stealth liposomes encapsulating zoledronic acid: a new opportunity to treat neuropathic pain

Stealth Liposomes Encapsulating Zoledronic Acid: A NewOpportunity To Treat Neuropathic PainMichele Caraglia,† Livio Luongo,‡ Giuseppina Salzano,§ Silvia Zappavigna,† Monica Marra,†

Francesca Guida,§ Sara Lusa,§ Catia Giordano,‡ Vito De Novellis,‡ Francesco Rossi,‡

Alberto Abbruzzese Saccardi,† Giuseppe De Rosa,*,§ and Sabatino Maione‡

†Department of Biochemistry and Biophysics “F. Cedrangolo” and ‡Department of Experimental Medicine, Second University ofNaples, Via Costantinopoli, 16 80138 Naples, Italy§Department of Pharmacy, University Federico II of Naples, Via Montesano, 49, 80131 Naples, Italy

ABSTRACT: In the pathogenesis of neuropathic pain, theconversion of astrocytes in the reactive state and the ras-dependent Erk-mediated pathway play an important role.Zoledronic acid (ZOL) is a potent inhibitor of the latterpathway, but its activity in neurological diseases is hamperedby its biodistribution that is almost exclusively limited to thebone. We have developed nanotechnological devices able toincrease the accumulation of ZOL in extra bone sites. In thiswork, we have evaluated the effects of ZOL-encapsulatingPEGylated liposomes (LipoZOL) on an animal model ofneuropathic pain. We have found that 2 iv administrations (10μg of ZOL, either as free or encapsulated into liposomes) atdays 2 and 4 after the injury markedly reduced mechanicalhypersensitivity at 3 and 7 days after nerve injury. On the other hand, free ZOL did not exert any significant alteration of themechanical threshold. Immunohistochemical analysis of spinal cord revealed that GFAP-labeled astrocytes appeared hypertrophicactivated cells in the ispilateral dorsal horn of spinal cord 7 days after SNI. LipoZOL significantly changed astrocyte morphology,by inducing a protective phenotype, without changing the total cell number. Moreover, the astrocytes of the spinal cord ofLipoZOL-treated mice were positive for interleukin-10. Delivery of ZOL into the CNS was confirmed by biodistribution offluorescently labeled liposomes. In particular, liposomes accumulated in the liver and kidney in both groups of normal andneuropathic animals; on the other hand, only in the case of neuropathic animals, a fluorescence increase in the brain and spinalcord occurred only in neuropathic animals at 30 min and 1 h. These data demonstrate that ZOL, only by using a delivery systemable to cross the altered BBB, could be a new opportunity to treat neuropathic pain.

KEYWORDS: stealth liposomes, zoledronic acid, aminobisphosphonates, neuropathic pain, liposome biodistribution

■ INTRODUCTION

Clinical management of chronic pain after nerve injury(neuropathic pain) and tumor invasion (cancer pain) is anobjective difficult to achieve because the cell mechanisms thattrigger and sustain chronic pain are not still definitivelycharacterized. Glial cells, such as microglia and astrocytes in thecentral nervous system (CNS), are suggested to play animportant role in the development and maintenance of chronicpain. In detail, chronic pain results from the occurrence ofneural plasticity in both peripheral nervous system (PNS) andCNS.1−3 Glial cells are more abundant than neurons in theCNS and can be divided in three different groups: (i)astrocytes, (ii) microglia, and (iii) oligodendrocytes. Microgliaare cells that display phenotypical signatures similar tomacrophages, thus playing a scavenger function in CNS. Themost present cell components for both number (about the halfof glial cells) and size are astrocytes.4 Both microglia andastrocytes, in physiological conditions, are relatively not in a

proliferating status.5 After external damage leading to apathological condition, they can pass toward a reactive status,participating in the processes leading to the occurrence ofneurological diseases.6−8

In fact, glial and microglial cells are involved in the neuronalsensitization occurring in the dorsal horn of the spinal cordafter peripheral nerve injury-induced neuropathic pain.9−11 Indetail, microglia and then astrocytes are activated by the fiberdamage and by the neuronal release of several nociceptivemediators such as ATP, glutamate, and substance P (SP), thusinducing a spinal wind up phenomenon responsible for theestablishment of mechanical allodynia, which represents theneurological symptom of chronic neuropathic pain. Once

Received: October 29, 2012Revised: December 27, 2012Accepted: January 17, 2013Published: January 17, 2013

Article

pubs.acs.org/molecularpharmaceutics

© 2013 American Chemical Society 1111 dx.doi.org/10.1021/mp3006215 | Mol. Pharmaceutics 2013, 10, 1111−1118

pubs.acs.org/molecularpharmaceutics

activated, glial cells are able to release pro-inflammatorycytokines and chemokines such as CCL2, IL-1β, and TNFα,thus continuously sensitizing neurons by interacting with theirspecific receptors. The pro-inflammatory state of these cells isalso associated with a change in morphology that becomeshypertrophic and/or round shape, retracting the processes.12

