechanism of action of dextrin–rhEGF in HEp2 cells found that conjugate exposed to α-amylase was able to, like free rhEGF, induce EGFR phosphorylation (Fig. 5). This suggests a direct effect on the EGFR and promotion on signal transduction. In addition, experiments in progress have shown that α-amylase-activated dextrin–rhEGF conjugates can also induce the migration of HaCaT cells and fibroblasts isolated from human chronic dermal wounds (Hardwicke et al., 2008; unpublished). Such enhanced migration and wound repopulation is key to the repair of chronic wounds in vivo. In conclusion, these studies demonstrate for the first time the potential of a bioresponsive polymer therapeutic as a nanomedicine for wound healing. The conjugate uses a clinically welltolerated polymer, dextrin, and rhEGF as a potential growth factor. The positive effects seen for dextrin–rhEGF in the in vitro studies, suggest that further evaluation of the conjugate towards in vivo and clinical trials is warranted. First clinical studies would certainly involve local, topical application of the conjugate, probably to diabetic or venous leg ulcers. These pathologies are causative of the majority of non-healing wounds and, as such, there is real clinical need for a new treatment. However, this conjugate also has potential for i.v. administration as it would be expected to exhibit targeted delivery to wounds developing new blood vessels by angiogenesis and thus, displaying localised vascular hyperpermeability. Given that recent studies have shown that polymer conjugates can be tailored to deliver anticancer combination therapy [38], the approach reported here has also has the versatility to deliver other growth factors (and or their combinations) to enhance wound healing. This study also verifies the broader therapeutic potential of the PUMPT concept, especially using the combination of dextrin and α-amylase as the enzyme is widely distributed in serum and extracellular fluids [39]. Acknowledgements J.H. was funded by a Fellowship from The Healing Foundation (Registered Charity number 1078666) and the Welsh Office for Research and Development (WORD) and the authors would like to acknowledge support from EPSRC platform grant No. EP/C 013220/1. References [1] J. Li, J. Chen, R. Kirsner, Pathophysiology of acute wound healing, Clin. Dermatol. 25 (2007) 9–18. [2] N.B. Menke, K.R. Ward, T.M. Witten, D.G. Bonchev, R.F. Diegelmann, Impaired wound healing, Clin. Dermatol. 25 (2007) 19–25. [3] P. Price, K. Harding, Cardiff Wound Impact Schedule: the development of a condition- specific questionnaire to assess health-related quality of life in patients with chronic wounds of the lower limb, Int. Wound J. 1 (2004) 10–17. [4] S.E. Herrick, P. Slo, an, M. McGurk, L. Freak, C.N. McCollum, M.W. Ferguson, Sequential changes in the histologic pattern and extra-cellular matrix deposition during the healing of chronic venous leg ulcers, Am. J. Pathol. 141 (1992) 1085–1095. [5] R. Blakytny, E. Jude, The molecular biology of chronic wounds and delayed healing in diabetes, Diabet. Med. 23 (2006) 594–608. [6] R. Goldman, Growth factors and chronic wound healing: past, present and future, Adv. Skin Wound Care 17 (2004) 24–35.
283
[7] N.J. Trengove, M.C. Stacey, S. MacAuley, N. Bennett, J. Gibson, F. Burslem, G. Murphy, G. Schultz, Analysis of the acute and chronic wound environments: the role of proteases and their inhibitors, Wound Rep. Regen. 7 (1999) 442–452. [8] S.M. Chen, S.I. Ward, O.O. Olutoye, R.F. Diegelmann, I. Kelman Cohen, Ability of chronic wound fluids to degrade peptide growth factors is associated with increased levels of elastase activity and diminished levels of proteinase inhibitors, Wound Rep. Regen. 5 (1997) 23–32. [9] R. Duncan, The dawning era of polymer therapeutics, Nat. Rev. Drug. Discov. 2 (2003) 347–360. [10] R. Duncan, Polymer conjugates as anticancer nanomedicines, Nat. Rev. Cancer 6 (2006) 688–701. [11] J.M. Harris, R.B. Chess, Effect of PEGylation on pharmaceuticals, Nature Rev. Drug Discov. 2 (2003) 214–221. [12] M. Francesco, F. Veronese, J. Milton Harris (Eds.), Peptide and protein PEGylation III: advances in chemistry and clinical applications, Adv Drug Delivery Rev., vol. 60, 2008, pp. 1–88. [13] S. Shaunak, S. Thomas, E. Gianasi, A. Godwin, E. Jones, I. Teo, K. Mireskandari, P. Luthert, R. Duncan, S. Patterson, P. Khaw, S. Brocchini, Polyvalent dendrimer glucosamine conjugates prevent scar tissue formation, Nat. Biotechnol. 