Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

TMP21 in Alzheimer's disease : biochemical and behavioural characterization of TMP21 Bromley-Brits, Kelley 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_bromley-brits_kelley.pdf [ 32.77MB ]
Metadata
JSON: 24-1.0103326.json
JSON-LD: 24-1.0103326-ld.json
RDF/XML (Pretty): 24-1.0103326-rdf.xml
RDF/JSON: 24-1.0103326-rdf.json
Turtle: 24-1.0103326-turtle.txt
N-Triples: 24-1.0103326-rdf-ntriples.txt
Original Record: 24-1.0103326-source.json
Full Text
24-1.0103326-fulltext.txt
Citation
24-1.0103326.ris

Full Text

TMP21 in Alzheimer’s Disease Biochemical and behavioural characterization of TMP21 by Kelley Bromley-Brits B.Sc. Hon., Memorial University of Newfoundland, 2005 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Neuroscience) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) May 2011 c© Kelley Bromley-Brits 2011 Abstract Alzheimer’s disease (AD) is the most common neurodegenerative disorder leading to dementia. The two major neuropathological hallmarks of AD are the deposition of amyloid-b (Ab) protein in neuritic plaques and the formation of neurofibrillary tangles. Ab is generated from a larger Ab precursor protein (APP) following sequen- tial cleavage by b- and g-secretase. APP can also be cleaved in a non-amyloidogenic pathway following sequential cleavage by a- and g-secretase. In addition to the pathogenic processing of APP, the g-secretase complex also cleaves a protein called Notch, which is essential for embryonic development and may be involved in learning and memory. Transmembrane emp24-like trafficking protein 10 (TMP21) is a 21 kDa trans- membrane protein involved in vesicular trafficking. Ubiquitously expressed, partic- ularly in the plasma membrane, endoplasmic reticulum, and Golgi, TMP is vital to development, and homozygous knockout mice are embryonic lethal. Recently, TMP21 was found to play a second, pivotal role as a regulatory member of the g-secretase complex involved in AD pathogenesis. Knockdown of TMP21 increased Ab production without affecting Notch cleavage, making it a seductive target for AD research (Chen et al., 2006). This thesis shows that, similar to other members of the g-secretase complex, TMP21 is also degraded by the ubiquitin-proteasome pathway, as treatment with proteasomal inhibitors increased TMP21 protein levels in both a time- and dose- dependent manner. Furthermore, overexpression of TMP21 shifted APP processing from the a-secretase to b-secretase pathway in cell culture, and b-secretase and TMP21 could coimmunoprecipitate. This suggests that TMP21 may not only affect AD pathogenesis through its modulatory role on g-secretase or its trafficking of APP (Vetrivel et al., 2007), but also through its influence on b-secretase, providing a novel enzymatic target for future study. Finally, this work presents the only in vivo study of the behavioural consequences of TMP21 suppression. Motor function, anxiety, and learning and memory were examined using a comprehensive test battery. Mice heterozygous for TMP21 were found to have slightly enhanced physical abilities, increased anxiety, and potential anxiety-augmented deficits in hippocampal learning and memory. This data will prove vital when examining future work regarding TMP21 suppression in a mouse model of AD. ii Preface Unless otherwise indicated, all text, figures, and data in this thesis were created by the author, Kelley Bromley-Brits. Chapter 1 A large portion of this chapter has been submitted as a review for publication. I am first-author on the submission and wrote the first draft. Dr. Weihong Song provided editorial assistance. Some of the material covered in section 1.1.2 was adapted from an unpublished review in which I was co-first author with Dr. Xiulian Sun. Dr. Weihong Song provided editorial assistance. Chapter 2 This work has been previously published (Liu, S., Bromley-Brits, K., Xia, K., Mittelholtz, J., Wang, R., and Song, W. (2008). Tmp21 degradation is mediated by the ubiquitin-proteasome pathway. Eur J Neurosci, 28(10):1980–1988). I was co-first author on the paper with a post-doctoral fellow from the Song lab, Dr. Shengchun Liu, and a collaborator from China, Dr. Kun Xia. Ms. Jill Mittelholtz was second author, and Dr. Ruitao Wang was third. Dr. Weihong Song was the principal investigator. I wrote the first draft of the paper. For this work, I cloned the TMP21 expression plasmids, created the stable cell lines HTM2 and NTM1, and performed the half-life experiments, in-cell western assay, sucrose gradients, and immunocytochemistry. For the proteasomal inhibitor work, I performed the experiments for the HTM2 cell line while Dr. Liu performed the work for the NTM1 cell line. For the lysosomal inhibitor work, I performed the experiments for the HTM2 cell line, while work on the NTM1 cell line was split equally between Dr. Liu and myself. There was also some collaboration on the lysosomal inhibitor experiments such that I would perform the experiment and Dr. Liu would run the western blot (Figure 2.4A, B). Dr. Xia generated the rabbit anti-TMP21 antibody in China. Dr. Liu character- ized the antibody (Figure 2.1A, B) and did the coimmunoprecipitation experiment with ubiquitin (Figure 2.7A). Ms. Mittelholtz and Dr. Wang performed additional experiments which were requested by the reviewers for publication, but were not iii Preface included in the paper or this thesis. Dr. Song was the principal investigator on the project, and as such provided the initial idea for the project and editorial expertise for the manuscript. Chapter 3 The pM-TMP and pVP16-TMP plasmids used in section 3.3.4 were created by Dr. Shengchun Liu. Dr. Jimmy Guo synthesized the TMP21 peptide array used in Figures 3.7 and 3.10 from a sequence I provided. The procedure for creating the peptide array described in section 3.2.7 was a slight re-wording of the procedure from Dr. Guo’s paper, used with his permission (Guo, J.-P., Petric, M., Campbell, W., and McGeer, P. L. (2004). Sars corona virus peptides recognized by antibodies in the sera of convalescent cases. Virology, 324(2):251–256). Chapter 4 Approximately 95% of the behaviour experiments shown in this thesis were per- formed by the author. Ms. Haiyan Zou, Dr. Fang Cai, Ms. Yu Deng, Ms. Xiaojie Zhang, and Ms. Rebecca Ko assisted with the remaining 5% of the experiments when I was unable to perform them due to scheduling conflicts. Ms. Zou was the primary individual responsible for maintaining the S2P23 and C57BL/6 colonies during the course of this experiment. The genotyping analysis was performed by Ms. Zou, with assistance from Ms. Mittelholtz, Ms. Fiona Zhang, and Mr. Odysseus Zis. Ms. Zou also assisted with the animal autopsies and brain tissue collection. Figure 4.1 and the included caption was originally published by Denzel et al. (2000) (Denzel, A., Otto, F., Girod, A., Pepperkok, R., Watson, R., Rosewell, I., Bergeron, J. J., Solari, R. C., and Owen, M. J. (2000). The p24 family member p23 is required for early embryonic development. Curr Biol, 10(1):55–58) and was used with permission from Elsevier (license #2604910209883). Permission to use the ANY-Maze TM behavioural apparatus images shown in Figures 4.5A, 4.3A, and 4.4A was obtained from Dr. Richard Mills, Chief Security Officer of the Neuroscience and Physiology Division of Stoelting Company. All procedures were approved by the University of British Columbia Animal Care Committee (Protocols A05-1888, A10-0040, and A06-0007). iv Table of Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x List of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii 1 General introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Alzheimer’s disease pathogenesis . . . . . . . . . . . . . . . . . . . . 1 1.1.1 Non-amyloidogenic pathway . . . . . . . . . . . . . . . . . . . 2 1.1.2 Amyloidogenic pathway . . . . . . . . . . . . . . . . . . . . . 5 1.2 The secretory pathway . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.2.1 Coat protein complex I and models of golgi transport . . . . 10 1.2.2 TMP21 in vesicle budding . . . . . . . . . . . . . . . . . . . . 13 1.2.3 TMP21 in Golgi structure . . . . . . . . . . . . . . . . . . . . 21 1.2.4 TMP21 in vesicle tethering . . . . . . . . . . . . . . . . . . . 23 1.3 TMP21 in Alzheimer’s disease . . . . . . . . . . . . . . . . . . . . . . 24 1.3.1 TMP21 as a component of the g-secretase complex . . . . . . 24 1.3.2 TMP21 and APP processing . . . . . . . . . . . . . . . . . . 25 1.3.3 Clinical relevance of TMP21 in Alzheimer’s disease . . . . . . 28 1.4 Overall goals of this research . . . . . . . . . . . . . . . . . . . . . . 29 1.4.1 Examine the biochemical properties of TMP21 . . . . . . . . 29 1.4.2 Investigate the effect of TMP21 on APP processing . . . . . . 29 1.4.3 Understand the behavioural consequences of TMP21 suppres- sion in C57BL/6 mice . . . . . . . . . . . . . . . . . . . . . . 30 v Table of Contents 2 TMP21 degradation is mediated by the ubiquitin-proteasome pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.2 cDNA constructs, cell cultures and transfection . . . . . . . . 32 2.2.3 Antibody generation and characterization . . . . . . . . . . . 33 2.2.4 Subcellular fractionation . . . . . . . . . . . . . . . . . . . . . 33 2.2.5 Pharmacological treatment . . . . . . . . . . . . . . . . . . . 34 2.2.6 Immunoprecipitation and immunoblotting . . . . . . . . . . . 34 2.2.7 In-cell western assay . . . . . . . . . . . . . . . . . . . . . . . 35 2.2.8 Immunocytochemistry . . . . . . . . . . . . . . . . . . . . . . 36 2.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 2.3.1 Antibody characterization and stable cell line creation . . . . 36 2.3.2 TMP21 has a half-life of approximately 3 hours . . . . . . . . 39 2.3.3 Lysosomal inhibition does not affect TMP21 protein levels . . 39 2.3.4 Proteasomal inhibition increases TMP21 protein levels . . . . 39 2.3.5 Proteasomal inhibition causes a time-dependent accumulation of TMP21 in the Golgi . . . . . . . . . . . . . . . . . . . . . . 42 2.3.6 TMP21 is ubiquitinated . . . . . . . . . . . . . . . . . . . . . 45 2.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 2.4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 2.4.2 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3 Overexpression of TMP21 alters APP processing . . . . . . . . . . 48 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.2.2 Cell culture and transfections . . . . . . . . . . . . . . . . . . 49 3.2.3 siRNA knockdown . . . . . . . . . . . . . . . . . . . . . . . . 49 3.2.4 Coimmunoprecipitation . . . . . . . . . . . . . . . . . . . . . 50 3.2.5 Immunoblotting . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.2.6 Immunocytochemistry . . . . . . . . . . . . . . . . . . . . . . 51 3.2.7 Dot-blot interaction array . . . . . . . . . . . . . . . . . . . . 52 3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.3.1 C83 production is lower when TMP21 is overexpressed . . . . 55 3.3.2 C99 production is higher when TMP21 is overexpressed . . . 57 3.3.3 TMP21 colocalizes and interacts with members of the g-secretase complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.3.4 TMP21 associates with BACE1 only when TMP21 is overex- pressed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 vi Table of Contents 3.4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 3.4.2 Results in context with the literature . . . . . . . . . . . . . . 69 3.4.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4 TMP21 heterozygous mice display heightened anxiety . . . . . . . 71 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.1.1 Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.1.2 The TMP21 heterozygous mouse, S2P23 . . . . . . . . . . . . 72 4.1.3 Behavioural testing . . . . . . . . . . . . . . . . . . . . . . . . 72 4.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.2.1 Animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.2.2 Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 4.2.3 General guidelines for behaviour testing . . . . . . . . . . . . 81 4.2.4 Motor testing . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 4.2.5 Open field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 4.2.6 Light-dark box . . . . . . . . . . . . . . . . . . . . . . . . . . 83 4.2.7 Y-maze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 4.2.8 Morris water maze . . . . . . . . . . . . . . . . . . . . . . . . 86 4.2.9 Contextual fear conditioning . . . . . . . . . . . . . . . . . . 88 4.2.10 Cued fear conditioning . . . . . . . . . . . . . . . . . . . . . . 88 4.2.11 Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.3.1 S2P23 appear grossly normal . . . . . . . . . . . . . . . . . . 91 4.3.2 With increasing age, S2P23 and control mice lose hang strength equally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.3.3 Three month old S2P23 mice have better rotarod performance than controls . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.3.4 Twelve month old S2P23 males travel more distance than con- trols, as a result of moving faster . . . . . . . . . . . . . . . . 94 4.3.5 S2P23 mice show heightened anxiety in the open field test . . 97 4.3.6 Six month old S2P23 males show heightened anxiety in the light-dark box test . . . . . . . . . . . . . . . . . . . . . . . . 102 4.3.7 S2P23 and control mice perform a similar number of sponta- neous alternations . . . . . . . . . . . . . . . . . . . . . . . . 102 4.3.8 S2P23 mice do not show deficiencies in spatial memory . . . 106 4.3.9 S2P23 mice show deficits in the contextual fear conditioning task . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 4.3.10 S2P23 females show heightened anxiety to novel environments at 12 months of age . . . . . . . . . . . . . . . . . . . . . . . 112 4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 4.4.1 Physical abilities . . . . . . . . . . . . . . . . . . . . . . . . . 114 4.4.2 Anxiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 vii Table of Contents 4.4.3 Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 4.4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 5 Conclusions and future directions . . . . . . . . . . . . . . . . . . . . 120 5.1 Overall conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 5.2 Significance of the research . . . . . . . . . . . . . . . . . . . . . . . 122 5.3 Strengths and weaknesses . . . . . . . . . . . . . . . . . . . . . . . . 122 5.3.1 Chapter 2: Degradation of TMP21 . . . . . . . . . . . . . . . 122 5.3.2 Chapter 3: TMP21’s effect on APP processing . . . . . . . . 124 5.3.3 Chapter 4: TMP21 in mouse behaviour . . . . . . . . . . . . 125 5.4 Potential applications in future research . . . . . . . . . . . . . . . . 126 5.4.1 The role of TMP21 in substrate processing . . . . . . . . . . 126 5.4.2 Further exploration of the S2P23 mouse . . . . . . . . . . . . 132 5.4.3 Effect of TMP21 suppression in an in vivo model of Alzheimer’s disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 viii List of Tables Table 3.1 Sequence information for the dot-blot interaction array . . . . 54 Table A.1 N values for growth curve data . . . . . . . . . . . . . . . . . . 168 Table A.2 N values for hanging wire data . . . . . . . . . . . . . . . . . . 168 Table A.3 N values for rotarod data . . . . . . . . . . . . . . . . . . . . . 169 Table A.4 N values for open field data . . . . . . . . . . . . . . . . . . . 169 Table A.5 N values for light-dark box data . . . . . . . . . . . . . . . . . 170 Table A.6 N values for Y-maze data . . . . . . . . . . . . . . . . . . . . . 170 Table A.7 N values for water maze data . . . . . . . . . . . . . . . . . . . 171 Table A.8 N values for contextual fear conditioning data . . . . . . . . . 171 Table A.9 N values for cued fear conditioning data . . . . . . . . . . . . 172 ix List of Figures Figure 1.1 APP processing . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Figure 1.2 Schematic diagram of the early stages of COPI vesicle assembly 14 Figure 1.3 Suggested model for the dual role of TMP21 in APP process- ing and trafficking . . . . . . . . . . . . . . . . . . . . . . . . 26 Figure 2.1 Antibody and stable cell line characterization . . . . . . . . . 37 Figure 2.2 TMP21-myc displays correct subcellular localization . . . . . 38 Figure 2.3 TMP21 has a half-life of approximately 3 hours . . . . . . . . 40 Figure 2.4 Lysosomal inhibition does not affect TMP21 protein levels . . 41 Figure 2.5 Proteasomal inhibition increases TMP21 protein levels . . . . 43 Figure 2.6 Proteasomal inhibition causes a time-dependent accumulation of TMP21 in the Golgi . . . . . . . . . . . . . . . . . . . . . . 44 Figure 2.7 TMP21 can be ubiquitinated . . . . . . . . . . . . . . . . . . 46 Figure 3.1 Sequence information for the dot-blot interaction array . . . . 53 Figure 3.2 Overexpression of TMP21 reduces C83 production . . . . . . 56 Figure 3.3 Overexpression of TMP21 increases C99 production . . . . . 58 Figure 3.4 Endogenous TMP21 colocalizes with members of the g-secretase complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Figure 3.5 Exogenous TMP21 colocalizes with members of the g-secretase complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Figure 3.6 BACE1 coimmunoprecipitates with TMP21 only when TMP21 is overexpressed . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Figure 3.7 Specific regions of TMP21 serve as potential interaction sites with Aph1, Pen2, Nicastrin, and PS2 . . . . . . . . . . . . . . 63 Figure 3.8 Specific regions of TMP21 may interact with Aph1, Nct, Pen2, and PS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Figure 3.9 TMP21 is more likely to colocalize with BACE1 when overex- pressed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Figure 3.10 The dot blot interaction array did not find an interaction site with BACE1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Figure 4.1 Generation of p23 mutant mice by gene targeting . . . . . . . 73 Figure 4.2 Behaviour testing schedule . . . . . . . . . . . . . . . . . . . . 81 Figure 4.3 Open field apparatus . . . . . . . . . . . . . . . . . . . . . . . 83 x List of Figures Figure 4.4 Light-dark box apparatus . . . . . . . . . . . . . . . . . . . . 84 Figure 4.5 Y-maze apparatus . . . . . . . . . . . . . . . . . . . . . . . . 85 Figure 4.6 Water maze apparatus . . . . . . . . . . . . . . . . . . . . . . 87 Figure 4.7 Fear conditioning apparatus . . . . . . . . . . . . . . . . . . . 89 Figure 4.8 S2P23 mice weigh more than controls at 12 months of age . . 92 Figure 4.9 With increasing age, S2P23 and C57BL/6 mice lose hang strength equally . . . . . . . . . . . . . . . . . . . . . . . . . 93 Figure 4.10 Three month old S2P23 mice have better rotarod performance than controls . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Figure 4.11 Twelve month old S2P23 males travel further and faster than controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Figure 4.12 C57BL/6 and S2P23 mice show different rearing tendencies . 98 Figure 4.13 Six month old S2P23 interact with the Inner zone less than controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Figure 4.14 Experienced, 6 month old S2P23 mice spend less time in the open field Inner zone . . . . . . . . . . . . . . . . . . . . . . . 101 Figure 4.15 Six month S2P23 males hesitate longer before entering the Light side and spend less time there overall . . . . . . . . . . 103 Figure 4.16 S2P23 and control mice make a similar number of entries to the Light side . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Figure 4.17 All mice have the same level of general locomotor activity in the Y-maze . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Figure 4.18 All mice perform a similar number of spontaneous alternations in the Y-maze . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Figure 4.19 S2P23 and C57BL/6 mice perform similarly in the water maze visible platform condition . . . . . . . . . . . . . . . . . . . . 108 Figure 4.20 S2P23 and C57BL/6 mice have similar path lengths in water maze hidden platform trials . . . . . . . . . . . . . . . . . . . 109 Figure 4.21 S2P23 and C57BL/6 mice have similar escape latencies in wa- ter maze hidden platform trials . . . . . . . . . . . . . . . . . 110 Figure 4.22 S2P23 and C57BL/6 mice spend a similar amount of time in the platform quadrant in the probe trial . . . . . . . . . . . . 111 Figure 4.23 S2P23 have a decreased conditioned fear response . . . . . . . 113 Figure 4.24 S2P23 females show heightened anxiety to novel environments at 12 months of age . . . . . . . . . . . . . . . . . . . . . . . 115 Figure 5.1 TMP21 partially colocalizes with APP . . . . . . . . . . . . . 128 Figure 5.2 Schematic of Notch processing . . . . . . . . . . . . . . . . . 130 Figure 5.3 TMP21 coimmunoprecipitates with Notch . . . . . . . . . . . 131 xi List of Abbreviations Ab Amyloid-b AD Alzheimer’s disease ALLM N-acetyl-L-leucinyl-L-leucinyl-L-methional ALLN N-acetyl-L-leucinyl-L-leucinyl-L -norleucinal ARF ADP-ribosylation factor Aph-1 Anterior pharynx-defective 1 AICD APP intracellular domain APP Amyloid-b precursor protein APPSwe APP Swedish mutation APPwt APP wildtype ARU Animal Research Unit BACE1 b-site APP Cleaving Enzyme 1 BACE2 b-site APP Cleaving Enzyme 2 BFA brefeldin A BIG1 BFA-inhibited guanine nucleotide exchange factor 1 BSA bovine serum albumin CHX Cycloheximide COPI Coat protein complex I COPII Coat protein complex II CTFa C-terminal fragment a CTFb C-terminal fragment b xii List of Abbreviations DCM dichloromethane DEPC diethylpyrocarbonate DMEM Dulbecco’s modified eagles’ medium DMF dimethylformamide ER endoplasmic reticulum FBS fetal bovine serum FMOC 9-fluorenylmethyloxycarbonyl GBF1 Golgi BFA resistance factor 1 GEFs guanine nucleotide exchange factors GM130 golgin subfamily A member 2 GRASP65 Golgi re-assembly stacking protein p65 HB homogenization buffer HEK human embryonic kidney HeLa human cervical HTM2 HEK TMP21-mycHis MG-132 N-carbobenzoxyl-L-leucinyl-L-leucinyl-L-leucinal N2a mouse neuroblastoma Nct Nicastrin NECD Notch extracellular domain NEXT Notch extracellular truncation NICD Notch intracellular domain NTM1 N2a TMP21-mycHis OBB Odyssey R© blocking buffer p115 general vesicular transport factor p115 Pen-2 Presenilin enhancer 2 PBS phosphate buffered saline xiii List of Abbreviations PBS-Tx PBS with 0.1% Triton X-100 PBS-T PBS with 0.1% Tween-20 PCR polymerase chain reaction PFA paraformaldehyde PNS post-nuclear supernatant PS1 Presenilin 1 PS2 Presenilin 2 PVDF-FL Immobilon R©-FL polyvinylidene difluoride RIPA-DOC radio-immunoprecipitation assay – deoxycholate sAPPb secretory APPb sAPPa secretory APPa SDS sodium dodecyl sulfate TACE tissue necrosis factor a converting enzyme SH-SY5Y human neuroblastoma SNAREs soluble NSF attachment protein receptors TAP transcytotic vesicle-associated protein TGN trans-Golgi network TMP21 transmembrane emp24-like trafficking protein 10 xiv Acknowledgments While the merit of my degree shall be judged on the contents herein, I feel my PhD encompasses much more beyond these pages. Throughout this journey I have had the support and guidance of many kind individuals, who have nurtured both my intellectual and personal development and have, as a result, shaped me into the person I am today. First and foremost, I would like to thank my supervisor, Dr. Weihong Song, for the excellent training I received while being a member of his laboratory. Dr. Song taught me to think critically yet creatively, and to work quickly yet efficiently. With Dr. Song’s support I was able to pursue multiple avenues of career development, taking advantage of many professional conferences and workshops offered at UBC, as well as nationally and internationally. I would especially like to thank him for his sensitive flexibility when I took maternity leave for the birth of my son, and for his understanding as I slowly transitioned back to life at the bench. For this I am very grateful. I would also like to thank the members of my supervisory committee, Dr. Ka- terina Dorovini-Zis, Dr. Catharine Rankin, and Dr. Peter Reiner, for their support and encouragement throughout my degree. Their membership in my committee made the annual supervisory meeting a process I looked forward to, both for their intellectual contributions and their geniality. I would especially like to thank Dr. Rankin, who introduced me to research life at UBC during my first PhD rota- tion, and has always been approachable for advice on topics ranging from my thesis project, to career planning, to balancing work and personal life. Many thanks. To all my colleagues in the Song lab, both past and present, thank you for your camaraderie. A special thanks to Haiyan Zhou for her hard work caring for the mice and helping any way she could, and to Fang Cai and Candy Deng for assisting with behaviour testing during scheduling conflicts. Thank you as well to Weihui Zhou for teaching me several techniques in his patient way, to Xiaojie Zhang for continuing this project in the future, and to Qian Xu and Yi Yang for being great bench mates. Also, a big thanks to Shuting Zhang, Xiaozhu Zhang, Mingming Zhang, Yili Wu, Philip Ly, Odysseus Zis, and Rebecca Ko for our therapeutic chats! Finally, a special thanks to Derrick Lee for our conversations on statistics, which were very helpful. I would also like to thank my Neuroscience classmates, especially Blair Duncan and Sharmin Hossain, for their company along the long and sometimes winding road to the PhD. Sharing the ups and downs with them gave me the extra fuel to carry on. Similarly, I’d like to thank my brother Sam, who has been through this process xv Acknowledgments before, for being a sounding board along the way. Last, but certainly not least, I would like to extend my deepest gratitude to my amazing husband, Lionel Brits, who provided the love, support, and encouragement needed to complete my journey. His understanding as I worked late nights in the lab or at home in the office, his encouragement when experiments failed, and his ability to help me refocus on the important things in life were just a few of the many acts for which I am grateful. I am also immensely proud of his selfless decision to leave his PhD in Physics to care for myself and my son. His sacrifice allowed our son to be raised by his father, and allowed me to devote myself more fully to pursuing my degree. His sacrifice for our family still amazes me. xvi Dedication For my husband, Lionel Charles Brits for his constant and unwavering support. For my son, Nikolas Thomas Brits for getting by with less Mommy-time than I would like. And for my grandmother, Mildred Jane Hapgood who died of Alzheimer’s in 2007. xvii Chapter 1 General introduction 1.1 Alzheimer’s disease pathogenesis Alzheimer’s disease (AD) is the most common neurodegenerative disorder leading to dementia. Patients with AD display progressive cognitive impairment and memory loss. One in 10 people over age 65 and 1 in 3 people over age 85 have AD (AA-USA, 2010; ASC, 2010). Each year we can expect to see more AD cases than the year previous due to a population-wide increase in the greatest risk factor for AD – age. The next generation will see more than double the current number of AD patients, straining families, social systems, and the already burdened economy. The two major neuropathological hallmarks of AD are the deposition of Amyloid- b (Ab) protein in the form of neuritic plaques (Glenner and Wong, 1984), and the formation of neurofibrillary tangles largely consisting of hyperphosphorylated tau (Grundke-Iqbal et al., 1986a; Wood et al., 1986; Kosik et al., 1986; Ihara et al., 1986; Grundke-Iqbal et al., 1986b). Ab is generated from a larger Amyloid-b pre- cursor protein (APP) following sequential cleavage by b-secretase and g-secretase (Figure 1.1). Despite the robust expression of the APP gene and the resulting high level of APP protein in vivo, Ab production through the amyloidogenic pathway of APP processing is a rare occurrence under normal conditions (Li et al., 2006). The majority of APP is cleaved first by a-secretase, rather than b-secretase, within the Ab domain in the non-amyloidogenic pathway, generating a secreted form of APP, secretory APPa (sAPPa), and C-terminal fragment C83 (C-terminal frag- 1 1.1. Alzheimer’s disease pathogenesis ment a (CTFa)) (Oltersdorf et al., 1990; Esch et al., 1990; Sisodia et al., 1990). C83 can be further cleaved by g-secretase, producing extracellular fragment p3 and intracellular APP intracellular domain (AICD) (Kim et al., 2003; Edbauer et al., 2003). 1.1.1 Non-amyloidogenic pathway a-secretase APP is a type I transmembrane protein with three major isoforms encoded by a single gene on chromosome 21: APP695 (Kang et al., 1987), APP751 (Ponte et al., 1988; Tanzi et al., 1988), and APP770 (Kitaguchi et al., 1988). APP695 is the only isoform lacking an extracellular Kunitz Protease Inhibitor domain and is the major isoform enriched in the neuronal cells of the brain. APP is processed through the secretory pathway, undergoing N -glycosylation in the endoplasmic reticulum (ER) and O-glycosylation in the Golgi (Dyrks et al., 1988; Weidemann et al., 1989), as well as phosphorylation and sulfation in the late Golgi and at the cell surface (Gandy et al., 1988; Weidemann et al., 1989; Buxbaum et al., 1990). Only a small proportion of holo-APP is detected at the cell-surface when com- pared to the total cellular quantity due to a cell-surface half-life of less than 10 minutes (Koo and Squazzo, 1994; Koo et al., 1996). At the cell-surface, a fraction (≈14%) of holo-APP undergoes cleavage between Lys16 and Leu17 within the Ab domain by a-secretase, precluding amyloidogenesis and producing a soluble frag- ment, sAPPa, and a membrane bound 10 kDa C-terminal fragment, C83 (CTFa), which has a half-life of 4 h (Selkoe et al., 1988; Oltersdorf et al., 1990; Esch et al., 1990; Sisodia et al., 1990; Caporaso et al., 1992). Radiolabelling indicates that ap- proximately 92% of C83 is produced by this intraamyloid cleavage in the secretory pathway (Caporaso et al., 1992). Cell-surface holo-APP, as well as C83, can also 2 1.1. Alzheimer’s disease pathogenesis Figure 1.1: APP processing. APP (gray rectangle) can be cleaved in two path- ways: the more common, non-pathogenic, non-amyloidogenic pathway (bottom), or the pathogenic, amyloidogenic pathway (top). In normal conditions, the majority of APP is cleaved within the Ab domain (small gray rectangle) by a-secretase to pro- duce sAPPa and membrane-bound C83. C83 can be further cleaved by g-secretase, producing extracellular fragment p3 and intracellular AICD. In the pathogenic path- way, APP is first cleaved by b-secretase at the start of the Ab domain, producing sAPPb and membrane-bound CTFb (C89/C99). Cleavage of CTFb by g-secretase yields pathogenic Ab fragments, and intracellular AICD. 3 1.1. Alzheimer’s disease pathogenesis be targeted to the endosomal/lysosomal pathway through clarithin-coated vesicles to undergo degradation (Haass et al., 1992a; Golde et al., 1992; Nordstedt et al., 1993). BACE2 as θ-secretase b-site APP Cleaving Enzyme 1 (BACE1) is a type I integral membrane protein and the b-secretase in vivo (Hussain et al., 1999; Sinha et al., 1999; Vassar et al., 1999; Yan et al., 1999; Cai et al., 2001; Luo et al., 2001; Roberds et al., 2001; Sun et al., 2005, 2006a,c). Although b-site APP Cleaving Enzyme 2 (BACE2) is the homolog of BACE1 (Yan et al., 1999; Acquati et al., 2000; Lin et al., 2000), neural tissue expression of BACE2 mRNA is low or undetectable, contrary to what would be expected of a b-secretase (Bennett et al., 2000). The promoter region, 5′UTR, and 3′UTR of BACE1 and BACE2 show no similarities, and the transcription of both proteins is distinctly regulated (Sun et al., 2005). Furthermore, BACE2 RNAi increases, and overexpression decreases, Ab production, suggesting BACE2 is anti- amyloidogenic (Basi et al., 2003; Sun et al., 2005). Lentiviral overexpression of BACE2 in neuronal cultures derived from Swedish mutant APP transgenic AD model mice also reduced Ab production (Sun et al., 2006b). Recently, BACE2 was found to cleave at a new θ-site in APP between F19 and F20 (KLVF*FAE), precluding Ab formation (Sun et al., 2006b). These data suggest that BACE2 is not functionally homologous to BACE1 and does not function as a b-secretase in vivo; rather, BACE2 functions as an θ-secretase, sharing cleavage similarities with a-secretase. 