In the last process, astrocytes overexpress intermediatefilament proteins, such as vimentin, nestin, and glial fibrillaryacidic protein (GFAP), proteoglycans, and other molecules thatinduce axon growth inhibition. The most important changefound in astrocytes is not induction of proliferation butcytoplasm enlargement and migration toward the sites wheredamage was generated.Epidermal growth factor (EGF) and transforming growth

factor α (TGFα) are likely involved in the wake up ofastrocytes as it has been described that EGF receptor can beupregulated and hyperactivated in astrocytes after damage tothe CNS.13 However, the downstream signal transductioncomponents activated by EGF in these conditions have notbeen completely defined.Recent findings suggest an important role played by mitogen-

activated protein kinases (MAPKs)ERK, p38, and JNKinthe development of neuropathic pain.14 Interestingly, the threedifferent MAPKs have a different timing of activation in spinalcord glial cells after nerve damage. In fact, p38 kinase ispersistently stimulated, ERK is activated only in the earlystages,15−17 while pERK induction has a late onset (more than21 days from the damage).18,19

The intracranial administration of a MEK inhibitorantagonizes both the late onset stimulation of ERK and theoccurrence of the associated mechanical allodynia, suggesting arole for astrocytic ERK in sustaining chronic pain.19 ERK-1/2and their cognate kinases can be under the control of the rasfamily proteins that usually trigger the MAPK cascades.20

Moreover, it was recently reported that the Rheb (a ras familymember)−mTOR pathway is up-regulated in reactive astro-cytes of the injured spinal cord.Zoledronic acid (ZOL) is a member of the pharmacological

agents named as aminobisphosphonates (NBPs) that are agentsindicated for the treatment of bone demineralization caused byboth osteoporotic conditions and tumor metastases. ZOL actsas a potent inhibitor of farnesyl pyrophosphate synthase andcompletely abolishes the synthesis of both farnesylpyrophos-phate and geranylgeranylpirophosphate, inhibiting isoprenyla-tion processes. Therefore, ZOL suppresses prenylation of allsmall GTPases, including Ras family proteins.20 Theprenylation process is needed for the compartmentalization ofras proteins at the inner side of the plasma membrane wherethey are activated by external signals.21,22

Unfortunately, one of the most important limitations of ZOLis its pharmacokinetic profile. In fact, pharmacokinetic studieshave demonstrated that ZOL, following a standard intravenousinfusion, is still detectable in the plasma only for 1−2 h beforeits accumulation in the bone.23,24 In these conditions, ZOLestimated distribution and elimination plasma half-lives are 15(t1/2) and 105 min (t1/2β), respectively, with a peak plasmaconcentration after the end of infusion (Cmax) of approximately1 μM.25

Biodistribution studies in rats and dogs with single ormultiple intravenous doses of 14C-labeled ZOL have shown thatZOL is avidly uptaken by bone, where it accumulates,maintaining high levels over the time. Moreover, ZOL iscontinuously and slowly released in the plasma from the bone,

achieving a terminal half-life of approximately 240 days.26 Onthis background, we designed a formulation based on stealthliposome-encapsulating ZOL (LipoZOL) to reduce the bindingof ZOL to bone and to increase its bioavailability inextraskeletal tissue. We have demonstrated that, with this newliposome-based formulation, ZOL had increased antitumorproperties as compared to standard (free) ZOL, using both invitro and in vivo models of different human cancers.27 It hasbeen well demonstrated that stealth liposomes can be used todeliver drugs into the CNS in pathological states in which BBBpermeability is altered.28 In the case of chronic neuropathicpain, BBB is partially or totally disrupted and could allow thepassage of nanovectors such as LipoZOL.In light of these considerations, we have investigated the

antinociceptive effect of ZOL, administered as free or asLipoZOL, in an animal model of neuropathic pain. Moreover,levels of proinflammatory cytokines and CNS accumulation ofLipoZOL were evaluated.

■ MATERIALS AND METHODSMaterials. Phosphatidylcholine from egg yolk (EPC) and

1,2-diacyl-sn-glycero-3-phosphoethanolamine-N-[methoxy-(polyethylene glycol)-2000] (DSPE-PEG2000) were a kind giftfrom Lipoid GmbH (Cam, Switzerland). 22-[N-[(7-Nitro-2-1,3-benzoxadiazol-4-yl)methyl]amino]-27-norcholesterol(NBD cholesterol) was obtained by Invitrogen (Paisley, UnitedKingdom). Tetrabutylammonium bromide (TBA), cholesterol(Chol), ammonium chloride, ammonium tiocyanate, potassiumphosphate dibasic, sodium phosphate dibasic, iron(III) chlorideanhydrous, and Sephadex G-150 were purchased from SigmaChemical Co. (St. Louis, MO). Lactose was obtained from NewFa.Dem (Naples, Italy). Analytical grade diethyl ether,methanol, chloroform, and 30% ammonia solution, as well asHPLC grade acetonitrile, were obtained from Carlo Erba(Milan, Italy). ZOL was a kind gift from Novartis (Novartis,Basel, Switzerland).

