Evaluation of Four Pyridine Analogs to Characterize 6-OHDA-Induced Modulation of mGluR5 Function in Rat Brain Using microPET Studies

Abstract

Micro-positron emission tomography imaging studies were conducted to characterize modulation of metabotropic glutamate subtype-5 receptor (mGluR5) function in a 6-hydroxydopamine (6-OHDA)-induced rat model of Parkinson's disease using four analogical PET ligands: 2-[¹¹C]methyl-6-(2-phenylethynyl) pyridine ([¹¹C]MPEP), 2-(2-(3-[¹¹C]methoxyphenyl)ethynyl)pyridine ([¹¹C]M-MPEP), 2-(2-(5-[¹¹C]methoxypyridin-3-yl)ethynyl)pyridine ([¹¹C]M-PEPy), and 3-[(2-[¹⁸F]methyl-1,3-thiazol-4-yl)ethynyl]pyridine ([¹⁸F]M-TEP). A total of 45 positron emission tomography (PET) imaging studies were conducted on nine male Sprague-Dawley rats within 4 to 6 weeks after unilateral 6-OHDA lesioning into the right medial forebrain bundle. The severity of the lesion was determined with [¹¹C]CFT ([¹¹C]2-β-carbomethoxy-3-β-(4-fluorophenyl)tropane), a specific and sensitive ligand for imaging dopamine transporter function. The binding potential (BP) images were processed on pixel-by-pixel basis by using a method of the distribution volume ratio with cerebellum as a reference tissue. The values for BP were determined on striatum, hippocampus, and cortex. [¹¹C]CFT binding was decreased on the lesioned (right) striatum by 35.4% ± 13.4% compared with the intact left striatum, indicating corresponding loss of presynaptic dopamine terminals. On the same areas of the lesioned striatum, three of the four tested mGluR5 ligands showed enhanced binding characteristics. The average differences between the right and left striatum were 4.4% ± 6.5% (P < 0.05) with [¹¹C]MPEP, 0.1% ± 1.7% (P > 0.05) with [¹¹C]M-MPEP, 3.9% ± 4.6% (P < 0.05) with [¹¹C]M-PEPy, and 6.6% ± 2.7% (P > 0.05) with [¹⁸F]M-TEP. The enhanced binding was also observed in the right hippocampus and cortex. These studies showed that glutamatergic neurotransmission might have a complementary role in dopaminergic degeneration, which can be evaluated by in vivo PET imaging.

Keywords

distribution volume dopamine transporter metabotropic glutamate receptor Parkinson's disease positron emission tomography striatum

Introduction

Metabotropic glutamate receptors (Catania et al, 1994) are functional components of the glutamatergic system (Krieger et al, 2000; Meldrum, 2000). These receptors are affected on several intracellular signal transduction mechanisms through G-proteins, having various physiological functions in the mammalian central nervous system (Bruno et al, 2001). According to their signal transduction mechanism, amino-acid sequence homologies, and pharmacological properties, metabotropic glutamate receptors have been identified as having eight subtypes, divided into three groups (Conn and Pin, 1997; Bhave et al, 2003; Schaffhauser et al, 2003). The metabotropic glutamate subtype-5 receptors (mGluR5s) are included in the first group. It has been reported that many neurological disorders such as Parkinson's disease (PD), Schizophrenia, mood disorders, drug addiction, and pain might be affected by the direct or indirect effects of mGluR5s. Metabotropic glutamate subtype-5 receptors have also been tested as potential neuroprotective drugs (Chiamulera et al, 2001; Marino et al, 2003; Sotgiu et al, 2003; Brody and Geyer, 2004).

Increased glutamatergic activity in the basal ganglia circuitry is involved in movement abnormalities in PD (Wichmann and DeLong, 1997). The mGluR5s are implicatively connected to the subthalamic nucleus from glutamatergic afferents (Awad et al, 2000). The mGluR5s are localized in cholinergic striatal neurons (Marti et al, 2001; Bell et al, 2002), whose excessive excitation was observed in the basal ganglia. Therefore, inhibition of mGluR5 binding has been proposed as a therapeutic approach for PD (Testa et al, 1994; Breysse et al, 2002; Coccurello et al, 2004).

In this study, modulation of mGluR5s was investigated in 6-hydroxydopamine (6-OHDA) lesioned rat brain by in vivo positron emission tomography (PET) imaging using four pyridine derivatives: [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy (Yu et al, 2005), and [¹⁸F]M-TEP.

