Professional Documents
Culture Documents
Pradal 2016
Pradal 2016
Pradal 2016
A R T I C L E I N F O A B S T R A C T
Article history: Intra-articular (IA) injection of extended drug release forms based on biodegradable microparticles holds
Received 6 September 2015 promise for the treatment of joint diseases. However, the fate of microparticles following intra-articular
Received in revised form 13 November 2015 injection is controversial and has not been thoroughly investigated. The aim of this work was therefore to
Accepted 7 December 2015
evaluate the biodistribution of fluorescent poly(lactic acid) particles of different sizes after IA injection in
Available online 10 December 2015
arthritic or healthy mice.
Regardless of the inflammatory status of the joint, 300 nm-nanoparticles leaked from the joint. Due to
Keywords:
inflammation and related increase of vascular permeability, 3 mm-microparticles that were retained in
Microparticles
Arthritic diseases
the non-inflamed synovial membrane leaked from the inflamed joint. Complete retention of 10 mm-
Hyaluronic acid microparticles was observed independently of the joint inflammatory status. Embedding particles in a
Synovial membrane hyaluronic acid gel prolonged the retention of the formulations only in inflamed joints. Depending on
In vivo biodistribution particle’s size, formulations were preferentially eliminated by blood vessels or lymphatic pathways. Poly
Intra-articular administration (lactic acid) particles of 3 mm were biocompatible and retained in knee joints at least for 6 weeks.
This work highlights the need to deliver hyaluronic acid-embedded particles of at least 3 mm to
guarantee their retention in inflamed joints. These results will contribute to the rational design of long-
lasting formulations to treat acute and chronic joint diseases.
ã 2015 Elsevier B.V. All rights reserved.
1. Introduction delivery of the active drug substance directly to the site of action
via intra-articular administration was demonstrated to be a
Rheumatic diseases are painful conditions usually caused by promising strategy. In most studies, rapid drug clearance from
inflammation. The inflammatory component can have variable the joint requires frequently repeated injections, which are at risk
degrees, such as in rheumatoid arthritis, or can be low as to induce infection (Albert et al., 2006). Thus, maintaining an
encountered in osteoarthritis (OA). Historically, treatments for effective local therapeutic concentration of the drug over longer
chronic arthritic diseases were essentially focused on pain relief by time periods is needed. This can be achieved using IA delivery of
oral dosage forms. However, the poor bioavailability of active drug biodegradable particles. Several studies report particles sizes
substances in the joints and the major side effects that result from ranging from nanometers (Rothenfluh et al., 2008) to hundreds of
distribution in non-target organs do not allow to efficiently micrometers (Liggins et al., 2004) for IA administration (Horisawa
alleviate the symptoms of the patients. Viscosupplementation and et al., 2002a; Liggins et al., 2004; Monkkonen et al., 1995; Pradal
et al., 2015; Ratcliffe et al., 1986; Rothenfluh et al., 2008; Setton,
2008). The formulation, technical and physiological parameters
Abbreviations: AIA, antigen-induced arthritis; CTL, control; DiD, near-red have to be taken into account to deliver sufficient therapeutic
DilC18(5) fluorescent dye; HA, hyaluronic acid; PLA, poly(D,L)lactide; H&E, doses to the site of action. Concerning formulation parameters,
hematoxylin and eosin; OA, osteoarthritis; PBS, phosphate buffered saline; SDS, administration of large particles may help reduce the amount
sodium dodecyl sulfate.
injected and provide extended release properties. Additionally,
* Corresponding author at: University of Geneva, Sciences II, 30 quai E.-Ansermet,
CH-1211 Geneva 4, Switzerland. drug encapsulation efficiency usually increases with the size of
E-mail address: eric.allemann@unige.ch (E. Allémann). particles, which is an asset. Small particles are easier to inject, but
http://dx.doi.org/10.1016/j.ijpharm.2015.12.015
0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
120 J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129
might be rapidly cleared from the joint cavity through fenestra- Staufen, Germany) in a PVAL solution (2% w/v) at an organic/
tions of the synovial membrane. The size of particles was shown to aqueous volume phase ratio of 1:3. Stirring speed and duration of
influence their fate in the joint (Horisawa et al., 2002a; Rothenfluh the emulsification were adjusted to obtain particles of 300 nm,
et al., 2008). To target cartilage or synovium, small nanoparticles 3 mm and 10 mm. The freshly prepared emulsions were then
should be formulated because larger particles would be phagocy- poured in milliQ water and stirred at 500 rpm for 2 h (Eurostar
tized by macrophages (Rothenfluh et al., 2008). Nevertheless, these Digital, IKA Labortechnik, Staufen, Germany).
studies remain unclear about the optimal size of particles to be Particles were centrifuged and washed once with PBS
used for extended residence time in the joints. Specifically, it is still supplemented with 0.1% SDS. The particles were centrifuged a
unclear in how far the joint inflammation itself may influence the second time and washed with milliQ water. Blank particles, i.e.,
biodistribution of nano- and microparticles. without the fluorescent dye, were also prepared using the same
In addition to selecting the appropriate size of particles, method and parameters and were used as control formulations.
embedding them in a hyaluronic acid (HA) gel may prolong their HA-embedded particles were prepared by mixing particles
persistence in the joints and provide viscosupplementation. suspended in PBS with an appropriate amount of concentrated
Injections of HA gels are frequently done in patients suffering HA gel at 3% (w/v) with a magnetic stirrer. The final concentration
from OA to alleviate pain via a cushioning effect. Although positive of HA was 0.6% (w/v) in these formulations.
results have been reported with the use of HA in OA patients
(Bellamy et al., 2006), the long-term clinical efficacy of IA HA
2.3. Characterization of particles
remains controversial (Arrich et al., 2005; Jubb et al., 2003).
