Targeted delivery of glucocorticoids to macrophages in a mouse model of multiple sclerosis using inorganic-organic hybrid nanoparticles

    Targeted delivery of glucocorticoids to macrophages in a mouse model of multiple sclerosis using inorganic-organic hybrid nanoparticles Elena Montes-Cobos, Sarah Ring, Henrike J. Fischer, Joachim Heck, Judith Strauß, Markus Schwaninger, Sybille D. Reichardt, Claus Feldmann, Fred L¨uhder, Holger M. Reichardt PII: DOI: Reference:

S0168-3659(16)31253-6 doi:10.1016/j.jconrel.2016.12.003 COREL 8559

To appear in:

Journal of Controlled Release

Received date: Revised date: Accepted date:

20 July 2016 23 November 2016 1 December 2016

Please cite this article as: Elena Montes-Cobos, Sarah Ring, Henrike J. Fischer, Joachim Heck, Judith Strauß, Markus Schwaninger, Sybille D. Reichardt, Claus Feldmann, Fred L¨ uhder, Holger M. Reichardt, Targeted delivery of glucocorticoids to macrophages in a mouse model of multiple sclerosis using inorganic-organic hybrid nanoparticles, Journal of Controlled Release (2016), doi:10.1016/j.jconrel.2016.12.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Targeted delivery of glucocorticoids to macrophages in a mouse model of multiple sclerosis

SC RI

PT

using inorganic-organic hybrid nanoparticles

Elena Montes-Cobosa,b, Sarah Ringa,#, Henrike J. Fischera,b, Joachim Heckc, Judith Straußb, Markus Schwaningerd, Sybille D. Reichardta, Claus Feldmannc, Fred Lühderb,‡ and Holger M.

a

a,‡,

*

NU

Reichardt

Institute for Cellular and Molecular Immunology, University Medical Center Göttingen,

MA

37073 Göttingen, Germany b

Institute for Multiple Sclerosis Research and Neuroimmunology, University Medical Center Göttingen, 37073 Göttingen, Germany

Institute of Inorganic Chemistry, Karlsruhe Institute of Technology, 76131 Karlsruhe,

ED

c

Germany

Institute of Experimental and Clinical Pharmacology and Toxicology, University of Lübeck,

PT

d

present address: Priority Area Infections, Research Center Borstel, 23845 Borstel,

AC

#

CE

23562 Lübeck, Germany

Germany

‡

these authors contributed equally to the work

* corresponding author: Prof. Dr. Holger Reichardt, Institute for Cellular and Molecular Immunology, University Medical Center Göttingen, Humboldtallee 34, 37073 Göttingen, Germany; Email: [email protected]

-1-

ACCEPTED MANUSCRIPT Abstract Glucocorticoids (GC) are widely used to treat acute relapses in multiple sclerosis (MS) patients, but their application is accompanied by side effects due to their broad spectrum of

PT

action. Here, we report on the therapeutic option to apply GC via inorganic-organic hybrid nanoparticles (IOH-NP) with the composition [ZrO]2+[(BMP)0.9(FMN)0.1]2- (designated BMP-

SC RI

NP with BMP: betamethasone phosphate; FMN: flavinmononucleotide). We found that these BMP-NP have an increased cell type-specificity compared to free GC while retaining full therapeutic efficacy in a mouse model of MS. BMP-NP were preferentially taken up by phagocytic cells and modulated macrophages in vivo more efficiently than T cells. When GC were

NU

applied in the form of BMP-NP, treatment of neuroinflammatory disease in mice exclusively depended on the control of macrophage function whereas effects on T cells and brain endo-

MA

thelial cells were dispensable for therapeutic efficacy. Importantly, BMP-NP were not only active in mice but also showed strong activity towards monocytes isolated from healthy human volunteers. We conclude that application of GC via IOH-NP has the potential to improve

ED

MS therapy in the future.

PT

Keywords

CE

hybrid nanoparticles, glucocorticoids, experimental autoimmune encephalomyelitis,

AC

multiple sclerosis, T cells, macrophages

-2-

ACCEPTED MANUSCRIPT Introduction Chronic inflammatory diseases are a group of highly variable pathogenic conditions that may affect any organ or tissue. Their total prevalence in western countries is currently around 5%

PT

and still rising [1]. A common denominator of these diseases is their immune-mediated pathomechanism. Loss of self-tolerance or the response to harmless foreign antigens lead to

SC RI

leukocyte infiltration into affected organs, and causes a misbalance of circulating cytokines and the release of cytotoxic mediators. Despite the availability of highly effective biologicals [2], application of synthetic GC remains the mainstay in the treatment of many of such inflammatory diseases. They possess strong immunosuppressive activity and potently amelio-

NU

rate clinical symptoms by inhibiting innate and acquired immune responses [3]. GC achieve these effects by employing a variety of mechanisms such as inhibiting cytokines, redirecting

MA

leukocyte migration, inducing T cell apoptosis, and altering macrophage polarization [4]. While the therapeutic efficacy of GC is unsurpassed, their use is accompanied by various complications [5]. Some patients are resistant to GC treatment while others experience side

ED

effects such as opportunistic infections, muscle wasting, and diabetes. Hence, it would be desirable to improve currently available GC-based therapeutic regimens to overcome these problems.

PT

MS is a chronic neuroinflammatory disease of autoimmune origin that affects more than two million people worldwide [6, 7]. It is initiated by self-reactive T cells that recognize antigens

CE

present in the central nervous system (CNS), subsequently leading to the recruitment of other cells of the innate and adaptive immune response resulting in destruction of neuronal

AC

cells. The most common form is relapsing-remitting MS, which is characterized by acute disease bouts and intervening periods that are largely free of symptoms. The mononuclear cell infiltrate found in the CNS of affected patients is mainly composed of T cells and macrophages, which closely cooperate in the pathogenesis of MS. The main role of pathogenic T cells is disease initiation and exacerbation by recognition of antigen, production of cytokines and induction of oligodendrocyte apoptosis, processes with all of which GC are able to interfere [8, 9]. Macrophages can commit to different phenotypes and thereby fulfill distinct functions in MS depending on their polarization [10]. When exposed to GC, they assume an antiinflammatory phenotype, which is characterized by reduced pro-inflammatory cytokine secretion, an upregulation of scavenger receptors, intensified phagocytosis, and diminished antigen presentation [11]. These changes eventually enable macrophages to reduce inflammation, limit the adaptive immune response, initiate repair mechanisms, and contribute to the resolution of disease symptoms. Although there are a number of new drugs on the market that slow down disease progression of MS [12], GC administration has persisted as the gold standard with which to interfere

-3-

ACCEPTED MANUSCRIPT with acute relapses and optic neuritis for the last 30 years [13]. Nevertheless, GC have also been tested as an add-on to disease-modifying drugs in relapsing-remitting MS, and in the treatment of primary and secondary progressive forms of MS [14-16]. Besides clinical trials, studies in mice have made an important contribution towards our current understanding of

PT

this therapeutic regimen, in particular the analysis of experimental autoimmune encephalomyelitis (EAE), an animal model that mimics many pathophysiological features of MS [17].