22 (2004) 977–984. [14] M.J. Vicent, E. Perez-Paya, Poly-L-glutamic acid (PGA) aided inhibitors of apoptotic protease activating factor 1 (Apaf-1): an antiapoptotic polymeric nanomedicine, J. Med. Chem. 49 (2006) 3763–3765. [15] R. Duncan, H.R.P. Gilbert, R.J. Carbajo, M.J. Vicent, Polymer Masked-Unmasked Protein Therapy (PUMPT) 1. Bioresponsive dextrin-trypsin and -MSH conjugates designed for a-amylase activation, Biomacromolecules (2008) Web release date: 19th March, DOI 10.1021/bm701073n. [16] Y.B. Kwon, H.W. Kim, D.H. Roh, S.Y. Yoon, R.M. Baek, J.Y. Kim, H. Kweon, K.G. Lee, Y.H. Park, J.H. Lee, Topical application of epidermal growth factor accelerates wound healing by myofibroblast proliferation and collagen synthesis in rat, J. Vet. Sci. 7 (2006) 105–109. [17] J.P. Hong, Y.W. Kim, H.D. Jung, K.I. Jung, The effect of various concentrations of human recombinant epidermal growth factor on split-thickness skin wounds, Int. Wound J. 3 (2006) 123–130. [18] J.P. Hong, H.D. Jung, Y.W. Kim, Recombinant human epidermal growth factor (EGF) to enhance healing for diabetic foot ulcers, Ann. Plast. Surg. 56 (2006) 394–398. [19] H. Lu, J. Chai, M. Li, B. Huang, C. He, R. Bi, Crystal structure of human epidermal growth factor and its dimerization, J. Biol. Chem. 276 (2001) 34913–34917. [20] G. Carpenter, S. Cohen, Human epidermal growth factor and the proliferation of human fibroblasts, J. Cell Physiol. 88 (1976) 227–237. [21] K. Hashimoto, Regulation of keratinocyte function by growth factors, J. Derm. Sci. 24 (2000) S46–S50. [22] R.C. Kurten, et al., Coordinating epidermal growth factor-induced motility promotes efficient wound closure, Am. J. Physiol. Cell Physiol. 288 (2005) C109–C121. [23] J.E. Frampton, G.L. Plosker, Icodextrin: a review of its use in peritoneal dialysis, Drugs 63 (2003) 2079–2105. [24] K.B. Hosie, D.J. Kerr, J.A. Gilbert, M. Downes, G. Lakin, G. Pemberton, K. Timms, A. Young, A. Stanley, A pilot study of adjuvant intraperitoneal 5-fluorouracil using 4% icodextrin as a novel carrier solution, Eur. J. Surg. Oncol. 29 (2003) 254–260. [25] D. Hreczuk-Hirst, D. Chicco, L. German, R. Duncan, Dextrins as potential carriers for drug targeting: tailored rates of dextrin degradation by introduction of pendant groups, Int. J. Pharm. 230 (2001) 57–66. [26] D. Hreczuk-Hirst, L. German, R. Duncan, Dextrins as carriers for drug targeting: reproducible succinoylation as a means to introduce pendant groups, J. Bioact. Compat. Polym. 16 (2001) 353–365. [27] P. Boukamp, R.T. Petrussevska, D. Breitkreutz, J. Hornung, A. Markham, N.E. Fusenig, Normal keratinisation in a spontaneously immortalised aneuploid human keratinocyte cell line, J. Cell Biol. 106 (1988) 761–771. [28] P. Branca, R.M. Rodriguez, J.T. Rogers, D.S. Ayo, J.P. Moyers, R.W. Light, Routine measurement of pleural fluid amylase is not indicated, Arch. Intern. Med. 161 (2001) 228–232. [29] J.R. Mekkes, M.A.M. Loots, A.C. Vanderwal, J.D. Bos, Causes, investigation and treatment of leg ulceration, Br. J. Dermatol. 148 (2003) 388–401. [30] W.M. Saltzman, W.L. Olbricht, Building drug delivery into tissue engineering, Nat. Rev. Drug Discov. 1 (2002) 177–186. [31] A. Tanaka, T. Nagate, H. Matsuda, Acceleration of wound healing by gelatin film dressings with epidermal growth factor, J. Vet. Med. Sci. 67 (2005) 909–913. [32] V. Falanga, W.H. Eaglstein, B. Bucalo, M.H. Katz, B. Harris, P. Carson, Topical use of human recombinant epidermal growth factor (h-EGF) in venous ulcers, J. Dermatol. Surg. Oncol, 18 (1992) 604–606. [33] D.M. Danilenko, et al., Growth factors in porcine full and partial thickness burn repair. Differing targets and effects of keratinocyte growth factor, platelet-derived growth factor-BB, epidermal growth factor, and neu differentiation factor, Am. J. Pathol. 147 (1995) 1261–1277. [34] T. James, M.A. Hughes, G.W. Cherry, R.P. Taylor, Evidence of oxidative stress in chronic venous ulcers, Wound Rep. Regen. 11 (2003) 172–176. [35] R. Moseley, J.E. Stewart, P. Stephens, R.J. Waddington, D.W. Thomas, Extracellular matrix metabolites as potential biomarkers of disease activity in wound fluid: lessons learned from other inflammatory diseases? Br. J. Dermatol. 150 (2004) 401–413. [36] J.A. Bauer, W. Rao, D.J. Smith, Evaluation of linear polyethyleneimine/nitric oxide adduct on wound repair: therapy versus toxicity, Wound Rep. Regen. 6 (1998) 569–577. [37] D.L. Steed, Clinical evaluation of recombinant human platelet-derived growth factor for the treatment of lower extremity ulcers, Plast. Reconstr. Surg. 117 (2006) S143–S149. [38] M.J. Vicent, F. Greco, R.I. Nicholson, R. Duncan, Polymer–drug conjugates as a novel combination therapy for the treatment of hormone-dependent cancers, Angew. Chem. Int. Ed. 44 (2005) 4061–4066. [39] K. Miyakasa, S.S. Rothman, Redistribution of amylase activity accompanying its secretion by the pancreas, Proc. Natl. Acad. Sci. 79 (1982) 5438–5442.
NANOMEDICINE
J. Hardwicke et al. / Journal of Controlled Release 130 (2008) 275–283