The aim of the present study was to determine the optimal size
The size of the microparticles was determined by laser light
of particles that are retained in the mouse joint in the presence or
diffraction (Mastersizer S, Malvern Instruments Ltd., Malvern, UK).
absence of inflammation. Poly(D,L-lactide) (PLA) nano- and micro-
A Nano ZS (Malvern Instruments Ltd., Malvern, UK) was used to
particles containing DiD as a fluorescent dye were formulated and
determine the size of nanoparticles. The Nano ZS was also used to
characterized. The compatibility of the particles with synovial
assess the zeta potential of micro- and nanoparticles. DiD loading
fibroblasts was investigated in vitro. Then, particles were injected
in particles was quantified by spectrofluorometry (Safire, Tecan,
to antigen-induced arthritis (AIA) mice and to naïve mice. Using
Männedorf, Switzerland) after dissolving the particles overnight in
intravital fluorescence and fluorescence quantification in organs,
acetone.
persistence and leakage of particles from mice joints could be
DiD in vitro release kinetics from particles were carried out in
determined. The optimal size of particles was determined, taking
sink conditions by suspending 20 mg of particles in 40 mL of PBS
into account the optional dispersion of particles in HA.
containing 0.5 % SDS (m/v), and also in ethyl oleate to mimic a
physiologic and cell membrane environment, respectively. Liquid
2. Materials and methods
synovial fluid from patients suffering from arthritis was also used
as a release medium. The in vitro release samples were placed on a
2.1. Materials
see-saw rocking plate (60 rpm) at 25 C. Homogenization of the
vials before sampling was realized by upside down shaking.
Poly(D,L)-lactide (PLA, inherent viscosity 0.91 dl g1; molecular
Aliquots of 50 mL were withdrawn and centrifuged at 26,500 g
weight: 153 kDa; PI: 1.6 determined by GPC) (Resomer1 R 207) was
using an Avanti 30 Centrifuge (Beckman, California, USA) before
provided by Boehringer Ingelheim Pharma GmbH (Ingelheim,
spectrofluorimetric assay of the supernatant. All experiments were
Germany). Hyaluronic acid (HA) (Intrinsic viscosity by Ubbelode
performed in triplicate.
viscometer: 2.76 m3 kg1, molecular weight: 1.8 MDa) was pur-
chased from HTL (Javène-Fougères, France).
2.4. Viability tests
DiD (1,10 -dioctadecyl-3,3,30 ,30 -tetramethylindodicarbocyanine,
4-chlorobenzenesulfonate salt) was purchased from Life Technol-
Toxicity of fluorescent particles was assessed by viability tests
ogies (Grand Island, New York, USA). Poly(vinyl alcohol) (PVAL)
performed on synovial fibroblasts from OA patients. For these
was a gift from Clariant GmbH (Frankfurt, Germany) (Mowiol1 4-
experiments, cells were deposited in a 96-well plate at a density of
88, hydrolysis degree 88%, molecular weight: 26 kDa). Dichloro-
30,000 cells/well. The following day, all aqueous suspensions of
methane and acetone were of analytical grade. Sodium dodecyl-
dye-loaded or blank particles, as well as particles dispersed in the
sulfate (SDS) and ethyl oleate were provided by Sigma–Aldrich
HA gel, were freshly prepared and adjusted to concentrations of
(Saint Louis, Missouri, United States).
1 mg/mL of polymer material. Then, 200 mL of the formulations
Human synovial fibroblasts were provided by the Division of
were added to the cells for 24 h at 37 C. Each experimental
Rheumatology from The University Hospitals of Geneva and were
condition was replicated 5 times. Then, supernatants were
obtained from a 76-year-old patient undergoing joint replacement
carefully aspirated. Viability was assessed by MTT tests. To perform
for osteoarthritis. Samples were obtained after appropriate
the tests, 50 mL of 0.1% MTT solution were added to the wells and
informed consent, and their use for research was approved by
left in contact for 3 h. Then, 200 mL of dimethyl sulfoxide was
the ethics committee of the University Hospital of Geneva. Cells
added before absorbance was measured at 595 nm.
were cultured in penicillin–streptomycin supplemented with
DMEM according to an established protocol (Pradal et al., 2015).
2.5. Localization of particles in naïve mice and in mice with antigen-
All cell experiments were carried out below passage 10. All cells
induced arthritis
were tested and maintained in mycoplasma-free conditions.
All in vivo experiments were performed in compliance with the
2.2. Formulation of particles Swiss Federal Law on the Protection of the Animals. The protocols
were accepted by the canton of Geneva Authority (Direction
PLA nano- and microparticles were prepared by a solvent- Générale de la Santé, Authorization number GE/4/14).
evaporation method in the dark. PLA R207 (20 mg mL1) and DiD For the arthritis model, 56 C57Bl/6 male 8–10 week old mice
(2% w/w) were first dissolved in dichloromethane and then (Charles Rivers, France) were divided into 8 groups of 7 mice each
emulsified (homogenizer T25 Ultra-Turrax, IKA Labortechnik, (Table 1). Inflammation was induced as previously described
J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129 121
Table 1
Animal groups for the in vivo studies. Description of injected formulations in inflamed and naïve knee joints with size details. Formulations were suspended in PBS or mBSA in
the case of the naïve model or inflamed mouse models, respectively.
