- Email: [email protected]
Dopaminergic imaging of nonmotor manifestations in a rat model of Parkinson's disease by fMRI
Neurobiology of Disease 49 (2013) 99–106
Contents lists available at SciVerse ScienceDirect
Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi
Dopaminergic imaging of nonmotor manifestations in a rat model of Parkinson's disease by fMRI Chiao-Chi V. Chen a, b, Yen-Yu I. Shih c, d, e,⁎, Chen Chang a, b,⁎⁎ a
Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan Functional and Micro-magnetic Resonance Imaging Center, Academic Sinica, Taipei, Taiwan Department of Neurology, University of North Carolina at Chapel Hill, NC, USA d Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, NC, USA e Department of Biomedical Engineering, University of North Carolina at Chapel Hill, NC, USA b c
a r t i c l e
i n f o
Article history: Received 4 May 2012 Revised 4 July 2012 Accepted 20 July 2012 Available online 27 July 2012 Keywords: Parkinson disease Non-motor Biomarker BOLD Vasoconstriction fMRI Neuroimaging In vivo
a b s t r a c t Nonmotor manifestations determine the life quality of patients with Parkinson's disease (PD). Identification of the nonmotor symptoms in PD as definite changes will represent a milestone in its diagnosis. Outcome measures that characterize nonmotor manifestations with specificity for dopaminergic deficiency are essential to that goal. Pain is a prevalent sensory disturbance in PD patients. The prevalence was reported to be up to 83%. Nociceptive stimuli under normal conditions elicit decreases in cerebral blood volume (CBV) in the striatum via dopaminergic neurotransmission. This nociception-induced CBV response is potentially to be defined as a characteristic of the pain symptom of PD. To validate this concept, steady-state CBV-weighted functional magnetic resonance imaging with iron oxide nanoparticles was employed to measure CBV changes in parkinsonian rats. Tyrosine hydroxylase immunohistology was used to identify the dopaminergic integrity to corroborate the imaging findings. Additional experiments that tested pain responses in parkinsonism were also carried out. The results revealed that the lesioned striatum exhibited a weakened CBV decrease in response to the nociceptive stimulus. This weakened CBV response occurred mainly in areas with dopaminergic denervation. A strong correspondence was observed between the distributions of the nociception-induced CBV responses and dopaminergic innervation. The persisting CBV signals in the striatum were abolished by the D2/D3 antagonist eticlopride. The findings of these behavioral, neuroimaging, immunohistological, and pharmacological experiments demonstrate that pain in a rat model of PD can be characterized by nociception induced striatal CBV signal changes with specificity for dopaminergic dysfunction. © 2012 Elsevier Inc. All rights reserved.
Introduction Parkinson's disease (PD) is characterized mainly by its motor symptoms, which include bradykinesia, rigidity, and tremors. However, the nonmotor manifestations – ranging from sensory disturbances and cognitive decline to emotional perturbations – determine the quality of life and life expectancy of the patients (Chaudhuri et al., 2007; Chaudhuri and Schapira, 2009; Kotz et al., 2009; Leroi et al., 2011; Martinez-Martin et al., 2011; Simuni and Sethi, 2008; Solla et al., 2011). Some of the nonmotor symptoms appear many years before the clinical diagnosis. It is increasingly accepted that clinically diagnosed PD is preceded by a ⁎ Correspondence to: Y.-Y.I. Shih, Department of Neurology, Biomedical Research Imaging Center, and Department of Biomedical Engineering, 130 Mason Farm Road, CB# 7513, University of North Carolina, Chapel Hill, NC 27599, USA. Fax: +1 919 843 4456. ⁎⁎ Correspondence to: C. Chang, N123, Institute of Biomedical Sciences, Academia Sinica, 128 Section 2, Academia Rd., Nankang, Taipei 11529, Taiwan. Fax: +886 2 2788 7641. E-mail addresses: [email protected] (Y.-Y.I. Shih), [email protected] (C. Chang). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nbd.2012.07.020
long latency period during which nonmotor symptoms gradually develop, antedating the motor dysfunctions (Berendse and Ponsen, 2009; Chaudhuri and Naidu, 2008; Chaudhuri and Schapira, 2009; Park and Stacy, 2009; Simuni and Sethi, 2008). If the prodromal nonmotor changes specific to PD can be identified as definite characteristics, PD may be arrested earlier or even in the preclinical stage (Maetzler et al., 2009; Tolosa and Pont-Sunyer, 2011). In particular, outcome measures of nonmotor manifestations with specificity for the pathology of PD are believed to represent a cornerstone of progress in PD diagnosis and treatment (Remy et al., 2005; Stiasny-Kolster et al., 2005). Pain is a sensory disturbance that is present in as many as 83% of PD patients (Beiske et al., 2009). Pain in PD is mostly left untreated, with analgesic medication only occasionally being used. The pain symptom often occurs before the clinical diagnosis is delivered, and manifests throughout the disease stages. Importantly, the neuropathy underlying the pain symptom in PD could be largely attributed to the similar aberrations responsible for the motor dysfunctions—deficient dopaminergic neurotransmission in the basal ganglia (Chaudhuri and Schapira, 2009; Chudler and Dong, 1995). Moreover, the D2/D3 receptors play a major role in the modulation of pain responses (Chudler and Dong, 1995; Magnusson and Fisher, 2000; Pertovaara et al., 2004). Levodopa
100
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
alleviates not only the motor symptoms but also the painful sensations in PD patients (Brefel-Courbon et al., 2005). Given the reasons stated above, pain may be a suitable nonmotor symptom that can be utilized to characterize PD. The present study applied a novel functional magnetic resonance imaging (fMRI) technique for depicting pain in PD with respect to dopaminergic deficiency. This new approach is potentially more specific than clinical rating scales in the identification of the pain symptom of PD. The concept is based upon a recent fMRI finding that nociceptive stimuli under normal conditions elicit bilateral decreases in cerebral blood volume (CBV) in the striatum (Shih et al., 2009, 2011, 2012; Zhao et al., 2012). Antagonizing dopamine D2/D3 receptors abolishes this response (Shih et al., 2009). Since these striatal CBV decreases are a result of nociception and are tightly coupled to dopaminergic neurotransmission, it is likely that this decreased CBV response is modified in PD, and that the disease-modified changes may thus be defined as a characteristic of the disease (Shih et al., 2009). The aim of the present study was to prove the concept that pain as a prevalent nonmotor symptom in PD can be characterized by fMRI with specificity for dopaminergic dysfunction. CBV-weighted fMRI was used to explore the nociception-triggered CBV decreases in a rat PD model induced by 6-hydroxy-dopamine (6-OHDA) (Mandeville et al., 2004; Shih et al., 2009). MRI is recognized for its lack of radioactivity, high resolution, sensitivity, reproducibility, and quantifiability (Baudrexel et al., 2010). The use of MRI to characterize disease symptoms is also advantageous for the development of homologous measurements that can be seamlessly translated from bench to bedside. The present study represents an endeavor toward the development of dopaminergic imaging for nonmotor symptoms in PD. Importantly, the use of MRI may greatly enhance the practical value of the assessment. Materials and methods Subjects Twenty-two male 8–10 weeks old Wistar rats from the National Laboratory Animal Center of Taiwan were used. Rats were housed in triplets in plastic cages with free access to food and water. The housing environment was specific-pathogen-free with a 12:12 hour light: dark cycle, controlled humidity and temperature. All experimental procedures were approved by the Institute of Animal Care and Utilization Committee at Academia Sinica, Taipei, Taiwan. The rat PD model The rat PD animal model was induced by lesioning the right substantia nigra (SN) with 6-OHDA at 30μg in a volume of 3μl dissolved in ice-cold 0.02% ascorbic acid preceded by an intraperitoneal injection of desipramine at 15 mg/kg. The coordinates for the right SN were 5.3 mm posterior, 2.1 mm lateral, and 8.2 mm ventral from the bregma. A total of sixteen parkinsonian rats and six sham operated (sham) rats were induced. Six parkinsonian and six sham were used for nociceptive tests while ten parkinsonian rats were for the fMRI experiment. Rotation test Two weeks after the injection of 6-OHDA, the rats were injected intraperitoneally with methamphetamine (2 mg/kg), and then the rotation behavior was analyzed. The recording began at 15 min postinjection and lasted for 60 min. The rotation rate was quantified as the number of ipsilateral turns minus the number of contralateral turns per minute. A rat was considered parkinsonian when the rotation reaches 6 turns/min, according to previous studies. The averaged rotation rate of the parkinsonian rats used in the present study was 9.0 ± 3.3 (mean ± SD) turns/min.