SC RI

These studies revealed that T cells are the major target of conventional GC therapy, and respond by reduced cytokine production, apoptosis induction and altered migration [18]. In contrast, encapsulation of GC in PEGylated liposomes caused their redirection to the myeloid cell compartment, which induced polarization of macrophages and monocytes to an anti-

NU

inflammatory phenotype [19]. Modifying the delivery method of GC thus makes it possible to alter their mechanism of action and accordingly, application of liposomal GC in animal mod-

MA

els of autoimmune diseases provided favorable results [19-21]. However, liposomes are also known to cause adverse effects, including the activation of the complement system, the latter which may lead to life-threatening hypersensitivity reactions [22]. Since the outcome of human trials to treat MS with liposomal GC has not been reported in literature, the clinical suc-

ED

cess of this approach is difficult to estimate. Meanwhile, phospholipid and polymer-based nanostructured carrier systems have been developed as alternative delivery vehicles for GC,

PT

which show potent anti-inflammatory activity and reduced side effects in mouse models [2325]. Whether the new GC formulations will be suitable to treat neuroinflammatory disorders,

CE

however, remains to be determined. Recently, we reported on novel inorganic-organic hybrid nanoparticles (IOH-NP) that could be used for the treatment of inflammatory diseases [26]. Precipitation of inorganic cations

AC

such as [ZrO]2+ with the negatively charged organic molecules betamethasone (BMZ) phosphate [BMP]2- and flavinmononucleotide [FMN]2- in aqueous solution results in the formation of nanoparticles with the composition [ZrO]2+[(BMP)0.9(FMN)0.1]2- with a hydrodynamic diameter of 30-40 nm. These BMP-NP reduce the production of pro-inflammatory cytokines and thus are able to act as immunosuppressive drugs [26]. In addition, they can be tracked due to a fluorescent dye contained therein. It is against this background that we investigated the mechanism of GC delivered in the form of BMP-NP in the modulation of the two major cell types involved in MS, and to characterize the features of BMP-NP in vivo in the treatment of this disease by using EAE, a relevant mouse model of MS. Our results indicate that BMP-NP strongly increase the target cell-specificity of GC for macrophages without compromising therapeutic efficiency. Furthermore, they are capable of modulating functions of human monocytes, suggesting that they will also be biologically active in patients. These findings feed the hope that it might become possible to improve GC therapy using this novel drug formulation in the future.

-4-

ACCEPTED MANUSCRIPT Materials and methods Nanoparticle preparation, characterization and application

PT

Synthesis of [ZrO]2+[(BMP)0.9(FMN)0.1]2- IOH-NP was performed as previously described [26]. The BMP-NP were composed of 90 mol-% BMP and 10 mol-% FMN. Synthesis was carried out by admixing a solution of ZrOCl2  8H2O (5 mg) in H2O (2.5 ml) to a solution of sodium

SC RI

betamethasone 21-phosphate (Na2(BMP), 50 mg) and sodium riboflavin-5’-mono-phosphate dihydrate (Na(HFMN), 5.1 mg) in H2O (50 ml). After nucleation, the IOH-NP were purified by repeated centrifugation and redispersion from/in water. Finally, they were redispersed (2.8 mg/ml) into HEPES buffer (30 mM) at pH 7.4 with an effective concentration of 4.4 µM of

NU

the pharmacologically active drug. The stoichiometry of the nanoparticles was confirmed by energy-dispersive X-ray spectroscopy and elemental analysis as described previously [26].

MA

Dynamic light scattering (DLS) data and Zeta potentials of [ZrO]2+[(BMP)0.9(FMN)0.1]2- IOHNP freshly dispersed in PBS are depicted in Supplementary Figure S1A,B. Empty nanoparticles not containing any BMP (designated EP-NP) were obtained through replacing [BMP]2-

ED

against [HPO4]2- by using Na2(HPO4) as a starting material instead of Na2(BMP) and used as a reference in most experiments.

The chemical stability of the BMP-NP in PBS was determined based on the carbon content

PT

of the nanoparticles after 12 and 24 hours stirring at 37°C. Elemental analysis was performed after centrifugation and purification. Accordingly, the as-prepared BMP-NP exhibited an initial

CE

carbon content of 42 ± 2 wt-%, which matches very well the expectation (44 wt-%). After 12 and 24 hours, the carbon contents were 40 ± 2 and 38 ± 2 wt-%, respectively. These data

AC

point towards a slow release of BMZ and FMN via hydrolytic cleavage of the phosphate ester bond [26]. It is noteworthy that the colloidal stability did not show any relevant changes over time. In terms of the significance of the experiment, the particle size and size distribution of BMP-NP freshly dispersed in PBS (Supplementary Figure S1) was similar compared to BMPNP stirred for 24 hours in PBS (data not shown). In vitro, BMP-NP were added to the cell culture such that the BMZ contained in the nanoparticles corresponded to a concentration of 10-7 or 10-6 M. Equal volumes of EP-NP were used for each concentration of BMP-NP to ensure that the amount of ZrO in control cultures was identical. In vivo, the volume of BMP-NP was calculated such that the dosage of the BMZ contained in the nanopartilces corresponded to 10 mg of active drug per kg of body weight. Equal volumes of EP-NP were injected as controls to ensure that the amount of ZrO applied to each mouse was the same. All calculations were based on the mass content of the BMPNP as mentioned above.

-5-

ACCEPTED MANUSCRIPT Macrophage isolation and culture Bone marrow-derived macrophages (BMDM) were obtained by culturing single-cell suspensions of bone marrow obtained from femur and tibiae of C57BL/6 wildtype mice for seven days in the presence L929-conditioned medium (LCCM) as described previously [27]. After

PT

completion of macrophage differentiation, BMDM were harvested, resuspended in DMEM medium with 10% FCS, and incubated for 24 hours with BMP-NP or EP-NP as a control.

SC RI

Alternatively, BMDM were treated with DEX or its corresponding vehicle PBS. BMDM were analyzed by FACS, bright field microscopy, or quantitative RT-PCR.

Peritoneal macrophages (PMΦ) were elicited by injecting 1 ml 4% thioglycolate solution i.p. four days prior to isolation. For in vivo analysis of nanoparticle effects, mice were injected on

NU

three consecutive days with BMP-NP, EP-NP, DEX, or PBS. One day after the last treatment a peritoneal lavage was performed using PBS with 0.1% BSA. The cells were seeded in 10

MA

cm plates and after 1 hour of incubation at 37°C, adherent macrophages were collected in PBS with 2 mM EDTA and analyzed by FACS or bright field microscopy. Alternatively, the cells were cultured at a density of 2×105 cells/ml in DMEM medium with 10% FCS for another

ED

48 hours with 20 ng/ml LPS and 50 ng/ml IFN followed by the analysis of TNF levels in the supernatant by ELISA. In case not only macrophages but also T and B cells contained in the

Cell viability test

PT

peritoneal lavage should be analyzed, the adherence step was omitted.

CE

1×105 non-adherent bone marrow cells were seeded in 96-well flat-bottom plates in a volume of 100 µl and differentiated into BMDM for seven days in the presence LCCM. After replacing the medium by DMEM with 10% FCS, BMP-NP were added at a concentration equivalent to

AC

10-7 M BMZ followed by 24 hours of incubation. For comparison, BMDM were cultured in the presence of the same concentration of DEX, equal volumes of EP-NP or PBS, or after adding 10% DMSO to the medium as a negative control. Cell viability was assessed by using an MTT cell proliferation assay according to the manufacturer’s instructions (Promega, Mannheim, Germany), which is based on the formation of a blue dye that can be measured by spectrophotometry. Cellular uptake of nanoparticles Splenocytes or total peritoneal cells were seeded at a concentration of 1×105 cells per 100 µl in 96-well plates and incubated with BMP-NP containing 10-6 M BMZ. After 6, 24 or 48 hours, splenocytes and peritoneal cells were collected. The uptake of BMP-NP into different cell types (CD3+, B220+ and CD11b+ cells) was determined by FACS analysis via quantification of the mean fluorescence intensity (MFI) of the FMN measured in the green channel after gating on each of the three subpopulations. Alternatively, 1×105 BMDM or PMΦ were seed-

-6-

ACCEPTED MANUSCRIPT ed in cover glass bottom 1.0 Imaging dishes (zell-kontakt, Nörten Hardenberg, Germany), cultured in DMEM medium with 10% FCS, and treated with BMP-NP containing 10-6 M BMZ for 24 hours. Afterwards, the macrophages were washed with PBS and the uptake of the BMP-NP was visualized by 2-photon microscopy using a Laser Scanning Microscope 710

PT

(Zeiss, Jena, Germany) or with the help of a confocal laser scanning microscope TCS SP2

Fluorescence-activated cell sorting (FACS)

SC RI

(Leica, Wetzlar, Germany).