(Brackertz et al., 1977; Pradal et al., 2015). Briefly, to induce the first emission wavelength of 720 nm was extracted from the set of
immunization, an intradermal injection at 3 weeks before Day images and further analyzed with Image J software (National
0 was performed at the tail root with 100 mL of 2 mg/mL Institutes of Health, Bethesda, Maryland, USA). The mean
methylated bovine serum albumin (mBSA) (Fluka, Switzerland) fluorescence (FU/pixel2) multiplied by the area of the regions of
emulsified 1:1 with Freund’s complete adjuvant (Sigma–Aldrich, interest (pixel2) was expressed as the fluorescence integrated
Switzerland) containing 1 mg/mL heat-inactivated and dried intensity (FU). Auto-fluorescence images were acquired on mice
Mycobacterium tuberculosis. At 2 weeks before Day 0, a second from the control group. This background was subtracted from the
immunization was performed by intradermal injection of 100 mL of corresponding subsequent images.
2 mg/mL mBSA emulsified with Freund’s incomplete adjuvant
(Sigma–Aldrich, Switzerland). On Day 0, inflammation was 2.8. Sacrifice and harvest of organs
induced by an intra-articular injection concomitantly with a
single treatment injection of 10 mL in the left knee. The At 4, 8 or 42 days after the induction of the inflammation, a
formulations of particles or HA-dispersed particles were dispersed number of animals was sacrificed. Practically, deep anesthesia was
in mBSA (10 mg/kg). In this way, multiple injections were avoided induced by isoflurane and the total blood was collected by intra-
and the risk of joint infection was reduced. For all experiments, the cardiac puncture. Then, the mice were sacrificed by spinal cord
left knee was used to test samples. The right knee was injected only dislocation. After animal dissection, organs and tissues of interest
with PBS (no mBSA) and considered as the control non-inflamed were harvested and rinsed with NaCl 0.9% and blotted dry. Ex vivo
joint (CTL). full organ fluorescence was quantified by imaging with the
To test samples in naïve knees, 49 mice were divided into Maestro with the same settings as previously indicated except
9 groups (Table 1). Formulations of particles or HA-dispersed the exposure time, which was 500 ms. Heart, kidney, liver, spleen,
particles (10 mL) were also injected into the left knee. All brain and lung were imaged. The organs and other tissues of
formulations contained particles at a final concentration of interest (lymph nodes) were then frozen at 20 C and kept for
36 mg/mL and in the case of HA-dispersed particles, HA was at a further analysis. Knees were collected for histology.
final concentration of 0.6% (w/v).
2.9. Histology
2.6. Quantification of inflammation
After fixation in 10% formaldehyde and decalcification in 4%
Joint inflammation was assessed by measuring the accumulation Tris–EDTA solution for 4 weeks, all knee joints were cut in the
of 99mTc sodium pertechnetate in the knee on Days 1 and 4 after sagittal plane. After embedding in paraffin, 4 mm sections were cut
induction of inflammation. Briefly, 10 mCi of 99mTc sodium pertech- with a microtome. Sections were stained with hematoxylin–eosin
netate was subcutaneously injected. Thirty minutes later, external (HE) or toluidine blue and graded by one pathologist (CAS) in a
gamma counting was used to measure accumulation of the isotope in blinded manner. The severity of the synovial inflammation,
both knees. The ratio of 99mTc accumulation in the inflamed knee to including synovial hyperplasia and the percentage of polymor-
the contralateral control knee was calculated. Joints were considered phonuclear cell infiltration, as well as the degree of cartilage
as inflamed when the ratio was greater than 1.1. erosion and bone destruction, were evaluated and scored from 0 to
4 (0 = normal, 1 = minimal, 2 = moderate, 3 = severe, 4 = very severe)
2.7. Fluorescence imaging (Camps et al., 2005). Particles in the joints were also analyzed by a
reverse fluorescent microscope.
Intravital fluorescence was carried out with a Maestro
M1 imaging system (PerkinElmer, Cambridge Research and 2.10. Biodistribution study
Instrumentation Inc., Massachusetts). Images were acquired with
a cooled CCD camera system at exposure times of 5 and 500 ms One milliliter of distilled water was added to 100 mL of freshly
from 680 to 950 nm (each 10 nm). Fluorescence emission scans harvested blood to induce hemolysis. Hemolyzed blood was freeze
were performed after 5 min, 1 day, 4 days, 8 days, and then, once a dried and 3 mL of (N,N-dimethylformamide (DMF)) were then
week for 6 weeks, using the provided excitation and emission filter added to extract the fluorescent dye. Mechanical agitation was
set (excitation wavelength between 615 and 665 nm, longpass performed overnight and was followed by centrifugation
emission filter to collect light above 700 nm). In most cases, the (2800 g, 20 min). Supernatants were collected and centrifuged
122 J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129
Fig. 1. Intravital fluorescence surface areas of knees after IA injection of particles on Day 0 (within 5 min after injection), and Days 1, 4 and 8 (n = 7 except D8: n = 5, SD). A)
naïve model, B) arthritic model. Significant differences were labeled with *(p < 0.05) and **(p < 0.01) and NS indicates a non-significant difference.
Fig. 2. Intravital mean fluorescence of knees after IA injections of particles on Day 0 (5 min after injection), Days 1, 4 and 8 after IA injections. Mean SD, n = 7, except at Day 8
(n = 5). (A) naïve model, (B) arthritic model.
spreading of particles was only observed as a significant decrease Based on the initial results (Figs. 1 and 2), with the perspective
of the mean fluorescence under inflamed conditions during the of the formulation of a delivery system with extended release
first 24 h (Fig. 2B, Day 1 vs Day 0, p = 3.104 for Nano + HA, p = 5.104 properties over weeks, formulation Micro 3 and Micro 3 + HA were
for Micro 3 + HA). However at Day 1, neither the fluorescent area selected for a longer term study in naïve mice. For that,
nor mean fluorescence were significantly different comparing biodistribution of particles was assessed for 6 weeks (Fig. 3).
particles versus HA-embedded particles, irrespective of the Spreading of the 3 mm particles, as observed by the increase of the
inflammatory status of the joint. fluorescent area, was also noted within the first day after the IA
Fig. 3. Intravital fluorescence (A) area and (B) integrated intensity from Micro 3 (circle) and Micro 3 embedded in HA (square) in knees of the naïve model over 6 weeks
(mean SD, n = 7).