Liposome Preparation. The liposomes were prepared by amodified reverse-phase evaporation technique as previouslydescribed.27 Briefly, an organic solution consisting of EPC/Chol/DSPE-PEG 2000 (1:0.32:0.30 weight ratio) in chloro-form/methanol (2:1 volume ratio) was placed in a round-bottom flask under nitrogen atmosphere, and the solvent wasremoved under vacuum in a rotary evaporator. Three millilitersof diethyl ether was then added to the lipid film, and theresulting solution was sonicated (bath-type sonicator, Branson3510, Danbury, United States) for 30 min in presence of 1 mLof ammonium chloride buffer at pH 9.5 containing 75 mMZOL and 58 mM lactose. Glass beads (Sigma) were also addedto the flask to facilitate the formation of an emulsion. Then, theorganic solvent was removed under vacuum at 30 °C by arotary evaporator (Laborota 4010 digital, Heidolph, Schwabach,Germany) in nitrogen atmosphere. Once a viscous gel wasobtained, the vacuum was broken, and the gel was agitated onvortex for about 1 min. The resulting dispersion was placedagain in the rotary evaporator for about 15 min under vacuum.The liposome suspension was repeatedly passed throughpolycarbonate membrane (Nucleopore Track Membrane 25mm, Whatman, Brentford, United Kingdom) with 0.4 μm poresize under nitrogen by using a thermobarrel extruder system(Northern Lipids Inc., Vancouver, BC, Canada). Then,unencapsulated ZOL was removed by passing the suspensionthrough Sephadex G-150 column where the liposomes wereeluted in an aqueous solution containing 58 mM lactose. After

Molecular Pharmaceutics Article

dx.doi.org/10.1021/mp3006215 | Mol. Pharmaceutics 2013, 10, 1111−11181112

preparation, the liposome suspension was quickly frozen inliquid nitrogen and lyophilized for 24 h. Blank liposomes wereprepared similarly. For FACS analysis, liposomes containingNBD cholesterol in a 1.7% weight ratio with respect to totalChol were prepared. All liposome preparations were stored at−20 °C. Each formulation was prepared in triplicate.Liposome Size. The liposome mean diameter and size

distribution were measured by photon correlation spectroscopy(PCS) (N5, Beckman Coulter, Miami, United States) at 20 °C.Briefly, liposomes were diluted in deionizer/filtered water, andthe measures were carried out with the detector set at a 90°angle. The particle size distribution was expressed aspolydispersity index (PI). For each batch, the results were themean of three measures. For each formulation, the meandiameter (reported in nm) and PI were calculated as the meanof three different batches.ZOL Encapsulation into Liposomes. ZOL loading into

liposomes was expressed as ZOL actual loading andencapsulation efficiency. ZOL actual loading was calculated asμg of ZOL/mg of lipids in the freeze-dried powder; ZOLencapsulation efficiency was obtained as the ratio between ZOLloaded into liposomes (LipoZOL) and ZOL theoreticalencapsulation. The phospholipid content of the liposomesuspension was determined by the Stewart’s assay.29 Briefly,liposomes were added to an aqueous ammonium ferrithiocya-nate solution (0.1 N) mixed with chloroform. The concen-tration of the phospholipids, namely, PC and DSPE-PEG, wascalculated by measuring the absorbance at a wavelength of 485nm into the organic layer. The quantitative analysis of ZOL wascarried out by reverse-phase chromatography (RP-HPLC) on aGemini 5 μm C18 column (250 mm × 4.60 mm, 110 Å,Phenomenex, Klwid, United States) coupled with a securityguard. ZOL was eluted in isocratic conditions (flow rate, 1 mL/min) with a mixture 20:80 (v/v) of acetonitrile/aqueoussolution (8 mM dipotassium hydrogen orthophosphate, 2 mMdisodium hydrogen orthophosphate, and 7 mM tetra-n-butylammonium hydrogen sulfate, adjusted to a pH of 7.0 withsodium hydroxide) at room temperature. The analysis wascarried out with an isocratic pump (LC-10A VP, Shimadzu,Kyoto, Japan) equipped with a 7725i injection valve(Rheodyne, Cotati, United States), SPV-10A UV−Vis detector(Shimadzu) set at a wavelength of 220 nm. Acquisition andanalyses of the chromatograms were carried out by a Class VPClient/Server 7.2.1 program (Shimadzu). ZOL dosage wascarried out as follows. Briefly, 100 μL of liposome suspensionwas mixed with 400 μL of water and 500 μL of chloroform.After mixing on vortex, the emulsion was centrifuged at 55g for10 min, and the surnatant was analyzed by RP-HPLC.Animals. Male CD1 mice (35−40 g) were positioned three

per cage under room temperature (20−22 °C) and humidity(55−60%) and under controlled illumination (12:12 hlight:dark cycle; light on 06.00 h) for at least 1 week beforeto begin the experimental procedures. Food and water wereprovided ad libitum. The experiments received the approval bythe Animal Ethics Committee of the Second University ofNaples. Animal care was carried out according to the IASP andEuropean Community (E.C. L358/1 18/12/86) guidelines forthe use and protection of animals in experimental research.Animal suffering and the number of mice used were reduced aspossible.Spared Nerve Injury (SNI). Mononeuropathy was induced

according to a previously described method of Decostered andWoolf.30 Mice were anaesthetized with sodium pentobarbital

(50 mg/kg, ip). The sciatic nerve was exposed. The tibial andcommon peroneal nerves were tightly ligated with 5.0 silkthread, leaving the sural nerve spared. Sham mice wereanaesthetized, and the sciatic nerve was exposed at the samelevel but not ligated. Mechanical allodynia was measured byusing the Dynamic Plantar Aesthesiometer (Ugo Basile, Varese,Italy). Mice were allowed to move freely in one of the twocompartments of the enclosure positioned on the metal gridsurface. A mechanical stimulus was delivered to the plantarsurface of the mouse hind paw by an automated steel filamentexerting an increasing force of 3 g per second. Nociceptiveresponses for mechanical sensitivity were measured in grams.Baseline thresholds were determined 6 days before starting thetreatments. Each mouse served as its own control, and theresponses were measured both before and after surgicalprocedures. The observer was blind to the treatments.