6-Hydroxydopamine is a selective catecholaminergic neurotoxin (Ungerstedt, 1968; Sachs and Jonsson, 1975) and it is structurally similar to dopamine (DA). 6-Hydroxydopamine can be taken up by the DA transporters and concentrated in DA cells, resulting in degeneration of DA neurons. Various PD models in different species such as rodents and small monkeys have been developed by the injection of 6-OHDA into the substantia nigra, medial forebrain bundle, or intrastriatal locations (Breese and Breese, 1998; Shults et al, 2000; Kirik et al, 2001; Betarbet et al, 2002; Zafar et al, 2003). A noticeable benefit of applying unilateral lesioning is that the loss of DA neurons in the lesion side can be investigated by comparing with the intact part on the contralateral side.

In this study, the rat model of PD was induced by unilateral stereotaxic 6-OHDA injection into the right side of the medial forebrain bundle. The severity of the lesion was determined by PET imaging of DA transporter binding using [¹¹C]CFT, a selective and sensitive ligand for DA transporter (Brownell et al, 1996). After PET imaging with mGluR5 ligands and [¹¹C]CFT binding potential images were processed using a modified graphical method of Logan (Logan et al, 1990, 1996; Logan, 2000). In addition, representative regions of interests (ROIs) were drawn to the left and right brain areas on the original images of activity distribution. Time-dependent activity curves were processed and used in the graphical method to determine local binding characteristics to investigate 6-OHDA-induced differences between the right and left brain areas, including striatum, hippocampus, and cortex.

Materials and methods

6-Hydroxydopamine Rat Model

Nine male Sprague–Dawley rats (weigh between 300 and 350 g) were anesthetized with ketamine (80 mg/kg intraperitoneal) and xylazine (10 mg/kg intraperitoneal), and placed into the Kopf stereotaxic frame. To restrict the effect of 6-OHDA on the dopaminergic neurons, the animals received desipramine (20 mg/kg intraperitoneal) 30 mins before the 6-OHDA injection. A unilateral injection of 6-OHDA (8 μg in 2 μL) was conducted into the medial forebrain bundle on the right side of the rat brain (coordinates: AP −4.5; ML +1.2; DV −7.6) using a Hamilton syringe (Cicchetti et al, 2002). All experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Policy by the Subcommittee on Research Animals of the Harvard Medical School and Massachusetts General Hospital.

Radioligands

The radiosyntheses of [¹¹C]MPEP, [¹¹C]M-MPEP, and [¹¹C]M-PEPy were performed according to Yu et al (2005, 1999). For the radiosynthesis of [¹⁸F]M-TEP, the precursor 2-bromo-5-(2-(2-methylthiazol-4-yl)ethynyl)pyridine (∼2.0 mg) was added to the dried [¹⁸F]fluoride/Kryptofix 2.2.2 vial with dry dimethylsulfoxide (0.5 mL). The vial was heated at 100°C for 10 mins. After cooling for 5 mins, a sample was taken to radio thin layer chromatography, showing that 72% of the radioactivity was in the desired product, 2-[¹⁸F]fluoro-5-(2-(2-methylthiazol-4-yl)ethynyl)-pyridine, that is, [¹⁸F]M-TEP, and the product was identified with an authentic reference standard. The reaction mixture was diluted with 3 ml ethyl acetate, loaded to a silica Sep-Pak, and eluted with an additional 5 ml ethyl acetate. The ethyl acetate solution was evaporated at 130°C under a N₂ flow. The residue (∼0.5 mL) was cooled for 5 mins, diluted with high-performance liquid chromatography solvent (1.5 mL), and transferred to a high-performance liquid chromatography injection vial. By using a μ-Bondapak column and a mobile phase of 40% methanol and 60% phosphate buffer (pH 7.4, 0.15 mmol/L), the pure tracer [¹⁸F]M-TEP was isolated.

Radiosynthesis of [¹¹C]CFT was performed according to Brownell et al (1996).

Positron Emission Tomography Imaging

Forty-five PET imaging studies were conducted in nine 6-OHDA-lesioned rats. After a rat was anesthetized by gas (1% to 1.5% isoflurane, 2% nitrous oxide, and oxygen flow of 3 L/min), the tail vein was catheterized for administration of radiolabeled ligands. The head was adjusted securely in a stereotaxic head holder equipped with a gas inhalation system. Then the rat was placed ventrally into the scanner at the center of the imaging field, where the spatial resolution is 1.8 mm (microPET, P4, Concorde Microsystems, Knoxville, TN, USA). A heated waterbed and an infrared light were used to maintain body temperature. Heart rate was monitored during the whole imaging time (HESKA Vet/Ox Plus 4800). Before administering radioactivity, a transmission imaging was conducted using a rotating germanium-68 pin source to create maps for attenuation correction. After that, a radiolabeled ligand was injected into the tail vein and volumetric dynamic data were acquired in list-mode format for 60 mins. Dopamine transporter binding was investigated with [¹¹C]CFT and studies were conducted twice on each animal. The severity of the lesion was investigated 2 weeks after 6-OHDA lesioning and 4 weeks later, at the time of the studies of mGluR5 function to accurately determine dopaminergic deficit. The mGluR5 binding was investigated with [¹¹C]MPEP in each animal and repeated in two rats, [¹¹C]M-MPEP in six rats, [¹¹C]M-PEPy in seven rats, and [¹⁸F]M-TEP in three rats.