Pain hot-plate and foot-shock tests Six parkinsonian and six sham rats were subjected to two pain tests (Bannon and Malmberg, 2007; Tassorelli et al., 2007). Before the tasks, the rats were allowed to acclimatize to the testing environment. The hot-plate test employed a temperature of 55 °C, with the latency of paw licking or jumping being recorded in three trials. The mean latency was used for statistical analysis. In the foot-shock test, the threshold to electrical stimulation was measured. This was defined as the lowest intensity at which a flinch or jump response was observed. Four trials were given. Unilateral testing was enabled by covering the untested paws with small sacks. The mean threshold was used for statistical analysis. fMRI experiments A total of ten parkinsonian rats were subjected to the fMRI experiment performed on a 4.7-T spectrometer (Biospec 47/40, Bruker, Germany) with a 72 mm volume coil as the RF transmitter and a quadrature surface coil as the receiver. Anesthesia was induced by an intravenous injection of α-chloralose (70 mg/kg) dissolved in heated 0.9% saline and 10% polyethylene glycol. According to the settings of previous studies (Shih et al., 2009, 2012), steady state CBV-weighted fMRI was acquired with the use of superparamagnetic iron oxide (SPIO; Resovist, Schering, Berlin, Germany) at a dose of 30 mg Fe/kg given intravenously as the contrast agent. Increases in regional CBV during enhanced neural activity increase the quantity of SPIO nanoparticles, leading to a lower signal intensity on images. Elevated signal intensity represents a decrease in regional CBV (Mandeville et al., 2004; Shih et al., 2009). For each fMRI scan, a time-series of 60 images on the axial plane were acquired. The first 20, middle 20, and last 20 time frames correspond to the off, on, and off phases of nociceptive electrical stimulation that was delivered to a unilateral forepaw via a pair of needle electrodes. Gradient-echo images were acquired in time-series with a repetition time of 150 ms, echo time of 15 ms, flip angle of 22.5°, field of view of 2.56 cm by 2.56 cm, slice thickness of 1.5 mm, acquisition matrix of 128 by 64 (zero-filled to 128 by 128), and temporal resolution of 9.6 s. The nociceptive electrical stimulation was delivered to a unilateral forepaw via a pair of needle electrodes. The stimulation intensity was 10 mA administered by a constant-current stimulator (AM system, model 2100, Carlsburg, WA, USA). The intensity was shown to induce nociception (Chang and Shyu, 2001; Liu et al., 2004; Zhao et al., 2008b). Such unilateral electrical stimulation is known to induce bilateral CBV decreases in the striatum with CBV increases in the contralateral S1 (Shih et al., 2009). Five parkinsonian rats were perfused immediately for immunohistology after scanning while the other five were further tested with Eticlopride (E101, Sigma-Aldrich, USA), a dopamine D2/D3 receptor antagonist. Eticlopride was given intravenously at a dose of 1 mg/kg (Shih et al., 2009). Steady state CBV weighted fMRI was performed again 10 min after the injection. Throughout the experiment, the body temperature was maintained at 37 °C using a warm-water blanket. The averaged end-tidal CO2 concentration was 3.0–3.5%. The baseline pCO2 was 41–43 mm Hg. The mean arterial blood pressure before stimulation was 95–105 mm Hg, and this increased during electrical stimulation by 5–10 mm Hg. Image processing and analysis Correlation maps were generated by plotting the correlation coefficient (CC) between the image signals and the off–on–off electrical stimulation paradigm on a voxel-by-voxel basis using the cross-correlation method (Bandettini et al., 1993). The cut-off points for the CC were r = ±0.2. The threshold was chosen empirically since the experimental condition (such as the lesion induced by 6-OHDA herein) affects the
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
correlations (Shih et al., 2009, 2011, 2012; Shmuel et al., 2002). This threshold was effective to identify spatial clusters in the striatum and the primary somatosensory cortex. The detected activation pixels within the striatum using this threshold show high correspondence with the tyrosine hydroxylase immunostaining results. The signal intensities expressed as % changes in relation to the baseline with a linear correction for the washout of the iron oxide nanoparticles were plotted along the time course. The % changes in signal intensities during the stimulation were averaged as a measurement for statistical analysis. The number of voxels showing CBV signals above the thresholds was also quantified from the CBV correlation maps. ANOVA followed by Fisher's post-hoc was used to discern statistical differences. The striatum was defined by its contrast to the surroundings. The striatum was bordered by the white matters, including the external capsule, and internal capsule. These structures appear as hypointensities on the images. Such signal contrast was utilized for identifying the striatum. In addition, the brain atlas was also used as a reference. Tyrosine hydroxylase (TH) immunohistochemistry The five parkinsonian rats were perfused by 4% paraformaldehyde and the brains were cryosectioned at 50 μm. The brain sections were pretreated with phosphate-buffered saline containing 0.3% H2O2 and 0.1% sodium azide, then incubated in anti-TH (1:3000; Sigma-Aldrich, USA) for 24 h at 4 °C, then in biotinylated goat anti-rabbit IgG (1:1000; Jackson Immunoresearch, USA) for 4 h at room temperature, then exposed to avidin–biotin–peroxidase complex (1:500; Vector, USA) for 3 h at room temperature, and then stained in 0.05 M Tris buffer with 0.025% diaminobenzidine and 0.024% H2O2 for 5–10 min until the desired brown color had developed. The sections were then washed, mounted, dehydrated, and coverslipped. The TH-stained sections were photographed under a light microscope (BX51, Olympus, Japan). TH signal to noise ratio (SNR) was defined as the ratio of TH immunoreactivity of the striatum to the adjacent corpus callosum where minimal immunoreactivity was observed. The photomicrographs of the TH-stained sections were inversely converted to a 256-level grayscale so as to associate stronger staining with a higher TH SNR. Results Changes in pain responses in PD The results of the hot-plate test shown in Fig. 1a indicate that the response latency for the nociceptive thermal stimulus was significantly
101
shorter in the parkinsonian rats (p b 0.05). The results of the foot-shock test shown in Fig. 1b indicate that the parkinsonian rats exhibited a lower nociceptive threshold (i.e., hypersensitivity) to the stimulus on the lesioned side of the body compared with the contralateral side, and with both sides of the sham-operated group (ANOVA: F(1,10) = 82.1, p b 0.0001; Fisher's post-hoc: all p values b 0.0001). Moreover, the body contralateral to the lesioned side exhibited a blunted sensitivity to nociceptive foot shock, as reported previously (Bannon and Malmberg, 2007; Tassorelli et al., 2007). A diminished pattern of nociception-induced CBV decreases in parkinsonism that corresponds geometrically with the distribution of dopaminergic fibers Fig. 2a presents the correlation map between the CBV signals and the stimulation paradigm. CBV decreases were observed in the intact striatum, whereas the lesioned striatum exhibited a diminished reaction, yet with residual responses. CBV increases remained intact in the primary somatosensory cortex (S1) contralateral to the stimulation side. Figs. 2b and c shows the time courses of the signal intensities of the left and right hemispheres, respectively, of the parkinsonian rats. In the left hemisphere, the striatal CBV decreases appeared as signal elevations during the stimulation time period (marked by the shading), whereas the increased CBV reaction occurring in S1 manifested as signal reductions. The signal increases in the striatum were smaller in the right (lesioned) hemisphere. Fig. 2d presents the averaged percentage signal changes in the bilateral striata during stimulation. There was a significant difference between the intact striatum and the lesioned striatum (p b 0.05). Fig. 3a shows a TH-stained brain section from a parkinsonian rat. It reveals that the intact striatum retained a high density of dopaminergic fibers, whereas the damaged striatum displayed only some residual fibers. Fig. 3b shows that the averaged TH SNR was significantly lower in the damaged striatum than in the intact striatum. Figs. 3d and c both show that the CBV responses were positively correlated with the striatal TH SNR. The correlation plot was based upon the data from both the intact and lesioned striata of the five parkinsonian rats that underwent both fMRI and TH immunohistology. The CBV response is quantified by the number of voxels for the striatal CBV response in Fig. 3d and by the averaged percentage signal changes in Fig. 3e. Fig. 4 demonstrates the geometric correspondence between the correlation map and the TH staining. In the left half of the rostral striatum, as shown in Fig. 4a, the CBV decreases were as abundant as those of the TH staining. By contrast, in the right half of the rostral
Fig. 1. PD alters behavioral responses to nociceptive stimuli. (a) The hot-plate test showed that the response latency to a nociceptive thermal stimulus was significantly shorter in PD, indicating nociceptive hypersensitivity caused by dopaminergic impairment (p b 0.05). (b) The foot-shock test further indicated that the hypersensitivity to pain stimuli mainly occurred on the side of the body ipsilateral to the lesion (R.). * indicates p b 0.0001.
102
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Figs. 5c, d, e, and f illustrates the time courses of the signal changes in the left and right striata before and after eticlopride, respectively, in parkinsonian rats. Prior to eticlopride injection, the striatal response in the intact striatum appeared as signal elevations during the period of stimulation. Signal changes were weaker in the lesioned striatum. Following eticlopride injection, both striata exhibited minimal signal changes during stimulation. Fig. 5g presents the averaged percentage of signal changes in each striatum before and after eticlopride. Two-way ANOVA revealed that administration of the dopamine D2/D3 blocker significantly inhibited the CBV decreases after eticlopride treatment [F(1,8) = 8.8, p b 0.05]. Fisher's post-hoc tests indicated that the averaged activity was significantly higher on the left than on the right side before the treatment (p b 0.01). The activity was low in both hemispheres after the treatment. Laterality of the fMRI responses The CBV increases in S1 occurred contralateral to the side receiving nociceptive electrical stimulation, regardless of the dopaminergic lesion or antagonism. The right and left forepaws were stimulated for Figs. 1 and 4, respectively. The striatal CBV decreases occurred bilaterally, irrespective of the side receiving the nociceptive stimulation. However, the reduced responsiveness of the striatal CBV reactions occurred ipsilateral to the lesion of nigrostriatal dopaminergic inputs. The lesion was made on the right SN in the present study. Discussion
Fig. 2. In PD, the lesioned striatum showed a diminished CBV response. (a) The CBV correlation maps of the striatum in response to nociceptive stimulation. (b), (c) The time course of the signal intensities in the left and right hemispheres of the parkinsonian rats, respectively. (d) The average % signal change of the bilateral striatum in relation to the baseline during stimulation.
striatum, as shown in Fig. 4b, minimal residual CBV reactions were observed in the ventral portion. This was consistent with the location of the remaining TH staining in the ventral area. In the caudal striatum, the CBV response of the left half was distributed as widely as the TH staining (Fig. 4c). The right half did not exhibit any CBV signals, consistent with the absence of TH staining (Fig. 4d). Dopamine D2/D3 antagonism abolished the remaining CBV decreases in parkinsonism Figs. 5a and b presents the correlation maps before and after treatment with eticlopride in parkinsonian rats, respectively. The injection of eticlopride abolished all of the striatal CBV signals on the correlation maps, but did not affect the CBV response of S1.