Immunohistochemistry. Under deep pentobarbital anes-thesia, mice were transcardially perfused with saline solutionfollowed by 4% paraformaldehyde (PFA) in 0.1% PBS. Thelumbar spinal cord was dissected, postfixed for 4 h in 4% PFA,cryoprotected for 72 h in 20% sucrose in 0.1 M phosphatebuffer, and frozen in O.C.T. embedding compound. Transversesections (20 μm) were cut by using a cryostat and thenmounted onto slides. Slides were incubated overnight withprimary antibody solutions for the glial cell marker rabbit polyclonal anti-GFAP (1:1000; Dako Cytomation, Denmark) orIL10 (goat anti- IL10 Santa Cruz, United States). Followingincubation, sections were washed and incubated for 3 h withsecondary antibody solution (donkey antirabbit or donkeyantigoat IgG-conjugated Alexa FluorTM 488 and 568; 1:500;Molecular Probes, United States). Slides were washed,coverslipped with Vectashield medium (Vector Laboratories,United States), and analyzed under a Leica fluorescencemicroscope.

Cell Preparation from Solid Tissues for FACS. Tissueswere excised for either 30 s or 1, 3, 6, or 24 h after injection andminced into 2−4 mm pieces using scissors or scalpel blade.Tissue pieces were trypsinized and incubated at 37 °C for 15min. Cells were dispersed by gentle pipetting and filteredthrough a cell strainer to eliminate clumps and debris. The cellsuspension was collected in a conical tube and centrifuged for4−5 min (300−400g) at 4 °C, discarding the supernatant.The cell pellet was washed in PBS to remove excess enzyme

solution and then resuspended in the same buffer to perform aFACS analysis (FACScan, Becton Dickinson). For each sample,2 × 104 events were acquired. Analysis was carried out bytriplicate determination on at least three separate experiments.CellQuest software (Becton Dickinson) was used to calculatemean fluorescence intensities (MFIs). The MFIs werecalculated by the formula (MFItreated/MFIcontrol), whereMFItreated is the fluorescence intensity of cells treated withLiposomes and MFIcontrol is the fluorescence intensity ofuntreated and unstained cells.

■ RESULTSLiposome Characteristics. Liposomes containing zole-

dronic acid (LipoZOL) had a mean diameter of about 241 ±36. ZOL was successfully encapsulated into liposomes at anactual loading of 101.41 ± 38.4, with about 70% of ZOL stillloaded into liposomes after freeze drying and followingreconstitution in water, as previously reported.27

LipoZOL Reduced Thermal Hyperalgesia and Me-chanical Allodynia in Mice with SNI. SNI was associated



with the development of ipsilateral mechanical and thermalhypersensitivity that was assessed up to 7 days after peripheralnerve injury (Figure 1). Both contralateral and sham-operated

thresholds remained unaltered (data not shown). Two ivadministrations (10 μg of ZOL encapsulated into liposomes) atdays 2 and 4 after the injury markedly reduced mechanical

hypersensitivity at 3 and 7 days after nerve injury (Figure 1).On the other hand, free ZOL (10 μg/dose) did not exert anysignificant alteration of the mechanical threshold (Figure 1).

Immunohistochemistry. In the present study, we haveevaluated the glial components associated with the establish-ment of spinal sensitization occurring in neuropathic pain.GFAP-labeled astrocytes appeared as hypertrophic-activatedcells in the ispilateral dorsal horn of the spinal cord 7 days afterSNI. Two LipoZOL administrations significantly changed theastrocyte morphology, by inducing a protective phenotype,without changing the total cell number (Figure 2). Moreover,double labeling revealed in the spinal cord of lipoZOL-treatedmice positive profiles for IL-10, which was expressed by GFAP-labeled astrocytes (Figure 2).

Liposome Biodistribution. To study the in vivobiodistribution of liposomes, we performed FACS analysis ofdifferent tissues collected at different times from mice injectedwith fluorescently labeled liposomes. Mice were randomizedinto two groups, neuropathic or not, and iv administered with asingle dose of fluorescently labeled LipoZOL. In detail, weanalyzed the LipoZOL content in cells from different solidtissues (liver, kidney, brain, spinal cord, lung, and ganglia) at 30s and 1, 3, 6, and 24 h after injection. Unlabeled liposomesdetermined a not significant fluorescence that was absolutelyoverlapping that one induced by free ZOL. On this light, wehave assumed unlabeled liposomes as negative control.

Figure 1. Effect of two intravenous treatments with 0.9% NaCl, whiteliposomes, ZOL, or lipozol (20 μg) on mechanical allodynia in shamand SNI mice. Data are reported as means ± SEMs from six to eightmice per group. * indicates a significant difference (P < 0.05) vs SNI/veh. Data were analyzed by two-way ANOVA, followed by StudentNeuman−Keuls’ posthoc test.

