Wx Reduced Binding of F Altanserin to Serotonin Type 2A Receptors in Aging

Brain Research 813Ž. 1998 167±171

Short communication Reduced binding ofw18 Fx altanserin to serotonin type 2A receptors in aging: persistence of effect after partial volume correction

Carolyn Cidis Meltzer a,b,), Gwenn Smith a,b, Julie C. Price a, Charles F. Reynolds III b,c, Chester A. Mathis a,d, Phil Greer a, Brian Lopresti a, Mark A. Mintun a,b, Bruce G. Pollock b, Doron Ben-Eliezer a, Michael N. Cantwell a, Walter Kaye b, Steven T. DeKosky b,c a Department of Radiology, UniÕersity of Pittsburgh, Pittsburgh, PA 15213, USA b Department of Psychiatry, UniÕersity of Pittsburgh, Pittsburgh, PA 15213, USA c Department of Neurology, UniÕersity of Pittsburgh, Pittsburgh, PA 15213, USA d Department of Pharmaceutical Sciences, UniÕersity of Pittsburgh, Pittsburgh, PA 15213, USA Accepted 25 August 1998

Abstract

The serotoninŽ. 5-HT neurotransmitter system, which has a widespread distribution in the central nervous system, has been implicated in regulating mood and many human behaviors. There is evidence from postmortem human studies and limited information from prior in vivo studies to support a decline in 5-HT2A receptor density with aging. We examined nine elderlyŽ. ages 61±76 and nine young Ž ages 18±29.Ž. healthy individuals with positron emission tomography PET andw18 Fx altanserin, a ligand with high affinity for the 5-HT2A binding site. The PET data were corrected for differences in brain tissue volume between the young and elderly subjects using a magnetic resonanceŽ. MR imaging-based partial volume correction method. Highly significant and widespread cortical reductions in 5-HT2A specific binding were demonstrated in the elderly group relative to young controls. Regional losses averaged 61% before and 57% following correction for effects of cerebral atrophy. This finding, which is consistent with prior postmortem and in vivo studies, has both etiological and potential therapeutic implications for behavioral changes commonly observed in the elderly, including geriatric depression. q 1998 Published by Elsevier Science B.V. All rights reserved.

Keywords: Emission tomography; Serotonin receptor; Aging; Partial volume correction

The serotoninŽ. 5-HT system is involved in the modula- for both 5-HT2A and dopamine receptor subtypes, in tion of mood and behaviors, such as sleep, that undergo tissue specimens of human frontal cortex. Autopsy studies characteristic alterations in agingwx 5,12 . Although there of the effect of aging on the human 5-HT2A receptor have are in excess of fourteen 5-HT receptor subtypes, the been limited by relative lack of tracer specificity, as well 5-HT2A is among the best characterized and the effect of as potential confounding factors intrinsic to postmortem aging on this receptor has been addressed in animal mod- studies, including issues of tissue handling, retrospective els and human postmortem studies. Robson et al.wx 18 subject characterization and the influence of concurrent found no age effect inw3 Hx ketanserin binding to 5-HT2A medical illness. receptors in rat cortex; however, others have demonstrated Initial positron emission tomographyŽ. PET studies reductions in receptor density and ligand affinity in post- demonstrating a decline in 5-HT2A binding with aging mortem human frontal cortexwx 1,4,7 . Similarly, Marcusson used tracers with affinities for both 5-HT2A and dopamine et al.wx 13 and Sparks wx 22 reported reduced density of D2 receptorswx 8,25 . PET radioligands that are more selec- binding sites forw3 Hx spiperone, a tracer with high affinity tive for 5-HT receptor subtypes were later developed and applied to investigate the influence of aging in vivowx 3 . In addition to the limited selectivity of ligands for the 5-HT2A receptor, no prior PET studies investigating aging changes ) Corresponding author. University of Pittsburgh Medical Center, PET in 5-HT receptor binding have attempted to correct for the Facility, B-938, 200 Lothrop Street, Pittsburgh, PA 15213-2582. Fax: potential confounding effect that age-related cerebral vol- q1-412-647-0700; E-mail: [email protected] ume loss may have on PET measurementswx 23 .