The aim of the present study was to determine whether pain as a prevalent symptom of the nonmotor manifestations in PD can be characterized with specificity for the cardinal deficiency of the dopaminergic system, based upon an objective fMRI approach. Behavioral assessment first confirmed that hyperalgesia is indeed a symptom of PD. fMRI data then revealed that the lesioned striatum exhibited a diminished CBV response to the nociceptive stimulus. Immunohistology revealed that the reduced CBV responsiveness occurred mainly in areas with dopaminergic denervation. A strong correspondence was observed between the distributions of the nociception-induced CBV responses and dopaminergic innervations. The remaining CBV signals in the striatum were abolished by the D2/D3 antagonist eticlopride. The findings from the behavioral, neuroimaging, immunohistological, and pharmacological experiments demonstrate that pain in PD can be characterized as striatal CBV signal changes with specificity for dopaminergic dysfunction. These changes are a potential candidate for the objective identification of the nonmotor symptoms in PD. Pain is an under-recognized but important symptom in PD (Beiske et al., 2009; Djaldetti et al., 2004; Ford, 2010; Hanagasi et al., 2011; Reichling and Levine, 2011). It is often viewed as a secondary consequence to the motor disabilities. However, as many as 43% of PD patients experience painful sensations in the periphery even before their motor symptoms are diagnosed (Reichling and Levine, 2011). Neurodegeneration is increasingly accepted as a major cause of peripheral pain. The nigrostriatal pathway is known to respond to nociceptive stimuli and to perform an antinociceptive function (Brooks, 2006; Chudler and Dong, 1995). Many previous studies have demonstrated that lesioning striatal dopaminergic neurons increases pain sensitivity (Lin et al., 1984), whereas activating these neurons inhibits pain responses (Chudler and Dong, 1995). In particular, the D2/D3 receptors – and not the D1 receptor – is the main contributor to this behavior (Magnusson and Fisher, 2000). According to human reports, D2 agonists such as pergolide, ropinirole, and pramipexole, which are sometimes used for treatment, produced reductions in pain responses in PD patients (Linazasoro, 2008; Stuginski-Barbosa et al., 2008). These findings suggest an intimate relationship among pain and dopamine D2/D3 receptors in PD.
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
103
Fig. 3. Quantitative analysis of the correspondence between nociception-induced CBV responses and tyrosine hydroxylase (TH) staining. (a) A striatal section of PD that is stained by TH. (b) The averaged TH SNR of the left, intact and right, damaged striatum. (c) TH SNR is correlated with the count of striatal voxels on the correlation maps of PD. (d) TH SNR is correlated with the averaged % signal changes in the striatum of PD.
Noxious electrical stimulation induces stress and related CNS opioid release at the substantia nigra, which could affect dopaminergic functioning at the striatum (Drolet et al., 2001; Ostrowski and Pert, 1995). The dense expression of μ-opioid receptors in the substantia nigra and the striatum is likely responsible for the interactions between the opioid and dopaminergic systems (Tempel and Zukin, 1987). Our
recent study showed that the activation of opioid neurotransmission strengthened the striatal CBV vasoconstriction while the blockage diminished it (Shih et al., 2012). It seems that in a scenario that involves noxious stimuli, the role of opioids should be considered in the modulation of the striatal dopaminergic functions including the nociception-induced CBV signals and the pain response.
Fig. 4. Spatial correspondence between the striatal CBV decreases and the dopaminergic innervations. (a) In the left half of the rostral striatum, the CBV decreases were as abundant as those of the TH staining. (b) In the right half of the rostral striatum, minimal residual CBV reactions were observed in the ventral portion. This was consistent with the location of the remaining TH staining in the ventral area. (c) In the caudal striatum, the CBV response of the left half was distributed as widely as the TH staining. (d) The right half of the caudal striatum did not exhibit any CBV signals, consistent with the absence of TH staining.
104
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Fig. 5. The role of dopaminergic D2/D3 receptors in the striatal CBV responses of PD. (a) The CBV correlation maps before the eticlopride treatment. (b) The CBV correlation maps after the eticlopride treatment. (c) The time course of the signal intensities of the left striatum before eticlopride. (d) The time course of the signal intensities of the right striatum before eticlopride. (e) The time course of the signal intensities of the left striatum after eticlopride. (f) The time course of the signal intensities of the right striatum after eticlopride. (g) The averaged % signal change of each striatum before and after eticlopride.
The nociception-induced striatal CBV response is a negative fMRI response involving decreases in the CBV. This reaction has been reported repeatedly by different groups (Shih et al., 2009; Zhao et al., 2008a). This striatal response per se is an intriguing phenomenon. The activation of the dopaminergic D2/D3 receptors plays a key role in the generation of such response. Dopaminergic neurotransmission may affect the vascular tone by the activation of receptors on astrocytes or directly on the vessels (Choi et al., 2006). When astrocytes are involved, the released dopamine binds to the dopaminergic receptors on the astrocytes. Astrocytes then use their endfeet to regulate the vascular activity. An alternative mechanism involves direct activation of the receptors expressed on the vessels. But the exact cellular mechanism responsible for the striatal CBV response remains to be elucidated. The development of definite diagnostic markers for symptoms in PD is urgently required among the numerous priorities in the treatment of
the disease. Characterization of the nonmotor manifestations is particularly important, since it will (1) identify under-recognized nonmotor signs, (2) redefine the disease progression and stages, (3) facilitate the development of biomarkers, and (4) serve as a standard for the evaluation of the effects of new drugs. Nonmotor symptoms in PD such as depression, anxiety, olfactory dysfunction, fatigue, and sleep disorders have been of great interest to the research community in the recent decade. Clinical assessment by means of questionnaires, rating scales, and behavioral tests remains the most convenient, economical, and immediate evaluation tool available to neurologists. However, since most of the nonmotor symptoms are not specific (i.e., occurring in other clinical circumstances), outcome measures with specificity for the cardinal abnormalities of the disease should be mandatory for objective end points (Maetzler et al., 2009). Neuroimaging modalities including positron-emission tomography (PET), single-photon-emission computed tomography (SPECT),
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
diffusion-tensor MRI, and structural MRI have emerged as promising approaches for PD diagnosis (Antonelli et al., 2010; Brooks, 2006; Brooks and Pavese, 2010; Loane and Politis, 2011; Pavese and Brooks, 2009; Remy et al., 2005; Stoessl, 2009). In particular, imaging methods specific to the detection of dopaminergic dysfunctions have received the greatest attention, with PET and SPECT being the primary modalities. However, they are limited by their spatial resolution and radioactivity in nature. As such, dopaminergic imaging by MRI appears as an appealing option. fMRI has been shown to have the capacity to image dopaminergic activity (Jenkins et al., 2004). The present study extended the use of this technique and is the first to demonstrate the utility of fMRI in dopaminergic imaging of nonmotor manifestations in PD using a rat model. The observation of dopaminergic innervations and denervations represented by CBV signals is a clear illustration of the strength of the high spatial resolution offered by MRI. A significant step that has been made recently in the prospective development of biomarkers for PD is the launch of the Parkinson's Progression Markers Initiative (Foundation, 2011; The Lancet, 2010). The aim of this program is to collect clinical, imaging, and biological data to identify promising biomarkers of PD progression. Unsurprisingly, the characterization of nonmotor manifestations is an important pursuit of the consortium. Recent research trends appear to indicate that dopaminergic imaging offered either by fMRI or PET (in addition to the conventional tools) will be used to assess nonmotor manifestations to enhance the specificity and accuracy of clinical evaluations. Biomarker development is an iterative process. Several nonmotor symptoms such as depression are believed to antedate the motor disabilities and to be directly linked to dopaminergic dysfunctions (Brooks, 2006; Thobois et al., 2010). The development of these nonmotor manifestations as biomarkers that can be used to detect any early deficiencies associated with the dopaminergic system is therefore particularly important. We are also interested in the potential application of dopaminergic imaging of pain in PD by fMRI in the clinical setting. The present study focused on the validation of that concept. However, there are some uncertainties that should be further elucidated, such as the correlation with disease severity, the ethical issues of inducing pain in PD patients, and translation of the fMRI method to the clinical setting. Additionally, in PD patients, inducing pain to trigger the CBV response could be complicated by the individual differences as well as disease-modified changes. Several milestones need to be reached before developing the dopaminergic imaging of pain in PD as a biomarker. Acknowledgment We acknowledge the technical support provided by Mr. Nai-Wei Yao and would like to thank the Functional and Micro-Magnetic Resonance Imaging Center supported by the National Research Program for Genomic Medicine, National Science Council, Taiwan, Republic of China, NSC100-3112-B-001-009. References Antonelli, F., et al., 2010. Imaging cognitive and behavioral symptoms in Parkinson's disease. Expert Rev. Neurother. 10, 1827–1838. Bandettini, P.A., et al., 1993. Processing strategies for time-course data sets in functional MRI of the human brain. Magn. Reson. Med. 30, 161–173. Bannon, A.W., Malmberg, A.B., 2007. Models of nociception: hot-plate, tail-flick, and formalin tests in rodents. Curr. Protoc. Neurosci. (Chapter 8, Unit 8 9). Baudrexel, S., et al., 2010. Quantitative mapping of T1 and T2* discloses nigral and brainstem pathology in early Parkinson's disease. Neuroimage 51, 512–520. Beiske, A.G., et al., 2009. Pain in Parkinson's disease: prevalence and characteristics. Pain 141, 173–177. Berendse, H.W., Ponsen, M.M., 2009. Diagnosing premotor Parkinson's disease using a two-step approach combining olfactory testing and DAT SPECT imaging. Parkinsonism Relat. Disord. 15 (Suppl. 3), S26–S30. Brefel-Courbon, C., et al., 2005. Effect of levodopa on pain threshold in Parkinson's disease: a clinical and positron emission tomography study. Mov. Disord. 20, 1557–1563.
