Latrepirdine Improves Cognition and Arrests Progression of Neuropathology in an Alzheimer’S Mouse Model

ORIGINAL ARTICLE Latrepirdine improves cognition and arrests progression of neuropathology in an Alzheimer’s mouse model

JW Steele1,2,3,21, ML Lachenmayer1,4,21,SJu5, A Stock1, J Liken5, SH Kim1,2, LM Delgado6, IE Alfaro7, S Bernales7,8, G Verdile9, P Bharadwaj9, V Gupta9, R Barr9, A Friss10, G Dolios10, R Wang10, D Ringe5, P Fraser11, D Westaway12, PH St George-Hyslop12, P Szabo13, NR Relkin13, JD Buxbaum2,10,14, CG Glabe15, AA Protter8, RN Martins9,16,17, ME Ehrlich1,10,18, GA Petsko5,13,ZYue1,19 and S Gandy1,2,20

Latrepirdine (Dimebon) is a pro-neurogenic, antihistaminic compound that has yielded mixed results in clinical trials of mild to moderate Alzheimer’s disease, with a dramatically positive outcome in a Russian clinical trial that was unconﬁrmed in a replication trial in the United States. We sought to determine whether latrepirdine (LAT)-stimulated amyloid precursor protein (APP) catabolism is at least partially attributable to regulation of macroautophagy, a highly conserved protein catabolism pathway that is known to be impaired in brains of patients with Alzheimer’s disease (AD). We utilized several mammalian cellular models to determine whether LAT regulates mammalian target of rapamycin (mTOR) and Atg5-dependent autophagy. Male TgCRND8 mice were chronically administered LAT prior to behavior analysis in the cued and contextual fear conditioning paradigm, as well as immunohistological and biochemical analysis of AD-related neuropathology. Treatment of cultured mammalian cells with LAT led to enhanced mTOR- and Atg5-dependent autophagy. Latrepirdine treatment of TgCRND8 transgenic mice was associated with improved learning behavior and with a reduction in accumulation of Ab42 and a-synuclein. We conclude that LAT possesses pro-autophagic properties in addition to the previously reported pro-neurogenic properties, both of which are potentially relevant to the treatment and/or prevention of neurodegenerative diseases. We suggest that elucidation of the molecular mechanism(s) underlying LAT effects on neurogenesis, autophagy and behavior might warranty the further study of LAT as a potentially viable lead compound that might yield more consistent clinical beneﬁt following the optimization of its pro-neurogenic, pro-autophagic and/or pro-cognitive activities.

Molecular Psychiatry (2013) 18, 889--897; doi:10.1038/mp.2012.106; published online 31 July 2012 Keywords: Alzheimer’s disease; amyloid; autophagy; therapeutics

INTRODUCTION function and calcium homeostasis;1 (3) modulation of Ab release from Latrepirdine (Dimebon; dimebolin) is a neuroactive compound cultured cells, isolated intact nerve terminals, and from hippocampal 10 with antagonist activity at histaminergic, a-adrenergic and neurons in living mouse brain; and (4) promotion of neurogenesis in 11 serotonergic receptors.1 Based on its effects on cognition in the murine hippocampus. We reported that latrepirdine (LAT) 10 rodents,2--6 taken in conjunction with its highly favorable safety stimulates APP catabolism and Ab secretion, an unexpected result proﬁle, the compound has formed the basis for clinical trials for for a drug that was claimed to beneﬁt AD. In pursuit of a parsi- both Alzheimer’s disease (AD)7 and Huntington’s disease (HD),8 monious subcellular mechanism underlying this unexpected result, despite a poor understanding of the molecular mechanisms we considered the possibility that LAT-stimulated Ab release might underlying its putative mnemoactive properties. occur via an unconventional secretory pathway associated with induc- Latrepirdine has been reported to possess several properties tion of macroautophagy (autophagy),12 a highly regulated process that are potentially relevant to the treatment of neurodegenera- that can be activated in response to various stressful conditions.13 tive diseases: (1) protection of cultured cells from the cytotoxicity Several laboratories have demonstrated that autophagy has a of amyloid-b (Ab) peptide;9 (2) stabilization of mitochondrial neuroprotective role in cell and animal models of neurodegene-

