Journal of Neuroimmune Pharmacology

, Volume 7, Issue 1, pp 231–242

[11C]DAC-PET for Noninvasively Monitoring Neuroinflammation and Immunosuppressive Therapy Efficacy in Rat Experimental Autoimmune Encephalomyelitis Model

Authors

  • Lin Xie
    • Division of Radiation Safety and Immune ToleranceNational Research Institute for Child Health and Development
    • Department of Advanced Technology for TransplantationOsaka University Graduate School of Medicine
  • Tomoteru Yamasaki
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
    • Graduate School of Pharmaceutical SciencesTohoku University
  • Naotsugu Ichimaru
    • Department of UrologyOsaka University Graduate School of Medicine
  • Joji Yui
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
  • Kazunori Kawamura
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
  • Katsushi Kumata
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
  • Akiko Hatori
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
  • Norio Nonomura
    • Department of UrologyOsaka University Graduate School of Medicine
    • Department of Molecular Probes, Molecular Imaging CenterNational Institute of Radiological Sciences
    • Division of Radiation Safety and Immune ToleranceNational Research Institute for Child Health and Development
  • Shiro Takahara
    • Department of Advanced Technology for TransplantationOsaka University Graduate School of Medicine
ORIGINAL ARTICLE

DOI: 10.1007/s11481-011-9322-3

Cite this article as:
Xie, L., Yamasaki, T., Ichimaru, N. et al. J Neuroimmune Pharmacol (2012) 7: 231. doi:10.1007/s11481-011-9322-3

Abstract

Neuroimaging measures have potential for monitoring neuroinflammation to guide treatment before the occurrence of significant functional impairment or irreversible neuronal damage in multiple sclerosis (MS). N-Benzyl-N-methyl-2-(7-[11C]methyl-8-oxo-2-phenyl-7,8-dihydro-9H-purin-9-yl) acetamide ([11C]DAC), a new developed positron emission tomography (PET) probe for translocator protein 18 kDa (TSPO), has been adopted to evaluate the neuroinflammation and treatment effects of experimental autoimmune encephalomyelitis (EAE), an animal model of MS. [11C]DAC-PET enabled visualization of neuroinflammation lesion of EAE by tracing TSPO expression in the spinal cords; the maximal uptake value reached in day 11 and 20 EAE rats with profound inflammatory cell infiltration compared with control, day 0 and 60 EAE rats. Biodistribution studies and in vitro autoradiography confirmed these in vivo imaging results. Doubling immunohistochemical studies showed the infiltration and expansion of CD4+ T cells and CD11b+ microglia; CD68+ macrophages were responsible for the increased TSPO levels visualized by [11C]DAC-PET. Furthermore, mRNA level analysis of the cytokines by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) revealed that TSPO+/CD4 T cells, TSPO+ microglia and TSPO+ macrophages in EAE spinal cords were activated and secreted multiple proinflammation cytokines to mediate inflammation lesions of EAE. EAE rats treated with an immunosuppressive agent: 2-amino-2-[2-(4-octylphenyl)ethyl] propane-1,3-diolhydrochloride (FTY720), which exhibited an absence of inflammatory cell infiltrates, displaying a faint radioactive signal compared with the high accumulation of untreated EAE rats. These results indicated that [11C] DAC-PET imaging is a sensitive tool for noninvasively monitoring the neuroinflammation response and evaluating therapeutic interventions in EAE.

Keywords

NeuroinflammationMultiple sclerosisExperimental autoimmune encephalomyelitisPositron emission tomographyTranslocator protein 18 kDaT cellsMicrogliaMacrophage

Introduction

Multiple sclerosis (MS) is a common and often disabling disease of the central nervous system (CNS) (Steinman 2001). The lesions at early phase are characterized by the presence of infiltrated inflammatory cells around venules and small veins. MS remains difficult to diagnose because of its profound heterogeneity in pathomorphology, clinical course and therapeutic response (Martin et al. 2001; Disanto et al. 2010). Trials have shown that an early diagnosis can make a marked difference to the efficacy of MS drug treatments (Hartung 2005; Heesen et al. 2010). The very simplistic view is that inflammation as the primary driving force of the disease process has to be profoundly reconsidered. Experimental evidence strongly supports the concept that inflammation in MS is aimed not only at promoting axonal and neuronal damage but also at inhibiting the regeneration and recovery of oligodendrocytes (McFarland and Martin 2007; Glass et al. 2010). A number of anti-inflammatory therapies, such as corticosteroids, IFN-β and immunosuppression, have been adopted clinically to ameliorate exacerbation and slow the progression of MS (Martin et al. 2001; Rivera 2001). Therefore, the development of a reproducible technique for monitoring the inflammation response in CNS would be helpful not only for the early diagnosis of MS but also for assessing the efficacy of anti-inflammatory treatments in these patients.

