Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate

YEXNR-11302; No. of pages: 16; 4C: Experimental Neurology xxx (2012) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr

4

F

Q1 3

Rina Aharoni a, Efrat Sasson b, Tamar Blumenfeld-Katzir b, Raya Eilam c, Michael Sela a, Yaniv Assaf d, Ruth Arnon a,⁎

5

a

7 8

c

Q5 6

O

2

Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate

Department of Immunology, The Weizmann Institute of Science, Rehovot, Israel Bioimage, MRI Research and Consulting, Tel Aviv, Israel Department of Veterinary Resources, The Weizmann Institute of Science, Rehovot, Israel d Department of Neurobiology, Wise Faculty of Life Science, Tel Aviv University, Tel Aviv, Israel b

R O

1

9

i n f o

a b s t r a c t

The roles of inflammation and degeneration as well as of gray matter abnormalities in multiple sclerosis (MS) and its animal model experimental autoimmune encephalomyelitis (EAE) are controversial. We analyzed the pathological manifestations in two EAE models, the chronic oligodendrocyte glycoprotein (MOG)-induced versus the relapsing–remitting proteolipid protein (PLP)-induced, along the disease progression, using advanced magnetic resonance imaging (MRI) parameters. The emphasis of this study was the overall assessment of the whole brain by histogram analysis, as well as the detection of specific affected regions by voxel based analysis (VBA) using quantitative T2, magnetization transfer ratio (MTR) and diffusion tensor imaging (DTI). Brains of EAE-inflicted mice from both models revealed multiple white and gray matter areas with significant changes from naïve mice for all MRI parameters. Ventricle swelling was more characteristic to the PLP-induced model. Decreased MTR values and increased apparent diffusion coefficient (ADC) were observed mainly in MOG-induced EAE, indicative of macromolecular loss and structural CNS damage involvement in the chronic disease. The MS drug glatiramer acetate (GA), applied either as prevention or therapeutic treatment, affected all the MRI pathological manifestations, resulting in reduced T2 values and ventricle volume, elevated MTR and decreased ADC, in comparison to untreated EAE-inflicted mice. In accord, immunohistochemical analysis indicated less histological damage and higher amount of proliferating oligodendrocyte progenitor cells after GA treatment. The higher brain tissue integrity reflected by the MRI parameters on the level of the whole brain and in specific regions supports the in situ anti-inflammatory and neuroprotective consequences of GA treatment. © 2012 Published by Elsevier Inc.

E

D

Article history: Received 18 August 2012 Revised 16 October 2012 Accepted 5 November 2012 Available online xxxx

P

a r t i c l e

E

C

T

Keywords: Multiple sclerosis (MS) Experimental autoimmune encephalomyelitis (EAE) Glatiramer acetate (GA) Magnetic resonance imaging (MRI) Magnetization transfer ratio (MTR) Diffusion tensor imaging (DTI) Neuroprotection

R

R

10 11 12 13 14 15 17 16 18 19 20 21 22 23 24 25 26 27

49

N C O

50

Introduction

52

Multiple sclerosis (MS) is a chronic disease of the central nervous system (CNS) in which pro-inflammatory processes that target self myelin constituents lead to multifocal demyelination (Lassmann, 2008; Lassmann et al., 2007). In addition, increasing evidence indicates that the axonal and neuronal pathology, which is crucial to the impairment associated with the progressive disease course, begins at early disease stage (Bruck, 2008; Trapp and Nave, 2008). Diffuse abnormalities in the gray matter and in normal-appearing brain tissue are currently recognized as central components of MS (Geurts and Barkhof, 2008). Yet,

53 54 55 56 57 58 59 60

U

51

Abbreviations: MS, Multiple sclerosis; EAE, Experimental autoimmune encephalomyelitis; GA, Glatiramer acetate; MOG, Oligodendrocyte glycoprotein; PLP, Proteolipid protein; CNS, Central nervous system; MTR, Magnetization transfer ratio; DTI, Diffusion tensor imaging; VBA, Voxel based analysis. ⁎ Corresponding author at: The Department of Immunology, The Weizmann Institute of Science, Rehovot, Israel, 76100. Fax: +972 8 9469712. E-mail address: [email protected] (R. Arnon).

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 48 47

the relative roles of the diverse pathological manifestations in this multifaceted disease are highly controversial. In this respect, deeper investigation of the various experimental autoimmune encephalomyelitis (EAE) models, induced by exposing susceptible mice strains to different myelin antigens and which mirror different aspects of MS, may lead to better understanding. Indeed, we have recently demonstrated that the proteolipid protein (PLP)-induced relapsing–remitting model is characterized mainly by widespread myelin damage, whereas in the chronic model induced by oligodendrocyte glycoprotein (MOG), axonal degeneration and neuronal loss are more prevalent (Aharoni et al., 2011). MS diagnosis and assessment has progressed dramatically by the application of magnetic resonance imaging (MRI). At present, MRI is an essential tool for treatment management and evaluation of therapeutic impact (Polman et al., 2011). In addition, MRI methodologies are employed to investigate various EAE models in order to study the detrimental processes occurring within the CNS and their relevance to the human disease, as they allow direct correlation of radiological

0014-4886/$ – see front matter © 2012 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.expneurol.2012.11.004

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 Q6 76 77 78

104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121

Q8 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144

151

C57BL/6 and (SJL/JxBALB/c)F1 mice were purchased from Harlan (Jerusalem, Israel). Female mice, 8–12 weeks of age, were used and kept under specific pathogen free (SPF) environment. All experiments were approved by the Institutional Animal Care and Use Committee of the Weizmann Institute.

152

Induction and evaluation of EAE

157

Chronic EAE was induced in C57BL/6 mice by injecting a peptide consisting of amino acids 35–55 of myelin oligodendrocyte glycoprotein (MOG), synthesis by Genscript (Piscataway, NJ, USA). Relapsing– remitting EAE was induced in (SJL/JxBALB/c)F1 mice by the peptide encompassing amino acids 139–151 of proteolipid protein (PLP) synthesized by Genscript. Mice were injected subcutaneously at the flank, with 200 μl emulsion containing 200–300 μg of the encephalitogenic peptide in incomplete Freund's adjuvant enriched with 3 mg/ml heat-inactivated Mycobacterium tuberculosis (Sigma, St. Louis, MO, USA). Pertussis toxin (Sigma), 250 μg/mouse, was injected intravenously immediately after the encephalitogenic injection and 48 h later. Mice were examined daily. EAE was scored as follows: 0—no disease, 1—limp tail, 2—hind limb paralysis, 3—paralysis of all limbs, 4— moribund condition, and 5—death.

158

Glatiramer acetate (GA, Copaxone, Copolymer 1)

172

GA consists of acetate salts of synthetic polypeptides containing four amino acids L-alanine, L-glutamate, L-lysine, and L-tyrosine (Teitelbaum et al., 1971). GA from batch 242905809, with an average molecular weight of 7.7 kDa, obtained from Teva Pharmaceutical Industries (Petah Tiqva, Israel) was used throughout the study. GA treatment was applied by consecutive 7–8 daily subcutaneous injections, in 0.1 ml phosphate buffered saline, either as a prevention treatment starting one day following disease induction, or as a suppression treatment beginning after the appearance of clinical manifestations. In the first experiment GA doses of 0.1, 0.5 and 2.0 mg per mouse were tested. In subsequent experiments the dose of 2 mg/mouse was used. Layouts of the GA treatment schedules are demonstrated in panel A of Figs. 3–5. Mice that were not treated by GA (untreated control) were similarly injected by PBS.

173 174

Magnetic resonance imaging (MRI)

187

Brain MRI measurements were performed using 9.4 Tesla BioSpec Magnet, at three time points along disease progression (TP1, TP2, TP3), namely at days 13, 20, 27 and days 11, 17, 27 for the MOG and the PLP models, respectively, 5–9 mice per group. During imaging, mice were anesthetized with 2–2.5% isofluorane and were placed in a 35 mm diameter birdcage coil. An MR compatible small animal monitoring system was used to monitor respiratory rate and body temperature of the anesthetized mouse was maintained during imaging. The MRI protocol included the following sequences:

188 189

Quantitative T2: T2 multi-slice multi-echo (MSME) sequence was performed with sixteen echoes collected at intervals of 10 ms from TE = 10 ms to TE = 160 ms, with TR = 3000 ms and with NA = 2, FOV 22 mm 2, 14 slices, slice thickness of 1 mm and a

197

F

Mice

O

102 103

150

R O

100 101

Materials and methods

P

98 99

145 146

D

96 97

treatment by GA, were analyzed both for the whole brain by histogram analysis and for specific areas indicated by voxel based analysis (VBA). We report herewith on differences in the MRI parameters characteristic to each EAE model and a beneficial effect of GA treatment in their restoration.