1Department of Neurology, Mount Sinai School of Medicine, New York, NY, USA; 2Department of Psychiatry and The Mount Sinai Alzheimer’s Disease Research Center, New York, NY, USA; 3Laboratory of Molecular and Cellular Neuroscience, Rockefeller University, New York, NY, USA; 4Department of Neurology, University of Bonn, Bonn, Germany; 5Departments of Biochemistry and Chemistry, Rosenstiel Basic Medical Sciences Research Center, Brandeis University, Waltham, MA, USA; 6Facultad de Ciencias Biolo´ gicas, Universidad Andre´s Bello, Santiago, Chile; 7Fundacioń Ciencia & Vida, Santiago, Chile; 8Medivation, Inc., San Francisco, CA, USA; 9Centre of Excellence for Alzheimer’s Disease Research & Care, School of Medical Sciences, Edith Cowan University, Perth, WA, Australia; 10Department of Genetics and Genomic Sciences, Mount Sinai School of Medicine, New York, NY, USA; 11Tanz Centre for Research in Neurodegenerative Diseases and Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada; 12Centre for Prions and Protein Folding Diseases, University of Alberta, Edmonton, AB, Canada; 13Department of Neurology and Neurosciences, Weill Medical College of Cornell University, New York, NY, USA; 14Department of Psychiatry, Seaver Autism Center for Research and Treatment, and The Friedman Brain Institute, Mount Sinai School of Medicine, New York, NY, USA; 15Department of Neurology, University of California Irvine School of Medicine, Irvine, CA, USA; 16School of Psychiatry and Clinical Neurosciences, University of Western Australia, Crawley, WA, Australia; 17Sir James McCusker Alzheimer’s Disease Research Unit, Hollywood Private Hospital, Nedlands, WA, Australia; 18Department of Pediatrics, Mount Sinai School of Medicine, New York, NY, USA; 19Department of Neuroscience, Mount Sinai School of Medicine, New York, NY, USA and 20James J Peters VA Medical Center, Bronx, NY, USA. Correspondence: Dr S Gandy, Departments of Neurology and Psychiatry, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1137, New York, NY 10029, USA. E-mail: [email protected] 21These authors contributed equally to this work. Received 24 April 2012; accepted 31 May 2012; published online 31 July 2012 Latrepirdine arrests amyloid-b neuropathology JW Steele et al 890 rative diseases, including AD, HD14 and Parkinson’s disease.15 test for homogeneity of variance were utilized for inclusion in parametric Converging data indicate that therapeutic manipulation of tests (P40.05 for Shapiro-Wilk and Levene’s tests). Independent samples autophagy with rapamycin can improve behavioral function and t-tests (parametric design) or Mann--Whitney U tests (nonparametric arrest neuropathology in at least two mouse models of AD.16--18 To design) were utilized to determine significant mean differences between this end, small molecule enhancers of rapamycin (SMERs; most two groups. One-way analysis of variance (parametric analysis) or Kruskal-- notably SMER-28) induced autophagy, improved cell viability Wallis test (nonparametric analysis) were used to compare three or more and promoted clearance of neurodegenerative disease- groups, depending on whether the data fit the assumptions of parametric related proteins including APP metabolites (among them Ab,19 analysis (see above). Two-way analyses of variance were used to analyze huntingtin20 and a-synuclein20) in cellular models. CHX time-course experiments with regard to effects of time and treatment, Herein, we report the following: (1) LAT modulates Atg5- and interactions. Bonferroni’s or Dunn’s correction for multiple compar- dependent autophagic activity in a dose-dependent manner and isons were utilized depending on whether data fit the assumptions of via the mTOR-signaling pathway; (2) LAT potentiates the parametric or nonparametric design, respectively. Significance for t-tests degradation of APP metabolites in cell culture and in mouse and analyses of variances are reported with a Pp0.05 using two-tailed brain; and (3) LAT improves the memory behavior of TgCRND8 tests with an a-level of 0.05. All statistical analyses were performed using mice while reducing the accumulation of insoluble Ab42. Given SPSS v18.0 and/or GraphPad Prism 5 (La Jolla, CA, USA). the pressing need for effective disease-modifying treatment for symptomatic AD and the current evidence that Ab-lowering 21 agents might only be effective for prophylaxis, we argue that RESULTS identification of the molecular basis of the pro-cognitive and anti- Acute latrepirdine treatment induced autophagy via the mTOR- neurodegeneration actions of LAT remains highly valuable. pathway in cultured cells We sought to determine whether LAT might regulate autophagy as one mechanism of its reported anti-neurodegeration activity. MATERIALS AND METHODS We treated HeLa cells stably expressing LC3 fused with EGFP Preparation and handling of latrepirdine 23 (eGFP-LC3) for 3 or 6 h in the absence or presence of 50 mM The synthesis and characterization of LAT was described previously.10 latrepirdine. Treatment with latrepirdine for 3 (data not shown) or Briefly, latrepirdine was purchased from SinoChemexper (Shanghai, China) 6 h (Figure 1a) markedly enhanced the number of eGFP-LC3 and purity of the compound was determined to be 499%, or provided punctae, indicating that LAT induced the formation of autophago- directly by Medivation (San Francisco, CA, USA). For use in vitro, somes. latrepirdine was dissolved directly into culture media to the desired Next, mouse N2a neuroblastoma cells were treated in the 10 concentration, as described in Steele et al. For administration in vivo, absence (vehicle) or presence of 5, 500 or 50 mM LAT for 3 or 6 h in latrepirdine was dissolved in 0.9% saline (vehicle) at a final concentration order to determine the effects of acute drug treatment on the of 3.5 mg mlÀ1 (made fresh every 2 days). regulation of autophagy. We observed a significant and dose- dependent increase in LC3-II levels in N2a cells following 3 or 6 h Cell culture experiments treatment with either 5, 500 nM or 50 mM LAT (Figure 1). Because N2a cells stably transfected with APP K670N, M671L (N2a SweAPP; the inhibition of autophagic clearance also results in accumulation generous gift of Dr Gopal Thinakaran, University of Chicago) or of LC3-II, we chose to assay the levels of p62, which is readily untransfected N2a cells, stable human cervical carcinoma (HeLa) cells degraded by autophagy, and which accumulates when auto- expressing EGFP-LC3 (kind gift from Aviva M Tolkovsky) and mouse phagic clearance is inhibited (see Supplementary Figure 1). The embryonic fibroblasts (MEFs) derived from wild-type mice or ATG5À/À p62 levels did not increase in the presence of 500 nM or 50 mM LAT; mice (kindly provided by Noboru Mizushima) were cultured and treated instead, 50 mM LAT caused a trend toward reduction of p62 levels with or without drug, and samples were prepared for analysis as in at both 3 and 6 h treatment (Figures 1a and c), suggesting that Supplementary Methods. LC3-II accumulation was associated with increased autophagosome formation and activity. We next examined the mTOR signaling pathway, which is a Experimental animals and drug treatment key regulator of autophagic activity.24 mTOR kinase activity can Animals were individually housed and maintained on a 12:12 light:dark be monitored by the levels of phosphorylated-mTOR (p-mTOR; cycle (lights on at 7am) with ad libitum access to food and water serine 2448) and the levels of the phosphorylated form of its throughout the course of the entire experiment. All experimental protocols substrate, ribosomal S6 protein kinase (p70S6K; p-S6K).25 We were conducted within NIH guidelines for animal research and were observed a significant decreaseofp-mTORandp-S6KfromN2a approved by the Institutional Animal Care and Use Committee (IACUC) at cells treated with 50 mM LAT for 3 h, whereas the total mTOR and Mount Sinai School of Medicine. See Supplementary Methods for specific p70S6K levels remained relatively constant (Figure 2). Based animal ages, group assignment and tissue preparation. on this result, we conclude that LAT likely induced autophagy via inhibition of the mTOR signaling pathway. The specific Behavioral testing molecular target of latrepiridine within this pathway remains Mice were trained and tested in operant chambers on three consecutive unknown. days in the cued and contextual fear conditioning paradigm, similar to that 22 described in Jacobsen et al. See Supplementary Methods for detailed Latrepirdine-induced degradation of p62 and APP metabolites protocol. Freezing behavior was recorded remotely and analyzed using was Atg5-dependent Stoelting ANY-MAZE Fear Conditioning Software (Stoelting, Wood Dale, We treated ATG5 þ / þ wild type (WT) or ATG5À/À MEFs for 3 h IL, USA). in the absence or presence of 50 mM LAT (Figure 2). 3 h LAT treatment of WT MEFs was associated with significant decreases in Statistical analysis p62 and p-S6K in WT MEFs (Figure 2a). No accumulation of LC3-II Integrated density of immunoreactive western blot bands or spots was was noted in this cell line following 3 h of LAT treatment, which measured using MultiGauge Software and normalized to % control (vehicle may be related to differences in cell line-specific regulation of or non-transgenic littermates (nTg), where indicated). Absolute concentra- LC3-II turnover. By comparison with untreated WT MEFs, untreated tions of monomeric Ab40 and Ab42 or oAb were normalized to initial ATG5À/À MEFs displayed significant Btwofold accumulation of tissue weight. In all instances, Shapiro-Wilk test of normality and Levene’s both p62 and APP-CTFs, suggesting that inhibition of autophagy