An inflammatory reaction involves a marked increase in the expression of a mitochondrial transmembrane protein, translocator protein 18 kDa (TSPO) (also known as peripheral-type benzodiazepine receptor) (Papadopoulos et al. 2006) . TSPO up-regulation is considered a hallmark of neuroinflammation (Rivera 2001). Radiolabeled imaging agents have permitted sensitive detection of TSPO when applied to autoradiographic and positron emission tomography (PET) techniques (Zhang et al. 2007a, b; Yanamoto et al. 2010; Yui et al. 2010). It has been reported that [11C] PK11195, the most commonly used TSPO probe, increased binding in patients with stroke (Chauveau et al. 2009; Thiel et al. 2010) and traumatic brain injury (Toyama et al. 2008; Folkersma et al. 2009), and in patients with chronic neurodegenerative conditions, including Huntington’s disease (Pavese et al. 2006; Tai et al. 2007), Parkinson’s disease (Gerhard et al. 2006; Bartels and Leenders 2007) and MS (Agnello et al. 2000; Banati et al. 2000). Unfortunately, [11C] PK11195 presents some limitations, including a high level of nonspecific binding and poor signal-to-noise ratio (Chauveau et al. 2008), a series of new and alternative radiolabeled ligands were developed to establish PET probes suitable for imaging TSPO in the living CNS. N-Benzyl-N-methyl-2-(7-[11C]methyl-8-oxo-2-phenyl-7,8-dihydro-9H-purin-9-yl) acetamide ([11C] DAC), a novel PET probe, has recently been developed and demonstrated a high signal/noise ratio for TSPO in brain ischemia and unilateral kainic-acid induced striatum lesion rats (Yanamoto et al. 2009; Yui et al. 2010). To clarify whether [11C] DAC enables to noninvasively evaluate neuroinflammation and immunosuppressive treatment efficacy in experimental autoimmune encephalomyelitis (EAE), an animal model reproduces the neuroinflammatory reactions seen in the early stage of human MS, was assessed by serial [11C] DAC-PET imaging.

Materials and methods

Animals

Adult female Lewis rats, weighing 90–110 g, were purchased from Shizuoka Laboratory Animal Center (Shizuoka, Japan), maintained under standard conditions, and fed rodent food and water. The animal experiment protocol was reviewed and approved by the Laboratory Animal Ethics Committee of the National Research Institute for Child Health & Development and National Institute of Radiological Sciences.

Induction of acute EAE and FTY720 treatment protocol

As previously described (Xie et al. 2009), in detail, Lewis rats were immunized in the hind of footpad with 25 μg guinea pig MBP 68–86 (YGSLPQKSQRSQDENPV) emulsified (1:1) in 100 μL complete Freund’s adjuvant (CFA; Difco, Detroit, MI) containing 2 mg/mL heat-inactivated mycobacterium tuberculosis (strain H37RA; Difco). The total dose of MBP 68–86 peptide was 50 μg/rat. CFA-only immunized rats were used as a control group. Clinical scores of EAE were evaluated for clinical signs in a blinded fashion by at least two investigators according to the following criteria (Fujino et al. 2003; Quintana et al. 2008): grade 0: no clinical signs; grade 0.5: partial loss of tail tonicity; grade 1: complete loss of tail tonicity; grade 2: flaccid tail and abnormal gait; grade 3: hind leg paralysis; grade 4: hind leg paralysis with hind body paralysis; grade 5: hind and foreleg paralysis; grade 6: death. For immunosuppressive therapy, rats were orally administered 2-amino-2- [2-(4-octylphenyl) ethyl] propane-1,3-diolhydrochloride (FTY720; a gift from Mitsubishi Tanabe Pharma, Osaka, Japan) at 1 mg/kg daily dissolved in physiological saline from day 0 to 14 after immunization with MBP 68–86 peptide/CFA. Rats in the untreated group received an equal volume of physiological saline solution.

Radiochemistry

The probe [11C] DAC was synthesized as previously described (Yamasaki et al. 2010). Briefly, 11C was produced by 14N (p, α) 11C nuclear reaction using a Cypris HM18 cyclotron (Sumitomo Heavy Industries, Tokyo, Japan). [11C] CH3I for radiosynthesis was synthesized from cyclotron-produced [11C] CO2. Starting from [11C] CO2 of 12–15 GBq, [11C] DAC of 700–1,380 MBq (n = 25) was obtained with >98% radiochemical purity and 105 ± 44 GBq/μmol specific activity at the end of synthesis.

PET imaging

PET scans were performed 0 (n = 8), 7 (n = 8), 11 (n = 16), 20 (n = 9), and 60 (n = 8) days after immunization with MBP/CFA or CFA only (n = 5, respectively), as well as for FTY720 treated or untreated day 11 EAE rats (n = 8, respectively), using a small animal PET scanner, Inveon (Siemens Medical Solutions USA, Knoxville, TN), which provides 159 transaxial slices 0.796 mm (center-to-center) apart, a 10 cm transaxial field of view (FOV), and a 12.7 cm axial FOV. Rats were anesthetized with 5% (v/v) isoflurane, and maintained thereafter by 1-2% (v/v) isoflurane. Body temperature was monitored by a rectal probe (RET-3; Neuroscience, Tokyo, Japan) and maintained at 37 ± 0.5°C by a heating pump and pad (T/Pump TP401; Gaymar Industries, Orchard Park, NY). To inject [11C]DAC, a 29-gauge needle with 12–15 cm PE 10 tubing was inserted into the tail vein. A bolus of about 18 MBq [11C] DAC in 200 μL saline was injected through the tail vein catheter. A dynamic emission scan in 3D list mode was performed for 30 min after [11C] DAC injection. PET dynamic images were reconstructed with filtered back-projection using a Hanning’s filter with a Nyquist cutoff of 0.5 cycles/pixel. PET images were reconstructed using ASIPro VMTM software, Analysis Tools and System Setup/Diagnostics Tool (Siemens Medical Solutions). Radioactivity was decay corrected for injection time and expressed as the standardized uptake value (SUV), normalized for injected radioactivity and body weight. SUV was calculated as (radioactivity per cm3 tissue/injected radioactivity) x g body weight.