T

94 95

C

92

Q7 93

E

90 91

R

88 89

R

86 87

O

85

C

83 84

N

81 82

and histopathological findings. Advance MRI parameters that are sensitive to the microstructural changes occurring in the tissue can serve as biomarkers for different aspects of the disease pathology (Vigeveno et al., 2012). Quantitative T2, which is sensitive to water content and tissue composition, has the potential to reflect pathological changes such as edema and myelin damage (MacKay et al., 2006). Increased T2-weighted signals were demonstrated in the corpus callosum of rats inflicted by MOG-induced EAE, which is manifested by focal lesions (Serres et al., 2009), and in a mouse model combining cuprizone-induced demyelination and MOG induced EAE (Boretius et al., 2011). Magnetization transfer ratio (MTR) — the ratio with and without magnetization transfer, is considered a measure of the macromolecular structure and myelin water fraction, both potential measures of the myelin sheath integrity (Dousset et al., 1992; Whittall et al., 1997; Wolff and Balaban, 1989). Studies using the focal MOG-induced EAE in rats revealed correspondence between reduced MTR values and demyelination within the lesions (Rausch et al., 2009; Serres et al., 2009). Reduction in MTR was found also in the corpus callosum in the cuprizone–MOG combined mouse model (Boretius et al., 2011), and in normal appearing white matter in brains of chronic-progressive EAE-induced guinea pig (Gareau et al., 2000). Diffusion tensor imaging (DTI), which measures the magnitude and directionality of water diffusion in the tissue, serves as an indicator for axonal integrity. Apparent diffusion coefficient (ADC) indicates the average diffusion in the tissue (Basser and Pierpaoli, 1998). Positive correlation between clinical scores and ADC values in the external capsule (Verhoye et al., 1996), as well as in the corpus callosum (Serres et al., 2009), were demonstrated in the rat brain using chronic-relapsing and focal EAE models, respectively. Both MTR and DTI have been investigated as measures of demyelination, axonal damage and myelin integrity (Budde et al., 2008, 2009; DeBoy et al., 2007; Filippi and Agosta, 2009; Serres et al., 2009; Vigeveno et al., 2012). As such they are linked to the neurodegenerative component of MS and complement the established MRI readouts of inflammation. The current study was prompted by the notion that further usage of these MRI measurements to study the whole brain, as well as various specific regions in different EAE models and the effect induced by treatment, may contribute to the elucidation of in situ pathological and therapeutic mechanisms. Glatiramer acetate (GA, Copaxone®), a synthetic polypeptide of L-alanine, L-lysine, L-glutamic acid and L-tyrosine (Teitelbaum et al., 1971), is an approved MS drug which is widely used as first-line disease-modifying therapy (Aharoni, 2010; Carter and Keating, 2010). In MS patients, GA treatment has been shown to modify various MRI parameters that indicate disease activity and severity. These include the reduction in mean number of gadolinium-enhancing lesions, the number of new enhancing lesions, the volume of enhancing lesions and the change in the volume and number of lesions on T2-weighted images (Carter and Keating, 2010; Comi et al., 2001; Sormani et al., 2005). The mechanism of action of GA was extensively studied in several EAE models. These studies attributed the therapeutic activity of GA to immunomodulation, mainly by the induction of anti-inflammatory Th2/3 and T-regulatory cells that penetrate the CNS and induce in situ bystander suppression (Aharoni et al., 2000, 2003, 2010; Farina et al., 2005). During recent years, cumulative results indicated that, in addition to its immunomodulatory activity, GA induces neuroprotective and repair processes within the CNS, as manifested by the decrease in neurological damage and demyelination as well as by increased expression of neurotrophic factors, neurogenesis and remyelination (Aharoni et al., 2005a, 2005b, 2008, 2011; Azoulay et al., 2005). In the present study we used these advanced MRI methodologies, namely quantitative T2, MTR and DTI, to investigate the two widely used MS models: the PLP-induced relapsing–remitting EAE and the chronic EAE form induced by MOG. The various manifestations revealed at different time points during disease course, as well as following

U

79 80

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

E

2

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

147 148 149

153 154 155 156

159 160 161 162 163 164 165 166 167 168 169 170 171

175 176 177 178 179 180 181 182 183 184 185 186

190 191 192 193 194 195 196 Q9

198 199 200

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

208 209 210 211 212 213 214 215 216 217 218

MRI analysis

220

The following MRI parameters were calculated: quantitative T2 maps, magnetization transfer ratio (MTR) and apparent diffusion coefficient (ADC). Lateral ventricle volume was measured manually using in-house software written in Matlab (©MathWorks, USA) on the quantitative T2 maps. Analysis was performed both for the entire brain by histogram analysis and on a voxel based analysis (VBA) for area of significance (p b 0.005). Lateral ventricle volume was measured on the T2 maps.

228 229 230 231 232 233 234 235

Image analysis MRI parameters were calculated using in-house software written in Matlab (©MathWorks, USA). Quantitative T2 was calculated using non-linear fit to produce a quantitative T2 map. DTI analysis and calculation of apparent diffusion coefficient (ADC) were performed as described previously (Basser and Pierpaoli, 1998). Magnetization transfer ratio (MTR) was calculated as MTR = Ms − M0 / M0 (M0— without saturation pulse; M0—with saturation pulse).

C

227

E

225 226

R

223 224

R

221 222

T

219

236 Histogram analysis Q11 237 The extracranial content was removed from all images. Histogram of

242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257

N C O

240 241

the T2, MTR and ADC parameters were computed with 0.2 ms wide bins, 0.01%, wide bins and 0.01 s/mm2 wide bins respectively. To correct for differences between mice in brain size, each bin was normalized by using the total number of pixels contributing to the histogram. Height of the histogram is the largest normalized frequency. Statistical analysis The statistical analysis included the following procedures:

U

238 239

1. Spatial registration and normalization of T2 maps, DTI b0, and MTR images to an in-house mouse brain atlas were performed using the SPM 2.0 software (Wellcome Trust Centre for Neuroimaging, London, UK). The registration and normalization parameters of the DTI b0 were applied on the ADC map. 2. Spatial smoothing of 3 mm FWHM was performed on the images. 3. Significant differences between naïve, EAE sham-treated and GA-treated mice were analyzed using VBA analysis of variance (ANOVA) and t test analysis using the SPM 2.0 software (Wellcome Trust Centre for Neuroimaging, London, UK) (p b 0.005, uncorrected). 4. Areas within the CSF and within the ventricles were excluded from the statistics due to the enlarged ventricle volume observed in the disease animals.

261

Mice were deeply anesthetized and perfused transcardially with 2.5% paraformaldehyde. Brains were removed, postfixed (48 h, 4 °C), paraffin embedded and sectioned coronaly by microtome (Leica, 4 μm). Paraffin sections were stained by hematoxylin and eosin (H&E) and by Luxol fast blue (LFB) for cellular structure and myelin, respectively. For immunostaining paraffin sections that had been deparaffinized and antigenretrieved (in 10 nM citric acid pH 6), were pre-incubated in PBS solution containing 20% normal horse serum and 0.2% Triton for 1 h, and then incubated overnight with primary antibodies. The following antibodies were used: rat anti-myelin basic protein (MBP, Abcam, Cambridge, United Kingdom), mouse anti-Alzheimer Precursor Protein A4 (APP, Millipore, Temecula, CA, USA), rabbit anti-chondroitin sulfate proteoglycan NG2 (Chemicon, Temecula, CA, USA), and mouse anti-proliferation cell nuclear antigen (PCNA, Abcam). The second antibody step was performed by labeling with species specific highly cross-absorbed cy2 or cy3 conjugated antibodies (Jackson ImmunoResearch, West Grove, PA, USA), 1:200 for 30–60 min. In some cases the signal was enhanced by incubation with biotinylated secondary antibodies for 90 min, followed by cy2 or cy3 conjugated Streptavidin (Jackson ImmunoResearch). Sections were stained with Hoechst 33258 (Molecular Probes, Eugene, OR, USA) for nuclear counter staining. Stained sections were examined and photographed by fluorescence microscope (E600, Nikon, Tokyo, Japan), equipped with Plan Fluor objectives connected to CCD camera (DMX1200F, Nikon). Digital images were collected and analyzed using Image-Pro Plus software (Media Cybernetics, Inc., Bethesda, MD, USA). Images were assembled using Adobe Photoshop (Adobe Systems).