Image Analysis

Dynamic imaging data were corrected for uniformity, attenuation, and decay factors, and processed by filtered back projection with the ramp filter (cutoff value of 0.5). The software, ASIPro 6.0, was provided by the manufacturer. The binding characteristics of the PET ligands were analyzed using a modified distribution volume (DV) method of Logan (Logan et al, 1990, 1996; Logan, 2000). The Logan linear model has become a standard analyzing method for receptor–ligand studies to determine the DV of ligands, which have reversible binding kinetics. If the metabolite corrected plasma data are available, the DVs can be calculated from the region of interest and reference blood data to obtain the distribution volume ratio (DVR). If metabolite corrected plasma curves are not available, the plasma curve can be replaced with the reference region curve C_r(t) as long as the reference region is considered receptor-free. The slope of the linear phase is DVR with the following formula:

\frac{\int_{0}^{T} C_{t} (t) dt}{C_{t} (T)} = D V R \times \frac{\int_{0}^{T} C_{r} (t) dt}{C_{t} (T)} + C^{'}

(1)

where C_t(t) is the radioactivity measured by the PET at time t in a specified area (single pixel or averaged ROI over the selected pixels) and C_r(t) is the radioactivity in reference region curve at time t. For time T > t*, the DVR can be found as the slope of a plot of

\int_{0}^{T} C_{t} (t) d t

versus

\int_{0}^{T} C_{r} (t) d t

and the ratio of C_t(t)/C_r(t) stays stable. The time t* can be determined as the time after which no further significant increase in the slope is observed. The binding potential (BP) values can be obtained as BP = DVR–1. In practice, considering the noncontinuousness of time-dependent activity curves obtained from dynamic PET images, equation (1) has to be modified as

\frac{\sum_{0}^{T} C_{t} (t) Δ t}{C_{t} (T)} = D V R \times \frac{\sum_{0}^{T} C_{r} (t) Δ t}{C_{t} (T)} + C^{'}

(2)

where Δt is the time interval for time-dependent activity curves. Assuming that cerebellum has 0 or almost 0 receptor density, the slopes, that is DVRs, can be obtained by graphical fitting (Figure 3).

When calculating BP images, the voxels outside the head were eliminated by using binary 3D images as space filters. Each filter was created separately from the coronal slices of the original dynamic volumetric images by adding images between 5 and 15 mins to create a static summary image. After thresholding, these images were converted to binary images having a value of 1 for the data inside the head and 0 outside the head. The original images were multiplied with this filter and used in further calculations of binding values on pixel-by-pixel bases by using equation (2) (Figure 3). The total number of the calculated voxels was 10^∧5 and the calculation speed was 406 voxels/sec by using x86-based PC with 1,395 MHz processor.

Two methods were applied for drawing the regions of interest on the PET images. If the studies with an mGluR5 ligand and [¹¹C]CFT were performed in the same imaging session, the ROIs were drawn on [¹¹C]CFT images and the same ROIs were directly loaded to the PET images of the mGluR5 ligand. This type of sequential imaging allows analyses of the same brain regions and enables accurate local comparison of the mGluR5 and DA transporter data, as long as there is no movement of animals during the acquisition of PET data.

The other method includes fusing the mGluR5 images to [¹¹C]CFT images on the same rat. The merging procedure was performed manually on a pixel-by-pixel basis by using the Harderian glands, the olfactory bulb, and the whole-brain contours as internal markers. After the fusion procedures of the mGluR5 and DA transporter images, ROIs were drawn on the [¹¹C]CFT images and applied to the PET images of the mGluR5 ligands. Regions of interests were drawn on the right and left striatum, cortex, hippocampus, and cerebellum, and the corresponding time-dependent activity curves were processed.