Previous PET studies demonstrated thatw18 Fx altanserin confounding influence of gender on differences in 5-HT2A binds with high specificity and selectivity to 5-HT2A binding between the young and elderly subject groups was receptors in animals and humanswx 20 . Although radiola- minimized by excluding post-menopausal women on estro- beled metabolites ofw18 Fx altanserin appear to cross the gen replacement therapy. Although two of the young blood±brain barrier, these metabolites have been demon- women were taking oral contraceptive medication, no gen- strated to have no significant specific binding to the 5- der effects in receptor binding were detected in either HT2A receptor in vivo and, therefore, contribute to free subject group. radioligand concentration and non-specific binding only Radiosynthesis ofw18 Fx altanserin was performed accord- wx14 . Rosier et al. wx 19 recently reported an inverse correla- ing to the method of Lemaire et al.wx 10 . Dynamic PET tion between age and 5-HT2A binding in human brain scanning was performed over 90 min following bolus usingw18 Fx altanserin and PET. However, this study was intravenous injection of 10 mCi high specific activity restricted to mid-life individualsŽ. 24±48 years . In the Ž.G 1.04 Cirmmolw18 Fx altanserin using a Siemens current study, we usedw18 Fx altanserin and PET to deter- 951Rr31 tomographŽ. CTI PET Systems, Knoxville, TN mine the magnitude and regional distribution of age-re- in two-dimensional imaging modeŽ. septa extended . Head lated changes in 5-HT2A binding in a subject population movement was minimized through the use of a thermoplas- ranging in age from 18 to 76 years. These data are tic mask and headholder system. Arterial blood sampling presented both with and without an magnetic resonance was performed throughout the scanning interval for deter- Ž.MR -based correction for partial volume effects due to mination ofw18 Fx altanserin kinetics in plasma used in the differences in brain tissue volume between the young and calculation of radioligand distribution volumeŽ. DV from elderly subjects. regional time±activity data. A 10-min transmission scan Healthy young and elderly subjects were recruited was acquired using rotating rods of w68 Grxw68 Gax immedi- through the use of advertisements. Subjects with a history ately prior tow18 Fx altanserin injection and used for attenua- of significant medical burden andror history of neurologi- tion correction of the emission data. PET data were also cal or psychiatric disease, or first-degree relative with a corrected for radioactive decay and scatter. psychiatric or neurodegenerative disorder, were excluded MR imaging, performed on a Signa 1.5 Tesla scanner from study participation. Individuals using psychotropic or Ž.GE Medical Systems, Milwaukee, WI using a standard other medications with known central effects, including head coil, was used to guide region-of-interestŽ. ROI selec- b-blockers, were also excluded. All elderly controls under- tion and for partial volume correction of the PET data. went a Mini-Mental State Examination to insure a mini- Following a scout T1-weighted sagittal sequence, a volu- mum score of 27r30, and Hamilton Depression Rating to metric spoiled gradient recalledŽ. SPGR sequence Ž TEs5, exclude depressive symptomatology. The study groups TRs25, flip angles408, NEXs1; field of views24 consisted of nine young healthy controlsŽ six women, three cm, image matrixs256=192 pixels. was acquired in the men.Ž. with a mean age of 23 years range: 18±29 years coronal plane. MR and PET image data were co-registered and nine elderly healthy individualsŽ six women, three using Automated Image RegistrationŽ. AIRwx 26 . men.Ž. with a mean age of 69 years range: 61±76 years . ROIs were hand drawn on the registered MR images Since postmenopausal decreases in estrogen levels may and selected for sampling of high, moderate and low contribute to alterations in circulating 5-HT and central 5-HT2A receptor density areasŽ. Table 1 . Regional time± 5-HT2A receptor densities in womenwx 2,6 , the potential activity data were averaged for paired right and left ROIs.