Molecular Psychiatry (2013), 889 -- 897 & 2013 Macmillan Publishers Limited Latrepirdine arrests amyloid-b neuropathology JW Steele et al 891

Figure 1. Latrepirdine activates autophagy via mTOR-pathway following 3 and 6 h treatment. (a) HeLa cells stably expressing eGFP-LC3 were treated in the absence or presence of 50-mM latrepirdine for 6 h. (b--f) N2a cells were treated in the absence (Vehicle; Ctrl) or presence of 5, 500 or 50 mM latrepirdine for 3 or 6 h. (b) Western blot analysis of lysates for levels of endogenous LC3-I/--II, p62, (p)mTOR, (p)S6K and total mTOR or S6K. (c--f) Quantification of western blot densitometry for experiments summarized in a and b. All figures are representative of three or more independent experiments, performed in duplicate or triplicate. Graphs are mean±s.e.m.; *Pp0.05; **Pp0.01; ***Pp0.001. by deletion of ATG5 alone is sufficient to cause accumulation result of autophagic activity that is partially relieved following of APP-CTFs (see also Supplementary Figure 1, Supplementary addition of LAT. In ATG5À/À MEFs, incubation with 50-mM LAT for Table 1). In contrast, incubation of WT MEFs with 50 mM LAT for 3 h 3 h did not result in any change in APP-CTF or p62 accumulation, resulted in a trend toward decreased APP-CTFs, suggesting that confirming that the LAT-related stimulation of clearance of APP the accumulation of APP-CTFs and a resultant pool of Ab may be a metabolites and p62 was Atg5-dependent.

& 2013 Macmillan Publishers Limited Molecular Psychiatry (2013), 889 -- 897 Latrepirdine arrests amyloid-b neuropathology JW Steele et al 892

Figure 2. Latrepirdine enhances Atg5-dependent autophagy and stimulates the degradation of intracellular APP metabolites in cultured cells. (a--e) Mouse embryonic fibroblasts (MEFs) derived from wild-type or ATG5À/À mice were treated for 3 h in the absence (VEH; Ctrl) or presence of 50-mM latrepirdine (LAT). Lysates were analyzed by western blot for levels of p-S6K, total S6K, p62 and LC3 and a/b-CTFs (using the pan- species pAb 369). (f--l) N2a cells stably overexpressing Swedish APP were treated for 3 h in the absence (VEH; Ctrl) or presence of 50-mM latrepirdine (LAT). Lysates were analyzed by western blot for levels of p-S6K, total S6K, p62 and LC3. Lysates and conditioned medium were also analyzed by western blot for levels of intracellular (IC Ab, holoAPP, C99-CTF) or secreted (sAPPa and sAb) APP metabolites, respectively, with the human APP-specific mAb 6E10. All figures are representative of three or more independent experiments, performed in duplicate or triplicate. Graphs are mean±s.e.m.; nsPp0.015; #Pp0.10; *Pp0.05; **Pp0.01; ***Pp0.001.