105
Brooks, D.J., 2006. Dopaminergic action beyond its effects on motor function: imaging studies. J. Neurol. 253 (Suppl. 4), IV8–IV15. Brooks, D.J., Pavese, N., 2010. Imaging non-motor aspects of Parkinson's disease. Prog. Brain Res. 184, 205–218. Chang, C., Shyu, B.C., 2001. A fMRI study of brain activations during non-noxious and noxious electrical stimulation of the sciatic nerve of rats. Brain Res. 897, 71–81. Chaudhuri, K.R., Naidu, Y., 2008. Early Parkinson's disease and non-motor issues. J. Neurol. 255 (Suppl. 5), 33–38. Chaudhuri, K.R., Schapira, A.H., 2009. Non-motor symptoms of Parkinson's disease: dopaminergic pathophysiology and treatment. Lancet Neurol. 8, 464–474. Chaudhuri, K.R., et al., 2007. The metric properties of a novel non-motor symptoms scale for Parkinson's disease: results from an international pilot study. Mov. Disord. 22, 1901–1911. Choi, J.K., et al., 2006. Brain hemodynamic changes mediated by dopamine receptors: role of the cerebral microvasculature in dopamine-mediated neurovascular coupling. Neuroimage 30, 700–712. Chudler, E.H., Dong, W.K., 1995. The role of the basal ganglia in nociception and pain. Pain 60, 3–38. Djaldetti, R., et al., 2004. Quantitative measurement of pain sensation in patients with Parkinson disease. Neurology 62, 2171–2175. Drolet, G., et al., 2001. Role of endogenous opioid system in the regulation of the stress response. Prog. Neuropsychopharmacol. Biol. Psychiatry 25, 729–741. Ford, B., 2010. Pain in Parkinson's disease. Mov. Disord. 25 (Suppl. 1), S98–S103. Foundation, T.M.J.F., 2011. The Parkinson Progression Marker Initiative (PPMI). Prog. Neurobiol. 95, 629–635. Hanagasi, H.A., et al., 2011. Pain is common in Parkinson's disease. Clin. Neurol. Neurosurg. 113, 11–13. Jenkins, B.G., et al., 2004. Mapping dopamine function in primates using pharmacologic magnetic resonance imaging. J. Neurosci. 24, 9553–9560. Kotz, S.A., et al., 2009. Non-motor basal ganglia functions: a review and proposal for a model of sensory predictability in auditory language perception. Cortex 45, 982–990. Leroi, I., et al., 2011. Behavioural disorders, disability and quality of life in Parkinson's disease. Age Ageing 40, 614–621. Lin, M.T., et al., 1984. Effects of kainic acid injections in the striatum on physiologic and behavioral functions in conscious rats. Exp. Neurol. 83, 71–83. Linazasoro, G., 2008. Conversion from dopamine agonists to cabergoline: an open-label trial in 128 patients with advanced Parkinson disease. Clin. Neuropharmacol. 31, 19–24. Liu, Z.M., et al., 2004. Imaging oxygen consumption in forepaw somatosensory stimulation in rats under isoflurane anesthesia. Magn. Reson. Med. 52, 277–285. Loane, C., Politis, M., 2011. Positron emission tomography neuroimaging in Parkinson's disease. Am. J. Transl. Res. 3, 323–341. Maetzler, W., et al., 2009. Progression of Parkinson's disease in the clinical phase: potential markers. Lancet Neurol. 8, 1158–1171. Magnusson, J.E., Fisher, K., 2000. The involvement of dopamine in nociception: the role of D(1) and D(2) receptors in the dorsolateral striatum. Brain Res. 855, 260–266. Mandeville, J.B., et al., 2004. Exogenous contrast agent improves sensitivity of gradient-echo functional magnetic resonance imaging at 9.4 T. Magn. Reson. Med. 52, 1272–1281. Martinez-Martin, P., et al., 2011. The impact of non-motor symptoms on health-related quality of life of patients with Parkinson's disease. Mov. Disord. 26, 399–406. Ostrowski, N.L., Pert, A., 1995. Substantia nigra opiate receptors on basal ganglia efferents. Brain Res. Bull. 38, 513–523. Park, A., Stacy, M., 2009. Non-motor symptoms in Parkinson's disease. J. Neurol. 256 (Suppl. 3), 293–298. Pavese, N., Brooks, D.J., 2009. Imaging neurodegeneration in Parkinson's disease. Biochim. Biophys. Acta 1792, 722–729. Pertovaara, A., et al., 2004. Striatal dopamine D2/D3 receptor availability correlates with individual response characteristics to pain. Eur. J. Neurosci. 20, 1587–1592. Reichling, D.B., Levine, J.D., 2011. Pain and death: neurodegenerative disease mechanisms in the nociceptor. Ann. Neurol. 69, 13–21. Remy, P., et al., 2005. Depression in Parkinson's disease: loss of dopamine and noradrenaline innervation in the limbic system. Brain 128, 1314–1322. Shih, Y.Y.I., et al., 2009. A new scenario for negative fMRI signals: endogenous neurotransmission. J. Neurosci. 29, 3036–3044. Shih, Y.Y., et al., 2011. Striatal and cortical BOLD, blood flow, blood volume, oxygen consumption, and glucose consumption changes in noxious forepaw electrical stimulation. J. Cereb. Blood Flow Metab. 31, 832–841. Shih, Y.Y., et al., 2012. Endogenous opioid–dopamine neurotransmission underlie negative CBV fMRI signals. Exp. Neurol. 234, 382–388. Shmuel, A., et al., 2002. Sustained negative BOLD, blood flow and oxygen consumption response and its coupling to the positive response in the human brain. Neuron 36, 1195–1210. Simuni, T., Sethi, K., 2008. Nonmotor manifestations of Parkinson's disease. Ann. Neurol. 64 (Suppl. 2), S65–S80. Solla, P., et al., 2011. Behavioral, neuropsychiatric and cognitive disorders in Parkinson's disease patients with and without motor complications. Prog. Neuropsychopharmacol. Biol. Psychiatry 35, 1009–1013. Stiasny-Kolster, K., et al., 2005. Combination of ‘idiopathic’ REM sleep behaviour disorder and olfactory dysfunction as possible indicator for alpha-synucleinopathy demonstrated by dopamine transporter FP-CIT-SPECT. Brain 128, 126–137. Stoessl, A.J., 2009. Functional imaging studies of non-motoric manifestations of Parkinson's Disease. Parkinsonism Relat. Disord. 15 (Suppl. 3), S13–S16. Stuginski-Barbosa, J., et al., 2008. Burning mouth syndrome responsive to pramipexol. J. Headache Pain 9, 43–45.
106
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Tassorelli, C., et al., 2007. Behavioral responses and Fos activation following painful stimuli in a rodent model of Parkinson's disease. Brain Res. 1176, 53–61. Tempel, A., Zukin, R.S., 1987. Neuroanatomical patterns of the mu, delta, and kappa opioid receptors of rat brain as determined by quantitative in vitro autoradiography. Proc. Natl. Acad. Sci. U. S. A. 84, 4308–4312. The Lancet, N., 2010. Biomarker promise for Parkinson's disease. Lancet Neurol. 9, 1139. Thobois, S., et al., 2010. Non-motor dopamine withdrawal syndrome after surgery for Parkinson's disease: predictors and underlying mesolimbic denervation. Brain 133, 1111–1127.
Tolosa, E., Pont-Sunyer, C., 2011. Progress in defining the premotor phase of Parkinson's disease. J. Neurol. Sci. 310, 4–8. Zhao, F., et al., 2008a. BOLD and blood volume-weighted fMRI of rat lumbar spinal cord during non-noxious and noxious electrical hindpaw stimulation. Neuroimage 40, 133–147. Zhao, F., et al., 2008b. BOLD study of stimulation-induced neural activity and restingstate connectivity in medetomidine-sedated rat. Neuroimage 39, 248–260. Zhao, F., et al., 2012. fMRI of pain processing in the brain: a within-animal comparative study of BOLD vs. CBV and noxious electrical vs. noxious mechanical stimulation in rat. Neuroimage 59, 1168–1179.