262

Results

289

Disease manifestations, as well as the effect of GA treatment, were investigated in two EAE models: the chronic model induced by MOG and the relapsing remitting model induced by PLP. Brain MRI was performed at three time points along disease progression (TP1, TP2, TP3), at days 13, 20, and 27 and days 11, 17, and 27 for the MOG and the PLP models, respectively. The following MRI parameters were measured: quantitative T2, magnetization transfer ratio (MTR) and diffusion tensor imaging (DTI). Lateral ventricle volume was calculated from T2 maps. In order to assess the overall changes in the diseased brain we employed histogram analysis that reflects the distribution of the various MRI parameters for the whole brain. To determine the specific brain regions that are affected by disease induction and treatment voxel based analysis (VBA) for areas of significance (pb 0.005) was performed. In addition, following the final MRI measurement (one month after disease induction), the mice were sacrificed and their brains were examined by histological procedures using various histological staining combinations.

290 291

F

207

Histopathology

O

206

R O

205

P

204

D

202

Q10 203

5. Post-hoc analysis on the significant voxels was performed using 258 STATISTICA software (StatSoft, Inc., USA) on clusters that exceeded 259 the statistical threshold of p b 0.005. 260

matrix of 256 × 128 which was zero-filled to 256 × 256, yielding an in-plane resolution of 86 × 86 (μm) 2. Magnetization transfer ratio: magnetization transfer was performed with two spin-echo images, and with and without off-resonance pulses were acquired for calculation of the magnetization transfer ratio. TR/TE=3000/10 ms, with 2 kHz off-resonance pulse, NA=2, FOV=22 mm2, 14 slices, slice thickness of 1 mm and a matrix of 256×128, which was zero-filled to 256×256, yielding an in-plane resolution of 86×86 μm2. Diffusion tensor imaging (DTI) acquisition was performed with a diffusion-weighted (DW) spin-echo echo-planar-imaging (EPI) pulse sequence having the following parameters: TR/TE= 3000/25 ms, Δ/δ = 10/4.5 ms, four EPI segments, and 15 noncollinear gradient directions with a single b-value shell at 1000 s/mm2 and one image with a b-value of 0 s/mm2 (referred to as b0). Geometrical parameters were: 12 slices, each 1 mm thick, FOV 19.2 mm2, matrix size 96× 128 which was zero-filled to 128 × 128, yielding an in-plane resolution of 150× 150 (μm)2.

E

201

3

263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288

292 293 294 295 296 297 298 299 300 301 302 303 304 305 306

Differences between EAE-induced and naïve mice by brain MRI parameters — 307 comparison between the MOG and the PLP models 308 MOG-EAE induction resulted in a chronic disease course with persistent kinetics. In contrast, in mice inflicted by PLP-induced EAE, initial disease exacerbation was followed by remission, with 1–2 relapses in a period of 45 days. Substantial variability within mice in the same group in the kinetic and severity of the relapses was typical to the PLP model. The daily clinical scores of the entire groups as well as of two individual representative mice for each model are depicted in Fig. 1A. MRI analyses performed for the EAE-induced mice of both models, in

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

309 310 311 312 313 314 315 316

340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374

400

O

F

The effect of GA treatment on MRI parameters

R O

338 339

P

336 337

375 376

D

334 335

signify the typical inflammation in EAE. For the MTR parameter, more areas with significant changes were depicted in PLP induced mice, indicative of more macromolecular/myelin damage in this model at the early disease phase. Differences in the ADC parameter were found in few brain regions at disease appearance, in both models, but at the later time point, multiple areas were revealed only in the MOG models, while in the PLP model no region with significant changes from naïve mice could be detected. This finding corroborates the overall brain ADC histogram analysis, indicative of the progressive structural/axonal damage in this model along disease development. Staining by hematoxylin and eosin (H&E) for detection of cellular accumulation and lesions and by Luxol fast blue (LFB) for myelin presence, revealed multifocal perivascular and parenchymal cellular infiltrations, multiple lesions and demyelination sites in various brain regions (demonstrated in brain stem, corpus callous and thalamus areas for both models in Fig. 2A). Immunohistochemical staining for myelin by antibodies to myelin basic protein (MBP) and for damaged axons by antibodies to amyloid precursor protein (APP) further revealed multiple demyelinated lesions surrounded by positive APP expressing axons in sites of cellular accumulation (indicated by Hoechst staining) in both models (demonstrated in the internal capsule and corpus callosum for the MOG model and in the thalamus and cerebellum for the PLP model, Fig. 2B). These findings corroborate the widespread inflammation, demyelination and axonal injury characteristic to EAE, shown by the various MRI parameters.

377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399

A. Prevention regimen GA treatment in MOG-induced mice GA treatment, by injection of 0.1, 0.5 or 2.0 mg for eight consecutive days, starting one day after MOG induction (prevention), blocked the development of the clinical manifestations of chronic EAE (Fig. 3A). In this experiment, at the time of clinical disease appearance (day 11), brains of EAE untreated mice revealed significant enlargement of the lateral ventricle volume and increased T2 signals. GA treatment led to reduction of ventricle volumes, significant difference from untreated-mice by GA 2.0 mg/day (Figs. 3B, C). The reduction in T2 histogram peak heights obtained by GA-treatment did not reach significance over untreated-mice, but in mice treated by GA 2.0 mg/day the T2 histogram peak height was not significantly different from that of the naive controls (Fig. 3D). VBA for area of T2 change in GA-treated compared to untreated mice revealed significant T2 reduction after GA treatment in various brain regions shown for representative regions – the caudate putamen, corpus callosum, subiculum, brain stem and cerebellum – in Figs. 3E, F. Since 2.0 mg/day GA was the most effective dosage, this treatment was used in the subsequent experiments.

401 402

B. Prevention GA treatment in PLP-induced mice GA treatment, 2.0 mg/mouse, starting one day after PLP induction for seven consecutive days, prevented disease exacerbation as well as its typical relapses (Fig. 4A). Although GA blocked the clinical symptoms, the ventricle enlargement characteristic to this model was manifested in treated as well as in untreated mice at the early time point — 11 days after EAE induction. But, with time, whereas in untreated mice ventricle swelling persisted, in GA-treated mice their size decreased (significant reduction from untreated mice at days 17 and 27, Figs. 4B, C). Furthermore, at the late time point ventricle volume in the GA-treated mice was similar to that of naïve controls.

419 420

T

332 333

C

330 331

E

328 329

R

326 327

R

324 325

O

323

C

321 322

N

319 320

comparison to naïve controls from their corresponding strains, along the disease course, are demonstrated in Figs. 1B–F. In the PLP model, the prominent phenomenon was the enlargement of the lateral ventricle volume, extending 2–7 folds from that of naïve controls, significant differences at all three time points, evident in spite of the high variability among the mice in this model (Fig. 1B). In contrast, in mice induced by the MOG model, lateral ventricle enlargement was less manifested. It should be noted that in one experiment significant ventricle enlargement was observed in MOG-induced mice (Figs. 3B, C), but to a lesser extent than that obtained in PLP-induced mice. Increased T2-values were found in mice from both models, yet on the level of the whole brain, the T2 histogram peak heights did not reach significance over normal brains (Fig. 1C). These results may indicate that considerable fraction of the brain was not affected by the disease as apparent by the T2 parameter. But, in addition, the expansion of the ventricles may have increased the density of the surrounding tissue, thus reducing the T2 value. Indeed when the area under the curve of the T2 histogram tail was calculated, a significant T2 increase was revealed in the PLP model, in which the sizes of ventricle were especially elevated (Fig. 1D). MTR values were reduced in EAE inflicted mice. The histogram peak heights of the overall brains for the MTR parameter were significantly reduced, in MOG-induced mice at all three time points, and in PLP-induced mice at the early time point, 18–23% reduction from the values of naïve controls (Fig. 1E). These changes are indicative of the widespread damage to the macromolecule/myelin structure. The insignificant change in the overall MTR values obtained for the PLP group at the late time points could be related to less structural damage or to the substantial variability between the mice typical to this model. The ADC parameter on the level of the overall brain was not affected at the early disease phase (Fig. 1F). But along disease progression, in MOG-induced mice, significant elevation in the histogram peak height was evident, 13% above the values of naïve controls by day 27, indicative of development of structural/axonal damage and decrease in tissue integrity in this model. To detect specific affected brain regions voxel based analysis was performed in EAE inflicted mice in comparison to syngeneic naïve controls. VBA for all the MRI parameters was carried out in both EAE models following disease exacerbation (TP1). In addition, for the ADC parameter, analysis was performed at the late disease stage (TP3), since at that time point histogram peak (in the MOG model) was at its highest level. Areas within the CSF and around the ventricles were excluded from the analysis to avoid artifact due to the enlarged ventricle volume. All comparisons revealed higher T2, lower MTR, and higher ADC in brains of the EAE-mice than in the naïve controls. The main brain regions depicted by this analysis are demonstrated in Table 1. Differences between EAE and naïve controls in all MRI parameters were found in structures of white matter (corpus callosum) and gray matter (thalamus, hippocampus and deep mesencephalic nucleus) in MOG as well as PLP-induced mice. Changes in motor (e.g. internal capsule, cerebellar peduncles, cerebellum), sensory (e.g. S1, visual and auditory cortex), hippocampal-limbic (e.g. entorinal cortex, subiculum) and brain stem structures were identified by one or more MRI parameters in both models. These results indicate the widespread distribution of disease manifestation in the CNS. The increased T2 value intensities found in multiple brain regions in both models