Percent differences of the right–left regions were calculated for striatum, hippocampus, and cortex by the following equation:

Φ = \frac{D V R_{r i g h t} - D V R_{l e f t}}{D V R_{l e f t}} \times 100

(3)

Thus, the percent difference (φ) of DVRs of [¹¹C]CFT binding in striatum is a measure of the deficit in the DA transporter function, as well as a measure of the severity of damage on presynaptic DA terminals. Correspondingly, the percent differences (φ) of DVRs of [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP binding are measures of the modulation of mGluR5 ligands in the striatum of PD-model of rats. The percent differences were calculated also on the cortex and hippocampus for the used PET ligands.

Statistical Analysis and P-value

The statistical significance of the changes between the right and left brain areas was calculated using a paired t-test (Whitley and Ball, 2002) (Table 1). The used t-test compares the observed mean difference with the hypothesized value of 0. The t-statistic for the paired t-test is

t = \frac{\bar{φ}}{s \cdot d \cdot \sqrt{N}}

(4)

Table 1

Sensitivity of the different mGluR5 ligands to express 6-OHDA-induced modulation of mGluR5 function determined as the average percent differences between the right and left striatum, hippocampus, and cortex. The [¹¹C]CFTPET studies indicate the severity of the 6-OHDA lesion

		The average percentage differences of the right–left ROIs of rat brain
ROIs		[¹¹C]CFT	[¹¹C]MPEP	[¹¹C]M-MPEP	[¹¹C]M-PEPy	[¹⁸F]M-TEP
Striatum	Mean	−35.4	4.4	−0.1	3.9	6.6
	s.d.	13.4	6.5	1.7	4.4	2.7
	P-value	3.5E-07	0.0495	0.8604	0.0498	0.0520
Hippocampus	Mean	−4.0	6.2	7.6	0.4	6.0
	s.d.	3.2	4.3	3.3	7.2	4.6
	P-value	5.5E-05	0.0007	0.0026	0.4500	0.1524
Cortex	Mean	2.6	10.2	7.4	6.9	5.8
	s.d.	5.9	11.3	6.2	7.4	6.4
	P-value	0.0768	0.0140	0.0329	0.0488	0.2571
PET studies for ligands		18	11	6	7	3

where s.d. is the standard deviation. Significance was determined based on t-distribution. Practically, the P-values were calculated directly using the t-test function in Microsoft Excel 2002 SP3.

Results

Figure 1 shows PET images of activity distribution of the DA transporter ligand [¹¹C]CFT and four mGluR5 ligands: [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP The left striatum and hippocampus have high [¹¹C]CFT accumulation, whereas mGluR5 ligands have high accumulation in the right striatum and hippocampus, where [¹¹C]CFT accumulation is decreased because of a 6-OHDA-induced defect on DA transporter function.

Figure 1

Accumulation of [¹¹C]CFT, [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP in the 6-OHDA-lesioned rat brain. Distribution of mGluR5 ligands is illustrated at 7 to 10 mins and [¹¹C]CFT at 25 to 40 mins after administration of the ligand (1.62 ± 0.69 mCi). The first column of the coronal images is at the striatum level, the second column is at the hippocampus level, and the third column is at the cerebellum level. The axial views show the distribution of the radioactivity around the midstriatal level.

Time–activity curves were generated from several brain areas including cerebellum, right and left striatum, hippocampus, and cortex (Figure 2A). The number of the voxels included to the ROIs was 155 ± 43 for the left striatum, 119 ± 39 for the right striatum, 103 ± 27 for the left hippocampus, 102 ± 22 for the right hippocampus, 66 ± 22 for the left cortex, 69 ± 20 for the right cortex, and 96 ± 20 for the cerebellum. The corresponding Logan plots of these curves are shown in Figure 2B. The straight lines (Logan plots) were obtained using equation (2) with T = 10 mins. For all compounds, the Logan plots show linearity starting at 10 mins after injection of the ligand.

Figure 2

(A) Time–activity curves illustrating time-dependent accumulation of [¹¹C]CFT, [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP in different brain areas (cerebellum, right and left striatum, and hippocampus) in the rat model of PD induced by 6-OHDA. (B) Logan plots for the time–activity curves in (A) show linearity: left striatum (l-str), right striatum (r-str), left hippocampus (l-hipp), right hippocampus (r-hipp), left cortex (l-cort) and right cortex (r-cort). Integration Y = DVR*IntegrationX + C, where Integration Y = Σ^T₀G_t(t)Δt/G_t(T), IntegrationX = Σ^T₀G_t(t)Δt/G_t(T). The values of the slopes (DVR) are included in the legends.

Figure 3 shows the BP images for the DA transporter ligand [¹¹C]CFT and four mGluR5 ligands [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP These images were created based on the activity distributions shown in Figure 1. High BP values of [¹¹C]CFT can be observed in the left striatum and hippocampus, whereas high BP values of the mGluR5 ligands can be observed in the right striatum and hippocampus, which is consistent with the PET images in Figure 1. By comparing Figures 1 and 3, it is observed that the BP images of each ligand have higher contrast than the corresponding PET images of activity distribution.