Splenocytes, BMDM or peritoneal cells were stained for extracellular antigens as described previously [28]. Analysis by six-color flow cytometry was performed using a FACS Canto II

NU

device (BD Biosciences, Heidelberg, Germany) in combination with FlowJo software (Tree Star, Ashland, OR). All antibodies and reagents were directly labelled with different fluoro-

MA

phores and obtained from BioLegend (Uithoorn, The Netherlands) unless otherwise indicated: anti-mouseCD3 (17A2), anti-mouseCD4 (RM4-5), anti-mouseCD8 (53-6.7), antimouseCD11b (M1/70), anti-mouseCD45R/B220 (RA3-6B2), anti-mouseCD86 (GL-1), anti-

AnnexinV (BD Biosciences).

ED

mouseMHC-class-II (H2b, AF6-120.1), anti-humanCD14 (HCD14), anti-humanCD16 (3G8),

PT

Quantitative RT-PCR and ELISA

Total RNA was isolated using the Quick-RNA MiniPrep Kit (Zymo, Irvine, CA) and cDNA was

CE

prepared with the iScript Kit (Bio-Rad, Munich, Germany). Quantitative RT-PCR was performed on an ABI 7500 instrument (Applied Biosystems, Darmstadt, Germany) using the SYBR mastermix from the same company. Results were normalized to the mRNA expres-

AC

sion of HPRT (mouse) or -actin (human), and evaluated using the Ct method. TNF levels in the cell culture supernatant were determined using a commercially available ELISA MAX™ kit (Biolegend) according to the manufacturer’s instructions. Assessment of T cell apoptosis in vitro and in vivo Splenocytes were cultured in RPMI 1640 medium with 10% FCS and treated with BMP-NP, EP-NP, DEX, or PBS for up to 20 hours. Induction of apoptosis was analyzed by FACS after incubation with an antibody against CD3 as well as AnnexinV. To evaluate GC-induced T cell apoptosis in vivo, C57BL/6 wildtype mice were treated for one or three days by daily i.p. injection of BMP-NP, EP-NP, DEX, or PBS. On the day after the last injection, spleens were removed and total splenocyte numbers determined by microscopic counting using a hematocytometer. The cells were analyzed by FACS after staining with antibodies against CD3, CD4 and CD8 to identify individual cellular subsets. Cell numbers in untreated mice served as controls. -7-

ACCEPTED MANUSCRIPT Animal experimentation Wildtype C57BL/6 mice and genetically modified mouse strains on the same background were bred in our own animal facilities at the University Medical Center Göttingen. Both male and female mice were used at 10-14 weeks of age. The following GC receptor (GR) mutant

PT

mouse strains were employed: GRflox/flox (Nr3c1tm2Gsc; designated GRflox), GRflox/flox;LysMCre (Nr3c1tm2Gsc Lyz2tm1(cre)lfo/J; designated GRlysM), GRflox/flox;LckCre (Nr3c1tm2GscTg(Lckdesignated

GRlck),

GRflox/flox;slco1c1CreERT2

SC RI

cre)548Jxm/J;

(Nr3c1tm2GscTg(Slco1c1-

icre/ERT2)1Mrks; designated GRslco1c1). Gene deletion in GRslco1c1 mice was achieved by tamoxifen treatment 4 weeks prior to EAE experiments. To this end, tamoxifen was dissolved in sunflower oil and administered as three consecutive doses of 3 mg every other day via

NU

oral gavage. All experiments were conducted according to Lower Saxony state regulations for animal experimentation and were approved by the responsible authority (LAVES, Olden-

MA

burg, Germany). Active induction of EAE in C57BL/6 mice

Mice were immunized with 50 µg myelin oligodendrocyte glycoprotein peptide 35-55

ED

(MOG35-55) in CFA and treated twice with 400 ng pertussis toxin in total as described [18]. Starting at day 9 after disease induction, the mice were weighed and scored daily. EAE se-

PT

verity was assessed according to a 10-grade scoring scale as previously described[18]: 0 = healthy; 1 = reduced tonus of the tail; 2 = limp tail; 3 = absent righting; 4 = gait ataxia; 5 =

CE

mild paresis of hind limbs; 6 = moderate paraparesis; 7 = severe paraparesis or paraplegia; 8 = tetraperesis; 9 = moribund; 10 = death.

AC

GC therapy of EAE

Treatment with DEX (Dexa-ratiopharm®; Ratiopharm, Ulm, Germany) or BMZ (Celestan®; Merck Sharp & Dohme, Kenilworth, NJ) was started once the mice had reached an average clinical score of 2 after EAE induction. The application protocol consisted of three daily i.p. injections of 10 mg/kg DEX or BMZ, or an equal volume of PBS as a vehicle control. BMPNP were applied at a dose equivalent to 10 mg/kg BMZ, which corresponds to a volume of approximately 100 µl of the undiluted nanoparticle suspension per mouse, following the same application scheme as for free GC. An equal volume of EP-NP served as the respective vehicle control. In one experiment, BMP-NP were applied only once by i.v. injection. Immunohistochemistry Analysis of PFA-fixed and paraffin-embedded spinal cord sections was performed according to standard protocols. In brief, 3 µm cross-sections were stained with an anti-humanCD3 antibody (1:200; Serotec, Düsseldorf, Germany) or an anti-mouseMAC3 antibody (1:200 BD

-8-

ACCEPTED MANUSCRIPT Biosciences) followed by incubation with a biotinylated rabbit anti-rat antibody (1:200; Vector Laboratories, Burlingame, CA). Antigen retrieval was achieved by pre-treating the sections in citrate buffer, pH = 6.0, for 15 minutes in a microwave oven at 850 W. 3,3'-Diaminobenzidine was used for visualization. Quantification was achieved by taking pictures with an Olympus

SC RI

stained areas with ImageJ (http://rsb.info.nih.gov/ij/).

PT

BX51 microscope at a 200-fold magnification and counting individual cells or quantifying the

Preparation of human monocytes

All experiments using human cells were approved by the local ethics committee and adhered to the Declaration of Helsinki. Informed consent was obtained from each subject prior to ma-

NU

terial collection. Blood samples were drawn from twelve healthy volunteers (males: 10; females: 2; median age: 30.0 +/- 2.4 years). Mononuclear cells were isolated from heparinized

MA

blood and subjected to a density gradient using LymphoprepTM (Axis-Shield, Oslo, Norway). Subsequently, monocytes further purified by magnetic cell sorting through negative depletion using the EasySepTM Human Monocyte Enrichment Kit without CD16 Depletion (StemCell Technologies, Grenoble, France). FACS analysis using antibodies against CD14 and CD16

PT

Boyden chamber assay

ED

indicated that the purity was routinely >95%.

Monocytes were cultured for 3 hours in RPMI 1640 medium supplemented with 0.5 % fatty

CE

acid-free BSA under serum-starved conditions, either alone or additionally treated with 10-7 M DEX or BMP-NP containing an equal amount of BMZ. A proportion of the cells were used for RNA isolation and the remaining cells served to assess their migratory capacity. To this end,

AC

5×105 monocytes per well were subjected to a transwell assay using a pore size of 5 µm (Corning Life Sciences, NY, USA) as previously described [29]. The cells were allowed to migrate against a gradient of 10 ng/ml CCL2 (ImmunoTools, Friesoythe, Germany) for one hour. The medium in the lower chamber was harvested and the transmigrated monocytes attached to the plate were incubated with 2 mM EDTA in PBS for 20 minutes at 37°C. Then, the detached cells were scratched off the well bottom and pooled with the harvested medium for analysis. Finally, the transmigrated cells were quantified by FACS analysis using Calibrite Beads (BD Bioscience). Statistical analysis Statistical analyses were performed using the student’s t-test for unpaired samples. The analysis of human monocytes was achieved with the Wilcoxon matched-pairs signed rank test, EAE disease curves were evaluated using the Mann-Whitney U test. GraphPad Prism software (San Diego, CA) was employed for all statistical analyses. Data are depicted as

-9-

ACCEPTED MANUSCRIPT mean ± SEM; measures of significance were as follows: n.s.: p>0.05; *: p<0.05; **: p<0.01;