U

317 318

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

E

4

Fig. 1. The consequences of EAE on MRI parameters — comparison between the MOG and the PLP-induced models along disease progression. EAE was induced in C57BL/6 mice by the MOG peptide 35–55 and in (SJL/JxBALB/c)F1 mice by the PLP peptide 139–151 (6–9 mice per group). MRI measurements were acquired at days 13, 20, and 27 and days 11, 17, and 28 after disease induction, in the MOG and the PLP models, respectively (indicated by arrows in panel A). A. Clinical disease manifestation shown as the averaged daily scores ± standard errors and as daily scores of individual mice, in the MOG model (left) and in the PLP model (right). B. Lateral ventricle volume calculated from T2 maps, presented as elevation folds from naïve control mice of the same strain. C–F. Histogram analysis of whole brain MRI parameters presented as percent change from naïve controls. T2 histogram peak heights (C), T2 histogram tail area under curve (D), MTR histogram peak heights (E), and ADC histogram peak heights (F), are presented. #Significant differences from naïve controls, (p b 0.005).

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418

421 422 423 424 425 426 427 428 429

5

U

N C O

R

R

E

C

T

E

D

P

R O

O

F

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488

T2

MTR

ADC

T2

*

*

*

*

*

F

* *

*

O

Sensory system S1 Sensory cortex Medullary lamina Piriform cortex Insular cortex Auditory cortex Visual cortex

R O

451 452

Motor system Internal capsule M1 Pyramidal tract/stria terminalis Cerebellar peduncle/spinocerebellar tract Cerebellum

Striatum Nucleus accumbens Caudate–putamen Sub-thalamic nucleus

* *

* *

*

* *

*

*

*

*

Limbic system Orbital cortex Anterior commissure Cingulate cortex Amygdala

*

* * * *

* * * * *

* * *

*

* * * *

*

* *

Hippocampal formation Hippocampal commissure Hippocampus Entorhinal cortex Subiculum

* * *

*

*

* * *

Brain stem regions Brain stem Cranial nerve roots Inferior superior colliculi

* * *

*

*

*

* *

* * * *

Other Thalamus Corpus callosum External capsule Deep mesencephalic nucleus Hypothalamus

ADC

*

*

* *

MTR

*

* *

* * * *

* * * *

* * *

* * * *

t1:1 t1:2 t1:3 t1:4 t1:5 t1:6

PLP model

P

449 450

MOG model

D

447 448

Area of significance

T

445 446

C. Suppression treatment in the MOG-induced model To study the consequences of GA on MRI parameters when treatment is applied by therapeutic regimen (suppression treatment), GA treatment was initiated after clinical disease exacerbation. The MOG-induced model was used due to its consistent disease course. Seven daily injections, applied at days 13–20 from disease induction, ameliorated the clinical manifestations of chronic MOG-induced EAE (Fig. 5A). MRI measurements were performed before treatment initiation (day 13, TP1) after treatment conclusion (day 20, TP2) and seven days after treatment conclusion (day 27, TP3). The lateral ventricle enlargement and the T2 elevation on the level of the whole brain histograms, which were not notably elevated in untreated mice in this model (Figs. 1B, C), were also not significantly affected by treatment (not shown). VBA for area of T2 change in GA-treated compared to untreated mice revealed several regions of significant differences (shown for representative regions in Figs. 5C, D). The opposite phenomenon, of elevated T2 in GA-treated mice compared to untreated mice was not found. Notably, in all these regions, significant T2 reduction in GA-treated mice was obtained for TP2 — at the time of treatment completion, whereas for TP3 – 7 days later – when the reduction in T2 values was evident also in untreated mice, T2 values of treated and untreated mice were similar. To further understand the consequences of GA therapeutic treatment we studied MTR and ADC parameters in brains of GA-treated versus untreated mice. MTR histogram analysis revealed significant reduction for untreated mice at all time points tested, as well as for GA treated mice at TP1, indicative of the widespread damage to the macromolecule/myelin structure before treatment initiation (Figs. 6A, B). At the day of treatment termination (TP2) MTR histogram values did not indicate significant impact of GA treatment compared to untreated mice. But, at the late time point — two weeks after treatment initiation, the histogram peak height of the GA-treated group was significantly higher than that of the untreated group and similar to the naive controls. VBA for area of significant changes for the late time point revealed similar patterns in multiple brain regions, namely reduced MTR in untreated EAE-mice at all the time points, as well as in GA-treated mice before and immediately after the treatment period, and significant elevation by GA treatment with restored MTR values at the late time point (Figs. 6C, D). The opposite phenomena, of reduced MTR values in GA-treated mice compared to untreated mice were not found. The MTR restoration, on the level of the whole brain as well as in specific brain areas, suggests the induction of repair processes to the myelin and to other cellular structures by GA-treatment. Histogram analysis of ADC indicated that, in contrast to the untreated EAE mice in which overall brain values increased along disease progression, in GA-treated mice the ADC parameter did not change at all the

C

443 444

E

441 442

R

439 440

R

437 438

Table 1 Voxel based analysis for area of significance for the T2, MTR and ADC parameters, performed at the first time point (TP1). Significant difference from naïve controls is indicated by *. For the ADC parameter analysis was performed also at the late time point (TP3), and significant difference from naïve at TP3 is indicated by (p b 0.005).

*

*

* * *

t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17 t1:18 t1:19 t1:20 t1:21 t1:22 t1:23 t1:24 t1:25 Q3 t1:26 t1:27 t1:28 t1:29 t1:30 t1:31 t1:32 t1:33 t1:34 t1:35 t1:36 t1:37 t1:38 t1:39 t1:40 t1:41 t1:42 t1:43 Q4 t1:44 t1:45 t1:46 t1:47 t1:48 t1:49 t1:50 t1:51

O

436

C

434 435

N

432 433

On the level of the whole brain, the T2 histogram peak heights, which were not notably elevated in untreated mice, were also not significantly affected by treatment (Fig. 1D). However, when the area under the curve of the T2 histogram tail was calculated, a significant decrease by GA treatment was revealed (Fig. 1E). In spite of the high variability within the groups, typical to this model, a few areas of significant T2 reduction in GA-treated mice, in comparison to untreated mice, were detected, demonstrated for the caudate putamen and the corpus callosum at the second time point in Figs. 4F, G. The opposite phenomenon, namely elevated T2 in GA-treated mice compared to untreated mice, was not found. These results obtained by the prevention treatment in both the MOG and PLP models indicate the ability of GA treatment to block the typical EAE inflammatory processes.

U

430 431

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

E

6

time points tested (Figs. 7A, B). Although at treatment initiation the ADC histogram peak height of the GA group was significantly higher than that of the untreated group, two weeks after treatment initiation their ADC values were significantly lower. Yet, it should be noted, that even at the latest time point the overall brain ADC value in the GA-treated mice was higher than in healthy controls. VBA exposed multiple brain regions in which the ADC values were significantly lower after GA treatment in comparison to untreated mice, attaining similar values to those of naïve controls (Figs. 7C, D). The opposite phenomena, of elevated ADC values in GA-treated mice compared to untreated mice were not found. In two regions of the hippocampal formation the decline from untreated mice reached to significant level already one week after treatment initiation. The decreased ADC, obtained on the

Fig. 2. Histopathology in brain regions of MOG and PLP-induced mice. A. Staining by hematoxylin and eosin (H&E) and by Luxol fast blue (LFB), demonstrating perivascular and parenchymal cellular infiltrations, multiple lesions and demyelination sites, in the brain stem, thalamus and corpus callous, in both models. B. Immunohistochemical staining by antibodies to myelin basic protein (MBP) and to amyloid precursor protein (APP) and by Hoechst, revealing demyelinated lesions, surrounded by positive APP expressing axons, as well as cellular infiltrations, in the internal capsule and corpus callous for the MOG model and in the thalamus and cerebellum for the PLP model. Representing images from 2 different mice for each model are demonstrated out of 5 mice analyzed.