Figure 3

Three coronal BP images of [¹¹C]CFT, [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP in the 6-OHDA-lesioned rat brain processed pixel-by-pixel basis using a modified Logan graphical method. The first column of the coronal images is at the striatum level, the second at the hippocampus level, and the third at the cerebellum level. The axial views illustrate the distribution of the radioactivity at the midstriatal level. All these levels are exactly the same as shown in Figure 1.

The percent differences between the right and left brain areas were calculated using equation (3), and the values (φ, s.d., and P) for the striatum, hippocampus, and cortex are shown in Figure 4 and Table 1. [¹¹C]CFT binding was decreased on the lesioned striatum (right side) by 35.4% ± 13.4% compared with the left intact side, indicating a corresponding loss of presynaptic DA terminals. In contrast, on the same area of the lesioned striatum, three of four tested mGluR5 ligands showed increased binding. The average differences of DVR between the right and left striatum were 4.4% ± 6.5% (P < 0.05) with [¹¹C]MPEP, −0.1% ± 1.7% (P > 0.05) with [¹¹C]M-MPEP, 3.9% ± 4.4% (P < 0.05) with [¹¹C]M-PEPy, and 6.6% ± 2.7% (P > 0.05) with [¹⁸F]M-TEP. In the hippocampus, the percent differences between the right and left sides were 6.2 ± 4.3 (P < 0.005) with [¹¹C]MPEP, 7.6 ± 3.3 (P < 0.005) with [¹¹C]M-MPEP, 0.4 ± 7.2 (P > 0.05) with [¹¹C]M-PEPy, and 6.0 ± 4.6 (P > 0.05) with [¹⁸F]M-TEP In the cortical area, the right–left side differences were 10.2% ± 11.3% (P < 0.05) with [¹¹C] MPEP, 7.4% ± 6.2% (< 0.05) with [¹¹C]M-MPEP, 6.9% ± 7.4% (P < 0.05) with [¹¹C]M-PEPy, and 5.8% ± 6.4% (P > 0.05) with [¹⁸F]M-TEP, as shown in Table 1.

Figure 4

The percent differences of DVRs between the right and left brain areas (including striatum, hippocampus, and cortex) in 6-OHDA-lesioned rat brain for DA transporter ligand [¹¹C]CFT and four mGluR5 ligands: [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP (*P < 0.05 and **P < 0.005).

Discussion

The pyridine derivatives used to investigate mGluR5 function have a reversible binding profile with maximum binding in 1 to 3 mins. Rapid changes in the accumulation of the activity in small brain areas create significant fluctuation of data and cause difficulties in direct mathematical fitting using a three-compartment model. For that reason, we developed a modified graphical model of the Logan method for data analysis (Logan et al, 1990, 1996; Logan, 2000). To increase the accuracy of the outcome measures from the same rat, we used cerebellum as a reference tissue, as the selection of the region is repeatable; based on in situ hybridization histochemical and immunohistochemical studies, adult rat brain has only weak expression of mGluR5 in Golgi cells and cerebellar nuclei in the cerebellum (Shigemoto and Mizuno, 2000). As observed in Figure 2A, each mGluR5 ligand has faster washout from each brain area than [¹¹C]CFT Owing to the low contrast in the PET images of mGluR5 ligands, it is difficult to draw representative anatomical borderlines of striatum, hippocampus, cortex, or cerebellum. Instead, using the [¹¹C]CFT images, it is easy to segment these areas as shown in Figure 1. For this reason, the ROIs in PET images of mGluR5 were obtained from [¹¹C]CFT studies. When the PET studies of DA transporters with [¹¹C]CFT were performed in the same imaging session as mGluR5 studies, the ROIs were extracted from [¹¹C]CFT images and used in both studies. When the studies were performed in different imaging sessions, the anatomical landmarks of Harderian glands, olfactory bulbs, and whole-brain contours were used as internal markers, and the images of mGluR5 ligands and [¹¹C]CFT were fused on pixel-by-pixel basis. The ROIs determined from [¹¹C]CFT images were used in both studies. This allows precisely the same anatomical localization to be applied for generation of the quantitative data in both studies.