[11C]DAC biodistribution study

Rats (n = 4, respectively) from each designated day were sacrificed by cervical dislocation after all PET scans were finished. Thoracic, lumbar, and sacral cords were removed quickly. Radioactivity in these tissues was measured using a 1480 Wizard autogamma scintillation counter (PerkinElmer, Waltham, MA) and is expressed as the percentage of the injected dose per gram of wet tissue (% ID/g). All radioactivity measurements were corrected for decay.

In vitro autoradiography

Day 0 naïve and day 11 EAE rats (n = 3) were sacrificed under ether anesthesia. Thoracic, lumbar, and sacral cords were harvested immediately and embedded in OCT compound (Sakura Finetek USA, Torrance, CA) and frozen in hexane (Wako Pure Chemical Industries, Osaka, Japan). Spinal sections (5 μm) were prepared using a cryotome, MICROM HM560 (Carl Zeiss, Jena, Germany), at a temperature of −20°C and mounted on Matsunami adhesive silane-coated glass slides (Matsunami Glass Industries, Kyoto, Japan). In accordance with the established procedure (Yui et al. 2011), the sections were pre-incubated in 50 mM Tris buffer (pH 7.4) at room temperature for 20 min and then incubated in the same buffer containing [11C] DAC (18 MBq/L) at room temperature for 30 min. After incubation, the sections were washed twice for 2 min each in 50 mM cold fresh Tris-HCl buffer and 10 s in distilled water. They were then dried with a warm air current and placed in contact with an imaging plate, BAS-MS 2325 (Fujifilm, Tokyo, Japan), for 60 min and analyzed using a Bio Imaging Analyzer System, BAS 5000 (Fujifilm). The radioactivity in the spinal sections was quantified and expressed as photo-stimulated luminescence per unit area (PSL/mm2). A ligand displacement study was performed by adding nonradioactive PK11195 (10 μM) to the reaction system. The rest of the protocol was as described above.

Histological analysis

Three rats from each designated day were sacrificed under ether anesthesia. The spinal cords were removed quickly and a portion was fixed in 10% neutral -buffered formalin and then cut into 5 μm thick sections. Tissues were stained with hematoxylin and eosin to assess tissue damage and inflammation.

Immunohistochemical analysis

To identify the cell types expressing TSPO, double immunostaining of the spinal cords was performed. Five micrometer-thick frozen sections of spinal cords were cut, stained with mouse-anti-rat CD4 (OX-38; Serotec, Oxford, UK, 1:100) /mouse-anti-rat CD8 (OX-8; Serotec, 1:100)/mouse-anti-rat CD11b (OX42; Serotec, 1:100)/mouse-anti-rat CD68 (ED-1; Serotec, 1:100), and then incubated with alkaline phosphatase (ALP)-conjugated anti-mouse Ig (Sigma-Aldrich, St Louis, MO, 1:100) and developed with Vector Blue (Vector Laboratories, Burlingame, CA). Thereafter, samples were treated with 0.2% Triton x-100 (Sigma-Aldrich) in PBS for 15 min, stained with rabbit-anti TSPO/PBR (NP155; a gift from Dr. M. Higuchi, National Institute of Radiological Sciences, 1:200), incubated with an ALP-labeled sheep F(ab’)2 to rabbit Ig (R&D Systems, Minneapolis, MN, 1:100) and developed with a Vector Red substrate kit (Vector Laboratories).

RNA preparation and quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR)

As described previously (Xie et al. 2009; Chen et al. 2011), the total tissue RNA was extracted from frozen spinal cords using Sepasol-RNA I Super (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s protocol. The quality of total RNA was checked by the ratio of 260/280 nm. Quantitative RT-PCR was performed using the TaqMan system on the Applied Biosystems PRISM 7900 (ABI Japan, Tokyo, Japan). The target-specific primers and probes were purchased from Applied Biosystems (ABI Japan).

Statistical analysis

All data are presented as the means ± SEM. Comparison data among designated days, naïve and corresponding control rats were performed using one-way ANOVA followed by Dunnett’s test. P < 0.05 was considered significant.