We tested whether LAT altered the accumulation of intracellular Twenty-four-hour LAT treatment arrested lysosomal degradation APP metabolites using stable SweAPP (APPK670N,M671L)-overexpres- of APP metabolites and p62 in cultured cells containing a sing N2a cells (the generous gift of Dr Gopal Thinakaran). N2a defective autophagy pathway SweAPP cells were treated for 3 h in the absence or presence of Unexpectedly, following a 24 h treatment of MEFs or N2a SweAPP 50-mM LAT. Cell lysates and conditioned media were separately cells with 50 mM LAT, we observed significant hyperaccumulation analyzed for APP metabolites by western blot. We observed a of LC3-II, p62 and APP metabolites, a profile consistent with an significant accumulation of LC3-II and without the accumulation of inhibition of autophagy, potentially via inhibition of end-stage p62, indicating enhanced autophagy. Moreover, we noted parallel lysosomal protease activity. This effect was observed in both the changes in levels of both intra- and extracellular APP metabolites presence (N2a SweAPP cells or WT MEFs) or absence (ATG5À/À (Figures 2f--l), in which the induction of autophagy was associated MEFs) of an intact autophagic pathway and was therefore with decreased intracellular Ab accumulation and a trend toward Atg5-independent. Although LAT inhibited mTOR signaling follow- increased secreted Ab. These changes are consistent with our ing brief drug treatments, no regulation of the mTOR pathway previous report that LAT stimulates the secretion of APP could be detected following 24 h exposure of cells to the drug metabolites,10 suggesting that LAT may reduce intracellular APP/ (Figures 3a and e). This effect appeared to be distinct from the Ab accumulation by stimulating both secretion of APP metabolites regulation of autophagic activity observed with short-term exposure and intracellular catabolism of APP/Ab via autophagy. to latepirdine.

Molecular Psychiatry (2013), 889 -- 897 & 2013 Macmillan Publishers Limited Latrepirdine arrests amyloid-b neuropathology JW Steele et al 893

Figure 3. 24 h incubation with 50-mM latrepirdine potentiates Atg5-independent accumulation of APP metabolites and p62. (a--d) Mouse embryonic ﬁbroblasts (MEFs) derived from either wild-type or ATG5À/À mice were incubated for 24 h in the absence or presence of 50-mM latrepirdine. Cell lysates were analyzed for levels of APP metabolites, as well as markers of autophagy including LC3, P62, S6K and pS6K by western blot. (e--k) N2a cells stably overexpressing Swedish APP (N2a SweAPP) were incubated for 24 h in the absence or presence of 50-mM latrepirdine. Cell lysates and conditioned media were analyzed for levels of APP metabolites, as well as markers of autophagy including LC3, P62, S6K and pS6K by Western blot. All ﬁgures are representative of three or more independent experiments, performed in duplicate or triplicate. Graphs are mean±s.e.m.; *Pp0.05; **Pp0.01; ***Pp0.001.

A related report demonstrated that some histamine-related APP metabolites, a-syn, LC3 and p62. Along this line, it is relevant compounds were associated with an off-target and receptor- that cortical accumulation of a-syn, a known autophagic substrate, independent vacuolar-ATPase-dependent inhibition of lysosomal is observed in 30--50% of AD patients.27 Here, we utilized the protonation when used at high concentrations and with long TgCRND8 (APPK670N/M671L/V717F) mouse model of Alzheimer’s incubation times (similar to the conditions used here).26 We cerebral amyloidosis.28 Male TgCRND8 mice or their nTg were propose that this effect may underlie our observation that 24 h killed at 3 or 4 months of age, and cerebral cortices were dissected incubation with 50-mM LAT potentiates the accumulation of typical for analysis of levels of the APP metabolites, a-syn, p62 and LC3 in p62 and APP-CTFs, even in the absence of an intact autophagic both soluble and insoluble fractions (for schematic, see Supple- pathway (Figures 3e--k). mentary Figure 2). By 3 months of age, TgCRND8 mice were impaired in acquisition and learning reversal in the reference memory version of the Morris water maze task.28 Impaired autophagic clearance of a-synuclein and LC3 occurred as At 3 months of age, TgCRND8 mice exhibited deposition of a delayed pathology following accumulation of insoluble Ab42 Ab42-immunoreacive cortical and hippocampal plaques and and p62 in the brains of TgCRND8 mice accumulation of insoluble p62, which was recovered along with An aging-dependent decrease in basal neuronal turnover of brain Ab42 in the 70% formic acid fraction of extracts of TgCRND8 autophagic substrates has been identiﬁed as a feature of the brain; however, neither LC3 nor a-syn were ever detectable in the molecular neuropathology associated with the progression of AD. formic acid fractions. When 3-month-old mice were studied, no Here, we hypothesized that autophagic/lysosomal failure leads to signiﬁcant difference was observed between genotypes in the the hyperaccumulation of typical autophagic substrates, including levels of soluble a-syn, LC3-I, LC3-II or p62 (Figure 4c), suggesting