Partial volume effect results from the limited spatial resolution of PET imaging. It may cause significant underestimation in the measurement of radioactivity concentration in small brain areas, and furthermore creates systematic errors in evaluating quantitative physiological parameters (Hoffman et al, 1979). In our studies, there are two aspects that may suppress the partial volume effect systematic errors: one is the use of cerebellum as the reference tissue in equation (2); the other is the comparison of the right (lesioned) and the left (control) brain regions of the same rats by calculating percent differences in equation (3). The average difference between the selected number of the voxels on the right and left brain regions is 1 to 4 voxels. In addition, the same ROIs on the left and right sides were used to obtain data of DA transporter and mGluR5 ligands. Therefore, partial volume effect is not currently considered as a major effect on our results.

As shown in Figure 2A, the accumulation of [¹¹C]MPEP, [¹¹C]M-MPEP, and [¹¹C]M-PEPy achieved the maximum value within 1 min after injection of the activity and then decreased to the same level as cerebellum in about 15 mins, whereas [¹⁸F]M-TEP achieved its peak accumulation within about 3 mins and decreased to the cerebellum level in about 18 mins. After 20 mins, mGluR5 ligands have only minor differences in accumulation of the right and left striatum and hippocampus. With the exception of the plots of the right striatum and hippocampus with [¹¹C]MPEP and [¹¹C]M-PEPy, the Logan plots showed good linearity when T was set to 10 mins. The areas of right-side striatum and hippocampus were directly affected by 6-OHDA lesioning, which might affect the binding mechanism of MPEP and M-PEPy. After setting T to 15 min or 20 mins, a better linearity of Logan plots was obtained in these brain areas.

The aim of these studies was to compare four mGluR5 ligands to characterize 6-OHDA-induced modulation of the function of the metabotropic glutamate subtype-5 receptors in rat brain. Based on Logan graphic method (equation (1)), the values of DVRs as shown in Figure 2B can express modulation of mGluR5 function as numeric binding parameters. Therefore, the differences in the accumulation between 6-OHDA lesioned and control brain areas can be obtained simply by calculating the percent difference as shown in Eq. (3).

As displayed in Figure 4 and Table 1, the percent differences of four mGluR5 ligands indicate enhanced glutamatergic function around the 6-OHDA-affected regions, whereas the percent differences of the DA transporter ligand indicate the severity of the 6-OHDA-induced lesion in the same site. These results are consistent with PET images of activity distribution (Figure 1) and BP (Figure 3). Taken together, these results indicate a complementary role of metabotropic glutamine subtype-5 receptors during or after dopaminergic degeneration. In other words, the 6-OHDA-lesioned areas in the rat brain have enhanced function of the metabotropic glutamine receptors, and this increase may compensate the effect of dopaminergic degeneration. Earlier in vitro studies have shown that rat striatal mGluR5 exerts a direct control on the phosphorylation of transmembrane DA transporter protein by decreasing the efficiency in re-uptake/release of DA at the presynaptic terminals (Page et al, 2001).

Although the same dose of 6-OHDA was injected into each rat, the loss of presynaptic DA terminals in individual rats was not the same. The s.d. was 13.4% as shown in Figure 4 and Table 1. It should be noted that the microPET P4 system is stable and the experimental imaging techniques are well defined and they do not contribute to such large deviation. The individual discrepancy of degeneration in presynaptic DA terminals might be caused by individual differences of the animals, including their head size, their health status, and ability to recover. It is reasonable to believe that the individual variances of the sizes of brains might cause a little excursion on the localization when researchers were positioning the injection sites for 6-OHDA by the Kopf stereotaxic frame.

The average percent differences between the brain areas were analyzed statistically and their P-values were compared with 0.05 or 0.005. The results are tabulated in Table 1. In addition, Figure 4 and Table 1 show that the mGluR5 ligands have different binding profiles in different brain areas. This might result from the different lipophilicity, molecular size, and polarity of the ligands (Cosford et al, 2003; Yu et al, 2005). All four mGluR5 ligands are pyridine derivatives with high specific radioactivity and high radiochemical purity. Three of the four ligands showed enhanced glutamatergic function in the right-side lesioned striatum, hippocampus, and cortex of the rat model of PD, although their binding values were different.

Based on the percent differences of the binding values between the right and left striatum, the investigated pyridine derivatives can be ranked in the following order: [¹⁸F]M-TEP ∼ [¹¹C]MPEP > [¹¹C]M-PEPy > [¹¹C]M-MPEP. However, studies with [¹⁸F]M-TEP were conducted in only three animals and the obtained percent differences were not significant (P > 0.05) for the three brain areas because of the small sample number. The advantage of [¹⁸F]M-TEP over the three other ligands is that the radiolabeled fluorine-18 (T½ = 110mins) allows transportation of the labeled ligand to users who do not have an accelerator in the proximity.