Results

Clinical characteristics of EAE and designated days of [11C] DAC-PET imaging

In Lewis rats, acute EAE was induced via a single administration of MBP 68–86 emulsified in CFA. The clinical severity of EAE was monitored from day 0 to day 60 after immunization. All immunized rats developed typical clinical signs of acute EAE. The first clinical signs occurred after 8–9 days, progressing to paraparesis, with severity peaking at day 11–12, showing a mean maximum clinical score of 3.73 ± 0.11. After EAE symptoms, persisting for about 9 days, the challenged rats developed a self-limiting disease and had recovered completely by 16 days post-immunization (Fig. 1a). To determine whether the quantification of TSPO in the spinal cord using [11C] DAC can be correlated with clinical symptoms or pathologic damage by EAE, rats were serially imaged by a small animal PET scanner at day 0 as naive control, day 7, that is, 1 day before the acute paralytic attack, day 11 as the peak of clinical severity of EAE, day 20 as clinical remission and day 60 as recovery from the disease in the course of EAE. The mean clinical score and designate days of PET imaging are shown in Fig. 1a.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig1_HTML.gif
Fig. 1

Clinical evaluation and designated days of PET imaging for EAE. (A) Mean maximum clinical scores for Lewis rats immunized with MBP 68–86 peptide/CFA or CFA only. Rats were serially imaged using [11C]DAC-PET on day 0 (n = 8), 7 (n = 8), 11 (n = 16), 20 (n = 9), and 60 (n = 8) after immunization with MBP/CFA or CFA only (n = 5). (B) Representative summed images. These images were generated by summed whole scan duration of 0–30 min after [11C]DAC injection. [11C]DAC-PET signal increased significantly in day 7, 11 and 20 EAE spinal cords

Induction of the EAE model requires the use of a strong adjuvant, CFA. It is conceivable that CFA-induced activation of innate immune pathways facilitates the access of activated T cells to the CNS by helping them to cross the blood–brain barrier. To determine whether we were really imaging autoimmunity rather than nonspecific adjuvant-driven activation of the innate immune system, as experimental controls, we also scanned rats that were immunized with CFA adjuvant without MBP peptides at each designated day. Also, all the rats immunized with CFA only were free of disease (Fig. 1a).

Higher accumulation of radioactivity was seen in EAE spinal cords by [11C] DAC-PET imaging

To determine the pathological changes in EAE rats, rats were scanned 0, 7, 11, 20 and 60 days after immunization. Dynamic emission scans were performed for 30 min after [11C] DAC injection. Figure 1b showed representative summed images of whole scans (0–30 min). A significant increase in radioactivity was observed in day 11 EAE rats with the peak clinical score, an increase in radioactivity was also detected in day 7 and 20 rats with the beginning and remission of EAE symptoms, whereas day 60 EAE rats and all control rats immunized with CFA only exhibited faint and relatively homogeneous distribution of radioactivity, similar to day 0 naïve rats. And we also determined the SUVs in the spinal cords at 30 min after [11C] DAC injection, SUVs for day 11 EAE were highest (0.726 ± 0.054, p ≤ 0.05). High SUVs were indicated at day 7 (0.623 ± 0.098) and 20 EAE (0.549 ± 0.074) rats compared to day 0 (0.432 ± 0.057) and corresponding control rats (0.462 ± 0.015). These results suggested that the alterations in radioactivity seen on the PET images reflected a specific autoimmunity activation and appeared to be correlated with the clinical process of EAE.

Higher radioactivity was determined in the spinal cords of EAE rats

To further pinpoint the anatomical location of [11C]DAC uptake in EAE rats, the radioactivity in all three major spinal sections (thoracic, lumbar and sacral cord) was measured after PET scans. Consistent with the findings of PET imaging, a high concentration of radioactivity (decay-corrected %ID/g) was shown in spinal column segments of day 11 and 20 EAE rats. However, different from the PET images, the radioactivity (%ID/g) was not up-regulated in day 7 EAE spinal cords compared to day 0, 60 EAE rats and control rats, which showed no obvious changes in the accumulation of radioactivity (Fig. 2). The highest radioactivity was found in the EAE sacral cords of day 11 (0.74 ± 0.11%ID/g, p ≤ 0.01) and day 20 (0.72 ± 0.13%ID/g, p ≤ 0.01) rats, followed by the lumbar cords of day 11 (0.68 ± 0.11%ID/g, p ≤ 0.01) and day 20 (0.65 ± 0.1%ID/g, p ≤ 0.01) rats, and the thoracic cord of day 11 (0.53 ± 0.07%ID/g, p ≤ 0.01) and day 20 (0.56 ± 0.09%ID/g, p ≤ 0.05) rats. The results confirmed the findings of [11C] DAC-PET imaging, and the uptake of radioactivity appeared to reflect the ascending nature of this inflammatory process in day 11 and 20 EAE spinal cords: sacral cord>lumbar cord>thoracic cord.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig2_HTML.gif
Fig. 2

Biodistribution of [11C]DAC in three spinal column segments at day 0, 7, 11, 20, 60 after immunization. Radioactivity in the thoracic, lumbar and sacral cords was measured after [11C]DAC-PET scanning. High concentration of radioactivity was observed in the spinal cords of day 11 and 20 EAE rats compared with control and day 0, 7, 60 EAE rats. Numbers of rats per data point derived from 4 animals respectively. Significant difference was calculated using one-way ANOVA, * p < 0.05. ** p < 0.01