AC

CE

PT

ED

MA

NU

SC RI

PT

***: p<0.001.

-10-

ACCEPTED MANUSCRIPT Results Synthesis and characterization of BMP-NP IOH-NP were recently presented as a new material concept for drug delivery [26]. Similar to

PT

sodium chloride composed of Na+ cations and Cl- anions, BMP-NP consist of [ZrO]2+ as an inorganic cation and [BMP]2- as a functional organic anion (Figure 1A). In addition, BMP-NP

SC RI

also contain [FMN]2- as a green emitting fluorescent marker (Figure 1B), which can be excited in a range of 300-550 nm and shows intense fluorescence in the green to yellow spectral range of 500-680 nm (Figure 1C,D) [30]. Since [ZrO]2+[(BMP)0.9(FMN)0.1]2- is insoluble in water, nanoparticles and colloidally stable suspensions can be directly formed by suitable injec-

NU

tion of an aqueous solution of ZrOCl28H2O to an aqueous solution of Na2(BMP) and Na(HFMN) [26]. The resulting nanoparticles exhibit a mean diameter of 30-40 nm (according to scanning electron microscopy, Figure 1E) and a hydrodynamic diameter of 60-80 nm with

MA

a negative Zeta potential of about -40 mV at neutral pH (according to dynamic light scattering, Supplementary Figure S1A,B). Such difference between scanning electron microscopy and dynamic light scattering data is often observed for aqueous suspensions and related to

ED

the high polarity of the solvent leading to extended solvent shells and the formation of weak agglomerates via hydrogen bonding. It is noteworthy that nanoparticles with the composition

PT

[ZrO]2+[(HPO4)0.9(FMN)0.1]2- were used as a reference not containing any BMP (designated

AC

CE

EP-NP) in most experiments .

2-

Figure 1. Composition and characterization of BMP-NP. (A,B) The two functional anions [BMP] 22+ 2and [FMN] as constituents of [ZrO] [(BMP)0.9(FMN)0.1] nanoparticles, designated BMP-NP. (C) Excitation and emission spectra of FMN contained in the BMP-NP. (D) Fluorescence of BMP-NP in water. (E) Analysis by scanning electron microscopy to illustrate the particle size of BMP-NP. Size bar: 200 nm.

-11-

ACCEPTED MANUSCRIPT BMP-NP are preferentially taken up by macrophages Initially, we tested whether the uptake of IOH-NP by cell types involved in the pathogenesis of MS was different in vitro. Splenocytes and peritoneal cells were isolated from wildtype mice and incubated with BMP-NP for 24 and 48 hours. Subsequently, their abundance in T

PT

cells, B cells and macrophages was quantified by FACS analysis based on the green fluorescence of FMN. The signal intensity increased in all three cell types over time, indicating

SC RI

that the BMP-NP accumulated within the cells (Figure 2A,B). Regardless of their cellular source, however, macrophages incorporated BMP-NP more efficiently than T and B cells

AC

CE

PT

ED

MA

NU

(Figure 2A,B and Supplementary Figure S2).

Figure 2. Uptake of BMP-NP by leukocyte subpopulations in vitro. (A,B) Splenocytes or perito-6 neal cells were cultured in the presence of BMP-NP containing 10 M BMZ, stained for CD3, B220 and CD11b, and analyzed by FACS. Uptake of BMP-NP was quantified after gating on T cells, B cells or macrophages (MΦ) on the basis of the mean fluorescence intensity (MFI) of FMN contained in the BMP-NP. N = 3. Values are depicted as mean ± SEM; statistical analysis was performed using the unpaired t-test (**: p<0.01; ***: p < 0.001; asterisks: T cells compared to MΦ, diamonds: B cells compared to MΦ). (C) BMDM were cultured in the presence of BMP-NP for 24 hours and their uptake visualized by 2-photon microscopy based on the fluorescence of FMN. One representative photograph out of three is depicted. Size bar: 75 m. (D) PMΦ were cultured with BMP-NP for 24 hours and their uptake visualized by confocal microscopy based on the fluorescence of FMN; the corresponding bright field picture is depicted alongside. Size bar: 15 m.

-12-

ACCEPTED MANUSCRIPT To further investigate the uptake of IOH-NP into macrophages, we incubated BMDM for 24 hours with BMP-NP. By using 2-photon microscopy we could demonstrate intensively green fluorescent cells (Figure 2C), indicating that BMP-NP were an efficient delivery vehicle. To obtain a better understanding of the intracellular localization of the IOH-NP, we incubated

PT

PMΦ for 24 hours with BMP-NP and analyzed them by confocal microscopy. Green fluorescence was found in small clusters within the cytosol of the PMΦ sparing the area of the nu-

SC RI

cleus (Figure 2D). This finding indicates that the BMP-NP were presumably localized in vesicles within the cell following endocytosis rather than sticking unspecifically to the outside of the membrane.

NU

BMP-NP alter macrophage polarization and are nontoxic

In the next step, we characterized BMP-NP concerning their ability to modulate macrophage function in vitro. BMDM were generated from wildtype mice and incubated with BMP-NP at

MA

concentrations equivalent to 10-6 or 10-7 M BMZ for 24 hours. Cells cultured in the presence of the same concentration of DEX, an equal amount of EP-NP, or PBS served as controls (Supplementary Figure S3). FACS analysis revealed that BMP-NP potently diminished sur-

ED

face levels of MHC class II and CD86, which is typical for an anti-inflammatory phenotype of macrophages (Figure 3A and Supplementary Figure S4). Analysis by quantitative RT-PCR

PT

demonstrated that BMP-NP also decreased mRNA levels of TNF and IL-1, while concomitantly increasing the ones of CD163 and Ym1 (Figure 3B). The effects of BMP-NP and DEX

CE

were similar at both concentrations, whereas EP-NP had no impact on any of the tested parameters (Figure 3A,B).

To investigate whether the ZrO contained in the IOH-NP showed any toxicity towards BMDM,

AC

we measured the metabolic activity of these cells after incubation for 24 hours with BMP-NP containing 10-7 M BMZ or an equal amount of EP-NP. In addition, 10-7 M DEX, an equal volume of PBS, or 10% DMSO were added to BMDM as controls (Supplementary Figure S5). EP-NP did not impact cell viability at all whereas DMSO drastically reduced it. This confirms that ZrO has no toxic effect on macrophages. DEX moderately reduced survival whereas BMP-NP was significantly better tolerated by BMDM. This observation indicates that the delivery of GC via IOH-NP even reduces the known anti-proliferative activity of this class of drugs. Collectively, our results confirm that IOH-NP are fully biocompatible.

-13-

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

Figure 3. Modulation of macrophage function by BMP-NP in vitro. BMDM were incubated for 24 -6 -7 hours with 10 or 10 M DEX or BMP-NP containing the same concentration of BMZ. Treatment with equal amounts of EP-NP or PBS served as controls. (A) Surface levels of MHC class II and CD86 + were analyzed by FACS on the basis of the MFI after gating on CD11b cells; the MFI of PBS-treated samples were arbitrarily set to 1. N = 9. (B) Relative mRNA levels of TNF, IL-1, CD163, and Ym1 were determined by quantitative RT-PCR after normalization to the housekeeping gene HPRT; PBStreated samples were arbitrarily set to 1. N = 3. All values are depicted as mean ± SEM; statistical analysis was performed by unpaired t-test and refers to the comparison of each GC-treated sample to the respective vehicle control (n.s. = not significant; *: p < 0.05; **: p<0.01; ***: p < 0.001).