4 1.1. Alzheimer’s disease pathogenesis 1.1.2 Amyloidogenic pathway b-secretase (BACE1) BACE1, the in vivo b-secretase, is located on chromosome 11, and BACE1 protein is most highly expressed in pancreas and brain (Yan et al., 1999). BACE1 gene expression is tightly regulated at the transcriptional level and can be enhanced by Sp1 and oxidative stress (Christensen et al., 2004; Tong et al., 2005); however, under normal conditions gene expression is low due to weak promoter activity and the negative effects of the 5′-untranslated region on translation initiation, resulting in only a minority of APP undergoing amyloidogenic processing (Li et al., 2006; Zhou and Song, 2006). APP is the best-characterized BACE1 substrate. The major cleavage site is located between Met596 and Asp597 of the APP695 isoform (Asp1 of Ab), produc- ing secretory APPb (sAPPb) and C-terminal fragment C99 (C-terminal fragment b (CTFb)). BACE1 also cleaves at a minor site between Tyr10 and Glu11 of Ab to release a lower level C89 fragment both in vivo (Masters et al., 1985; Näslund et al., 1994) and in vitro (Haass et al., 1992b). Subsequent cleavage of C99 by g-secretase results in Ab production. The human APP Swedish mutation (APPSwe), in which Lys595-Met596 (KM) is mutated to Asn595-Leu596 (NL), increases b-secretase activ- ity and Ab deposition, resulting in early onset AD (Citron et al., 1992). BACE1 is localized in the Golgi and endosomes (Vassar et al., 1999). Wildtype APP cleavage is suggested to primarily occur in endosomal compartments, as disrup- tion of the YENP endocytosis signal in the cytoplasmic tail of APP diminishes Ab production, as does incubating cells in low-potassium media and thereby inhibiting endocytosis via inhibition of clathrin-coat assembly (Koo and Squazzo, 1994; Perez et al., 1999). Ab is therefore generated intracellularly, but is quickly secreted in both neuronal and non-neuronal cell lines (Wertkin et al., 1993; Perez et al., 1996). 5 1.1. Alzheimer’s disease pathogenesis Treatment with brefeldin A (BFA), which inhibits anterograde transport of APP in the secretory pathway, can inhibit intracellular Ab production and Ab secretion in both APP wildtype (APPwt) and APPSwe cells, suggesting that cleavage does not occur in the proximal Golgi (Haass et al., 1993; Martin et al., 1995). Contrarily, treatment with monensin, an ionophore which inhibits trans-Golgi protein matura- tion and transport through the trans-Golgi network (TGN) (Tartakoff, 1983), can inhibit Ab secretion in both APPwt and APPSwe cells, but only partially inhibit intracellular Ab production in APPSwe cells, suggesting that APPSwe is cleaved in the TGN (Haass et al., 1993; Martin et al., 1995; Stephens and Austen, 1996). g-secretase complex PS1 and PS2 The presenilin proteins, Presenilin 1 (PS1) and Presenilin 2 (PS2), were the first two identified components of the g-secretase complex. Notch under- goes intramembranous S3 cleavage by the g-secretase complex to generate Notch intracellular domain (NICD), which can then translocate to the cell nucleus and function in gene transcription (Kopan et al., 1996; Schroeter et al., 1998). PS1 knockout (PS1-/-) mice die at embryonic day 19 and display a Notch null pheno- type, such as deficits in somatogenesis and spinal ganglia patterning (Shen et al., 1997; Wong et al., 1997), while PS1-/-PS2-/- double knockout mice die at embryonic day 9.5 with deficits in somite segmentation, loss of midbrain mesenchyme cells, ventral neural tube defects, and other problems associated with loss of Notch sig- nalling (Donoviel et al., 1999). Furthermore, the production of NICD is diminished in PS1-deficient cells (Song et al., 1999; De Strooper et al., 1999), suggesting PS contains the g-secretase activity. 6 1.1. Alzheimer’s disease pathogenesis Missense mutations in PS1 and PS2 are also a major cause of familial, early onset AD (Mullan et al., 1992; Schellenberg et al., 1992; George-Hyslop et al., 1992; Levy- Lahad et al., 1995a,b; Li et al., 1995; Rogaev et al., 1995; Sherrington et al., 1995), and these mutations accelerate Ab deposition and neurodegeneration in model mice (Borchelt et al., 1996; Scheuner et al., 1996; Borchelt et al., 1997), suggesting the PS proteins play a major role in APP processing. PS1 null cells show decreased Ab production and accumulation of APP CTFs (De Strooper et al., 1998; Naruse et al., 1998), dominant negative aspartyl mutants of PS1 and PS2 inhibit Ab production (Steiner et al., 1999; Wolfe et al., 1999; Kimberly et al., 2000), and transition- state inhibitors of g-secretase can directly bind to PS1 (Esler et al., 2000; Li et al., 2000b). Furthermore, both NICD and Ab generation are completely inhibited in PS1-/-PS2-/- cells, indicating that the PSs are absolutely required for both g- and e- site cleavages (Herreman et al., 2000; Zhang et al., 2000). The g- and e-site cleavages seem to be independently regulated however, as certain g-secretase inhibitors can affect APP cleavage without affecting Notch cleavage (Petit et al., 2001), and PS mutations can increase Ab-42 production while decreasing Notch cleavage (Song et al., 1999; Chen et al., 2002). While the PS proteins seem to contain the enzymatic active site of g-secretase, overexpression of full-length PS1 does not increase g-secretase activity, suggesting that other limiting factors are required (Thinakaran et al., 1996, 1997; Ratovit- ski et al., 1997). Holo-PS undergoes endoproteolysis to produce stable N- and C-terminal fragments (NTF and CTF, respectively); PS1 is cleaved to produce a 27-28-kDa NTF and a 16-17-kDa CTF (Steiner et al., 1998), while PS2 is cleaved to produce a 34-kDa NTF and a 20-kDa CTF (Kim et al., 1997). These fragments colocalize in the ER and Golgi (Kovacs et al., 1996), and remain associated with each other (Seeger et al., 1997; Capell et al., 1998). Overexpression of full-length 7 1.1. Alzheimer’s disease pathogenesis PS1 results in accumulation of the PS1 holoprotein, as the level of PS1 CTF and NTF is tightly regulated (Thinakaran et al., 1996, 1997; Ratovitski et al., 1997), sug- gesting that further factors are necessary for reconstitution of g-secretase activity. These factors contribute to the high molecular weight complex in which g-secretase is active (Li et al., 2000a). Nicastrin Nicastrin (Nct) was the first of these limiting factors to be identified. Nct RNAi in C. elegans resulted in a Notch phenotype similar to that seen following disruption of C. elegans PS homologs sel-12 and hop-1 (Yu et al., 2000b), and later work showed similar findings in Drosophila (Chung and Struhl, 2001; Hu et al., 2002; López-Schier and Johnston, 2002). Nct interacts directly with active g-secretase (Esler et al., 2002; Kimberly et al., 2002), and PS and Nct co-regulate each other; Nct deficiency destabilizes PS (Edbauer et al., 2002; Hu et al., 2002), while PS is required for Nct maturation (Edbauer et al., 2002; Kimberly et al., 2002; Leem et al., 2002). Similar to overexpression of PS, Nct overexpression results in an accumulation of immature Nct with no increase in g-secretase activity (Kimberly et al., 2002; Yang et al., 2002), suggesting a further limiting factor is necessary for reconstitution of g-secretase activity. Aph-1 and Pen-2 Anterior pharynx-defective 1 (Aph-1) and Presenilin enhancer 2 (Pen-2) were identified in a C. elegans genetic screen where aph-1 and pen-2 mutants displayed a subset of the Notch-defective phenotype, while aph-1 mutants also showed mislocalization of Nct, similar to that seen in sel-12 and hop-1 mutants (Goutte et al., 2002; Francis et al., 2002). Overexpression of Aph-1, Nct, and PS in mammalian cells resulted in increased levels of relatively stable holo-PS with no increase in g-secretase activity. When Pen-2 was also expressed, levels of holo- PS1 decreased while PS1 CTFs and NTFs increased along with g-secretase activity 8 1.1. Alzheimer’s disease pathogenesis (Kim et al., 2003; Luo et al., 2003). Similar results were found in Drosophila, suggesting that Nct and Aph-1 stabilize PS1 holoprotein, while Pen-2 is required for PS1 cleavage (Takasugi et al., 2003). Nct is also suggested to act as the substrate receptor for g-secretase (Shah et al., 2005). In Drosophila S2 cells, RNAi of PS, Nct, Aph-1, or Pen-2 can disrupt g-secretase activity, and in mammalian cells Pen-2, Nct, PS, and Aph-1 coprecipitate with each other (Steiner et al., 2002; Edbauer et al., 2003; Kimberly et al., 2003; Takasugi et al., 2003). Thus, PS, Nct, Aph-1, and Pen-2 comprise the core-limiting factors of the g-secretase complex. Modifiers of g-secretase activity While no other critical PS-interacting pro- teins were identified in a thorough C. elegans genome scan (Francis et al., 2002), modifiers of the g-secretase complex have been identified. Upon further examina- tion, some of these proteins do not seem to directly modify g-secretase activity. CD147, for example, was identified as a component of the g-secretase complex in HeLa cells, and RNAi of CD147 increased Ab production, suggesting CD147 func- tioned as a negative regulator of g-secretase activity (Zhou et al., 2005); however, it was later shown that CD147 stimulates production of matrix metalloproteinases which degrade extracellular Ab, thus modifying Ab levels indirectly (Vetrivel et al., 2008b). Recently, transmembrane emp24-like trafficking protein 10 (TMP21) was iden- tified as a new member of the g-secretase complex which could negatively regulate g-secretase activity (Chen et al., 2006). Unlike CD147, it is likely that TMP21 is a direct modifier, as TMP21 immunoprecipitation could pull down all four core g-secretase complex members, and results were reproducible in a cell-free system (Chen et al., 2006). TMP21 is degraded through the ubiquitin-proteasome pathway (Liu et al., 2008), similar to other proteins involved in AD pathogenesis, including the core members of the g-secretase complex, PS1 (Fraser et al., 1998; Steiner et al., 9 1.2. The secretory pathway 1998; Honda et al., 1999), PS2 (Kim et al., 1997), Nct (He et al., 2007), Pen-2 (Bergman et al., 2004; Crystal et al., 2004), and Aph-1 (He et al., 2006), as well as other proteins implicated in AD, such as tau (David et al., 2002; Keck et al., 2003), the APP C-terminus (Nunan et al., 2001), and BACE1 (Qing et al., 2004; Zhou et al., 2004). The role of TMP21 in AD pathology has just recently been explored when com- pared to its role in protein trafficking. This introduction will focus on the dual roles of TMP21 and the current knowledge on its new role as a regulator of g-secretase activity. 1.2 The secretory pathway As secretory proteins are synthesized they pass through the ER and the Golgi ap- paratus, to then be released into the extracellular space in secretory granules, or vesicles; this is the secretory pathway (Palade, 1975). Proper vesicle transport re- quires four essential steps: budding of vesicles from the donor membrane, movement to the acceptor membrane either by diffusion or along the cytoskeletal track, teth- ering to the acceptor membrane, and finally fusion of the vesicle with the acceptor membrane. Plasma membrane, endosomal, and lysosomal proteins are processed through the early secretory pathway, which includes the ER and Golgi. 1.2.1 Coat protein complex I and models of golgi transport Transport within the Golgi can be described by two models. Cisternal progression, the earliest model, describes a progressive movement and maturation of Golgi cister- nae from the cis- to trans-face, with membranes from the ER forming new cisternae at the cis-face, which then progress through the Golgi stack to be fragmented and disassembled at the trans-Golgi (Beams and Kessel, 1968). This model is supported 10 1.2. The secretory pathway by current evidence that procollagen undergoes anterograde transport from the cis- to trans-Golgi without leaving the lumen of the cisternae (Bonfanti et al., 1998); however, it does not explain why secretory proteins transverse the Golgi in minutes while Golgi resident proteins have half-lives of hours and remain polarized at either the cis- or trans-face (Farquhar, 1985; Dunphy and Rothman, 1985). If the stack progressed as a whole we would expect Golgi resident proteins and secretory proteins to transverse at the same rate. This polarization was accounted for by the retro- grade transport of Golgi resident proteins by Coat protein complex I (COPI) coated vesicles (Lewis and Pelham, 1992; Letourneur et al., 1994; Cosson and Letourneur, 1994). COPI-coated vesicles were originally identified in the Golgi cisternae (Balch et al., 1984) and were shown to have a coat which was distinct from previously iden- tified clathrin-coated vesicles (Orci et al., 1986). When transport within the Golgi was blocked by non-hydrolysable GTP analogue GTPgS, COPI vesicles accumulated from the donor Golgi and could not fuse with acceptor Golgi membranes, suggesting that COPI vesicles mediated anterograde transport rather than retrograde transport as previously suggested by the cisternal progression model (Melançon et al., 1987). These data lead to the formation of the vesicular model of Golgi transport, in which Golgi cisternae were static structures which proteins transversed through the bud- ding and fusion of vesicles in the anterograde direction. Purification of these COPI-coated vesicles revealed two major components, am- phipathic protein ADP-ribosylation factor (ARF) and cytosolic protein coatomer (Malhotra et al., 1989; Serafini et al., 1991a; Waters et al., 1991; Donaldson et al., 1992a). Addition of ARF and coatomer were sufficient to induce budding in Golgi preparations, while absence of either component significantly reduced vesicle num- bers (Orci et al., 1993b,a), indicating both are vital to vesicle budding. The membrane- 11 1.2. The secretory pathway mediated budding hypothesis states that Golgi membranes have an intrinsic bud- ding ability, with vesicles budding off from the tips of Golgi tubules. This would predict that a reduction in coat proteins would cause tubule extension to produce compensatory vesicles; however, this does not occur, indicating that budding is coat- dependent and not an intrinsic Golgi membrane property (Orci et al., 1993b,a). While there are data to support both models, the cisternal progression model does not account for COPI vesicles which travel at different rates (Karrenbauer et al., 1990), or in the anterograde direction (Pepperkok et al., 1993; Peter et al., 1993), and the vesicular transport model does not account for proteins which are too large to be transported by vesicles, yet still transverse the Golgi stack (Bonfanti et al., 1998). When it was discovered that COPI vesicles could travel bidirectionally (Fiedler et al., 1996; Orci et al., 1997), a fusion percolating vesicle model was proposed (Orci et al., 2000). This was supported by the discovery of an anterograde COPI vesicle population containing Golgi-restricted v-SNARE GOS28, and a retrograde COPI vesicle population containing ER-directed KDEL receptors (Orci et al., 1997, 2000). Each cisternae in the Golgi stack was found to contain both GOS28 and its t-SNARE partner syntaxin 5, suggesting anterograde cargo can fuse with either the more cis- or more trans-cisternae, resulting in a random walk pattern (Orci et al., 2000). Proteins would continue to progress through the stack due to the steady influx of cargo from the cis-Golgi and its exit at the trans-Golgi. This model allows both retrograde transport of ER proteins and anterograde transport of cargo and accounts for the varying transport rates observed, allowing large proteins to transverse the stack via cisternal progression and others to transverse via COPI vesicle populations (Orci et al., 2000). 12 1.2. The secretory pathway 1.2.2 TMP21 in vesicle budding ADP-ribosylation factor ARF was first described as the in vitro requirement for ADP ribosylation of the Gs regulatory subunit of adenylate cyclase by cholera toxin (Schleifer et al., 1982; Kahn and Gilman, 1984), and was later shown to be a GTP binding protein (Kahn and Gilman, 1986). Six ARFs have currently been identified in mammals; these are divided into three classes based on gene structure, amino acid sequence, size, and phylogenetic analysis: Class I (ARFs 1, 2, and 3), class II (ARFs 4 and 5), and class III (ARF6) (Tsuchiya et al., 1991). Class I ARFs, in particular ARF1, are the most studied and are involved not only in vesicle budding through recruitment of coatomer (Serafini et al., 1991a; Donaldson et al., 1992a), but also in recruitment of the AP-1 clathrin adaptor complex to Golgi membranes (Stamnes and Rothman, 1993; Traub et al., 1993) and the AP-3 adaptor complex to the TGN and/or endosomes (Ooi et al., 1998). ARF1 is an amphipathic protein which has a low affinity for binding to Golgi membranes in the ARF-GDP state and a higher affinity for binding following gua- nine nucleotide exchange (Serafini et al., 1991a; Helms and Rothman, 1992; Franco et al., 1993; Helms et al., 1993). ARF-GDP can interact loosely with Golgi mem- branes through its myristoylated N-terminus; however, this myristoylation is not responsible for the GTP-dependent increase in membrane binding affinity (Franco et al., 1993). When converted to the GTP-bound state by guanine nucleotide ex- change factors (GEFs), an N-terminal alpha-helix is released from ARF’s protein core, exposing hydrophobic residues which aid in membrane binding (Antonny et al., 1997) (Figure 1.2). It is now known that fungal metabolite BFA, long used in trafficking studies to disrupt Golgi transport (Lippincott-Schwartz et al., 1989; Doms et al., 1989), exerts 13 1.2. The secretory pathway Figure 1.2: Schematic diagram of the early stages of COPI vesicle assem- bly. Despite having a hydrophobic myristoyl group (zig-zag line) which can interact with membrane lipids, ARF-GDP has a low membrane binding affinity. Interactions between ARF-GDP and membrane-bound TMP21 facilitate guanine-nucleotide ex- change on ARF by GEFs (1). This causes a conformational change in ARF which allows TMP21 to dissociate and exposes a hydrophobic alpha helix which can aid membrane binding (2). Cytosolic coatomer can bind in a bimodal fashion, via its b-COP and g-COP subunits, to both ARF-GTP and TMP21, enhancing polymer- ization of the COPI vesicle coat. 14 1.2. The secretory pathway its effect by interfering with guanine nucleotide exchange on ARF (Donaldson et al., 1992b; Helms and Rothman, 1992). Several ARF-GEFs have been described, all possessing an approximately 200 amino acid Sec7 domain necessary for its activity. These GEFs can be subdivided into two major categories: those of high (>100 kDa) molecular weight, and those of low (45-50 kDa) molecular weight. Low molecular weight GEFs have a pleckstrin homology domain in addition to a Sec7 domain, are not BFA-sensitive, and are thought to be involved in cytoskeletal remodeling and endosomal recycling (Donaldson, 2003, for review). High molecular weight mam- malian GEFs include BFA-inhibited guanine nucleotide exchange factor 1 (BIG1), also known as p200 (Narula et al., 1992; Morinaga et al., 1996, 1997, 1999), BIG2 (Togawa et al., 1999), and Golgi BFA resistance factor 1 (GBF1) (Claude et al., 1999). All are found in the Golgi and are involved in membrane trafficking (Man- sour et al., 1999; Yamaji et al., 2000; Claude et al., 1999), and all except GBF1 are BFA-sensitive (Morinaga et al., 1997; Togawa et al., 1999; Claude et al., 1999). BIG2 and BIG1 are associated with the TGN and membrane recruitment of AP- 1 (Shinotsuka et al., 2002; Narula and Stow, 1995; Ishizaki et al., 2008), while GBF1 is associated with the cis-Golgi and membrane recruitment of COPI (Zhao et al., 2002; Kawamoto et al., 2002), yet they both activate the same ARF. This suggests that the GEF, rather than the ARF itself, determines the localization and type of coat recruited (Kawamoto et al., 2002). Golgi preparations from cells overexpressing GBF1 showed that it could function as a GEF for class I ARFs; however, in vitro overexpression, while allowing cells to grow in the presence of BFA, found GBF1 functioning as a GEF for class II ARFs (Claude et al., 1999). Subsequent in vivo studies showed GBF1 could facilitate exchange on both classes of ARFs and may be influenced by membrane lipid composition (Kawamoto et al., 2002). GBF1 rapidly cycles on and off membranes, becoming stabilized on the membrane when complexed 15 1.2. The secretory pathway with ARF-GDP (Szul et al., 2005; Niu et al., 2005). Following its exchange of GDP for GTP on ARF, GBF1 dissociates from the membrane, suggesting that continuous cycling of GBF1 on the membrane is necessary for ARF activation and vesicle formation (Szul et al., 2005). Coatomer Coatomer is a cytosolic heteroheptameric complex composed of a (≈160 kDa), b (107 kDa), g (98 kDa), d (61 kDa) (Waters et al., 1991; Serafini et al., 1991b; Duden et al., 1991), b’ (102 kDa) (Stenbeck et al., 1993; Harter et al., 1993), e (≈36 kDa) (Hara-Kuge et al., 1994), and z (20 kDa) (Kuge et al., 1993) subunits. Later, additional isoforms of g-COP and z-COP, labelled g2 and z2, were identified and the original isoforms were relabelled as g1 and z1 (Blagitko et al., 1999; Futatsumori et al., 2000; Yamasaki et al., 2000). Each subunit of coatomer is present in only a single copy, suggesting that the subunit isoform may affect coatomer function; in support of this, g2/z1-COP is the predominant form on COPI vesicles (Wegmann et al., 2004). Furthermore, while g2/z1- and g1/z2-coatomer colocalize with total coatomer, there are areas where the isoforms do not colocalize with each other (Futatsumori et al., 2000; Wegmann et al., 2004). Approximately 70% and 80% of g1- and z2- COP are localized in the early cis-Golgi, respectively, while g2-COP has a <40% cis and >60% trans distribution. Comparatively, overall coatomer, as measured by detecting b’-COP, is distributed equally over the Golgi, with <55% cis and >45% trans (Moelleken et al., 2007). The major mammalian coatomer subcomplexes g1/z1, g1/z2, and g2/z1 occur with a ratio of approximately 2 : 1 : 2 in rat HepG2 cell liver cytosol (Wegmann et al., 2004) and 10:3:5 in mouse 3T3/NIH fibroblast cytosol (Moelleken et al., 2007). g2/z2-COP, if present, was below detection levels 16 1.2. The secretory pathway and <5% of the total (Moelleken et al., 2007). It was originally hypothesized that coatomer was incorporated into vesicles en bloc (Waters et al., 1991); this was supported by an invariant molar ratio of coatomer subunits in purified coatomer, Golgi-bound coatomer, and coated vesicle coatomer, with the exception of z-COP, which was enriched in assembled coats (Hara-Kuge et al., 1994). Furthermore, coatomer which has been disassociated under high salt conditions can spontaneously reassemble in vitro following dialysis and be incorpo- rated onto membranes (Lowe and Kreis, 1995). Assembly of coatomer takes approx- imately 1 to 2 hours and under normal conditions no partial complexes are detected; once assembled, the heteroheptameric complex has a half-life of ≈26 hours (Lowe and Kreis, 1996). Under defined in vitro conditions coatomer subcomplexes can be isolated and a-, b’-, and e-COP remain associated and bind Golgi membranes; the in- teraction between g- and z-COP is somewhat weaker, while the interaction between b- and d-COP is the weakest, although b/d-COP can undergo ARF1-dependent binding to Golgi membranes (Lowe and Kreis, 1995; Faulstich et al., 1996; Pavel et al., 1998). Two-hybrid studies show that the interactions between g- and z-COP and b- and d-COP are mediated by the N-terminal regions of g-COP and b-COP, respectively (Eugster et al., 2000). In vivo pulse-chase and immunoprecipitation studies showed ordered interactions between a-, b’- and d-COP, b- and d-COP, and g-, z-, and d-COP (Lowe and Kreis, 1996); suggesting these come together and share a d-COP subunit, after which e- COP is recruited last and membrane binding can occur (Lowe and Kreis, 1996). In support of this, studies in yeast have found that a-COP is necessary for incorporation of e-COP into the coatomer complex (Eugster et al., 2000), and dimers of a- and e-COP have been detected (Faulstich et al., 1996; Pavel et al., 1998). Furthermore, while e-COP does not seem to be required for coatomer function in yeast, it is 17 1.2. The secretory pathway necessary to stabilize the coatomer complex, possibly by preventing degradation of a-COP (Duden et al., 1998). Studies in mammalian systems have found that e- COP is essential, but it is not yet known whether this role is through its interaction with a-COP (Guo et al., 1994, 1996). The question which remains is how the subcomplexes of a/b’/d-COP, b/d-COP, and γ/ζ/δ-COP interact with each other to enable assembly of heteroheptameric coatomer. a- and b’-COP can interact with the carboxy terminus of b-COP, and b’-COP can also interact g-COP (Eugster et al., 2000). Furthermore, the appendage domain of g-COP can interact with a/b’/e- COP, possibly through the interaction of g-COPs C-terminus with e-COP (Eugster et al., 2000; Watson et al., 2004). More research is necessary to fully elucidate the mechanics of coatomer assembly. Coatomer cannot independently bind to Golgi membranes and requires ARF- GTP for membrane recruitment, a process which is saturable (Donaldson et al., 1992a,b; Helms and Rothman, 1992; Palmer et al., 1993). Site-directed photocrosslink- ing showed a GTP-dependent interaction of ARF with b-COP (Zhao et al., 1997), and later at the interface of the b- and g-subunits through the ARF1-GTP switch I region (Zhao et al., 1999; Sun et al., 2007). Later, two-hybrid studies showed an interaction between e-COP and ARF-GTP (Eugster et al., 2000), supporting the hypothesis that e-COP acts as a keystone for coatomer assembly and recruitment (Guo et al., 1994, 1996). Finally, photocrosslinking showed an interaction between ARF and the d-subunit of coatomer; however, this was not GTP-dependent and therefore may not be essential for vesicle formation (Sun et al., 2007). Cargo receptors and TMP21 While coatomer, ARF, and GTP are sufficient to produce vesicle budding in Golgi membrane preparations (Orci et al., 1993b; Ostermann et al., 1993), a third com- 18 1.2. The secretory pathway ponent is required to reproduce budding in a chemically defined in vitro system. Artificial lipid bilayers with mammalian lipid compositions require a dilysine KKXX motif, from either membrane cargo proteins or the C-terminal cytoplasmic tails of the p24 family of cargo receptors, to produce vesicles (Bremser et al., 1999). The KKXX motif functions as an ER-retention signal, allowing ER proteins which ma- ture through the Golgi to be recaptured and directed to the ER (Nilsson et al., 1989; Jackson et al., 1990). Coatomer was found to directly interact with this mo- tif through KKXX affinity chromatography, which isolated the a/b’/e-subcomplex (Cosson and Letourneur, 1994; Lowe and Kreis, 1995). Supporting this, yeast mu- tants lacking functional a- or b’-COP, but not g-COP, could no longer bind KKXX motifs in vitro (Letourneur et al., 1994). Members of the p24 family, including p23 (TMP21), can also bind coatomer through the KKXX motif (Stamnes et al., 1995; Sohn et al., 1996), as well as through a C-terminal diphenylalanine motif, which can bind to the b-, g-, and z-COP subunits according to GST-fusion assays (Fiedler et al., 1996). Mutation of the diphenylalanine motif impeded anterograde transport of a CD8-p24 fusion protein, suggesting a mechanism through which vesicles can carry both retrograde and anterograde cargo (Fiedler et al., 1996). In support of this, the diphenylalanine motif was shown to bind Sec23 on Coat protein complex II (COPII) (Dominguez et al., 1998). In contrast to affinity studies, a photolinking approach suggested both dilysine and diphenylalanine motifs bind coatomer through interactions with the g- subunit, and suggests that previous studies involving a-COP interactions were due to the presence of detergent in the coatomer isolates (Harter et al., 1996; Harter and Wieland, 1998). ARF’s low affinity for binding to Golgi membranes in the ARF-GDP state sug- gests that a membrane receptor for ARF-GDP may be the first step in the recruit- 19 1.2. The secretory pathway ment of vesicle coat proteins to the Golgi membrane (Figure 1.2). A direct in- teraction of membrane bound p23, also known as TMP21, with soluble ARF-GDP suggests that TMP21 serves as this receptor, recruiting ARF-GDP to the membrane where GEFs convert ARF-GDP to the higher membrane affinity ARF-GTP, caus- ing a conformational change which leads to the dissociation of TMP21 (Gommel et al., 2001; Majoul et al., 2001). Coatomer can then bind in a bimodal fashion to both ARF-GTP through its b-subunit, and the KKXX motif in the cytoplasmic tail of the now dissociated TMP21 through its g-subunit, enhancing polymeriza- tion of the COPI vesicle coat (Fiedler et al., 1996; Fligge et al., 2000). Members of the p24 family can also form oligomers (Füllekrug et al., 1999; Gommel et al., 1999; Marzioch et al., 1999), and oligomers of TMP21 can induce a conformational change in coatomer which triggers coatomer aggregation in a membrane-free system (Reinhard et al., 1999; Fligge et al., 2000; Weidler et al., 2000), also suggesting that TMP21 contributes to coat protein polymerization. This is further supported by the finding that TMP21 is concentrated approximately 20 fold in COPI vesicles (Sohn et al., 1996). Human TMP21 is an ubiquitously transcribed type I transmembrane protein with a large N-terminal luminal domain and short C-terminal cytoplasmic tail, with higher protein levels in the pancreas and intestines (Blum et al., 1996; Sohn et al., 1996). While some TMP21 is found on the plasma membrane, the majority is located in the Golgi (Sohn et al., 1996; Blum et al., 1996). If TMP21 were a member of the medial- or trans-Golgi, treatment with BFA would cause TMP21 to relocate to the ER; however, this was not shown to occur (Rojo et al., 1997; Blum et al., 1999). This, in addition to an accumulation of TMP21 in the intermediate compartment (IC) at 15◦C, suggests TMP21 is primarily a member of the IC (Rojo et al., 1997; Blum et al., 1999; Gommel et al., 1999). TMP21 was found to cycle through the 20 1.2. The secretory pathway early secretory pathway between the IC and the cis-Golgi face (Nickel et al., 1997; Rojo et al., 1997; Blum et al., 1999; Gommel et al., 1999). There is some debate whether this cycling is mediated by COPI vesicles (Nickel et al., 1997; Gommel et al., 1999) or microtubule-dependent pre-Golgi carriers (Rojo et al., 1997; Blum et al., 1999). While the cytoplasmic tail is required for retrieval back to the ER, the luminal domain seems to be responsible for its steady-state localization in the Golgi (Nickel et al., 1997). 