Radioactivity in EAE spinal cords was specific to TSPO

To confirm the PET results, in vitro autoradiography studies were performed. Thoracic, lumbar, and sacral cords of day 0 and day 11 EAE rats were harvested and the spinal sections were prepared and incubated with [11C] DAC (18 MBq/L). Autoradiography images showed preferential localization of the radioactivity in the spinal cords of day 11 EAE rats (peak of the disease) compared to day 0 rats (naïve) (Fig. 3). Consistent with the biodistribution results, radioactivity accumulation was higher in the sacral cords than in the lumbar and thoracic cords in the day 11 EAE rats.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig3_HTML.gif
Fig. 3

In vitro autoradiographic images of [11C]DAC distribution in thoracic, lumbar and sacral cords of day 0 and 11 EAE rats. Longitudinal sections of thoracic (a), lumbar (b), and sacral cords (c) of day 0 and day 11 EAE rats (n = 3) were incubated with [11C]DAC (18 MBq/L). To confirm the specific binding of [11C]DAC for TSPO, the three spinal sections of day 11 EAE rats were co-incubated with nonradioactive PK11195 (10 μM) and [11C]DAC (18 MBq/L)

To confirm that [11C] DAC binding in EAE spinal cords was specific to TSPO, a competition binding experiment was performed in the three spinal sections of day 11 EAE rats. The sections were co-incubated with nonradioactive PK11195 (10 μM) and [11C] DAC (18 MBq/L). Radioactivity in the spinal cords showed a marked decrease in the signal by PK11195 displacement (Fig. 3); the highest displacement of the signal by PK11195 was found in the sacral section, suggesting the specificity of [11C] DAC binding for TSPO in the spinal sections.

Pathological features of EAE

To explore the correlation between pathological changes and accumulation of radioactivity in EAE rats, we examined the morphologic changes of spinal cords at all designated days. At the peak of clinical severity, H&E staining showed that day 11 EAE rats had widespread inflammatory cell infiltration and multifocal perivascular cuffs in the spinal cords (Fig. 4d), while high uptake of [11C] DAC was also found at this point (Figs. 1b, and 2). EAE entering day 20, which was recognized as clinical remission (Fig. 1a), still had multiple inflammatory lesions and perivascular cuffs in the spinal cords that were not coincident with the clinical manifestation (Fig. 4e), while high accumulation of [11C] DAC was also shown in day 20 EAE spinal cords (Figs. 1b, and 2). At day 0 (Fig. 4b), day 7 (Fig. 4c), and day 60 (Fig. 4f) EAE and control rats (Fig. 4a) showed no obvious inflammation or pathological damage to the spinal cord. Consistent with the morphologic results, day 0, 7, 60 EAE and control rats showed faint and relatively homogeneous uptake of radioactivity (Fig. 2). The results suggested that the inflammatory lesions were highly concurrent with the up-regulation of [11C] DAC accumulation in EAE spinal cords.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig4_HTML.gif
Fig. 4

Immunopathological findings in the EAE spinal cords. H&E staining (af) and double Immunostaining including TSPO (red) and CD4 for infiltrating CD4 T cells (gl)/CD11b for microglia (N-R)/CD68 for macrophages (sx) (blue) were performed at the designated days (n = 3). (b, h, m, t) TSPO+ cells were minimal without any inflammatory lesions in day 0 naïve spinal cords. In control, day 7 and 60 EAE rats, only a few infiltrating T cells (g, i, l), microglia (n, q, r) and macrophages (s, u, x) were found, and total numbers of TSPO+ cells didn’t appear to be markedly up-regulated compared to day 0 naïve rats. In day 11 spinal cords, (d) widespread inflammatory lesions and multifocal perivascular cuffs were revealed. Inflammatory foci consisted mainly of CD4+/TSPO+ T lymphocytes (j), CD11b+/TSPO+ microglia (p) and CD68+/TSPO+ macrophages (v). Day 20 rats still had some inflammatory lesions and perivascular cuffs in spinal cords (e), the CD4+/TSPO+ T lymphocytes including CD11b+/TSPO+ microglia and CD68+/TSPO+ macrophages extensively remained (k, q, w). Scale bars = 200 μm

Overexpression of TSPO correlated with increased neuroinflammatory cells in EAE spinal cords

CD4 T lymphocytes and activated microglia/macrophages represent most of the inflammatory cells in EAE (Steinman 2001; McFarland and Martin 2007). To confirm TSPO+cell identity, double immunostaining with multiple markers was employed at each designated day for EAE and control rats. CD4 mAb stained for CD4+ T lymphocytes, CD11b/OX-42 mAb for microglia, and CD68/ED-1 mAb for macrophages. On day 0, although no CD4+, CD11b+, or CD68+ cells were observed by double immunostaining, minimal TSPO+cells were present (Fig. 4h, m, t). On day 7 after immunization, a few inflammatory CD4+ T cells, CD11b+ microglia, and CD68+ macrophages were found, and these cells were immunopositive for TSPO (Fig. 4i, q, u), but the total numbers of TSPO+ cells didn’t show a significant increase over day 0 rats. In day 11 EAE spinal cords, the inflammation reaction was severe (Fig. 4d), and [11C] DAC showed high accumulation (Figs. 1b, and 2), and severely enhanced TSPO positivity was also found in CD11b+microglia, CD68+ macrophages and infiltrating CD4+ T lymphocytes (Fig. 4j, p, v). Immunopositive cells were localized mainly in inflammatory and perivascular cuffs. Although the clinical signs had disappeared in day 20 EAE rats, the CD4+/TSPO+ T lymphocytes, CD11b+/TSPO+ microglia and CD68+/TSPO+ macrophages extensively remained (Fig. 4k, q, w), whereas the day 60 EAE rats, similar to the day 7 and control rats, had just a few CD11b+/TSPO+ microglia, CD68+/TSPO+ macrophages and CD4+/TSPO+ infiltration T cells in the spinal cords, and TSPO expression was very low. Otherwise, little or no expression was observed in CD8+ T cells during EAE progression (data not shown). The results suggested that TSPO upregulation was closely correlated with the infiltration and expansion of CD4+ T cells, microglia and macrophages, which were the main source of [11C] DAC signals in EAE spinal cords.

