in vivo spect imaging of 5‐ht1a receptors with [123i] p‐mppi in nonhuman primates

SYNAPSE 24:273-281 (1996)

In Vivo SPECT Imaging of 5=HTlA Receptors With [1231]p-MPPI in

Nonhuman Primates HANK F. KUNG, DANA FWDERICK, HEE-JOUNG KIM, WILLIAM McELGIN, MEI-PING KUNG,

STEVEN A. KUSHNER, AM) ZHI-PING ZHUANG MU MU, P. DAVID MOZLEY, JANET M. VESSOTSKIE, D. ANDREW STEVENSON,

Departments of Radiology (H.RK., D.R, H.-J.K., W.M., M.-PK., M.M., PD.M., D.A.S., S.A.K., Z.-PZ.) and Pharmacology (H.l?K., J.M.V), Unversity of Pennsylvania, Philadelphia, Pennsylvania 19104

KEY WORDS Hippocampus, Serotonin, Antagonists

ABSTRACT The in vivo imaging of a novel iodinated phenylpiperazine derivative for 5-HTIA receptors, [1231Jp-MPP1 (4-(2'-methoxy-)phenyl-l-[2'-(n-2"-pyridinyl)-p-iodobenzamido-]ethyl-piperazine), using single photon emission computed tomography (SPECT), was evaluated in nonhuman primates. After an i.v. injection, [1231]p-MPPI penetrated the blood-brain barrier quickly and localized in brain regions where 5-HTIA receptor density is high (hippocampus, frontal cortex, cingulate gyrus, entorhinal cortex). Maximum ratio of hippocampus to cerebellum was 3 to 1 at 50 min postinjection. The specific binding of the radioligand in the hippocampal region, an area rich in 5-HTlA receptor density, was blocked by a chasing dose of (?) 8-OH-DPAT (2 mgkg, i.v.) or non-radioactive p-MPPI (1 mgkg, Lv.1, whereas the regional distribution of [1231]p-MPPI was unaffected by treatment with non 5-HTIA agents, such as ketanserin. Ex vivo and in vitro autoradiographic studies using monkey brain further confirmed that the specific binding of [Iz3I Jp- MPPI is associated with 5-HTIA receptor sites. However, the initial attempt a t [Iz3I]p- MPPI human imaging studies did not display specific localization of 5-HTIA receptors. This discrepancy observed for [1231]p-MPPI may be due to a dramatic difference in meta- bolic pathways between humans and monkeys. o 1996 WiIey-Liss, Inc.

INTRODUCTION Many important CNS functions, such as emotion,

feeding, sleep cycle, sexual behavior, and affective states, are regulated or mediated by 5-HTlA receptors (Hoyer et al., 1994). Multiple secondary messenger link- ages, such as adenylyl cyclase, inositol phosphate, and K' channels, have been observed in brain tissues and in cloned cells expressing this receptor. However, it is generally accepted that in the hippocampus, the 5-HTIA receptor is negatively coupled to G-protein (G,,-inhibitory effect) (De Vivo and Maayani, 1988; Gettys et al., 1994; Gozlan et al., 1995; Kung et al., 1995). The receptor most likely exists in a high or low affinity state and a dynamic regulatory mechanismb) may be responsible for controlling the activation or deactivation ofthe affn- ity state (Mahle et al., 1992). Since 5-HTIA receptors are localized in dorsal raphe and hippocampus as presynaptic (autoreceptor) and postsynaptic receptors, respectively, they may have distinctive regulators and functional roles. Several recent reports also suggest that 5-HTIA receptors mediate the effects of antidepressants (Meltzer and Maes, 1994; Newman et al., 1993; 0 1996 WILEY-LISS, INC.

Stahl, 1994). The pharmacological mechanisms and regional selectively of the regulation of serotonin release by 5-HTIA autoreceptors and antidepressants have been characterized (Kreiss and Lucki, 1994; Lucki et al., 1994). The loss of 5-HTIA receptors was observed in normal aging (Burnet e t al., 1994; Robson et al., 1993) and, more prominently, in patients with Alzheimer's disease (Cross et al., 1988).

Due to their potential as pharmacological agents, a large number of antagonists for 5 -HTI~ receptors have been reported in the literature (Cliffe et al., 1993; Fletcher et al., 1993; Glennon, 1992). Several of the

Received November 1, 1995; accepted m revised form April 11, 1996. Address reprint requests to Hank F. Kung, Ph.D., 3700 Market Street, Room

305, Philadelphia, P A 19104. E-Mail. [email protected]. Abbreviations used: 8-OH-DPAT, 8-hydroxy-2-N,N-(di-npropyl)aminotetralin;

8-0H-PIPAT, trans-8-hydroxy-2-(N-n-propyl-N-3'-iodo-2'-propenyl~aminotetralin; NAN-190, l-~2-methoxyphenyl~-4-~4-~2-phthalimido)butyllpiperazine; BMY7378, 8-l2-~4-~2-metboxypheny1)-l-piperaz~nyllethyl1-8-azaspirol-l4,51-dec~ ane-7,9-dione; SDZ-216-525, methyl 4(4-L4-(1,1,3-trioxo-2H-l,2-benzoisothiazol- 2-yl)butylI-1-piperazinyll 1H-indole-2-carboxylate; WAY100635, 4-(2'-methoxy- pheny1)- l-I2'-~N-2-pyridyl)-cyelohexanecarboxamidol-ethyl-piperazine; (SI-UH- 301, (S~-5-fluoro-8-hydroxy-2-(dipropylamino~-tetralin; p-MPPI, 4-Wmethoxy- phenyl). 1-1 2'-(N-Z-pyridyl )-p-iodobenzam~dol-ethyl-piperazine.