1.2.3 TMP21 in Golgi structure TMP21 accounts for approximately 30% of integral membrane proteins in the cis- Golgi network (Rojo et al., 1997). This suggests that in addition to its role in transport, TMP21 may play a role in maintaining Golgi structure. When TMP21 was overexpressed it was shown to accumulate in the smooth ER, rather than the cis-Golgi, where it expanded ER membranes to produce regular, flat morphology similar to TMP21-rich Golgi cisternae in structure (Rojo et al., 2000). Exogenous TMP21 expression also caused a shift in endogenous TMP21 distribution from the cis-Golgi to the ER, with a corresponding reduction in Golgi stack size compared to controls; proteins which codistribute with TMP21, such as KDEL receptor ERD2, did not show this distribution shift, suggesting the effect is specific to TMP21 (Rojo et al., 2000). Similarly, overexpression of a GFP-TMP21 fusion protein caused accumulation of large membraneous perinuclear structures and altered localization of IC-cis-Golgi marker ERGIC-53, trans-Golgi marker g-adaptin, and cis/medial- Golgi marker mannosidase II, effectively destroying Golgi structure (Blum et al., 1999). A fusion protein of GFP with the cytoplasmic tail signal of TMP21 did not have this effect, indicating that overexpression of GFP and ER-directed localization were not responsible, and that the effect was specific to TMP21 overexpression 21 1.2. The secretory pathway (Blum et al., 1999). Interestingly, neither anterograde nor retrograde transport was altered following TMP21 overexpression (Rojo et al., 2000). Similar to overexpression, underexpression of TMP21 can also interfere with Golgi structure. Although knockout of all eight members of the p24 family result in viable yeast (Springer et al., 2000), homozygous TMP21 knockout mice are embry- onic lethal prior to the blastocyst stage, suggesting an essential role for TMP21 in early mammalian development (Denzel et al., 2000). Reduction of TMP21 steady- state levels can also disrupt the Golgi, as seen in mice heterozygous for TMP21 which have dilated Golgi cisternae and increased vacuoles surrounding the Golgi in liver and kidney tissues (Denzel et al., 2000). Overall, these data suggest that TMP21 plays a prominent role in the maintenance of Golgi structure and that this maintenance is dependent on an optimal level of TMP21 expression. Several other proteins play a role in the structural organization of the Golgi. While treatment with BFA disrupts Golgi transport and results in Golgi resident proteins being shuttled to the ER (Lippincott-Schwartz et al., 1989; Doms et al., 1989; Lippincott-Schwartz et al., 1990; Strous et al., 1991), certain Golgi matrix proteins, including general vesicular transport factor p115 (p115), Golgi re-assembly stacking protein p65 (GRASP65), golgin subfamily A member 2 (GM130), and giantin, remain in a Golgi-like remnant structure (Hendricks et al., 1992; Hidalgo et al., 1992; Lemos-Chiarandini et al., 1992; Nakamura et al., 1995; Seemann et al., 2000). Analysis of this remnant structure reveals a ribbon-like reticulum in the correct cellular location; however, the stacked cisternae are not well-defined on an electron micrograph level (Seemann et al., 2000). Thus the Golgi must also depend on interactions with integral membrane proteins for structural maintenance, and TMP21 may fill this role. 22 1.2. The secretory pathway 1.2.4 TMP21 in vesicle tethering Docking of COPI vesicles to specific receptor membranes is mediated by soluble NSF attachment protein receptors (SNAREs) (Söllner et al., 1993). SNAREs are categorized by their location; v-SNAREs are incorporated into vesicles, while tar- get (t)-SNAREs are localized on the recipient membrane. Preceding docking and SNARE assembly, vesicles must become tethered to the potential target membrane (Shorter et al., 2002). Within the Golgi this is accomplished by members of the golgin protein family, including p115, GM130, and giantin. GM130 is a peripheral cytoplasmic protein primarily bound to cis-Golgi mem- branes at the C-terminus (Nakamura et al., 1995). p115, also known as transcytotic vesicle-associated protein (TAP) (Barroso et al., 1995), is a 115-kD peripheral Golgi membrane protein consisting of two polypeptides, each having a globular head and a coiled-coil tail. p115 is required for intra-Golgi transport (Sztul et al., 1991; Waters et al., 1992; Sztul et al., 1993) and stacking of Golgi cisternae (Shorter and War- ren, 1999). p115 can also bind to GM130 on the target membrane (Levitan et al., 1996; Nakamura et al., 1997), as well as to giantin, a 350-kD Golgi membrane protein with a large cytoplasmic domain, found on COPI vesicles (Linstedt and Hauri, 1993; Sönnichsen et al., 1998). Therefore, p115 serves as a bridge between COPI vesicles and Golgi membranes, tethering vesicles to Golgi cisternae. GM130 can also bind to membrane-associated GRASP65, and it is suggested that the GRASP65-GM130- p115 complex can act in both cisternal stacking and vesicle tethering (Barr et al., 1997, 1998). In turn, GRASP65 was shown to bind to members of the p24 family of cargo receptors, including p24a (Barr et al., 2001). The p24 family forms hetero- oligomeric complexes with each other which include p24a and TMP21 (Füllekrug et al., 1999), suggesting a mechanism by which changes in steady-state TMP21 lev- els could reverberate through a series of interactions to disrupt Golgi structure and 23 1.3. TMP21 in Alzheimer’s disease vesicular trafficking. 1.3 TMP21 in Alzheimer’s disease 1.3.1 TMP21 as a component of the g-secretase complex The vast majority of research on TMP21 has pertained to its role in protein traffick- ing; however, mass spectrometry has recently identified TMP21 as a new member of the PS complex (Chen et al., 2006). TMP21 was found to selectively regulate cleavage at the g-site without affecting e-site cleavage; siRNA knock down of TMP21 caused an increase in Ab-40 and Ab-42 but no change in Notch cleavage. Further- more, no changes were seen in the protein levels of PS1, Nct, Aph-1, or Pen-2. These findings suggest that a deficiency in TMP21 may exacerbate AD pathology. TMP21 was identified as a regulator of the g-secretase complex following im- munoprecipitation from cells overexpressing PS1 and PS2 using a PS1 N-terminal polyclonal antibody (Chen et al., 2006). TMP21 immunoprecipitation could also successfully pull down all four members of the g-secretase complex (Chen et al., 2006). Previous immunoprecipitation work in cells expressing multiple tagged g- secretase subunits failed to note this interaction, possibly due to nonessential com- plex components being titrated out when essential subunits are highly expressed (Fraering et al., 2004). Later groups also failed to confirm this interaction when isolating the g-secretase complex using g-secretase active-site inhibitor Merck C (Vetrivel et al., 2007; Winkler et al., 2009). Furthermore, chemical cross-linking was unable to find an interaction between g-secretase and TMP21 (Winkler et al., 2009). It is important to note that these findings do not exclude the possibility that TMP21 associates with the g-secretase complex in a transient manner, however; in- deed, TMP21’s role as a negative regulator of g-secretase activity implies a minimal 24 1.3. TMP21 in Alzheimer’s disease presence in the active g-secretase complexes which would be isolated by Merck C. Later studies have shown that TMP21 is associated with the g-secretase complex in rat brain (Teranishi et al., 2009). 1.3.2 TMP21 and APP processing The conclusion that TMP21 functions as a regulator of g-secretase activity is fur- ther supported by TMP21 suppression in human embryonic kidney (HEK)-293 and human neuroblastoma (SH-SY5Y) cells, where no change was found in N’O’- glycosylated holo-APP protein levels (Chen et al., 2006). While TMP21 suppression does not increase APP synthesis in mouse neuroblastoma (N2a) and human cervi- cal (HeLa) cells, it does increase the stability of both immature and mature APP, resulting in a 2-fold increase in cell-surface APP (Vetrivel et al., 2007). The au- thors suggest that TMP21 suppression compromises bidirectional transport in the ER/Golgi, resulting in more mature APP which can undergo amyloidogenic cleavage in endocytic compartments (Vetrivel et al., 2007); however, the 2-fold increase in both cell-surface and sAPP suggests that the majority of the additional cell-surface APP was processed through the non-amyloidogenic pathway. Furthermore, suppres- sion of TMP21 in cell-free assays can increase Ab production, suggesting TMP21’s modulatory role is independent of trafficking, and that TMP21 may directly modu- late g-secretase activity (Chen et al., 2006) (Figure 1.3). It is possible that TMP21’s role in trafficking and in the direct modulation of g-secretase activity may not be mutually exclusive, however. Up to 30% of PS1 was found to be located between the ER and Golgi (pre-Golgi) in COPI-coated mem- branes (Réchards et al., 2003). Mutation of the first Asp residue (D257A) can reduce COPI-associated PS1 to 15%, while increasing plasma membrane PS1 from 16 to 36% (Réchards et al., 2003); similar findings were seen upon mutation of the second 25 1.3. TMP21 in Alzheimer’s disease Figure 1.3: Suggested model for the dual role of TMP21 in APP pro- cessing and trafficking. Knockdown of TMP21 increases Ab levels (Chen et al., 2006). While it was initially suggested that this was due to the selective modulation of g-secretase activity by TMP21, it is likely that TMP21’s role in trafficking is also involved. Knockdown of TMP21 was shown to increase nascent APP stability and cause accumulation of mature APP on the cell surface, suggesting that TMP21 may normally inhibit APP trafficking to the plasma membrane (PM) (Vetrivel et al., 2007). Permutations in the distribution of PS1 to COPI vesicles can also cause PS1 to accumulate on the plasma membrane (Réchards et al., 2003). While it has been shown that TMP21 interacts with PS1 (Pardossi-Piquard et al., 2009) and it is known that TMP21 is vital for COPI vesicle formation, it remains to be seen whether TMP21 knockdown can reduce PM PS1 levels. Taken together, these find- ings suggest that knockdown of TMP21 increases cell surface APP and perhaps PS1, while also increasing g-secretase activity, with a net result of increased Ab accumulation. (Figure adapted from Vetrivel et al. (2007)). 26 1.3. TMP21 in Alzheimer’s disease Asp residue (Kaether et al., 2002). This suggests that COPI retention could medi- ate the downstream localization of PS1. The multi-protein high molecular weight complex which may contain regulators of g-secretase activity is not formed in cells with the D257A mutation (Yu et al., 2000a). TMP21 is part of the high molecular weight g-secretase complex, and could therefore be the component necessary for the pre-Golgi retention of g-secretase. Reduction of TMP21 may therefore result in higher cell-surface g-secretase through TMP21’s role in trafficking, while an as of yet undetermined mechanism may function to selectively regulate g-site, but not e-site, cleavage (Figure 1.3). Although the original study of TMP21 as a regulator of g-secretase activity showed that overexpression of TMP21 had no discernible effect on protein levels of PS1, Nct, Aph-1, or Pen-2, or the production of Ab (Chen et al., 2006), studies since then have disagreed. The ability of TMP21 to regulate g-secretase cleavage can be further evaluated using a truncated form of APP (APPe) which lacks the AICD do- main and terminates at Ab-49. APPe can be cleaved by a-, b-, and g-secretase, but lacks the e-cleavage site, allowing the examination of g-secretase activity without confounding e-site cleavage (Lefranc-Jullien et al., 2006). When TMP21 is over- expressed with APPe, but not wtAPP, Ab species are produced which are similar to those made when cells are treated with g-secretase inhibitor Compound E, sug- gesting that membrane embedded Ab1-49 may be a target for TMP21-mediated g-secretase inhibition (Pardossi-Piquard et al., 2009). Recent findings in our lab also suggest that overexpression of TMP21 can affect APP processing (see chapter 3). Recently, knockdown of another member of the p24 protein family, p24a2, was also found to increase Ab production while not af- fecting AICD generation, while overexpression suppressed Ab production (Hasegawa et al., 2010). For TMP21, it appears that the transmembrane domain is required 27 1.3. TMP21 in Alzheimer’s disease for association of TMP21 with mature Nct and PS1-NTF, as well as for g-secretase inhibition (Pardossi-Piquard et al., 2009). Additionally, both the conserved dily- sine ER-retrieval motifs of p24a2 and TMP21 seem to be necessary for g-secretase inhibition (Hasegawa et al., 2010). 1.3.3 Clinical relevance of TMP21 in Alzheimer’s disease In conclusion, while the biochemical mechanism of TMP21’s regulation of APP processing is still being deduced, some clinically relevant data is available. The temporal and spatial distribution of TMP21 in brain has recently been described (Vetrivel et al., 2008a). While TMP21 is expressed throughout the brain, it shows highest expression in the hippocampus, an area of the brain essential for learning and memory and which shows AD pathology at the mid-stage of the disease (Braak and Braak, 1991; Braak et al., 1993). Interestingly, TMP21 levels are highest during embryonic development and decline with age. In humans, individuals with AD show lower total levels of TMP21 in the frontal cortex and hippocampus, but not in the cerebellum, which is less susceptible to AD changes (Vetrivel et al., 2008a). Why is TMP21 lower in AD brain? Our lab has recently characterized the pro- moter of human TMP21 and found that it contains an NFAT1 response element (Liu et al., 2011). Interestingly, NFAT1 activation is briefly increased early in cognitive decline and falls to below normal levels as the disease progresses (Abdul et al., 2009). Further work is necessary to determine whether TMP21 also shows a brief surge in expression early in AD as part of a possible compensatory protective mechanism. These findings suggest that TMP21 levels are significantly altered in AD and further suggest that TMP21 may play an integral, clinically relevant role in AD pathogenesis. 28 1.4. Overall goals of this research 1.4 Overall goals of this research The overall goals of this thesis are to further characterize TMP21 biochemically through an investigation of its degradation, to expand our understanding of TMP21’s role in AD pathogenesis, and to understand how TMP21 contributes to mouse be- haviour as a guideline for future work in a TMP21-heterozygous × AD model mouse cross. 1.4.1 Examine the biochemical properties of TMP21 Several proteins involved in AD pathology are degraded by the ubiquitin-proteasome pathway, including tau (David et al., 2002; Keck et al., 2003), the APP C-terminus (Nunan et al., 2001) and BACE1 (Qing et al., 2004; Zhou et al., 2004). All four critical members of the g-secretase complex are also degraded in this pathway, in- cluding PS1 (Fraser et al., 1998; Steiner et al., 1998; Honda et al., 1999), PS2 (Kim et al., 1997), Nct-1 (He et al., 2007), Pen-2 (Bergman et al., 2004; Crystal et al., 2004), and Aph-1 (He et al., 2006). This suggests that the g-secretase complex may be degraded as a whole rather than as its component parts, a perspective which has yet to be explored in the context of AD pathogenesis. TMP21 has been identified as the newest member of the g-secretase complex, yet thus far no studies have ex- amined how TMP21 undergoes degradation. This thesis examines whether TMP21 is degraded in a similar manner as other members of the g-secretase complex. 1.4.2 Investigate the effect of TMP21 on APP processing The majority of previous research has focused on how TMP21 suppression or over- expression affects Ab production (Chen et al., 2006; Vetrivel et al., 2007; Dolcini et al., 2008; Hasegawa et al., 2010), while the effect of TMP21 on APP CTF forma- tion has been largely overlooked. Those studies which did examine APP CTFs did 29 1.4. Overall goals of this research not examine C99, and thus could not determine whether TMP21’s effect on APP processing was solely due to its influence on g-secretase, or whether b-secretase was also involved. Thus, this thesis investigates the effect of TMP21 on APP processing, with an emphasis on APP CTF formation and BACE1. 1.4.3 Understand the behavioural consequences of TMP21 suppression in C57BL/6 mice While several studies have utilized biochemical techniques to study the function of TMP21 in trafficking and Golgi structure (see section 1.2), the behavioural role of TMP21 has yet to be examined. Without an understanding of how TMP21 suppression affects mouse behaviour in general, we cannot properly interpret results from a cross with AD model mice in our future research. Thus this thesis will examine the behavioural consequences of TMP21 suppression in the C57BL/6 mouse background. 30 Chapter 2 TMP21 degradation is mediated by the ubiquitin-proteasome pathway 2.1 Introduction Thus far no studies have examined how TMP21 undergoes degradation. While APP was being characterized, several groups showed that the ubiquitin-proteasome path- way was involved in AD pathogenesis (Mori et al., 1987; Perry et al., 1987; Cole and Timiras, 1987; Shaw and Chau, 1988; Manetto et al., 1988). The ubiquitin- proteasome pathway is fundamental to many cellular processes, including signal transduction, development, apoptosis, and antigen processing, as well as the degra- dation of cytosolic, nuclear, and membrane proteins (Hershko and Ciechanover, 1998, for review). It also plays an important role in the degradation of misfolded proteins, a hallmark of many neurodegenerative disorders (Hershko and Ciechanover, 1998, for review). Several proteins involved in AD pathology are degraded by the ubiquitin-proteasome pathway, including tau (David et al., 2002; Keck et al., 2003), the APP C-terminus (Nunan et al., 2001), BACE1 (Qing et al., 2004; Zhou et al., 2004), PS1 (Fraser et al., 1998; Steiner et al., 1998; Honda et al., 1999), PS2 (Kim et al., 1997), Nct-1 (He et al., 2007), Pen-2 (Bergman et al., 2004; Crystal et al., 2004), and Aph-1 (He et al., 2006). Here we show that degradation of TMP21, the newest member of the g-secretase complex, is also mediated by the ubiquitin-proteasome pathway. 31 2.2. Methods 2.2 Methods 2.2.1 Materials Dulbecco’s modified eagles’ medium (DMEM) was purchased from Invitrogen. Pro- teasomal inhibitors b-lactone, N-carbobenzoxyl-L-leucinyl-L-leucinyl-L-leucinal (MG-132), N-acetyl-L-leucinyl-L-leucinyl-L-methional (ALLM), and N-acetyl-L-leucinyl-L-leucinyl- L -norleucinal (ALLN), as well as lysosomal inhibitors NH4Cl and chloroquine, were obtained from Calbiochem. Cycloheximide (CHX) and b-actin antibody AC-15 were obtained from Sigma. Odyssey R© blocking buffer (OBB), IRDyeTM680-labeled goat anti-rabbit, and IRDye TM 800CW-labeled goat anti-mouse antibodies were obtained from LI-COR Biosciences. 2.2.2 cDNA constructs, cell cultures and transfection Human TMP21 cDNA was amplified by polymerase chain reaction (PCR) using for- ward primer 5′–cg ggatcc gccacc atg tctggtttgt ctggcccac–3′ and reverse primer 5′–g gaattc ctcaatcaatttcttggccttg–3′ and cloned into expression vector pcDNA4-mycHisA (Invitrogen) at the BamHI and EcoRI sites to create mammalian expression plasmid pTMP21-mycHis, which expresses a fusion protein of TMP21 with a C-terminal myc epitope followed by a 6xHis tag. N2a and HEK cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 1% sodium pyruvate, 1% L-glutamine, and 1% penicillin/streptomycin (Invitrogen). All cells were maintained at 37◦C in an incubator containing 5% CO2. Transient trans- fections were performed using the calcium phosphate method. To generate the stable cell line HEK TMP21-mycHis (HTM2), HEK cells were transfected with pTMP21- mycHis using Lipofectamine 2000 (Invitrogen) as per the manufacturer’s protocol. Transfected cells were selected using 200 mg/mL zeocin and clones were selected 32 2.2. Methods for several generations. Stably transfected colonies were confirmed by immunoblot using 9E10 antibody and maintained in 60 mg/mL zeocin. Stable cell line N2a TMP21-mycHis (NTM1) was similarly created from N2a cells. 2.2.3 Antibody generation and characterization TMP21 antibodies T21 and T51 were generated by inoculating two rabbits with synthetic peptide hkdllvtgayeihk or ymkkreeemrdtnestntrvym correspond- ing to amino acids 49 to 60 and 167 to 186 of TMP21, respectively. Both peptides share 100% sequence homology with mouse and human TMP21. To determine an- tibody specificity, stable cell lysates were analysed by western blot with preimmune sera and following preabsorbtion. For preabsorbtion, equal concentrations of anti- body and peptide were diluted into phosphate buffered saline (PBS) and precleared with CL-4B for 1 h at 4◦C. Samples were centrifuged at 14,000×g for 10 minutes at 4◦C and the supernatent was used for protein detection at a 1:500 concentration. 2.2.4 Subcellular fractionation HEK cells were transfected with pTMP21-mycHis or negative control by the calcium- phosphate method and harvested 72 h later in PBS. Samples were centrifuged at 1000×g for 2 min to pellet the cells, which were then resuspended in 0.4mL homogenization buffer (HB) containing 5 mM HEPES pH 7.4, 125 mM EDTA, 3 mM imidazole, 0.25 M sucrose, and Roche protease inhibitor Complete. Cells were homogenized with 15 strokes of a Kontes Dounce homogenizer and centrifuged at 1000×g for 15 min to produce a post-nuclear supernatant (PNS), which was layered on top of a discontinuous sucrose gradient consisting of 0.8 mL 2 M, 1.2 mL 1.3 M, 1.2 mL 1.0 M, and 0.8 mL 0.5 M sucrose in HB. Tubes were balanced with the addition of HB to ±0.1 g and spun at 280,000×g for 2 h at 4◦C. Fractions were 33 2.2. Methods manually collected from the top of the tube (3, 0.4 mL and 16, 0.2 mL) and frozen at -80◦C until analysis by immunoblotting on 16% tris-tricine gels. 2.2.5 Pharmacological treatment To study lysosomal involvement, stable cells were treated with 100 mM chloroquine or 50 mM NH4Cl for 24 hours. To test for a dosage effect, stable cells were treated with 0, 10, 25, 50, or 100 mM chloroquine or 0, 5, 10, 25, or 50 mM NH4Cl for 24 hours. Proteasomal involvement was tested using 25 mM ALLM, 25 mM ALLM, 5 mM MG-132, or 10 mM b-lactone for 24 hours. Time dependence was examined by treating with 10 mM b-lactone for 0, 2, 4, 8, or 12 hours, while dose dependence was examined by treating with 0, 0.5, 1, 2.5, 5, or 10 mM b-lactone for 24 hours. Protein half life was determined using a 100 mg/mL CHX chase for 0, 0.5, 1, 2, 3, 4, 6, 8, 12, 16, 20, or 24 hours as previously described (Schoenfeld et al., 2000; Touitou et al., 2001; Liu et al., 2004). Cells were lysed with radio-immunoprecipitation assay – deoxycholate (RIPA-DOC) buffer containing 50 mM TrisHCl (pH 7.2), 150 mM NaCl, 1% deoxycholate, 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), and protease inhibitor cocktail Complete (Roche). TMP21 protein levels were analysed by immunoblot assay. 2.2.6 Immunoprecipitation and immunoblotting HTM2 cells were transfected with ubiquitin plasmid pHisUbi plasmid (Qing et al., 2004), lysed 48 h later in NP-40 buffer (10 mM Hepes, pH 7.5, 142.4 mM KCl, 5 mM MgCl2, 1 mM EDTA, 1% NP-40, and Roche protease inhibitor cocktail Complete), and sonicated. Cell debris was removed by centrifugation at 14,000×g at 4◦C for 20 min, and the supernatent was cleared with CL-4B for 1 hour followed by centrifuga- tion at 14,000×g at 4◦C for 10 min. The supernatent was incubated with primary 34 2.2. Methods capturing antibody at 4◦C for 1 h and then with protein A/G sepharose beads (Santa Cruz) overnight. The following day the immunoprecipitates were washed three times with NP-40 buffer and one time with PBS on ice. To elute captured proteins, 2× sample loading buffer was added to the pellets and samples were boiled for 5 min. For western blot analysis samples were loaded onto 16% Tris-Tricine gels and transferred to Immobilon R©-FL polyvinylidene difluoride (PVDF-FL) membranes. Membranes were blocked for 1 h in 1:1 OBB:PBS and incubated with primary antibodies 9E10, T21, and/or AC-15 in 1:1 OBB:PBS with 0.1% Tween-20 at 4◦C overnight. The next day membranes were rinsed in PBS with 0.1% Tween-20 (PBS-T), incubated with IRDye TM 680-labeled goat anti-rabbit and IRDye TM 800CW-labeled goat anti-mouse antibodies in 1:1 OBB:PBS with 0.1% Tween-20 and 0.01% SDS at room temper- ature for 1 h, further rinsed in PBS-T, and visualized on the LI-COR Odyssey R© system. 2.2.7 In-cell western assay HTM2 cells were seeded onto 24 well plates and treated the following day with 100 mg/mL CHX for 0, 0.5, 1, 2, 4, 6, 8, 12, or 24 hours. Following each time point cells were fixed in 4% paraformaldehyde (PFA) for 20 min at room temperature and placed at 4◦C until all time points were obtained. Cells were permeabolized by rinsing 5 times in PBS with 0.1% Triton X-100 (PBS-Tx) for 5 min, blocked with 1:1 OBB:PBS for 1.5 h, and incubated in primary antibody 9E10 overnight at 4◦C. The next day cells were rinsed in PBS-T and incubated with IRDye TM 800CW-labeled goat anti-mouse secondary antibody in 1:1 OBB:PBS with 0.5% Tween-20 for 1 h, rinsed in PBS-Tx, and scanned on the LI-COR Odyssey R© system. Background plate fluoresence was controlled for by omitting the primary antibody. 35 2.3. Results 2.2.8 Immunocytochemistry HTM2 cells were seeded onto glass coverslips in 24 well plates. The following day cells were rinsed in PBS, fixed in 4% PFA for 20 min, permeabolized with PBS-Tx for 30 min, blocked in 5% bovine serum albumin (BSA) in PBS-Tx for 30 min, and incubated with primary antibodies in 1% BSA in PBS-Tx overnight at 4◦C. The next day cells were rinsed, incubated for 1 h with goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red), and mounted using Fluoromount-G (Southern Biotech). Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. 2.3 Results 2.3.1 Antibody characterization and stable cell line creation Incubation of HTM2 stable cell lysates with pre-immune sera did not result in TMP21 detection. Antibody preabsorption with inoculating peptide also prevented detection of both exogenous and endogenous TMP21. Both the T21 and T51 anti- bodies showed robust, specific detection of TMP21 and TMP21-myc (Figure 2.1A,B). Dual-colour western blots comparing N2a and NTM1 cell lysates, as well as HEK and HTM2 cell lysates, showed that both stable cell lines have robust expression of exogenous TMP21. Furthermore, the TMP21-myc band can be simultaneously detected by both anti-myc antibody 9E10, N-terminal TMP21 antibody T21, and C-terminal antibody T51 (Figure 2.1C,D), indicating that the myc tag may have little influence on protein structure. Immunofluorescent and sucrose gradient local- ization studies also showed that exogenous TMP21-myc has a similar subcellular expression profile as its endogenous counterpart (Figure 2.2). 36 2.3. Results Figure 2.1: Antibody and stable cell line characterization. Equal amounts of total protein from HEK and pTMP21-mycHis stably transfected HTM2 cell lysates were used to determine antibody specificity. Pre-immune sera were diluted 1:500. Antibody pre-absorption was performed by diluting equal volumes of antibody and immunizing peptide in PBS and clearing with CL-4B for 1 h at 4◦C; the supernatant was then used for detection. TMP21 was detected by both TMP21 N-terminal antibody T21 (A) and C-terminal antibody T51 (B). (C) NTM1 and (D) HTM2 stable cell lysates were examined by dual-colour western blots on 16% tris-tricine gels. Exogenous TMP21-myc expression was detected using anti-myc antibody 9E10 or endogenous TMP21 antibodies T21 or T51. HTM2 and NTM1 cells showed robust TMP21-myc expression, which could also be recognized by the endogenous antibody. 37 2.3. Results Figure 2.2: TMP21-myc displays correct subcellular localization. HTM2 cell homogenate was subjected to sucrose gradient fractionation to isolate the ER and Golgi. TMP21 colocalized with Golgi marker Golph4 (A) and ER marker Grp78 (B), similar to the localization profile of endogenous TMP21. (C) HTM2 cells were grown on glass coverslips in 24 well plates and analysed 24 h later by immunofluorescent labeling with either rabbit anti-Golph4 (Golgi marker) or rabbit anti-KDEL (ER marker) and mouse anti-myc (TMP21-myc) antibodies, which were detected using goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red). The merged image clearly shows that TMP21-myc colocalizes with both the ER and Golgi, as predicted from the literature. 38 2.3. Results 2.3.2 TMP21 has a half-life of approximately 3 hours CHX is a protein synthesis inhibitor produced by the Streptomyces griseus bac- terium (Kerridge, 1958; Kay and Korner, 1966). Protein half-life can be determined by halting protein synthesis with 100 mg/mL CHX and measuring the amount of protein remaining at various time points (Schoenfeld et al., 2000; Touitou et al., 2001; Liu et al., 2004). Using this method we found that TMP21 has a half-life of approximately 3 to 4 h (Figure 2.3A and B). To confirm this finding in situ we performed a quantitative in-cell western assay as previously described (Arredondo et al., 2006). The resulting half-life agreed with the traditional western blot data (Figure 2.3C). 2.3.3 Lysosomal inhibition does not affect TMP21 protein levels Chloroquine inhibits lysosomal protein degradation by raising lysosomal pH and interfering with receptor recycling (Ohkuma and Poole, 1978; Gonzalez-Noriega et al., 1980), while NH4Cl inhibits phagosome-lysosome fusion (Gordon et al., 1980; Amenta and Brocher, 1980). To examine whether the degradation of TMP21 is mediated by the lysosome, NTM1 and HTM2 cells were treated with varying doses of lysosomal inhibitors NH4Cl or chloroquine. TMP21 protein levels did not show a cumulative rise with increasing dosage of lysosomal inhibitors in either cell line (Figure 2.4A-D), suggesting that the lysosome is not involved in TMP21 degrada- tion. Furthermore, immunofluorescent labelling showed no colocalization of TMP21 with lysosomes in either cell line (Figure 2.4E). 2.3.4 Proteasomal inhibition increases TMP21 protein levels ALLN and MG-132 are peptide aldehydes which inhibit proteasome-dependent pro- tein degradation, while ALLM is a much weaker proteasome inhibitor which does 39 2.3. Results Figure 2.3: TMP21 has a half-life of approximately 3 hours. (A) HTM2 cells were treated with 100 mg/mL CHX for 0, 0.5, 1, 2, 3, 4, 6, 8, 12, 16, 20, or 24 hours for a total of 32 plates. Cells were lysed in equal volumes of RIPA-DOC and protein level was plotted as a percentage of TMP21 remaining compared to the 0 hour amount. (B) Quantification of the CHX assays. Values represent mean ±sem. (C) In-cell western analysis. HTM2 cells were seeded onto 24 well plates and treated the following day with 100 mg/mL CHX for 0, 0.5, 1, 2, 4, 6, 8, 12, or 24 hours. Following each time point cells were fixed and TMP21-myc was detected using 9E10 and visualized with IRDye TM 800CW-labeled goat anti-mouse secondary antibody using the LI-COR Odyssey R© system to determine the in situ half life of TMP21. Background plate fluorescence was controlled for by omitting the primary antibody (Ab). 40 2.3. Results Figure 2.4: Lysosomal inhibition does not affect TMP21 protein levels. HTM2 cells were treated with varying doses of NH4Cl (A) or chloroquine (B) for 24 h and analysed by immunoblot. No dose-dependent increase in exogenous or endogenous TMP21 levels was seen, indicating that the lysosome is not involved in TMP21 degradation. (C and D) NTM1 cells were treated as per A and B above, with similar results. (E) HTM2 and NTM1 cells were grown on glass coverslips in 24 well plates and analysed 24 h later by immunofluorescent labeling with rabbit anti-Lamp2 (lysosome marker) and mouse anti-myc (TMP21-myc) antibodies, which were detected using goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red). The merged image clearly shows that TMP21 does not colocalize with lysosomes. 41 2.3. Results not show inhibition at 25 mM (Rock et al., 1994; Jensen et al., 1995; Ward et al., 1995). b-lactone, a Streptomyces microbial metabolite, is an irreversible inhibitor of the 20S and 26S proteasome (Omura et al., 1991; Ward et al., 1995; Fenteany et al., 1995). HTM2 and NTM1 cells treated with 25 mM ALLN, 5 mMMG-132, or 10 mM b-lactone for 24 hours showed an increase in TMP21 protein levels, while ALLM treatment did not show this effect, suggesting the proteasome is involved in TMP21 degradation (Figure 2.5E, F). Moreover, TMP21 protein levels increased in both a dose- (Figure 2.5A, B) and time-dependent (Figure 2.5C, D) manner in both HTM2 and NTM1 stable cells. 