& 2013 Macmillan Publishers Limited Molecular Psychiatry (2013), 889 -- 897 Latrepirdine arrests amyloid-b neuropathology JW Steele et al 894

Figure 4. Autophagic clearance of a-synuclein and LC3 occurs as a delayed pathology following accumulation of insoluble amyloid-b42 (Ab42) and p62 in the brains of TgCRND8 mice. 90-day-old and 120-day-old male TgCRND8 mice or their non-transgenic littermates (nTg) were killed and brains were analyzed for levels of soluble and insoluble APP metabolites, a-syn, LC3 and p62. (a and b) Composite micrograph of 90-day- old non-transgenic littermates (nTg) (a) or TgCRND8 (b) brain immunostaining with LC3 and Ab42 and DAPI as counterstain (as labeled). Ab42- immunoreactive amyloid plaques are visible in the cortex and hippocampus of TgCRND8 (b) mice, but not nTg (a) littermates. Scale bar is 1.0 mm. (c) Western blot analysis of soluble and insoluble (formic acid) APP metabolites, LC3 and p62 from the gross cerebral cortex of 90- or 120-day-old mice (genotype as indicated). (d--i) Analysis of western blot integrated density, following normalization to Actin of p62 (d-90-days; e-120-days), LC3 (f-90-days; g-120-days) and a-syn (h) 90-days; (i) 120-days). Only 120-day-old TgCRND8 mice accumulated soluble LC3-I, LC3-II, p62 and a-syn by comparison with nTg littermates (c, e and g). Graphs are mean % nTg expression±s.e.m.; *Pp0.05; **Pp0.01; ***Pp0.001; ****Pp0.0001.

that autophagic clearance of these substrates is functionally intact Chronic LAT administration also promoted autophagy and the at that age in TgCRND8 mice. However, by 4 months of age, we degradation of a-syn in the brains of wild-type mice.30 On the observed a significant increase in accumulation of soluble a-syn, basis of these results, we postulated that a once-daily i.p. bolus total LC3 (LC3-I and LC3-II), and p62 and as insoluble p62 of 3.5 mg kgÀ1 LAT for 31 days might enhance autophagy and (Figure 4c), suggesting the development of an aging-dependent stimulate the clearance of autophagic substrates while avoiding impairment of autophagic clearance among TgCRND8 mice. Taken the potential confounding of prolonged exposure to high together, these results indicate that accumulation of insoluble concentrations of LAT (Figure 3). Ab42 and p62 temporally precedes the impairment of autophagic Male, 90-day-old TgCRND8 mice or their wild-type littermates clearance, and may lead to autophagic/lysosomal failure. In future (nTg) received 31 consecutive once-daily i.p. injections of either studies, we plan to investigate the temporal relationship between 3.5 mg kgÀ1 LAT or 0.9% saline (vehicle). At the culmination of insoluble Ab42 and p62 accumulation and alterations in treatment, mice were tested for cued and contextual fear autophagic/lysosomal stasis. Regardless of the cause and effect, conditioning using a paradigm that has been widely accepted these data indicate that autophagic/lysosomal failure may lead to for evaluating learning and memory deficits in APP transgenic the hyperaccumulation of autophagic substrates, accelerating mice.31 No difference was noted for cued or contextual memory disease progression. between vehicle- or LAT-treated nTg mice. We observed a significant increase in cued memory only among LAT- versus vehicle-treated TgCRND8 mice (P ¼ 0.01; Figure 5). A weak, Chronic LAT therapy was associated with improved cognition nonsignificant trend toward an improvement in contextual and reduced neuropathology in TgCRND8 mice memory among LAT- versus vehicle-treated mice (P ¼ 0.099) was Based on the total daily intake of latrpeirdine described by Doody also observed. Neither performance in the cued nor contextual et al.,7 an equivalent single intraperitoneal (i.p.) bolus in memory tasks by LAT-treated TgCRND8 mice were sufficient to an adult mouse was determined to be 3.5 mg kgÀ1 per day, and correct the animals to the level of performance observed for their we previously reported that acute administration of this dose nTg littermates. However, because TgCRND8 mice are known to increased the secretion of Ab40 into the interstitial fluid of the be impaired in related tasks at 3 months of age,28 we suggest that hippocampus of living Tg2576 mice.10 The dosing regimen for the the effect of improved cued memory may be related to arrest of current study was chosen based on two primary considerations: disease progression, rather than improved cognition per se. (1) in vitro studies (herein), indicating that sustained high levels Brain extracts were analyzed for soluble and insoluble APP of LAT may inhibit autophagic clearance and (2) the observation metabolites, LC3, p62 and a-syn by western blot. No differences that 21 consecutive (once daily) injections of 3.5 mg kgÀ1 LAT were observed between treatment groups for levels of soluble were sufficient to enhance (not inhibit) autophagy in young APP metabolites (including Ab, Figure 5). We further assayed the TgCRND8 mice (Supplementary Figures 3 and 4). In related studies brain extracts for concentrations of soluble oligomeric oAb (oAb) in Saccharomyces cerevisiae, we observed that LAT enhanced subtypes that were dimeric or larger in structure. We observed autophagy and promoted the clearance of Ab4229 or a-syn.30 a trend toward decreased mean soluble oAb levels among