Contents lists available at SciVerse ScienceDirect
Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi
Dopaminergic imaging of nonmotor manifestations in a rat model of Parkinson's disease by fMRI Chiao-Chi V. Chen a, b, Yen-Yu I. Shih c, d, e,⁎, Chen Chang a, b,⁎⁎ a
Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan Functional and Micro-magnetic Resonance Imaging Center, Academic Sinica, Taipei, Taiwan Department of Neurology, University of North Carolina at Chapel Hill, NC, USA d Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, NC, USA e Department of Biomedical Engineering, University of North Carolina at Chapel Hill, NC, USA b c
a r t i c l e
i n f o
Article history: Received 4 May 2012 Revised 4 July 2012 Accepted 20 July 2012 Available online 27 July 2012 Keywords: Parkinson disease Non-motor Biomarker BOLD Vasoconstriction fMRI Neuroimaging In vivo
a b s t r a c t Nonmotor manifestations determine the life quality of patients with Parkinson's disease (PD). Identification of the nonmotor symptoms in PD as definite changes will represent a milestone in its diagnosis. Outcome measures that characterize nonmotor manifestations with specificity for dopaminergic deficiency are essential to that goal. Pain is a prevalent sensory disturbance in PD patients. The prevalence was reported to be up to 83%. Nociceptive stimuli under normal conditions elicit decreases in cerebral blood volume (CBV) in the striatum via dopaminergic neurotransmission. This nociception-induced CBV response is potentially to be defined as a characteristic of the pain symptom of PD. To validate this concept, steady-state CBV-weighted functional magnetic resonance imaging with iron oxide nanoparticles was employed to measure CBV changes in parkinsonian rats. Tyrosine hydroxylase immunohistology was used to identify the dopaminergic integrity to corroborate the imaging findings. Additional experiments that tested pain responses in parkinsonism were also carried out. The results revealed that the lesioned striatum exhibited a weakened CBV decrease in response to the nociceptive stimulus. This weakened CBV response occurred mainly in areas with dopaminergic denervation. A strong correspondence was observed between the distributions of the nociception-induced CBV responses and dopaminergic innervation. The persisting CBV signals in the striatum were abolished by the D2/D3 antagonist eticlopride. The findings of these behavioral, neuroimaging, immunohistological, and pharmacological experiments demonstrate that pain in a rat model of PD can be characterized by nociception induced striatal CBV signal changes with specificity for dopaminergic dysfunction. © 2012 Elsevier Inc. All rights reserved.
Introduction Parkinson's disease (PD) is characterized mainly by its motor symptoms, which include bradykinesia, rigidity, and tremors. However, the nonmotor manifestations – ranging from sensory disturbances and cognitive decline to emotional perturbations – determine the quality of life and life expectancy of the patients (Chaudhuri et al., 2007; Chaudhuri and Schapira, 2009; Kotz et al., 2009; Leroi et al., 2011; Martinez-Martin et al., 2011; Simuni and Sethi, 2008; Solla et al., 2011). Some of the nonmotor symptoms appear many years before the clinical diagnosis. It is increasingly accepted that clinically diagnosed PD is preceded by a ⁎ Correspondence to: Y.-Y.I. Shih, Department of Neurology, Biomedical Research Imaging Center, and Department of Biomedical Engineering, 130 Mason Farm Road, CB# 7513, University of North Carolina, Chapel Hill, NC 27599, USA. Fax: +1 919 843 4456. ⁎⁎ Correspondence to: C. Chang, N123, Institute of Biomedical Sciences, Academia Sinica, 128 Section 2, Academia Rd., Nankang, Taipei 11529, Taiwan. Fax: +886 2 2788 7641. E-mail addresses: [email protected] (Y.-Y.I. Shih), [email protected] (C. Chang). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nbd.2012.07.020
long latency period during which nonmotor symptoms gradually develop, antedating the motor dysfunctions (Berendse and Ponsen, 2009; Chaudhuri and Naidu, 2008; Chaudhuri and Schapira, 2009; Park and Stacy, 2009; Simuni and Sethi, 2008). If the prodromal nonmotor changes specific to PD can be identified as definite characteristics, PD may be arrested earlier or even in the preclinical stage (Maetzler et al., 2009; Tolosa and Pont-Sunyer, 2011). In particular, outcome measures of nonmotor manifestations with specificity for the pathology of PD are believed to represent a cornerstone of progress in PD diagnosis and treatment (Remy et al., 2005; Stiasny-Kolster et al., 2005). Pain is a sensory disturbance that is present in as many as 83% of PD patients (Beiske et al., 2009). Pain in PD is mostly left untreated, with analgesic medication only occasionally being used. The pain symptom often occurs before the clinical diagnosis is delivered, and manifests throughout the disease stages. Importantly, the neuropathy underlying the pain symptom in PD could be largely attributed to the similar aberrations responsible for the motor dysfunctions—deficient dopaminergic neurotransmission in the basal ganglia (Chaudhuri and Schapira, 2009; Chudler and Dong, 1995). Moreover, the D2/D3 receptors play a major role in the modulation of pain responses (Chudler and Dong, 1995; Magnusson and Fisher, 2000; Pertovaara et al., 2004). Levodopa
100
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
alleviates not only the motor symptoms but also the painful sensations in PD patients (Brefel-Courbon et al., 2005). Given the reasons stated above, pain may be a suitable nonmotor symptom that can be utilized to characterize PD. The present study applied a novel functional magnetic resonance imaging (fMRI) technique for depicting pain in PD with respect to dopaminergic deficiency. This new approach is potentially more specific than clinical rating scales in the identification of the pain symptom of PD. The concept is based upon a recent fMRI finding that nociceptive stimuli under normal conditions elicit bilateral decreases in cerebral blood volume (CBV) in the striatum (Shih et al., 2009, 2011, 2012; Zhao et al., 2012). Antagonizing dopamine D2/D3 receptors abolishes this response (Shih et al., 2009). Since these striatal CBV decreases are a result of nociception and are tightly coupled to dopaminergic neurotransmission, it is likely that this decreased CBV response is modified in PD, and that the disease-modified changes may thus be defined as a characteristic of the disease (Shih et al., 2009). The aim of the present study was to prove the concept that pain as a prevalent nonmotor symptom in PD can be characterized by fMRI with specificity for dopaminergic dysfunction. CBV-weighted fMRI was used to explore the nociception-triggered CBV decreases in a rat PD model induced by 6-hydroxy-dopamine (6-OHDA) (Mandeville et al., 2004; Shih et al., 2009). MRI is recognized for its lack of radioactivity, high resolution, sensitivity, reproducibility, and quantifiability (Baudrexel et al., 2010). The use of MRI to characterize disease symptoms is also advantageous for the development of homologous measurements that can be seamlessly translated from bench to bedside. The present study represents an endeavor toward the development of dopaminergic imaging for nonmotor symptoms in PD. Importantly, the use of MRI may greatly enhance the practical value of the assessment. Materials and methods Subjects Twenty-two male 8–10 weeks old Wistar rats from the National Laboratory Animal Center of Taiwan were used. Rats were housed in triplets in plastic cages with free access to food and water. The housing environment was specific-pathogen-free with a 12:12 hour light: dark cycle, controlled humidity and temperature. All experimental procedures were approved by the Institute of Animal Care and Utilization Committee at Academia Sinica, Taipei, Taiwan. The rat PD model The rat PD animal model was induced by lesioning the right substantia nigra (SN) with 6-OHDA at 30μg in a volume of 3μl dissolved in ice-cold 0.02% ascorbic acid preceded by an intraperitoneal injection of desipramine at 15 mg/kg. The coordinates for the right SN were 5.3 mm posterior, 2.1 mm lateral, and 8.2 mm ventral from the bregma. A total of sixteen parkinsonian rats and six sham operated (sham) rats were induced. Six parkinsonian and six sham were used for nociceptive tests while ten parkinsonian rats were for the fMRI experiment. Rotation test Two weeks after the injection of 6-OHDA, the rats were injected intraperitoneally with methamphetamine (2 mg/kg), and then the rotation behavior was analyzed. The recording began at 15 min postinjection and lasted for 60 min. The rotation rate was quantified as the number of ipsilateral turns minus the number of contralateral turns per minute. A rat was considered parkinsonian when the rotation reaches 6 turns/min, according to previous studies. The averaged rotation rate of the parkinsonian rats used in the present study was 9.0 ± 3.3 (mean ± SD) turns/min.