274 H.F. KUNG ET AL

- Partial Agonist “Pure” An tag onis ts

Kd=O. l nM K, = 0.3 nM Ki = 0.6 nM I

early contenders for pure antagonists are arylpiperazine derivatives, such as NAN-190 (1-(2-methoxyphenyl)-4- [4-(2-phthalimido)butyl]piperazine), which displayed high 5-HTlA affinity (K, = 0.6 nM), but also had high po- tency for the a1 receptor (Greuel and Glaser, 1992; Per- rone et al., 1995; Raghupathi et al., 1991). In addition, NAN-190 displayed partial agonist-like activity in a radioligand binding assay (Fornal et al., 1994; Rydelek- Fitzgeraldetal., 1990; Wozniak et al., 1991). BMY7378 (8-[2-[4-(2-methoxyphenyl)-l-piperazinyl] ethyl]-8-azaspirol-[4,5]-decane-7,9-dione) appeared to exhibit the same partial agonist property (Greuel and Glaser, 1992). In order to develop radioiodinated antagonists for the in vivo imaging of 5-HTIA receptors, we have reported a series of new arylpiperazine-benzamido derivatives, including p-MPPI (4-(2’-methoxy-phenyl)- 1 - [ 2’- (N - 2”-pyridy1)-p -iodobenzamido]-ethyl -pipera - zine) (Zhuang et al., 1994b). The iodinated derivative, [1231]p-MPPI, demonstrated a high affinity and selectivity toward 5-HTlA receptors (& =: 0.36 nM) (Kung et al., 1995; H.F. Kung et al., 1994) (\see Scheme 1). This ligand is a derivative of another arylpiperazine, WAY100635 (4-(2’-methoxy-phenyl)-l-[2’-(N-2”-pyridyl)- cyclo - hexanecar- boxamido] - ethyl -piperazine), which also displays high binding affinity (& = 0.10 nM) and selectivity (Gozlan et al., 1995; Laporte et al., 1994). Results of in vitro binding and adenylyl cyclase assays for p-MPPI demonstrated that it is an excellent “pure” antagonist for ~ - H T ~ A receptors (Kung et al., 1995; H.F. Kung et al., 1994). In vivo behavioral studies ofp-MPPI in rats clearly demonstrated that it is a pure antagonist for 5-HT1A receptors (Thielen and Frazer, 1995; Allen et al., in press). This novel radioiodinated compound, [‘2511p-MPPI, is the first selective 6-HTlA receptor antagonist labeled with 1-125 (tliz = 60 d, y energy 30-65 KeV) suitable for in vitro binding studies, and the first labeled with 1-123 (tl,z = 13 hr, y energy 159 KeV) suitable for in vivo imaging studies.

reserve; ii) targets of drug action, such as mechanism of action for antidepressants and anxiolytics; iii) rela- tionship of regional serotonin release; and iv) regulation of receptor affinity states in normal aging and in patients with various psychiatric diseases. Recent reports have indicated that [3H] or [“C]WAY100635, a close analog of p-MPPI, displayed excellent in vivo biodistribution properties suitable for imaging with positron emission tomography (PET) (Hume et al., 1994; Mathis et al., 1994). Regional distribution of ~ - H T ~ A receptors using [“C]WAY100635 in normal human brain was reported (Hume et al., 1994; Pike et al., 1995; Ritter and Jones, 1995). As expected, [“C]WAY100635 exhibited high density in the hippocampal region of human brain, where the 5-HTIA receptors are concentrated.

In order to evaluate the possible utility of the radioiodinated derivative ofp-MPPI as a potential SPECT imaging agent for 5-HTlA receptors, we prepared [1231]p- MPPI. In vivo binding characteristics of [1231]p-MPPI to 5-HT,A receptors using single photon emission computed tomography (SPECT) were evaluated in nonhuman primates and normal human subjects, and the preliminary results are reported herein.

MATERIALS AND METHODS Chemicals

No-carrier-added [12311 and [lZ5I]NaI in a 0.1N NaOH solution were purchased from Nordion (Ottawa, Can- ada) and DuPontNEN Research Products (Boston, MA), respectively. Ketanserin and (t )8-OH-DPAT were obtained from Research Biochemicals International (Na- tick, MA). Unlabeledp-MPPI was synthesized in our lab- oratory, as reported previously (Zhuang et al., 1994b).

Preparation of [‘231]p-MPPI No-carrier-added [12311p-MPPI was prepared by an io-

dodestannylation reaction similar to the procedure reported previously (Kung et al., 1990). Hydrogen perox-

Investigation of the complex function of 5-HTlA receptors requires the development of new selective radioactive ligands, such as [1z311p-MP€’I, for in vivo imaging. The iodinated agonists and antagonists will be useful to investigate the following important questions on the function of ~ - H T ~ A receptors using in vitro binding and in vivo SPECT imaging studies: i) functional roles of the receptors, such as in vivo regulation and receptor

ide (50 PI, 3% w/v) was added to a mixture containing 50 kg of the corresponding tributyltin precursor (dissolved in 50 pl ofethanol), 50 pl of IN HCl and 1-123 (10-20 mCi) in a sealed vial. The reaction was allowed to continue for 10 min at room temperature. It was then terminated by the addition of 0.1 ml ofsaturated sodium bisulfite. The reaction mixture was extracted with ethyl acetate (3 X 1 ml) after neutralization with a saturated

275 IN VIVO IMAGING OF p-MPPI

NaHC03 solution. The extracted ethyl acetate layers were evaporated to dryness, and the remaining residue was dissolved in 100 p1 of EtOH and purified by HPLC using a reverse phase column (PRP-1 column, Hamilton Co., Reno, NV) eluted with an isocratic solvent of 90% acetonitrile-10% buffer (5 mM 3,3'-dimethylglutaric acid, pH 7.0); the retention time was 9 min (1 mumin). The fractions containing the desired product were col- lected, condensed and re-extracted with ethyl acetate (3 x 1 ml). The no-carrier-added product (yield 60-70%; purity >98%), was evaporated to dryness and redis- solved in 100 ~1 of 50% EtOH with 100 pg of ascorbic acid added as an anti-oxidant. The final product, [12311p- MPPI, was diluted with saline before animal distribution studies. Radioiodination with the 1-125 isotope was carried out in a similar fashion.