Figure 2. SNI induces an increase in the number of hypertrophyc astrocytes in the ipsilateral dorsal horn of the spinal cord as compared to thecontralateral side (A, upper panel). LipoZOL treatment significantly reduces the number of hypertrophyc astrocytes in the ipsilateral side of thedorsal horn (A, lower panel). B represents LipoZOL-induced overexpression of IL-10 in the GFAP-labeled astrocytes. C represents quantitativeanalysis of GFAP-positive profiles. Data are expressed as the mean ± SEM. ANOVA Tukey test. Scale bar, 100 μm.



As expected, FACS analysis revealed an increased LipoZOLuptake in liver and kidney of both mice groups, neuropathic ornot. In detail, after 30 min, LipoZOL uptake was about 210 and200% increased as compared to negative control in liver andkidney of both neuropathic and not neuropathic mice,respectively (Figure 3A,B). LipoZOL accumulation in theliver of both animal groups remained almost unchanged for allof the observation period, while liposome accumulation inkidney decreased in a time-dependent manner, reaching about70−80% increase as compared to negative controls (Figure3A,B). On the other hand, there were not significant changes inthe LipoZOL content as compared to negative controls in thelungs of both animal groups (Figure 3C). The accumulation ofthe fluorescence associated to LipoZOL in the brain of normalanimals was not significantly increased as compared to negativecontrols for all of the time of the observation. On the otherhand, the increase of fluorescently labeled LipoZOL was about100 and 150% increased in the brain of neuropathic mice at 30min and 1 h after injection, respectively (Figure 4A).Thereafter, the fluorescence associated with LipoZOLdecreased in a time-dependent manner reaching an about35% increase at 24 h from the injection (Figure 4A). In the case

of spinal cord, the increase of fluorescence associated withLipoZOL in neuropathic animals occurred at a later time pointand at a lesser extent than that one recorded in brain: in fact, anabout 190% increase of the fluorescence was recorded only at 1h from the injection (Figure 4B). Similarly to the brain, thefluorescence in spinal cord of neuropathic animals decreased ina time-dependent manner becoming not significantly alteredafter 24 h from the injection (Figure 4B). Once again,fluorescence was not significantly increased in the spinal cord ofnormal mice (Figure 4B). The fluorescence associated withLipoZOL was evaluated also in ganglia of normal andneuropathic animals, and the results indicated no specificuptake of the liposomes in this specific tissue at both 6 and 24 hfrom the injection (Figure 4C). In the latter case, thefluorescence was evaluated only at two time points due tothe poorness of the biological material associated with theganglia of the mice.In conclusion, these results suggested a significant accumu-

lation of the ZOL-containing liposomes in brains and spinalcords of neuropathic mice while their accumulation did notoccur at all in the same tissues of normal mice.

Figure 3. Right panels: NBD-cholesterol-labeled LipoZOL fluorescence determined in cells from liver (A), kidney (B), and lung (C) of neuropathic(SNI) and not neuropathic (CTR) animals expressed as a % of increase of the mean fluorescence intensity (MFI) as compared to that one inducedby nonfluorescent liposomes (blank liposomes). FACS analysis of the LipoZOL content at 30 s and 1, 3, 6, and 24 h after injection of a single dose ofLipoZOL. The MFIs were calculated, as described in the Materials and Methods. Values are the means of three independent experiments (±SD).Left panels: Representative histograms relative to fluorescence associated to NBD-cholesterol-labeled LipoZOL in cells derived from liver (A),kidney (B), and lung (C) of neuropathic (SNI) and not neuropathic (CTR) animals. Full blue histograms represent the fluorescence of the negativecontrols (nonfluorescent or blank liposomes).



■ DISCUSSION

In this work, a new approach for the treatment of theneuropathic pain was proposed and investigated. In particular,our hypothesis was that ZOL, an aminobisphosphonate used inthe clinical setting to prevent skeletal-related events in bonemetastasis, osteoporosis, or Paget’s disease, could be a new andpowerful pharmacological agent to control neuropathic pain.However, following iv administration, ZOL rapidly accumulatesinto the bone (about 55% of the administered dose), with verylow concentrations in extraskeletal tissues.23−25 In our previousworks, we demonstrated that the use of nanovectors allows theescape from bone accumulation, thus increasing the ZOLconcentration in nonskeletal tissues, that is, in extrabone sitesof different kinds of human tumors.27,31,32 In fact, we havepreviously demonstrated efficient inhibition of tumor growth indifferent experimental cancer models, that is, prostate cancerand multiple myeloma, only when using stealth nanovectorsencapsulating ZOL, while this effect was negligible with freeZOL. These data supported our hypothesis that the use ofstealth nanovectors can change ZOL pharmacokinetics.