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

489 490 491 492 493 494 495 496 497 498 499 500 501

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

F O R O P D E T C E R R N C O

504

In accord with the findings obtained by the MRI parameters, im- 505 munohistochemical staining revealed cellular infiltration, myelin 506 loss and APP expression in brain regions of EAE-untreated mice, 507

level of the whole brain as well as in specific brain areas, suggests reduced structural/axonal damage and increased tissue integrity after GA-treatment.

U

502 503

7

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

U

N

C

O

R

R

E

C

T

E

D

P

R O

O

F

8

Fig. 3. The effect of prevention treatment by different GA doses on chronic MOG-induced EAE. GA treatment, 0.1, 0.5 or 2.0 mg/day per mouse, was applied by 8 daily injections, starting one day after MOG peptide 35–55 EAE induction (5 mice per treatment group). MRI measurements were acquired 11 days after disease induction. A. Clinical disease manifestations depicted by the averaged daily scores ± standard errors. B. Representative T2 relaxation maps. C. Lateral ventricle volume calculated from the T2 maps, presented as elevation folds from naïve controls. D. Assessment of whole brain T2 indicated by histogram peak heights. E. Voxel based analysis indicating areas of significant differences in T2 between naïve, EAE untreated and EAE treated groups. The significant regions are superimposed on T2 relaxation maps (p b 0.005). F. Averaged T2 values at significant brain regions. #Significant differences from naïve controls, *significant differences from untreated mice (p b 0.005).

508 509 510 511 512

indicative of the in situ inflammation, demyelination and axonal degeneration processes. In contrast, in brains of EAE mice treated by GA (suppression treatment) cell infiltration, demyelination and axonal damages were drastically lower as demonstrated in Fig. 8 for the inferior cerebral peduncle (A) and the corpus callosum (B),

corroborating the consequences of GA treatment found in these regions by the T2, MTR and DTI parameters. To find out if the consequences of GA treatment can be related to its effect on the oligodendrocyte population we performed double staining for the oligodendrocyte progenitor marker NG2 and the proliferation cell

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

513 514 515 516 517

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

531

529 530

C

T

E

D

P

R O

O

F

Inflammation, demyelination and neuroaxonal injury as well as their 532 distribution in the CNS are main constituents of the heterogeneous 533

E

525 526

Discussion

R

524

527 528

R

522 523

vicinity of damage sites. These findings suggest that the reduced myelin damages in mice treated by GA after disease exacerbation, observed by both the MRI and the histology approaches, may result from increased repair processes such as oligodendrocyte mediated remyelination.

N C O

520 521

nuclear antigen PCNA, which is expressed at high levels in proliferating cells (Celis et al., 1986). As demonstrated in Fig. 8C, in brains of EAE untreated mice, in sites of damage, most of the PCNA expressing cells did not express NG2, indicating the limited repair capability at the chronic disease stage (one month after MOG induction). As revealed by the accumulation of Hoechst labeled nuclei, these cells could indicate the proliferation of inflammatory cell. In brains of GA-treated mice higher amounts of NG2+ oligodendrocyte progenitor cells as well as double positive NG2/PCNA proliferating oligodendrocytes were found in the

U

518 519

9

Fig. 4. The effect of GA prevention treatment on relapsing remitting PLP-induced EAE. GA treatment, 2.0 mg/day per mouse, was applied by 7 daily injections, starting one day after PLP peptide 139–151 EAE induction (6 mice per group). MRI measurements were acquired 11, 17 and 28 days after disease induction (indicated by arrows in A). A. Clinical disease manifestations depicted by the averaged daily scores ± standard errors. B. Representative T2 relaxation maps acquired at day 11. C. Lateral ventricle volume calculated from the T2 maps, presented as elevation folds from naïve controls. D. Assessment of whole brain T2 indicated by histogram peak heights. E. T2 histogram tail area under curve F. Voxel based analysis indicating areas of significant differences in T2 between EAE untreated and EAE treated groups on day 28. The significant regions are superimposed on T2 relaxation maps (p b 0.005). G. Averaged T2 values at significant brain regions. #Significant differences from naïve controls, *significant differences from untreated mice (p b 0.005).

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

Q2

U

N

C

O

R

R

E

C

T

E

D

P

R O

O

F

10

Fig. 5. The effect of GA suppression treatment on chronic MOG-induced EAE. GA treatment, 2.0 mg/day per mouse, was applied by 7 daily injections, starting 13 days after MOG peptide 35–55 EAE induction (7–9 mice per group). MRI measurements were acquired 13, 20, and 27 days after disease induction (indicated by arrows in A). A. Clinical disease manifestations depicted by the averaged daily scores ± standard errors. B. Representative T2 relaxation maps acquired at day 20. C. Lateral ventricle volume calculated from the T2 maps, presented as elevation folds from naïve controls. D. Assessment of whole brain T2 indicated by histogram peak heights. E. Voxel based analysis indicating areas of significant differences in T2 between EAE untreated and EAE treated groups on day 20. The significant regions on superimposed on T2 relaxation maps (p b 0.005). F. Averaged T2 values at significant brain regions. #Significant differences from naïve controls, *significant differences from untreated mice (p b 0.005).

Fig. 6. The effect of GA suppression treatment on magnetization transfer ratio (MTR) in chronic MOG-induced EAE. GA treatment, 2.0 mg/day per mouse, was applied by 7 daily injections, starting 13 days after MOG peptide 35–55 EAE induction (7–9 mice per group). MRI measurements were acquired 13, 20, and 27 days after disease induction. A. Histogram of whole brain MTR and B. their peak heights indicating overall brain MTR values. C. Voxel based analysis indicating areas of significant differences in MTR between EAE untreated and EAE treated groups on day 27. The significant regions are superimposed on T2 relaxation maps (p b 0.005). D. Averaged MTR values at significant brain regions. #Significant differences from naïve controls, *significant differences from untreated mice (p b 0.005).

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

11

U

N C O

R

R

E

C

T

E

D

P

R O

O

F

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

U

N

C

O

R

R

E

C

T

E

D

P

R O

O

F

12

Fig. 7. The effect of GA suppression treatment on apparent diffusion coefficient (ADC) in chronic MOG-induced EAE. GA treatment, 2.0 mg/day per mouse, was applied by 7 daily injections, starting 13 days after MOG peptide 35–55 EAE induction (7–9 mice per group). MRI measurements were acquired 13, 20, and 27 days after disease induction. A. Histogram of whole brain ADC and B. their peak heights indicating overall brain ADC values. C. Voxel based analysis indicating areas of significant differences in ADC between EAE untreated and EAE treated groups on day 27. The significant regions are superimposed on T2 relaxation maps (p b 0.005). D. Averaged ADC values at significant brain regions. #Significant differences from naïve controls, *significant differences from untreated mice (p b 0.005).

534 535 536 537 538 539

pathology of MS and are thus essential targets for disease assessment and therapy. In this study we applied various MRI parameters, namely quantitative T2, MTR and DTI, to investigate the pathological manifestations in two widely used clinically different MS models, along the disease course as well as following GA treatment. The emphasis of this study was the overall assessment of the brain by histogram analysis, as well as the

detection of specific affected regions by voxel based analysis, in order to characterize the changes induced by the disease and by its therapy. Analysis of disease-inflicted mice in comparison to naïve controls revealed distinct pathological pattern for each EAE model. In the relapsing remitting model induced by PLP, a prominent phenomenon was the enlargement of the lateral ventricle volume. This finding

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

540 541 542 543 544 545

13

U

N C O

R

R

E

C

T

E

D

P

R O

O

F

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

Fig. 8. The effect of GA treatment on brain histopathology in chronic MOG-induced EAE. Immunohistochemical staining by antibodies to myelin basic protein (MBP) and to amyloid precursor protein (APP) and by Hoechst, in A. the inferior cerebral peduncle and B. the corpus callosum, one month after MOG peptide 35–55 induction, revealing demyelinated lesions with positive APP expressing axons and cellular infiltrations in untreated mice, and less damages in mice treated by GA, suppression treatment started 13 days after disease induction. C. The effect of GA treatment on oligodendrocyte proliferation. Staining by antibodies to the oligodendrocyte progenitor marker NG2 and the proliferation marker PCNA, and by Hoechst, in the midbrain. Higher amounts of NG2 expressing oligodendrocyte progenitor cells and double positive NG2/PCNA proliferating oligodendrocytes are depicted in sites of cellular infiltration in GA-treated mice than in EAE untreated mice. Representing images from 5 mice analyzed from each group.