SPECT imaging Three cynomolgus monkeys (-5 kg) were the subjects

of 10 SPECT imaging studies. Prior to imaging, the animals were fasted, immobilized with ketamine (10-20 mgkg, i.m.1 and xylazine (2-3 mgkg, i.m.), intubated and maintained on a 1.5-2.0% isofluoraneJ98.5% oxy- gen mixture (flow rate of 200-500 cc/min). The animals were injected with glycopyrrolate (10 pgkg, s.c.), an anticholinergic drug that does not cross the blood-brain barrier, in order to decrease digestive and respiratory secretions. Body temperature was maintained using a hot water circulating heating pad and was monitored with a rectal thermometer. The animal's head was immobilized using a vacuum-packed bean-bag device that hardens upon evacuation when molded around the head.

For control experiments, no-carrier-added [12311p- MPPI (5-10 mCi) was administered as an i.v. bolus in the saphenous vein. Immediately after injection, se- quential 5 min dynamic SPECT scans were acquired on a triple-head Picker Prism 3000 camera (FWHM: 7 mm) equipped with fan beam collimators for 2 h. The acquisition parameters were a 20% energy window at 159 KeV, 120 projection angles over 360 degrees, a 128 x 128 matrix, and a zoom factor of 1.78 in a slice thickness of 2 mm. The projection data was recon- structed with a count dependent 3-D Wiener filter. Chang's first order correction method was used to com- pensate for the 1-123 photon attenuation.

Kmetic analyses were performed by outlining the regions of interest (ROIs) which outline the right and left basal ganglia, occipital cortex, frontal cortex, cerebellum, and hippocampus. The activity was expressed as an average cpdpixel and the rates of washout of brain activity were determined for the 30-60 min period following peak hippocampal levels as a percentage of the control.

In vivo competitive binding of various compounds with the regional uptake of [1231]p-MPPI was investi- gated by injecting the animals with (?)8-OH-DPAT (1.7

mgkg, i.v.), non-radioactive p-MPPI (1.7 mgkg, i.v.) or ketanserin (0.83 mgkg, i.v.) at 50 min post-tracer dose injection.

Biodistribution A monkey was fasted, anesthetized, injected with

['2311p-MPPI and subjected to the same SPECT imaging protocol as described previously. At 60 min post i.v. injection, the animal was euthanized with pentobarbi- tal(50 mgkg, i.v.1. The major organs were removed and weighed, and samples were placed in pre-weighed test tubes. The brain was removed and bisected down the sagittal midline into two hemispheres. One half was microdissected into various regions with the tissue placed into pre-weighed test tubes, the other half was quickly frozen with powdered dry ice for ex vivo studies, and the same sections were subjected to in vitro autoradiography after the complete decay of the 1-123. The tissue radioactivity was counted using a gamma counter (Packard 5000). Percentage dose per organ was calculated by comparing the tissue counts to 100 times diluted aliquots of the injected material measured at the same time. Total activities of blood and muscle were calculated assuming they were 7 and 40%, respectively, of total body weight.

Magnetic resonance imaging and co-registration

Magnetic resonance imaging (MRI) scans of the brain were acquired on each monkey with a 1.5 Tesla machine (GE Medical Systems, Milwaukee, WI). The spoiled GRASS acquisition parameters included a repetition time (TR) of 5 msec and a flip angle of 35 degrees, which produced 1 mm thick slices. The data were reinterpo- lated in 256 x 256 matrices with cubic voxels of 2 mm per side to match the SPECT images to produce Fig- ure 2.

The MPPI-SPECT images were summed together and reformatted to match the MRI scans. Both data sets were imported into a software package developed in- house for co-registration. The program simultaneously displayed three orthogonal views of either the MRI, the SPECT scans, or the fused images of both sets. In each window, an operator could toggle through each set of images. Mouse control allowed the SPECT image sets to be dragged and dropped in a new x-y coordinate position on top of the MRI scans. An operator could also rotate the SPECT image sets within the plane. The effect of any maneuver in a given window of operation became apparent in the other two windows showing the orthogonally oriented plans. By heuristically manipu- lating the images in all three windows and then scroll- ing through the results in all three planes, visually satisfying results could be produced consistently. The parameters that were used to rotate and translate the summed image sets were then applied to each individ- ual 10-min scan automatically without further operator

276 H.F. KUNG ET AL.

interaction. Regions of interest were placed on the MRI scans and transposed on the SPECT image sets. Mean counts per pixel in each region were converted into units of concentration with experimentally measured calibration factors for the camera.