TSPO+/CD4+ T cells and TSPO+ microglia/macraphages initiated the inflammatory cascade in EAE rats

Infiltrating CD4+ T cells develop into four major subtypes of cells, known as Th1, Th2, Th17 and regulatory T (Treg) cells (Dong 2008; Ouyang et al. 2008). The subtypes migrate into the CNS and secrete inflammatory cytokines, while these inflammatory cytokines have crucial roles in initiating and perpetuating CNS inflammation in EAE/MS. To determine the events potentially involved in CD4+/ TSPO+ T cells and TSPO+ microglia/macraphages in EAE rats, genetic profiling of proinflammatory cytokines, such as IFN-γ (secreting by Th1 cells), IL-17 (by Th17 cells), IL-6 and IL-10 (by Th2 cells), TGF-β and IL-10 (by Foxp3+ Treg cells), iNOS and osteopontin (by activated microglia/macrophages), was performed by quantitative RT-PCR analysis. We found a parallel marked increase in IL-6 (mean 12.8-fold), TGF-β (13.4-fold), Foxp3 (33.6-fold), iNOS (438.1-fold), osteopontin (7.5-fold) mRNA levels in day 11 EAE spinal cords, and IL-6 (1.8-fold), TGF-β (2-fold), Foxp3 (14.3-fold), iNOS (13.8-fold), and osteopontin (2.6-fold) in day 20 EAE spinal cords compared to day 0 naïve rats (Fig. 5), whereas in control, day 7 and day 60 rats, the levels of proinflammation cytokines were almost identical to day 0 normal states, except for Foxp3 (8.2-fold) in day 60 EAE spinal cords. IFN-γ, IL-17 and IL-10, which are considered the most important immunological contributors in EAE/MS, were not detectable by qRT-PCR when EAE signs were absent in day 0, 7 and control spinal cords (Fig. 5). The mRNA expression levels in day 11 rats were up-regulated markedly in IFN-γ (95.4-fold), IL-17 (147.3-fold), and IL-10 (44.4-fold) than in day 60 rats, in which inflammatory lesions had almost resolved by immunoregulatory mechanisms. In day 20 spinal cords, with self-limitation of EAE symptoms, the cytokines were quickly down-regulated, but a high expression of proinflammation cytokines was still found with IFN-γ (5.5-fold), IL-17 (5.9-fold), and IL-10 (7.1-fold) (Fig. 5) compared to day 60 rats. The results indicated that TSPO+/CD4 T cells and TSPO+ microglia/macrophages in EAE spinal cords were activated and secreted multiple proinflammation cytokines to mediate inflammation lesions of EAE.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig5_HTML.gif
Fig. 5

mRNA expression of inflammatory cytokines in the EAE spinal cords (n = 3). The expression levels of IFN-γ (a), IL-17 (b), IL-6 (c), iNOS (d), osteopontin (e), Foxp3 (f), TGF-β (g), and IL-10 (h) were quantified by real-time RT-PCR. Higher mRNA levels of proinflammation cytokines were indicated in day 11 and 20 EAE rats compared with control and 0, 7, 60 day rats

[11C] DAC-PET identified amelioration of EAE due to FTY720 immunosuppressive treatment

To further investigate the potential use of [11C] DAC-PET in EAE, we sought to determine whether this technique could be used to monitor the effects of systemic immunosuppressive therapies known to block the development of neurological manifestations in this animal model. FTY720 administration almost entirely prevented the development of EAE in the experiment (Fig. 6a), and the spinal cords of rats administered FTY720 exhibited a complete absence of inflammatory cell infiltrates compared with the severe inflammatory lesions in untreated rats (Fig. 6b–d). Rats were scanned on day 11 after immunization with or without FTY720 (Fig. 6a). Uptake of [11C] DAC was clearly increased only in day 11 rats that were not treated with FTY720 compared with treated and naïve rats (Fig. 6e). Consistent with PET imaging, high uptake of [11C] DAC was confirmed in various spinal column segments of untreated rats; however, [11C] DAC uptake in FTY720-treated rats was faint and homogeneous, similar to naïve rats (Fig. 6f).
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-011-9322-3/MediaObjects/11481_2011_9322_Fig6_HTML.gif
Fig. 6