124 J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129
injection. Then, a slight decrease of the fluorescent areas was the smallest particles injected into naïve joints (Fig. 6 A, Nano vs
observed during the first two weeks for both formulations. Nano + HA, p = 0.01). DiD concentration in the lymph nodes after
Although larger areas were initially obtained with microparticles injection of Micro 3 was lower in naïve joints than in inflamed
embedded in HA, on Day 14, the fluorescent areas converged ones.
toward similar values for both formulations tested. From Day 14 to
Day 28, the areas remained almost constant for both formulations
(Fig. 3A). The integrated intensity of the fluorescence followed a
similar trend. Fluorescence levels then decreased from Day 28 until
the end of the experiment on Day 42 (Fig. 3B). Clearly, the retention
of 3 mm particles injected in knee joints of mice could be observed
for at least 6 weeks.
Fig. 5. Percentage of injected dose of DiD assessed by fluorescence spectroscopy in the main organs of (A) naïve and (B) inflamed mice model, 8 days after IA injection.
(mean SD, n = 5, except for both formulations of Nano in the naïve model where n = 4).
3.6. Histological analysis revealed the presence of particles in all healthy joints (Table 4,
Fluorescence columns and Fig. 8A–C). Regarding the long-term
The histological analysis of inflamed and naïve joints was study, particles were still observed even after 42 days and
performed to confirm and complete results obtained by gamma associated histological scores were below 1 (Fig. 8G).
counting and fluorescence methods. Mice were sacrificed on Days Strong and acute inflammation (synovial hyperplasia and
8 (n = 5) and 42 (n = 7). Scores ranging from 0 to 4 were attributed inflammatory infiltration (granulation tissue)) and focal abscesses
for levels of inflammation, erosion of cartilage, erosion of bone and of the control mBSA group (Fig. 7B) were observed and confirmed
presence of polymorphonuclear cells (PMN) in each treated knee by an inflammatory score of 3.5 (Fig. S3), consistently with 99mTc
cross-section (Fig. 7). Absolutely no signs of injury were observed scintillation counting (Fig. S2). In no case particles increased
in contralateral knees (Fig. 7A). inflammation and induced higher erosion in inflamed joints
When nano- or microparticles were injected into naïve joints, compared with the mBSA group. By contrast, inflammation on Day
all histological scores were below or equal to 1, which corre- 8 was slightly lower in knees injected with Nano or Micro 3,
sponded to a minor inflammatory reaction (Table 4). Cartilage was compared with knees injected with Micro 10, HA or only mBSA
preserved, as indicated by homogenous toluidine blue staining (Fig. S3). Unlike the reduction of inflammation in the presence of
(data not shown). A thin synovial membrane, foamy macrophages HA revealed by gamma counting, no effect of the gel was observed
in small foci and no inflammatory cell infiltration were observed as by histological scoring on Day 8 (Fig. 7C and Fig. S3). Irrespective of
indicated by HE staining (Fig. 7D–F). Although foamy macro- the particle size, no foamy macrophages were observed by
phages, indicative for ingested particles, were observed in histology in all inflamed joints (Table 4, Histology columns).
approximately 60% of the cases irrespective of the particle size, However, fluorescence allowed the identification of particles
only Micro 10 particles were identified by histology in the articular (Fig. 8D–F).
cavity (Table 4, Histology columns) after 8 days (Fig. 7I). However, Supplementary material related to this article found, in the
observation of histological sections by fluorescence microscopy online version, at http://dx.doi.org/10.1016/j.ijpharm.2015.12.015.
Fig. 6. Percentage of injected dose of DiD assessed by fluorescence spectroscopy in lymph nodes in naïve model (A) and arthritic model (B) on Day 8 after inflammation
induction (mean SD, n = 5, except for both formulations of Nano in the naïve model where n = 4).
126 J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129
Fig. 7. Histology of mouse knee joints 8 days after intra-articular injection as specified below and stained with H&E: (A) PBS (negative control): normal joint: smooth cartilage,
no inflammatory infiltrate either in the synovial tissue or in the joint cavity, (B) mBSA (positive control) and (C) HA: joint with AIA: strong inflammation of the synovial tissue
and the joint cavity, Injection of (D) Nano, (E) Micro 3, (F) Micro 10 particles (scale bar = 500 mm) in the non-inflammatory setting: smooth cartilage surface, thin synovial
membrane, no inflammatory infiltrate. Enlargement of rectangles drawn on slides B, C and F are, respectively, presented in the following slides: (G) mBSA (scale bar = 200 mm)
and (H) HA (scale bar = 100 mm): thickened synovial membrane, inflammation rich in neutrophils (I) Micro 10 ingested by tissue macrophages organized in a small centralized
area close to the synovial capsule (scale bar = 50 mm). Particles and HA were injected in naïve joints. Mice were sacrificed on Day 8. Magnification: 5 (A-F), 20 (G and H) and
40 (I).
Table 4
Histological analysis of mouse knee joints on Days 8 and 42 after intra-articular administration of nano- and microparticles. Inflammation, foamy macrophages and particles
were analyzed by conventional light microscopy and complemented with fluorescence microscopy.