Ex vivo and in vitro autoradiography of monkey brain

After equilibrium to - 15”C, consecutive 30 pm coronal sections were cut on a cryostat microtome (Hacker Instruments, Fairfield, NJ), thaw-mounted on gelatin- coated microscope slides, and air-dried at room temperature. For in vitro autoradiographic studies, the same sections were used after complete decay (approximately 10 days) while being maintained at -70°C. Prior to the experiment, sections were thawed, dried a t room temperature, and preincubated for 30 min at room temperature in a buffer containing 50 mM Tris-HC1 (pH 7.4) and 2 mM MgClz. Following preincubation, the sections were removed from the buffer and allowed to dry a t room temperature. Labeling was initiated by covering each section with 800 pl of preincubation buffer containing [1251]p-MPPI (0.14 nM) for 90 rnin at room temperature, followed by rinsing with cold buffer for 20 min, with one change of buffer. Nonspecific binding was determined in the presence of 10 pM (+)8-OH-DPAT. After the rinsing was completed, the sections were dipped in ice-cold distilled water t,o remove buffer salts, then air-dried at room temperature. These slides, containing the brain sections, and slides containing 20 pm thick 1-125 standards (Amersham, Arlington Heights, IL), were simultaneously exposed to DuPont x-ray film in an autoradiographic cassette for 5 days for ex vivo autoradiographic experiments, and 18 h €or in vitro autoradiographic experiments. The exposed film was developed with a Kodak automatic film processor. The optical densities were determined with an image analysis system developed by NIH (Image 1.47).

Binding assays Measurement of 5-HTlA binding sites with [lz5I]p-

MPPI were carried out in rat and human hippocampal homogenates as previously described (Kung et al., 1995). Frozen hippocampal tissues from postmortem human brain were obtained from the University of Mi- ami Brain Endowment Bank. The preparation of human hippocampal homogenates was similar to the preparation used for rat hippocampal homogenates (Kung et al., 1995). Binding assays were performed in glass tubes (12 X 75 mm) in a final volume of 0.25 ml. In saturation experiments, aliquots (100 pl corresponding to 20-30 pg of protein) of membrane suspensions were mixed with 50 mM Tris-HC1, pH 7.4, and 2 mM MgCl- I) containing 0.05-2.0 nM [12511p-MPPI. Incubation was carried out for 20 min a t 37°C and then terminated by separation of bound from free radioligand by filtration through glass fiber filters (Schleicher & Schuell No. 25,

Keene, NH) presoaked with 1% polyethylenimine. The filters were then washed three times with 3 ml of ice- cold 50 mM Tris buffer and counted in a gamma counter (Packard 5000) with 70% efficiency. Protein determinations were performed using Lowry’s method (Lowry et al., 19511, with bovine serum albumin as a standard. The results of the saturation experiments were subjected to nonlinear regression analysis using EBDA (Macpherson, 1983) to obtain I(d and B,,,.

Association experiments were performed in the presence of 0.2 nM [12’I]p-MPPI a t 37°C. Dissociation was initiated by the addition of (?)8-OH-DPAT to the incubation mixture (final concentration 1 pM) after equilibrium was reached (30 rnin). The rate of association and dissociation was calculated according to the literature (Weiland and Molinoff, 1981).

RESULTS Radioiodination

Radioiodinated p-MPPI was successfully obtained with high specific activity (2.4 x lo5 and 2.2 x lo3 Ci/ mmol for 1-123 and 1-125, respectively) using a region specific iodo-tin exchange reaction. The desired product, radioiodinated p-MPPI, was easily separated by HPLC from the corresponding tributyltin material based on difference in retention times (9 min for the iodinated product vs. 30 min for the tributyltin starting material) with the same solvent system. The identity of the radio- labeled compound was verified with t,he nonradioactive authentic standard showing an identical retention time on HPLC (H.F. Kung et al., 1994).

SPECT imaging SPECT images of CNS 5-HTlA receptors in monkey

brain were evaluated. Cynomolgus monkeys were injected with a tracer dose of [1231Jp-MPPI (5-12 mCi, no- carrier-added), and a series of transaxial images was taken (Fig. 1). The MPPI-SPECT images clearly showed selective brain uptake in frontal cortex, cingulate gyrus, hippocampus, and entorhinal cortex, regions with high 5-HTlA receptor density. They co-registered well with MRI images of the same monkey (Fig. 2). Target to background ratios were about 3 to 1 at 50 min postinjection. In vivo competition experiments in monkey using SPECT demonstrated that hippocampal localization is displaceable with nonradioactive p-MPPI (Fig. 3B). Similar displacement was observed using (+)&OH- DPAT, a 5-HTlA receptor agonist (Fig. 3C). However, when ketanserin, a selective 5-HTz receptor antagonist, was used as the chasing agent, SPECT images displayed no in vivo displacement (Fig. 3D). These data suggest that the in vivo hippocampal binding of [lZ3I]p- MPPI is selective to 5-HTIA receptors.

Monkey biodistribution The biodistribution and regional brain uptake of

[12311p-MPPI in one monkey is shown in Table I. The

IN VIVO IMAGING OF p-MPPI 277

Fig. 1. Transaxial SPECT images (1.34 mm thick) of monkey brain at 51-60 min after an i.v. injection of 12 mCi of [1231]p-MPPI. SPECT images were acquired by a Picker T3000 scanner. High accumulation was observed in frontal cortex, cingulate gyrus, and hippocampus, areas of the brain where 5-HT,, receptors are concentrated.