In healthy organisms, long circulating liposomes are not ableto across the BBB.33 On the other hand, stealth nanovectors,such as PEGylated liposomes, can be efficiently used to deliverdrug into the CNS, in the case of diseases characterized by analtered BBB.29 This has been shown in the case of anexperimental model of tumor,34 multiple sclerosis,35,36 brainischemia,37 and metastases.38

Taking into account the alterations of BBB found inexperimental models of neuropathic pain,39 we hypothesizedthat stealth liposomes could allow the delivery of ZOL intoCNS of SNI animal, to use ZOL as a modulator of neuropathicpain. FACS analysis has been recently proposed by differentauthors to follow the in vivo biodistribution of drug deliverysystems, when radiolabeled drugs are not available.40−42

Intriguingly, through FACS analysis approach, we foundsignificant fluorescence associated with liposomes encapsulatingZOL in the spinal cord in the case of SNI animals. On the otherhand, fluorescence was not found in the case of healthy animals.Therefore, the second step was to confirm the hypothesis ofZOL delivery into the CNS by measuring the effect ofLipoZOL on the mechanical allodynia. We found a significantreduction of mechanical hypersensitivity at 3 and 7 days after

Figure 4. Right panels: NBD-cholesterol-labeled LipoZOL fluorescence determined in cells from brain (A), spinal cord (B), and ganglia (C) ofneuropathic (SNI) and not neuropathic (CTR) animals expressed as a % of increase of the mean fluorescence intensity (MFI) as compared to thatone induced by nonfluorescent liposomes (blank liposomes). FACS analysis of the LipoZOL content at 30 s and 1, 3, 6, and 24 h after injection of asingle dose of LipoZOL. The MFIs were calculated, as described in the Materials and Methods. Values are the means of three independentexperiments (±SD). Left panels: Representative histograms relative to fluorescence associated to NBD-cholesterol-labeled LipoZOL in cells derivedfrom brain (A), spinal cord (B), and ganglia (C) of neuropathic (SNI) and not neuropathic (CTR) animals. Full blue histograms represent thefluorescence of the negative controls (nonfluorescent or blank liposomes).



nerve injury, while no effect was found in the case of free ZOL,administered at the same dose, that is, 10 μg of ZOL at days 2and 4 after the injury. Moreover, it has been recently reportedthat cortical plasticity is mandatory for maintaining tactileallodynia.43 Therefore, the balancing of the non-neuronal cellphenotypes in this pivotal area of the CNS is important forneuropathic pain management. Peripheral nerve injury is alsoassociated with BBB alterations39 due to an overall inflamma-tory process that involves several cytotypes. Interestingly, inthis study, the analgesic effect of LipoZOL occurred togetherwith the restoration of normal glial architecture of the dorsalhorn of the spinal cord, while free ZOL was not able to induceany restoring effect. These effects on the astrocyte shape towarda protective phenotype were not associated to changes in theirtotal number. The phenotypical shift of astrocytes is alsoconsistent with the increased expression of the anti-inflammatory cytokine IL-10 induced by the LipoZOLtreatment. In fact, in healthy SNI mice treated with vehicle aswell as SNI animals treated with free ZOL, the IL-10 wasundetectable immunohistochemically (not shown).Taken together, our results confirm that sciatic nerve

damage-induced BBB alterations could promote the infiltrationof the LipoZOL in the dorsal horn of spinal cord, thus leadingto the right ZOL concentrations in the CNS able to modulatethe phenotypical shift of glial cells and to abrogate neuropaticpain. Our study opens new perspectives for the treatment ofneuropathic pain, providing a new and powerful approachbased on the combined use of a potent aminobisphosphonateand stealth liposomes.

■ AUTHOR INFORMATIONCorresponding Author*Tel: +39(0)81 678 666. Fax: +39(0)81 678 630. E-mail:[email protected].

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSM.C. received a contribution from the Italian Ministry ofEducation and Research (MIUR, PRIN 2009), from the ItalianAssociation for Cancer Research (AIRC) for a project entitled“Liposomes encapsulating zoledronic acid: a new experimentaltherapeutic for the treatment of brain tumors”, and fromRegione Campania for “Laboratori Pubblici” Hauteville. Thiswork is dedicated to the memory of the beloved Prof. AlbertoAbbruzzese.

■ REFERENCES(1) Costigan, M.; Scholz, J.; Woolf, C. J. Neuropathic pain: Amaladaptive response of the nervous system to damage. Annu. Rev.Neurosci. 2009, 32, 1−32.(2) Ji, R. R.; Kohno, T.; Moore, K. A.; Woolf, C. J. Centralsensitization and LTP: do pain and memory share similarmechanisms? Trends Neurosci. 2003, 26, 696−705.(3) Basbaum, A. I.; Bautista, D. M. G.; Scherrer; Julius, D. Cellularand molecular mechanisms of pain. Cell 2009, 139, 267−284.(4) Aldskogius, H.; Kozlova, E. N. Central neuron-glial and glial-glialinteractions following axon injury. Prog. Neurobiol. 1998, 55, 1−26.(5) Nimmerjahn, A.; Kirchhoff, F.; Helmchen, F. Resting microglialcells are highly dynamic surveillants of brain parenchyma in vivo.Science 2005, 308, 1314−1318.(6) Rossi, D. J.; Brady, J. D.; Mohr, C. Astrocyte metabolism andsignaling during brain ischemia. Nat. Neurosci. 2007, 10, 1377−1386.