Table 1 Regional binding potential estimates Region Uncorrected BP % Group Atrophy corrected BP % Group Young Elderlydifference Young Elderly difference Medial orbitofrontal 1.70Ž. 0.17 0.56 Ž. 0.14 ))) 67.1 1.97 Ž. 0.21 0.68 Ž. 0.18 ))) 65.5 Lateral temporal 1.65Ž. 0.16 0.59 Ž. 0.13 ))) 64.2 1.92 Ž. 0.18 0.80 Ž. 0.20 ))) 58.3 Anterior cingulate 1.70Ž. 0.19 0.53 Ž. 0.18 ))) 68.8 1.76 Ž. 0.25 0.62 Ž. 0.22 ))) 64.8 Occipital 1.69Ž. 0.46 0.81 Ž. 0.18 )) 52.1 1.84 Ž. 0.52 0.98 Ž. 0.23 ) 46.7 Parietal 1.63Ž. 0.29 0.65 Ž. 0.14 ))) 60.1 1.77 Ž. 0.36 0.82 Ž. 0.17 ))) 53.7 Lateral orbitofrontal 1.42Ž. 0.18 0.55 Ž. 0.14 ))) 61.3 1.65 Ž. 0.22 0.71 Ž. 0.19 ))) 57.0 Sensorimotor 1.29Ž. 0.21 0.52 Ž. 0.15 ))) 59.7 1.43 Ž. 0.27 0.68 Ž. 0.23 ))) 52.4 Amygalarhippocampus 0.84Ž. 0.06 0.27 Ž. 0.16 ))) 67.9 0.91 Ž. 0.09 0.31 Ž. 0.18 ))) 65.9 Striatum 0.39Ž. 0.09 0.19 Ž. 0.12 ) 51.3 0.41 Ž. 0.09 0.21 Ž. 0.13 ) 48.8

Logan-derived BP values are shown uncorrected and corrected for partial volume effects due to atrophy. Standard deviation values included in parentheses. ) pF0.005. )) pF0.0001. ))) pF0.00001. C.C. Meltzer et al.rBrain Research 813() 1998 167±171 169

Due to its relatively low 5-HT2A receptor concentrations, cation of a previously published MR-based methodwx 15 . A the cerebellum was selected as a reference region and used trend toward a higher mean cerebellar DV value, represen- to approximate the free concentration and non-specific tative of the free radioligand concentration and non-specific binding ofw18 Fxwx altanserin 17 . Plasma data were corrected binding, was observed in the elderlyŽ. 1.6"0.5 S.D. for the presence of radiolabeled metabolites ofw18 Fx al- compared to the young subjectsŽ.Ž 1.3"0.3 S.D. unpaired tanserin as determined by high-performance liquid chro- two-tailed t-test, ps0.08. . By normalizing regional DV matography analyses. estimates to the cerebellar DV, the BP parameter serves to In this work, the Logan graphical analysiswx 11 was minimize the influence of potential groups differences in applied to the regional time±activity dataŽ. 10 to 90 min to non-specific binding. Plasma protein binding ofw18 Fx al- obtain estimates of the DV and apparent binding potential tanserin, available in six of the nine elderly subjects and all s r y Ž.BP , where BP ŽDVROIDV Cerebellum .1. Correction of nine of the young subjects, was not significantly different Logan BP values for partial volume effects due to age-re- Ž.ps0.10 between the young Ž 98.9%"0.7% S.D. . and lated cerebral volume loss was performed using a modifi- elderlyŽ. 98.2%"0.9% S.D. groups. Also, no significant