546 547 548 549 550 551 552

should be attributed to increased inflammation and swelling rather than to atrophy or tissue loss, since the ADC parameter which is sensitive to structural damage was not affected in this model. Disrupted MTR parameter obtained for the whole brain by histogram analysis, as well as the multiple regions depicted by the VBA, indicates the widespread damage to the macromolecule/myelin structure in this model at the time of disease appearance. At the later time points

MTR changes were insignificant, which could be due to the high variability in this model. But they could also signify restoration, namely the occurrence of myelin repair, which was manifested clinically by the disease remission typical to this model. These results are in accord with our previous findings, obtained by electron microscopy analysis, which showed substantial demyelination and remyelination as a characteristic feature of the PLP model (Aharoni et al., 2011).

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

553 554 555 556 557 558 559

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

P

R O

O

F

MRI changes could be attributed to disrupted myelin and cell membrane integrity within gray matter similar to the finding in post mortem examinations of brains of MS patients (Kutzelnigg and Lassmann, 2006). Increased diffusivity in gray matter of MS patients was also attributed to the loss of cellular integrity (Peterson et al., 2001). Most of the previous studies in the EAE system focused on white matter, whereas MTR and DTI were less studied in the gray matter. However, in MS patients, reduced MTR and increased diffusion values were found in the gray matter and even in normal appearing gray matter (Davies et al., 2004; Ramio-Torrenta et al., 2006; Vrenken et al., 2006). Furthermore, changes in MTR and DTI in the gray matter correlated with MS severity and could predict disability worsening (Agosta et al., 2006; Rovaris et al., 2006). Several explanations were suggested to gray matter deterioration in MS patients as reflected in these MRI parameters. It was suggested that even within the gray matter, changes in MTR and ADC reflect changes in axon–myelin structure (Magliozzi et al., 2007). But other notions claim that MTR and ADC changes within the gray matter are secondary to white matter lesions and may be a result of Wallerian degeneration or anterograde trans-synaptic changes (Geurts and Barkhof, 2008). Be that as it may, pathology in gray matter structures such as cortical regions hippocampus and thalamus is currently recognized as central component of MS that underlines neurological impairment such as cognitive dysfunction in MS patients (Cercignani et al., 2001; Geurts and Barkhof, 2008). The findings obtained in this study for the two models by the various MRI parameters as well as by the histological staining establish the occurrence of gray matter pathology in EAE and confirm the widespread disease distribution in the CNS. The consequences of EAE therapy by GA were investigated in this study on the various MRI parameters, in both models using prevention and therapeutic (suppression) regimens. GA treatment by the prevention regimen significantly reduced the lateral ventricle enlargement associated with the inflammatory disease course. Concerning the T2 parameter, the consequences of GA treatment (prevention) were revealed by reduction in the histogram peak height in the MOG model (when elevation in untreated mice was apparent), and by significant reduction in the area under the curve of the histogram tail — in the PLP model. In addition, following GA treatment (both by prevention and suppression regimens) several brain regions with significant T2 reduction compared to untreated mice were depicted. These results are in accord with various studies that demonstrated the immunomodulatory effect of GA on the typical EAE/MS inflammation (Aharoni et al., 2003, 2010; Farina et al., 2005). Both the MTR and the DTI parameters, which are linked to the neurodegenerative component of the disease (Filippi and Agosta, 2009; Vigeveno et al., 2012), were improved by GA treatment. Thus, GA restored the MTR values to the normal level, and significantly reduced the ADC values compared to those of untreated mice. The effect of GA on the ADC parameter was particularly prominent since at treatment initiation the GA-group manifested significantly more damage than the sham-treated group. The histogram findings for the overall brain were confirmed by the voxel analysis that depicted multiple individual regions with MTR and ADC patterns similar to that of the histograms. Furthermore, this is supported by the immunohistological analysis that demonstrated drastically less inflammation, demyelination and axonal damage in the brain regions of GA treated mice. These combined results indicate reduced structural myelin and axonal damage and increased tissue integrity after GA-treatment. The increase in MTR and decrease in ADC in comparison to untreated mice, were obtained when GA treatment was initiated after disease exacerbation (suppression regimen), when substantial injury was already manifested, as evident by their values before treatment initiation. This suggests that GA induces not only prevention of tissue damage but also genuine repair processes. These findings are in accord with several previous indications for increased repair processes after GA treatment, e.g. the elevation in the CNS of neurotrophic factors such as

U

N

C

O

R

R

E

C

T

In the model induced by MOG peptide, the elevation in the ventricle volume and the T2 histogram peak height compared to naïve con562 trols, were significant in one experiment, yet they did not reach Q12 563 significant level in another experiment. This could result from the 564 differences in the time points in which the measurements were 565 performed, as well as from variability between individual experi566 ments. The results obtained for the T2 parameter in this study may 567 reflect inconsistency or insensitivity of this measure when it is evaluat568 ed on the level of the whole brain. Concerning the changes in ventricle 569 volume, even when significant differences were evident, they were less 570 prominent in MOG-induced than in the PLP-induced mice, emphasizing 571 the differences between the models as reflected by this parameter. 572 The chronic MOG model was characterized by consistent reduction 573 in the MTR histogram peak heights at all the time points tested, indicat574 ing damage to macromolecule/myelin structure all along the disease 575 duration. In addition, with time, 20 and 27 days after disease induction, 576 significant increase in the ADC histogram values was revealed in this 577 model. Furthermore, at the latest time point, in this model the ADC 578 voxel analysis depicted multiple areas of significant changes from 579 naïve mice, while in the PLP model no such region could be detected. 580 Changes in DTI parameters have been reported before in the optic 581 nerve and tract at late stages of chronic MOG-induced EAE and were re582 lated to axonal injury (Sun et al., 2007). These combined results are in583 dicative of structural/axonal damage and decrease in tissue integrity as 584 a primary pathological mechanism in MOG-induced EAE, in accord with 585 previous studies that demonstrated axonal degeneration and neuronal 586 loss as a primary pathological process in this model using electron mi587 croscopy and histological methods (Aharoni et al., 2005a, 2011). The 588 patterns obtained in this study for the MTR and the ADC parameters 589 may account for the different clinical manifestations, namely the pro590 gressive chronic course of MOG-induced EAE, in contrast to the occur591 rence of remissions in the PLP-induced model. This is in accordance 592 with findings obtained in clinical studies comparing relapsing–remitting 593 and progressive cases of MS. For instance, newly enhancing lesions and 594 normal appearing white matter in patients with secondary progressive 595 MS had significantly lower MTR values than those with relapsing– 596 remitting MS (Rocca et al., 1999), and the mean lesion ADC in sec597 ondary progressive MS was significantly higher than in relapsing remit598 ting MS (Phuttharak et al., 2006). Furthermore, a number of correlative 599 pathological and MRI studies have recently confirmed the association be600 tween pathological and MRI finding in MS (Filippi et al., 2012). Accord601 ingly, the findings in our study support the notion that MTR and ADC 602 are useful parameters to elucidate the heterogeneous pathological pat603 terns of EAE and MS. 604 It should however be noted, that to a certain extent, the lower 605 MTR and higher ADC values obtained for the MOG model could reflect 606 the higher disease severity, manifested by the MOG-induced mice at 607 all time points, compared to the PLP-induced mice. This is in accor608 dance with studies in MS patients demonstrating that MRI changes 609 (especially MTR) are correlated with reduction in neurological func610 tions such as mobility and cognitive impairment, as well as with the 611 structural CNS damage (Ranjeva et al., 2005; Vigeveno et al., 2012; 612 Zivadinov et al., 2001). 613 The findings of reduced MTR values in both EAE models at early dis614 ease stages and their connection to the myelin damage demonstrated 615 histologically, are in accord with other studies that demonstrated MTR 616 changes in the corpus callosum in a rat EAE model (Serres et al., 2009) 617 and in normal appearing white matter in a guinea pig EAE model 618 (Gareau et al., 2000), which preceded myelin loss and reflected struc619 tural myelin changes. In MS patients as well, reduction in MTR in the 620 normal appearing white matter precedes the appearance of enhanced 621 lesions (Vigeveno et al., 2012). 622 Notably, the voxel based analysis performed in this study for the dif623 ferent MRI parameters depicted in both models multiple regions with 624 significant changes, not only in white matter structures but also in the 625 gray matter, including the thalamus, hippocampus and cortex. These

D

560 561

E

14

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 Q13 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

698 699 700 701 702 703 704 705 706 707 708 709

F

696 697

Q14 710 Uncited reference

Acknowledgment

713

716

This study was supported by a grant from Teva Pharmaceuticals (to RA, MS, RA). We thank Dr. Inbal Biton for performing the MRI measurements, Mrs. Renana From for helping in the immunohistochemical staining and Mrs. Haya Avital for the graphic work.