2.3.5 Proteasomal inhibition causes a time-dependent accumulation of TMP21 in the Golgi TMP21 cycles through the early secretory pathway between the intermediate and cis-Golgi compartments (Nickel et al., 1997; Blum et al., 1999; Gommel et al., 1999). We used immunofluorescent labeling to determine whether proteasomal inhibition causes a time-dependent accumulation of TMP21 in the Golgi. HTM2 cells were seeded onto 24 well plates and treated the next day with 5 mM MG-132 for 0, 8, or 12 hours. Antibodies Golph4 and 9E10 were used to detect the Golgi and TMP21, respectively. Exposure times and post-experimental level adjustments were performed equally on all slides. TMP21 showed a time-dependent accumulation in the Golgi with increased exposure time to MG-132 (Figure 2.6), indicating that proteasomal inhibition does not interfere with intracellular localization of TMP21 and that TMP21 is degraded by the proteasome. 42 2.3. Results Figure 2.5: Proteasomal inhibition increases TMP21 protein levels. (A) HTM2 and (B) NTM1 cells were treated with various doses of b-lactone for 24 h. b-actin was used as control. TMP21 protein levels increased in a dose-dependent manner. (C) HTM2 and (D) NTM1 cells were treated with 10 mM b-lactone for various times. b-actin was used as control. TMP21 protein levels increased in a time-dependent manner. (E) HTM2 and (F) NTM1 cells were treated with con- trol solution, 25 mM ALLM, 25 mM ALLN or 5 mM MG-132 for 24 h and analysed by western blot with T51 and b-actin antibodies. Treatment with the proteaso- mal inhibitors ALLN or MG-132 increased TMP21 protein levels in HTM2 cells and NTM1 cells, whereas non-proteasome-specific ALLM treatment had no effect on TMP21 protein level. (G) Quantification of TMP21 protein levels in HTM2 cells. Values represent mean ±sem (n = 3). **p < 0.01 by one-way ANOVA with Bonferroni’s multiple comparison post-hoc test. 43 2.3. Results Figure 2.6: Proteasomal inhibition causes a time-dependent accumulation of TMP21 in the Golgi. HTM2 cells were seeded onto 24 well plates and treated the next day with 5 mM MG-132 for 0, 8, or 12 hours. Rabbit anti-Golph4 and mouse anti-myc antibodies were detected using goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red). Exposure times and post- experimental level adjustments were performed equally on all slides. TMP21 showed a time-dependent accumulation in the Golgi with increased exposure time to MG-132 (TMP21 high contrast), suggesting that TMP21 degradation is mediated by the proteasome. 44 2.4. Discussion 2.3.6 TMP21 is ubiquitinated As the above data suggest proteasomal involvement in TMP21 degradation, we fur- ther examined whether TMP21 is ubiquitinated. HTM2 cells were transfected with pUbi-mycHis and lysed with NP-40 lysis buffer 48 h later. Immunoprecipitation with anti-ubiquitin antibody could pull down TMP21-ubiquitin conjugates, as could anti- TMP21 antibody immunoprecipitation (Figure 2.7A). Additionally, immunofluores- cent labeling of both HTM2 and NTM1 cells showed colocalization of TMP21 with ubiquitin (Figure 2.7B). These results, taken together, suggest that TMP21 is ubiq- uitinated and that degradation of TMP21 is mediated by the ubiquitin-proteasome pathway. 2.4 Discussion 2.4.1 Summary The ubiquitin-proteasome pathway was previously shown to play a role in AD patho- genesis (Mori et al., 1987; Perry et al., 1987; Cole and Timiras, 1987; Shaw and Chau, 1988; Manetto et al., 1988). Indeed, all four critical components of the g-secretase complex, including PS1 (Fraser et al., 1998; Steiner et al., 1998; Honda et al., 1999), Nct (He et al., 2007), Pen-2 (Bergman et al., 2004; Crystal et al., 2004), and Aph-1 (He et al., 2006), are degraded by this pathway. Here we show that degradation of TMP21, one of the newest members of the g-secretase complex, is also mediated by the ubiquitin-proteasome pathway. Treatment of HTM2 cells with CHX showed that TMP has a half-life of ap- proximately 3 hours. This was further confirmed in situ using an in-cell western assay (Figure 2.3). While lysosomal inhibitors initially seemed to cause an increase in TMP21 in HTM2 cells, NTM1 cells did not show this effect, and both HTM2 45 2.4. Discussion Figure 2.7: TMP21 can be ubiquitinated. (A) HTM2 cells were transfected with pUbi-mycHis and lysed with NP-40 lysis buffer 48 h later. TMP21-ubiquitin conjugates could be detected by performing ubiquitin immunoprecipitation with anti-TMP21 blot or vice versa. (B) HTM2 and NTM1 cells were seeded on glass coverslips in 24-well plates. Immunofluorescent labelling was performed 24 h later with rabbit anti-ubiquitin and mouse anti-myc antibodies, which were detected with goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red). TMP21 and ubiquitin colocalize (white arrowheads). 46 2.4. Discussion and NTM1 cells failed to show a dose-dependent increase in TMP21 levels when treated with lysosomal inhibitors. Furthermore, immunofluorescent labeling showed no colocalization of TMP21 with lysosomes in either cell line (Figure 2.4); thus TMP21 does not undergo lysosomal degradation. To determine whether TMP21 was degraded by the proteasome, HTM2 and NTM1 cells were treated with proteasome inhibitors ALLN, MG-132, and b-lactone for 24 hours, and with non-proteasome specific inhibitor ALLM as a control. In- hibitor treatment increased TMP21 protein levels in both cell lines, an effect which was both time- and dose-dependent (Figure 2.5). This effect was confirmed in situ, where treatment of HTM2 cells with MG-132 caused a time-dependent accumulation of TMP21 in the Golgi (Figure 2.6). To determine whether TMP21 was ubiquiti- nated we conducted coimmunoprecipitation experiments, which showed that ubiq- uitin IP could pull down TMP21, and vice versa. Furthermore, immunofluorescent labeling showed colocalization of TMP21 with ubiquitin (Figure 2.7). 2.4.2 Conclusion The above results suggest that TMP21 degradation is mediated by the ubiquitin- proteasome pathway. As all four critical members of the g-secretase complex are degraded by this pathway, it suggests that the g-secretase complex may be degraded as a whole rather than as its component parts. Further work is needed to determine whether this is the case, and what effect such coordinated degradation would have on AD pathogenesis. 47 Chapter 3 Overexpression of TMP21 alters APP processing 3.1 Introduction TMP21 was first associated with AD when it was linked to the AD3 gene locus on chromosome 14, a locus associated with an aggressive form of AD (Sherrington et al., 1995). It received little attention in the AD field until it was discovered that suppression of TMP21 increased Ab production without affecting Notch cleavage (Chen et al., 2006). TMP21 was found to coimmunoprecipitate with components of the g-secretase complex both in vitro (Chen et al., 2006) and in vivo (Teranishi et al., 2009), although other groups failed to see this interaction (Fraering et al., 2004; Vetrivel et al., 2007; Winkler et al., 2009). While TMP21 suppression could increase Ab production in the test-tube, it was also found to increase the stability of nascent APP in vitro, suggesting TMP21 affected APP processing through its role in both modulation and trafficking (Chen et al., 2006; Vetrivel et al., 2007). The majority of previous research has focused on how TMP21 suppression or overexpression affects Ab production (Chen et al., 2006; Vetrivel et al., 2007; Dolcini et al., 2008; Hasegawa et al., 2010), while the effect of TMP21 on APP CTF forma- tion has been largely overlooked. Those studies which did examine APP CTFs did not examine C99, and thus could not determine whether TMP21’s effect on APP processing was solely due to its influence on g-secretase, or whether b-secretase was also involved. The following work examines the effect of TMP21 on APP processing, 48 3.2. Methods with an emphasis on APP CTF formation and BACE1. 3.2 Methods 3.2.1 Materials DMEM, FBS, zeocin, and Lipofectamine TM 2000 were purchased from Invitrogen. b-actin antibody AC-15 was obtained from Sigma. siRNA reagents were obtained from Dharmacon, and IRDye TM 680-labeled goat anti-rabbit, and IRDye TM 800CW- labeled goat anti-mouse antibodies were obtained from LI-COR Biosciences. 3.2.2 Cell culture and transfections HTM2 cells were created as previously described in section 2.2.2. All cells were cultured in DMEM supplemented with 10% FBS, 1% sodium pyruvate, 1% L- glutamine, and 1% penicillin/streptomycin (Invitrogen), and were maintained at 37◦C in an incubator containing 5% CO2. HTM2 cell media also contained 60 mg/mL zeocin. Transient transfections were performed using the calcium phosphate method, and cells were allowed to grow for an additional 48 to 72 h until confluent. Un- less otherwise stated, all cells were harvested in PBS and lysed by sonication with RIPA-DOC buffer containing 50 mM TrisHCl (pH 7.2), 150 mM NaCl, 1% de- oxycholate, 1% Triton X-100, 0.1% SDS, and protease inhibitor cocktail Complete (Roche). 3.2.3 siRNA knockdown HEK cells were seeded onto 35 mM plates and transfected with Lipofectamine TM 2000 the following day with various pools of 4 siRNAs which targeted either human TMP21 (TMED10 siGENOME SMARTpool, M-003718-01-0005, Dharmacon), hu- 49 3.2. Methods man p24a (TMED10 siGENOME SMARTpool, M-008074-01-0005, Dharmacon), or were non-targeting (siGENOME Non-Targeting siRNA Pool #2, D-001206-14-05, Dharmacon). siRNA pools were diluted in RNase-free siRNA buffer (Dharmacon) containing 300 mM KCl, 30 mM HEPES (pH 7.5), and 1.0 mM MgCl2 in diethylpy- rocarbonate (DEPC) treated water, to a concentration of 20 mM. Cells were given fresh media, and siRNA was prepared with Lipofectamine TM 2000 in Opti-media as per the manufacturer’s protocol (Invitrogen). Cells were harvested 48 to 72 h later to access knockdown. 3.2.4 Coimmunoprecipitation HEK and HTM2 cells were seeded onto 10 cm plates and transfected the follow- ing day with myc-tagged expression plasmids of either PS1, PS2, Aph-1, Pen-2, Nct, BACE1, BACE2, the S2-cleavage product of Notch (MV), or with untagged expression plasmids of either TMP21, APPwt , APPSwe, C99 or C83 by the cal- cium phosphate method. Cells were harvested 48 h later in PBS and centrifuged at 1,000×g for 2 min at 4◦C. Cell pellets were lysed in 0.4 to 0.8 mL NP-40 buffer (10 mM HEPES pH 7.5, 142.4 mM KCl, 5 mM MgCl2, 1 mM EDTA, 1% NP-40) by sonication for 10 × 0.6 sec bursts with 1 sec rest in between. Samples were spun at 2,000×g for 2 min and supernatents were transferred to a new tube. Immuno- precipitation antibody was added and tubes were gently mixed at 4◦C for 1 h, after which 25 mL of protein A/G coupled to agarose beads was added and samples were allowed to mix overnight at 4◦C. The following day, beads were rinsed three times in PBS and striped with addition of 2× tricine sample loading buffer (Invitrogen) and boiling for 5 min. Samples were frozen at -80◦C until analysis by immunoblotting. For HEK cell experiments, cells transfected with myc-tagged expression plasmids were immunoprecipitated with anti-myc antibody 9E10, APP and CTF samples were 50 3.2. Methods immunoprecipitated with anti-APP-CTF antibody C20, and a positive control sam- ple was immunoprecipitated with anti-TMP21 antibody T21. Immunoblotting was performed using 9E10. For HTM2 cell experiments, samples were immunoprecipi- tated with anti-TMP21 antibody T51 and immunoblotted with anti-myc antibody 9E10. 3.2.5 Immunoblotting For immunoblot analysis, samples were loaded onto 16% Tris-Tricine gels and trans- ferred to PVDF-FL membranes. Membranes were blocked for 1 h in 5% milk in PBS and incubated with the required primary antibodies in 5% milk in PBS-T at 4◦C overnight. Common primary antibodies were mouse anti-myc antibody 9E10, mouse anti-actin antibody AC-15, rabbit anti-TMP21 antibodies T21 or T51, and rabbit anti-APP-CTF antibody C20. The next day membranes were rinsed in PBS-T, incubated with IRDye TM 680-labeled goat anti-rabbit and IRDye TM 800CW-labeled goat anti-mouse antibodies in PBS-T at room temperature for 1 h, further rinsed in PBS-T, and visualized on the LI-COR Odyssey R© system. All quantification measurements were based on original CCD optical density. 3.2.6 Immunocytochemistry HTM2 and HEK cells were seeded onto glass coverslips in 24 well plates. The following day cells were transfected with expression plasmids of either PS1, PS2, Aph-1, Pen-2, Nct, BACE1, or BACE2, or with untagged expression plasmids of either TMP21, APPwt , APPSwe, C99 or C83 by the calcium phosphate method. After 48 to 72 h, cells were rinsed in PBS, fixed in 4% PFA for 20 min, permeabolized with PBS-Tx for 30 min, blocked in 5% BSA in PBS-Tx for 30 min, and incubated with or without primary antibodies in 1% BSA in PBS-Tx overnight at 4◦C. The 51 3.2. Methods next day cells were rinsed, incubated for 1 h with goat anti-rabbit Alexa Fluor 488 (green) and goat anti-mouse Alexa Fluor 568 (red), rinsed in PBS-Tx, and mounted using Fluoromount-G (Southern Biotech). Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. Images were assembled in Adobe Photoshop (versions CS2 or CS3), cropped to equal size, and the relative brightness levels for each channel were adjusted to allow visualization of both antibodies in the same image. Slides on which the primary antibody was omitted were adjusted in the same way as slides with antibody. 3.2.7 Dot-blot interaction array Human TMP21 is composed of 220 amino acids. The amino acid sequence of TMP21 was synthesized to create a dot-blot array of 21 dots of 20-mer peptides, which cover the entire TMP21 sequence with a 10-mer overlap (Figure 3.1; Ta- ble 3.1). Peptides were synthesized by Jimmy Guo using an AutoSpot robot (In- tavis Bioanalytical Instruments) programmed using software from DIGEN (Ger- many). 9-fluorenylmethyloxycarbonyl (FMOC) amino acids (Intavis Bioanalytical Instruments) were dissolved in dimethylformamide (DMF) to 0.25 M and and ac- tivated with 1-hydroxybenzotriazole and N,N’-disopropylcarbodiimide (Sigma) for at least 15 min. FMOC amino acids were delivered to a nitrocellulose membranes derivatized with a polyethylene linker and free amino terminal (Intavis Bioanalytical Instruments) in 60 nL aliquots per spot by the robotic synthesizer. Fifteen minutes after the completion of each cycle, the membranes were removed from the apparatus and treated with 2% acetic anhydride in DMF to acetylate any free remaining amino groups. Membranes were then washed and further treated with 20% piperidine in DMF to remove the FMOC protecting group, followed by washing with DMF and methanol, and finally by air drying. The membrane was then precisely repositioned 52 3.2. Methods on the robotic apparatus to initiate the next coupling cycle. After the final cycle, the side chain protecting groups were removed by treatment with a solution of 5 mL 50% trifluoracetic acid, 5 mL dichloromethane (DCM), 300 mL triisopropyl silane (Sigma), and 200 mL water. Membranes were then washed twice with DCM, twice with DMF, and twice with methanol. After air drying, membranes were stored in a sealed bag at 4◦C until used. Figure 3.1: Sequence information for the dot-blot interaction array. The entire amino acid sequence of human TMP21 was represented as a series of 20-mer synthetic peptides with a 10-mer overlap. “D-#” represents dot number. HEK cells were transfected with expression plasmids of Aph1-myc, Nct1-myc, Pen2-myc, PS1-myc, PS2-myc, BACE1-myc, or a negative calcium-phosphate only control. Cells were allowed to grow for 48 to 72 h, after which they were rinsed in PBS, harvested in PBS, and lysed in either RIPA-DOC or NP-40 lysis buffer and frozen at -80◦C until further analysis. Membranes were briefly soaked in methanol, rinsed in PBS, and blocked in either 1:1 OBB:PBS or 5% milk in PBS. The use 53 3.2. Methods Dot Number Start AA End AA Sequence Row One 1 1 20 MSGLSGPPARRGPFPLALLL 2 11 30 RGPFPLALLLLFLLGPRLVL 3 21 40 LFLLGPRLVLAISFHLPINS 4 31 50 AISFHLPINSRKCLREEIHK 5 41 60 RKCLREEIHKDLLVTGAYEI 6 51 70 DLLVTGAYEISDQSGGAGGL 7 61 80 SDQSGGAGGLRSHLKITDSA 8 71 90 RSHLKITDSAGHILYSKEDA 9 81 100 GHILYSKEDATKGKFAFTTE 10 91 110 TKGKFAFTTEDYDMFEVCFE 11 101 120 DYDMFEVCFESKGTGRIPDQ 12 111 130 SKGTGRIPDQLVILDMKHGV Row Two 13 121 140 LVILDMKHGVEAKNYEEIAK 14 131 150 EAKNYEEIAKVEKLKPLEVE 15 141 160 VEKLKPLEVELRRLEDLSES 16 151 170 LRRLEDLSESIVNDFAYMKK 17 161 180 IVNDFAYMKKREEEMRDTNE 18 171 190 REEEMRDTNESTNTRVLYFS 19 181 200 STNTRVLYFSIFSMFCLIGL 20 191 210 IFSMFCLIGLATWQVFYLRR 21 201 220 ATWQVFYLRRFFKAKKLIE – – – Spacer, no peptide 22 myc control EQKLISEEDLNMHTG Table 3.1: Sequence information for the dot-blot interaction array. The entire amino acid (AA) sequence of human TMP21 was represented as a series of 20-mer synthetic peptides with a 10-mer overlap. Some membranes also contained a positive control dot for the myc-epitope. 54 3.3. Results of different lysis buffers and blocking agents allowed a more thorough inspection of protein interaction. Bound proteins possessing a myc-tag were detected using mouse anti-myc antibody 9E10. Antibodies were visualized using IRDye TM 800CW-labeled goat anti-mouse secondary antibodies, and scanned on the LI-COR Odyssey R© sys- tem. Potential interactions were explored by quantitative analysis using the LI-COR Odyssey R© analysis tools. Dot 21 showed the strongest response, so all dots were nor- malized to dot 21 optical density. Samples were then compared to levels seen in the control HEK lysate probed membrane. Dots which were unable to be distinguished from background were removed from the analysis. Due to the high background present in the PS1-myc probed membrane, it was excluded from the analysis. Al- pha and beta regions, hydrophobicity, and average charge were determined using the Protean program in the DNAStar software package. 3.3 Results 3.3.1 C83 production is lower when TMP21 is overexpressed Transient transfection of APPwt or APPSwe expression plasmids into TMP21 sta- ble cell line HTM2 caused a significant reduction in APP C-terminal fragment C83, as compared to HEK cell transfections (Figure 3.2 A, B). Similar results were seen when both TMP21 and APPwt or APPSwe were transiently overexpressed (Fig- ure 3.2C). In the complementary condition, TMP21 was transiently overexpressed in APPwt (HAW) or APPSwe (20E2) stable cell lines, with no effect on C83 levels (Figure 3.2E). Similarly, siRNA knockdown of TMP21 had no effect on C83 levels (Figure 3.2D). 55 3.3. Results Figure 3.2: Overexpression of TMP21 reduces C83 production. HEK and HTM2 cells were transfected with APPwt (A) or APPSwe (B) by the calcium phosphate method, harvested 48 to 72 h later in PBS, and lysed in RIPA-DOC. Equal quantities of total protein were loaded onto 16% tricine gels for immunoblot analysis to detect holoAPP, b-actin, TMP21-myc, and APP C-terminal fragment C83. When TMP21 was stably overexpressed, there was a significant reduction in C83 levels (**p < 0.01, ***p < 0.001, plotted as means ±sem). (C) HEK cells were transfected with expression plasmids either APPwt or APPSwe and a TMP21 or empty vector. Transient overexpression of untagged TMP21 also caused a reduction in C83 levels. (D) siRNA knockdown of TMP21 in APPwt (HAW) or APPSwe (20E2) stable cells did not influence C83 levels. (E) Overexpression of either tagged or untagged TMP21 did not affect C83 production in the HAW or 20E2 cell lines. 56 3.3. Results 3.3.2 C99 production is higher when TMP21 is overexpressed Transient transfection of BACE1-myc and either APPwt or APPSwe expression plasmids into TMP21 stable cell line HTM2 caused a significant increase in APP C-terminal fragment C99, as compared to HEK cell transfections (Figure 3.3 A, B). Conversely, overexpression of either tagged or untagged TMP21 did not affect C99 production in APPwt+BACE1-myc stable cells (BAW) or APPSwe+BACE1-myc stable cells (2EB2). Furthermore, siRNA knockdown of TMP21 did not influence C99 production in 2EB2 cells. 3.3.3 TMP21 colocalizes and interacts with members of the g-secretase complex HEK cells were transfected with expression plasmids of myc-tagged Aph1, Nct, Pen2, PS1, or PS2, with or without a TMP21 expression plasmid. After 48 to 72 h, immunofluorescent labeling was used to examine colocalization of TMP21 with AD- related proteins. Myc-tagged proteins were detected with primary antibody 9E10 (mouse anti-myc) and secondary antibody goat anti-mouse Alexa Fluor 568 (red), while TMP21 was detected with primary antibody T21 (rabbit anti-TMP21) and secondary antibody goat anti-rabbit Alexa Fluor 488 (green). Both endogenous (Fig- ure 3.4) and exogenous (Figure 3.5) TMP21 were found to colocalize with members of the g-secretase complex. Coimmunoprecipitation was used to determine whether there was a direct in- teraction between TMP21 and members of the g-secretase complex in vivo, as previously suggested (Chen et al., 2006). To examine interactions of endogenous TMP21 with g-secretase complex components, HEK cells were transfected with myc-tagged expression plasmids of Aph1, Pen2, Nct1, PS1, or PS2, immunoprecipi- tated with anti-myc antibody 9E10, and analysed by immunoblot using anti-TMP21 57 3.3. Results Figure 3.3: Overexpression of TMP21 increases C99 production. (A) HEK and HTM2 cells were transfected with APPwt or APPSwe expression plasmids by the calcium phosphate method, harvested 48 to 72 h later in PBS, and lysed in RIPA-DOC. Equal quantities of total protein were loaded onto 16% tricine gels for immunoblot analysis to detect holoAPP, BACE1-myc, b-actin, TMP21-myc, and the APP C-terminal fragments C99, C89, and C83. When TMP21 was stably overexpressed, there was a significant increase in C99 levels, as quantified in (B) (***p < 0.001, plotted as means ±sem). (C) Overexpression of either tagged or untagged TMP21 did not affect C99 production in APPwt+BACE1-myc stable cells (BAW) or APPSwe+BACE1-myc stable cells (2EB2). (D) siRNA knockdown did not influence C99 production in 2EB2 cells. 58 3.3. Results Figure 3.4: Endogenous TMP21 colocalizes with members of the g- secretase complex. HEK cells were transfected with expression plasmids of myc- tagged Aph1, Nct, Pen2, PS1, or PS2. After 48 to 72 h, cells were fixed in 4% PFA, permeabolized, blocked with 5% BSA, and incubated with or without primary anti- bodies 9E10 (mouse anti-myc) and T21 (rabbit anti-TMP21) overnight at 4◦C. The next day cells were rinsed and primary antibodies were detected by a 1 h incubation with goat anti-mouse Alexa Fluor 568 (red) and goat anti-rabbit Alexa Fluor 488 (green), rinsed, and mounted onto slides with Fluoromount-G. Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. Images were assembled in Adobe Photoshop, cropped to equal sizes, and the rela- tive brightness levels for each channel were adjusted to allow visualization of both antibodies in the same image. Slides on which the primary antibody was omitted were adjusted in the same way as slides with antibody. 59 3.3. Results Figure 3.5: Exogenous TMP21 colocalizes with members of the g- secretase complex. HEK cells were transfected with expression plasmids of TMP21 in addition to either myc-tagged Aph1, Nct, Pen2, PS1, or PS2. After 48 to 72 h, cells were fixed in 4% PFA, permeabolized, blocked with 5% BSA, and incubated with or without primary antibodies 9E10 (mouse anti-myc) and T21 (rabbit anti-TMP21) overnight at 4◦C. The next day cells were rinsed and primary antibodies were detected by a 1 h incubation with goat anti-mouse Alexa Fluor 568 (red) and goat anti-rabbit Alexa Fluor 488 (green), rinsed, and mounted onto slides with Fluoromount-G. Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. Images were assembled in Adobe Photo- shop, cropped to equal sizes, and the relative brightness levels for each channel were adjusted to allow visualization of both antibodies in the same image. Slides on which the primary antibody was omitted were adjusted in the same way as slides with antibody. 60 3.3. Results antibody T21. Interactions of exogenous TMP21 with g-secretase complex com- ponents was examined by similarly transfecting HTM2 cells, immunoprecipitating with anti-TMP21 antibody T51, and immunoblotting with 9E10. No interaction was seen between TMP21 and g-secretase complex components in either condition (Figure 3.6). While previous groups have demonstrated a direct interaction between TMP21 and the g-secretase complex in vitro, others have been unable to replicate this find- ing (Fraering et al., 2004; Vetrivel et al., 2007; Winkler et al., 2009). However, this does not exclude the possibility that TMP21 may exert its effects through a transient association with the g-secretase complex. To examine potential inter- action sites between members of the g-secretase complex and TMP21 in situ, a dot-blot array of the entire TMP21 protein sequence was synthesized as a series of 20-mer peptides with a 10-mer overlap. HEK cells were transfected with expres- sion plasmids of Aph1-myc, Nct1-myc, Pen2-myc, PS1-myc, PS2-myc, or a negative calcium-phosphate only control, to create probing lysates. Several variations of the dot-blot were performed, using different blocking agents and lysis buffers to allow a more complete analysis of protein interaction. Bound proteins were detected using 9E10 anti-myc antibody and IRDye TM 800CW-labeled goat anti-mouse (Figure 3.7). 61 3.3. Results Figure 3.6: BACE1 coimmunoprecipitates with TMP21 only when TMP21 is overexpressed. (A) HTM2 cells were transfected with myc-tagged expression plasmids of Aph1, Pen2, Nct1, PS1, PS2, or BACE1, or negative control by the calcium phosphate method. After 48 to 72 h, cells were harvested in PBS and lysed in NP-40 buffer. Lysates were immunoprecipitated with anti-TMP21 an- tibody T51 and analysed by immunoblot using anti-myc antibody 9E10. BACE1 coimmunoprecipitated with TMP21 when TMP21 was stably overexpressed. (B) HEK cells were transfected as per (A) above, or with untagged TMP21 as a positive control. Cells were harvested 48 to 72 h later in PBS and lysed in NP-40 lysis buffer. The positive control was immunoprecipitated with anti-TMP21 antibody T21, while the remaining samples were immunoprecipitated with anti-myc antibody 9E10. Im- munoblotting for TMP21 with antibody T21 revealed no interaction with BACE1 when TMP21 is expressed at endogenous levels, despite robust immunoprecipitation of the positive control. 62 3.3. R esu lts Figure 3.7: Specific regions of TMP21 serve as potential interaction sites with Aph1, Pen2, Nicastrin, and PS2. A dot-blot array of the entire TMP21 protein sequence was synthesized as a series of 20-mer peptides with a 10-mer overlap, as per table 3.1. HEK cells were transfected with expression plasmids of Aph1-myc, Nct1-myc, Pen2-myc, PS1-myc, or PS2-myc or a negative calcium-phosphate only control. Membranes were blocked with either OBB or 5% milk in PBS, and probed with the transfected cell lysates, which were lysed in either RIPA-DOC or NP-40 lysis buffer. The use of different blocking agents and lysis buffers allowed a more complete analysis of the protein interaction. Bound proteins were detected using 9E10 anti-myc antibody and IRDye TM 800CW-labeled goat anti-mouse, and visualized on the LiCor Odyssey R© system. (A) Membranes blocked in OBB, cells lysed with RIPA-DOC. (B) Membranes blocked in 5% milk, cells lysed with RIPA-DOC. (C) Membranes blocked with 5% milk, cells lysed with NP-40. 63 3.3. Results Dot-blot results were quantified in LI-COR Odyssey R© software. Dot 21 resulted in a strong background signal in all samples and was used as a calibration point. All dots were normalized to Dot 21, and then normalized to the corresponding HEK dot value to determine specific binding (Figure 3.8). Aph-1, Pen-2, Nct, and PS2 showed potential interaction with the large N-terminal domain of TMP21, while Aph-1 also showed a potential interaction with the smaller, cytosolic C-terminal of TMP21. This suggests there is interaction potential between TMP21 and members of the g-secretase complex, despite inconsistent coimmunoprecipitation data in the literature. 3.3.4 TMP21 associates with BACE1 only when TMP21 is overexpressed For overexpression of TMP21 to affect APP cleavage, something must differ between the overexpressed and endogenous states. As the g-secretase complex did not yield any clues, BACE1, the other player in the pathogenic pathway of AD, was examined. BACE1-myc was found to coimmunoprecipitate with TMP21 in vitro when it was transiently overexpressed in HTM2 cells (Figure 3.6A). Interestingly, BACE1 did not coimmunoprecipitate when TMP21 was at endogenous levels (Figure 3.6B). This suggests that the alteration seen in APP processing when TMP21 is overexpressed could be due to TMP21’s effect on BACE1. BACE1 is localized in the Golgi and endosomes (Vassar et al., 1999). Immunoflu- orescent labelling was used to determine whether colocalization of TMP21 with BACE1 was dependent on TMP21 expression levels. HEK cells were transfected with expression plasmids of BACE1-myc, or BACE1-myc and TMP21. After 48 to 72 h, BACE1 was detected with primary antibody 9E10 and goat anti-mouse Alexa Fluor 568 (red), while TMP21 was detected with primary antibody T21 and sec- 64 3.3. Results Figure 3.8: Specific regions of TMP21 may interact with Aph1, Nct, Pen2, and PS2. (A) A dot-blot array of the entire TMP21 protein sequence was synthesized as a series of 20-mer peptides with a 10-mer overlap, as per table 3.1 and seen in Figure 3.7. An interaction was found in blots blocked with 1:1 OBB:PBS and probed with cell lysates. These dots were quantified in the LI-COR Odyssey R© software. Dot 21 resulted in a strong background signal in all samples and was used as a calibration point. All dots were normalized to Dot 21. Dots were then nor- malized to the corresponding HEK dot value to determine specific binding (arrows). Red arrows indicate binding to the large N-terminal domain of TMP21, while black arrows correspond to binding to the smaller, cytosolic C-terminal of TMP21. Aph1 showed a strong interaction with dots 8 and 20, while Pen2, Nct1, and PS2 showed a strong interaction with dot 17. These dots correspond to the regions of TMP21 indicated in (B). 65 3.3. Results ondary antibody goat anti-rabbit Alexa Fluor 488 (green). TMP21 was more likely to colocalize with BACE1 when overexpressed (Figure 3.9). Figure 3.9: TMP21 is more likely to colocalize with BACE1 when over- expressed. HEK cells were transfected with expression plasmids of BACE1-myc, or BACE1-myc and TMP21. After 48 to 72 h, cells were fixed in 4% PFA, perme- abolized, blocked with 5% BSA, and incubated with or without primary antibodies 9E10 (mouse anti-myc) and T21 (rabbit anti-TMP21) overnight at 4◦C. The next day cells were rinsed and primary antibodies were detected by a 1 h incubation with goat anti-mouse Alexa Fluor 568 (red) and goat anti-rabbit Alexa Fluor 488 (green), rinsed, and mounted onto slides with Fluoromount-G. Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. Images were assembled in Adobe Photoshop, cropped to equal sizes, and the relative brightness levels for each channel were adjusted to allow visualization of both antibodies in the same image. Slides in which the primary antibody was omitted were adjusted in the same way as slides with antibody. To determine the specific site of interaction between TMP21 and BACE1, a TMP21 dot-blot array was performed as previously described (section 3.3.3) using BACE1-myc cell lysates. No interaction site was found, although this technique has limitations which will be further discussed in section 3.4.1. 