Imaging microglial activation in Huntington's disease

Brain Research Bulletin 72 (2007) 148–151

Imaging microglial activation in Huntington’s disease Yen F. Tai a , Nicola Pavese a , Alexander Gerhard a , Sarah J. Tabrizi b , Roger A. Barker c , David J. Brooks a , Paola Piccini a,∗ a

Division of Neuroscience and Psychological Medicine, Hammersmith Hospital, Imperial College London, United Kingdom b Department of Neurodegenerative Disease, Institute of Neurology, London, United Kingdom c Cambridge Centre for Brain Repair, Forvie Site, Cambridge, United Kingdom Available online 27 November 2006

Abstract Activated microglia have been proposed to play a major role in the pathogenesis of Huntington’s Disease (HD). PK11195 is a ligand which binds selectively to peripheral benzodiazepine binding sites, a type of receptor selectively expressed by activated microglia in the central nervous system. Using 11 C-(R)-PK11195 positron emission tomography (PET), we have recently shown in vivo evidence of increased microglial activation in both symptomatic and presymptomatic HD gene carriers and that the degree of microglial activation in the striatum correlates with the severity of striatal dopamine D2 receptor dysfunction measured with 11 C-raclopride PET. Our findings indicate that microglial activation is an early process in the HD pathology, occurring before the onset of symptoms. The close spatial and temporal relationship between microglial activation and neuronal dysfunction lends further support to the pathogenic link between the two processes in HD. Further longitudinal studies are needed to fully elucidate this link. © 2006 Elsevier Inc. All rights reserved. Keywords: Huntington’s disease; Microglial activation; Emission tomography; PK11195

1. Introduction Huntington’s disease (HD) is an autosomal dominantly inherited neurodegenerative disease. The underlying genetic mutation has been identified as a CAG-repeat expansion in the IT15 gene of chromosome 4. This leads to the formation of a mutant huntingtin protein with an elongated N-terminal polyglutamine chain [19]. The exact mechanisms linking the formation of the mutant huntingtin to neuronal cell death, particularly in the striatum, are as yet unclear but fibril formation leading to excitotoxicity, caspase activation and apoptosis, mitochondrial dysfunction, and RNA dysregulation have all been implicated [11]. Recent evidence suggests that microglial activation is also an integral part of HD pathogenesis. 2. Activated microglia Microglia are the major intrinsic immunocompetent phagocytic cells in the central nervous system. They comprise 10–20% ∗ Corresponding author at: Cyclotron Building, MRC Clinical Sciences Centre, Imperial College London, Hammersmith Hospital, Du Cane Road, London W12 0NN, United Kingdom. Tel.: +44 20 83833773; fax: +44 20 83833172. E-mail address: [email protected] (P. Piccini).

0361-9230/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2006.10.029

of the white cell population and are normally found in a quiescent state with spidery processes. Upon exposure of the brain to any form of insult, such as trauma, infection or ischaemia, the microglia rapidly become activated. They proliferate and surround the site of injury, stripping and remodelling synapses. In the case of HD, the insult could be the presence of dying neurones or of abnormal fibrillar mutant huntingtin [16]. Microglial activation often precedes reactions of other cell types in the brain, and hence is a very sensitive marker of neuronal insult [10]. 2.1. Microglial activation in Huntington’s disease Activated microglia have been proposed to play a major role in the pathogenesis of a range of neurodegenerative diseases including Alzheimer’s disease, Parkinson’s disease, and HD. While microglial activation is unlikely to be the initiating event in these neurodegenerative diseases, it may cause cell death via various pathways. When activated, microglia produce cytotoxic substances including pro-inflammatory cytokines (e.g. TNF-␣ and IL-1␤) and reactive oxygen species (e.g. hydrogen peroxide, superoxide) [9]. These cytokines in turn cause further activation of microglia, resulting in a self-propagating inflammatory cascade. In an in vitro model of dopaminergic neurodegen-

Y.F. Tai et al. / Brain Research Bulletin 72 (2007) 148–151

149

3. In vivo imaging of microglial activation using 11 C-(R)-PK11195 PET

Fig. 1. When microglia become activated, they express peripheral benzodiazepine binding sites (PBBS) on their mitochondrial membrane. PK11195, an isoquinoline, is a ligand that binds selectively to PBBS. When labelled with 11 C (arrow), it is used as a PET tracer for in vivo imaging of activated microglia.