Fig. 1. Representative time±activity curveŽ. A, B and Logan graphical analysis Ž. C, D results of regional PET data in a young Ž 18 years old .Ž. A, C and elderlyŽ 61 years old . subject Ž.Ž. B, D . A, B Percent of injected radiotracer dose per liter as a function of time post-injection was plotted for a cortical area Ž.medial orbitofrontal cortex, MOF with an expected high density of 5-HT2A receptors, the striatum Ž STR; a region of known lower 5-HT2A receptor density.Ž and the cerebellum CER; reference region . . A notably lower fraction of the injected dose was observed in the cortical region of the elderly relative to the young subject.Ž. C, D Linear regression Ž solid lines . was performed on the regional time±activity data over 10±90 min.w ROIŽ.T represents the concentration of radioactivity within the ROI at time T;CpŽ.t represents the plasma radioactivity concentration ofw18 Fx altanserinŽ plasma input function.x 170 C.C. Meltzer et al.rBrain Research 813() 1998 167±171 difference in the rate of metabolism ofw18 Fx altanserin in brain changes on PET measurements; however, this find- plasma between the young and elderly subjects was ob- ing cannot predict whether the relationship between recep- servedŽ. time-series ANOVA, ps0.69 . tor density and age is linear across the lifespan, or perhaps In this study, a highly significant reduction inw18 Fx al- maintained until a critical age after which there is a more tanserin binding was found in the elderly in all brain precipitous decline. Marcusson et al.wx 13 observed de- regions examinedŽ. Table 1, Figs. 1 and 2 . Overall, re- creased binding ofw3 Hx spiperone to postmortem frontal gional specific binding estimates in the young and elderly cortex predominately after age 60. In vivo studies, how- followed the known rank order of 5-HT2A receptor den- ever, have suggested either a linear decline from early sity; however, uncorrected BP values in the elderly aver- adulthood to old agewx 25 or higher rates of decline in aged 61.3%"6.6% S.D. lower than those of young con- young adultswx 3 . trols. A similar magnitude reduction of approximately The 5-HT system is a widespread neurotransmitter sys- two-thirds in cortical binding has been reported using PET tem with known modulatory effects on dopaminergic, g- andw18 Fxwx setoperone 3 in subjects 49 to 86 years of age. aminobutyric acidŽ. GABAergic and cholinergic neuronal This is also in accord with post-mortem data demonstrat- function, and cerebrovascular tonewx 16,21,24 . Specifically, ing overall losses of 50% inw3 Hx spiperone binding to the 5-HT2A receptor may play an important role in modu- frontal cortexwx 13 . Following partial volume correction, lation of cortical function, since this receptor subtype has group differences in BP values averaged 57.0%"7.2% been localized to cortical pyramidal cellswx 9 . Serotonergic S.D.Ž. Table 1 . Although the MR-derived regional tissue dysfunction with loss of 5-HT modulation has been impli- correction factors indicated a significant difference in cor- cated in conditions that frequently affect the elderly, in- tical brain volume between the young and elderly subjects cluding depression, sleep disturbance and memory loss. in all regions except striatum, correction for the effects of Using PET andw18 Fx altanserin, this study demonstrated a atrophy on the PET data accounted for less than 5% of the global, marked reduction in central 5-HT2A specific bind- large between-group difference in specific binding of ing with normal aging. This finding may have etiologic w18 Fx altanserin. Thus, although important, the effect of importance for behavioral, mood and memory changes atrophy was small relative to the large change in 5-HT2A encountered in the geriatric population. We speculate that receptor binding with age. the large aging decline in 5-HT2A binding may impart These data demonstrate a marked reduction in 5-HT2A increased susceptibility to mood and behavioral alterations receptor in the elderly relative to young adults even after in older persons. Further, since the 5-HT2A receptor is the accounting for dilutional effects of age-related structural site of action of atypical antipsychotic and several antide-