Reduced uptake of [11C]DAC in EAE by FTY720 treatment. a Mean maximum clinical scores for FTY720 treated or untreated day 11 EAE rats (n = 8). FTY720 administration almost entirely prevented EAE clinical symptoms. b-d Pathological findings. H&E staining of day 11 spinal sections indicated that rats with FTY720 treatment exhibited a complete absence of inflammatory cell infiltrates compared with severe inflammatory lesions in untreated rats (n = 3). e Representative summed images. [11C]DAC-PET scans were performed for 11 day EAE rats treated or untreated with FTY720 (n = 8, respectively). Uptake of [11C]DAC was clearly increased only in day 11 untreated rats, whereas uptake in FTY720-treated rats remained low and unchanged, similar to naïve rats. f Distribution of radioactivity in thoracic, lumbar and sacral cords (n = 4). Compared with the high concentration of [11C]DAC radioactivity in untreated EAE rats, [11C]DAC uptake in FTY720-treated rats was similar to naïve rats in the thoracic, lumbar and sacral cords. Significant difference was calculated using one-way ANOVA, * p < 0.05. ** p < 0.01. Scale bars = 200 μm

Discussion

Neuropathological studies indicate that neuroinflammatory responses begin prior to significant degeneration and loss of neuronal populations in the progression of MS (Goertsches et al. 2006; Gold et al. 2006). The aim of this study is to determine whether [11C] DAC-PET may be useful for monitoring neuroinflammation in EAE. Increased [11C] DAC radioactivity in day 11 and 20 EAE rats with comprehensive inflammation lesions was indicated by PET images, and upregulation of TSPO expression visualized by [11C] DAC-PET was associated with the infiltration and expansion of CD4+ T cells, CD11b+ microglia, CD68+ macrophages in EAE spinal cords. In addition, the uptake of [11C] DAC in EAE rats treated with FTY720 remained low and stable during treatments. These results indicated that [11C] DAC-PET could be used to noninvasively diagnose neuroinflammation and to monitor therapeutic effectiveness in EAE.

The diagnostic criteria of MS have evolved from being solely clinical symptom-based to integrating evoked potentials, spinal fluid analysis and magnetic resonance imaging assessments (Fox and Cohen 2001; Kalkers et al. 2002); however, MS is still very difficult to diagnose and evaluate because of its profound heterogeneity in pathomorphology, clinical course and therapeutic response (McFarland and Martin 2007; Disanto et al. 2010). Abundant evidence strongly supports that neuroinflammation lesions occur early during the disease course and that they are prominent from the onset of MS (McFarland and Martin 2007; Glass et al. 2010). PET imaging, which is sensitive to changes at the cellular or molecular level, may bring a big breakthrough in the early diagnosis of MS. TSPO is a ubiquitous 18 kDa transmembrane protein that participates in diverse cell functions, including cholesterol transport, mitochondrial respiration, chemotaxis and cellular immunity, cell proliferation and apoptosis (Chen et al. 2004). Radiolabeled imaging agents targeting TSPO have permitted the sensitive detection of inflammation associated with a variety of brain insults by PET (Venneti et al. 2006; Ji et al. 2008; van der Laken et al. 2008). [11C] PK11195 was the first to enable PET measurement of TSPO in diverse CNS pathologies, and other radiolabeled ligands were developed, such as [18F] FE-DAA1106 and [11C]AC-5216, to establish PET probes suitable for imaging TSPO (Fujimura et al. 2006; Zhang et al. 2007a, b). EAE in Lewis rats is a reliable animal model to study the initiating events of MS as it mimics neuroinflammation progression, a critical step in the development of MS. In the present study, [11C] DAC, a novel analog derivative of AC-5216 for TSPO, was evaluated in the course of EAE (Figs. 1, and 2). Imaging results revealed that [11C] DAC radioactivity was highly accumulated in day 11 and 20 EAE spinal cords with severe inflammation lesions, whereas day 60 and control rats only exhibited a faint and relatively homogeneous distribution of radioactivity, similar to day 0 naïve rats. The results are consistent with 3H(R)-PK11195 uptake in brain tissues with MS and adoptive transfer EAE models (Banati et al. 2000). EAE in Lewis rats manifests itself as an ascending paralysis that is most severe in the sacral cord and progresses up the thoracic cord with decreasing intensity. The uptake of [11C] DAC reflected this change of EAE manifestation; that is, the ranking order of [11C] DAC radioactivity was the sacral cord>lumbar cord>thoracic cord. Autoradiography also confirmed the in vivo data. The specificity of [11C] DAC binding was demonstrated by co-treatment of day 11 EAE spinal sections with non-radioactive PK11195 (Fig. 3). This co-incubation completely blocked the radioactivity of [11C] DAC in the severely affected thoracic, lumbar, and sacral cords in day 11 EAE rats. The distribution pattern of [11C] DAC was in agreement with TSPO distribution in the spinal cords of EAE.