To confirm our findings on the modulation of macrophage function in vivo, wildtype mice received three consecutive daily injections of 10 mg/kg DEX or BMP-NP containing an equivalent dose of BMZ. Equal volumes of PBS or EP-NP were applied as vehicle controls (Supplementary Figure S6). Analysis of PMΦ by FACS revealed that BMP-NP and DEX reduced the percentages of MHC class II+ and CD86+ cells to a similar extent (Figure 4A,B and Supplementary Figure S7). To determine whether BMP-NP and DEX exerted their antiinflammatory effect via the GR, PMΦ were isolated from GRflox and GRlysM mice that had been treated in vivo with PBS, DEX, BMP-NP, or EP-NP. Of note, GRlysM mice specifically lack the GR in macrophages and other myeloid cell types. FACS analysis showed that the -14-

ACCEPTED MANUSCRIPT percentage of MHC class II+ cells amongst GRlysM macrophages was neither reduced by BMP-NP nor DEX (Figure 4C). Finally, we cultured PMΦ from wildtype mice treated as above in the presence of IFN and LPS. Cells obtained from the DEX or BMP-NP groups secreted less TNF than cells from control mice having received PBS or EP-NP (Figure 4D).

PT

Taken together, GC delivered via BMP-NP efficiently polarize macrophages towards an anti-

CE

PT

ED

MA

NU

SC RI

inflammatory phenotype in vitro and in vivo in a GR-dependent manner.

AC

Figure 4. Impact of BMP-NP on macrophage function in vivo. PMΦ were collected from mice having received daily injections of 10 mg/kg DEX or BMP-NP containing the same dose of BMZ for three consecutive days. Mice treated with equal amounts of EP-NP or PBS served as vehicle controls. (A,B) PMΦ were stained for CD11b, MHC class II and CD86 and analyzed by FACS. Percent+ + + ages of MHC class II cells (A) or CD86 cells (B) were determined after gating on CD11b macroflox lysM phages. N = 6-8. (C) PMΦ were isolated from GR or GR mice treated in a similar manner as in + + panel A and analyzed for the percentage of MHC class II cells amongst CD11b macrophages by FACS. N = 3-6. (D) PMΦ obtained from mice treated as in panel A were stimulated with LPS and IFN or left untreated. The concentration of TNFα in the supernatant was determined by ELISA. N = 3. All values are depicted as mean ± SEM; statistical analysis was performed by unpaired t-test and refers to the comparison of each GC-treated sample to the respective vehicle control (n.s. = not significant; *: p < 0.05; **: p<0.01; ***: p < 0.001).

The capacity of BMP-NP to target T cells is reduced compared to free GC The impact of GC on T cells was assessed on the basis of their ability to induce apoptosis. Splenocytes were isolated from wildtype mice, incubated in vitro for up to 20 hours in the presence of 10-6 or 10-7 M DEX, or BMP-NP containing equal amounts of BMZ. Corresponding volumes of EP-NP and PBS were added to control cultures. FACS analysis indicated that GC delivered using IOH-NP induced apoptosis with similar or slightly reduced efficacy compared to free GC (Figure 5A and Supplementary Figure S8). When we repeated the experi-15-

ACCEPTED MANUSCRIPT ment with GRlck mice specifically lacking the GR in T cells, neither DEX nor BMP-NP were able to induce apoptosis, indicating that the latter exerted their effect via the GR in T cells in

PT

ED

MA

NU

SC RI

PT

a similar manner as free GC (Figure 5B).

AC

CE

Figure 5. Induction of T cell apoptosis by BMP-NP in vitro. (A) Splenocytes were incubated for up -6 -7 to 20 hours with 10 or 10 M DEX or BMP-NP containing the same concentrations of BMZ. Treatment with equal amounts of EP-NP or PBS served as reference values. The percentages of live T cells were determined by FACS after staining for CD3 and AnnexinV. Ratios between live T cells in the DEX and PBS group, and between the BMP-NP and the EP-NP group were calculated; ratios in samples analyzed before treatment were arbitrarily set to 1. N = 6. (B) Splenocytes were isolated flox lck -7 from GR or GR mice and incubated for up to 20 hours with 10 M DEX or BMP-NP containing the same concentrations of BMZ. Treatment with EP-NP or PBS served as reference values. DEX/PBS and BMP-NP/EP-NP ratios of live T cells were calculated, and ratios in samples analyzed before treatment were arbitrarily set to 1. N = 6. All values are depicted as mean ± SEM; statistical analysis was performed using the unpaired t-test and refers to the comparison between different treatment groups at each time point (n.s. = not significant; *: p < 0.05; **: p<0.01; ***: p < 0.001).

To test apoptosis induction in vivo, wildtype mice were injected on three consecutive days with 10 mg/kg DEX or BMP-NP containing the same dose of BMZ. Control mice received equal volumes of PBS and EP-NP. On day one and three, spleens were removed and the numbers of CD4+ and CD8+ T cells determined. DEX strongly reduced T cell counts at both time points as expected, whereas BMP-NP had only little effect on the numbers of both T cell subsets compared to the vehicle control (Figure 6A). Notably, neither DEX nor BMP-NP induced T cell apoptosis in GRlck mice (Figure 6B). Taken together, the use of IOH-NP for the delivery of GC reduces their capacity to target T cells. -16-

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Figure 6. Impact of BMP-NP on splenic T cell numbers in vivo. (A) Splenocytes were collected from mice having received daily injections of 10 mg/kg DEX or BMP-NP containing the same dose of BMZ for three consecutive days. Mice treated with equal amounts of EP-NP or PBS served as vehicle + controls. Splenocytes were stained for CD3, CD4, CD8, and AnnexinV, and analyzed by FACS. CD4 + + and CD8 cells were gated on CD3 T cells, and absolute cell numbers were multiplied with the percentages of each cell population. Ratios between numbers of live cells in the DEX and PBS group, and between the BMP-NP and EP-NP group were calculated; ratios in samples analyzed before flox lck treatment were arbitrarily set to 1. N = 3-5. (B) Splenocytes were isolated from GR or GR mice treated in a similar manner as in panel A and analyzed after staining for CD3 and AnnexinV by FACS. Absolute numbers of splenic T cells are depicted for each treatment and genotype. N = 3-6. All values are depicted as mean ± SEM; statistical analysis was performed using the unpaired t-test and refers to the comparison between different treatment groups at each time point and in each genotype (n.s. = not significant; *: p < 0.05; ***: p < 0.001).

BMP-NP are a suitable nanoformulation for GC therapy of EAE Having established that BMP-NP act in a cell type-specific manner in vivo, we next aimed to determine their efficacy in modulating CNS autoimmune responses. EAE was induced by immunization of C57BL/6 mice with MOG35-55 peptide in CFA (Figure 7A). Once the mice had reached an average clinical score of 2, they were treated on three consecutive days with 10 mg/kg DEX, 10 mg/kg BMZ, or BMP-NP containing the same dose of BMZ. Control mice received equal volumes of EP-NP or PBS. Importantly, treatment efficacy was undistinguishable regardless of whether free GC or BMP-NP were administered (Figure 7B). EP-NP had no effect on the disease course at all (Figure 7B). These findings confirm that a potent therapeutic effect can be achieved when using IOH-NP for GC delivery. Interestingly, a single i.v. injection of 10 mg/kg BMP-NP turned out to be as efficient in ameliorating EAE as three consecutive i.p. injections of the same dose (Supplementary Figure S9). This observation sug-17-

ACCEPTED MANUSCRIPT gests that altering the application protocol may allow further improvements in therapeutic efficacy to be made. Notably, the dose of 10 mg/kg GC used in all EAE experiments corresponds very well to the therapeutic regimen employed in treatment of MS patients, who generally receive daily pulses of 1-2 g of methylprednisolone [31].

PT

The two major target cells of GC in EAE therapy were quantified in spinal cord sections by immunohistochemistry one day after the last injection of BMP-NP containing 10 mg/kg BMZ

SC RI

or an equal volume of EP-NP as a control. T cell numbers in the CNS of mice of both treatment groups were similar (Figure 7C). In contrast, macrophage infiltration into the spinal cord was significantly reduced by administration of BMP-NP compared to EP-NP as determined by electronic quantification of the area occupied by MAC3+ cells (Figure 7C,D). This observa-

NU

tion provides evidence that macrophages are the primary target of BMP-NP in the treatment of EAE.

MA

Macrophages are critical targets of BMP-NP in EAE therapy As our findings up to this point indicated that BMP-NP predominantly impact myeloid cells, we aimed to confirm this notion by inducing EAE in cell type-specific GR knock-out mice.

