Encapsulation of the p38 MAPK inhibitor GSK 678361A in nanoparticles for inflammatory‐based disease states

Inhibitors of p38 mitogen‐activated protein kinase (MAPK) are currently being pursued as therapeutics in inflammatory conditions, but many candidates have demonstrated limited efficacy or toxicity issues to date. Nanoformulation of p38 MAPK inhibitors may overcome these challenges, by enabling controlled release and targeted delivery. Thus, the aim of this study was to develop a nanoformulation of the p38 MAPK inhibitor GSK 678361A and subsequently validate its anti‐inflammatory efficacy in vitro, versus the drug in its free format. Poly(lactic‐co‐glycolic acid) nanoparticles encapsulating GSK 678361A were prepared via a salting‐out method and characterised by photon correlation spectroscopy, scanning electron microscopy and high‐performance liquid chromatography. The anti‐inflammatory effect of both free and nanoformulated GSK 678361A was evaluated in cultures of lipopolysaccharide‐stimulated macrophages, with subsequent enzyme‐linked immunosorbent assay analysis of TNF‐α and IL‐6 providing readouts of efficacy. A controlled release nanoformulation of GSK 678361A was successfully developed, with physicochemical characterisation revealing an average particle diameter of 115.5 ± 3.5 nm and polydispersity index of 0.13 ± 0.03, indicative of a homogeneous size distribution. GSK 678361A loading was quantified at 10.1 ± 0.4 µg per mg of poly(lactic‐co‐glycolic acid), equating to an entrapment efficiency of approximately 50%. When tested in cultures of lipopolysaccharide‐stimulated macrophages, GSK 678361A nanoparticles inhibited the production of pro‐inflammatory cytokines to an extent that was largely comparable with the free drug, although superior efficacy of the nanoformulation was observed at selected doses. These studies indicate that GSK 678361A may be successfully nanoformulated without loss of drug activity, warranting further evaluation in models of inflammation in vivo. © 2016 The Authors. Journal of Interdisciplinary Nanomedicine published by John Wiley & Sons Ltd and the British Society for Nanomedicine


Introduction
p38 Mitogen-activated protein kinase (MAPK) is a serinethreonine kinase that forms part of the mammalian MAPK family, alongside other members such as extracellular signal-regulated kinase and c-Jun NH 2 -terminal kinase (Kim and Choi, 2010). It exists in several isoforms, namely, p38-α, -β, -γ and -δ, which are activated in response to environmental stresses or inflammatory cytokines (Cuenda and Rousseau, 2007). Upon activation, p38 MAPK mediates a diverse array of functions, including the regulation of cell differentiation, proliferation and apoptosis (Coulthard et al., 2009). Moreover, it also performs a fundamental role in initiating pro-inflammatory cytokine production and has been shown to modulate cytokine expression at both transcriptional and translational levels (Cargnello and Roux, 2011). As such, p38 MAPK inhibitors are currently being pursued as potential treatment candidates in conditions underpinned by inflammation.
To overcome such limitations, p38 MAPK inhibitors may be integrated within targeted drug delivery systems such as nanoparticles. This strategy affords numerous advantages, including enhanced bioavailability and a reduction in toxicity arising from off-target effects. Moreover, nanoformulation enables sustained release of the payload with a subsequent reduction in dosing schedules and enhanced patient compliance (Gelperina et al., 2005;Parveen et al., 2012).
In this study, we aimed to develop a polymeric nanoparticle platform encapsulating the p38 MAPK inhibitor GSK 678361A (Triantaphyllopoulos et al., 2010), with controlled release properties. We examined the ability of this nanoparticle to inhibit pro-inflammatory cytokine production by murine macrophages in both a timedependent and dose-dependent manner.
GSK 678361A entrapment efficiency GSK 678361A entrapment efficiency was determined using high-performance liquid chromatography (HPLC). The purified NP pellet was dissolved in DMSO and acetonitrile (ACN) (1:3) and measured at 254 nm wavelength using a validated HPLC method. HPLC comprised a Varian Prostar Spectra system with a C-18 HPLC column (50 × 4.6 mm diameter, 5 μM pore size) (Phenomenex, UK). The mobile phase (flow rate of 1 mL/min) consisted of solvent A (H 2 O) and solvent B (ACN + 0.05% TFA) with elution over a linear gradient of solvent B 10-100% over 5 min. Results were compared to a standard calibration curve of GSK 678361A solution in DMSO : ACN (1:3). Entrapment efficiency was calculated as follows:

Nanoparticle characterisation
Nanoparticle size and polydispersity index (PDI) were measured in PBS using photon correlation spectroscopy (Zetasizer Nano ZS, Malvern instruments, UK). Measurements were carried out at room temperature. Each sizing determination was carried out in triplicate, and an average particle size was expressed as the mean diameter (Z ave ). Routine size measurements and surface morphology were confirmed by scanning electron microscopy (SEM). Nanoparticles were mounted onto copper tape covering an aluminium stub, sputter coated with gold and visualised by SEM (Jeol 6500 field emission gun, Japan).

Evaluation of GSK 678361A release from poly (lactic-co-glycolic acid) nanoparticles
Release profiles of GSK 678361A were determined using a dialysis-based method. A mass of nanoparticles equivalent to approximately 150 μg GSK 678361A was dispersed in 2 mL growth media (Dulbecco's modified Eagle's medium supplemented with 10% low endotoxin FCS, 1% penicillin and streptomycin). The nanoparticle solution was injected into a pre-wetted Slide-A-Lyzer® dialysis cassette with a 7 K molecular weight cut-off (Pierce Biotechnology, USA) using a 21 gauge bevelled needle and immersed into a reservoir of 28 mL pre-warmed growth media. The release study was performed at 37°C under conditions of continuous stirring. Samples (200 μL) were removed from the reservoir at defined time points. GSK 678361A concentration was measured using a validated HPLC protocol as described previously and compared with a calibration curve for standard GSK 678361A solutions prepared under similar conditions.

Cell preparation and culture
Bone marrow-derived macrophages (BMDM) were isolated from the femur and tibia of C57BL/6 female mice. The femur and tibia were removed from the female donors, and marrow was flushed out using a 23 gauge needle and syringe containing growth media. The bone marrow was pelleted at 650 rcf for 8 min at 4°C, and lysis buffer was used to lyse red blood cells. Cells were differentiated into macrophages by culturing them in the presence of growth media containing 20% L929 supernatant as a source of macrophage-colony stimulating factor. Culture medium was changed on day three to remove non-adherent cells. After seven days, cells were counted and seeded into 96 well plates at a density of 5 × 10 4 cells/well and allowed to rest overnight in non-L929 supplemented growth media.
Effect of GSK 678361A on lipopolysaccharide-stimulated cytokine production The efficacy of GSK 678361A in either the free or nanoencapsulated form was assessed using murine BMDM. Cells were stimulated with 100 ng/mL E. coli R515 LPS and left to incubate for 1 h prior to further treatment with 2 μM free GSK 678361A or 2 μM GSK 678361A NP reconstituted in low endotoxin PBS. Cells were incubated for a further 5 h, after which supernatants were collected and analysed by ELISA.

GSK 678361A dose response
Bone marrow-derived macrophages (BMDM) were pre-treated with LPS as described previously and then further treated with increasing concentrations (ranging from 10 À4 to 10 μM/225 μL) of GSK 678361A suspended in DMSO or GSK 678361A NP reconstituted in low endotoxin PBS. Cells were incubated for a further 5 h, after which supernatants were collected and analysed by ELISA. Cytokine concentrations were compared to those of cells treated with the respective positive controls, DMSO or Blank NP, as a means of calculating therapeutic percentage inhibition of cytokine levels.

Determination of GSK 678361A effect over time
Bone marrow-derived macrophages (BMDM) were treated with 2 μM free GSK 678361A or 2 μM GSK 678361A NP reconstituted in low endotoxin PBS at 1 h post-LPS treatment as described previously. Supernatants were collected at pre-determined time points and analysed by ELISA.

Statistical analysis
Statistical significance between data groups was determined using Student's t-test or one way analysis of variance (ANOVA) with Tukey post-hoc test. Where statistical significance was achieved, it was represented by asterisks in the appropriate figures and defined as: ***P < 0.001, **P < 0.01, *P < 0.05. All analyses were performed using GraphPad Prism software (GraphPad software, USA).