PET studies of DA transporter function were performed twice on each animal. The first study was performed 2 weeks after 6-OHDA administration to determine the immediate extent of the induced lesion and the second study 4 weeks later to accurately determine dopaminergic deficit at the time of the mGluR5 studies. There were no significant differences between the two imaging studies of DA transporter function, indicating that the 6-OHDA-induced lesion was stable during the 4-week research period.

In summary, the graphical analyses based on Logan plots can be used to analyze binding characteristics of mGluR5 ligands: [¹¹C]MPEP, [¹¹C]M-MPEP, [¹¹C]M-PEPy, and [¹⁸F]M-TEP. [¹¹C]CFT biding was reduced in striatum and hippocampus of the lesion side. Decrease of [¹¹C]CFT binding in the striatum indicates 6-OHDA-induced degeneration of presynaptic DA terminals. The enhanced binding of mGluR5 ligands in the lesion side might be an indication of the compensatory role of metabotropic glutamate receptors in dopaminergic neurodegeneration.

References

Awad

Hubert

Smith

Levey

Conn

(2000) Activation of metabotropic glutamate receptor 5 has direct excitatory effects and potentiates NMDA receptor currents in neurons of the subthalamic nucleus. J Neurosci 20:7871–9

Bell

Richardson

Lee

(2002) Functional and molecular characterization of metabotropic glutamate receptors expressed in rat striatal cholinergic inter-neurons. J Neurochem 81:142–9

Betarbet

Sherer

Greenamyre

(2002) Animal models of Parkinson's disease. Bioessays 24:308–18

Bhave

Nadin

Brasier

Glauner

Shah

Heinemann

Karim

Gereau

IV (2003) Membrane topology of a metabotropic glutamate receptor. J Biol Chem 278:30294–301

Breese

(1998) The use of neurotoxin to lesion catecholamine neurons to model clinical disorder. In: Highly selective neurotoxin: basic and clinical applications ( Kostrzewa

, ed) Totowa, NJ: Humana Press, 19–73

Breysse

Baunez

Spooren

Gasparini

Amalric

(2002) Chronic but not acute treatment with a metabotropic glutamate 5 receptor antagonist reverses the akinetic deficits in a rat model of parkinsonism. J Neurosci 22:5669–78

Brody

Geyer

(2004) Interactions of the mGluR5 gene with breeding and maternal factors on startle and prepulse inhibition in mice. Neurotox Res 6:79–90

Brownell

Elmaleh

Meltzer

Shoup

Brownell

Fischman

Madras

(1996) Cocaine congeners as PET imaging probes for dopamine terminals. J Nucl Med 37:1186–92

Bruno

Battaglia

Copani

D'Onofrio

Di Iorio

De Blasi

Melchiorri

Flor

Nicoletti

(2001) Metabotropic glutamate receptor subtypes as targets for neuroprotective drugs. J Cereb Blood Flow Metab 21:1013–33

10.

Catania

Landwehrmeyer

Testa

Standaert

Penney

Jr Young

(1994) Metabotropic glutamate receptors are differentially regulated during development. Neuroscience 61:481–95

11.

Chiamulera

Epping-Jordan

Zocchi

Marcon

Cottiny

Tacconi

Corsi

Orzi

Conquet

(2001) Reinforcing and locomotor stimulant effects of cocaine are absent in mGluR5 null mutant mice. Nat Neurosci 4:873–4

12.

Cicchetti

Brownell

Williams

Chen

Livni

Isacson

(2002) Neuroinflammation of the nigrostriatal pathway during progressive 6-OHDA dopamine degeneration in rats monitored by immunohistochemistry and PET imaging. Eur J Neurosci 15:991–8

13.

Coccurello

Breysse

Amalric

(2004) Simultaneous blockade of adenosine A2A and metabotropic glutamate mGlu5 receptors increase their efficacy in reversing Parkinsonian deficits in rats. Neuropsychopharmacology 29:1451–61

14.

Conn

(1997) Pharmacology and functions of metabotropic glutamate receptor. Annu Rev Pharmacol Toxicol 37:205–37

15.

Cosford

Roppe

Tehrani

Schweiger

Seiders

Chaudary

Rao

Varney

(2003) [3H]-methoxymethyl-MTEP and [3H]-methoxy-PEPy: potent and selective radioligands for the metabotropic glutamate subtype 5 (mGlu5) receptor. Bioorg Med Chem Lett 13:351–4

16.

Hoffman

Huang

Phelps

(1979) Quantitation in positron emission computed tomography: 1. Effect of object size. J Comput Assist Tomogr 3:299–308

17.

Kirik

Georgievska

Rosenblad

Bjorklund

(2001) Delayed infusion of GDNF promotes recovery of motor function in the partial lesion model of Parkinson's disease. Eur J Neurosci 13:1589–99

18.