717

References

T

R

R

E

C

Agosta, F., Rovaris, M., Pagani, E., Sormani, M.P., Comi, G., Filippi, M., 2006. Magnetization transfer MRI metrics predict the accumulation of disability 8 years later in patients with multiple sclerosis. Brain 129, 2620–2627. Aharoni, R., 2010. Immunomodulatory drug treatment in multiple sclerosis. Expert Rev. Neurother. 10, 1423–1436. Aharoni, R., Teitelbaum, D., Leitner, O., Meshorer, A., Sela, M., Arnon, R., 2000. Specific Th2 cells accumulate in the central nervous system of mice protected against experimental autoimmune encephalomyelitis by copolymer 1. Proc. Natl. Acad. Sci. U. S. A. 97, 11472–11477. Aharoni, R., Kayhan, B., Eilam, R., Sela, M., Arnon, R., 2003. Glatiramer acetate-specific T cells in the brain express T helper 2/3 cytokines and brain-derived neurotrophic factor in situ. Proc. Natl. Acad. Sci. U. S. A. 100, 14157–14162. Aharoni, R., Arnon, R., Eilam, R., 2005a. Neurogenesis and neuroprotection induced by peripheral immunomodulatory treatment of experimental autoimmune encephalomyelitis. J. Neurosci. 25, 8217–82128. Aharoni, R., Eilam, R., Domev, H., Labunsky, G., Sela, M., Arnon, R., 2005b. The immunomodulator glatiramer acetate augments the expression of neurotrophic factors in brains of experimental autoimmune encephalomyelitis mice. Proc. Natl. Acad. Sci. U. S. A. 102, 19045–19050. Aharoni, R., Herschkovitz, A., Eilam, R., Blumberg-Hazan, M., Sela, M., Bruck, W., Arnon, R., 2008. Demyelination arrest and remyelination induced by glatiramer acetate treatment of experimental autoimmune encephalomyelitis. Proc. Natl. Acad. Sci. U. S. A. 105, 11358–11363. Aharoni, R., Eilam, R., Stock, A., Vainshtein, A., Shezen, E., Gal, H., Friedman, N., Arnon, R., 2010. Glatiramer acetate reduces Th-17 inflammation and induces regulatory T-cells in the CNS of mice with relapsing–remitting or chronic EAE. J. Neuroimmunol. 225, 100–111. Aharoni, R., Vainshtein, A., Stock, A., Eilam, R., From, R., Shinder, V., Arnon, R., 2011. Distinct pathological patterns in relapsing–remitting and chronic models of experimental autoimmune enchephalomyelitis and the neuroprotective effect of glatiramer acetate. J. Autoimmun. 37, 228–241. Azoulay, D., Vachapova, V., Shihman, B., Miler, A., Karni, A., 2005. Lower brain-derived neurotrophic factor in serum of relapsing remitting MS: reversal by glatiramer acetate. J. Neuroimmunol. 167, 215–218. Basser, P.J., Pierpaoli, C., 1998. A simplified method to measure the diffusion tensor from seven MR images. Magn. Reson. Med. 39, 928–934. Boretius, S., Escher, A., Dallenga, T., Wrzos, C., Tammer, R., Bruck, W., Nessler, S., Frahm, J., Stadelmann, C., 2011. Assessment of lesion pathology in a new animal model of MS by multiparametric MRI and DTI. Neuroimage 59, 2678–2688. Bruck, W., 2008. Evidence for primary neurodegeneration in MS. Mult. Scler. (Suppl. 1), 9–13. Budde, M.D., Kim, J.H., Liang, H.F., Russell, J.H., Cross, A.H., Song, S.K., 2008. Axonal injury detected by in vivo diffusion tensor imaging correlates with neurological disability in a mouse model of multiple sclerosis. NMR Biomed. 21, 589–597.

N C O

718 719 720 721 722 723 724 725 Q15 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761

U

714 715

E

712

P

Fazekas et al., 2002

D

711

Budde, M.D., Xie, M., Cross, A.H., Song, S.K., 2009. Axial diffusivity is the primary correlate of axonal injury in the experimental autoimmune encephalomyelitis spinal cord: a quantitative pixelwise analysis. J. Neurosci. 29, 2805–2813. Carter, N.J., Keating, G.M., 2010. Glatiramer acetate: a review of its use in relapsing– remitting multiple sclerosis and in delaying the onset of clinically definite multiple sclerosis. Drugs 70, 1545–1577. Celis, J.E., Madsen, P., Nielsen, S., Celis, A., 1986. Nuclear patterns of cyclin (PCNA) antigen distribution subdivide S-phase in cultured cells — some applications of PCNA antibodies. Leuk. Res. 10, 237–249. Cercignani, M., Bozzali, M., Iannucci, G., Comi, G., Filippi, M., 2001. Magnetisation transfer ratio and mean diffusivity of normal appearing white and grey matter from patients with multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 70, 311–317. Comi, G., Filippi, M., Wolinsky, J.S., 2001. European/Canadian multicenter double-blind randomized placebo-controlled study of the effects of glatiramer acetate on magnetic resonance imaging measured disease activity and burden in patients with relapsing multiple sclerosis. European/Canadian Glatiramer Acetate Study Group. Ann. Neurol. 49, 290–297. Davies, G.R., Ramio-Torrenta, L., Hadjiprocopis, A., Chard, D.T., Griffin, C.M., Rashid, W., Barker, G.J., Kapoor, R., Thompson, A.J., Miller, D.H., 2004. Evidence for grey matter MTR abnormality in minimally disabled patients with early relapsing–remitting multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 75, 998–1002. DeBoy, C.A., Zhang, J., Dike, S., Shats, I., Jones, M., Reich, D.S., Mori, S., Nguyen, T., Rothstein, B., Miller, R.H., Griffin, J.T., Kerr, D.A., Calabresi, P.A., 2007. High resolution diffusion tensor imaging of axonal damage in focal inflammatory and demyelinating lesions in rat spinal cord. Brain 130, 2199–2210. Dousset, V., Grossman, R.I., Ramer, K.N., Schnall, M.D., Young, L.H., Gonzalez-Scarano, F., Lavi, E., Cohen, J.A., 1992. Experimental allergic encephalomyelitis and multiple sclerosis: lesion characterization with magnetization transfer imaging. Radiology 182, 483–491. Farina, C., Weber, M.S., Meinl, E., Wekerle, H., Hohlfeld, R., 2005. Glatiramer acetate in multiple sclerosis: update on potential mechanisms of action. Lancet Neurol. 4, 567–575. Fazekas, F., Ropele, S., Enzinger, C., Seifert, T., Strasser-Fuchs, S., 2002. Quantitative magnetization transfer imaging of pre-lesional white-matter changes in multiple sclerosis. Mult. Scler. 8, 479–484. Filippi, M., Agosta, F., 2009. Magnetic resonance techniques to quantify tissue damage, tissue repair, and functional cortical reorganization in multiple sclerosis. Prog. Brain Res. 175, 465–482. Filippi, M., Rocca, M., Barkhof, B., Bruck, W., Chen, J.T., Comi, G., DeLuca, G., De Stefano, N., Erickson, B.J., Evangelou, N., Fazekas, F., Geurts, J.J., Lucchinetti, C., Miller, D.H., Pelletier, D., Popescu, B.F., Lassmann, H., 2012. Attendees of the correlation between pathological MRI findings in MS workshop. Association between pathological and MRI findings in multiple sclerosis. Lancet Neurol. 11, 349–360. Gareau, P.J., Rutt, B.K., Karlik, S.J., Mitchell, J.R., 2000. Magnetization transfer and multicomponent T2 relaxation measurements with histopathologic correlation in an experimental model of MS. J. Magn. Reson. Imaging 11, 586–595. Geurts, J.J., Barkhof, F., 2008. Grey matter pathology in multiple sclerosis. Lancet Neurol. 7, 841–851. Kutzelnigg, A., Lassmann, H., 2006. Cortical demyelination in multiple sclerosis: a substrate for cognitive deficits? J. Neurol. Sci. 245, 123–126. Lassmann, H., 2008. Mechanisms of inflammation induced tissue injury in multiple sclerosis. J. Neurol. Sci. 274, 45–47. Lassmann, H., Bruck, W., Lucchinetti, F., 2007. The immunopathology of MS: an overview. Brain Pathol. 17, 210–218. Lessman, V., Gottmann, K., Malcangio, M., 2003. Neurotrophin secretion: current facts and future prospect. Prog. Neurobiol. 69, 341–374. MacKay, A., Laule, C., Vavasour, I., Bjarnason, T., Kolind, S., Madler, B., 2006. Insights into brain microstructure from the T2 distribution. Magn. Reson. Imaging 24, 515–525. Magliozzi, R., Howell, O., Vora, A., Serafini, B., Nicholas, R., Puopolo, M., Reynolds, R., Aloisi, F., 2007. Meningeal B-cell follicles in secondary progressive multiple sclerosis associate with early onset of disease and severe cortical pathology. Brain 130, 1089–1104. Peterson, J.W., Bo, L., Mork, S., Chang, A., Trapp, B.D., 2001. Transected neurites apoptotic neurons and reduced inflammation in cortical multiple sclerosis lesions. Ann. Neurol. 50, 389–400. Phuttharak, W., Galassi, W., Laopaiboon, V., Laopaiboon, M., Hesselink, J.R., 2006. ADC measurements in various patterns of multiple sclerosis lesions. J. Med. Assoc. Thai. 89, 196–204. Polman, C.H., Reingold, S.C., Banwell, B., Clanet, M., Cohen, J.A., Filippi, M., Fujihara, K., Havrdova, E., Hutchinson, M., Kappos, L., Lublin, F.D., Montalban, X., O'Connor, P., Sandberg-Wollheim, M., Thompson, A.J., Waubant, E., Weinshenker, B., Wolinsky, J.S., 2011. Diagnostic criteria for multiple sclerosis: revisions to the McDonald criteria. Ann. Neurol. 292–302. Ramio-Torrenta, L., Sastre-Garriga, J., Ingle, G.T., Davies, G.R., Ameen, V., Miller, D.H., Thompson, A.J., 2006. Abnormalities in normal appearing tissues in early primary progressive multiple sclerosis and their relation to disability: a tissue specific magnetisation transfer study. J. Neurol. Neurosurg. Psychiatry 77, 40–45. Ranjeva, J.P., Audoin, B., Au Duong, M.V., Ibarrola, D., Confort-Gouny, S., Malikova, I., Soulier, E., Viout, P., Ali-Chérif, A., Pelletier, J., Cozzone, P., 2005. Local tissue damage assessed with statistical mapping analysis of brain magnetization transfer ratio: relationship with functional status of patients in the earliest stage of multiple sclerosis. AJNR Am. J. Neuroradiol. 26, 119–127. Rausch, M., Tofts, P., Lervik, P., Walmsley, A., Mir, A., Schubart, A., Seabrook, T., 2009. Characterization of white matter damage in animal models of multiple sclerosis