ED

Initially, we induced EAE in GRlck mice and GRflox control mice, and treated them with BMP-NP or EP-NP in a similar manner as in the previous experiment. BMP-NP ameliorated

PT

the disease course regardless of the genotype (Figure 8A). Hence, they are able to modulate EAE independently of the presence of the GR in T cells, which is in line with their limited ca-

CE

pacity to modulate this cell type in vivo. Next we immunized GRlysM mice and GRflox control mice with MOG35-55 peptide, and treated them with BMP-NP or EP-NP as a control. BMP-NP ameliorated EAE symptoms in GRflox

AC

mice as expected (Figure 8B). The disease course in GRlysM mice was slightly aggravated but, most importantly, it was not improved by administration of BMP-NP (Figure 8B). This observation suggests that macrophage polarization to the anti-inflammatory phenotype is crucial for therapeutic efficacy of BMP-NP in the treatment of EAE. Finally, we tested the administration of BMP-NP in GRslco1c1 mice, a strain which allows an inducible ablation of the GR in brain endothelial cells that form the blood-brain-barrier and protect the CNS from leukocyte infiltration. Gene disruption was achieved by repeated application of tamoxifen, and four weeks later, EAE was induced in GRslco1c1 mice as well as similarly treated GRflox control mice. BMP-NP ameliorated clinical symptoms with comparable efficacy in mice of both genotypes (Figure 8C), suggesting that the modulation of brain endothelial cells was not essential for therapeutic efficacy of BMP-NP.

-18-

AC

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

Figure 7. Impact of BMP-NP on the disease course of EAE and spinal cord infiltration in mice. (A) Scheme of the employed mouse model. (B) EAE was induced in C57BL/6 mice, and after reaching an average clinical score of 2, they received three consecutive daily i.p. injections of 10 mg/kg DEX, 10 mg/kg BMZ, or BMP-NP containing the same dose of BMZ (black arrows). Injection of equal amounts of EP-NP or PBS served as controls; day 0 corresponds to the first day of treatment in each individual mouse. N = 5-9. (C,D) C57Bl/6 mice suffering from EAE were treated daily on three consecutive days with BMP-NP containing 10 mg/kg BMZ or an equal amount of EP-NP. One day after the last injection the spinal cords was isolated and paraffin sections stained for CD3 or MAC3 to iden+ tify T cells and macrophages, respectively. (C) Counting of CD3 cells and quantification of the area + occupied by MAC3 macrophages were performed using ImageJ software. N = 5/6. (D) Representative pictures of MAC3-stained spinal cord sections from one mouse treated with BMP-NP and one treated with EP-NP from the same experiment as shown in panel C. Magnification: 10x and 40x. Statistical significance in panel B was calculated using the Mann-Whitney U test, and in panel C using the unpaired t-Student test; all values are depicted as mean ± SEM (n.s. = not significant; *: p < 0.05; **: p<0.01).

-19-

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

Figure 8. Efficacy of BMP-NP therapy of EAE in cell type-specific GR knock-out mice. EAE was lck lysM induced in mice lacking the GR in T cells (GR ), macrophages (GR ), or brain endothelial cells slco1c1 flox lck (GR ), as well as in each of the corresponding GR littermate controls: (A) GR (N = 7/6/8/8), lysM slco1c1 (B) GR (N = 8/3/6/5), and (C) GR (N = 6/5/5/5). BMP-NP containing 10 mg/kg BMZ or an equal dose of EP-NP were applied i.p. once per day for three consecutive days starting when the mice had reached an average clinical score of 2 (black arrows). Disease severity was evaluated on a daily basis for 6 days after the beginning of the treatment. All values are depicted as mean ± SEM; statistical analysis was performed using the Mann-Whitney U test (n.s. = not significant; *: p < 0.05; **: p<0.01).

BMP-NP impact human monocyte function To determine whether BMP-NP were also able to modulate human cells, we isolated monocytes from peripheral blood of healthy human volunteers (Figure 9A). Initially, we sought to measure chemokine-directed monocyte migration towards CCL2 in vitro after incubation with 10-7 M DEX or BMP-NP containing the same concentration of BMZ. Treatment with both GC formulations increased the monocytes’ migratory capacity, although BMP-NP were slightly more potent than DEX (Figure 9B). Gene expression of CCR2, the receptor for CCL2, remained unaltered within this time frame (Figure 9C). To test whether BMP-NP polarized monocytes towards the anti-inflammatory phenotype in a similar manner as murine macro-20-

ACCEPTED MANUSCRIPT phages, we performed a quantitative RT-PCR analysis. Expression of the anti-inflammatory genes CD163, CD206, and Arg1 was increased by DEX and BMP-NP although with partially different magnitudes (Figure 9D-F). Concomitantly, mRNA levels of the pro-inflammatory cytokine IL-1were reduced (Figure 9G). Taken together, GC delivered via IOH-NP induce

PT

changes in chemotaxis and polarization of human monocytes with comparable potency as

AC

CE

PT

ED

MA

NU

SC RI

free GC, suggesting that they might also be therapeutically active in patients.

Figure 9. Modulation of human peripheral blood monocytes by BMP-NP. (A) Scheme of the ex-7 perimental setup. Monocytes from the blood of healthy human volunteers were incubated with 10 M DEX, BMP-NP containing the same concentration of BMZ, or left untreated (con). (B) The migratory capacity towards CCL2 was determined using a Boyden chamber assay. (C-G) Gene expression of CCR2, CD163, CD206, Arg1, and IL-1 was analyzed by quantitative RT-PCR and normalized to mRNA levels of -actin. N = 12 (each symbol represents an individual blood donor). All values are depicted as mean ± SEM; statistical analysis was performed using the Wilcoxon matched-pairs signed rank test (n.s. = not significant; *: p < 0.05; **: p<0.01; ***: p < 0.001).

-21-

ACCEPTED MANUSCRIPT Discussion High-dose GC therapy has been the gold standard for the treatment of acute disease bouts in MS patients for the last 30 years [13]. However, recent studies in mice and rats have re-

PT

vealed that application of GC encapsulated in PEGylated liposomes might be superior to the use of the conventional drug as it allows to reduce the dose and application frequency while

SC RI

retaining full therapeutic efficacy [19, 32]. Notwithstanding the favorable features of PEGylated liposomes, side effects have also been reported. The most important complication appears to be the activation of the complement system, which was demonstrated in cell culture, animal models, and clinical studies, and may occasionally lead to life-threatening hypersensi-

NU

tivity reactions [22, 33-36]. More recently developed phospholipid and polymer-based nanocarrier systems could be an alternative to PEGylated lipsomes for the targeted delivery

MA

of GC as they possess potent anti-inflammatory activity in the absence of any overt side effects [23-25]. These findings indicate that nanoformulations in general are a promising strategy for improving the treatment of MS and other chronic inflammatory disorders with GC,

ED

although PEGylated liposomes might not be the optimal choice. The use of IOP-NP is a new attractive option for GC delivery.