Fig. 2. Co-registered MR and parametricw18 Fx altanserin PET images at three brain levels with pixel-by-pixel representation of the Logan BP values Ž.uncorrected for atrophy acquired in an 18-year-old woman Ž.Ž. left and 61-year-old man right . The age-related global reduction in specific binding of w18 Fx altanserin is evident. The MR images in the older subject demonstrate mild cerebral volume loss typically associated with normal aging. C.C. Meltzer et al.rBrain Research 813() 1998 167±171 171 pressant medications, age-related changes in this receptor ity measurements applied to w N-C-11 methylx-Ž.y cocaine PET studies in human subjects, J. Cereb. Blood Flow Metab. 10Ž. 1990 7407. system may influence the treatment response of neuropsy- wx chiatric disorders in the elderly. 12 I. Lucki, The spectrum of behaviors influenced by serotonin, Biol. PsychiatryŽ. 1988 in press. wx13 J. Marcusson, D. Morgan, B. Winblad, C. Finch, Serotonin-2 binding sites in human frontal cortex and hippocampus. Selective loss of Acknowledgements S-2A sites with age, Brain Res. 311Ž.Ž 11 1984 . 51±56. wx14 N. Mason, Y. Huang, D. Holt, J. Perevuznik, B. Lopresti, C. Mathis, This work was supported by PHS grants AG5133 Synthesis and characterization ofwx F-18 4-Ž. 4-fluoro-benzoyl piperi- wx Ž.ŽAlzheimer's Disease Research Center , MH52247 Mental dine, an F-18 altanserin metabolite, J. Nucl. Med. 38Ž. 1997 56P. wx15 C.C. Meltzer, J.P. Leal, H.S. Mayberg, H.J. Wagner, J.J. Frost, Health Clinical Research Center for Late-Life Mood Disor- Correction of PET data for partial volume effects in human cerebral ders. , MH49936, and MH01210, and the Radiological cortex by MR imaging, J. Comput. Assist. Tomogr. 14Ž. 1990 Society of North America Research and Education Fund. 561±570. We also wish to thank N. Scott Mason, Daniel Holt and wx16 D. Morilak, S. Garlow, R. Ciaranello, Immunocytochemical localiza- Norman Simpson for their contributions to this work. tion and description of neurons expressing serotonin-2 receptors in the rat brain, Neuroscience 54Ž. 1993 701±717. wx17 A. Pazos, A. Probst, J.M. Palacios, Serotonin receptors in the human brain: IV. Autoradiographic mapping of serotonin-2 receptors, Neu- References roscience 21Ž. 1987 123±139. wx18 L. Robson, A. Gower, D. Kendall, C. Marsden, Age-related be- wx1 B. Arranz, A. Eriksson, E. Mellerup, P. Plenge, J. Marcusson, Effect havioural, neurochemical and radioligand binding changes in the of aging in human cortical pre- and postsynaptic serotonin binding central 5-HT system of Sprague±Dawley rats, Psychopharmacology sites, Brain Res. 620Ž. 1993 163±166. 113Ž. 1993 274±281. wx2 F. Biver, F. Lotstra, M. Monclus, D. Wikler, P. Damhaut, J. wx19 A. Rosier, P. Dupont, J. Peuskens, G. Bormanns, R. Vandenberghe, Mendlewicz, S. Goldman, Sex difference in 5HT2 receptors in the M. Maes, T. de Groot, C. Schiepers, A. Verbruffen, L. Mortelmans, living human brain, Neurosci. Lett. 204Ž. 1996 25±28. Visualisation of loss of 5-HT2A receptors with age in healthy wx3 J. Blin, J.C. Baron, B. Dubois, C. Crouzel, M. Fiorelli, D. Attar-Levy, volunteers usingw18 Fx and positron emission tomographic imaging, B. Pillon, D. Fournier, M. Vidailhet, Y. Agid, Loss of brain 5HT2 Psychiatr. Res. 68Ž. 1996 