Fig. 2. MRI images of a monkey brain were co-registered with MPPI-SPECT images of the same monkey. It is evident that the highly radioactive areas of the MPPI-SPECT images (i.e., cortex, cingulate gyrus, and hippocampus, as well as entorhinal cortex) match with the same areas on the MRI images.

kidney and liver had the highest relative concentration of radioactivity a t 60 min postinjection of [1231]p-MPPI. The total brain uptake was 0.76% dose. Radioactivity levels in the brain were highest in regions known to

have high 5-HTlA receptor density, such as hippocampus, cingulate gyrus, and cortex, and were low in regions that do not contain 5-HTIA receptors, like cerebellum and striatum. The hippocampus to cerebellum, cingu-


250 r 300 r-

200

100

n 0 20 40 60 80 100 0 20 40 60 80 100 120

fiME(rnin) TIME (min)

C .- $ g X

D C 3

3

D 100 120 R Ketanserin 80

60

40

20 0

0 20 40 60 80 100 120 0 20 40 60 80 100 120 TIME (min) TIME (MIN)

Fig. 3. Time course of regional brain uptakes as measured by SPECT imaging following i.v. injection of [12311p-MPPI in (A) control monkey, (B) a dose of nonradioactivep-MPPI (1.7 mg/Kg, i.v. injection) a t 50 min post-tracer injection, (C) a dose of 8-OH-DPAT (1.7 mg/ Kg, i.v. injection) at 50 min post-tracer injection, and (D) a dose of ketanserin (0.83 mg/Kg, i.v. injection) at 50 min post-tracer injection. The uptake in hippocampus (HP) and cerebellum (CB) is expressed as counts/pixel/min. ['2311p-MPPI displayed selective regional brain uptake with the highest concentration in the hippocampal region (HP). The selective uptake in brain was displaced by a chaser dose of nonradioactive p-MPPI and a selective 5-HTIA agonist, ( 5 )8-OH-DPAT, while the 5-HTa receptor antagonist ketanserin showed no effects on the specific uptake.

late gyrus to cerebellum, and entorhinal cortex to cerebellum ratios were 3.5,6.4, and 3.3, respectively. These ratios are higher than those observed using SPECT imaging (Figs. 1,2).

Ex vivo and in vitro autoradiography Ex vivo and in vitro autoradiography further con-

firmed the results obtained by regional brain dissection. The regional brain distribution was examined at 60 min after an [12311p-MPPI injection (Fig. 4). Intense labeling was observed in distinct regions, such as hippocampus and entorhinal cortex, areas known to have high densities of 5-HTIA receptors (Pazos et a]., 1985). These re-

sults are consistent with that reported in rat brain with [*251]p-MPPI (Kung et al., 1995). Other areas with few or no 5-HTIA receptors showed very little labeling, indi- cating that in vivo ['231]p-MPPI displayed low nonspecific binding. The same section was examined after complete decay for in vitro labeling. The regions labeled with [1251]p-MPPI were superimposed with the areas labeled with ['231]p-MPPI in vivo (Fig. 4).

The initial imaging studies attempted in normal healthy human subjects with ['231]p-MPPI were not suc- cessful (data not shown). The brain uptake was rela- tively low and there was no specific regional localization corresponding to the 5-HTlA receptor distribution over a period of 2 h. To characterize this ligand further in human brain tissue, the in vitro binding parameters, such as &, association and dissociation rates, of [12311p- MPPI for 5-HT,A receptors in hippocampus of human brain were examined.

In vitro binding The dissociation constant (&) of [1251]p-MPPI ob-

tained for human 5-HTIA receptors was similar to that of rat 5-HTlA receptors (0.30 vs. 0.32 nM; Table 11). The density found in rat hippocampal homogenates (Bma = 315 t 60 fmol/mg protein) was higher than the value observed for human hippocampal homogenates (81 ? 5 fmol/mg protein). Factors such as fresh vs. frozen tissues and the discrete vs. diffused hippocampal areas could be explanations for the density difference. Similar association and dissociation rates of [12511p- MPPI for rat 5-HTlA and human ~ - H T ~ A receptors were observed (Table 11).

DISCUSSION There are many CNS functions and disease states,

including anxiety, eating, sexual behaviors, and depres- sion, in which 5-HTlA receptors are believed to play an important role (Fletcher e t al., 1993). With the aid of suitable tracers and noninvasive imaging techniques (PET or SPECT), it is possible to investigate changes in 5-HTIA receptor density in living human brain during the course of psychiatric and neurological illness. Unfor- tunately, progress in this area has been hindered by

T M L E I. Biodistribution and regional distribution in one monkey after an L.U. cnjection ~ f [ ~ ~ ~ l l p - M P P l

Organ' 60 min Region2 60 min

Blood Heart Muscle Lung Kidney Spleen Liver Skin Thyroid Brain

4.49 -C 0.55 0.91 i 0.09 7.877 1.153 C 0.207

11.48 ? 03.115 0.170 ? 0.085

14.469 i- 9.514 5.41 0.06 i- 0.14 0.758 i 0.22

Cerebellum CinguIate gyrus Striatum Hippocampus Cortex Ent. cortedcortex

0.00896 0.0573 0.0119 0.0317 0.0298 0.0387

'Biodistribution (percentage doseiorgan, average of >3 samples i SD) 2Regional distribution (percentage dose/g)

IN VIVO IMAGING OF p-MPPI 279

In Vitro Ex Vivo

Fig. 4. Autoradiography of MPPI in monkey brain sections: ex vivo autoradiography of coronal brain section 1 h after uptake o f Pz3I1p- MPPI, and the corresponding section after complete decay, labeled in vitro with [12511p-MPPI. Both images display essentially the same regional labeling, suggesting that the SPECT in vivo image is consistent with the 5-HT,, receptor distribution.

the lack of suitable tracers for in vivo imaging of 5-HTIA receptors. Ideal tracers should have several desirable properties, including high binding affinity (typically with I(d or K, value in nM range), high receptor binding selectivity and low in vivo metabolism (Sedvall et al., 1986). In addition, they should show adequate brain uptake [i.e., blood-brain barrier (BBB) penetration] and favorable target to nontarget ratio.