O

694 695

BDNF, NT3 and NT4 (Aharoni et al., 2005b; Azoulay et al., 2005), which are known to promote axonal growth, remyelination and regeneration (Lessman et al., 2003). In addition, increased neurogenesis manifested by proliferation and differentiation of neuronal progenitor cells and their migration to injury sites was found in GA-treated mice (Aharoni et al., 2005a). We have recently demonstrated enhanced remyelination (Aharoni et al., 2011) as well as proliferation and survival of the oligodendrocyte population (Aharoni et al., 2008) after GA treatment. This is corroborated in the present study by the elevation in NG2 expressing oligodendrocyte progenitor cells, as well as in double positive NG2/PCNA proliferating oligodendrocytes at damage sites, in GA treated in comparison to untreated mice. Notably, whereas the differences between GA-treated and untreated mice for T2-burden were manifested one week after treatment initiation, the effects of GA on the MTR and the ADC parameters were revealed only at the latest time point, suggesting that the anti-inflammatory activity of GA precedes its effect on structural recovery. These consequences of GA treatment provide further support to its long-term therapeutic effect.

R O

692 693

15

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 Q16 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847

by magnetization transfer ratio and quantitative mapping of the apparent bound proton fraction. Mult. Scler. 15, 16–27. Rocca, M.A., Mastronardo, G., Rodegher, M., Comi, G., Filippi, M., 1999. Long-term changes of magnetization transfer-derived measures from patients with relapsing–remitting and secondary progressive multiple sclerosis. AJNR Am. J. Neuroradiol. 20, 821–827. Rovaris, M., Judica, E., Gallo, A., Benedetti, B., Sormani, M.P., Caputo, D., Ghezzi, A., Montanari, E., Bertolotto, A., Mancardi, G., Bergamaschi, R., Martinelli, V., Comi, G., Filippi, M., 2006. Grey matter damage predicts the evolution of primary progressive multiple sclerosis at 5 years. Brain 129, 2628–2634. Serres, S., Anthony, D.C., Anthony, D.C., Jiang, Y., Campbell, S.J., Broom, T.A., Khrapitchev, A., Sibon, N.R., 2009. Comparison of MRI signatures in pattern I and II multiple sclerosis models. NMR Biomed. 22, 1014–1024. Sormani, M.P., Bruzzi, P., Comi, G., Filippi, M., 2005. The distribution of the magnetic resonance imaging response to glatiramer acetate in multiple sclerosis. Mult. Scler. 11, 447–449. Sun, S.W., Liang, H.F., Schmidt, R.E., Cross, A.H., Song, S., 2007. Selective vulnerability of cerebral white matter in a murine model of multiple sclerosis detected using diffusion tensor imaging. Neurobiol. Dis. 28, 30–38. Teitelbaum, D., Meshorer, M., Hirshfeld, T., Arnon, R., Sela, M., 1971. Suppression of experimental allergic encephalomyelitis by a synthetic polypeptide. Eur. J. Immunol. 1, 242–248.

Trapp, B.D., Nave, K.A., 2008. Multiple sclerosis an immune or neurodegenerative disease. Annu. Rev. Neurosci. 31, 247–269. Verhoye, M.R., Gravenmade, E.J., Raman, E.R., Van Reempts, J., Van der Linden, A., 1996. In vivo noninvasive determination of abnormal water diffusion in the rat brain studied in an animal model for multiple sclerosis by diffusion-weighted NMR imaging. Magn. Reson. Imaging 14, 521–532. Vigeveno, M.R., Wiebenga, O.T., Wattjes, M.K., Geurts, J.J.G., Barkhof, F., 2012. Shifting imaging targets in multiple sclerosis: from inflammation to neurodegeneration. J. Magn. Reson. Imaging 36, 1–19. Vrenken, H., Pouwels, P.J., Geurts, J.J., Knol, D.L., Polman, C.H., Barkhof, F., Castelijns, J.A., 2006. Altered diffusion tensor in multiple sclerosis normal-appearing brain tissue: cortical diffusion changes seem related to clinical deterioration. J. Magn. Reson. Imaging 23, 628–636. Whittall, K.P., MacKay, A.L., Graeb, D.A., Nugent, R.A., Li, D.K., Paty, D.W., 1997. In vivo measurement of T2 distributions and water contents in normal human brain. Magn. Reson. Med. 37, 34–43. Wolff, S.D., Balaban, R.S., 1989. Magnetization transfer contrast (MTC) and tissue water proton relaxation in vivo. Magn. Reson. Med. 10, 135–144. Zivadinov, R., De Masi, R., Nasuelli, D., Bragadin, L.M., Ukmar, M., Pozzi-Mucelli, R.S., Grop, A., Cazzato, G., Zorzon, M., 2001. MRI techniques and cognitive impairment in the early phase of relapsing–remitting multiple sclerosis. Neuroradiology 43, 272–278.

F

848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868

R. Aharoni et al. / Experimental Neurology xxx (2012) xxx–xxx

O

16

U

N

C

O

R

R

E

C

T

E

D

P

R O

891

Please cite this article as: Aharoni, R., et al., Magnetic resonance imaging characterization of different experimental autoimmune encephalomyelitis models and the therapeutic effect of glatiramer acetate, Exp. Neurol. (2012), http://dx.doi.org/10.1016/j.expneurol.2012.11.004

869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890