Molecular Psychiatry (2013), 889 -- 897 & 2013 Macmillan Publishers Limited Latrepirdine arrests amyloid-b neuropathology JW Steele et al 895

Figure 5. Chronic latrepirdine therapy enhanced autophagy and arrested the progression of behavioral impairment and neuropathology among TgCRND8 mice. 90-day-old male TgCRND8 mice and their wild-type littermates received 31 consecutive once-daily intraperitoneal (i.p.) injections of either 3.5 mg kgÀ1 latrepirdine (n ¼ 10; LAT) or 0.9% saline (n ¼ 10; VEH). Mice were tested on both contextual (a) and cued (b) fear conditioning tasks, as described in Materials and methods. (c and d) Representative coronal sections comparing the CA1 region of the hippocampus and cortex from vehicle-treated (c) or latrepirdine-treated (d) animals at low magniﬁcation. Green punctae were determined to be amyloid-b42 (Ab42)-immunoreactive amyloid plaques (Pan1G6 anti-Ab42 antibody), red staining is LC3, DAPI was used as counterstain. Scale Bars ¼ 1.0 mm. (e--h) Western blot analysis of soluble and insoluble fractions from TgCRND8 mouse brains for p62, LC3, a-syn and APP metabolites. Western blots and brain sections are representative littermates from each treatment group. The noted decrease in absolute concentration of insoluble Ab42 (i) is in agreement with our observation of decreased Ab42-immunoreactive plaques among latrepirdine- treated mice (C-Veh versus D-Lat). Graphs are mean±s.e.m.; n.s.P ¼ 0.099 (non-signiﬁcant trend); *Pp0.05; **Pp0.01; ***Pp0.001.