Although [1251]8-OH-PIPAT is an ideal in vitro pharmacological tool for 5-HTlA receptor studies, with high and selective agonist binding properties (Zhuang et al., 1993), [lz3I]8-0H-PIPAT, the 5-HTlA receptor radioiodinated ligand developed previously, has not been success- ful as an in vivo imaging agent in nonhuman primates (Kung et al., unpublished data). The competition of the endogenous neurotransmitter (5-HT) with the agonist tracer [1231]8-OH-PIPAT for binding sites or the unfavor- able in vivo kinetic properties of this tracer may explain the nonspecific regional brain uptake ratio observed with in vivo SPECT imaging.

The radioiodinated ligandp-MPPI is a specific ~ - H T ~ A

receptor antagonist with high affinity (& = 0.35 nM in rat hippocampal homogenates) (H.F. Kung et al., 1994). Detailed in vitro and in vivo binding characteristics, functional and behavior studies in rats have been reported (Kung et al., 1995; Thielen and Frazer, 1995; Allen et al., in press). Biodistribution studies in rats indicated that the initial brain uptake of [lZ31]p-MPPI (1.22% dose/organ at 2 min post i.v. injection) is high enough to warrant further study of [1231]p-MPPI as a potential imaging agent for SPECT (M.-P. Kung et al., 1994). The SPECT imaging results of the present study using [12311p-MPPI in monkeys clearly indicated slower activity washout from hippocampus regions compared to the cerebellum (Fig. 3A). This specific localization, confirmed by autoradiography (Fig. 4), can be displaced by either nonradioactive p-MPPI or &OH-DPAT, indi- cating that the binding of [1231]p-MPPI in the hippocampus was reversible and associated with 5-HTlA receptors.

To our suprise, preliminary studies in humans showed that, a t 20 min after an i.v. injection of 5 mCi of ['Z31]p-MPPI, there was little brain uptake (Mozley and Kung, unpublished data). Initial SPECT images (0-20 min) were similar to those of brain perfusion images. Apparently, [1Z31]p-MPPI has a very fast kinetic washout from the target regions of brain that may sig- nificantly lower its usefulness. Several lines of evidence suggest that in vivo metabolism may be a key factor in producing the unsuccessful results. First, binding properties, such as dissociation constant (&), density (Bma) and kon/kaf rate, of [ 1231]p-MPPI to 5-HTIA receptors in hippocampal homogenates of human and rat brain are essentially the same (see Table 11). These data suggest that there is no apparent difference in the receptor binding properties between rats and humans; therefore, the lack of localization in human brain did not result from different receptor binding affinity. Apparent differ- ences in B,, between human and rat (81 vs. 315 fmol/ mg protein, respectively) may be due to the delays in obtaining postmortem samples of human brain, while the hippocampal samples of rat brain were freshly obtained. Second, [1231]p-MPPI appears to be very stable in human blood. Incubating ['231]p-MPPI in a human blood sample (in vitro) for >4 h did not show any change in radiochemical purity (Kung et al., unpublished data). Third, initial whole body monitoring of [1231]p-MPPI kinetics in humans exhibited a rapid liver uptake, high

TABLE II. Comparison of dissociation constant (KJ, density (Bm& and k , , l k , rates of [''511p-MPPI to 5-HTIA receptors in hippocampal homogenates of human and rat brain'

Onloff rate I% (nM) B,, (fmolimg protein) Kn (nM 'min ') %rt.(min '1

Human 0.30 i 0.05 81 2 5 Rat 0.32 2 0.04 315 i 60

0.384 0.080 0.264 0.097

YLZSI)p-MPPI (0.2 nM) was used for the association and dissociation experiments. Each value represents the mean 2 SD of three to five determinations.


f 1

in vivo

metabolism p-MPPI Jx-l* Major Metabolites

* mm~* + bioconjugates 0 -

OMe

Scheme 2

renal excretion rate, and no thyroid uptake, all of which suggest that the major metabolite(s) may be the p-iodo- benzoic acid and its bioconjugates (see Scheme 2 above). Metabolite analysis of monkey blood samples after an i.v. injection of [123T]p-MPPI suggests that the parent compound was quickly metabolized. At 10 min postinjection, less than 15% of the radioactivity was associated with the parent compound ((Kung et al., unpublished data).

Similar in vivo metabolism, for example, cleavage of amide bond, has been reported with the compound WAY100635, which shares a common moiety with p-MPPI that could be susceptible to in vivo cleavage (Pike et al., 1995). This in vivo metabolism appears to complicate the future utility of ["C]WAY100635 for imaging 5-HT1* receptors using positron emission tomography (PET), as well as the potential for using [1231]p-MPPI to image human 5-HT1* receptors using SPECT.

In conclusion, [12311p-MPPI, a radioiodinated ligand with desirable in vivo characteristics for imaging 5-HTIA receptors, was developed. [1231]p-MPPI, with its high affinity, selectivity, good brain penetration and in vivo kinetics, may be ideal for SPECT imaging of 5-HTIA receptors. Despite the initial failure of [lZ3I]p- MPPI in humans, it is likely that alternative analogs with higher retention times (less or slower in vivo metabolism) can be developed in the future for human use.

ACKNOWLEDGMENTS The authors thank Ms. Susan West for her assistance

in preparing this manuscript. This work was supported by NIH (MH48125).

REFERENCES Allen, A.R., Singh, A., Zhuang, Z.P., Kimg, M.2 , Kung, H.F., and

Lucki, I. (in press) The 5-HTIA receptor antagonist p-MPPI blocks responses mediated by presynaptic and postsynaptic 5-HTlA receptors. Pharmacol. Biochem. Behav.