66 3.4. Discussion Figure 3.10: The dot blot interaction array did not find an interaction site with BACE1. (A) A dot-blot array of the entire TMP21 protein sequence was synthesized as a series of 20-mer peptides with a 10-mer overlap, as per table 3.1. HEK cells were transfected with a BACE1-myc expression plasmid or a negative calcium-phosphate only control. Membranes were blocked with 5% milk in PBS, and probed with the transfected cell lysates, which were lysed in either RIPA-DOC (B) or NP-40 (C) lysis buffer. The use of different lysis buffers allowed a more complete analysis of the protein interaction. Bound proteins were detected using 9E10 anti- myc antibody and IRDye TM 800CW-labeled goat anti-mouse, and visualized on the LI-COR Odyssey R© system. 67 3.4. Discussion 3.4 Discussion 3.4.1 Summary Both stable and transient overexpression of TMP21 decreased C83 production in cells transiently transfected with APPwt or APPSwe; however, transient overex- pression of TMP21 in cells stably expressing APPwt or APPSwe did not show the same effect (section 3.3.1). Similarly, C99 production was enhanced in TMP21 sta- ble cells transiently transfected with BACE1-myc and either APPwt or APPSwe, but was unaffected by transient overexpression of TMP21 in APPwt+BACE1-myc or APPSwe+BACE1-myc stable cells (section 3.3.2). Overall, these results suggest that overexpression of TMP21 shifts APP process- ing away from a-secretase cleavage, as indicated by the decrease in C83, and towards b-secretase cleavage, as indicated by the rise in C99. This suggests that in regards to TMP21 overexpression, BACE1, rather than g-secretase, may be the major player. Supporting this theory, while TMP21 colocalizes with members of the g-secretase complex (Figures 3.4 and 3.5), TMP21 and g-secretase complex components did not coimmunoprecipitate (Figure 3.6), similar to reports from other groups (Fraering et al., 2004; Vetrivel et al., 2007; Winkler et al., 2009). Also, while specific regions of TMP21 showed potential interaction sites with Aph-1, Pen-2, Nct, and PS2 (Fig- ure 3.7), one of these were on the C-terminal side of TMP21, which is involved in coat polymerization (Figure 1.2). As the g-secretase complex is found within the ER and Golgi, one would expect interactions with TMP21 to occur solely with its luminal N-terminus. Interestingly, TMP21 was found to coimmunoprecipitate with BACE1, specif- ically when BACE1 was overexpressed (Figure 3.6). Furthermore, TMP21 and BACE1 localization overlap (Figure 3.9). Despite this, a direct interaction be- 68 3.4. Discussion tween TMP21 with BACE1 could not be found using the TMP21 dot-blot array (Figure 3.7), although this system does have its caveats. The dot-blot array is a very artificial setup to explore protein interactions, as it is sensitive to blocking and lysate conditions and the presented epitope is an unfolded 20 amino acid peptide. APPwt cleavage is suggested to primarily occur in endosomal compartments (Koo and Squazzo, 1994; Perez et al., 1999), while APPSwe is cleaved in the TGN (Haass et al., 1993; Martin et al., 1995; Stephens and Austen, 1996). As both APPwt and APPSwe were equally affected by TMP21 overexpression, subcellular fractionation is being performed to further examine how BACE1 is affecting this process. 3.4.2 Results in context with the literature Thus far, the effect of TMP21 overexpression on APP CTFs has received little atten- tion, although the consensus is that overexpression does not affect Ab levels (Chen et al., 2006; Dolcini et al., 2008; Hasegawa et al., 2010). The one study which briefly addressed CTFs found that overexpression of a TMP21-FLAG fusion construct in- creased CTFs in HEK cells transfected with APPwt , and caused a corresponding decrease in Ab (Pardossi-Piquard et al., 2009). Although the CTF was not dis- tinctly identified, one can assume it was C83, as BACE1 was not simultaneously overexpressed. This suggests that overexpression of TMP21 enhances the a-secretase pathway, resulting in less b-secretase cleavage and therefore less Ab; however, this cannot be concluded without examining C99 levels as well. This thesis has found that TMP21 overexpression decreases C83 levels, with a corresponding increase in C99, results which are self-consistent, and suggest BACE1 is involved. An exami- nation of Ab levels would provided further insight on a potential mechanism for this effect. 69 3.4. Discussion As the majority of previous research has focused on how TMP21 suppression increases Ab production (Chen et al., 2006; Vetrivel et al., 2007; Dolcini et al., 2008), the effect on APP CTF formation has only received minor attention. Interestingly, similar to overexpression, suppression of TMP21 caused an increase in APP CTFs in HeLa cells stably expressing APPSwe, which was attributed to increased stability of nascent APP (Vetrivel et al., 2007). Again, as BACE1 was not overexpressed, we can assume the CTF was C83. From an enzymatic standpoint, a higher level of Ab coupled with a higher level of the non-pathogenic cleavage product C83 is inconsistent, unless the cause of the increased Ab is an overall increase in APP available to be cleaved, or a corresponding increase in C99 as well. This thesis shows that siRNA knockdown of TMP21 in HEK cells stably expressing both APPSwe and BACE1 have similar C99 levels (Figure 3.3D), lending support to the theory that suppression of TMP21 increases Ab production through its effect on APP trafficking (Vetrivel et al., 2007). It should be kept in mind, however, that suppression of TMP21 in cell-free assays can also increase Ab production (Chen et al., 2006), suggesting that TMP21 may affect APP processing through both modulation and trafficking. 3.4.3 Conclusion In conclusion, evidence from the literature suggests that TMP21 has both a mod- ulatory impact on g-secretase as well as a role in APP trafficking, which together result in Ab overexpression when TMP21 is suppressed (Chen et al., 2006; Vetrivel et al., 2007). Studies of TMP21 overexpression have had mixed findings. This the- sis makes the novel suggestion that TMP21 may additionally affect APP processing through its effect on BACE1. Further research is necessary to elucidate the precise mechanism of this effect. 70 Chapter 4 TMP21 heterozygous mice display heightened anxiety 4.1 Introduction 4.1.1 Rationale Knockdown of TMP21 increases Ab production in vitro (Chen et al., 2006) and human AD patients have less TMP21 protein in their brain (Vetrivel et al., 2008a). However, from these findings alone we cannot say that decreased levels of TMP21 contributes to AD pathogenesis; we can only say that there is a correlation between low TMP21 levels and the disease state. One method to truly determine whether a deficit in TMP21 can exacerbate AD pathology in vivo is to compare two groups of AD model mice: one in which TMP21 expression is at wildtype levels, and the other in which TMP21 is deficient. Examination of AD pathology, as well as the behavioural consequences of altered levels of TMP21 in an AD model, would provide a robust, causal linkage between TMP21 protein levels and the disease state. While several studies have utilized biochemical techniques to study the function of TMP21 in trafficking and Golgi structure (see section 1.2), the behavioural role of TMP21 has yet to be examined. Without an understanding of how TMP21 suppression affects mouse behaviour in general, we cannot properly interpret results from a cross with AD model mice. Thus this chapter will examine the behavioural consequences of TMP21 suppression in the C57BL/6 mouse background. 71 4.1. Introduction 4.1.2 The TMP21 heterozygous mouse, S2P23 TMP21 heterozygous mouse strain S2P23 was created by Denzel et al. (2000) using a gene targeting technique which replaced the first exon of TMP21 with the neomycin- resistance gene (Figure 4.1). Complete knockdown of TMP21 resulted in embryonic lethality prior to the blastocyst stage, as no homozygotes could be isolated at 3.5 days post coitus, suggesting that TMP21 is of vital importance at the earliest stages of mammalian development (Denzel et al., 2000). Heterozygous mice appeared grossly normal, but upon microscopic evaluation displayed dilated Golgi cisternae and increased vacuoles surrounding the Golgi in liver and kidney tissues (Denzel et al., 2000). This is in accordance with previous research which demonstrated TMP21’s role in Golgi structural maintenance (see section 1.2.3). 4.1.3 Behavioural testing As the S2P23 strain had not been previously characterized, a full set of behavioural testing was required. This included tests of motor function, anxiety, and learning and memory. Multiple tests were used for each category to obtain a thorough examination of behaviour. Motor testing Hanging wire The hanging wire test is a simple test of motor strength which is capable of detecting neuromuscular abnormalities (Takahashi et al., 2009). The mouse is placed on a wire grid within an area cordoned off with tape. The mouse is made to grip the grid, which is then inverted over a height sufficient to prevent the mouse from easily climbing down, but not great enough to cause injury in the case of a fall. Latency to fall is taken as a measure of muscle strength. 72 4.1. Introduction Figure 4.1: Generation of p23 mutant mice by gene targeting. Originally published by Denzel et al. (2000) in Current Biology and used with permission from Elsevier (license #2604910209883). The following text is the original caption from the paper: (a) Schematic diagram of the p23 locus, the targeting vector and the structure of the targeted allele. ES cells were electroporated with the targeting vec- tor and correctly targeted, G418-resistant colonies were identified by Southern blot analysis of BamHI-digested genomic DNA. Of 576 G418-resistant colonies screened, 3 were correctly targeted. Two different ES cell clones, in which homologous re- combination events had occurred, were independently injected into C57BL/6 blas- tocysts. Both clones transmitted through the germline. (b) Southern blot analysis showing correct 3’ targeting of the p23 locus. Genomic tail DNA from wild-type (WT) and heterozygous (HE) mice was digested with BamHI and hybridized to probe A. The resulting 8.5 kb and 3.3 kb fragments correspond to the wild-type and mutated (M) genotype, respectively. A, ApaI; B, BamHI; E, EcoRI; H, HindIII; K, KpnI; S, SmaI; X, XhoI. 73 4.1. Introduction Rotarod The rotarod test is a common test of motor coordination and balance in rodents which has been successfully adapted to mice (Jones and Roberts, 1968a). The test involves placing the mouse on a rotating cylinder; the mouse must contin- ually walk to avoid falling off the apparatus. Latency to fall is taken as the measure of motor coordination and balance. While a constant speed apparatus was often used in the past, more modern equipment utilizes an accelerating rotarod, which has been shown to produce more reliable results (Jones and Roberts, 1968b). The rotarod test can detect cerebellar abnormalities, as test performance can be significantly impaired by cerebellar defects (Cendeĺın et al., 2008, 2010). The background strain of the mice also significantly affects results (Homanics et al., 1999; Võikar et al., 2001; Bearzatto et al., 2005), thus it is important to use wildtype littermate controls in any novel characterizations. Open field The open field test is the most widely used test to measure sponta- neous locomotor activity (Crawley, 2000a). The first open field apparatus consisted of a large arena approximately 1 m2, with 25, 22 cm2 blocks drawn on the floor; a human observer quantified the number of squares a rat crossed in a given time period as a measure of spontaneous locomotion (Broadhurst, 1961). The technique was also successfully applied to studies of locomotor activity in the mouse (DeFries et al., 1966). Since then, the open field apparatus has decreased in size and become fully automated. Digital tracking software can calculate the path length of each mouse, and infrared photobeam arrays track rearing behaviour. Similar to other behavioural tests, the background strain of the inbred mouse can greatly affect performance (Crawley et al., 1997). For example, C57BL/6 mice display heightened open field locomotion compared to other inbred strains (Crabbe, 1986). Thus it is important to use wildtype littermate controls when characterizing novel knockout or transgenic 74 4.1. Introduction mice. Anxiety testing Behaviour tests of anxiety are mostly exploratory-based, and as such can often be confounded by differences in locomotion such as hypo- or hyperactivity (Holmes, 2001). These confounding factors can be minimized by interpreting anxiety test results in the context of general locomotion, and by performing a variety of tests from which to draw conclusions. The basic principal of most exploratory-based anxiety tests is the opposition between the rodent’s natural tendency to explore a novel environment and its cau- tionary avoidance of aversive environments (Holmes, 2001). Therefore, for this the- sis, heightened anxiety will be defined as decreased exploration of novel, aversive environments or stimuli compared to controls. For example, a rodent which dis- plays a higher rate of exploration in an exposed, brightly lit environment, which is traditionally considered aversive, would be considered to have low anxiety, while a rodent with decreased exploration of this environment would be considered to have high anxiety. Open field While commonly used to measure locomotor activity (see page 74), measures from the open field test are also used to assess anxiety in rodents. The brightly lit, exposed central/Inner zone of the open field arena (see section 4.2.5, page 82) is considered an aversive environment. As such, mice with higher anxiety levels avoid the Inner zone (Holmes, 2001). Rearing behaviour in the open field is also taken as a measure of anxiety. When a mouse rears, it stands on its hind limbs to examine the environment, yet this exposes the animal to increased risk. As such, a mouse which rears less than controls is said to have a greater level of anxiety (Weisstaub et al., 2006). 75 4.1. Introduction Light-dark box Similar to the open field apparatus, the light-dark box measures the tendency to explore an aversive environment. The apparatus consists of a square arena divided equally into a dark, covered compartment and a brightly lit, exposed compartment (see section 4.2.6, page 83). Most mouse strains show a strong pref- erence for the dark environment, thus the light environment is considered aversive (Holmes, 2001). The traditional anxiety measure is the number of entries to the Light side, and the total time spent in the Light side. Similar to other behavioural tests, different inbred strains of mice show differ- ences in the light-dark box test (van Gaalen and Steckler, 2000; Bouwknecht and Paylor, 2002). Littermate controls should thus be used whenever possible. Learning and memory testing Y-maze The Y-maze has been used to study mouse behaviour for over 30 years (Kokkinidis and Anisman, 1976). The modern apparatus consists of a symmetrical Y-maze which the mouse is free to explore while being tracked with a digital tracking system (see section 4.2.7, page 85). The test measures spatial working memory based on the principal that a mouse prefers to explore the maze arm it has not recently occupied (Hughes, 2004). For example, if the arms of the maze are labelled A, B, and C, and the mouse begins in arm A and proceeds to arm B, the mouse must remember which arms it had previously explored in order to successfully explore the next novel arm, C. This measure is called spontaneous alternation, and is defined as the successive entries into three different arms in overlapping triplet sets (Hughes, 2004). For example, the 7-step sequence abcabcb consists of 4 sets: (1) abc, (2) bca, (3) cab, and (4) abc, out of a possible 5 sets (total number of entries minus 2), for an alternation rate of 80%. Previous studies have shown that spontaneous alternation behaviour in the Y-maze is highly hippocampal dependent (Reisel et al., 76 4.1. Introduction 2002; Dillon et al., 2008). By comparing alternation rates, differences in spatial working memory can be elucidated. Morris water maze The Morris water maze is the most frequently used test of learning and memory in rodents. Originally developed to test spatial memory in rats, the test was shown to be highly dependent on hippocampal functioning in both rats (Morris, 1981; Morris et al., 1982) and mice (Logue et al., 1997). While in general mice have impaired water maze performance compared to rats, it is suggested that this is due to the better swimming ability of rats rather than problems with spatial learning in mice, as C57BL/6 mice perform equally well as Long-Evans rats in tests of spatial memory on dry land (Whishaw and Tomie, 1996). The test has since been commonly used in the characterization of transgenic and knockout mice, as well as in studies of neurodegeneration. The Morris water maze test consists of a white pool 150 cm in diameter placed in a room which contains several visible, high contrast spatial cues. While usually considered a visuospatial task, blind rats have also successfully learned the task, suggesting non-spatial cues such as tactile cues, odors, and/or sounds may also be used in task acquisition (Lindner et al., 1997). One of the strengths of the Morris water maze is the ability to distinguish be- tween the visible and hidden platform conditions. On the first day the pool is filled with clear water and a raised, flagged, visible platform is placed in the pool, with a position which varies across trials. Mice are released into the water from variable starting positions and their progress is automatically tracked and recorded. The purpose of the visible platform task is to ensure the mice can perform the proce- dures necessary for the task, such as using spatial cues to find the flagged platform, and possess the swimming endurance necessary to swim the complete 60 sec trial. The hidden platform condition begins the following day, and allows a measure 77 4.1. Introduction of task acquisition. The platform is submerged under an opaque water surface and is fixed in one position. The mice are released into the pool from variable starting locations and must use the provided spatial cues to escape the maze. Over the course of several days, the mice gradually reduce their escape latency and path length to find the platform. Mice which perform normally in the visible platform trial, but have deficits in the hidden platform condition, are interpreted as having a learning and memory deficit (Crawley, 2000b). Finally, a probe trial is conducted on the last day of the experiment. Often, the probe trial is considered the true test of task acquisition, as mice can display normal task acquisition during the hidden platform trials, yet fail the probe trial (Crawley, 2000b). In this test, the platform is removed and the amount of time the mouse spent in the quadrant which previously held the platform is measured. Mice which have successfully acquired the task will spend more than the average amount of time in the platform quadrant. Fear conditioning The contextual and cued fear conditioning paradigms take advantage of a fundamental, common response to fear in many species – freezing. Freezing is defined as the absence of all movements, excluding respiratory-related movements, while the animal has a stereotypical crouching posture (Blanchard and Blanchard, 1969). Contextual and cued fear conditioning both involve exposing the mouse to a novel fear conditioning apparatus on Day 1, in which the mouse receives an unconditioned foot shock stimulus, resulting in the unconditioned response of freezing. In the contextual task, the context of the testing chamber serves as the neutral stimulus. On Day 2, the mouse is placed in the same testing chamber but does not receive a foot shock; the context of the testing environment now serves as the conditioned stimulus. Freezing is taken as a measure of the conditioned response. 78 4.2. Methods In the cued fear conditioning task, the unconditioned stimulus on Day 1 is pre- sented with an auditory cue neutral stimulus. On Day 2 the mouse is placed in an altered testing environment. Freezing behavior in the new context should be minimal. The auditory cue, now the conditioned stimulus, is then presented and freezing behaviour in the altered environment is taken as a measure of the condi- tioned response. To successfully compare mice in the fear conditioning task, certain assumptions must be made. As the task depends on the mouse experiencing an aversive stimulus, the mice being compared must possess similar pain thresholds (Crawley, 2000b). As pain thresholds vary with background strain of mice, wildtype littermates provide the best control option. While lesions to the amygdala interfere with both cued and contextual fear condi- tioning, hippocampal lesions only affect performance in contextual fear conditioning tested 24 hr later (Kim and Fanselow, 1992; Phillips and LeDoux, 1992; Kim et al., 1993). The combination of both cued and contextual fear conditioning therefore allows an examination of both amygdalar and hippocampal function. 4.2 Methods 4.2.1 Animals This study used the S2P23 strain of mice, which display heterozygous knockdown of TMP21 in a C57BL/6 background. Three pairs of breeder mice were shipped to our facility from Transgenic Services at Cancer Research UK. These founder mice were bred with C57BL/6 mice by Ms. Haiyan Zou to create a mouse colony located in the Animal Research Unit (ARU) at the University of British Columbia Hospital. At weaning, animals were genotyped by PCR using genomic DNA isolated from 79 4.2. Methods tail biopsies. This study used a total of 151 mice, 67 females (27 C57BL/6, 40 S2P23) and 84 males (34 C57BL/6, 50 S2P23), which were maintained until at least 12 months of age. Previous research showed similar behavioural results in all four stages of the estrous cycle, thus estrous cycle was not synchronized for female mice (Meziane et al., 2007). All procedures used in this study were in accordance with guidelines established by the Canadian Council on Animal Care and approved by the University of British Columbia Animal Care Committee. 4.2.2 Genotyping After weaning at 3 weeks of age, mice were anesthetized with isoflurane and ear marked. A tissue sample was taken from the ear punch or the tip of the tail. Tis- sue was digested in 300 mL of lysis buffer (10 mM Tris-HCl pH 8.0, 10 mM EDTA pH 8.0, 150 mM NaCl, 0.5% SDS) with 3 mL of 10 mg/mL Proteinase K (Fisher) at 55◦C overnight. The next day, samples were centrifuged and DNA was isolated from the supernatent by precipitation with 0.7× volume of isopropanol. DNA was pelleted by centrifugation at 16,100×g for 15 min, washed twice in 70% ethanol, dried, and resuspended in distilled water. Genotyping was performed by PCR with forward primer G-TMP21mice-F (5′–ccggactctaggtccgccaa–3′), and re- verse primers G-TMP21mice-R (5′–tctggtttgtttggcccactctccg–3′) and G- TMP21mice-Neo (5′–aattcgccaatgacaagacgct–3′). S2P23 negative (C57BL/6) mice displayed one band, while S2P23 hemi mice displayed two bands, with the pos- itive band being 260 bp in size. 80 4.2. Methods 4.2.3 General guidelines for behaviour testing Mice were divided into 7 cohorts based on birth date, such that testing usually began within 7 days of the 3 month, 6 month, and 12 month time points. Mice received standard husbandry care during testing, including cage enrichment (paper hut, half a Nestlet square and/or shredded paper, and PVC pipe) and ad. lib. access to food and water. All testing occurred during the light cycle. The majority of behaviour testing occurred on a 14-day schedule in the following order: Y-maze, open field, light-dark box, motor testing, rest, water maze (6 days), rest, contextual fear conditioning (2 days) (Figure 4.2). While it is traditional to wait a week between behaviour tests, rapid testing batteries reduce operating costs and do not significantly affect results (Paylor et al., 2006). The majority of animals were tested at 3 months, 6 months, and 12 months of age, providing a longitudinal study of behaviour. The cued fear conditioning task was an exception to this; animals used in this task were tested only once in their lifetime and received no further behaviour testing. For all behaviour tests excluding motor testing, mice were transferred from their holding facility in the ARU to the testing site via covered push cart, and allowed to rest for a minimum of 30 min before testing. Motor testing was performed within the ARU unit itself. Figure 4.2: Behaviour testing schedule. Testing occurred on a 14-day schedule, in the following order: Y-maze, open field, light-dark box, motor testing, rest, water maze (6 days), rest, contextual fear conditioning (2 days). Cued fear conditioning was performed separately on näıve mice who did not receive this training schedule. 81 4.2. Methods 4.2.4 Motor testing Hanging wire The hanging platform for hanging wire tests consisted of a section of 1 cm by 1 cm wire mesh, with a 9 cm by 12 cm area isolated by clear plastic tape. Mice were allowed to grip the center of the mesh and were swiftly inverted over a black container 50 cm in height. A black, cushioned pad was placed at the bottom of the container to ensure that falling mice were not injured. Latency to fall was measured, with a maximum hang time of 60 sec. Mice were returned to their home cage for approximately 30 min, after which they were re-tested. The average of the two trials was taken as a measure of hang strength. Rotarod Mice were placed on a single station, standard mouse rotarod (ENV-576M, Med Associates Inc., USA) with a shaft diameter of 3.2 cm, lane width of 5.7 cm, fall height of 16.5cm, and divider diameter of 24.8 cm. The rod was accelerated from 20 to 20,000 rpm over a 300 sec period. Latency to fall was measured. Mice were returned to their home cage for approximately 30 min, after which they were re- tested. The average of the two trials was taken as a measure of balance and motor coordination. 4.2.5 Open field Open field testing occurred in a bare arena measuring 40 cm by 40 cm with 35 cm high black, infrared transparent Perspex walls. Outside the walls and out of view of the mice was an infrared photo-beam array (ANY-Maze) connected to the ANY- Maze Interface, which detected mouse rearing upon beam breakage. The apparatus was calibrated to include an Inner zone defined by a distance of 5.8 cm from the 82 4.2. Methods walls of the apparatus (Figure 4.3). Figure 4.3: Open field apparatus. (A) The open field apparatus consists of a bare arena measuring 40 cm by 40 cm with 35 cm high black walls. A photobeam array is position outside the apparatus to detect rearing via beam breakage. A tracking camera views the arena from above (image c© Stoelting Co., used with permission). (B) Calibration image as viewed in ANY-Maze TM Video Tracking Software (version 4.63). Orange lines are zone boundaries. The blue area is defined as the Inner zone, while the perimeter is defined as the Outer zone. Mice were placed in a corner of the arena and were tracked using ANY-Maze TM Video Tracking Software (version 4.63) with a UniBrain FireI BBW digital camera. Mice were allowed to explore the maze for 5 min, after which they were returned to their home cage. The maze was cleaned before commencing the next mouse. Distance and average speed were analyzed as measures of motor ability. The number of rears, latency to first rear, latency to first enter the Inner zone, time in the Inner zone, and number of entries into the Inner zone were taken as measures of anxiety. 4.2.6 Light-dark box Light-dark box testing occurred in a 40 cm by 40 cm apparatus which was divided into two segments. The Dark side consisted of a 20 cm × 40 cm area enclosed by 35 cm high black, infrared transparent Perspex walls and a lid. The Light side 83 4.2. Methods consisted of a 20 cm by 40 cm area enclosed by 35 cm high clear Perspex walls. Dividing the two areas was a 40 cm by 35 cm sheet of black, infrared transparent Perspex with a small opening to allow access to both sides (Figure 4.4). Figure 4.4: Light-dark box apparatus. (A) The light-dark box apparatus consists of a 40 cm by 40 cm arena divided into two, 20 cm by 20 cm sections: an uncovered Light side, and a covered Dark side. A tracking camera views the arena from above (image c© Stoelting Co., used with permission). (B) Calibration image as viewed in ANY-Maze TM Video Tracking Software (version 4.63). Orange lines are zone boundaries. The Dark side is defined as a hidden zone, while the Light side is designated by the boundary lines. Entries to the Light side are included once 50% of the animal’s length has exited the hidden zone. Mice were placed in the Dark side of the apparatus and tracking commenced after the lid was added to the chamber. Tracking was achieved using ANY-Maze TM Video Tracking Software (version 4.63) with a UniBrain FireI BBW digital camera. Mice were allowed to explore the apparatus for 10 min, after which they were returned to their home cage. The maze was cleaned before commencing the next mouse. Latency to enter the Light side, total number of entries to the Light side, total time spent in the Light side, and mean length of visit to the Light side were taken as measures of anxiety. 84 4.2. Methods 4.2.7 Y-maze Y-maze testing occurred in a Y-maze with three arms, each 5 cm wide and 35 cm in length, and shielded with 10 cm high walls (Figure 4.5). Each arm was differen- tiated using visual cues attached to the inner top portion of the wall: White (white squares), Red (red triangles), or Blue (blue dots). Figure 4.5: Y-maze apparatus. (A) Image of the Y-maze apparatus with spec- ifications. x = 5 cm; y = 35 cm; z = 10 cm (image c© Stoelting Co., used with permission). (B) Calibration image as viewed in ANY-Maze TM Video Tracking Software (version 4.63). Orange lines represent zone boundaries between the White, Red, and Blue zones. The center zone is undefined. Tracking is achieved through an analog black and white camera and a RTV24 Digitizer (ANY-Maze). Mice were placed in the White arm and tracking of the mice commenced auto- matically using ANY-Maze TM Video Tracking Software (version 4.63) with an analog black and white camera and a RTV24 Digitizer (ANY-Maze). Mice were allowed to explore the maze for 8 min, after which they were returned to their home cage. The maze was cleaned before commencing the next mouse. Distance and average speed were analyzed as measures of motor ability, while spontaneous alternation was analyzed as a measure of spatial working memory. Percent spontaneous alternation was calculated as the percentage of the number of alternations performed out of the total number of alternations possible, based on the number of arm entries per 85 4.2. Methods mouse. 