Pain hot-plate and foot-shock tests Six parkinsonian and six sham rats were subjected to two pain tests (Bannon and Malmberg, 2007; Tassorelli et al., 2007). Before the tasks, the rats were allowed to acclimatize to the testing environment. The hot-plate test employed a temperature of 55 °C, with the latency of paw licking or jumping being recorded in three trials. The mean latency was used for statistical analysis. In the foot-shock test, the threshold to electrical stimulation was measured. This was defined as the lowest intensity at which a flinch or jump response was observed. Four trials were given. Unilateral testing was enabled by covering the untested paws with small sacks. The mean threshold was used for statistical analysis. fMRI experiments A total of ten parkinsonian rats were subjected to the fMRI experiment performed on a 4.7-T spectrometer (Biospec 47/40, Bruker, Germany) with a 72 mm volume coil as the RF transmitter and a quadrature surface coil as the receiver. Anesthesia was induced by an intravenous injection of α-chloralose (70 mg/kg) dissolved in heated 0.9% saline and 10% polyethylene glycol. According to the settings of previous studies (Shih et al., 2009, 2012), steady state CBV-weighted fMRI was acquired with the use of superparamagnetic iron oxide (SPIO; Resovist, Schering, Berlin, Germany) at a dose of 30 mg Fe/kg given intravenously as the contrast agent. Increases in regional CBV during enhanced neural activity increase the quantity of SPIO nanoparticles, leading to a lower signal intensity on images. Elevated signal intensity represents a decrease in regional CBV (Mandeville et al., 2004; Shih et al., 2009). For each fMRI scan, a time-series of 60 images on the axial plane were acquired. The first 20, middle 20, and last 20 time frames correspond to the off, on, and off phases of nociceptive electrical stimulation that was delivered to a unilateral forepaw via a pair of needle electrodes. Gradient-echo images were acquired in time-series with a repetition time of 150 ms, echo time of 15 ms, flip angle of 22.5°, field of view of 2.56 cm by 2.56 cm, slice thickness of 1.5 mm, acquisition matrix of 128 by 64 (zero-filled to 128 by 128), and temporal resolution of 9.6 s. The nociceptive electrical stimulation was delivered to a unilateral forepaw via a pair of needle electrodes. The stimulation intensity was 10 mA administered by a constant-current stimulator (AM system, model 2100, Carlsburg, WA, USA). The intensity was shown to induce nociception (Chang and Shyu, 2001; Liu et al., 2004; Zhao et al., 2008b). Such unilateral electrical stimulation is known to induce bilateral CBV decreases in the striatum with CBV increases in the contralateral S1 (Shih et al., 2009). Five parkinsonian rats were perfused immediately for immunohistology after scanning while the other five were further tested with Eticlopride (E101, Sigma-Aldrich, USA), a dopamine D2/D3 receptor antagonist. Eticlopride was given intravenously at a dose of 1 mg/kg (Shih et al., 2009). Steady state CBV weighted fMRI was performed again 10 min after the injection. Throughout the experiment, the body temperature was maintained at 37 °C using a warm-water blanket. The averaged end-tidal CO2 concentration was 3.0–3.5%. The baseline pCO2 was 41–43 mm Hg. The mean arterial blood pressure before stimulation was 95–105 mm Hg, and this increased during electrical stimulation by 5–10 mm Hg. Image processing and analysis Correlation maps were generated by plotting the correlation coefficient (CC) between the image signals and the off–on–off electrical stimulation paradigm on a voxel-by-voxel basis using the cross-correlation method (Bandettini et al., 1993). The cut-off points for the CC were r = ±0.2. The threshold was chosen empirically since the experimental condition (such as the lesion induced by 6-OHDA herein) affects the
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
correlations (Shih et al., 2009, 2011, 2012; Shmuel et al., 2002). This threshold was effective to identify spatial clusters in the striatum and the primary somatosensory cortex. The detected activation pixels within the striatum using this threshold show high correspondence with the tyrosine hydroxylase immunostaining results. The signal intensities expressed as % changes in relation to the baseline with a linear correction for the washout of the iron oxide nanoparticles were plotted along the time course. The % changes in signal intensities during the stimulation were averaged as a measurement for statistical analysis. The number of voxels showing CBV signals above the thresholds was also quantified from the CBV correlation maps. ANOVA followed by Fisher's post-hoc was used to discern statistical differences. The striatum was defined by its contrast to the surroundings. The striatum was bordered by the white matters, including the external capsule, and internal capsule. These structures appear as hypointensities on the images. Such signal contrast was utilized for identifying the striatum. In addition, the brain atlas was also used as a reference. Tyrosine hydroxylase (TH) immunohistochemistry The five parkinsonian rats were perfused by 4% paraformaldehyde and the brains were cryosectioned at 50 μm. The brain sections were pretreated with phosphate-buffered saline containing 0.3% H2O2 and 0.1% sodium azide, then incubated in anti-TH (1:3000; Sigma-Aldrich, USA) for 24 h at 4 °C, then in biotinylated goat anti-rabbit IgG (1:1000; Jackson Immunoresearch, USA) for 4 h at room temperature, then exposed to avidin–biotin–peroxidase complex (1:500; Vector, USA) for 3 h at room temperature, and then stained in 0.05 M Tris buffer with 0.025% diaminobenzidine and 0.024% H2O2 for 5–10 min until the desired brown color had developed. The sections were then washed, mounted, dehydrated, and coverslipped. The TH-stained sections were photographed under a light microscope (BX51, Olympus, Japan). TH signal to noise ratio (SNR) was defined as the ratio of TH immunoreactivity of the striatum to the adjacent corpus callosum where minimal immunoreactivity was observed. The photomicrographs of the TH-stained sections were inversely converted to a 256-level grayscale so as to associate stronger staining with a higher TH SNR. Results Changes in pain responses in PD The results of the hot-plate test shown in Fig. 1a indicate that the response latency for the nociceptive thermal stimulus was significantly
101
shorter in the parkinsonian rats (p b 0.05). The results of the foot-shock test shown in Fig. 1b indicate that the parkinsonian rats exhibited a lower nociceptive threshold (i.e., hypersensitivity) to the stimulus on the lesioned side of the body compared with the contralateral side, and with both sides of the sham-operated group (ANOVA: F(1,10) = 82.1, p b 0.0001; Fisher's post-hoc: all p values b 0.0001). Moreover, the body contralateral to the lesioned side exhibited a blunted sensitivity to nociceptive foot shock, as reported previously (Bannon and Malmberg, 2007; Tassorelli et al., 2007). A diminished pattern of nociception-induced CBV decreases in parkinsonism that corresponds geometrically with the distribution of dopaminergic fibers Fig. 2a presents the correlation map between the CBV signals and the stimulation paradigm. CBV decreases were observed in the intact striatum, whereas the lesioned striatum exhibited a diminished reaction, yet with residual responses. CBV increases remained intact in the primary somatosensory cortex (S1) contralateral to the stimulation side. Figs. 2b and c shows the time courses of the signal intensities of the left and right hemispheres, respectively, of the parkinsonian rats. In the left hemisphere, the striatal CBV decreases appeared as signal elevations during the stimulation time period (marked by the shading), whereas the increased CBV reaction occurring in S1 manifested as signal reductions. The signal increases in the striatum were smaller in the right (lesioned) hemisphere. Fig. 2d presents the averaged percentage signal changes in the bilateral striata during stimulation. There was a significant difference between the intact striatum and the lesioned striatum (p b 0.05). Fig. 3a shows a TH-stained brain section from a parkinsonian rat. It reveals that the intact striatum retained a high density of dopaminergic fibers, whereas the damaged striatum displayed only some residual fibers. Fig. 3b shows that the averaged TH SNR was significantly lower in the damaged striatum than in the intact striatum. Figs. 3d and c both show that the CBV responses were positively correlated with the striatal TH SNR. The correlation plot was based upon the data from both the intact and lesioned striata of the five parkinsonian rats that underwent both fMRI and TH immunohistology. The CBV response is quantified by the number of voxels for the striatal CBV response in Fig. 3d and by the averaged percentage signal changes in Fig. 3e. Fig. 4 demonstrates the geometric correspondence between the correlation map and the TH staining. In the left half of the rostral striatum, as shown in Fig. 4a, the CBV decreases were as abundant as those of the TH staining. By contrast, in the right half of the rostral
Fig. 1. PD alters behavioral responses to nociceptive stimuli. (a) The hot-plate test showed that the response latency to a nociceptive thermal stimulus was significantly shorter in PD, indicating nociceptive hypersensitivity caused by dopaminergic impairment (p b 0.05). (b) The foot-shock test further indicated that the hypersensitivity to pain stimuli mainly occurred on the side of the body ipsilateral to the lesion (R.). * indicates p b 0.0001.
102
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Figs. 5c, d, e, and f illustrates the time courses of the signal changes in the left and right striata before and after eticlopride, respectively, in parkinsonian rats. Prior to eticlopride injection, the striatal response in the intact striatum appeared as signal elevations during the period of stimulation. Signal changes were weaker in the lesioned striatum. Following eticlopride injection, both striata exhibited minimal signal changes during stimulation. Fig. 5g presents the averaged percentage of signal changes in each striatum before and after eticlopride. Two-way ANOVA revealed that administration of the dopamine D2/D3 blocker significantly inhibited the CBV decreases after eticlopride treatment [F(1,8) = 8.8, p b 0.05]. Fisher's post-hoc tests indicated that the averaged activity was significantly higher on the left than on the right side before the treatment (p b 0.01). The activity was low in both hemispheres after the treatment. Laterality of the fMRI responses The CBV increases in S1 occurred contralateral to the side receiving nociceptive electrical stimulation, regardless of the dopaminergic lesion or antagonism. The right and left forepaws were stimulated for Figs. 1 and 4, respectively. The striatal CBV decreases occurred bilaterally, irrespective of the side receiving the nociceptive stimulation. However, the reduced responsiveness of the striatal CBV reactions occurred ipsilateral to the lesion of nigrostriatal dopaminergic inputs. The lesion was made on the right SN in the present study. Discussion
Fig. 2. In PD, the lesioned striatum showed a diminished CBV response. (a) The CBV correlation maps of the striatum in response to nociceptive stimulation. (b), (c) The time course of the signal intensities in the left and right hemispheres of the parkinsonian rats, respectively. (d) The average % signal change of the bilateral striatum in relation to the baseline during stimulation.
striatum, as shown in Fig. 4b, minimal residual CBV reactions were observed in the ventral portion. This was consistent with the location of the remaining TH staining in the ventral area. In the caudal striatum, the CBV response of the left half was distributed as widely as the TH staining (Fig. 4c). The right half did not exhibit any CBV signals, consistent with the absence of TH staining (Fig. 4d). Dopamine D2/D3 antagonism abolished the remaining CBV decreases in parkinsonism Figs. 5a and b presents the correlation maps before and after treatment with eticlopride in parkinsonian rats, respectively. The injection of eticlopride abolished all of the striatal CBV signals on the correlation maps, but did not affect the CBV response of S1.