Inflammatory Naïve joint (Day 8) Inflamed joint (Day 8) Naïve joint (Day 42)
status
Method Histology Fluorescence Histology Fluorescence Histology Fluorescence
Observation Inflammation Foamy Particles Particles Inflammation Foamy Particles Particles Inflammation Foamy Particles Particles
score macro- score macro- score macro-
phages phages phages
PBS 0.0 0/1 0/1 0/1 0.0 0/2 0/2 0/2
mBSA 3.5 0.6 0/5 0/5 0/5
HA 3.5 0.6 0/5 0/5 0/5
Nano 0.6 0.3 2/4 0/4 4/4 2.6 0.5 0/5 0/5 2/5
Nano + HA 0.8 0.3 4/4 0/4 4/4 2.8 0.5 0/5 0/5 3/5
Micro 3 0.8 0.7 4/5 0/5 5/5 2.5 0.5 0/5 0/5 5/5 0.5 0.1 7/7 7/7 7/7
Micro 3 + HA 0.3 0.2 3/5 0/5 5/5 2.5 0.5 0/5 0/5 4/5 0.4 0.2 6/7 6/7 7/7
Micro 10 0.7 0.3 4/5 2/5 5/5 3.6 0.8 0/5 0/5 4/5
Micro 10 + HA 1.0 0.0 5/5 4/5 5/5 3.6 0.4 0/5 0/5 5/5
J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129 127
Fig. 8. Histology of mouse knee joints under fluorescence microscopy: 8 days after intra-articular injection : (row 1) in healthy knee joints at Day 8, (row 2) in arthritic knee
joints at Day 8, (row 3) in healthy knee joints at Day 42, (column 1) Nano, (column 2) Micro 3, (column 3) Micro 10. Magnification 40.
inflamed and naïve joints (Fig. 7F and I, naïve joint and Table 4). addition to considering their size for the development of particles
This was also confirmed by very low DiD levels in the liver and providing extended drug release properties, their biocompatibility
demonstrates retention of particles in the joint cavity as a function is of importance. In our study, the scores remaining low even after
of their size. In short, Nano escaped from inflamed and naïve joints, 6 weeks confirms the biocompatibility of PLA particles with the IA
whereas Micro 3 escaped only from inflamed joints. The capillary joint tissues (Table 4, Fig. S3 and Fig. 8G).
permeability is thought to be the reason for this joint efflux The size of particles is important not only with regard to
(Kushner and Somerville, 1971). Synovial fluid and serum tend to retention or leakage from the joint, but also with regard to
equilibrate (Simkin, 1995) through synovium intercellular gaps interaction with cells present in the joint. According to Edwards
(Knight and Levick, 1984) under naïve conditions. This physiologi- and Dong (Dong et al., 2013; Edwards et al., 2007), the size of
cal turnover drives the leakage of the Nano and the limiting particles needs to be less than 10 mm in order to be phagocytized
permeability (or “mesh”) of the healthy synovium prevents the by macrophages, which delays their clearance from the joint cavity
escape of Micro 3. Actually, retention of Micro 3 into the healthy and increases the drug concentration directly in the inflammatory
joint cavity for at least 6 weeks was confirmed by both intravital cells to be treated (Butoescu et al., 2009). After sacrifice of the
imaging and by fluorescence microscopy (Fig. 8G). treated mice, histological observations showed that the presence
Joint inflammation is characterized by enhancement of the of Nano HA, Micro 3 HA and Micro 10 HA in the joint,
capillary permeability. In our study, this was demonstrated by the regardless of arthritic status, did not increase inflammation scores,
significant accumulation of 99mTc in inflamed joints (Fig. S2). which indicates their innocuity in this particular site (Fig. S3). As
Inflammation also leads to an increase of the synovial membrane previously observed by the authors (Pradal et al., 2015) and also
volume and a rise of the metabolic activity (Loeuille et al., 2009). reported by Horisawa et al. (2002a), administration of particles
However, the high amount of inflammatory cells limited the significantly larger than 10 mm (e.g., 20–50 mm) led to the
detection of particle carrying cells by classical histology. Nano- and production of giant cells and foreign body reactions (Anderson
microparticles could still be detected by fluorescence microscopy and Shive, 2012). However, in the present study, the presence of
in all specimens. These physiopathological modifications were giant cells around 10 mm particles was not observed.
distinctly illustrated by the spreading of Nano and Micro 3 out of In addition to an appropriate selection of the particles’ sizes,
the synovial cavity (Fig. 1A: Nano HA and Fig. 1B: Nano HA and embedding them into hydrogel could be beneficial in prolonging
Micro 3 HA). We deduced that the synovial fluid/serum the drug residence time in the joints and to reduce sedimentation
equilibrium was displaced toward the serum in cases of and possible aggregation of particles in the joint cavity. Although
inflammation as also suggested earlier by Simkin (1995). In the long term favorable effect of HA itself for OA patients remains
128 J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129
controversial (Altman et al., 2000; Brandt et al., 2000; Felson and preferentially drained through the iliac route (Fig. 6). Afterwards,
Anderson, 2002; Pendleton et al., 2000), the commercial for- both ways drain in the rostral direction into the subclavian vein via
mulations are well tolerated by patients. In our study it did not the thoracic duct (Harrell et al., 2008). Although Micro 3 were read
alter the viability of synovial fibroblasts (Fig. S1). Moreover, a non- (Fig. 4A and B) or quantified (Fig. 5A and B) in similar amounts in
significant trend toward decreased 99Tc sodium pertechnetate the liver, they seemed to escape differently from the joint
level was observed, only for HA without particles (Fig. S2). This according to the inflammatory status. In the case of inflamed
might suggest a decreased inflammation, consistent with some conditions, the elimination of Micro 3 by the lymphatic pathway
literature reports (Fujii et al., 2001; Mccourt et al., 1994). was increased. As discussed above, 10 mm particles were massively
Additionally, HA down-regulates TNF-alpha and proinflammatory retained in the joint but still some escaped and were found in the
cytokines (Takahashi et al. 1999; Wang et al., 2006). Intravital lymph nodes (Fig. 6). This is not surprising considering lymphatic
imaging indicated that HA gels improved the retention of small drainage, which is physiologically able to remove cartilage debris
particles that tend to leak in the inflamed knee model (Fig. 4A and from the joint (Horwitz, 1948; Saito et al., 2002).