Pharmacological probes that bind TSPO have been used extensively to study receptor-binding parameters in brain tissues (Pavese et al. 2006; Venneti et al. 2006; Toyama et al. 2008). It is necessary to determine the cell types responsible for increased probe binding. In the majority of reports, cellular localization of increased TSPO expression is specific to activated microglia elements (Cagnin et al. 2001; Debruyne et al. 2003; Wilms et al. 2003; Chen et al. 2004; James et al. 2006; Venneti et al. 2007). We differentiated the cell types expressing TSPO during EAE progression by double immunostaining (Fig. 4). The results showed that not only CD11b+ cells representing microglia but also all CD4+ cells representing pathological CD4+ T cells and CD68+ cells representing macrophages were responsible for increased TSPO expression in the inflammatory lesions of day 11 and 20 EAE spinal cords. Consistent with previous reports (Chen and Guilarte 2008), under normal physiological conditions TSPO levels are low, and low levels were also revealed in control rats and day 7 and 60 EAE rats that showed an absence of inflammation lesions.

Some studies have indicated that TSPO plays a role in modulating the activation of microglia (Rey et al. 2000; Trincavelli et al. 2002). Treating primary human embryonic microglia with PK11195 decreases the expression of COX2, TNF-α and intracellular calcium levels (Choi et al. 2002). Furthermore, PK11195 decreases microglia activation, iNOS, IL-1β, IL-6, TNF-α levels and neuronal damages in quinolinic acid-injected rats (Agnello et al. 2000). CD4+ T cell phenotypes, including Th1, Th2, Th17, Treg, and microglia and macrophages all had TSPO+ expression, playing important roles in the induction of disease and MS lesion pathogenesis through the secretion of proinflammatory cytokines. We quantified the gene expression of proinflammatory cytokines (Fig. 5) in the course of EAE by qRT-PCR, such as IFN-γ (secretion by Th1 cells), IL-17 (by Th17 cells), IL-6 and IL-10 (by Th2 cells), TGF-β and IL-10 (by Foxp3+ Treg cells), iNOS and osteopontin (by activated microglia and macrophages) (Dong 2008; El-behi et al. 2010). Gene expressions of IL-6, TGF-β, iNOS, and osteopontin showed a parallel increase in day 11 and 20 EAE spinal cords compared with day 0 naïve rats, whereas in control, day 7 and day 60 rats, levels of proinflammation cytokines were almost identical to day 0 normal states. In particular, IFN-γ, IL-17 and IL-10, which are considered the most important immunological contributors to EAE/MS (Stromnes et al. 2008; El-behi et al. 2010), were not detectable by qRT-PCR in 0, 7 days and control spinal cords. Their mRNA levels were up-regulated markedly in IFN-γ, IL-17, IL-10 in day 11 and 20 spinal cords compared to day 60 rats, although the functional consequences of increased TSPO expression in various cell types are unknown. Our data strongly supported that the neuroinflammation cascade was highly concurrent with TSPO up-regulation in EAE; therefore TSPO can be a biomarker for monitoring neuroinflammatory lesions in EAE.

Was simple inspection of the images adequate for identifying the medical treatment response in EAE? To answer this, FTY720, an immunosuppressive agent known to exert pleiotropic and potent immunosuppressive effects on T cells, treatment was given to EAE rats from day 0 to day 14 after immunization. As previously reported (Fujino et al. 2003; Chun and Hartung 2010; Papadopoulos et al. 2010), FTY720 administration almost entirely prevented the development of EAE and exhibited a complete absence of inflammatory cell infiltrates compared with the severe inflammatory lesions in untreated rats. Radioactivity of [11C] DAC was clearly increased only in day 11 untreated rats, whereas FTY720-treated rats showed a low level of radioactivity, similar to naïve rats (Fig. 6). These results together indicate that [11C] DAC-PET imaging could be used to monitor the therapeutic efficacy of EAE in response to anti-inflammatory treatments.

Some potential problems of [11C] DAC-PET imaging in EAE need to be considered for future applications: PET images and their SUVs in day 7 EAE rats were highly accumulated in radioactivity compared to day 0 naïve and control rats, but the distribution of radioactivity was exhibited a faint and relatively homogeneous level, and pathological studies and mRNA level analysis confirmed inflammation lesions were absent in day 7 spinal cords, similar to day 0 naïve and control rats. The cause of this discrepancy is unclear. Although this imaging can detect neuroinflammation changes in EAE spinal cords, we were unable to correlate [11C] DAC imaging with disease severity as represented by the clinical score (Fig. 1). In addition, the threshold of inflammation that [11C] DAC-PET imaging can detect is still unknown. Although this study was not designed to address that question, we surmise from day 20 EAE data that [11C] DAC-PET imaging is sensitive enough to detect changes even in these low levels of induced inflammation. High [11C] DAC accumulation in CNS could indicate the need to explore the nature of the abnormality and proceed to appropriate treatment. Furthermore, monitoring treatment efficacy by [11C] DAC-PET may yield important information for the management of MS patients.

Acknowledgements

The authors are grateful to Dr. M. Higuchi for the gift of antibody NP155 and Dr. H. Kimura for his critical comments and useful suggestions. We also thank the staff of the National Institute of Radiological Sciences for support with the cyclotron operation, radioisotope production, radiosynthesis, and animal experiments. This study was supported by research grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Grants-in-Aid 20390349, 21659310), and in part by the Japan China Medical Association.

Conflicts of interest

The authors declare no competing financial interests.

Copyright information

© Springer Science+Business Media, LLC 2011