Krieger

Hellgren-Kotaleski

Kettunen

El Manira

(2000) Interaction between metabotropic and ionotropic glutamate receptors regulates neuronal network activity. J Neurosci 20:5382–91

19.

Logan

Fowler

Volkow

Wang

Ding

Alexoff

(1996) Distribution volume ratios without blood sampling from graphical analysis of PET data. J Cereb Blood Flow Metab 16:834–40

20.

Logan

Fowler

Volkow

Wolf

Dewey

Schlyer

MacGregor

Hitzemann

Bendriem

Gatley

Christman

(1990) Graphical analysis of reversible radioligand binding from time-activity measurements applied to [N-11C-methyl]-(−)-cocaine PET studies in human subjects. J Cereb Blood Flow Metab 10:740–7

21.

Logan

(2000) Graphical analysis of PET data applied to reversible and irreversible tracers. Nucl Med Biol 27:661–70

22.

Marino

Valenti

Conn

(2003) Glutamate receptors and Parkinson's disease: opportunities for intervention. Drugs Aging 20:377–97

23.

Marti

Paganini

Stocchi

Bianchi

Beani

Morari

(2001) Presynaptic group I and II metabotropic glutamate receptors oppositely modulate striatal acetylcholine release. Eur J Neurosci 14:1181–4

24.

Meldrum

(2000) Glutamate as a neurotransmitter in the brain: review of physiology and pathology. J Nutr 130:1007s–15s

25.

Page

Peeters

Najimi

Maloteaux

J-M

Hermans

(2001) Modulation of the neuronal dopamine transporter activity by the metabotropic glutamate receptor mGluR5 in rat striatal synaptosomes through phosphorylation mediated process. J Neurochem 76:1282–90

26.

Sachs

Jonsson

(1975) Mechanisms of action of 6-hydroxydopamine. Biochem Pharmacol 24:1–8

27.

Schaffhauser

Rowe

Morales

Chavez-Noriega

Yin

Jachec

Rao

Bain

Pinkerton

Vernier

Bristow

Varney

Daggett

(2003) Pharmacological characterization and identification of amino acids involved in the positive modulation of metabotropic glutamate receptor subtype 2. Mol Pharmacol 64:798–810

28.

Shigemoto

Mizuno

(2000) Metabotropic glutamate receptors — immunocytochemical and in situ hybridization analyses. In: Handbook of chemical neuroanatomy: glutamate ( Ottersen

Storm-Mathisen

, eds) vol. 18. Amsterdam: Elsevier Science BV, pp 63–98

29.

Shults

Ray

Tsuboi

Gage

(2000) Fibroblast growth factor-2-producing fibroblasts protect the nigrostriatal dopaminergic system from 6-hydroxydopamine. Brain Res 883:192–204

30.

Sotgiu

Bellomi

Biella

GEM

(2003) The mGluR5 selective antagonist 6-methyl-2-(phenylethynyl)-pyridine reduces the spinal neuron pain-related activity in mononeuropathic rats. Neurosci Lett 342:85–8

31.

Testa

Standaert

Young

Penney

Jr (1994) Metabotropic glutamate receptor mRNA expression in the basal ganglia of the rat. J Neurosci 14:3005–18

32.

Ungerstedt

(1968) 6-Hydroxy-dopamine induced degeneration of central monoamine neurons. Eur J Pharmacol 5:107–10

33.

Whitley

Ball

(2002) Statistics review 5: comparison of means. Crit Care 6:424–8

34.

Wichmann

DeLong

(1997) Physiology of the basal ganglia and pathophysiology of movement disorders of basal ganglia origin. New York: McGraw-Hill, 87–97

35.

Nagren

Halldin

Swahn

Helfenbein

Guilloteau

(1999) Synthesis of p-methyl-[11C]RTI-32: a new tool for the in vivo evaluation of the metabolism of phenyltropane dopamine reuptake compounds. J Labelled Cpd Radiopharm 42(suppl 1):S469–71

36.

Tueckmantel

Wang

Zhu

Kozikowski

Brownell

(2005) Methoxyphenylethynyl, methoxypyridylethynyl and phenylethynyl derivatives of pyridine: synthesis, radiolabeling and evaluation of new PET ligands for metabotropic glutamate subtype 5 receptors. Nucl Med Biol 32:631–40

37.

Zafar

Siddiqui

Sayeed

Ahmad

Salim

Islam

(2003) Dose-dependent protective effect of selenium in rat model of Parkinson's disease: neurobehavioral and neurochemical evidences. J Neurochem 84:438–46