Nanoparticle characterisation
Nanoparticle size analysis was performed via photon correlation spectroscopy, revealing an average diameter of 115.5 ± 3.5 nm for GSK 678361A NP (Fig. 1A). The size distribution of the GSK 678361A NP formulation was relatively monodisperse, as indicated by a low PDI value of 0.13 ± 0.03 (Fig. 1A). Using validated HPLC methods, the entrapment efficiency of GSK 678361A within PLGA nanoparticles was measured at 50.6 ± 2.2%, equating to 10.1 ± 0.4 μg drug per mg of polymer (Fig. 1A). SEM analysis also confirmed the presence of a homogeneous population of nanoparticles, with similar diameters to those obtained via photon correlation spectroscopy (Fig. 1B). Finally, the drug release from these particles was examined at 37°C under continuous stirring conditions, demonstrating a typical biphasic release kinetic, with an initial burst release of 75.7 ± 8.4% of the drug during the first 24 h, followed by a slower sustained release lasting until the 144 h time point (Fig. 2).

Effect of GSK 678361A on lipopolysaccharide-stimulated cytokine production
Following optimisation of the GSK 678361A NP formulation, functional assays were performed to assess the anti-inflammatory efficacy of both GSK 678361A NP and free GSK 678361A in LPS-stimulated BMDM cultures. As anticipated, LPS stimulation significantly increased TNF-α and IL-6 levels to 427.1 ± 33.6 and 397.7 ± 4.5 pg/mL, respectively ( Fig. 3A and B). Subsequent treatment with 2 μM GSK 678361A was found to significantly reduce cytokine levels to 113.0 ± 20.4 pg/mL (TNF-α) and 235.9 ± 39.6 pg/mL (IL-6) when used in its free form, and to 132.8 ± 8.4 pg/mL (TNF-α) and 216.2 ± 4.3 pg/mL (IL-6) upon nano-encapsulation. No significant difference in cytokine levels was observed between free and nano-encapsulated GSK 678361A, indicating that both treatments were comparable in efficacy and that drug activity was preserved upon loading into nanoparticles.

GSK 678361A dose response
Having confirmed anti-inflammatory efficacy of free and nano-encapsulated GSK 678361A in BMDM cultures, we next examined the dose-dependency of these effects. Both TNF-α and IL-6 levels were inhibited by GSK 678361A NP in a dose-dependent manner ( Fig. 4A and B). Maximal inhibition was observed at a concentration of 10 μM, where TNF-α and IL-6 levels were significantly reduced by the GSK 678361A NP (70.9 ± 3.1% and 61.4 ± 1.4%, respectively), when compared with free GSK 678361A (50.2 ± 2.0% and 25.1 ± 6.0%, respectively).

Determination of GSK 678361A effect over time
The inhibitory activity of GSK 678361A in both its free and nano-encapsulated form was next evaluated as a function of time in LPS-stimulated BMDM. At all time points, both free and nano-encapsulated GSK 678361A significantly reduced TNF-α levels when compared with the LPS control (Fig. 5A). The inhibitory effect of GSK 678361A NP on TNF-α production was comparable to that of the free drug for up to and including the 6 h time point. Similar findings were also noted for IL-6 at 4 and 6 h, although the anti-inflammatory effect of both free and nano-encapsulated GSK 678361A had diminished by 8 h (Fig. 5B).