& 2013 Macmillan Publishers Limited Molecular Psychiatry (2013), 889 -- 897 Latrepirdine arrests amyloid-b neuropathology JW Steele et al 896 LAT- (1225 pg gÀ1) versus vehicle-treated (1318 pg gÀ1) TgCRND8 CONFLICT OF INTEREST mice that did not reach statistical significance (P ¼ 0.0851). AP is Vice President of Preclinical Development for Medivation. SG holds research We also noted a significant decrease in soluble a-syn monomer grant support from Amicus Pharmaceuticals and is a consultant to the Pfizer-Janssen (P ¼ 0.0042) in the brains of TgCRND8 mice, which was associated Alzheimer’s Immunotherapy Alliance. PS and NR hold research grant support from with significantly elevated LC3-II accumulation (P ¼ 0.0095) and Baxter Healthcare (both PS and NR) and Pfizer (NR). NR is a consultant for Bristol without accumulation of soluble p62 (Figure 5), confirming Meyer Squibb and Eisai Research. GAP is on the scientific advisory boards of Amicus our results from pre-depositing 3-month-old TgCRND8 mice Pharmaceuticals and Neurophage. (Supplementary Figures 3 and 4). Formic acid fractions (Triton X-100-‘insoluble’ material) were analyzed by Ab40- or Ab42-specific sandwich ELISA, and a ACKNOWLEDGEMENTS significant decrease in the levels of insoluble Ab42 were observed The work in this manuscript was used in a dissertation by JWS as partial requirement when LAT-treated mice were compared with vehicle-treated for the fulfillment of the PhD degree. JWS is a trainee in the Integrated mice (P ¼ 0.046; Figure 5). Western blot analysis of insoluble p62 Pharmacological Sciences Training Program supported by grant T32GM062754 from levels revealed a modest, but statistically significant, decrease the National Institute of General Medical Sciences. MLL was supported by the among LAT- versus vehicle-treated mice (P ¼ 0.0334). These results Deutsche Forschungsgemeinschaft. SG and CGG are members of the Oligomer Research Consortium of the Cure Alzheimer’s Fund. We acknowledge the generous suggest that the cognitive benefits of LAT were associated with support of the NH&MRC (APP1009295 to RNM, GV, SG), the McCusker Alzheimer’s decreases in insoluble Ab42 and p62, and may be due to Research Foundation (RNM, GV), Conicyt (PFB-16 to SB), the Fidelity Biosciences increased autophagic substrate clearance. Research Initiative (SJ, JL, DR, GAP), Cure Alzheimer’s Fund (SG; CGG), the US Department of Veterans Affairs (SG), the NIH (P01AG10491 to SG; U01AG01483 to PS; NS045283 to CGG, R01NS060123; U54RR022220 to ZY; P30 NS061777 and S10 DISCUSSION RR022415 to RW; and P50AG05138 to Mary Sano), the Canadian Institutes of Health Protein aggregates are resistant to degradation and thus Research and Alzheimer Society of Ontario (PF), and Baxter Healthcare (PS and NRR). inherently long-lived;32 therefore, maintenance of efficient We would also like to thank Rosilyn Kazanjian for the gift in memory of Powel intracellular and extracellular proteolysis is required in order to Kazanjian that supported the purchase of the Luminex xMAP100/200 system. We would like to thank Loren E Khan and Justine Bonet for technical assistance in animal prevent accumulation of toxic aggregates. Following logically, colony management and Dr Yun Zhong for technical support. from this consideration, is the notion that maintenance or restoration of optimal levels of autophagic flux may be of therapeutic benefit to several neurodegenerative diseases.33 Indeed, recent reports indicate that activation of autophagy REFERENCES reduces intracellular accumulation of protein aggregates and 1 Wu J, Li Q, Bezprozvanny I. Evaluation of Dimebon in cellular model of improves cognition in three different mouse models of Huntington’s disease. Mol Neurodegener 2008; 3:15. 2 Lermontova NN, Lukoyanov NV, Serkova TP, Lukoyanova EA, Bachurin SO. AD.16--18,34,35 At the cellular level, the work presented here, and 29,30 Dimebon improves learning in animals with experimental Alzheimer’s disease. through collaborative works, indicates that activation of Bull Exp Biol Med 2000; 129: 544--546. mTOR-dependent autophagy by rapamycin neutralizes the 3 Bachurin S, Bukatina E, Lermontova N, Tkachenko S, Afanasiev A, Grigoriev V et al. 30 29 proteotoxicity induced in S. cerevisiae by a-syn or Ab42. Antihistamine agent Dimebon as a novel neuroprotector and a cognition When taken together with a wealth of published evidence, the enhancer. Ann N Y Acad Sci 2001; 939: 425--435. principle emerges that pharmacological regulation of proteos- 4 Grigorev VV, Dranyi OA, Bachurin SO. Comparative study of action mechanisms of tasis in general, and of autophagy in particular, represents dimebon and memantine on AMPA- and NMDA-subtypes glutamate receptors in attractive potential interventions for treatment of some neuro- rat cerebral neurons. Bull Exp Biol Med 2003; 136: 474--477. degenerative diseases. 5 Schaffhauser H, Mathiasen JR, Dicamillo A, Huffman MJ, Lu LD, McKenna BA et al. Given the evidence that LAT has biological activity in TgCRND8 Dimebolin is a 5-HT6 antagonist with acute cognition enhancing activities. mice that includes activation of autophagy (reported herein), and Biochem Pharmacol 2009; 78: 1035--1042. 6 Giorgetti M, Gibbons JA, Bernales S, Alfaro IE, Drieu La Rochelle C, Cremers T that mTOR modulation of autophagy is much more effective as et al. Cognition-enhancing properties of Dimebon in a rat novel object prophylaxis than as therapy for established AD in mouse recognition task are unlikely to be associated with acetylcholinesterase inhibition 17 models, then (as appears to be the case with other therapies or N-methyl-D-aspartate receptor antagonism. J Pharmacol Exp Ther 2010; 333: 21 directed at Ab ) LAT might be most useful in presymptomatic 748--757. AD or in cases of mild cognitive impairment. However, if the 7 Doody RS, Gavrilova SI, Sano M, Thomas RG, Aisen PS, Bachurin SO et al. Effect of current data are relevant to the clinical benefit observed in the dimebon on cognition, activities of daily living, behaviour, and global function in Russian trial of LAT in mild-to-moderate AD, this would imply patients with mild-to-moderate Alzheimer’s disease: a randomised, double-blind, that symptomatic AD subjects might improve or have their placebo-controlled study. Lancet 2008; 372: 207--215. disease progression arrested. 8 Kieburtz K, McDermott MP, Voss TS, Corey-Bloom J, Deuel LM, Dorsey ER et al. It is worth noting that we have not investigated the recently A randomized, placebo-controlled trial of latrepirdine in Huntington disease. reported pro-neurogenic actions of LAT,11 so we cannot Arch Neurol 2010; 67: 154--160. 9 Lermontova NN, Redkozubov AE, Shevtsova EF, Serkova TP, Kireeva EG, comment on whether the autophagy modulation that we have Bachurin SO. Dimebon and tacrine inhibit neurotoxic action of beta-amyloid observed has a role in the drug’s reported pro-neurogenesis in culture and block L-type Ca(2+) channels. Bull Exp Biol Med 2001; 132: properties, although examination of this possible connection 1079--1083. would be worth undertaking. However, as we are contemplating 10 Steele JW, Kim SH, Cirrito JR, Verges DK, Restivo JL, Westaway D et al. the treatment of neurodegenerative disorders, the putative pro- Acute dosing of latrepirdine (Dimebon), a possible Alzheimer therapeutic, neurogenic action only makes the drug a more intriguing target elevates extracellular amyloid-beta levels in vitro and in vivo. Mol Neurodegener for further investigation, as does its clean safety profile and 2009; 4: 51. blood-brain-barrier penetrability. The evidence reported herein 11 Pieper AA, Xie S, Capota E, Estill SJ, Zhong J, Long JM et al. Discovery of a suggest that, despite mixed results in clinical trials, LAT possesses proneurogenic, neuroprotective chemical. Cell 2010; 142: 39--51. potentially relevant biological activities and, as such, the drug 12 Abrahamsen H, Stenmark H. Protein secretion: unconventional exit by exophagy. Curr Biol 2010; 20: R415--R418. may yet be useful as a lead scaffold for developing clinically safe, 13 Boland B, Nixon RA. Neuronal macroautophagy: from development to degenera- blood-brain-barrier-penetrating, pro-autophagic, pro-neurogenic tion. Mol Aspects Med 2006; 27: 503--519. drugs for treatment and/or prevention of cerebral proteinopa- 14 Yue Z, Friedman L, Komatsu M, Tanaka K. The cellular pathways of neuronal thies, including a-synucleinopathies and Alzheimer’s cerebral autophagy and their implication in neurodegenerative diseases. Biochim Biophys amyloidosis. Acta 2009; 1793: 1496--1507.