Burnet, P.W.J., Eastwood, S.L., and Harrison, P.J. (1994) Detection and quantitation of 5-HTIA and 5-HTa receptor messenger-RNAs in human hippocampus using a reverse-transcriptase polymerase chain-reaction (RT-PCR) technique and their correlation with binding-site densities and age. Neurosci. Lett., 178%-89.

Cliffe, LA., Brightwell, C.I., Fletcher, A., Forster, E.A., Mansell, H.L., Reilly, Y., Routledge, C., and White, A..C. (1993) (S)-N-tert-butyl-3- (4-~2-methoxyphenyl)-piperazin-l-yl)-~!-phenylpropanamide [(S)- WAY1001351: a selective antagonist at presynaptic and postsynaptic 5-HTlA receptors. J. Med. Chem., 36:1509-1510.

Cross, A.J., Slater, P., Perry, E.K., and Perry, R.H. (1988) An autoradiographic analysis of serotonin receptors in human temporal cortex: changes in Alzheimer's-type dementia. Neurochem. Int., 13:89-96.

De Vivo. M.. and Maavani. S. (1988) 5-HT receptors coupled to adenvl- ate cyclase. In: The Serotonin Receptors. E:Sanders:Bush, ed. The Humana Press, Clifton, NJ, pp. 141-179.

Fletcher, A., Cliffe, LA., and Dourish, C.T. (1993) Silent 5-HTIA receptor antagonists: utility as research tools and therapeutic agents. Trends Pharmacol. Sci., 14:441448.

Fornal, C.A., Marrosu, F., Metzler, C.W., Tada, K., and Jacobs, B.L. (1994) Effects of the putative 5-hydroxytryptaminelA antagonist BMY-7378, NAN-190 and (- )-propranolol on serotonergic dorsal raphe unit-activity in behaving cats. J. Pharmacol. Exp. Ther.,

Gettys, T.W., Fields, T.A., and Raymond, J.R. (1994) Selective activation of inhibitory g-protein alpha-subunits by partial agonists of the human 5-HTIA receptor. Biochemistry, 33:42834290.

Glennon, R.A. (1992) Concepts for the design of 5-HTlA serotonin agonists and antagonists. Drug Dev. Res., 26:251-274.

Gozlan, H., Thibault, S., Laporte, A.M., Lima, L., and Hamon, M. (1995) The selective 5-HTIA antagonist radioligand 13HJWAY100635 labels both g-protein-coupled and free 5-HTlA receptors in rat brain membranes. Eur. J. Pharmacol., 288:173-186.

Greuel, J.M., and Glaser, T. (1992) The putative 5-HTlA receptor antagonists NAN-190 and BMY 7378 are partial agonists in the rat dorsal raphe nucleus in vitro. Eur. J. Pharmacol., 211:211-219.

Hoyer, D., Clarke, D.E., Fozard, J.R., Hartig, P.R., Martin, G.R., My- lecharane, E.J., Saxena, PR., and Humphrey, P.P.A. (1994) Interna- tional Union of Pharmacology classification of receptors for 5-hy- droxytryptamine (serotonin). Pharmacological Rev., 46:157- 203.

Hume, S.P., Ashworth, S., Opacka-Juffry, J., Ahier, R.G., Lammert- sma, A.A., Pike,VW., Cliffe, I.A., Fletcher, A,, and White, A.C. (1994) Evaluation of [o-methyl-H-3lWAY100635 as an in vivo radioligand for 5-HTlA receptors in rat brain. Eur. J. Pharmacol., 271515-523,

Kreiss, D.S., and Lucki, I. (1994) Differential regulation of serotonin (5-HT) release in the striatum and hippocampus by 5-HTIA autoreceptors of the dorsal and median raphe nuclei. J . Pharmacol. Exp. Ther., 269: 1268-1279.

Kung, H.F., Kung, M.-P., Clarke, W., Maayani, S., and Zhuang, Z.-P. (1994) A potential 5-HTlA receptor antagonist: p-MPPI. Life Sci., 551459-1462.

Kung, M.-P., Kung, H.F., Billings, J.J., Yang, Y.Y., Murphy, R.A., and Alavi, A. (1990) The characterization of IBF as a new selective dopamine Dz receptor imaging agent. J. Nucl. Med., 31:648-654.

Kung, M.-P., Zhuang, Z.-P., Frederick, D., and Kung, H.F. (1994) In vivo binding of ~'23Il4-~2'-methoxyphenyl~-1-[2'-~n-2-py~dinyl)-~-iodobenzamido-lethyl-piperazine, p-MPPI, to 5-HTlA receptors in rat brain. Synapse, 18:359-366.

Kung, M.-P., Frederick, D., Mu, M., Zhuang, Z.-P. and Kung, H.F. (1995) I'2511p-MPPI (4-(2'-methoxy-phenyl)-1-[2'-(n-2-pyridinyl)-p- iodobenzamidol-ethyl-piperazine) as a new selective imaging radioligand of serotonin-1A sites in rat brain. In vitro binding and autoradiographic studies. J. Pharmacol. Exp. Ther., 272:429-437.

Laporte, A.M., Lima, L., Gozlan, H., and Hamon, M. (1994) Selective in vivo labeling of brain 5-HTIA receptors by L3HIWAY 100635 in the mouse. Eur. J. Pharmacol., 271505-514.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurement with folin phenol reagent. J. Biol. Chem.,

Lucki, I., Singh, A,, and Kreiss, D.S. (1994) Antidepressant-like behavioral-effects of serotonin receptor agonists. Neurosci. Biobehav. Rev., 18:85-95.