(7) Hanisch, U. K.; Kettenmann, H. Microglia: Active sensor andversatile effector cells in the normal and pathologic brain. Nat.Neurosci. 2007, 10, 1387−1394.(8) Scholz, J.; Woolf, C. J. The neuropathic pain triad: neurons,immune cells and glia. Nat. Neurosci. 2007, 10, 1361−1368.(9) Clark, A. K.; Yip, P. K.; Grist, J.; Gentry, C.; Staniland, A. A.;Marchand, F.; Dehvari, M.; Wotherspoon, G.; Winter, J.; Ullah, J.;Bevan, S.; Malcangio, M. Inhibition of spinal microglial cathepsin S forthe reversal of neuropathic pain. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,10655−10660.(10) Luongo, L.; Palazzo, E.; Tambaro, S.; Giordano, C.; Gatta, L.;Scafuro, M. A.; Rossi, F. S.; Lazzari, P.; Pani, L.; de Novellis, V.;Malcangio, M.; Maione, S. 1-(2′,4′-Dichlorophenyl)-6-methyl-N-cyclohexylamine-1,4-dihydroindeno[1,2-c]pyrazole-3-carboxamide, anovel CB2 agonist, alleviates neuropathic pain through functionalmicroglial changes in mice. Neurobiol. Dis. 2010, 37, 177−185.(11) Luongo, L.; Sajic, M.; Grist, J.; Clark, A. K.; Maione, S.;Malcangio, M. Spinal changes associated with mechanical hyper-sensitivity in a model of Guillain-Barre syndrome. Neurosci. Lett. 2008,437, 98−102.(12) Hains, B. C.; Waxman, S. G. Activated microglia contribute tothe maintenance of chronic pain after spinal cord injury. J. Neurosci.2006, 26, 4308−4317.(13) Liu, L.; Cash, T. P.; Jones, R. G.; Keith, B.; Thompson, C. B.;Simon, M. C. Hypoxia-induced energy stress regulates mRNAtranslation and cell growth. Mol. Cell 2006, 21, 521−531.(14) Ji, R. R.; Gereau, R. W.; Malcangio, M.; Strichartz, G. R. MAPkinase and pain. Brain Res. Rev. 2009, 60, 135−148.(15) Zhuang, Z. Y.; Wen, Y. R.; Zhang, D. R.; Borsello, T.; Bonny,C.; Strichartz, G. R.; Decosterd, I.; Ji, R. R. A peptide c-Jun N-terminalkinase (JNK) inhibitor blocks mechanical allodynia after spinal nerveligation: respective roles of JNK activation in primary sensory neuronsand spinal astrocytes for neuropathic pain development andmaintenance. J. Neurosci. 2006, 26, 3551−3560.(16) Jin, S. X.; Zhuang, Z. Y.; Woolf, C. J.; Ji, R. R. p38 mitogen-activated protein kinase is activated after a spinal nerve ligation inspinal cord microglia and dorsal root ganglion neurons and contributesto the generation of neuropathic pain. J. Neurosci. 2003, 23, 4017−4022.(17) Tsuda, M.; Mizokoshi, A.; Shigemoto-Mogami, Y.; Koizumi, S.;Inoue, K. Activation of p38 mitogen- activated protein kinase in spinalhyperactive microglia contributes to pain hypersensitivity followingperipheral nerve injury. Glia 2004, 45, 89−95.(18) Ma, W.; Quirion, R. Partial sciatic nerve ligation inducesincrease in the phosphorylation of extracellular signal-regulated kinase(ERK) and c-Jun N-terminal kinase (JNK) in astrocytes in the lumbarspinal dorsal horn and the gracile nucleus. Pain 2002, 99, 175−184.(19) Zhuang, Z. Y.; Gerner, P.; Woolf, C. J.; Ji, R. R. ERK issequentially activated in neurons, microglia, and astrocytes by spinalnerve ligation and contributes to mechanical allodynia in thisneuropathic pain model. Pain 2005, 114, 149−159.(20) Caraglia, M.; Budillon, A.; Tagliaferri, P.; Marra, M.;Abbruzzese, A.; Caponigro, F. Isoprenylation of intracellular proteinsas a new target for the therapy of human neoplasms: preclinical andclinical implications. Curr. Drug Targets 2005, 6, 301−23.(21) Hancock, J. F.; Magee, A. I.; Childs, J. E.; Marshall, C. J. All rasproteins are polyisoprenylated but only some are palmitoylated. Cell1989, 57, 1167−1177.(22) Aznar, S.; Lacal, J. C. Rho signals to cell growth and apoptosis.Cancer Lett. 2001, 165, 1−10.(23) Chen, T. Pharmacokinetics and pharmacodynamics ofzoledronic acid in cancer patients with bone metastases. J. Clin.Pharmacol. 2002, 42, 1228−1236.(24) Caraglia, M.; Santini, D.; Marra, M.; Vincenzi, B.; Tonini, G.;Budillon, A. Emerging anti-cancer molecular mechanisms of amino-bisphosphonates. Endocr. Relat. Cancer 2006, 13, 7−26.(25) Caraglia, M.; Marra, M.; Naviglio, S.; Botti, G.; Addeo, R.;Abbruzzese, A. Zoledronic acid: An unending tale for an antiresorptiveagent. Expert Opin. Pharmacother. 2010, 11, 141−154.



mailto:[email protected]