11±22. receptors in Alzheimer's disease, Brain 116Ž. 1993 497±510. wx20 B. Sadzot, C. Lemaire, P. Maquet, E. Salmon, A. Plenevaux, C. wx 4 S. Cheetham, M. Crompton, C. Katona, R. Horton, Brain 5-HT2 Degueldre, J. Hermanne, M. Guillaume, R. Cantineau, D. Comar, G. receptor binding sites in depressed suicide victims, Brain Res. 443 Franck, Serotonin 5HT2 receptor imaging in the human brain using Ž.1988 272±280. positron emission tomography and a new radioligand,w18 Fx al- wx5 S. DeKosky, A. Palmer, Neurochemistry of aging, in: M. Albert, J. tanserin: results in young normal controls, J. Cereb. Blood Flow KnoefelŽ. Eds. , Clinical Neurology of Aging, Oxford Press, New Metab. 15Ž. 1995 787±797. York, 1994, pp. 79±101. wx21 G. Smith, S.L. Dewey, J.D. Brodie, J. Logan, S. Vitkun, P. wx6 G. Gonzales, C. Carillo, Blood serotonin levels in postmenopausal Simkowitz, R. Schloesser, D. Alexoff, A. Hurley, T. Cooper, N. women: effects of age and serum oestradiol levels, Maturitas 17 Volkow, The serotonin±dopamine interaction measured with positron Ž.1993 23±29. emission tomographyŽ. PET andw11 Cx -raclopride in normal human wx7 R. Gross-Isseroff, D. Salama, M. Isreli, A. Biegon, Autoradio- subjects, Am. J. Psychiatry 154Ž. 1997 490±496. graphic analysis ofw3 Hx ketanserin binding in the human brain post- wx22 D. Sparks, Aging and Alzheimer's Disease. Altered cortical seroton- mortem: effect of suicide, Brain Res. 507Ž. 1990 208±215. ergic binding, Arch. Neurol. 46Ž. 1989 138±140. wx8 M. Iyo, T. Yamasaki, The detection of age-related decrease of wx23 N.K. Tanna, M.I. Kohn, D.N. Horwich, P.R. Jolles, R.A. Zimmer- dopamine D1, D2 and serotonin 5-HT2 receptors in living human man, W.M. Alves, A. Alavi, Analysis of brain and cerebrospinal brain, Prog. Neuro-Psychopharmacology Biol. Psychiatr. 17Ž. 1993 fluid volumes with MR imaging: impact on PET data correction for 415±421. atrophy: Part II. Aging and Alzheimer dementia, Radiology 178 wx9 R. Jakab, P. Goldman-Rakic, 5-Hydroxytryptamine 2A-serotonin Ž.1991 123±130. receptors in the primate cerebral cortex: possible site of action of wx24 N. Wiernsperger, Serotonin 5-HT2 receptors and brain circulation, J. hallucinogenic and antipsychotic drugs in pyramidal cell apical Cardiovasc. Pharmacol. 16Ž. 1990 S20±S24. dendrites, Proc. Natl. Acad. Sci. U.S.A. 95Ž. 1998 735±740. wx25 D. Wong, H.J. Wagner, R. Dannals, J. Links, J. Frost, H. Ravert, A. wx10 C. Lemaire, R. Cantineau, M. Guillaume, A. Plenevaux, L. Christi- Wilson, A. Rosenbaum, A. Gjedde, K. Douglass, J. Petronis, M. aens, Fluorine-18 altanserin: a radioligand for the study of serotonin Folstein, J. Toung, H. Burns, M. Kuhar, Effects of age on dopamine receptors with PET: radiolabeling and in vivo biologic behavior in and serotonin receptors measured by positron tomography in the rats, J. Nucl. Med. 32Ž.Ž 12 1991 . 2266±2272. living human brain, Science 226Ž. 1984 1393±1396. wx11 J. Logan, J. Fowler, N. Volkow, A. Wolf, S. Dewey, D. Schlyer, R. wx26 R.P. Woods, J.C. Mazziotta, S.R. Cherry, MRI-PET registration MacGregor, R. Hitzemann, B. Bendriem, S. Gatley, D. Christman, with automated algorithm, J. Comput. Assist. Tomogr. 17Ž. 1993 Graphical analysis of reversible radioligand binding from time activ- 536±546.