C). Transient improvement of HA was observed on the retention of
the particles over 1 week (Fig. 3). The semi-quantitative imaging of 5. Conclusion
the full organs following sacrifice and DiD quantification both
confirmed a favorable effect of HA against the leakage of particles Intra-articular administration of drug-releasing particles has
(Nano and Micro 3) in the inflamed model on Days 4 and 8 (Figs. 4 B encountered growing interest, prompting the need to better
and C and 5 B and Fig. S4). All HA embedded formulations injected understand particle retention in the joints and biodistribution in
in an arthritic joint were found in lower concentrations in remote organs.
abdominal organs compared with formulations of the same sizes In this report, size-dependent transport of the particles was
suspended in saline (Fig. 5). This improved retention may be linked observed and is consistent with an inflammation-dependent
to specific interaction of HA with CD44 expressed by chondrocytes capillary permeability. Nanoparticles may diffuse rapidly out of
and, to a lesser extent, by synoviocytes (Elron-Gross et al., 2009; the joint in both inflamed and naïve models, whereas large 10 mm-
Laroui et al., 2007). This might also be related to HA-induced particles are essentially retained. The mid-size 3 mm-particles are
viscosification leading to reduced turnover. well retained for 6 weeks in the naïve model. Embedding particles in
Supplementary material related to this article found, in the hyaluronic acid favored the retention of particles. Depending on the
online version, at http://dx.doi.org/10.1016/j.ijpharm.2015.12.015. size, the presence of HA and the inflammatory status of the joint,
The biodistribution of particles in various organs after IA biodistribution through blood circulation or lymphatic pathways
administration was assessed and is discussed with regard to was modulated. Long-term administration of DiD-loaded PLA
particles’ size. Material elimination from the joints can go through particles demonstrated biocompatibility and dye entrapment over
the microvasculature and/or through the lymphatic pathway. 6 weeks that opens possibilities for the development of drug delivery
Independently of the elimination route, particulate material tends systems for chronic arthritic diseases.
to accumulate in the liver, which was effectively observed (Fig. 4).
After injecting particles into inflamed knees, buildup of particles in
References
the liver remained very low but was slightly higher for the smallest
particles. This confirmed that leakage from the joint is dependent Albert, C., Brocq, O., Gerard, D., Roux, C., Euller-Ziegler, L., 2006. Septic knee arthritis
on the particles' sizes (Figs. 4 B and C and 5). Presence of DiD in after intra-articular hyaluronate injection: two case reports. Joint Bone Spine
distant organs was assessed. Concentrations in brain and lungs 73, 205–207.
Altman, R., Hochberg, M., Moskowitz, R., Schnitzer, T., 2000. Recommendations for
were below the quantification limit (data not shown). Actually, the medical management of osteoarthritis of the hip and knee: 2000 update.
since none of the investigated particles had a smaller size than Arthritis Rheum. 43, 1905–1915.
300 nm, nothing was expected to cross the blood–brain barrier but Anderson, J.M., Shive, M.S., 2012. Biodegradation and biocompatibility of PLA and
PLGA microspheres. Adv. Drug Deliv. Rev. 64 (Suppl), 72–82.
deserved to be confirmed with regard to safety (Kreuter, 2001; Arrich, J., Piribauer, F., Mad, P., Schmid, D., Klaushofer, K., Mullner, M., 2005. Intra-
Kreuter, 2014; Olivier, 2005). articular hyaluronic acid for the treatment of osteoarthritis of the knee:
The microvascular pathway represents the main elimination systematic review and meta-analysis. CMAJ 172, 1039–1043.
Bastiat, G., Pritz, C.O., Roider, C., Fouchet, F., Lignieres, E., Jesacher, A., Glueckert, R.,
pathway for very small particles from the naïve joint. In our study, Ritsch-Marte, M., Schrott-Fischer, A., Saulnier, P., Benoit, J.P., 2013. A new tool to
on Day 8, all nanoparticles were cleared from the blood (Fig. 5). ensure the fluorescent dye labeling stability of nanocarriers: a real challenge for
This clearance is attributed to efficient opsonization and uptake by fluorescence imaging. J. Control Release 170, 334–342.
Bellamy, N., Campbell, J., Robinson, V., Gee, T., Bourne, R., Wells, G., 2006.
the mononuclear phagocyte system (Harashima et al., 1994; Ishida
Intraarticular corticosteroid for treatment of osteoarthritis of the knee.
et al., 2002; Moghimi et al., 2001). Because of this rapid clearance Cochrane Database Syst. Rev. CD005328.
from the blood, the microvascular pathway may be discussed only Bonanomi, M.H., Velvart, M., Weder, H.G., 1987. Fate of different kinds of liposomes
on indirect facts such as the amount of DiD detected in the liver and containing dexamethasone palmitate after intra-articular injection into rabbit
joints. J. Microencapsulation 4, 189–200.
in the lymph nodes (Figs. 4 and 5). Although no particles were Bozdag, S., Calis, S., Kas, H.S., Ercan, M.T., Peksoy, I., Hincal, A.A., 2001. In vitro
observed in muscles or knee-surrounding tissues at sacrifice, the evaluation and intra-articular administration of biodegradable microspheres
very small amount of Nano detected in the lymph nodes (Fig. 6) containing naproxen sodium. J. Micoencapsulation 18, 443–456.