The aim of the present study was to determine whether pain as a prevalent symptom of the nonmotor manifestations in PD can be characterized with specificity for the cardinal deficiency of the dopaminergic system, based upon an objective fMRI approach. Behavioral assessment first confirmed that hyperalgesia is indeed a symptom of PD. fMRI data then revealed that the lesioned striatum exhibited a diminished CBV response to the nociceptive stimulus. Immunohistology revealed that the reduced CBV responsiveness occurred mainly in areas with dopaminergic denervation. A strong correspondence was observed between the distributions of the nociception-induced CBV responses and dopaminergic innervations. The remaining CBV signals in the striatum were abolished by the D2/D3 antagonist eticlopride. The findings from the behavioral, neuroimaging, immunohistological, and pharmacological experiments demonstrate that pain in PD can be characterized as striatal CBV signal changes with specificity for dopaminergic dysfunction. These changes are a potential candidate for the objective identification of the nonmotor symptoms in PD. Pain is an under-recognized but important symptom in PD (Beiske et al., 2009; Djaldetti et al., 2004; Ford, 2010; Hanagasi et al., 2011; Reichling and Levine, 2011). It is often viewed as a secondary consequence to the motor disabilities. However, as many as 43% of PD patients experience painful sensations in the periphery even before their motor symptoms are diagnosed (Reichling and Levine, 2011). Neurodegeneration is increasingly accepted as a major cause of peripheral pain. The nigrostriatal pathway is known to respond to nociceptive stimuli and to perform an antinociceptive function (Brooks, 2006; Chudler and Dong, 1995). Many previous studies have demonstrated that lesioning striatal dopaminergic neurons increases pain sensitivity (Lin et al., 1984), whereas activating these neurons inhibits pain responses (Chudler and Dong, 1995). In particular, the D2/D3 receptors – and not the D1 receptor – is the main contributor to this behavior (Magnusson and Fisher, 2000). According to human reports, D2 agonists such as pergolide, ropinirole, and pramipexole, which are sometimes used for treatment, produced reductions in pain responses in PD patients (Linazasoro, 2008; Stuginski-Barbosa et al., 2008). These findings suggest an intimate relationship among pain and dopamine D2/D3 receptors in PD.
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
103
Fig. 3. Quantitative analysis of the correspondence between nociception-induced CBV responses and tyrosine hydroxylase (TH) staining. (a) A striatal section of PD that is stained by TH. (b) The averaged TH SNR of the left, intact and right, damaged striatum. (c) TH SNR is correlated with the count of striatal voxels on the correlation maps of PD. (d) TH SNR is correlated with the averaged % signal changes in the striatum of PD.
Noxious electrical stimulation induces stress and related CNS opioid release at the substantia nigra, which could affect dopaminergic functioning at the striatum (Drolet et al., 2001; Ostrowski and Pert, 1995). The dense expression of μ-opioid receptors in the substantia nigra and the striatum is likely responsible for the interactions between the opioid and dopaminergic systems (Tempel and Zukin, 1987). Our
recent study showed that the activation of opioid neurotransmission strengthened the striatal CBV vasoconstriction while the blockage diminished it (Shih et al., 2012). It seems that in a scenario that involves noxious stimuli, the role of opioids should be considered in the modulation of the striatal dopaminergic functions including the nociception-induced CBV signals and the pain response.
Fig. 4. Spatial correspondence between the striatal CBV decreases and the dopaminergic innervations. (a) In the left half of the rostral striatum, the CBV decreases were as abundant as those of the TH staining. (b) In the right half of the rostral striatum, minimal residual CBV reactions were observed in the ventral portion. This was consistent with the location of the remaining TH staining in the ventral area. (c) In the caudal striatum, the CBV response of the left half was distributed as widely as the TH staining. (d) The right half of the caudal striatum did not exhibit any CBV signals, consistent with the absence of TH staining.
104
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Fig. 5. The role of dopaminergic D2/D3 receptors in the striatal CBV responses of PD. (a) The CBV correlation maps before the eticlopride treatment. (b) The CBV correlation maps after the eticlopride treatment. (c) The time course of the signal intensities of the left striatum before eticlopride. (d) The time course of the signal intensities of the right striatum before eticlopride. (e) The time course of the signal intensities of the left striatum after eticlopride. (f) The time course of the signal intensities of the right striatum after eticlopride. (g) The averaged % signal change of each striatum before and after eticlopride.
The nociception-induced striatal CBV response is a negative fMRI response involving decreases in the CBV. This reaction has been reported repeatedly by different groups (Shih et al., 2009; Zhao et al., 2008a). This striatal response per se is an intriguing phenomenon. The activation of the dopaminergic D2/D3 receptors plays a key role in the generation of such response. Dopaminergic neurotransmission may affect the vascular tone by the activation of receptors on astrocytes or directly on the vessels (Choi et al., 2006). When astrocytes are involved, the released dopamine binds to the dopaminergic receptors on the astrocytes. Astrocytes then use their endfeet to regulate the vascular activity. An alternative mechanism involves direct activation of the receptors expressed on the vessels. But the exact cellular mechanism responsible for the striatal CBV response remains to be elucidated. The development of definite diagnostic markers for symptoms in PD is urgently required among the numerous priorities in the treatment of
the disease. Characterization of the nonmotor manifestations is particularly important, since it will (1) identify under-recognized nonmotor signs, (2) redefine the disease progression and stages, (3) facilitate the development of biomarkers, and (4) serve as a standard for the evaluation of the effects of new drugs. Nonmotor symptoms in PD such as depression, anxiety, olfactory dysfunction, fatigue, and sleep disorders have been of great interest to the research community in the recent decade. Clinical assessment by means of questionnaires, rating scales, and behavioral tests remains the most convenient, economical, and immediate evaluation tool available to neurologists. However, since most of the nonmotor symptoms are not specific (i.e., occurring in other clinical circumstances), outcome measures with specificity for the cardinal abnormalities of the disease should be mandatory for objective end points (Maetzler et al., 2009). Neuroimaging modalities including positron-emission tomography (PET), single-photon-emission computed tomography (SPECT),
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
diffusion-tensor MRI, and structural MRI have emerged as promising approaches for PD diagnosis (Antonelli et al., 2010; Brooks, 2006; Brooks and Pavese, 2010; Loane and Politis, 2011; Pavese and Brooks, 2009; Remy et al., 2005; Stoessl, 2009). In particular, imaging methods specific to the detection of dopaminergic dysfunctions have received the greatest attention, with PET and SPECT being the primary modalities. However, they are limited by their spatial resolution and radioactivity in nature. As such, dopaminergic imaging by MRI appears as an appealing option. fMRI has been shown to have the capacity to image dopaminergic activity (Jenkins et al., 2004). The present study extended the use of this technique and is the first to demonstrate the utility of fMRI in dopaminergic imaging of nonmotor manifestations in PD using a rat model. The observation of dopaminergic innervations and denervations represented by CBV signals is a clear illustration of the strength of the high spatial resolution offered by MRI. A significant step that has been made recently in the prospective development of biomarkers for PD is the launch of the Parkinson's Progression Markers Initiative (Foundation, 2011; The Lancet, 2010). The aim of this program is to collect clinical, imaging, and biological data to identify promising biomarkers of PD progression. Unsurprisingly, the characterization of nonmotor manifestations is an important pursuit of the consortium. Recent research trends appear to indicate that dopaminergic imaging offered either by fMRI or PET (in addition to the conventional tools) will be used to assess nonmotor manifestations to enhance the specificity and accuracy of clinical evaluations. Biomarker development is an iterative process. Several nonmotor symptoms such as depression are believed to antedate the motor disabilities and to be directly linked to dopaminergic dysfunctions (Brooks, 2006; Thobois et al., 2010). The development of these nonmotor manifestations as biomarkers that can be used to detect any early deficiencies associated with the dopaminergic system is therefore particularly important. We are also interested in the potential application of dopaminergic imaging of pain in PD by fMRI in the clinical setting. The present study focused on the validation of that concept. However, there are some uncertainties that should be further elucidated, such as the correlation with disease severity, the ethical issues of inducing pain in PD patients, and translation of the fMRI method to the clinical setting. Additionally, in PD patients, inducing pain to trigger the CBV response could be complicated by the individual differences as well as disease-modified changes. Several milestones need to be reached before developing the dopaminergic imaging of pain in PD as a biomarker. Acknowledgment We acknowledge the technical support provided by Mr. Nai-Wei Yao and would like to thank the Functional and Micro-Magnetic Resonance Imaging Center supported by the National Research Program for Genomic Medicine, National Science Council, Taiwan, Republic of China, NSC100-3112-B-001-009. References Antonelli, F., et al., 2010. Imaging cognitive and behavioral symptoms in Parkinson's disease. Expert Rev. Neurother. 10, 1827–1838. Bandettini, P.A., et al., 1993. Processing strategies for time-course data sets in functional MRI of the human brain. Magn. Reson. Med. 30, 161–173. Bannon, A.W., Malmberg, A.B., 2007. Models of nociception: hot-plate, tail-flick, and formalin tests in rodents. Curr. Protoc. Neurosci. (Chapter 8, Unit 8 9). Baudrexel, S., et al., 2010. Quantitative mapping of T1 and T2* discloses nigral and brainstem pathology in early Parkinson's disease. Neuroimage 51, 512–520. Beiske, A.G., et al., 2009. Pain in Parkinson's disease: prevalence and characteristics. Pain 141, 173–177. Berendse, H.W., Ponsen, M.M., 2009. Diagnosing premotor Parkinson's disease using a two-step approach combining olfactory testing and DAT SPECT imaging. Parkinsonism Relat. Disord. 15 (Suppl. 3), S26–S30. Brefel-Courbon, C., et al., 2005. Effect of levodopa on pain threshold in Parkinson's disease: a clinical and positron emission tomography study. Mov. Disord. 20, 1557–1563.