eration, activated microglia-mediated neurotoxicity was found to be linked to caspase-3 activation and apoptosis [23]. Activated microglia also release excitotoxins such as glutamate and quinolinic acid, and contribute to NMDA-mediated excitotoxicity [21]. Minocycline is a tetracycline derivative which inhibits microglial activation and expression of caspase-3, a known effector of apoptosis. It has been shown to delay disease progression and mortality in the R6/2 transgenic mouse model of HD [5] and protect against microglia-mediated NMDA neurotoxicity [21]. However, these beneficial effects of minocycline have not been replicated by other groups [6,17]. In addition, there is also evidence to suggest that activated microglia might play a protective role in the central nervous system by secreting neurotrophic factors and promoting neurogenesis (for review, see ref. [9]). In a post-mortem study, Sapp et al. [16] found a significant accumulation of activated microglia in HD brains, particularly in regions which are known to be targeted such as the striatum, globus pallidus and frontal cortex. The density of activated microglia correlated with the severity of neuronal loss. The close temporal and spatial association between microglial activation and neuronal pathology in HD further suggests a possible pathogenic link.

When microglia become activated, they express peripheral benzodiazepine binding sites (PBBS) on their mitochondrial membrane. PBBS are functionally and structurally distinct from central benzodiazepine receptors associated with ␥aminobutyric acid (GABA)-regulated chloride channels. PBBS are found in abundance in peripheral organs (liver and adrenals) and haematogenous cells, but are present at only very low levels in the normal central nervous system [1]. PK11195, an isoquinoline, binds selectively to the PBBS (Fig. 1). Therefore, in the absence of invading blood borne cells or severe focal leakage of blood-brain barrier, the increased PK11195 binding to PBBS provides a marker of microglial activation in the central nervous system. This increase in PK11195 binding indicates the transition of microglia from a resting to an activated state, and is due to an increase in the number, rather than the affinity, of PBBS [2]. When labelled with 11 C and used as a positron emission tomography (PET) radiotracer, PK11195 can serve as an in vivo marker of microglial activation, as previously demonstrated in Alzheimer’s disease [4] and Parkinson’s disease [7,14]. 3.1.

11 C-(R)-PK11195

PET in Huntington’s disease

A 3 H-PK11195 autoradiography study of HD brains showed increased binding in the putamen and frontal cortex [12], consistent with the post-mortem findings of the distribution of activated microglia in HD brains [16]. However, post-mortem studies measure only single time-points in the pathological process, usually the end stages. PET has the advantage of allowing us to investigate the evolution and progression of HD pathology. We have recently studied 11 HD patients using 11 C-(R)PK11195 and 11 C-raclopride PET. The latter is a marker of striatal dopamine D2 receptor availability which in turn is a sensitive marker of striatal neuronal dysfunction [22]. We have found, for the first time, in vivo evidence of increased 11 C(R)-PK11195 binding in the striatum and cortical regions in symptomatic HD compared to healthy controls, and that the striatal 11 C-(R)-PK11195 binding correlates with clinical sever-

Fig. 2. Voxel-based analysis using statistical parametric mapping showing the similar topographic distribution of increases in decreases in 11 C-raclopride binding (B) in HD patients compared to healthy controls (p < 0.01, corrected for cluster size).