(26) Weiss, H. M.; Pfaar, U.; Schweitzer, A.; Wiegand, H.; Skerjanec,A.; Schran, H. Biodistribution and plasma protein binding ofzoledronic acid. Drug Metab. Dispos. 2008, 36, 2043−2049.(27) Marra, M.; Salzano, G.; Leonetti, C.; Tassone, P.; Scarsella, M.;Zappavigna, S.; Calimeri, T.; Franco, R.; Liguori, G.; Cigliana, G.;Ascani, R.; La Rotonda, M. I.; Abbruzzese Saccardi, A.; Tagliaferri, P.;Caraglia, M.; De Rosa, G. Nanotechnologies to use bisphosphonates aspotent anticancer agents: the effects of zoledronic acid encapsulatedinto liposomes. Nanomedicine 2011, 7, 955−964.(28) De Rosa, G.; Salzano, G.; Caraglia, M.; Abbruzzese Saccardi, A.Nanotechnologies: a strategy to overcome blood-brain barrier. Curr.Drug Metab. 2012, 13, 61−69.(29) Stewart, J. C. M. Colorimetric determination of phospholipidswith ammonium ferrothiocyanate. Anal. Biochem. 1980, 104, 10−14.(30) Decostered, I.; Woolf, C. J. Spared nerve injury: An animalmodel of persistent peripheral neuropathic pain. Pain 2000, 87 (2),149−158.(31) Salzano, G.; Marra, M.; Porru, M.; Zappavigna, S.; AbbruzzeseSaccardi, A.; La Rotonda, M. I.; Leonetti, C.; Caraglia, M.; De Rosa, G.Self-assembly nanoparticles for the delivery of bisphosphonates intotumors. Int. J. Pharm. 2011, 403, 292−297.(32) Marra, M.; Salzano, G.; Leonetti, C.; Porru, M.; Franco, R.;Zappavigna, S.; Liguori, G.; Botti, G.; Chieffi, P.; Lamberti, M.; Vitale,G.; Abbruzzese Saccardi, A.; La Rotonda, M. I.; De Rosa, G.; Caraglia,M. New self-assembly nanoparticles and stealth liposomes for thedelivery of zoledronic acid: A comparative study. Biotechnol. Adv. 2012,30, 302−309.(33) Huwyler, J.; Wu, D.; Pardridge, W. M. Brain drug delivery ofsmall molecules using immunoliposomes. Proc. Natl. Acad. Sci. U.S.A.1996, 93, 14164−14169.(34) Siegal, T.; Horowitz, A.; Gabizon, A. Doxorubicin encapsulatedin sterically stabilized liposomes for the treatment of a brain tumormodel: biodistribution and therapeutic efficacy. J. Neurosurg. 1995, 83,1029−1037.(35) Schmidt, J.; Metselaar, J. M.; Wauben, M. H.; Toyka, K. V.;Storm, G.; Gold, R. Drug targeting by long-circulating liposomalglucocorticosteroids increases therapeutic efficacy in a model ofmultiple sclerosis. Brain 2003, 126, 1895−1904.(36) Kizelsztein, P.; Ovadia, H.; Garbuzenko, O.; Sigal, A.; Barenholz,Y. Pegylated nanoliposomes remote-loaded with the antioxidanttempamine ameliorate experimental autoimmune encephalomyelitis.J. Neuroimmunol. 2009, 213, 20−25.(37) Fukumoto, D.; Kawaguchi, A. T.; Haida, M.; Yamano, M.;Ogata, Y.; Tsukada, H. Liposome-encapsulated hemoglobin reducesthe size of cerebral infarction in rats: Effect of oxygen affinity. Artif.Organs 2009, 33, 159−163.(38) Caraglia, M.; Addeo, R.; Costanzo, R.; Montella, L.; Faiola, V.;Marra, M.; Abbruzzese Saccardi, A.; Palmieri, G.; Budillon, A.;Grillone, F.; Venuta, S.; Tagliaferri, P.; Del Prete, S. Phase II study ofTemozolomide plus pegylated liposomal doxorubicin in the treatmentof brain metastases from solid tumours. Cancer Chemother. Pharmacol.2006, 57, 34−39.(39) Echeverry, S.; Shi, X. Q.; Rivest, S.; Zhang, J. Peripheral nerveinjury alters blood-spinal cord barrier functional and molecularintegrity through a selective inflammatory pathway. J. Neurosci. 2011,31, 10819−10828.(40) Hagtvet, E.; Evjen, T. J.; Nilssen, E. A.; Olsen, D. R. Assessmentof liposome biodistribution by non-invasive optical imaging: Afeasibility study in tumour-bearing mice. J. Nanosci. Nanotechnol.2012, 12, 2912−2918.(41) Hinton, T. M.; Monaghan, P.; Green, D.; Kooijmans, S. A.; Shi,S.; Breheney, K.; Tizard, M.; Nicolazzo, J. A.; Zelikin, A. N.; Wark, K.Biodistribution of polymer hydrogel capsules for the delivery oftherapeutics. Acta Biomater. 2012, 8, 3251−3260.(42) Merian, J.; Gravier, J.; Navarro, F.; Texier, I. Fluorescentnanoprobes dedicated to in vivo imaging: from preclinical validationsto clinical translation. Molecules 2012, 17, 5564−5591.(43) Giordano, C.; Siniscalco, D.; Melisi, D.; Luongo, L.; Curcio, A.;Soukupova, M.; Palazzo, E.; Marabese, I.; De Chiaro, M.; Rimoli, M.

G.; Rossi, F.; Maione, S.; de Novellis, V. The galactosylation of N(ω)-nitro-L-arginine enhances its anti-nocifensive or anti-allodynic effectsby targeting glia in healthy and neuropathic mice. Eur. J. Pharmacol.2011, 656, 52−62.



stealth liposomes encapsulating zoledronic acid: a new opportunity to treat neuropathic pain

Documents