4.2.8 Morris water maze Water maze testing occurred in a pool 1.5 m in diameter with a white interior, with a water temperature of 22 ± 1◦C. Mice received visible platform trials on Day 1, followed by 4 days of hidden platform trials, and a probe trial on Day 6. Tracking of animals was achieved using ANY-Maze TM Video Tracking Software (version 4.63). Mice which floated for greater than 20 sec during any trial were eliminated from the analysis. In visible platform trials, a 10 cm white platform was marked with a flag and positioned 1 cm above a clear water surface. The position of platform and the starting direction varied across trials (Figure 4.6). Mice were tested for 5 contiguous trials with an inter-trial interval of 1h15. Mice were allowed to swim for at most 60 s before climbing onto the platform. Failed mice were gently guided to the platform and remained there for 20 s before being sent back to their home cage. Escape latency and path length to reach the platform were analyzed as a measure of visual acuity and swimming ability. In hidden platform trials, the platform was submerged 1 cm below an opaque water surface in a fixed position in the southwestern/third quadrant. Water was made opaque by adding non-toxic white paint. Mice were trained for 5 trials with an inter-trial interval of 1h15. Mice were allowed to swim for at most 60 s before climbing onto the platform, and again, failed mice were gently guided to the platform and remained there for 20 s before being sent back to their home cage. Escape latency and path length to reach the platform were analyzed as a measure of spatial learning and memory. In the probe trial, the platform was removed and mice were allowed to swim for 86 4.2. Methods Figure 4.6: Water maze apparatus. (A) Image of the water maze pool as used for visible platform trials. The 10 cm platform was raised 1 cm above the clear water surface and flagged. The position of the platform varied across trials (NW, NE, SW, SE, or Center), as did the starting direction of the mouse (N, S, E, W). As shown, the platform is in the SW quadrant. On hidden platform days, the flag was removed and the platform was submerged under an opaque water surface in the SW quadrant. (B) Calibration image as viewed in ANY-Maze TM Video Tracking Software (version 4.63). Orange lines represent zone boundaries. The platform zone was defined as a moving zone with 5 possible locations: NW, NE, SW, SE, or Center. Tracking was achieved through an analog black and white camera and a RTV24 Digitizer (ANY- Maze). On hidden platform days the platform was submerged under an opaque water surface and fixed in the SW quadrant. During the probe trial, the platform was removed. As shown, the platform is in the SW quadrant. 87 4.2. Methods 60 sec in the pool. The percentage of time spent in the third quadrant was analyzed. Escape latency and path length to reach the platform were analyzed as a measure of spatial learning and memory. 4.2.9 Contextual fear conditioning A simplified contextual fear conditioning paradigm was used. On the first day, mice were placed in the conditioning chamber for 5 min. The walls of the chamber consisted of a combination of plexiglass and steel. The floor of the chamber consisted of stainless steel rods, 2 mm in diameter, spaced 5 mm apart and connected to a shock generator. At the beginning of the 3rd min, mice received a foot-shock unconditioned stim- ulus (1mA, 50Hz, 3s). On the second day, mice were placed into the same chamber without receiving a foot-shock for 4 min. In each trial, freezing behaviour was recorded on a second-by-second basis using Freeze Frame TM (ActiMetrics Software). Freezing was used as an index of conditioned fear, and was defined as the absence of all movements, excluding respiratory-related movements, while the animal had a stereotypical crouching posture (Blanchard and Blanchard, 1969). Total test time was divided into 60 sec bins. The conditioned response was measured by analyzing the fold-increase in the percentage of time spent freezing on Day 2 versus Day 1. This was achieved by taking the average of all 60 sec bins on Day 2, and dividing by the average of the first 3 bins (0-120 sec, pre-shock) of Day 1. 4.2.10 Cued fear conditioning On the first day of fear conditioning, the apparatus was similar to that seen in con- textual fear conditioning (see section 4.2.9), with the addition of a pair of computer speakers and a light in the testing chamber (Figure 4.7A, B). An alcohol olfactory 88 4.2. Methods Figure 4.7: Fear conditioning apparatus. (A) For contextual fear condition (left), the apparatus was as shown with the exception that the computer speakers and light were absent. The apparatus remained the same on both days. For cued fear conditioning, the apparatus included a pair of computer speakers and a light in the testing chamber on Day 1 (left). On Day 2 (right), the texture and patten of the walls and flooring were altered, and an almond extract olfactory cue was added to the apparatus. The position of the light was also changed from the back of the apparatus to the front. (B) View from the overhead tracking camera for contextual fear conditioning on Day 1 of cued fear conditioning (top), and Day 2 of cued fear conditioning (bottom). Notice that the colors chosen for the walls on Day 2 show contrast in the black and white view. 89 4.2. Methods cue was wiped inside the chamber between each mouse. Mice were placed in the conditioning chamber for 360 sec. After 150 sec, a hiss sound was played for 30 sec. The last 3 sec of the sound were accompanied by a foot-shock conditioned stimulus (1mA, 50Hz, 3s). On Day 2, both the visual and olfactory cues in the chamber were altered (Fig- ure 4.7A, B). The walls were changed to display various textures and patterns. The floor of the chamber was replaced with a soft, solid foam flooring. Almond extract olfactory cues were distributed throughout the chamber, and the position of the light was changed compared to Day 1. Mice were left in the chamber for 360 sec. After 150 sec, the hiss sound from Day 1 was played for 30 sec. Freezing was used as an index of conditioned fear. Total test time was divided into 30 sec bins. Fear to the novel environment was measured by analyzing the fold-increase in the percentage of time spent freezing on Day 2 versus Day 1 during the first 150 sec (pre-cue). Fear to the conditioned stimulus was measured by analyzing the fold-increase in the percentage of time spent freezing on Day 2 post-cue vs. Day 2 pre-cue. 4.2.11 Statistics All results are expressed as means ±sem. Two-way ANOVAs with Bonferroni com- parisons were used to determine whether differences existed based on experience level. One-way ANOVAs with Bonferroni comparisons were performed for each age group to examine between strain differences. Statistical significance was set at p < 0.05. Specific details on the statistical analyses used for each test can be found in the figure captions. If data displayed a high level of variability, a test for outliers was performed. Mild outliers were defined as values lying between 1.5 and 3 times the interquartile 90 4.3. Results range below the 1st quartile or above the 3rd quartile, while extreme outliers were defined as values lying more than 3 times beyond this range. 4.3 Results 4.3.1 S2P23 appear grossly normal Similar to previous descriptions of S2P23 mice by Denzel et al. (2000), the mice used in this study appeared grossly normal. Overall, most mice remained healthy for the 12 months of the study. No obvious differences were observed during animal husbandry in regards to animals suffering from dermatitis (2 C57BL/6 and 1 S2P23) or animals suffering from serious illnesses which resulted in sacrifice or death (2 C57BL/6 and 3 S2P23). S2P23 mice grew as well as control mice in the first 6 months of life, and by 12 months were slightly heavier than controls (12m males: 51.8 ±0.79g vs. 48.6 ±1.00g, p < 0.01; 12m females: 40.0 ±1.03g vs. 36.7 ±1.44g, p < 0.05; means ±sem; Figure 4.8). 4.3.2 With increasing age, S2P23 and control mice lose hang strength equally The hanging wire test was used as a general measure of motor function. To evaluate whether experience significantly influenced results, mice were grouped by experi- ence level and each age group was subjected to two-way ANOVA with Bonferroni correction. Prior experience did not significantly impact results, thus näıve and experienced mice were combined. With increasing age, S2P23 and control mice lost hang strength equally (Figure 4.9). 91 4.3. Results Figure 4.8: S2P23 mice weigh more than controls at 12 months of age. Mice were weighed at 3 months (3m), 6 months (6m), and 12 months (12m) of age. Two-way ANOVA with Bonferroni correction found no significant differences at 3m or 6m of age; however, by 12m of age both male and female S2P23 mice weighed significantly more than controls (*p < 0.05; **p < 0.01; data shown as means ±sem). 92 4.3. Results Figure 4.9: With increasing age, S2P23 and C57BL/6 mice lose hang strength equally. Hanging wire test. Data shown as means ±sem. (A) Latency to fall. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. Previous experience did not affect test performance at any age, thus näıve and experienced mice were combined. No significant differences in latency to fall were seen at any age (p > 0.05). (B) Strength loss with age. Average latency to fall was plotted longitudinally. While male and female mice performed differently, within each sex no differences were seen between C57BL/6 and S2P23 mice over time. 93 4.3. Results 4.3.3 Three month old S2P23 mice have better rotarod performance than controls The rotarod test was used as a general measure of balance and motor coordination. To evaluate whether experience significantly influenced results, mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. Prior experience only significantly affected performance in 6m old female S2P23 (p < 0.05), thus näıve and experienced mice were only examined independently at the 6m time point. At 3m of age, female S2P23 mice had better motor coordination than controls, as measured by increased latency to fall from the rotarod (Figure 4.10). No significant differences were found between strains at 6m or 12m of age (p > 0.05). 4.3.4 Twelve month old S2P23 males travel more distance than controls, as a result of moving faster General locomotor activity was examined by looking at total distance travelled and average speed in the open field test. To evaluate whether experience significantly influenced results, mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. At 6m of age, total distance travelled by both controls and S2P23 was negatively affected by experience, thus these groups were analysed independently. Similarly, S2P23 males with prior experience travelled less distance at 12m (Figure 4.11A). When experience was accounted for, no significant differences were seen between strains at 6m; however, by 12m, less experienced S2P23 males travelled a further distance than controls (p < 0.05). The differences seen in distance travelled were a result of differences in average speed. Mice which travelled shorter distances were slower moving, while mice which 94 4.3. Results Figure 4.10: Three month old S2P23 mice have better rotarod perfor- mance than controls. Data shown as means ±sem. Mice were grouped by expe- rience level and each age group was subjected to two-way ANOVA with Bonferroni correction. Prior experience only significantly affected performance in 6m old female S2P23 (#p < 0.05), thus näıve and experienced mice were only examined indepen- dently at the 6m time point. At 3m of age, female S2P23 mice had better motor coordination than controls, as measured by increased latency to fall from the ro- tarod (*p < 0.05). No significant differences were found between strains at 6m or 12m of age (p > 0.05). 95 4.3. Results Figure 4.11: Twelve month old S2P23 males travel further and faster than controls. Open field. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction (*p < 0.05, #p < 0.05, ##p < 0.01, ###p < 0.001). (A) Total distance. All mice with previous experience travelled less distance at 6m of age. At 12m of age, previous experience only affected S2P23 males. Thus, both näıve and experienced mice were analysed independently. When experience was accounted for, no significant differences were seen between strains at 6m; however, by 12m, less experienced S2P23 males travelled more distance than controls. (B) Average speed. All mice with previous experience exhibited slower speed at 6m of age. At 12m of age, previous experience only slowed S2P23 males. When experience was accounted for, no significant differences were seen between strains at 6m; however, by 12m, less experienced S2P23 males moved faster than controls. 96 4.3. Results travelled longer distances were faster (Figure 4.11B). Thus, as land-distance travelled was affected by physical locomotor differences between the groups, it was excluded as a viable measure in all future tests. 4.3.5 S2P23 mice show heightened anxiety in the open field test The open field test was used to measure anxiety in C57BL/6 and S2P23 mice. Decreased rearing or interaction with the Inner zone was taken as a measure of heightened anxiety, as described on page 75. To evaluate whether experience signifi- cantly influenced results, mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. Rearing behaviour was examined using the latency to first rear and the number of rears (Figure 4.12). Due to high variability, latency to first rear data were analysed for outliers, resulting in a 6.3% discard rate. There were 7 mild outliers (1 6m- F-S2P23-Näıve, 1 6m-F-S2P23-Exp, 1 6m-M-C57BL/6-Exp, 2 6m-M-S2P23-Exp, 1 12m-F-C57BL/6-Näıve, 1 12m-M-C57BL/6-Exp ) and 7 extreme outliers (1 3m- F-C57BL/6, 1 6m-M-S2P23-Exp, 1 12m-F-S2P23-Näıve, 1 12m-M-C57BL/6-Näıve, 1 12m-F-C57BL/6-Exp, 1 12m-M-C57BL/6-Exp, 1 12m-M-S2P23-Exp) out of 221 data points. For both the 6m and 12m age group, male C57BL/6 mice with greater experience had a significantly shorter latency to rear (p < 0.05), thus mice with different experience levels were examined independently. Näıve, 6m S2P23 males had a significantly increased latency to first rear than controls (p < 0.01), while no differences were seen at 12m. When the number of rears was examined, experience level had a significant effect at 6m of age (p < 0.05), thus näıve and experienced mice were analysed independently; experience did not affect mice at 12m of age, thus these groups were combined (p > 0.05). No significant differences in the number of rears were found 97 4.3. Results Figure 4.12: C57BL/6 and S2P23 mice show different rearing tendencies. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. (A) Latency to first rear. For both the 6m and 12m age groups, male S2P23 mice with greater experience had a significantly shorter latency to rear (#p < 0.05), thus mice with different experience levels were examined independently. Näıve, 6m S2P23 males had a significantly increased latency to first rear than controls (**p < 0.01). (B) Number of rears. Increased experience led to a decreased number of rears in all groups at 6m, thus 6m näıve and experienced mice were analysed independently (#p < 0.05, 6m). No differences were seen by 12m of age, so these groups were combined for further analysis (p > 0.05). No significant differences were found between C57BL/6 and S2P23 mice at any age group. 98 4.3. Results between C57BL/6 and S2P23 mice in any age group (Figure 4.12). Level of interaction with the Inner zone was measured using three variables: latency to enter the Inner zone, number of entries into the Inner zone, and time spent in the Inner zone. Due to high variability, latency to first enter the Inner zone data were analysed for outliers, resulting in a 6.8% discard rate. There were 10 mild outliers (1 3m-F-S2P23, 2 3m-M-S2P23, 1 6m-M-C57BL/6-Exp, 2 6m-M-S2P23- Exp, 2 12m-M-C57BL/6-Exp, 2 12m-M-S2P23-Exp) and 5 extreme outliers (1 3m- F-S2P23, 1 6m-F-S2P23-Exp, 1 12m-M-C57BL/6-Näıve, 1 12m-F-C57BL/6-Exp, 1 12m-M-S2P23-Exp) out of 219 data points. Six month S2P23 males with prior testing experience waited significantly longer before entering the Inner zone (p < 0.05, Figure 4.13), thus näıve and experienced mice were analysed independently. Prior experience did not affect performance at 12m of age (p > 0.05). Only 6m S2P23 males with prior experience waited longer than controls to enter the Inner zone (p < 0.01). When time spent in the Inner zone was analysed, if was also shown that expe- rience only affected S2P23 mice, with experienced mice showing heightened anxiety when compared to näıve mice, as measured by less time spent in the Inner zone (p < 0.05, 6m females; p < 0.01, 6m and 12m males; Figure 4.14). Since differ- ences were found, näıve and experienced mice were analysed independently. Both male and female 6 month old S2P23 mice showed heightened anxiety compared to controls, as measured by spending less time in the Inner zone, only when they had previous testing experience (p < 0.05, 6m females; p < 0.001, 6m males). 99 4.3. Results Figure 4.13: Six month old S2P23 interact with the Inner zone less than controls. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. (A) Latency to enter Inner zone. Six month S2P23 males with prior experience waited significantly longer before entering the Inner zone (#p < 0.05), thus näıve and experienced mice were analysed independently. Prior experience did not affect performance at 12m of age (p > 0.05). Only experienced male S2P23 mice waited significantly longer before entering the inner zone at 6 months (**p < 0.01), but not 12 months of age (p > 0.05). (B) Number of entries into Inner zone. Previous experience only affected S2P23 mice, with experienced mice showing fewer entries into the Inner zone (##p < 0.01, ###p < 0.001). Since differences were found, näıve and experienced mice were analysed independently. Only 6m experienced S2P23 males showed heightened anxiety compared to controls, as measured by a decreased number of entries into the Inner zone (**p < 0.01). 100 4.3. Results Figure 4.14: Experienced, 6 month old S2P23 mice spend less time in the open field Inner zone. Time spent in Inner zone. To evaluate whether experience significantly influenced results, mice were grouped by experience level and sex/strain, and each age group was subjected to two-way ANOVA with Bonferroni correction. Previous experience only affected S2P23 mice, with experienced mice showing heightened anxiety when compared to näıve mice, as measured by less time spent in the Inner zone (#p < 0.05, 6m females; ##p < 0.01, 6m and 12m males). Since differences were found, näıve and experienced mice were analysed independently. Six month old S2P23 mice with prior experience showed heightened anxiety compared to controls, as measured by spending less time in the Inner zone, only when they had previous testing experience (*p < 0.05, 6m females; ***p < 0.001, 6m males; data shown as means ±sem). 101 4.3. Results 4.3.6 Six month old S2P23 males show heightened anxiety in the light-dark box test The light-dark box test was used to measure rodent anxiety, as defined on page 76. Due to variability, data regarding latency to enter the Light side were subjected to outlier detection. Of 208 data points, there were 6 mild outliers (1 3m-F-S2P23, 3 6m-M-S2P23-Exp. 1 12m-F-S2P23-Exp, 1 12m-M-C57BL/6-Exp) and 8 extreme outliers (1 3m-F-C57BL/6, 1 3m-M-S2P23, 1 6m-F-S2P23-Exp, 2 6m-M-C57BL/6- Exp, 1 6m-M-S2P23-Exp, 2 12m-M-S2P23-Exp), for a discard rate of 6.7%. Prior experience affected 6m S2P23 males, thus näıve and experienced mice were analysed independently (p < 0.01). By 12m of age, experience no longer affected performance, so mice of differing experience levels were grouped (p > 0.05). At 6m of age, näıve S2P23 males waited significantly longer before entering the Light side, and all S2P23 males spent significantly less time there overall (Figure 4.15), despite making the same number of entries to the Light side as controls (Figure 4.16). 4.3.7 S2P23 and control mice perform a similar number of spontaneous alternations The Y-maze was used to examine spatial working memory. As general locomotor activity was previously shown to differ across groups, total distance travelled and average speed was examined. When prior experience was taken into account, no differences in total distance travelled or average speed were found (Figure 4.17). Thus these factors were unlikely to confound the Y-maze results. The percentage of spontaneous alternations, as calculated by the number of alternations performed out of all alternations possible with a given number of arm entries, was the unit of measure. Prior experience had no effect on performance, as determined by a two-way ANOVA with Bonferroni correction (p > 0.05), thus näıve 102 4.3. Results Figure 4.15: Six month S2P23 males hesitate longer before entering the Light side and spend less time there overall. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two- way ANOVA with Bonferroni correction. (A) Latency to enter Light side. Prior experience affected 6m S2P23 males, thus näıve and experienced mice were analysed independently (##p < 0.01). At 6m of age, naive S2P23 males waited significantly longer before entering the Light side (**p < 0.01). (B) Total time spent in Light side. Prior experience did not affect test performance at any age, thus näıve and experienced mice were combined (p > 0.05). At 6m of age, male S2P23 mice spent significantly less time in the Light side (*p < 0.01). 103 4.3. Results Figure 4.16: S2P23 and control mice make a similar number of entries to the Light side. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni cor- rection. Prior experience did not affect test performance at any age, thus näıve and experienced mice were combined (p > 0.05). C57BL/6 and S2P23 mice made the same number of entries to the Light side (p > 0.05). 104 4.3. Results Figure 4.17: All mice have the same level of general locomotor activity in the Y-maze. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. (A) Average speed: Y-maze. While at 6m of age, experienced female C57BL/6 and male S2P23 mice moved slower (#p < 0.05), the effect of prior experience disappeared by 12m of age. Thus, näıve and experienced mice were analysed inde- pendently at 6m, and together at 12m of age. No difference in average speed were found between strains at 6m or 12m of age (p > 0.05). (B) Total distance: Y-maze. While at 6m of age, experienced female C57BL/6 and male S2P23 mice travelled less distance (#p < 0.05), the effect of prior experience disappeared by 12m of age. Thus, näıve and experienced mice were analysed independently at 6m, and together at 12m of age. No difference in total distance travelled were found between strains at 6m or 12m of age (p > 0.05). 105 4.3. Results and experienced mice were combined. Heterozygous knockdown of TMP21 had no effect on the percentage of spontaneous alternations for any age group (Figure 4.18). Figure 4.18: All mice perform a similar number of spontaneous alter- nations in the Y-maze. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bon- ferroni correction. Prior experience did not affect performance at any age (p > 0.05), thus näıve and experienced mice were combined. No differences in spontaneous al- ternation behaviour was seen between C57BL/6 and S2P23 mice (p > 0.05). 4.3.8 S2P23 mice do not show deficiencies in spatial memory The Morris water maze was used to examine potential differences in spatial learning and memory between C57BL/6 and S2P23 mice. The Day 1 visible platform test was used to gage differences in swimming ability and vision. The hidden platform tests on Days 2 through 5 examined task acquisition, while the probe trial on Day 6 examined recall. Previous experience in the water maze test affected visible platform testing of 106 4.3. Results all six month old female mice (Figure 4.19). Therefore, to avoid confounding factors due to differential experience levels, näıve and experienced mice were analysed as separate groups for all days of the experiment, for all ages. No difference were found between C57BL/6 and S2P23 mice in the visible platform condition. Task acquisition was examined over Days 2 to 5 in the hidden platform condition. No significant differences were found in regards to path length (Figure 4.20) or escape latency (Figure 4.21) in either the less experienced or more experienced mice. Furthermore, both S2P23 and control mice showed a similar performance in the probe trial on Day 6 (Figure 4.22). 4.3.9 S2P23 mice show deficits in the contextual fear conditioning task The percentage of time spent freezing on Day 1 was taken as a general measure of anxiety to the fear conditioning apparatus. As the data displayed much variabil- ity, a test for outliers was performed as previously described (section 4.2.11). For the percentage of time spent freezing on Day 1, 5 mild outliers (1 3m-F-S2P23, 1 3m-M-S2P23, 1 6m-F-S2P23-Exp, 1 6m-M-S2P23-Exp, 1 12m-F-S2P23-Näıve) and 1 extreme outlier (3m-F-C57BL/6) were found out of 212 data points, for a discard rate of 2.8%. For fold increase in fear, 6 mild outliers (1 3m-F-S2P23, 1 6m-M- C57BL/6-Näıve, 1 6m-M-S2P23-Exp, 1 12m-M-S2P23-Näıve, 2 12m-M-S2P23-Exp) and 4 extreme outliers (1 6m-F-S2P23-Exp, 1 6m-M-S2P23-Exp, 1 12m-F-S2P23- Näıve, 1 12m-F-S2P23-Exp) were found out of 212 data points, for a discard rate of 4.7%. At 3m of age, näıve S2P23 males froze significantly longer than controls on Day 1 (p > 0.05). At 6m and 12m of age, previous experience caused a signif- icant increase in freezing for all mice except 6m S2P23 females and 12m females (Figure 4.23A). This indicated that at 6m of age, mice could recall their previous 107 4.3. Results Figure 4.19: S2P23 and C57BL/6 mice perform similarly in the water maze visible platform condition. (A) Path Length on Day 1. (B) Latency on Day 1. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. Previous experience resulted in a significantly shorter path length (A) and escape latency (B) in all females at 6m of age (###p < 0.001, #p < 0.05), thus näıve and experienced mice were analysed independently for all ages. No significant differences were found in visible platform performance between strains (p > 0.05). 108 4.3. Results Figure 4.20: S2P23 and C57BL/6 mice have similar path lengths in wa- ter maze hidden platform trials. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. No significant differences were seen between S2P23 and C57BL/6 mice, in either the less experienced (A) or more experienced (B) conditions (p > 0.05). 109 4.3. Results Figure 4.21: S2P23 and C57BL/6 mice have similar escape latencies in water maze hidden platform trials. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. No significant differences were seen between S2P23 and C57BL/6 mice, in either the less experienced (A) or more experienced (B) conditions (p > 0.05). 110 4.3. Results Figure 4.22: S2P23 and C57BL/6 mice spend a similar amount of time in the platform quadrant in the probe trial. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction. No significant differences were seen between S2P23 and C57BL/6 mice, in either the less experienced (green bars) or more expe- rienced (gray bars) conditions (p > 0.05). 111 4.3. Results test 3 months prior; however, by 12m of age, only male mice were affected by their experience 6 months prior, suggesting that male mice have a more robust very-long term recall of aversive experiences. The fold increase in the percentage of time spent freezing on Day 2 vs. Day 1 was used as a measure of conditioned fear. Three month old, male S2P23 mice showed a decreased conditioned response (p < 0.01; Figure 4.23B); however, since 3m S2P23 males also showed heightened anxiety on Day 1, which served as the denominator in the calculation of conditioned fear, the results are confounded. No such confounding factors exist for the 12 month data, however. Less experienced, twelve month old female S2P23 mice had equivalent reactions to the apparatus on Day 1, but on Day 2, S2P23 females showed a decreased conditioned response (p < 0.001), suggesting a memory deficit may exist for S2P23 mice in regards to conditioned fear. 4.3.10 S2P23 females show heightened anxiety to novel environments at 12 months of age The cued fear conditioning paradigm measured two properties: fear of a novel en- vironment, and memory retention of a conditioned stimulus. Anxiety to the novel environment was examined by comparing the percentage of time spent freezing be- fore the cue on Day 2, with the same measures on Day 1, pre-cue. Memory retention of the conditioned stimulus was examined by examining the fold increase in the per- centage of time spent freezing on Day 2 post-cue vs. Day 2 pre-cue. As the data displayed variability, a test for outliers was performed as previously described (section 4.2.11). No outliers were found for pre-cue measures, while 1 mild outlier (12m-M-S2P23) was found out of 55 data points for the fold increase in fear on Day 2, for a discard rate of 1.8%. Three month and twelve month data were analysed by one-way ANOVA, while 6m data was analysed by Student’s t-test. 112 4.3. Results Figure 4.23: S2P23 have a decreased conditioned fear response. Data shown as means ±sem. Mice were grouped by experience level and each age group was subjected to two-way ANOVA with Bonferroni correction (#p < 0.01, ##p < 0.01, ###p < 0.001). (A) Percent freezing on Day 1. Previous experience caused a significant increase in freezing for all mice except 6m S2P23 females and 12m females, thus näıve and experienced mice were analysed independently. Male S2P23 mice froze more than controls at 3m (*p < 0.05), but not 6m or 12m of age (p > 0.05). (B) Fold increase in freezing. Previous experience decreased the conditioned response in all mice except 12m S2P23 females, thus näıve and experienced mice were analysed independently. At 3m of age, S2P23 males showed a decreased conditioned response compared to controls (**p < 0.01); however, this is confounded by the increased freezing these mice displayed on Day 1. At 12m of age, less experienced S2P23 females showed a decreased conditioned response compared to controls (***p < 0.001). 113 4.4. Discussion By 12m of age, S2P23 females showed greater fear of the Novel environment than controls (Figure 4.24A). No differences were seen between S2P23 and controls in regards to memory of the conditioned stimulus (Figure 4.24B). 4.4 Discussion 4.4.1 Physical abilities S2P23 mice appeared grossly normal over the course of the study and gained weight at a satisfactory rate throughout their lifetime; by 12 months of age, S2P23 weighed an average of 6 to 8% more than controls when fed ad. lib. (section 4.3.1). Despite being heavier, S2P23 mice had a similar hang time as controls, suggesting motor strength was not hindered (section 4.3.2). S2P23 mice also displayed age-dependent motor advantages over wildtype mice. Three month old, female S2P23 mice had better motor coordination as determined by the rotarod test, an effect which disappeared with age (section 4.3.3). By 12 months of age S2P23 mice were also faster moving than controls, with less expe- rienced males having a higher average speed in the open field test of spontaneous motor activity, and as a result travelled more distance (section 4.3.4). Thus, as land-distance travelled was affected by physically locomotor differences between the groups, it was excluded as a viable measure in the remaining tests. 4.4.2 Anxiety Effect of experience within strains Both the open field test and light-dark box test were used to assess anxiety in the S2P23 mouse. Interestingly, prior experience with behaviour testing significantly affected mouse performance in the open field test in a strain-specific manner: S2P23 114 4.4. Discussion Figure 4.24: S2P23 females show heightened anxiety to novel environ- ments at 12 months of age. Data shown as means ±sem. Three month and twelve month data were analysed by one-way ANOVA, while 6m data were anal- ysed by Student’s t-test. (A) Difference in percentage of time spent freezing pre-cue. By 12m of age, S2P23 females showed greater fear of the Novel environment than controls (***p < 0.001). (B) Memory retention of conditioned stimulus. No signifi- cant differences were seen for any age group (p > 0.05). 