Brackertz, D., Mitchell, G.F., Mackay, I.R., 1977. Antigen-induced arthritis in mice. I.
indicates that the main elimination route for nanoparticles is Induction of arthritis in various strains of mice. Arthritis Rheum. 20, 841–850.
vascular. For Micro 3 under inflamed conditions, the permeability Brandt, K.D., Smith Jr., G.N., Simon, L.S., 2000. Intraarticular injection of hyaluronan
of the capillary bed occurred together with a relatively high as treatment for knee osteoarthritis: what is the evidence? Arthritis Rheum. 43,
1192–1203.
accumulation in the lymph nodes (Fig. 6). This suggests that Micro Butoescu, N., Jordan, O., Doelker, E., 2009. Intra-articular drug delivery systems for
3 are eliminated by both microvascular and lymphatic elimination the treatment of rheumatic diseases: a review of the factors influencing their
pathways. performance. Eur. J. Pharm. Biopharm. 73, 205–218.
Camps, M., Rückle, T., Ji, H., Ardissone, V., Rintelen, F., Shaw, J., Ferrandi, C., Chabert,
The lymphatic pathway removes particles and debris from the C., Gillieron, C., Françon, B., Martin, T., Gretener, D., Perrin, D., Lerroy, D., Vitte, P.
joint space independently of their size (Wallis et al., 1987). Harell A., Hirsch, E., Wymann, M., Cirillo, R., Schwarz, M., Rommel, C., 2005. Blockade of
et al. highlighted that approximately half of the hind leg lymphatic PI3K suppresses joint inflammation and damage in mouse models of
rheumatoid arthritis. Nat. Med. 11, 936–943.
drainage enters the inguinal lymphatic node and the remaining
Dong, J., Jiang, D., Wang, Z., Wu, G., Miao, L., Huang, L., 2013. Intra-articular delivery
goes through the iliac lymphatic nodes near the aortic bifurcation. of liposomal celecoxib–hyaluronate combination for the treatment of
In our study, particles and HA-dispersed particles were osteoarthritis in rabbit model. Int. J. Pharm. 441, 285–290.
J. Pradal et al. / International Journal of Pharmaceutics 498 (2016) 119–129 129
Edwards, S.H.R., Cake, M.A., Spoelstra, G., Read, R.A., 2007. Biodistribution and biocompatibility and efficacy determinations in rabbits. Inflamm. Res. 53, 363–
clearance of intra-articular liposomes in a large animal model using a 372.
radiographic marker. J. Liposome Res. 17, 249–261. Loeuille, D., Rat, A.C., Goebel, J.C., Champigneulle, J., Blum, A., Netter, P., Gillet, P., Chary-
Elron-Gross, I., Glucksam, Y., Margalit, R., 2009. Liposomal dexamethasone– Valckenaere, I., 2009. Magnetic resonance imaging in osteoarthritis: which
diclofenac combinations for local osteoarthritis treatment. Int. J. Pharm. 376, method best reflects synovial membrane inflammation?: correlations with
84–91. clinical, macroscopic and microscopic features. Osteoarthr. Cartil. 17, 1186–1192.
Felson, D.T., Anderson, J.J., 2002. Hyaluronate sodium injections for osteoarthritis: Mccourt, P.A.G., Ek, B., Forsberg, N., Gustafson, S., 1994. Intercellular-adhesion
hope, hype, and hard truths. Arch. Intern. Med. 162, 245–247. molecule-1 is a cell-surface receptor for hyaluronan. J. Biol. Chem. 269, 30081–
Fujii, K., Fujii, Y., Hubscher, S., Tanaka, Y., 2001. CD44 is the physiological trigger of 30084.
Fas up-regulation on rheumatoid synovial cells. J. Immunol. 167, 1198–1203. Moghimi, S.M., Hunter, A.C., Murray, J.C., 2001. Long-circulating and target-specific
Harashima, H., Sakata, K., Funato, K., Kiwada, H., 1994. Enhanced hepatic uptake of nanoparticles: theory to practice. Pharmacol. Rev. 53, 283–318.
liposomes through complement activation depending on the size of liposomes. Monkkonen, J., Liukkonnen, J., Taskinen, M., Heath, T.D., Urtti, A., 1995. Studies on
Pharm. Res. 11, 402–406. liposome formulations for intra-articular delivery of clodronate. J. Control
Harrell, M.I., Iritani, B.M., Ruddell, A., 2008. Lymph node mapping in the mouse. J. Release 35, 145–154.
Immunol. Methods 332, 170–174. Nishide, M., Kamei, S., Takakura, Y., Tamai, S., 1999. Fate of biodegradable D,L-lactic
Horisawa, E., Kubota, K., Tuboi, I., Sato, K., Yamamoto, H., Takeuchi, H., Kawashima, acid oligomer microspheres in the articulus. J. Bioact. Compat. Polym. 14, 385–
Y., 2002a. Size-dependency of DL-lactide/glycolide copolymer particulates for 398.
intra-articular delivery system on phagocytosis in rat synovium. Pharm. Res. 19, Olivier, J.C., 2005. Drug transport to brain with targeted nanoparticles. NeuroRx 2,
132–139. 108–119.