Macpherson, G.A. (1983) A practical computer-based approach to the analysis of radioligand binding experiments. Comput. Programs Bi- omed., 17:107-114.

Mahle, C.D., Wiener, H.L.,Yocca, F.D., andMaayani, S. (1992) Alloste- ric interactions between the binding sites of receptor agonists and guanine nucleotides: a comparative study of the 5-hydroxytrypta-

270~1359-1366.

193:265-275.

IN VIVO IMAGING OF p-MPPI 28 1

minelA and adenosineA, receptor systems in rat hippocampal membranes. 3. Pharmacol. Exp. Ther., 263:1275-1284.

Mathis, C.A., Simpson, N.R., Mahmood, K., Kinahan, P.E., and Min- tun, M.A. (1994) [C-lllWAYl00635: a radioligand for imaging 5-HTIA receptors with positron emission tomography. Life Sci., 55:PL403-PL407.

Meltzer, H.Y., and Maes, M. (1994) Effect of pindolol on the 1-5-HTP- induced increase in plasma prolactin and cortisol concentrations in man. Psychopharmacology, 114:635-643.

Newman, M.E., Lerer, B., and Shapira, B. (1993) 5-HTIA receptor- mediated effects of antidepressants [review]. Prog. Neuropsycho- pharmacol. Biol. Psychiatry, 17:l-19.

Pazos, A., Cartes, R., andPalacios, J.M. (1985) Quantitative autoradiographic mapping of serotonin receptors in the rat brain. 11. Seroto- nin-2 receptors. Brain Res., 346:231-249.

Perrone, R., Berardi, F., Colabufo, N.A., Leopoldo, M., Tortorella, V., Fiorentini, F., Olgiati, V., Ghiglieri, A., and Govoni, S. (1995) High affinity and selectivity on 5-HTIa receptor of l-aryl-4-[( 1-tetralinlal- kyllpiperazines. 2. J. Med. Chem., 38:942-949.

Pike, V.W., Hume, S.P., McCarron, J.A., Ashworth, S., Opacka-Juffry, J., Lammertsma, A.A., Osman, S., Bench, C.J., Poole, K.G., Grasby, P., Cliffe, LA., Fletcher, A,, and White, A.C. 11995) i”ClWAY100635: a radioligand for the study of central 5-HTlA receptors in viva Iabstrl. J. Nucl. Med., 36:58P.

Raghupathi, R.K., Rydelek-Fitzgerald, L., Teitler, M., and Glennon, R.A. (1991) Analogues of the 5-HT,A serotonin antagonist 1-(2-methoxyphenyl)-4-[4-(2-phthalimido)butyllpiperazine with reduced alpha 1-adrenergic affinity. J. Med. Chem., 34:2633-2638.

Ritter, J.M., and Jones, T. (1995) PET a symposium highlighting its clinical and pharmacological potential. Trends Pharmacol. Sci., 16:117-119.

Robson, L., Gower, A.J., Kendall, D.A., and Marsden, C.A. (1993) Age- related behavioral, neurochemical and radioligand binding changes

in the central 5-HT system of Sprague-Dawley rats. Psychopharma- cology, 113:274-281.

Rydelek-Fitzgerald, L., Teitler, M., Fletcher, P.W., Ismaiel, A.M., and Glennon, R.A. (1990) NAN-190: agonist and antagonist interactions with brain 5-HTIA receptors. Brain Res., 532:191-196.

Sedvall, G., Farde, L., Persson, A., and Wiesel, F.-A. (1986) Imaging of neurotransmitter receptors in the living human brain. Arch. Gen. Psychiatry, 43:995-1005.

Stahl, S. (1994) Is serotonin receptor down-regulation linked to the mechanism of action of antidepressant drugs? Psychopharmacol. Bull., 30:3943.

Thielen, R.J., and Frazer, A. (1995) Effects of novel 5-HTIA receptor antagonists on measures of postsynaptic 5-HTIA receptor activation in viva. Life Sci., 56:PL163-PL168.

Weiland, G.A., and Molinoff, P.B. (1981) Quantitative analysis ofdrug- receptor interaction: I. Determination of kinetics and equilibrium properties. Life Sci., 29:313-330.

Wozniak, K.M., Durcan, M.J., and Linnoila, M. (1991) Is NAN-190 an effective antagonist of the hypothermia and hyperglycemia induced by the 5-HTlA receptor agonist, 8-OH-DPAT? Eur. J. Pharmacol., 193:253-256.

Zhuang, Z.-P., Kung, M.-P., and Kung, H.F. (1993) Synthesis of (R,S)truns-8-hydroxy-2-(N-n-propyl-N-3’-iodo-2’-propenylf aminotetralin (trans-8-OH-PIPAT): a new 5 -HTl~ receptor ligand. J. Med. Chem., 36:3161-3165.

Zhuang, Z.-P., Kung, M.-P., Chumpradit, S., Mu, M., and Kung, H.F. (1994a) Derivatives of 4-(2’-methoxy-phenyl)-l-[2’-(N-2”-pyridinylf- p-iodobenzamidolethyl-piperazine (p-MPPI) as 5-HTlA ligands. J. Med. Chem., 37:45724575.

Zhuang, Z.-P., Kung, M.-P., and Kung, H.F. (199413) Synthesis and evaluation of 4-(2’-methoxy-phenyl)-1-[2’-(N-2-pyridinyl~-p-iodobenzamidolethyl-piperazine (p-MPPI): a new iodinated 5-HTIA ligand. J. Med. Chem., 37:1406-1407.

in vivo spect imaging of 5‐ht1a receptors with [123i] p‐mppi in nonhuman primates

Documents