11 C-(R)-PK11195

binding (A) and

150

Y.F. Tai et al. / Brain Research Bulletin 72 (2007) 148–151

ity of HD rated with Unified Huntington’s Disease Rating Scale motor scores, and inversely with striatal 11 C-raclopride binding [15]. When interrogated with a voxel-based approach using statistical parametric mapping, there is a closely matched topographical distribution of regions with abnormal 11 C-(R)PK11195 and 11 C-raclopride binding in HD patients (Fig. 2). The cortical microglial activation is likely to indicate the involvement of cortical neurons in HD, a well recognised phenomenon as the disease progresses. Overall, our findings are highly consistent with post-mortem pathological and autoradiographical studies [12,16], which provide a validation of the technique we have used. We have extended the study to presymptomatic HD gene carriers, and again demonstrated increased striatal 11 C-(R)PK11195 binding which correlates with the severity of striatal neuronal dysfunction reflected by 11 C-raclopride PET [18]. However, a minority of the presymptomatic HD gene carriers studied did not exhibit abnormal striatal 11 C-(R)-PK11195 or 11 C-raclopride binding. It may be that some of these carriers were further away from neuronal dysfunction and disease onset. Further longitudinal data is needed to elucidate the evolution of, and the temporal relationship between, microglial activation and neuronal dysfunction in these subjects. Nevertheless, the results indicate that microglial activation is an early process in HD pathogenesis and is present in the HD brains before the onset of symptoms. The close association of microglial activation and neuronal dysfunction, in terms of spatial distribution and severity, seen in both presymptomatic and symptomatic HD patients further supports the theory that activated microglia are contributory to neuronal death in HD. The ongoing and progressive neurodegeneration in HD is mirrored by the correspondingly progressive and persistent activation of microglia. 4. Clinical implications Despite the inconsistent animal data on the efficacy of minocycline [5,6,17] and some evidence suggesting beneficial effects of microglial activation [9], several studies investigating the safety and efficacy of minocycline in human HD patients are currently underway [3,8,20]. Our finding of significant association between microglial activation and striatal neuronal pathology in HD provides further support for the rationale of these trials. Should the trials in symptomatic HD patients prove to be positive, it would be reasonable to extend the trials to include presymptomatic gene carriers, since we have shown that microglial activation is present at a very early stage in the disease process and intervention at this point is likely to yield greater benefits. 11 C-(R)-PK11195 PET may be employed as an in vivo marker of disease activity and response to treatment in such trials. However, before this can be undertaken, we need a greater understanding of the temporal relationship between microglial activation and underlying pathology. We also need to investigate the length of time for microglial activation to subside with resolution of pathology, as an excitotoxic model of neuronal injury has shown marked and persistent microglial activation long after the initial injury [13].

5. Conclusion 11 C-(R)-PK11195

PET provides an important in vivo marker of microglial activation in HD. Using this technique, we have shown significant microglial activation in HD patients and asymptomatic carriers, and the close association between microglial activation and striatal neuronal dysfunction. Although both detrimental and beneficial roles of activated microglia in neurodegenerative diseases have been reported, the weight of evidence in the literature supports the former. Longitudinal clinical studies using 11 C-(R)-PK11195 PET will shed more light on the relationship between microglial activation and HD pathology. Acknowledgements Y. F. Tai was funded by the Wellcome Trust. The study was funded by the Medical Research Council, UK. References [1] R.B. Banati, Visualising microglial activation in vivo, Glia 40 (2002) 206–217. [2] R.B. Banati, J. Newcombe, R.N. Gunn, A. Cagnin, F. Turkheimer, F. Heppner, G. Price, F. Wegner, G. Giovannoni, D.H. Miller, G.D. Perkin, T. Smith, A.K. Hewson, G. Bydder, G.W. Kreutzberg, T. Jones, M.L. Cuzner, R. Myers, The peripheral benzodiazepine binding site in the brain in multiple sclerosis: quantitative in vivo imaging of microglia as a measure of disease activity, Brain 123 (2000) 2321–2337. [3] R.M. Bonelli, C. Heuberger, F. Reisecker, Minocycline for Huntington’s disease: an open label study, Neurology 60 (2003) 883–884. [4] A. Cagnin, D.J. Brooks, A.M. Kennedy, R.N. Gunn, R. Myers, F.E. Turkheimer, T. Jones, R.B. Banati, In-vivo measurement of activated microglia in dementia, Lancet 358 (2001) 461–467. [5] M. Chen, V.O. Ona, M. Li, R.J. Ferrante, K.B. Fink, S. Zhu, J. Bian, L. Guo, L.A. Farrell, S.M. Hersch, W. Hobbs, J.P. Vonsattel, J.H. Cha, R.M. Friedlander, Minocycline inhibits caspase-1 and caspase-3 expression and delays mortality in a transgenic mouse model of Huntington disease, Nat. Med. 6 (2000) 797–801. [6] E. Diguet, R. Rouland, F. Tison, Minocycline is not beneficial in a phenotypic mouse model of Huntington’s disease, Ann. Neurol. 54 (2003) 841–842. [7] A. Gerhard, N. Pavese, G. Hotton, F. Turkheimer, M. Es, A. Hammers, K. Eggert, W. Oertel, R.B. Banati, D.J. Brooks, In vivo imaging of microglial activation with [(11)C](R)-PK11195 PET in idiopathic Parkinson’s disease, Neurobiol. Dis. 21 (2006) 404–412. [8] Huntington Study Group, Minocycline safety and tolerability in Huntington disease, Neurology 63 (2004) 547–549. [9] S.U. Kim, J. de Vellis, Microglia in health and disease, J. Neurosci. Res. 81 (2005) 302–313. [10] G.W. Kreutzberg, Microglia: a sensor for pathological events in the CNS, Trends Neurosci. 19 (1996) 312–318. [11] J. Leegwater-Kim, J.H. Cha, The paradigm of Huntington’s disease: therapeutic opportunities in neurodegeneration, NeuroRx. 1 (2004) 128–138. [12] K. Messmer, G.P. Reynolds, Increased peripheral benzodiazepine binding sites in the brain of patients with Huntington’s disease, Neurosci. Lett. 241 (1998) 53–56. [13] J. Mitchell, L.E. Sundstrom, H.V. Wheal, Microglial and astrocytic cell responses in the rat hippocampus after an intracerebroventricular kainic acid injection, Exp. Neurol. 121 (1993) 224–230. [14] Y. Ouchi, E. Yoshikawa, Y. Sekine, M. Futatsubashi, T. Kanno, T. Ogusu, T. Torizuka, Microglial activation and dopamine terminal loss in early Parkinson’s disease, Ann. Neurol. 57 (2005) 168–175.