115 4.4. Discussion mice with previous behaviour testing experience showed higher anxiety levels than inexperienced S2P23 mice, as seen in increased latencies to enter the Inner zone, less entries into the Inner zone, and decreased time spent in the Inner zone overall (section 4.3.5), while wildtype mice were unaffected by previous experience. This suggests that S2P23 mice may have better very long term recall than wildtype mice, which affected their performance. Conversely, of the three anxiety measures used in the light-dark box test, two (number of entries to the Light side and total time spent there) were not affected by experience, while one (latency to enter the Light side) showed that S2P23 mice with prior experience had less anxiety than näıve S2P23 mice, as shown by decreased latencies (section 4.3.6). Furthermore, latency to rear in the open field test showed a similar opposing trend. One should note, however, that latency measures could depend on the general level of locomotor activity, and S2P23 mice were shown to move faster than controls, which could potentially result in the decreased latencies shown in these tests. Overall, when comparing anxiety measures within strains with varying experi- ence, there is evidence to suggest that S2P23 may possess a higher level of long-term recall which could affect their anxiety-like behaviours, as C57BL/6 mice with previ- ous experience did not show a similar trend. Further anxiety-specific tests, such as the elevated plus maze, could help resolve the suspected conflicts between the open field and light-dark box data, and potentially support the open field findings. Differences in behaviour between strains When S2P23 mice were compared to controls, S2P23 mice displayed increased anx- iety, as defined in section 4.1.3. Näıve, 6 month old S2P23 males had a significantly increased latency to first rear than näıve controls in the open field test, although 116 4.4. Discussion there were no differences in the number of rears. At 6 months of age, S2P23 mice also waited longer before entering the Inner zone, made fewer total entries, and spent less time there overall, all indicators of heightened anxiety. Results of the light-dark box agreed with the open field findings. Six month old S2P23 males showed heightened anxiety, as measured by an increased latency to enter the Light side and less total time spent there; however, no differences were seen in the total number of entries to the Light side. These findings suggest that TMP21 plays a significant role in mouse behaviour, specifically in terms of anxiety. 4.4.3 Memory While heterozygous knockdown of TMP21 did not significantly alter performance in hippocampal-dependent learning and memory tasks such as the y-maze task (sec- tion 4.3.7) or the Morris water maze task (section 4.3.8), differences in performance were seen in tasks dependent on both the hippocampus and the amygdala. Effect of experience within strains Regardless of strain, the majority of mice with previous experience in the fear con- ditioning apparatus showed heightened levels of anxiety on Day 1, as shown by increased freezing behaviour, as well as a suppressed fear conditioning response on Day 2, suggesting that mice can recall previous aversive events up to 3 months prior (section 4.3.9). The effect of re-testing in mice who were only tested at 6 months and 12 months of age, as opposed to 3 months, 6 months, and 12 months of age, was more muted (Figure 4.23A, 12 months, green bars). 117 4.4. Discussion Differences in behaviour between strains Three month old, S2P23 males showed a decreased contextually-conditioned re- sponse; however this was confounded by heightened anxiety on Day 1, which served as the denominator in the calculation of conditioned fear (section 4.3.9). The data at 12 months of age were not confounded, as twelve month old S2P23 females had equivalent reactions to the apparatus on Day 1, but showed a decreased contextually- conditioned response on Day 2, suggesting a memory deficit may exist for S2P23 mice in regards to contextually-conditioned fear. All mice used in the cued fear conditioning task had received no prior behavioural testing. Anxiety to the novel environment was examined by comparing the percent- age of time spent freezing before the cue on Day 2 with the same measures on Day 1, pre-cue. Twelve month old S2P23 females showed greater fear of the Novel envi- ronment than controls. When cue-conditioned fear was analysed by examining the fold increase in the percentage of time spent freezing on Day 2 post-cue vs. Day 2 pre-cue, no differences were seen between strains (section 4.3.10). The combined results of the contextual and cued fear conditioning paradigms must be interpreted with consideration of the heightened level of anxiety displayed by S2P23 mice, as differences in anxiety to the novel environment in the cued fear conditioning task agree with an overall increase in anxiety in these mice. Meanwhile, the lack of differences in cued-conditioned fear between S2P23 mice and controls, combined with the deficits in contextually-conditioned fear in S2P23 mice, suggest a stronger hippocampal than amygdalar involvement. This agrees well with the expression level of TMP21, which is higher in the hippocampus (Vetrivel et al., 2008a). A genetic relationship exists between contextual fear conditioning performance and heightened anxiety in mice, such that mice bred for better performance are 118 4.4. Discussion also more anxious (Ponder et al., 2007). While S2P23 mice in this study show heightened anxiety but a decreased contextually-conditioned response to fear, it is possible that contextual fear conditioning may have detected potential hippocampal deficits where Y-maze and water maze tests could not, as it is a more aversive test which may have been augmented by a natural sensitivity of S2P23 mice to anxious stimuli. 4.4.4 Conclusion Overall, S2P23 mice show slightly enhanced physical abilities, increased anxiety, and potentially anxiety-augmented deficits in hippocampal learning and memory. As TMP21 is involved in the trafficking of several proteins, it is not surprising that suppression of TMP21 would have behavioural consequences. Should these mice be crossed with AD model mice to determine the in vivo effect of TMP21 suppression on AD pathogenesis and behavioural outcomes, the behavioural characteristics of the S2P23 strain should be carefully considered. Sev- eral groups use the Morris water maze to evaluate AD model mice, and in this case the effect of TMP21 suppression alone should not confound results. As homozygous TMP21 knockouts are embryonic lethal, the ideal mouse model of TMP21 suppression in AD would be an inducible hippocampal knockout crossed with an AD model mouse. Ms. Xiaojie Zhang from our laboratory is currently engineering an inducible TMP21 knockout mouse which could be used for this ap- plication. 119 Chapter 5 Conclusions and future directions 5.1 Overall conclusions The overall goals of this thesis were to further characterize TMP21 biochemically through an investigation of its degradation, to expand our understanding of TMP21’s role in AD pathogenesis, and to understand how TMP21 contributes to mouse be- haviour as a guideline for future work in a TMP21-heterozygous × AD model mouse cross. Similar to other members of the g-secretase complex, including PS1 (Fraser et al., 1998; Steiner et al., 1998; Honda et al., 1999), Nct (He et al., 2007), Pen-2 (Bergman et al., 2004; Crystal et al., 2004), and Aph-1 (He et al., 2006), TMP21 was found to be degraded by the ubiquitin-proteasome pathway. A novel technique was used to visualize TMP21 degradation in situ using protein inhibitor CHX and the in-cell western protocol, and it was determined that TMP21 has a half-life of approximately 3 h, similar to the half-life of g-secretase component Pen-2 (Bergman et al., 2004). In examining the effect of TMP21 on AD pathogenesis, this thesis largely focused on the consequences of TMP21 overexpression. While several groups have exam- ined how TMP21 suppression increases Ab production (Chen et al., 2006; Vetrivel et al., 2007; Dolcini et al., 2008), the effect of TMP21 overexpression on APP CTF production has received significantly less attention (Pardossi-Piquard et al., 2009). 120 5.1. Overall conclusions In examining CTFs, this thesis made the novel discovery that in addition to its role in modulating g-secretase activity and affecting APP trafficking (Chen et al., 2006; Vetrivel et al., 2007), TMP21 may also affect BACE1 processing of APP. Overex- pression of TMP21 decreased C83 and increased C99 production, suggesting over- expression shifts APP processing from the non-amyloidogenic pathway dominated by a-secretase, to the pathogenic, amyloidogenic pathway dominated by b-secretase. Furthermore, this thesis showed that BACE1 and TMP21 can coimmunoprecipi- tate when TMP21 is overexpressed. Overall, this suggests that TMP21’s role in AD pathogenesis is multi-faceted, and that its effects on BACE1 can offer a new, as-of-yet unexplored mechanism of action. Finally, the behavioural consequences of TMP21 suppression were examined in preparation for future research using an in vivo mouse model of TMP21 suppression in AD. While knockdown of TMP21 increases Ab production in vitro (Chen et al., 2006) and human AD patients have less TMP21 protein in their brain (Vetrivel et al., 2008a), this is insufficient evidence to suggest that decreased levels of TMP21 contributes to AD pathogenesis; we can only say these events are correlated. More concrete evidence can be obtained from demonstrating that TMP21 suppression exacerbates AD pathology in an AD model mouse; however, without understanding how TMP21 suppression affects mouse behaviour in general, we would be unable to interpret data resulting from a double cross. This thesis was the first work which examined the behavioural consequences of TMP21 suppression. Here it is shown that S2P23 mice have slightly enhanced physical abilities, increased anxiety, and potentially anxiety-augmented deficits in hippocampal learning and memory. As TMP21 is involved in the trafficking of several proteins, it is not surprising that suppression of TMP21 would have be- havioural consequences. Fortunately, performance of S2P23 mice in the hidden and 121 5.2. Significance of the research probe conditions of the Morris water maze task, a task often used to assess learn- ing and memory in AD model mice, was similar to controls, thus this test would prove valuable in evaluating the consequences of a double cross (see section 5.4.3 for preliminary work in this area). 5.2 Significance of the research This thesis presented several novel findings which could significantly impact the field of AD research, including: 1. A novel method of visualizing protein degradation in situ through the combi- nation of the in-cell western assay and CHX protein inhibition. 2. Evidence that TMP21 overexpression is relevant in the context of AD patho- genesis, contrary to previous findings (Chen et al., 2006). 3. Evidence that TMP21 may not only affect AD pathogenesis through its mod- ulatory role on g-secretase or its trafficking of APP, but also through its in- fluence on BACE1, providing a new enzymatic target for future study. 4. Behavioural characterization of TMP21 in a mouse model, the first time the in vivo consequences of TMP21 suppression have been observed. 5.3 Strengths and weaknesses 5.3.1 Chapter 2: Degradation of TMP21 This chapter presented a novel method of visualizing protein half-life in vitro using an in-cell western assay with protein inhibitor CHX. The traditional method of determining protein half-life is the pulse-chase experiment, where cells are pulsed 122 5.3. Strengths and weaknesses with 35S-Met and analysed at various time points thereafter to examine protein degradation. CHX takes a different yet complementary approach, in that it halts all protein synthesis and protein degradation is measured at different time intervals following CHX exposure. This method has been successfully employed in numerous studies (Schoenfeld et al., 2000; Touitou et al., 2001; Liu et al., 2004), including determination of the half-life of Pen-2 (Bergman et al., 2004), and offers several advantages such as an increased safety profile and the ability to combine CHX with other established protocols, such as the in-cell western used here. A disadvantage of this technique is that it is based on the assumption that protein synthesis is completely inhibited, while pulse-chase experiments offer a firmly defined labelling event from which to proceed. One of the strengths of the pharmacological treatments selected to study protea- somal and lysosomal degradation was their varied methods of action. For the lysoso- mal study, chloroquine inhibited degradation by raising lysosomal pH and interfering with receptor recycling (Ohkuma and Poole, 1978; Gonzalez-Noriega et al., 1980), while NH4Cl inhibited phagosome-lysosome fusion (Gordon et al., 1980; Amenta and Brocher, 1980). For the proteasome study, ALLN and MG-132 were peptide aldehydes which inhibited proteasome-dependent protein degradation, while ALLM, a much weaker proteasome inhibitor which does not show inhibition at 25 mM, pro- vided a peptide aldehyde control (Rock et al., 1994; Jensen et al., 1995; Ward et al., 1995). Finally, b-lactone acted as an irreversible inhibitor of the 20S and 26S pro- teasome (Omura et al., 1991; Ward et al., 1995; Fenteany et al., 1995). By utilizing several inhibitors and observing the same outcome, one can confidently conclude that the result is specific to lysosomal or proteasomal degradation, rather than a coincidence deriving from the dirty nature of a drug. Similarly, the subcellular fractionation work shown in this chapter complimented 123 5.3. Strengths and weaknesses the immunocytochemical data, again providing multiple viewpoints of the same phenomenon. While Figure 2.6 may have benefited from supporting fractionation evidence, the overwhelming support from the rest of the chapter suggests this would be redundant, and thus it was omitted. 5.3.2 Chapter 3: TMP21’s effect on APP processing The potential involvement of TMP21 in the altered processing of APP by BACE1 presents an exciting and novel finding which could be explored for several more years if time allowed (see section 5.4.1). While several replications showed that overexpression of APP with or without BACE1 in HTM2 cells altered APP processing, these results could not be repeated in APP or APP+BACE1 stable cells. The argument could be raised that TMP21 stable cells have a high level of TMP21 expression in every cell, such that when APP and/or BACE1 are expressed transiently, TMP21 is guaranteed to be overexpressed simultaneously. Conversely, in APP or APP+BACE1 stable cells, while every cell should possess APP or APP and BACE1, the transient transfection efficiency of TMP21 may not increase TMP21 to a sufficient level to alter APP processing; however, repeating the experiment with Lipofectamine TM 2000 transfection reagent obtained similar results. On the other hand, the above theory does not explain why transient overex- pression of all three constructs successfully replicated the HTM2 cell results. While HTM2 cells stably express TMP21-myc, the triple transient transfections used un- tagged TMP21. This suggests that the results are not specific to TMP21 stable cells, but due to overexpression of TMP21 itself. This evidence, combined with the large number of replicates, lends confidence to the findings. Examining protein-protein interactions with a dot-blot peptide array should not 124 5.3. Strengths and weaknesses be the first choice for mapping protein interaction sites, as it is a very artificial system. In this work, it was employed only after previous studies observed an inter- action using more traditional methods such as coimmunoprecipitation. As protein- protein interactions are often dependent on protein conformation, and the dot-blot array consists of bare 20-mer peptides, the likelihood of success is low; however, as in the case here, it may reveal insights which could be further explored at a later point. This chapter provides the foundation for further work in our laboratory on TMP21’s role in APP processing, and will be continued by Ms. Xiaojie Zhang. 5.3.3 Chapter 4: TMP21 in mouse behaviour One of the strengths of this chapter was that multiple tests were used to measure the same behaviour, thus ensuring the results were not test-specific. For example, motor function was examined with both the hanging wire and rotarod tests, anxiety was examined with the open field and light-dark box tests, and learning and memory were examined with Y-maze, water maze, and fear conditioning tests. A goal of this study was to examine behaviour through different stages of mouse maturity. Mouse ages can be converted to their human counterpart on the ratio of 10 mouse days for 1 human year (JaxLabs, 2009). Mice in this study were examined in youth (3 months; 9 human years), adulthood (6 months; 18 human years), and mid-life (12 months; 37 human years). By choosing multiple time points we were able to examine age-related changes in behaviour. In consideration of animal housing limitations as well as operating costs, it was decided to re-test mice at the chosen time points rather than maintain them from birth to a single test date. It was assumed that a 3 to 6 month time span between testing would not interfere with further results. As shown in this work, experience 125 5.4. Potential applications in future research did have an effect; however, this serendipitously enriched the data set as it would have been otherwise impossible to discover that S2P23 mice are affected by anxiety- inducing stimuli three months prior, while control animals are not, suggesting S2P23 mice have better very long-term recall of aversive events. While this turned out to be a benefit, splitting the data retroactively resulted in very small n numbers for some groups (see Appendix). In future experiments, it would be prudent to arrange for some mice to be re-tested, and others to be näıve, in an effort to maintain a sizable n as well as discover differences in very-long term recall. 5.4 Potential applications in future research 5.4.1 The role of TMP21 in substrate processing Effects on BACE1 This thesis has found that TMP21 overexpression decreases C83 levels, with a cor- responding increase in C99, results which are self-consistent and suggest BACE1 may be affected. An interaction was found between BACE1 and TMP21 by coim- munoprecipitation, and BACE1 may colocalize with TMP21 more when TMP21 is highly expressed (section 3.3.4). To fully explore these initial findings, subcellular fractionation is being performed to determine whether BACE1 localization changes upon TMP21 overexpression. Subcellular fractionation experiments should also be performed to confirm the im- munocytochemical findings of increased colocalization. If BACE1 localization is changing, live-cell imaging can be used to explore how TMP21 affects BACE1 traf- ficking in real-time. For a complete view of the data, Ab ELISA assays should also be used to de- termine whether the heightened levels of C99, or lowered levels of C83, lead to a 126 5.4. Potential applications in future research corresponding increase in Ab. Effects on APP processing TMP21 could potentially exert its influence on APP processing on three fronts: (1) by affecting g-secretase through modulation (Chen et al., 2006); (2) by affect- ing b-secretase through an unknown mechanism; or (3) by affecting the substrate. TMP21 and APP have the potential to colocalize, as both can be found on the plasma membrane (see sections 1.1.1 and 1.2.2). To address point (3), immunofluo- rescent labelling was performed on HEK cells transfected with expression plasmids of APPwt , APPSwe, C99, or C83, with TMP21-myc. TMP21 was shown to colocalize with APP (Figure 5.1). It has been previously shown that TMP21 suppression causes an increase in cell- surface APP (Vetrivel et al., 2007); however, no studies have examined the effect of TMP21 overexpression. It would be worthwhile to examine this side of the APP trafficking story through subcellular fractionations and membrane preparations of HEK and HTM2 cells. Effects on Notch processing Background In addition to APP, substrates for g-secretase include Notch (De Strooper et al., 1999; Song et al., 1999; Zhang et al., 2000), Jagged and Delta (Ikeuchi and Sisodia, 2003; LaVoie and Selkoe, 2003; Six et al., 2003), E-cadherin (Marambaud et al., 2002), N-cadherin (Marambaud et al., 2003), ErbB-4 (Ni et al., 2001), Nectin-1α (Kim et al., 2002), CD44 (Lammich et al., 2002), and LRP (May et al., 2002). After APP, Notch is the next most highly studied substrate. The Notch receptor begins as a 300 to 350 kDa single-pass transmembrane protein, which is cleaved at 127 5.4. Potential applications in future research Figure 5.1: TMP21 partially colocalizes with APP. HEK cells were trans- fected with expression plasmids of APPwt , APPSwe, C99, or C83, with TMP21- myc. After 48 to 72 h, cells were fixed in 4% PFA, permeabolized, blocked with 5% BSA, and incubated with or without primary antibodies 9E10 (mouse anti-myc) and C20 (rabbit anti-APP-CTF) overnight at 4◦C. The next day cells were rinsed and primary antibodies were detected by a 1 h incubation with goat anti-mouse Alexa Fluor 568 (red) and goat anti-rabbit Alexa Fluor 488 (green), rinsed, and mounted onto slides with Fluoromount-G. Cells were imaged with a 40× objective lens on a Carl Zeiss Axiovert-200 epifluorescent microscope. Images were assembled in Adobe Photoshop, cropped to equal sizes, and the relative brightness levels for each channel were adjusted to allow visualization of both antibodies in the same image. Slides on which the primary antibody was omitted were adjusted in the same way as slides with antibody. 128 5.4. Potential applications in future research S1 in the trans-Golgi network by a furin-like convertase prior to expression on the plasma membrane (Blaumueller et al., 1997; Logeat et al., 1998). This results in an extracellular N-terminal domain, and a C-terminal domain consisting of an ectoderm segment, the transmembrane domain, and the cytoplasmic domain (Figure 5.2). The N- and C-terminal domains are non-covalently linked via a calcium ion to form the functional, heterodimeric receptor (Rand et al., 2000). Following its expression on the plasma membrane and subsequent ligand binding, the Notch receptor is cleaved extracellularly at the S2 cleavage site, releasing the Notch extracellular domain (NECD) and leaving a membrane-bound fragment called the Notch extracellular truncation (NEXT) (Mumm et al., 2000). S2 cleavage is performed by Sup-17 in C. elegans (Wen et al., 1997) and ADAM-10 or ADAM-17 (also known as tissue necrosis factor a converting enzyme (TACE)) in mammals (Dallas et al., 1999; Brou et al., 2000; Hartmann et al., 2002). TACE C. elegans homolog Adm-4 later showed a redundant activity to Sup-17 (mammalian ADAM- 10), suggesting functional overlap in Notch S2 site cleavage (Jarriault, 2005). Following S2 cleavage, Notch undergoes intramembranous S3 cleavage by PS1- associated g-secretase, causing the release of NICD, which can translocate to the nucleus and interact with DNA binding proteins (Schroeter et al., 1998). This process is inhibited in PS1-deficient cells (De Strooper et al., 1999; Song et al., 1999), and both NICD and Ab generation are completely inhibited in PS1/2 double-KO cells, indicating that the PSs are absolutely required for both cleavages (Herreman et al., 2000; Zhang et al., 2000). Rationale Thus far, Notch cleavage has only been studied in the context of TMP21 suppression, where it was found that NICD production was unaffected (Chen et al., 2006). It was therefore assumed that TMP21 exerted its effects through se- lective modulation of g-secretase activity; however, just as it was later found that 129 5.4. Potential applications in future research Figure 5.2: Schematic of Notch processing. The extracellular domain of the Notch receptor contains 36 epidermal growth factor (EGF)-like repeats, with a ligand-binding (LB) EGF-like motif within this, followed by 3 lin-12/Notch repeats (LNR) which keep the extracellular and intracellular fragments together with the help of calcium following S1 cleavage. The intracellular domain contains a RBP- Jk association module (RAM) which allows NICD to interact with DNA-binding proteins, a series of CDC10/Ankyrin (ANK) repeats which also aids with protein interactions, nuclear localization signals (NLS), a transactivation domain (TAD), and a Pro-Glu-Ser-Thr (PEST) motif involved in degradation (Kopan, 2002; Selkoe and Kopan, 2003). S1 cleavage occurs in the trans-Golgi network by a furin-like convertase prior to expression on the plasma membrane. S2 cleavage occurs upon ligand binding, releasing NECD and leaving membrane-bound NEXT. S3 cleavage of NEXT by g-secretase releases NICD, which can translocate to the nucleus and interact with DNA binding proteins. 130 5.4. Potential applications in future research TMP21 also affects APP trafficking (Vetrivel et al., 2007), it may be that overex- pression could also affect Notch processing and/or trafficking. Preliminary data HTM2 cells were transfected with a myc-tagged expression plasmid of the S2-cleavage product of Notch, which can undergo g-secretase cleav- age, or negative calcium phosphate control. After 48 to 72 h, cells were harvested in PBS and lysed in NP-40 buffer. Lysates were immunoprecipitated with anti- TMP21 antibody T51 and analysed by immunoblot using anti-myc antibody 9E10. Notch coimmunoprecipitated with TMP21 when TMP21 was stably overexpressed (Figure 5.3). Figure 5.3: TMP21 coimmunoprecipitates with Notch HTM2 cells were transfected with a myc-tagged expression plasmid of the S2-cleavage product of Notch, which undergoes g-secretase cleavage, or negative calcium phosphate control. After 48 to 72 h, cells were harvested in PBS and lysed in NP-40 buffer. Lysates were immunoprecipitated with anti-TMP21 antibody T51 and analysed by immunoblot using anti-myc antibody 9E10. Notch coimmunoprecipitated with TMP21 when TMP21 was stably overexpressed. These data raise many questions: Will this interaction depend on the expres- sion level of TMP21, similar to the BACE1 results in Figure 3.6? Where is the 131 5.4. Potential applications in future research potential binding site? How does TMP21 overexpression affect Notch cleavage? Is Notch trafficking affected? These and other questions can be answered using sim- ilar techniques to those used in chapter 3, as well as the experiments proposed in section 5.4.1 on page 127. 5.4.2 Further exploration of the S2P23 mouse This thesis found that TMP21 heterozygous mice have heightened levels of anxiety, better very long-term recall of aversive events, and may have deficits in hippocampal- dependent fear conditioning. To determine whether the deficits in contextual fear conditioning are affected by heightened levels of anxiety in S2P23 mice, the procedure could be repeated with groups given either a placebo or an anti-anxiety drug. If S2P23 mice given the drug perform similarly to controls given the drug, one can suggest that heightened anxiety inhibited task performance, and that the deficit may not be hippocampal in nature. 5.4.3 Effect of TMP21 suppression in an in vivo model of Alzheimer’s disease Conveniently, the S2P23 mouse was created on the same C57BL/6 background as the most common AD model mouse, APP23. APP23 transgenic mice express the human APP751 isoform with the APPSwe mutation under control of a neuron- specific Thy-1 promoter fragment (Sturchler-Pierrat et al., 1997). The strongest expressing line expresses human APP 7× higher than endogenous mouse APP and displays plaque deposition by 6 months of age, with the majority of deposits located deep in the cortex and in the hippocampus (Sturchler-Pierrat et al., 1997). While plaques become visible by 6 months, heterozygotes show behavioural deficits in the 132 5.4. Potential applications in future research hidden platform condition of the Morris water maze task by 3 months of age, with progressive decline thereafter (Dam et al., 2003; Kelly et al., 2003). To examine the effect of TMP21 suppression in an in vivo model of AD, Ms. Haiyan Zou was asked to breed a APP23/S2P23 cross by breeding APP23 heterozy- gotes with S2P23 hemizygotes. In 6 months we generated 50 pups from 2 breeding cages. The predicted genotyping ratios were 1:1:1:1 APP23 : S2P23 : C57BL/6 : APP23/S2P23. In practice, we observed a 0.48 : 1.92 : 0.8 : 0.8 ratio. In total, 10 double crossed mice were produced: 7 female and 3 male. Double mice appeared smaller than their littermate controls, and of the 10, 5 died just before 3 months of age, while the other genotypes appeared normal. Due to limited numbers, no behavioural tests were conducted; however, simply looking at the fatality of the double cross one can see that heterozygous knockdown of TMP21, when combined with APPSwe overexpression, is detrimental. While preliminary immunohistochemical staining with rabbit anti-human Ab antibody 4G8 was unable to detect Ab plaques in either double or APP23 mice, this does not exclude the possibility of a behavioural deficit, as Morris water maze deficits are present in APP23 mice up to 3 months before plaque detection (Dam et al., 2003; Kelly et al., 2003). The lethality of the APP23/S2P23 cross demands further attention. While we may be unable to assess behaviour at 3 months of age, almost all double mice survived to 2 months of age, and if the effect of TMP21 suppression is sufficiently strong it may be possible to detect differences in the Morris water maze task between double crossed mice and APP23, S2P23, or control mice at this age. As double mice appear smaller than their littermates, to properly interpret the data growth rate and motor function should be examined at one month and 2 months of age. Finally, when tests are complete, brain levels of Ab should be examined by ELISA assay 133 5.4. Potential applications in future research to determine the effect of in vivo TMP21 suppression on AD pathology, as plaques were unable to be detected immunohistochemically. While APP23 mice have APPSwe expressed specifically in neurons, S2P23 mice are heterozygous for TMP21 in all tissues. As double crossed mice only expressed both conditions in their neurons, it is likely that the observed lethality is neurolog- ical in origin; however, from this information alone we cannot determine if this is a developmental issue or something which develops later in life. As such, Ms. Xiaojie Zhang in our lab has designed an inducible TMP21 knockout mice, which she will cross with a Cre mouse to achieve an inducible neuronal-TMP21 knockout. This mouse can then be bred with the APP23 mouse. The ability to turn off neuronal expression of TMP21 on demand in an AD mouse model would provide an incred- ibly powerful tool in the study of TMP21 suppression in AD pathogenesis in the upcoming years. 134 Bibliography AA-USA (2010). Alzheimer’s Disease Facts and Figures. Technical report, Alzheimer’s Association. Abdul, H. M., Sama, M. A., Furman, J. L., Mathis, D. M., Beckett, T. L., Weidner, A. M., Patel, E. S., Baig, I., Murphy, M. P., LeVine, H., Kraner, S. D., and Norris, C. M. (2009). Cognitive decline in Alzheimer’s disease is associated with selective changes in calcineurin/NFAT signaling. J Neurosci, 29(41):12957–12969. Acquati, F., Accarino, M., Nucci, C., Fumagalli, P., Jovine, L., Ottolenghi, S., and Taramelli, R. (2000). The gene encoding DRAP (BACE2), a glycosylated transmembrane protein of the aspartic protease family, maps to the down critical region. FEBS Lett, 468(1):59–64. Amenta, J. S. and Brocher, S. C. (1980). Role of lysosomes in protein turnover: catch-up proteolysis after release from NH4Cl inhibition. J Cell Physiol, 102(2):259–266. Antonny, B., Beraud-Dufour, S., Chardin, P., and Chabre, M. (1997). N-terminal hydrophobic residues of the G-protein ADP-ribosylation factor-1 insert into mem- brane phospholipids upon GDP to GTP exchange. Biochemistry, 36(15):4675– 4684. Arredondo, J., Chernyavsky, A. I., Jolkovsky, D. L., Pinkerton, K. E., and Grando, S. A. (2006). Receptor-mediated tobacco toxicity: cooperation of the Ras/Raf- 1/MEK1/ERK and JAK-2/STAT-3 pathways downstream of alpha7 nicotinic re- ceptor in oral keratinocytes. FASEB J, 20(12):2093–2101. ASC (2010). Rising Tide: The Impact of Dementia on Canadian Society. Technical report, Alzheimer Society of Canada. Balch, W. E., Dunphy, W. G., Braell, W. A., and Rothman, J. E. (1984). Re- constitution of the transport of protein between successive compartments of the Golgi measured by the coupled incorporation of N-acetylglucosamine. Cell, 39(2 Pt 1):405–416. Barr, F. A., Nakamura, N., and Warren, G. (1998). Mapping the interaction between GRASP65 and GM130, components of a protein complex involved in the stacking of Golgi cisternae. EMBO J, 17(12):3258–3268. 