Horisawa, E., Kubota, K., Tuboi, I., Sato, K., Yamamoto, H., Takeuchi, H., Kawashima, Pendleton, A., Arden, N., Dougados, M., Doherty, M., Bannwarth, B., Bijlsma, J.W.J.,
Y., 2002b. Size-dependency of DL-lactide/glycolide copolymer particulates for Cluzeau, F., Cooper, C., Dieppe, P.A., Günther, K.-P., Hauselmann, H.J., Herrero-
intra-articular delivery system on phagocytosis in rat synovium. Pharm. Res. 19, Beaumont, G., Kaklamanis, P.M., Leeb, B., Lequesne, M., Lohmander, S., Mazieres,
132–139. B., Mola, E.-M., Pavelka, K., Serni, U., Swoboda, B., Verbruggen, A.A., Weseloh, G.,
Horwitz, T., 1948. Bone and cartilage debris in the synovial membrane; its Zimmermann-Gorska, I., 2000. EULAR recommendations for the management
significance in the early diagnosis of neuro-arthropathy. J. Bone Joint Surg. Am. of knee osteoarthritis: report of a task force of the Standing Committee for
30A, 579–588. International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann.
Ishida, T., Harashima, H., Kiwada, H., 2002. Liposome clearance. Biosci. Rep. 22, 197– Rheum. Dis. 59, 936–944.
224. Pradal, J., Zuluaga, M.-F., Maudens, P., Seemayer, C.A., Doelker, E., Gabay, C., Jordan,
Jubb, R.W., Piva, S., Beinat, L., Dacre, J., Gishen, P., 2003. A one-year, randomised, O., Allémann, E., 2015. Intra-articular bioactivity of a p38 MAPK inhibitor, VX-
placebo (saline) controlled clinical trial of 500–730 kDa sodium hyaluronate 745, and development of an extended-release system. Eur. J. Pharm. Biopharm.
(Hyalgan) on the radiological change in osteoarthritis of the knee. Int. J. Clin. 93, 110–117.
Pract. 57, 467–474. Ratcliffe, J.H., Hunneyball, I.M., Wilson, C.G., Smith, A., Davis, S.S., 1986. Albumin
Knight, A.D., Levick, J.R., 1984. Morphometry of the ultrastructure of the blood–joint microspheres for intra-articular drug-delivery: investigation of their retention
barrier in the rabbit knee. Q. J. Exp. Physiol. 69, 271–288. in normal and arthritic knee joints of rabbits. J. Pharm. Pharmacol. 39, 290–295.
Kreuter, J., 2001. Nanoparticulate systems for brain delivery of drugs. Adv. Drug Rothenfluh, D.A., Bermudez, H., O'Neil, C.P., Hubbell, J.A., 2008. Biofunctional
Deliv. Rev. 47, 65–81. polymer nanoparticles for intra-articular targeting and retention in cartilage.
Kreuter, J., 2014. Drug delivery to the central nervous system by polymeric Nat. Mater. 7, 248–254.
nanoparticles: what do we know? Adv. Drug Deliv. Rev. 71, 2–14. Saito, I., Koshino, T., Nakashima, K., Uesugi, M., Saito, T., 2002. Increased cellular
Kushner, I., Somerville, J.A., 1971. Permeability of human synovial membrane to infiltrate in inflammatory synovia of osteoarthritic knees. Osteoarthr. Cartil. 10,
plasma proteins. Relationship to molecular size and inflammation. Arthritis 156–162.
Rheum. 14, 560–570. Setton, L., 2008. Peptide-functionalized polymer nanoparticles target and bind to
Laroui, H., Grossin, L., Leonard, M., Stoltz, J.F., Gillet, P., Netter, P., Dellacherie, E., articular cartilage tissue, making them promising drug-delivery vehicles. Nat.
2007. Hyaluronate-covered nanoparticles for the therapeutic targeting of Mater. 7, 172–174.
cartilage. Biomacromolecules 8, 3879–3885. Simkin, P.A., 1995. Synovial perfusion and synovial fluid solutes. Ann. Rheum. Dis.
Levick, J.R., 1984. Blood flow and mass transport in synovial joints. Handbook of 54, 424–428.
Physiology. Section 2. The Cardiovascular System. American Physiology Society, Takahashi, K., Goomer, R.S., Harwood, F., Kubo, T., Hirasawa, Y., Amiel, D., 1999. The
Bethesda, pp. 917–947. effects of hyaluronan on matrix metalloproteinase-3 (MMP-3), interleukin-1b
Liang, L.S., Jackson, J., Min, W., Risovic, V., Wasan, K.M., Burt, H.M., 2004. (IL-1b), and tissue inhibitor of metalloproteinase-1 (TIMP-1) gene expression
Methotrexate loaded poly(L-lactic acid) microspheres for intra-articular during the development of osteoarthritis. Osteoarthr. Cartil. 7, 182–190.
delivery of methotrexate to the joint. J. Pharm. Sci. 93, 943–956. Wallis, W.J., Simkin, P.A., Nelp, W.B., 1987. Protein traffic in human synovial
Liang, L.S., Wong, W., Burt, H.M., 2005. Pharmacokinetic study of methotrexate effusions. Arthritis Rheum. 30, 57–63.
following intra-articular injection of methotrexate loaded poly(L-lactic acid) Wang, C.T., Lin, Y.T., Chiang, B.L., Lin, Y.H., Hou, S.M., 2006. High molecular weight
microspheres in rabbits. J. Pharm. Sci. 94, 1204–1215. hyaluronic acid down-regulates the gene expression of osteoarthritis-
Liggins, R.T., Cruz, T., Min, W., Liang, L., Hunter, W.L., Burt, H.M., 2004. Intra-articular associated cytokines and enzymes in fibroblast-like synoviocytes from patients
treatment of arthritis with microsphere formulations of paclitaxel: with early osteoarthritis. Osteoarthr. Cartil. 14, 1237–1247.