Discussion
In this current work, we have examined the ability to encapsulate the p38 MAPK inhibitor GSK 678361A, a biphenylamide originally characterised to produce anti-inflammatory effects in a murine collagen-induced arthritis model (Aston et al., 2009;Triantaphyllopoulos et al., 2010). We have demonstrated in this current work that the inhibitor can be successfully formulated into a consistent PLGA nanoparticle formulation, producing anti-inflammatory effects towards LPSstimulated macrophages. The p38 MAPKs play central roles in the progression of signaling pathways that promote a number of pathological conditions such as rheumatoid arthritis and asthma (Schett et al., 2008;Coulthard et al., 2009;Chung, 2011). GSK 678361A was originally developed as an inhibitor of p38-α, although it and the structurally related clinical lead compound, GW856553X (Losmapimod), were both found to selectively inhibit p38-β (Aston et al., 2009), blocking pro-inflammatory cytokine release. Consistent with this earlier work, here, we found that GSK 678361A was able to block release of TNF-α and IL-6 from murine BMDM in both its free and nano-encapsulated form.
Nanoparticle encapsulation represents an area of formulation science aimed at improving the bioavailability and/or delivery of drug cargos to the disease site. Various materials can be employed for the manufacture of such nanoformulations including lipids for the production of liposomes, which remains the most widely used nanomedicine approach applied clinically (Chang and Yeh, 2012). However, other  nanoformulation approaches can employ biodegradable and biocompatible polymer systems, which degrade more slowly than liposomes, providing a sustained or controlled release window (Kumari et al., 2010;Cheng et al., 2015). These include polymer systems such as naturally derived alginate or chitosan nanoparticles, or synthetic PLGA nanoparticles (Dhar et al., 2008;Deacon et al., 2015;Fredman et al., 2015). We have previously worked extensively with PLGA-based nanosystems, demonstrating the ability to enhance the targeting of nanoparticles through attachment of antibodies and other ligands on their surface, generating therapeutically useful effects (Schmid et al., 2014;Spence et al., 2015).
We have also previously examined polymeric nanoparticle systems to provide sustained release of active pharmaceutical ingredients (APIs), demonstrating enhanced effectiveness of aminoglycoside antibiotics (Abdelghany et al., 2012;Deacon et al., 2015). In this current work, we found that nanoformulation of GSK 678361A provided a typical biphasic controlled release under in vitro conditions, releasing 75% of the drug by 48 h, cumulating in 100% by 4-5 days. Using a range of concentrations, we comparatively examined the ability Figure 5. Time-dependent inhibition of inflammatory cytokine production by GSK 678361A NP: (A) Murine bone marrow-derived macrophages were pre-treated with 100 ng/mL lipopolysaccharide (LPS) for 1 h prior to stimulation with 2 μM free GSK 678361A or the equivalent amount of GSK 678361A NP. Culture supernatants were collected after (i) 4, (ii) 6 and (iii) 8 h and assayed for TNF-α by ELISA. (B) Murine bone marrow-derived macrophages were treated as in (A), and culture supernatants were assayed for IL-6 by enzyme-linked immunosorbent assay. Statistical significance was assessed by one way ANOVA with Tukey post-hoc test (**P < 0.01, ***P < 0.001). Data expressed as mean + SEM.
of both free and nano-encapsulated drugs to inhibit TNF-α and IL-6 production, showing superior efficacy of nanoformulated GSK 678361A at 10 μM in particular. These findings indicated that nanoformulation did not abrogate the bioavailability of the drug to the cells. Finally, we examined the persistence of the antiinflammatory action of GSK 678361A, revealing that both free and nano-encapsulated drug inhibited the production of inflammatory cytokines similarly up to 6 h in vitro. Although TNF-α inhibition was also maintained at 8 h, a loss of efficacy of both free and nanoencapsulated drug was observed for IL-6 at this time point. These findings highlight potential differences in the manner by which p38 MAPK controls individual cytokine production and warrants further investigation. Beneficial effects of the nano-encapsulation of GSK 678361A were only observed at higher concentrations, although this is not surprising as usefulness of the controlled release effect of the formulation can be difficult to determine in simplistic closed incubation conditions and is in keeping with what we and others have observed previously (Abdelghany et al., 2012;Kolate et al., 2015). However, it is possible that the nanoparticles may elicit further enhanced therapeutic effects using in vivo study models, where the pharmacokinetics of the API can be altered significantly by nanoencapsulation. Furthermore, it is well established that PLGA nanoparticles are relatively hydrophobic and susceptible to opsonisation by macrophages in circulation; given the key role that these cells play in mediating inappropriate inflammation, nanoformulations of GSK 678361A may enhance delivery of the drug to this target cell to provide enhanced effects.
In summary, here, we have developed and characterised a PLGA nanoparticle formulation encapsulating the anti-inflammatory drug GSK 678361A. We have shown that the activity of this API when nanoformulated is equivalent to the free drug and warrants its further evaluation in models underpinned by acute and chronic inflammation.