Our findings confirmed that BMP-NP are as potent as free DEX in the regulation of macro-

PT

phages and T cells at standard drug concentrations in vitro, which is possibly due to the large excess of nanoparticles present in the medium. Nonetheless, this observation was somewhat

CE

surprising since the uptake of IOH-NP by macrophages was more efficient than by T cells. The situation in vivo where nanoparticles are limited, however, was different. BMP-NP and

AC

free DEX were equally powerful in repressing MHC class II expression and TNF secretion by macrophages, while induction of T cell apoptosis by BMP-NP was strongly reduced. Consequently, IOH-NP have a clear cell type specificity in vivo. It is noteworthy that BMP-NP act exclusively via the GR, thus eliminating the possibility that nanoparticles affect immune cell functions through unspecific mechanisms. Confocal microscopy confirmed that IOH-NP are localized within the cell after incubation, but the exact mechanism via which they are taken up remains unknown. Yet, it is very likely that phagocytic cells recognize the inorganic component of the nanoparticles as we didn’t notice any difference between BMP-NP and EP-NP concerning uptake efficacy. Collectively, our data indicate that IOH-NP are suitable to deliver GC to macrophages, which is in line with the notion that they are taken up by endocytosis and exhibit a strong selectivity for myeloid cells [37]. Any new strategy aimed at improving CNS inflammation in MS patients must prove that it is better than currently available drugs: either its efficacy or specificity has to be superior over standard therapies. Our findings indicate that BMP-NP were as potent in reducing clinical symptoms of EAE as free DEX and free BMZ, two GC that a very similar and only differ in -22-

ACCEPTED MANUSCRIPT the stereoisomery of the C16 methyl group. Furthermore, the disease course in mice receiving EP-NP was the same as in those treated with PBS, indicating that IOH-NP themselves do not impact EAE. While having similar therapeutic efficacy, BMP-NP employ a different cellular mechanism to ameliorate disease symptoms, which makes them potentially suitable to

PT

overcome complications associated with standard GC therapy. Previously published data show that free GC reduce both T cell and macrophage infiltration into the spinal cord [18].

SC RI

The findings presented here revealed that BMP-NP only affected the latter. Experiments performed with conditional GR knock-out mice further supported our assumption that BMP-NP act in a cell type-specific manner. They attenuated clinical symptoms in mice lacking the GR in T cells or brain endothelial cells, but mice lacking the GR in myeloid cells were refractory

NU

to BMP-NP therapy. Interestingly, the dependency of BMP-NP on the presence of the GR in myeloid cells is higher than previously observed for liposomal GC [19]. This finding not only

MA

reconfirms a superior specificity of IOH-NP compared to PEGylated liposomes, but also excludes the potential release of free BMP from the nanoparticles by hydrolysis before they are taken up by their target cells.

The aim of preclinical studies is to eventually translate findings into human patients. As a first

ED

approach we tested whether BMP-NP were suitable to modulate the function of monocytes from healthy human volunteers. In the course of inflammatory diseases, these cells alter their

PT

migratory behavior and adapt a different gene expression profile. Many of these processes are affected by GC, which is known to contribute to their therapeutic efficacy. Our finding that

CE

BMP-NP were able to increase monocyte migration towards CCL2, reduce IL-1 expression, and upregulate CD163, CD206, and Arg1 mRNA expression in vitro indicates that IOH-NP allow the delivery of GC to human myeloid cells. Therefore it is likely that BMP-NP are useful

AC

for clinical applications.

Newly developed drug formulations not only have to prove therapeutic efficacy but they also need to be nontoxic. Previous publications have demonstrated good tolerability of ZrO in different cell lines and organisms [38-40]. In contrast, ZrO has also been reported to induce pro-inflammatory cytokine expression and release from macrophages [41]. Our own findings have so far not provided any hint that constituents of the employed IOH-NP, in particular the ZrO, might be toxic. They neither reduced the viability of BMDM in vitro nor did we observe any abnormalities or worsened clinical symptoms in vivo, even after several days. We therefore believe that IOH-NP can be applied safely to animals and presumably humans as well, although more systematic studies need to be conducted. In summary, our findings indicate that the delivery of GC via IOH-NP might be attractive new option for the treatment of acute relapses in MS patients as these nanoparticles have an improved target cell-specificity, full therapeutic efficacy, and no obvious toxicity. Therefore future investigations on BMP-NP in clinical trials appear warranted.

-23-

ACCEPTED MANUSCRIPT Acknowledgements

We thank Amina Bassibas, Birgit Curdt, and Martina Weig for expert technical help, Michael

PT

Engelke for his help with confocal microscopy, and Cathy Ludwig for text editing. This work was supported by the German Research Foundation (DFG) through grants RE 1631/15-1

Conflict of interest

AC

CE

PT

ED

MA

The authors declare no conflict of interest.

NU

SC RI

and LU 634/9-1.

-24-

ACCEPTED MANUSCRIPT References

H. El-Gabalawy, L.C. Guenther, C.N. Bernstein, Epidemiology of immune-mediated inflammatory diseases: incidence, prevalence, natural history, and comorbidities, J Rheumatol Suppl, 85 (2010) 2-10.

[2]

V. Bruno, G. Battaglia, F. Nicoletti, The advent of monoclonal antibodies in the treatment of chronic autoimmune diseases, Neurol Sci, 31 Suppl 3 (2011) 283-288.

[3]

H.M. Reichardt, R. Gold, F. Lühder, Glucocorticoids in multiple sclerosis and experimental autoimmune encephalomyelitis, Expert Rev Neurother, 6 (2006) 16571670.

[4]

F. Lühder, H.M. Reichardt, Traditional concepts and future avenues of glucocorticoid action in experimental autoimmune encephalomyelitis and multiple sclerosis therapy, Crit Rev Immunol, 29 (2009) 255-273.

[5]

U. Baschant, J. Tuckermann, The role of the glucocorticoid receptor in inflammation and immunity, J Steroid Biochem Mol Biol, 120 (2010) 69-75.

[6]

A. Compston, A. Coles, Multiple sclerosis, Lancet, 359 (2002) 1221-1231.

[7]

A. Nylander, D.A. Hafler, Multiple sclerosis, J Clin Invest, 122 (2012) 1180-1188.

[8]

M.J. Herold, H.M. Reichardt, Glucocorticoid-induced apoptosis in animal models of multiple sclerosis, Crit Rev Immunol, 33 (2013) 183-202.

[9]

H.J. Fischer, N. Schweingruber, F. Lühder, H.M. Reichardt, The potential role of T cell migration and chemotaxis as targets of glucocorticoids in multiple sclerosis and experimental autoimmune encephalomyelitis, Mol Cell Endocrinol, 380 (2013) 99-107.

PT

ED

MA

NU

SC RI

PT

[1]

[10] D.M. Mosser, J.P. Edwards, Exploring the full spectrum of macrophage activation, Nat Rev Immunol, 8 (2008) 958-969.

CE

[11] A. Mantovani, A. Sica, S. Sozzani, P. Allavena, A. Vecchi, M. Locati, The chemokine system in diverse forms of macrophage activation and polarization, Trends Immunol, 25 (2004) 677-686.

AC

[12] R.A. Du Pasquier, D.D. Pinschewer, D. Merkler, Immunological mechanism of action and clinical profile of disease-modifying treatments in multiple sclerosis, CNS Drugs, 28 (2014) 535-558. [13] N.M. Milligan, R. Newcombe, D.A. Compston, A double-blind controlled trial of high dose methylprednisolone in patients with multiple sclerosis: 1. Clinical effects, J Neurol Neurosurg Psychiatry, 50 (1987) 511-516. [14] J.A. Cohen, P.B. Imrey, P.A. Calabresi, K.R. Edwards, T. Eickenhorst, W.L. Felton, 3rd, E. Fisher, R.J. Fox, A.D. Goodman, C. Hara-Cleaver, G.J. Hutton, B.F. Mandell, T.F. Scott, H. Zhang, C. Apperson-Hansen, G.J. Beck, P.L. Houghtaling, M.T. Karafa, M. Stadtler, Results of the Avonex Combination Trial (ACT) in relapsing-remitting MS, Neurology, 72 (2009) 535-541. [15] A. Pato-Pato, J.M. Prieto, M. Lema, D. Dapena-Bolano, J. Abella-Corral, J.M. Pumar, [Cerebral atrophy in multiple sclerosis patients treated periodically with boluses of methylprednisolone], Rev Neurol, 37 (2003) 501-506. [16] D.E. Goodkin, R.P. Kinkel, B. Weinstock-Guttman, S. VanderBrug-Medendorp, M. Secic, D. Gogol, J.E. Perryman, M.M. Uccelli, L. Neilley, A phase II study of i.v. methylprednisolone in secondary-progressive multiple sclerosis, Neurology, 51 (1998) 239-245.

-25-

ACCEPTED MANUSCRIPT [17] R. Gold, C. Linington, H. Lassmann, Understanding pathogenesis and therapy of multiple sclerosis via animal models: 70 years of merits and culprits in experimental autoimmune encephalomyelitis research, Brain, 129 (2006) 1953-1971.