Molecular Psychiatry (2013), 889 -- 897 & 2013 Macmillan Publishers Limited Latrepirdine arrests amyloid-b neuropathology JW Steele et al 897 15 Butler D, Nixon RA, Bahr BA. Potential compensatory responses through 26 Morissette G, Lodge R, Bouthillier J, Marceau F. Receptor-independent, vacuolar autophagic/lysosomal pathways in neurodegenerative diseases. Autophagy ATPase-mediated cellular uptake of histamine receptor-1 ligands: possible origin 2006; 2: 234--237. of pharmacological distortions and side effects. Toxicol Appl Pharmacol 2008; 229: 16 Caccamo A, Majumder S, Richardson A, Strong R, Oddo S. Molecular interplay 320--331. between mammalian target of rapamycin (mTOR), amyloid-beta, and Tau: effects 27 Broe M, Shepherd CE, Mann DM, Milward EA, Gai WP, Thiel E et al. Insoluble alpha- on cognitive impairments. J Biol Chem 2010; 285: 13107--13120. synuclein in Alzheimer’s disease without Lewy body formation. Neurotox Res 17 Majumder S, Richardson A, Strong R, Oddo S. Inducing autophagy by rapamycin 2005; 7: 69--76. before, but not after, the formation of plaques and tangles ameliorates cognitive 28 Chishti MA, Yang DS, Janus C, Phinney AL, Horne P, Pearson J et al. Early-onset deficits. PLoS One 2011; 6: e25416. amyloid deposition and cognitive deficits in transgenic mice expressing a double 18 Spilman P, Podlutskaya N, Hart MJ, Debnath J, Gorostiza O, Bredesen D et al. mutant form of amyloid precursor protein 695. J Biol Chem 2001; 276: Inhibition of mTOR by rapamycin abolishes cognitive deficits and reduces 21562--21570. amyloid-b. PLoS One 2010; 5: e9979. 29 Bharadwaj P, Verdile G, Barr RK, Gupta V et al. Latrepirdine (Dimebont) enhances 19 Tian Y, Bustos V, Flajolet M, Greengard P. A small-molecule enhancer of auto- autophagy and reduces intracellular GFP-Ab42 levels in yeast. J Alzheimer’s phagy decreases levels of Abeta and APP-CTF via Atg5-dependent autophagy Disease (in press). pathway. FASEB J 2011; 25: 1934--1942. 30 Steele JW, Ju S, Lachenmayer L, Liken J, Stock A, Kim SH et al. Latrepirdine 20 Sarkar S, Perlstein EO, Imarisio S, Pineau S, Cordenier A, Maglathlin RL et al. Small stimulates autophagy and reduces accumulation of a-synuclein in cells and in molecules enhance autophagy and reduce toxicity in Huntington’s disease mouse brain. Mol Psychiatry (in press). models. Nat Chem Biol 2007; 3: 331--338. 31 Wang J, Ferruzzi MG, Varghese M, Qian X, Cheng A, Xie M et al. Preclinical study of 21 Lane RF, Shineman DW, Fillit HM. Beyond amyloid: a diverse portfolio of novel dimebon on beta-amyloid-mediated neuropathology in Alzheimer’s disease. Mol drug discovery programs for Alzheimer’s disease and related dementias. Neurodegener 2011; 6:7. Alzheimer’s Res Ther 2011; 3: 36. 32 Knauer MF, Soreghan B, Burdick D, Kosmoski J, Glabe CG. Intracellular 22 Jacobsen JS, Wu CC, Redwine JM, Comery TA, Arias R, Bowlby M et al. Early-onset accumulation and resistance to degradation of the Alzheimer amyloid A4/beta behavioral and synaptic deficits in a mouse model of Alzheimer’s disease. protein. Proc Natl Acad Sci USA 1992; 89: 7437--7441. Proc Natl Acad Sci USA 2006; 103: 5161--5166. 33 Nixon RA, Yang DS. Autophagy failure in Alzheimer’s disease-locating the primary 23 Bampton ET, Goemans CG, Niranjan D, Mizushima N, Tolkovsky AM. The dynamics defect. Neurobiol Dis 2011; 43: 38--45. of autophagy visualized in live cells: from autophagosome formation to fusion 34 Yang DS, Stavrides P, Mohan PS, Kaushik S, Kumar A, Ohno M et al. Reversal of with endo/lysosomes. Autophagy 2005; 1: 23--36. autophagy dysfunction in the TgCRND8 mouse model of Alzheimer’s disease 24 Yang Z, Klionsky DJ. Mammalian autophagy: core molecular machinery and ameliorates amyloid pathologies and memory deficits. Brain 2011; 134: 258--277. signaling regulation. Curr Opin Cell Biol 2010; 22: 124--131. 35 Caccamo A, Maldonado MA, Majumder S, Medina DX, Holbein W, Magr´ı A et al. 25 Hara K, Yonezawa K, Weng QP, Kozlowski MT, Belham C, Avruch J. Amino acid Naturally secreted amyloid-beta increases mammalian target of rapamycin sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common (mTOR) activity via a PRAS40-mediated mechanism. J Biol Chem 2011; 286: effector mechanism. J Biol Chem 1998; 273: 14484--14494. 8924--8932.

Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)

& 2013 Macmillan Publishers Limited Molecular Psychiatry (2013), 889 -- 897