Y.F. Tai et al. / Brain Research Bulletin 72 (2007) 148–151 [15] N. Pavese, A. Gerhard, Y.F. Tai, A.K. Ho, F. Turkheimer, R.A. Barker, D.J. Brooks, P. Piccini, Activated microglia accompany neuronal dysfunction in Huntington’s disease. An 11 C-(R)-PK11195 PET study, Neurology 66 (2006) 1638–1643. [16] E. Sapp, K.B. Kegel, N. Aronin, T. Hashikawa, Y. Uchiyama, K. Tohyama, P.G. Bhide, J.P. Vonsattel, M. DiFiglia, Early and progressive accumulation of reactive microglia in the Huntington disease brain, J. Neuropathol. Exp. Neurol. 60 (2001) 161–172. [17] D.L. Smith, B. Woodman, A. Mahal, K. Sathasivam, S. Ghazi-Noori, P.A. Lowden, G.P. Bates, E. Hockly, Minocycline and doxycycline are not beneficial in a model of Huntington’s disease, Ann. Neurol. 54 (2003) 186–196. [18] Y.F. Tai, N. Pavese, A. Gerhard, S.J. Tabrizi, R.A. Barker, D.J. Brooks, P. Piccini, Microglial activation in presymptomatic Huntington’s disease gene carriers: a 11 C-PK11195 PET study, J. Neurol. Neurosurg. Psychiatry 76 (2005) A18.

151

[19] The Huntington’s Disease Collaborative Research Group, A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes, Cell 72 (1993) 971–983. [20] M. Thomas, T. Ashizawa, J. Jankovic, Minocycline in Huntington’s disease: a pilot study, Mov. Disord. 19 (2004) 692–695. [21] T.M. Tikka, J.E. Koistinaho, Minocycline provides neuroprotection against N-methyl-d-aspartate neurotoxicity by inhibiting microglia, J. Immunol. 166 (2001) 7527–7533. [22] J.C. van Oostrom, R.P. Maguire, C.C. Verschuuren-Bemelmans, L. Veenma-van der Duin, J. Pruim, R.A. Roos, K.L. Leenders, Striatal dopamine D2 receptors, metabolism, and volume in preclinical Huntington disease, Neurology 65 (2005) 941–943. [23] X. Wang, S. Chen, G. Ma, M. Ye, G. Lu, Involvement of proinflammatory factors, apoptosis, caspase-3 activation and Ca(2+) disturbance in microglia activation-mediated dopaminergic cell degeneration, Mech. Ageing Dev. 126 (2005) 1241–1254.