PT

[18] S. Wüst, J. van den Brandt, D. Tischner, A. Kleiman, J.P. Tuckermann, R. Gold, F. Lühder, H.M. Reichardt, Peripheral T cells are the therapeutic targets of glucocorticoids in experimental autoimmune encephalomyelitis, J Immunol, 180 (2008) 8434-8443.

SC RI

[19] N. Schweingruber, A. Haine, K. Tiede, A. Karabinskaya, J. van den Brandt, S. Wüst, J.M. Metselaar, R. Gold, J.P. Tuckermann, H.M. Reichardt, F. Lühder, Liposomal encapsulation of glucocorticoids alters their mode of action in the treatment of experimental autoimmune encephalomyelitis, J Immunol, 187 (2011) 4310-4318.

NU

[20] D.H. Lee, C. Rotger, C.C. Appeldoorn, A. Reijerkerk, W. Gladdines, P.J. Gaillard, R.A. Linker, Glutathione PEGylated liposomal methylprednisolone (2B3-201) attenuates CNS inflammation and degeneration in murine myelin oligodendrocyte glycoprotein induced experimental autoimmune encephalomyelitis, J Neuroimmunol, 274 (2014) 96101.

MA

[21] R.M. Schiffelers, M. Banciu, J.M. Metselaar, G. Storm, Therapeutic application of longcirculating liposomal glucocorticoids in auto-immune diseases and cancer, J Liposome Res, 16 (2006) 185-194. [22] J.M. van den Hoven, R. Nemes, J.M. Metselaar, B. Nuijen, J.H. Beijnen, G. Storm, J. Szebeni, Complement activation by PEGylated liposomes containing prednisolone, Eur J Pharm Sci, 49 (2013) 265-271.

ED

[23] C.Y. Zhao, T.T. Fan, Y. Yang, M.L. Wu, L. Li, Z. Zhou, Y.L. Jian, Q. Zhang, Y. Huang, Preparation, macrophages targeting delivery and anti-inflammatory study of pentapeptide grafted nanostructured lipid carriers, Int J Pharmaceut, 450 (2013) 11-20.

CE

PT

[24] K. Ren, A. Dusad, F. Yuan, H.J. Yuan, P.E. Purdue, E.V. Fehringer, K.L. Garvin, S.R. Goldring, D. Wang, Macromolecular prodrug of dexamethasone prevents particleinduced peri-implant osteolysis with reduced systemic side effects, Journal of Controlled Release, 175 (2014) 1-9.

AC

[25] K. Ren, H.J. Yuan, Y.J. Zhang, X. Wei, D. Wang, Macromolecular glucocorticoid prodrug improves the treatment of dextran sulfate sodium-induced mice ulcerative colitis, Clinical Immunology, 160 (2015) 71-81. [26] J.G. Heck, J. Napp, S. Simonato, J. Mollmer, M. Lange, H.M. Reichardt, R. Staudt, F. Alves, C. Feldmann, Multifunctional phosphate-based inorganic-organic hybrid nanoparticles, J Am Chem Soc, 137 (2015) 7329-7336. [27] A. Kleiman, S. Hübner, J.M. Rodriguez Parkitna, A. Neumann, S. Hofer, M.A. Weigand, M. Bauer, W. Schmid, G. Schütz, C. Libert, H.M. Reichardt, J.P. Tuckermann, Glucocorticoid receptor dimerization is required for survival in septic shock via suppression of interleukin-1 in macrophages, Faseb J, 26 (2012) 722-729. [28] D. Tischner, A. Weishaupt, J. van den Brandt, N. Müller, N. Beyersdorf, C.W. Ip, K.V. Toyka, T. Hünig, R. Gold, T. Kerkau, H.M. Reichardt, Polyclonal expansion of regulatory T cells interferes with effector cell migration in a model of multiple sclerosis, Brain, 129 (2006) 2635-2647. [29] N. Schweingruber, H.J. Fischer, L. Fischer, J. van den Brandt, A. Karabinskaya, V. Labi, A. Villunger, B. Kretzschmar, P. Huppke, M. Simons, J.P. Tuckermann, A. Flügel, F. Lühder, H.M. Reichardt, Chemokine-mediated redirection of T cells constitutes a critical mechanism of glucocorticoid therapy in autoimmune CNS responses, Acta Neuropathol, 127 (2014) 713-729. [30] M. Roming, H. Lunsdorf, K.E. Dittmar, C. Feldmann, ZrO(HPO(4))(1-x)(FMN)(x): quick and easy synthesis of a nanoscale luminescent biomarker, Angew Chem Int Ed Engl, 49 (2010) 632-637. -26-

ACCEPTED MANUSCRIPT [31] R.L. Oliveri, P. Valentino, C. Russo, G. Sibilia, U. Aguglia, F. Bono, F. Fera, A. Gambardella, M. Zappia, K. Pardatscher, A. Quattrone, Randomized trial comparing two different high doses of methylprednisolone in MS: a clinical and MRI study, Neurology, 50 (1998) 1833-1836.

PT

[32] J. Schmidt, J.M. Metselaar, M.H. Wauben, K.V. Toyka, G. Storm, R. Gold, Drug targeting by long-circulating liposomal glucocorticosteroids increases therapeutic efficacy in a model of multiple sclerosis, Brain, 126 (2003) 1895-1904.

SC RI

[33] A. Chanan-Khan, J. Szebeni, S. Savay, L. Liebes, N.M. Rafique, C.R. Alving, F.M. Muggia, Complement activation following first exposure to pegylated liposomal doxorubicin (Doxil): possible role in hypersensitivity reactions, Ann Oncol, 14 (2003) 1430-1437. [34] J. Szebeni, Complement activation-related pseudoallergy: a new class of drug-induced acute immune toxicity, Toxicology, 216 (2005) 106-121.

MA

NU

[35] J. Szebeni, L. Baranyi, S. Savay, J. Milosevits, R. Bunger, P. Laverman, J.M. Metselaar, G. Storm, A. Chanan-Khan, L. Liebes, F.M. Muggia, R. Cohen, Y. Barenholz, C.R. Alving, Role of complement activation in hypersensitivity reactions to doxil and hynic PEG liposomes: experimental and clinical studies, J Liposome Res, 12 (2002) 165-172. [36] J. Szebeni, G. Storm, Complement activation as a bioequivalence issue relevant to the development of generic liposomes and other nanoparticulate drugs, Biochem Biophys Res Commun, 468 (2015) 490-497.

ED

[37] R. Weissleder, M. Nahrendorf, M.J. Pittet, Imaging macrophages with nanoparticles, Nat Mater, 13 (2014) 125-138.

PT

[38] E. Demir, D. Burgucu, F. Turna, S. Aksakal, B. Kaya, Determination of TiO2, ZrO2, and Al2O3 nanoparticles on genotoxic responses in human peripheral blood lymphocytes and cultured embyronic kidney cells, J Toxicol Environ Health A, 76 (2013) 990-1002.

CE

[39] G. Karunakaran, R. Suriyaprabha, P. Manivasakan, R. Yuvakkumar, V. Rajendran, N. Kannan, Screening of in vitro cytotoxicity, antioxidant potential and bioactivity of nanoand micro-ZrO2 and -TiO2 particles, Ecotoxicol Environ Saf, 93 (2013) 191-197.

AC

[40] V. Saxena, A. Diaz, A. Clearfield, J.D. Batteas, M.D. Hussain, Zirconium phosphate nanoplatelets: a biocompatible nanomaterial for drug delivery to cancer, Nanoscale, 5 (2013) 2328-2336. [41] G.A. Obando-Pereda, L. Fischer, D.R. Stach-Machado, Titanium and zirconia particleinduced pro-inflammatory gene expression in cultured macrophages and osteolysis, inflammatory hyperalgesia and edema in vivo, Life Sci, 97 (2014) 96-106.

-27-

SC RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

Graphical abstract

-28-