105
Brooks, D.J., 2006. Dopaminergic action beyond its effects on motor function: imaging studies. J. Neurol. 253 (Suppl. 4), IV8–IV15. Brooks, D.J., Pavese, N., 2010. Imaging non-motor aspects of Parkinson's disease. Prog. Brain Res. 184, 205–218. Chang, C., Shyu, B.C., 2001. A fMRI study of brain activations during non-noxious and noxious electrical stimulation of the sciatic nerve of rats. Brain Res. 897, 71–81. Chaudhuri, K.R., Naidu, Y., 2008. Early Parkinson's disease and non-motor issues. J. Neurol. 255 (Suppl. 5), 33–38. Chaudhuri, K.R., Schapira, A.H., 2009. Non-motor symptoms of Parkinson's disease: dopaminergic pathophysiology and treatment. Lancet Neurol. 8, 464–474. Chaudhuri, K.R., et al., 2007. The metric properties of a novel non-motor symptoms scale for Parkinson's disease: results from an international pilot study. Mov. Disord. 22, 1901–1911. Choi, J.K., et al., 2006. Brain hemodynamic changes mediated by dopamine receptors: role of the cerebral microvasculature in dopamine-mediated neurovascular coupling. Neuroimage 30, 700–712. Chudler, E.H., Dong, W.K., 1995. The role of the basal ganglia in nociception and pain. Pain 60, 3–38. Djaldetti, R., et al., 2004. Quantitative measurement of pain sensation in patients with Parkinson disease. Neurology 62, 2171–2175. Drolet, G., et al., 2001. Role of endogenous opioid system in the regulation of the stress response. Prog. Neuropsychopharmacol. Biol. Psychiatry 25, 729–741. Ford, B., 2010. Pain in Parkinson's disease. Mov. Disord. 25 (Suppl. 1), S98–S103. Foundation, T.M.J.F., 2011. The Parkinson Progression Marker Initiative (PPMI). Prog. Neurobiol. 95, 629–635. Hanagasi, H.A., et al., 2011. Pain is common in Parkinson's disease. Clin. Neurol. Neurosurg. 113, 11–13. Jenkins, B.G., et al., 2004. Mapping dopamine function in primates using pharmacologic magnetic resonance imaging. J. Neurosci. 24, 9553–9560. Kotz, S.A., et al., 2009. Non-motor basal ganglia functions: a review and proposal for a model of sensory predictability in auditory language perception. Cortex 45, 982–990. Leroi, I., et al., 2011. Behavioural disorders, disability and quality of life in Parkinson's disease. Age Ageing 40, 614–621. Lin, M.T., et al., 1984. Effects of kainic acid injections in the striatum on physiologic and behavioral functions in conscious rats. Exp. Neurol. 83, 71–83. Linazasoro, G., 2008. Conversion from dopamine agonists to cabergoline: an open-label trial in 128 patients with advanced Parkinson disease. Clin. Neuropharmacol. 31, 19–24. Liu, Z.M., et al., 2004. Imaging oxygen consumption in forepaw somatosensory stimulation in rats under isoflurane anesthesia. Magn. Reson. Med. 52, 277–285. Loane, C., Politis, M., 2011. Positron emission tomography neuroimaging in Parkinson's disease. Am. J. Transl. Res. 3, 323–341. Maetzler, W., et al., 2009. Progression of Parkinson's disease in the clinical phase: potential markers. Lancet Neurol. 8, 1158–1171. Magnusson, J.E., Fisher, K., 2000. The involvement of dopamine in nociception: the role of D(1) and D(2) receptors in the dorsolateral striatum. Brain Res. 855, 260–266. Mandeville, J.B., et al., 2004. Exogenous contrast agent improves sensitivity of gradient-echo functional magnetic resonance imaging at 9.4 T. Magn. Reson. Med. 52, 1272–1281. Martinez-Martin, P., et al., 2011. The impact of non-motor symptoms on health-related quality of life of patients with Parkinson's disease. Mov. Disord. 26, 399–406. Ostrowski, N.L., Pert, A., 1995. Substantia nigra opiate receptors on basal ganglia efferents. Brain Res. Bull. 38, 513–523. Park, A., Stacy, M., 2009. Non-motor symptoms in Parkinson's disease. J. Neurol. 256 (Suppl. 3), 293–298. Pavese, N., Brooks, D.J., 2009. Imaging neurodegeneration in Parkinson's disease. Biochim. Biophys. Acta 1792, 722–729. Pertovaara, A., et al., 2004. Striatal dopamine D2/D3 receptor availability correlates with individual response characteristics to pain. Eur. J. Neurosci. 20, 1587–1592. Reichling, D.B., Levine, J.D., 2011. Pain and death: neurodegenerative disease mechanisms in the nociceptor. Ann. Neurol. 69, 13–21. Remy, P., et al., 2005. Depression in Parkinson's disease: loss of dopamine and noradrenaline innervation in the limbic system. Brain 128, 1314–1322. Shih, Y.Y.I., et al., 2009. A new scenario for negative fMRI signals: endogenous neurotransmission. J. Neurosci. 29, 3036–3044. Shih, Y.Y., et al., 2011. Striatal and cortical BOLD, blood flow, blood volume, oxygen consumption, and glucose consumption changes in noxious forepaw electrical stimulation. J. Cereb. Blood Flow Metab. 31, 832–841. Shih, Y.Y., et al., 2012. Endogenous opioid–dopamine neurotransmission underlie negative CBV fMRI signals. Exp. Neurol. 234, 382–388. Shmuel, A., et al., 2002. Sustained negative BOLD, blood flow and oxygen consumption response and its coupling to the positive response in the human brain. Neuron 36, 1195–1210. Simuni, T., Sethi, K., 2008. Nonmotor manifestations of Parkinson's disease. Ann. Neurol. 64 (Suppl. 2), S65–S80. Solla, P., et al., 2011. Behavioral, neuropsychiatric and cognitive disorders in Parkinson's disease patients with and without motor complications. Prog. Neuropsychopharmacol. Biol. Psychiatry 35, 1009–1013. Stiasny-Kolster, K., et al., 2005. Combination of ‘idiopathic’ REM sleep behaviour disorder and olfactory dysfunction as possible indicator for alpha-synucleinopathy demonstrated by dopamine transporter FP-CIT-SPECT. Brain 128, 126–137. Stoessl, A.J., 2009. Functional imaging studies of non-motoric manifestations of Parkinson's Disease. Parkinsonism Relat. Disord. 15 (Suppl. 3), S13–S16. Stuginski-Barbosa, J., et al., 2008. Burning mouth syndrome responsive to pramipexol. J. Headache Pain 9, 43–45.
106
C.-C.V. Chen et al. / Neurobiology of Disease 49 (2013) 99–106
Tassorelli, C., et al., 2007. Behavioral responses and Fos activation following painful stimuli in a rodent model of Parkinson's disease. Brain Res. 1176, 53–61. Tempel, A., Zukin, R.S., 1987. Neuroanatomical patterns of the mu, delta, and kappa opioid receptors of rat brain as determined by quantitative in vitro autoradiography. Proc. Natl. Acad. Sci. U. S. A. 84, 4308–4312. The Lancet, N., 2010. Biomarker promise for Parkinson's disease. Lancet Neurol. 9, 1139. Thobois, S., et al., 2010. Non-motor dopamine withdrawal syndrome after surgery for Parkinson's disease: predictors and underlying mesolimbic denervation. Brain 133, 1111–1127.
Tolosa, E., Pont-Sunyer, C., 2011. Progress in defining the premotor phase of Parkinson's disease. J. Neurol. Sci. 310, 4–8. Zhao, F., et al., 2008a. BOLD and blood volume-weighted fMRI of rat lumbar spinal cord during non-noxious and noxious electrical hindpaw stimulation. Neuroimage 40, 133–147. Zhao, F., et al., 2008b. BOLD study of stimulation-induced neural activity and restingstate connectivity in medetomidine-sedated rat. Neuroimage 39, 248–260. Zhao, F., et al., 2012. fMRI of pain processing in the brain: a within-animal comparative study of BOLD vs. CBV and noxious electrical vs. noxious mechanical stimulation in rat. Neuroimage 59, 1168–1179.