135 Bibliography Barr, F. A., Preisinger, C., Kopajtich, R., and Körner, R. (2001). Golgi matrix proteins interact with p24 cargo receptors and aid their efficient retention in the Golgi apparatus. J Cell Biol, 155(6):885–891. Barr, F. A., Puype, M., Vandekerckhove, J., and Warren, G. (1997). GRASP65, a protein involved in the stacking of Golgi cisternae. Cell, 91(2):253–262. Barroso, M., Nelson, D. S., and Sztul, E. (1995). Transcytosis-associated protein (TAP)/p115 is a general fusion factor required for binding of vesicles to acceptor membranes. Proc Natl Acad Sci U S A, 92(2):527–531. Basi, G., Frigon, N., Barbour, R., Doan, T., Gordon, G., McConlogue, L., Sinha, S., and Zeller, M. (2003). Antagonistic effects of beta-site amyloid precursor protein- cleaving enzymes 1 and 2 on beta-amyloid peptide production in cells. J Biol Chem, 278(34):31512–31520. Beams, H. W. and Kessel, R. G. (1968). The Golgi apparatus: structure and function. Int Rev Cytol, 23:209–276. Bearzatto, B., Servais, L., Cheron, G., and Schiffmann, S. N. (2005). Age depen- dence of strain determinant on mice motor coordination. Brain Res, 1039(1-2):37– 42. Bennett, B. D., Denis, P., Haniu, M., Teplow, D. B., Kahn, S., Louis, J. C., Citron, M., and Vassar, R. (2000). A furin-like convertase mediates propeptide cleavage of BACE, the Alzheimer’s beta -secretase. J Biol Chem, 275(48):37712–7. 0021-9258 Journal Article. Bergman, A., Hansson, E. M., Pursglove, S. E., Farmery, M. R., Lannfelt, L., Lendahl, U., Lundkvist, J., and Näslund, J. (2004). Pen-2 is sequestered in the endoplasmic reticulum and subjected to ubiquitylation and proteasome-mediated degradation in the absence of presenilin. J Biol Chem, 279(16):16744–16753. Blagitko, N., Schulz, U., Schinzel, A. A., Ropers, H. H., and Kalscheuer, V. M. (1999). gamma2-COP, a novel imprinted gene on chromosome 7q32, defines a new imprinting cluster in the human genome. Hum Mol Genet, 8(13):2387–2396. Blanchard, R. J. and Blanchard, D. C. (1969). Crouching as an index of fear. J Comp Physiol Psychol, 67(3):370–375. Blaumueller, C. M., Qi, H., Zagouras, P., and Artavanis-Tsakonas, S. (1997). In- tracellular cleavage of Notch leads to a heterodimeric receptor on the plasma membrane. Cell, 90(2):281–291. Blum, R., Feick, P., Puype, M., Vandekerckhove, J., Klengel, R., Nastainczyk, W., and Schulz, I. (1996). Tmp21 and p24A, two type I proteins enriched in 136 Bibliography pancreatic microsomal membranes, are members of a protein family involved in vesicular trafficking. J Biol Chem, 271(29):17183–17189. Blum, R., Pfeiffer, F., Feick, P., Nastainczyk, W., Kohler, B., Schäfer, K. H., and Schulz, I. (1999). Intracellular localization and in vivo trafficking of p24A and p23. J Cell Sci, 112 ( Pt 4):537–548. Bonfanti, L., Mironov, A. A., Mart́ınez-Menárguez, J. A., Martella, O., Fusella, A., Baldassarre, M., Buccione, R., Geuze, H. J., Mironov, A. A., and Luini, A. (1998). Procollagen traverses the Golgi stack without leaving the lumen of cisternae: evidence for cisternal maturation. Cell, 95(7):993–1003. Borchelt, D. R., Ratovitski, T., van Lare, J., Lee, M. K., Gonzales, V., Jenkins, N. A., Copeland, N. G., Price, D. L., and Sisodia, S. S. (1997). Accelerated amyloid deposition in the brains of transgenic mice coexpressing mutant presenilin 1 and amyloid precursor proteins. Neuron, 19(4):939–945. Borchelt, D. R., Thinakaran, G., Eckman, C. B., Lee, M. K., Davenport, F., Ra- tovitsky, T., Prada, C. M., Kim, G., Seekins, S., Yager, D., Slunt, H. H., Wang, R., Seeger, M., Levey, A. I., Gandy, S. E., Copeland, N. G., Jenkins, N. A., Price, D. L., Younkin, S. G., and Sisodia, S. S. (1996). Familial Alzheimer’s disease-linked presenilin 1 variants elevate Abeta1-42/1-40 ratio in vitro and in vivo. Neuron, 17(5):1005–1013. Bouwknecht, J. A. and Paylor, R. (2002). Behavioral and physiological mouse assays for anxiety: a survey in nine mouse strains. Behav Brain Res, 136(2):489–501. Braak, H. and Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol, 82(4):239–259. Braak, H., Braak, E., and Bohl, J. (1993). Staging of Alzheimer-related cortical destruction. Eur Neurol, 33(6):403–408. Bremser, M., Nickel, W., Schweikert, M., Ravazzola, M., Amherdt, M., Hughes, C. A., Söllner, T. H., Rothman, J. E., and Wieland, F. T. (1999). Coupling of coat assembly and vesicle budding to packaging of putative cargo receptors. Cell, 96(4):495–506. Broadhurst, P. L. (1961). Analysis of maternal effects in the inheritance of behaviour. Animal Behaviour, 9(3-4):129 – 141. Brou, C., Logeat, F., Gupta, N., Bessia, C., LeBail, O., Doedens, J. R., Cumano, A., Roux, P., Black, R. A., and Israël, A. (2000). A novel proteolytic cleavage involved in Notch signaling: the role of the disintegrin-metalloprotease TACE. Mol Cell, 5(2):207–216. 137 Bibliography Buxbaum, J. D., Gandy, S. E., Cicchetti, P., Ehrlich, M. E., Czernik, A. J., Fracasso, R. P., Ramabhadran, T. V., Unterbeck, A. J., and Greengard, P. (1990). Process- ing of Alzheimer beta/A4 amyloid precursor protein: modulation by agents that regulate protein phosphorylation. Proc Natl Acad Sci U S A, 87(15):6003–6006. Cai, H., Wang, Y., McCarthy, D., Wen, H., Borchelt, D. R., Price, D. L., and Wong, P. C. (2001). BACE1 is the major beta-secretase for generation of Abeta peptides by neurons. Nat Neurosci, 4(3):233–234. Capell, A., Grünberg, J., Pesold, B., Diehlmann, A., Citron, M., Nixon, R., Beyreuther, K., Selkoe, D. J., and Haass, C. (1998). The proteolytic fragments of the Alzheimer’s disease-associated presenilin-1 form heterodimers and occur as a 100-150-kDa molecular mass complex. J Biol Chem, 273(6):3205–3211. Caporaso, G. L., Gandy, S. E., Buxbaum, J. D., Ramabhadran, T. V., and Green- gard, P. (1992). Protein phosphorylation regulates secretion of Alzheimer beta/A4 amyloid precursor protein. Proc Natl Acad Sci U S A, 89(7):3055–3059. Cendeĺın, J., Korelusová, I., and Vozeh, F. (2008). The effect of repeated rotarod training on motor skills and spatial learning ability in Lurcher mutant mice. Behav Brain Res, 189(1):65–74. Cendeĺın, J., Voller, J., and Vozeh, F. (2010). Ataxic gait analysis in a mouse model of the olivocerebellar degeneration. Behav Brain Res, 210(1):8–15. Chen, F., Gu, Y., Hasegawa, H., Ruan, X., Arawaka, S., Fraser, P., Westaway, D., Mount, H., and George-Hyslop, P. S. (2002). Presenilin 1 mutations activate gamma 42-secretase but reciprocally inhibit epsilon-secretase cleavage of amyloid precursor protein (APP) and S3-cleavage of notch. J Biol Chem, 277(39):36521– 36526. Chen, F., Hasegawa, H., Schmitt-Ulms, G., Kawarai, T., Bohm, C., Katayama, T., Gu, Y., Sanjo, N., Glista, M., Rogaeva, E., Wakutani, Y., Pardossi-Piquard, R., Ruan, X., Tandon, A., Checler, F., Marambaud, P., Hansen, K., Westaway, D., George-Hyslop, P. S., and Fraser, P. (2006). TMP21 is a presenilin complex component that modulates gamma-secretase but not epsilon-secretase activity. Nature, 440(7088):1208–1212. Christensen, M. A., Zhou, W., Qing, H., Lehman, A., Philipsen, S., and Song, W. (2004). Transcriptional regulation of BACE1, the beta-amyloid precursor protein beta-secretase, by Sp1. Mol Cell Biol, 24(2):865–874. Chung, H. M. and Struhl, G. (2001). Nicastrin is required for Presenilin-mediated transmembrane cleavage in Drosophila. Nat Cell Biol, 3(12):1129–1132. 138 Bibliography Citron, M., Oltersdorf, T., Haass, C., McConlogue, L., Hung, A. Y., Seubert, P., Vigo-Pelfrey, C., Lieberburg, I., and Selkoe, D. J. (1992). Mutation of the beta- amyloid precursor protein in familial Alzheimer’s disease increases beta-protein production. Nature, 360(6405):672–674. Claude, A., Zhao, B. P., Kuziemsky, C. E., Dahan, S., Berger, S. J., Yan, J. P., Armold, A. D., Sullivan, E. M., and Melançon, P. (1999). GBF1: A novel Golgi- associated BFA-resistant guanine nucleotide exchange factor that displays speci- ficity for ADP-ribosylation factor 5. J Cell Biol, 146(1):71–84. Cole, G. M. and Timiras, P. S. (1987). Ubiquitin-protein conjugates in Alzheimer’s lesions. Neurosci Lett, 79(1-2):207–212. Cosson, P. and Letourneur, F. (1994). Coatomer interaction with di-lysine endo- plasmic reticulum retention motifs. Science, 263(5153):1629–1631. Crabbe, J. C. (1986). Genetic differences in locomotor activation in mice. Pharmacol Biochem Behav, 25(1):289–292. Crawley, J. N. (2000a). What’s wrong with my mouse? Behavioural Phenotyping of Transgenic and Knockout Mice, pages 47–63. Wiley-Liss. Crawley, J. N. (2000b). What’s wrong with my mouse? Behavioural Phenotyping of Transgenic and Knockout Mice, pages 83–129. Wiley-Liss. Crawley, J. N., Belknap, J. K., Collins, A., Crabbe, J. C., Frankel, W., Hender- son, N., Hitzemann, R. J., Maxson, S. C., Miner, L. L., Silva, A. J., Wehner, J. M., Wynshaw-Boris, A., and Paylor, R. (1997). Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacology (Berl), 132(2):107–124. Crystal, A. S., Morais, V. A., Fortna, R. R., Carlin, D., Pierson, T. C., Wilson, C. A., Lee, V. M.-Y., and Doms, R. W. (2004). Presenilin modulates Pen-2 levels posttranslationally by protecting it from proteasomal degradation. Biochemistry, 43(12):3555–3563. Dallas, D. J., Genever, P. G., Patton, A. J., Millichip, M. I., McKie, N., and Skerry, T. M. (1999). Localization of ADAM10 and Notch receptors in bone. Bone, 25(1):9–15. Dam, D. V., D’Hooge, R., Staufenbiel, M., Ginneken, C. V., Meir, F. V., and Deyn, P. P. D. (2003). Age-dependent cognitive decline in the APP23 model precedes amyloid deposition. Eur J Neurosci, 17(2):388–396. David, D. C., Layfield, R., Serpell, L., Narain, Y., Goedert, M., and Spillantini, M. G. (2002). Proteasomal degradation of tau protein. J Neurochem, 83(1):176– 185. 139 Bibliography De Strooper, B., Annaert, W., Cupers, P., Saftig, P., Craessaerts, K., Mumm, J. S., Schroeter, E. H., Schrijvers, V., Wolfe, M. S., Ray, W. J., Goate, A., and Kopan, R. (1999). A presenilin-1-dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature, 398(6727):518–522. De Strooper, B., Saftig, P., Craessaerts, K., Vanderstichele, H., Guhde, G., Annaert, W., von Figura, K., and Leuven, F. V. (1998). Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein. Nature, 391(6665):387–390. DeFries, J. C., Hegmann, J. P., and Weir, M. W. (1966). Open-field behavior in mice: evidence for a major gene effect mediated by the visual system. Science, 154(756):1577–1579. Denzel, A., Otto, F., Girod, A., Pepperkok, R., Watson, R., Rosewell, I., Bergeron, J. J., Solari, R. C., and Owen, M. J. (2000). The p24 family member p23 is required for early embryonic development. Curr Biol, 10(1):55–58. Dillon, G. M., Qu, X., Marcus, J. N., and Dodart, J.-C. (2008). Excitotoxic lesions restricted to the dorsal CA1 field of the hippocampus impair spatial memory and extinction learning in C57BL/6 mice. Neurobiol Learn Mem, 90(2):426–433. Dolcini, V., Dunys, J., Sevalle, J., Chen, F., Guillot-Sestier, M.-V., George-Hyslop, P. S., Fraser, P. E., and Checler, F. (2008). TMP21 regulates Abeta production but does not affect caspase-3, p53, and neprilysin. Biochem Biophys Res Commun, 371(1):69–74. Dominguez, M., Dejgaard, K., Füllekrug, J., Dahan, S., Fazel, A., Paccaud, J. P., Thomas, D. Y., Bergeron, J. J., and Nilsson, T. (1998). gp25L/emp24/p24 protein family members of the cis-Golgi network bind both COP I and II coatomer. J Cell Biol, 140(4):751–765. Doms, R. W., Russ, G., and Yewdell, J. W. (1989). Brefeldin A redistributes resident and itinerant Golgi proteins to the endoplasmic reticulum. J Cell Biol, 109(1):61–72. Donaldson, J. G. (2003). Multiple roles for Arf6: sorting, structuring, and signaling at the plasma membrane. J Biol Chem, 278(43):41573–41576. Donaldson, J. G., Cassel, D., Kahn, R. A., and Klausner, R. D. (1992a). ADP- ribosylation factor, a small GTP-binding protein, is required for binding of the coatomer protein beta-COP to Golgi membranes. Proc Natl Acad Sci U S A, 89(14):6408–6412. Donaldson, J. G., Finazzi, D., and Klausner, R. D. (1992b). Brefeldin A inhibits Golgi membrane-catalysed exchange of guanine nucleotide onto ARF protein. Na- ture, 360(6402):350–352. 140 Bibliography Donoviel, D. B., Hadjantonakis, A. K., Ikeda, M., Zheng, H., Hyslop, P. S., and Bernstein, A. (1999). Mice lacking both presenilin genes exhibit early embryonic patterning defects. Genes Dev, 13(21):2801–2810. Duden, R., Griffiths, G., Frank, R., Argos, P., and Kreis, T. E. (1991). Beta-COP, a 110 kd protein associated with non-clathrin-coated vesicles and the Golgi complex, shows homology to beta-adaptin. Cell, 64(3):649–665. Duden, R., Kajikawa, L., Wuestehube, L., and Schekman, R. (1998). epsilon-COP is a structural component of coatomer that functions to stabilize alpha-COP. EMBO J, 17(4):985–995. Dunphy, W. G. and Rothman, J. E. (1985). Compartmental organization of the Golgi stack. Cell, 42(1):13–21. Dyrks, T., Weidemann, A., Multhaup, G., Salbaum, J. M., Lemaire, H. G., Kang, J., Müller-Hill, B., Masters, C. L., and Beyreuther, K. (1988). Identification, trans- membrane orientation and biogenesis of the amyloid A4 precursor of Alzheimer’s disease. EMBO J, 7(4):949–957. Edbauer, D., Winkler, E., Haass, C., and Steiner, H. (2002). Presenilin and nicastrin regulate each other and determine amyloid beta-peptide production via complex formation. Proc Natl Acad Sci U S A, 99(13):8666–8671. Edbauer, D., Winkler, E., Regula, J. T., Pesold, B., Steiner, H., and Haass, C. (2003). Reconstitution of gamma-secretase activity. Nat Cell Biol, 5(5):486–488. Esch, F. S., Keim, P. S., Beattie, E. C., Blacher, R. W., Culwell, A. R., Oltersdorf, T., McClure, D., and Ward, P. J. (1990). Cleavage of amyloid beta peptide during constitutive processing of its precursor. Science, 248(4959):1122–1124. Esler, W. P., Kimberly, W. T., Ostaszewski, B. L., Diehl, T. S., Moore, C. L., Tsai, J. Y., Rahmati, T., Xia, W., Selkoe, D. J., and Wolfe, M. S. (2000). Transition- state analogue inhibitors of gamma-secretase bind directly to presenilin-1. Nat Cell Biol, 2(7):428–434. Esler, W. P., Kimberly, W. T., Ostaszewski, B. L., Ye, W., Diehl, T. S., Selkoe, D. J., and Wolfe, M. S. (2002). Activity-dependent isolation of the presenilin- gamma -secretase complex reveals nicastrin and a gamma substrate. Proc Natl Acad Sci U S A, 99(5):2720–2725. Eugster, A., Frigerio, G., Dale, M., and Duden, R. (2000). COP I domains required for coatomer integrity, and novel interactions with ARF and ARF-GAP. EMBO J, 19(15):3905–3917. Farquhar, M. G. (1985). Progress in unraveling pathways of Golgi traffic. Annu Rev Cell Biol, 1:447–488. 141 Bibliography Faulstich, D., Auerbach, S., Orci, L., Ravazzola, M., Wegchingel, S., Lottspeich, F., Stenbeck, G., Harter, C., Wieland, F. T., and Tschochner, H. (1996). Architecture of coatomer: molecular characterization of delta-COP and protein interactions within the complex. J Cell Biol, 135(1):53–61. Fenteany, G., Standaert, R. F., Lane, W. S., Choi, S., Corey, E. J., and Schreiber, S. L. (1995). Inhibition of proteasome activities and subunit-specific amino- terminal threonine modification by lactacystin. Science, 268(5211):726–731. Fiedler, K., Veit, M., Stamnes, M. A., and Rothman, J. E. (1996). Bimodal in- teraction of coatomer with the p24 family of putative cargo receptors. Science, 273(5280):1396–1399. Fligge, T. A., Reinhard, C., Harter, C., Wieland, F. T., and Przybylski, M. (2000). Oligomerization of peptides analogous to the cytoplasmic domains of coatomer receptors revealed by mass spectrometry. Biochemistry, 39(29):8491–8496. Fraering, P. C., Ye, W., Strub, J.-M., Dolios, G., LaVoie, M. J., Ostaszewski, B. L., van Dorsselaer, A., Wang, R., Selkoe, D. J., and Wolfe, M. S. (2004). Purifica- tion and characterization of the human gamma-secretase complex. Biochemistry, 43(30):9774–9789. Francis, R., McGrath, G., Zhang, J., Ruddy, D. A., Sym, M., Apfeld, J., Nicoll, M., Maxwell, M., Hai, B., Ellis, M. C., Parks, A. L., Xu, W., Li, J., Gurney, M., Myers, R. L., Himes, C. S., Hiebsch, R., Ruble, C., Nye, J. S., and Curtis, D. (2002). aph-1 and pen-2 are required for Notch pathway signaling, gamma- secretase cleavage of betaAPP, and presenilin protein accumulation. Dev Cell, 3(1):85–97. Franco, M., Chardin, P., Chabre, M., and Paris, S. (1993). Myristoylation is not required for GTP-dependent binding of ADP-ribosylation factor ARF1 to phos- pholipids. J Biol Chem, 268(33):24531–24534. Fraser, P. E., Levesque, G., Yu, G., Mills, L. R., Thirlwell, J., Frantseva, M., Gandy, S. E., Seeger, M., Carlen, P. L., and George-Hyslop, P. S. (1998). Presenilin 1 is actively degraded by the 26S proteasome. Neurobiol Aging, 19(1 Suppl):S19–S21. Füllekrug, J., Suganuma, T., Tang, B. L., Hong, W., Storrie, B., and Nilsson, T. (1999). Localization and recycling of gp27 (hp24gamma3): complex formation with other p24 family members. Mol Biol Cell, 10(6):1939–1955. Futatsumori, M., Kasai, K., Takatsu, H., Shin, H. W., and Nakayama, K. (2000). Identification and characterization of novel isoforms of COP I subunits. J Biochem, 128(5):793–801. 142 Bibliography Gandy, S., Czernik, A. J., and Greengard, P. (1988). Phosphorylation of Alzheimer disease amyloid precursor peptide by protein kinase C and Ca2+/calmodulin- dependent protein kinase II. Proc Natl Acad Sci U S A, 85(16):6218–6221. George-Hyslop, P. S., Haines, J., Rogaev, E., Mortilla, M., Vaula, G., Pericak- Vance, M., Foncin, J. F., Montesi, M., Bruni, A., and Sorbi, S. (1992). Genetic evidence for a novel familial Alzheimer’s disease locus on chromosome 14. Nat Genet, 2(4):330–334. Glenner, G. G. and Wong, C. W. (1984). Alzheimer’s disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun, 120(3):885–890. Golde, T. E., Estus, S., Younkin, L. H., Selkoe, D. J., and Younkin, S. G. (1992). Processing of the amyloid protein precursor to potentially amyloidogenic deriva- tives. Science, 255(5045):728–730. Gommel, D., Orci, L., Emig, E. M., Hannah, M. J., Ravazzola, M., Nickel, W., Helms, J. B., Wieland, F. T., and Sohn, K. (1999). p24 and p23, the major trans- membrane proteins of COPI-coated transport vesicles, form hetero-oligomeric complexes and cycle between the organelles of the early secretory pathway. FEBS Lett, 447(2-3):179–185. Gommel, D. U., Memon, A. R., Heiss, A., Lottspeich, F., Pfannstiel, J., Lechner, J., Reinhard, C., Helms, J. B., Nickel, W., and Wieland, F. T. (2001). Recruitment to Golgi membranes of ADP-ribosylation factor 1 is mediated by the cytoplasmic domain of p23. EMBO J, 20(23):6751–6760. Gonzalez-Noriega, A., Grubb, J. H., Talkad, V., and Sly, W. S. (1980). Chloroquine inhibits lysosomal enzyme pinocytosis and enhances lysosomal enzyme secretion by impairing receptor recycling. J Cell Biol, 85(3):839–852. Gordon, A. H., Hart, P. D., and Young, M. R. (1980). Ammonia inhibits phagosome- lysosome fusion in macrophages. Nature, 286(5768):79–80. Goutte, C., Tsunozaki, M., Hale, V. A., and Priess, J. R. (2002). APH-1 is a multi- pass membrane protein essential for the Notch signaling pathway in Caenorhab- ditis elegans embryos. Proc Natl Acad Sci U S A, 99(2):775–779. Grundke-Iqbal, I., Iqbal, K., Quinlan, M., Tung, Y. C., Zaidi, M. S., and Wisniewski, H. M. (1986a). Microtubule-associated protein tau. A component of Alzheimer paired helical filaments. J Biol Chem, 261(13):6084–6089. Grundke-Iqbal, I., Iqbal, K., Tung, Y. C., Quinlan, M., Wisniewski, H. M., and Binder, L. I. (1986b). Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A, 83(13):4913–4917. 143 Bibliography Guo, Q., Penman, M., Trigatti, B. L., and Krieger, M. (1996). A single point muta- tion in epsilon-COP results in temperature-sensitive, lethal defects in membrane transport in a Chinese hamster ovary cell mutant. J Biol Chem, 271(19):11191– 11196. Guo, Q., Vasile, E., and Krieger, M. (1994). Disruptions in Golgi structure and membrane traffic in a conditional lethal mammalian cell mutant are corrected by epsilon-COP. J Cell Biol, 125(6):1213–1224. Haass, C., Hung, A. Y., Schlossmacher, M. G., Teplow, D. B., and Selkoe, D. J. (1993). beta-Amyloid peptide and a 3-kDa fragment are derived by distinct cel- lular mechanisms. J Biol Chem, 268(5):3021–3024. Haass, C., Koo, E. H., Mellon, A., Hung, A. Y., and Selkoe, D. J. (1992a). Targeting of cell-surface beta-amyloid precursor protein to lysosomes: alternative processing into amyloid-bearing fragments. Nature, 357(6378):500–503. Haass, C., Schlossmacher, M. G., Hung, A. Y., Vigo-Pelfrey, C., Mellon, A., Os- taszewski, B. L., Lieberburg, I., Koo, E. H., Schenk, D., and Teplow, D. B. (1992b). Amyloid beta-peptide is produced by cultured cells during normal metabolism. Nature, 359(6393):322–325. Hara-Kuge, S., Kuge, O., Orci, L., Amherdt, M., Ravazzola, M., Wieland, F. T., and Rothman, J. E. (1994). En bloc incorporation of coatomer subunits during the assembly of COP-coated vesicles. J Cell Biol, 124(6):883–892. Harter, C., Draken, E., Lottspeich, F., and Wieland, F. T. (1993). Yeast coatomer contains a subunit homologous to mammalian beta’-COP. FEBS Lett, 332(1- 2):71–73. Harter, C., Pavel, J., Coccia, F., Draken, E., Wegehingel, S., Tschochner, H., and Wieland, F. (1996). Nonclathrin coat protein gamma, a subunit of coatomer, binds to the cytoplasmic dilysine motif of membrane proteins of the early secretory pathway. Proc Natl Acad Sci U S A, 93(5):1902–1906. Harter, C. and Wieland, F. T. (1998). A single binding site for dilysine retrieval motifs and p23 within the gamma subunit of coatomer. Proc Natl Acad Sci U S A, 95(20):11649–11654. Hartmann, D., De Strooper, B., Serneels, L., Craessaerts, K., Herreman, A., An- naert, W., Umans, L., Lübke, T., Illert, A. L., von Figura, K., and Saftig, P. (2002). The disintegrin/metalloprotease ADAM 10 is essential for Notch signalling but not for alpha-secretase activity in fibroblasts. Hum Mol Genet, 11(21):2615– 2624. 144 Bibliography Hasegawa, H., Liu, L., and Nishimura, M. (2010). Dilysine retrieval signal-containing p24 proteins collaborate in inhibiting gamma-cleavage of amyloid precursor pro- tein. J Neurochem. He, G., Qing, H., Cai, F., Kwok, C., Xu, H., Yu, G., Bernstein, A., and Song, W. (2006). Ubiquitin-proteasome pathway mediates degradation of APH-1. J Neurochem, 99(5):1403–1412. He, G., Qing, H., Tong, Y., Cai, F., Ishiura, S., and Song, W. (2007). Degradation of nicastrin involves both proteasome and lysosome. J Neurochem, 101(4):982–992. Helms, J. B., Palmer, D. J., and Rothman, J. E. (1993). Two distinct populations of ARF bound to Golgi membranes. J Cell Biol, 121(4):751–760. Helms, J. B. and Rothman, J. E. (1992). Inhibition by brefeldin A of a Golgi membrane enzyme that catalyses exchange of guanine nucleotide bound to ARF. Nature, 360(6402):352–354. Hendricks, L. C., McClanahan, S. L., McCaffery, M., Palade, G. E., and Farquhar, M. G. (1992). Golgi proteins persist in the tubulovesicular remnants found in brefeldin A-treated pancreatic acinar cells. Eur J Cell Biol, 58(2):202–213. Herreman, A., Serneels, L., Annaert, W., Collen, D., Schoonjans, L., and De Strooper, B. (2000). Total inactivation of gamma-secretase activity in presenilin- deficient embryonic stem cells. Nat Cell Biol, 2(7):461–462. Hershko, A. and Ciechanover, A. (1998). The ubiquitin system. Annu Rev Biochem, 67:425–479. Hidalgo, J., Garcia-Navarro, R., Gracia-Navarro, F., Perez-Vilar, J., and Velasco, A. (1992). Presence of Golgi remnant membranes in the cytoplasm of brefeldin A-treated cells. Eur J Cell Biol, 58(2):214–227. Holmes, A. (2001). Targeted gene mutation approaches to the study of anxiety-like behavior in mice. Neurosci Biobehav Rev, 25(3):261–273. Homanics, G. E., Quinlan, J. J., and Firestone, L. L. (1999). Pharmacologic and behavioral responses of inbred C57BL/6J and strain 129/SvJ mouse lines. Phar- macol Biochem Behav, 63(1):21–26. Honda, T., Yasutake, K., Nihonmatsu, N., Mercken, M., Takahashi, H., Murayama, O., Murayama, M., Sato, K., Omori, A., Tsubuki, S., Saido, T. C., and Takashima, A. (1999). Dual roles of proteasome in the metabolism of presenilin 1. J Neu- rochem, 72(1):255–261. Hu, Y., Ye, Y., and Fortini, M. E. (2002). Nicastrin is required for gamma-secretase cleavage of the Drosophila Notch receptor. Dev Cell, 2(1):69–78. 145 Bibliography Hughes, R. N. (2004). The value of spontaneous alternation behavior (SAB) as a test of retention in pharmacological investigations of memory. Neurosci Biobehav Rev, 28(5):497–505. Hussain, I., Powell, D., Howlett, D. R., Tew, D. G., Meek, T. D., Chapman, C., Gloger, I. S., Murphy, K. E., Southan, C. D., Ryan, D. M., Smith, T. S., Simmons, D. L., Walsh, F. S., Dingwall, C., and Christie, G. (1999). Identification of a novel aspartic protease (Asp 2) as beta-secretase. Mol Cell Neurosci, 14(6):419–427. Ihara, Y., Nukina, N., Miura, R., and Ogawara, M. (1986). Phosphorylated tau pro- tein is integrated into paired helical filaments in Alzheimer’s disease. J Biochem, 99(6):1807–1810. Ikeuchi, T. and Sisodia, S. S. (2003). The Notch ligands, Delta1 and Jagged2, are substrates for presenilin-dependent ”gamma-secretase” cleavage. J Biol Chem, 278(10):7751–7754. Ishizaki, R., Shin, H.-W., Mitsuhashi, H., and Nakayama, K. (2008). Redun- dant Roles of BIG2 and BIG1, Guanine-Nucleotide Exchange Factors for ADP- Ribosylation Factors in Membrane Traffic between the trans-Golgi Network and Endosomes. Mol Biol Cell, 19(6):2650–2660. Jackson, M. R., Nilsson, T., and Peterson, P. A. (1990). Identification of a consen- sus motif for retention of transmembrane proteins in the endoplasmic reticulum. EMBO J, 9(10):3153–3162. Jarriault, S & Greenwald, I. (2005). Evidence for functional redundancy between C. elegans ADAM proteins SUP-17/Kuzbanian and ADM-4/TACE. Dev Biol, 287:1–10. JaxLabs (2009). Jackson Lab, collaborators extend lifespan of aging mice with transplant drug. Press Release. Jensen, T. J., Loo, M. A., Pind, S., Williams, D. B., Goldberg, A. L., and Riordan, J. R. (1995). Multiple proteolytic systems, including the proteasome, contribute to CFTR processing. Cell, 83(1):129–135. Jones, B. J. and Roberts, D. J. (1968a). A rotarod suitable for quantitative mea- surements of motor incoordination in naive mice. Naunyn Schmiedebergs Arch Exp Pathol Pharmakol, 259(2):211. Jones, B. J. and Roberts, D. J. (1968b). The quantiative measurement of motor inco-ordination in naive mice using an acelerating rotarod. J Pharm Pharmacol, 20(4):302–304. 146 Bibliography Kaether, C., Lammich, S., Edbauer, D., Ertl, M., Rietdorf, J., Capell, A., Steiner, H., and Haass, C. (2002). Presenilin-1 affects trafficking and processing of be- taAPP and is targeted in a complex with nicastrin to the plasma membrane. J Cell Biol, 158(3):551–561. Kahn, R. A. and Gilman, A. G. (1984). Purification of a protein cofactor required for ADP-ribosylation of the stimulatory regulatory component of adenylate cyclase by cholera toxin. J Biol Chem, 259(10):6228–6234. Kahn, R. A. and Gilman, A. G. (1986). The protein cofactor necessary for ADP- ribosylation of Gs by cholera toxin is itself a GTP binding protein. J Biol Chem, 261(17):7906–7911. Kang, J., Lemaire, H. G., Unterbeck, A., Salbaum, J. M., Masters, C. L., Grzeschik, K. H., Multhaup, G., Beyreuther, K., and Müller-Hill, B. (1987). The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell-surface receptor. Nature, 325(6106):733–736. Karrenbauer, A., Jeckel, D., Just, W., Birk, R., Schmidt, R. R., Rothman, J. E., and Wieland, F. T. (1990). The rate of bulk flow from the Golgi to the plasma membrane. Cell, 63(2):259–267. Kawamoto, K., Yoshida, Y., Tamaki, H., Torii, S., Shinotsuka, C., Yamashina, S., and Nakayama, K. (2002). GBF1, a guanine nucleotide exchange factor for ADP-ribosylation factors, is localized to the cis-Golgi and involved in membrane association of the COPI coat. Traffic, 3(7):483–495. Kay, J. E. and Korner, A. (1966). Effect of cycloheximide on protein and ribonucleic acid synthesis in cultured human lymphocytes. Biochem J, 100(3):815–822. CHX is effective in human cells. Keck, S., Nitsch, R., Grune, T., and Ullrich, O. (2003). Proteasome inhibition by paired helical filament-tau in brains of patients with Alzheimer’s disease. J Neurochem, 85(1):115–122. Kelly, P. H., Bondolfi, L., Hunziker, D., Schlecht, H.-P., Carver, K., Maguire, E., Abramowski, D., Wiederhold, K.-H., Sturchler-Pierrat, C., Jucker, M., Bergmann, R., Staufenbiel, M., and Sommer, B. (2003). Progressive age-related impairment of cognitive behavior in APP23 transgenic mice. Neurobiol Aging, 24(2):365–378. Kerridge, D. (1958). The effect of actidione and other antifungal agents on nucleic acid and protein synthesis in Saccharomyces carlsbergensis. J Gen Microbiol, 19(3):497–506. First showed that CHX stops protein synthesis. Kim, D. Y., Ingano, L. A. M., and Kovacs, D. M. (2002). Nectin-1alpha, an immunoglobulin-like receptor involved in the formation of synapses, is a substrate for presenilin/gamma-secretase-like cleavage. J Biol Chem, 277(51):49976–49981. 147 Bibliography Kim, J. J. and Fanselow, M. S. (1992). Modality-specific retrograde amnesia of fear. Science, 256(5057):675–677. Kim, J. J., Rison, R. A., and Fanselow, M. S. (1993). Effects of amygdala, hip- pocampus, and periaqueductal gray lesions on short- and long-term contextual fear. Behav Neurosci, 107(6):1093–1098. Kim, S.-H., Ikeuchi, T., Yu, C., and Sisodia, S. S. (2003). Regulated hyperaccu- mulation of presenilin-1 and the ”gamma-secretase” complex. Evidence for differ- ential intramembranous processing of transmembrane subatrates. J Biol Chem, 278(36):33992–34002. Kim, T. W., Pettingell, W. H., Hallmark, O. G., Moir, R. D., Wasco, W., and Tanzi, R. E. (1997). Endoproteolytic cleavage and proteasomal degradation of presenilin 2 in transfected cells. J Biol Chem, 272(17):11006–11010. Kimberly, W. T., LaVoie, M. J., Ostaszewski, B. L., Ye, W., Wolfe, M. S., and Selkoe, D. J. (2002). Complex N-linked glycosylated nicastrin associates with active gamma-secretase and undergoes tight cellular regulation. J Biol Chem, 277(38):35113–35117. Kimberly, W. T., LaVoie, M. J., Ostaszewski, B. L., Ye, W., Wolfe, M. S., and Selkoe, D. J. (2003). Gamma-secretase is a membrane protein complex com- prised of presenilin, nicastrin, Aph-1, and Pen-2. Proc Natl Acad Sci U S A, 100(11):6382–6387. Kimberly, W. T., Xia, W., Rahmati, T., Wolfe, M. S., and Selkoe, D. J. (2000). The transmembrane aspartates in presenilin 1 and 2 are obligatory for gamma- secretase activity and amyloid beta-protein generation. J Biol Chem, 275(5):3173– 3178. Kitaguchi, N., Takahashi, Y., Tokushima, Y., Shiojiri, S., and Ito, H. (1988). Novel precursor of Alzheimer’s disease amyloid protein shows protease inhibitory activ- ity. Nature, 331(6156):530–532. Kokkinidis, L. and Anisman, H. (1976). Dissociation of the effects of scopolamine and d-amphetamine on a spontaneous alternation task. Pharmacol Biochem Be- hav, 5(3):293–297. Koo, E. H. and Squazzo, S. L. (1994). Evidence that production and release of amy- loid beta-protein involves the endocytic pathway. J Biol Chem, 269(26):17386– 17389. Koo, E. H., Squazzo, S. L., Selkoe, D. J., and Koo, C. H. (1996). Trafficking of cell- surface amyloid beta-protein precursor. I. Secretion, endocytosis and recycling as detected by labeled monoclonal antibody. J Cell Sci, 109 ( Pt 5):991–998. 148 Bibliography Kopan, R. (2002). Notch: a membrane-bound transcription factor. J Cell Sci, 115(Pt 6):1095–1097. Kopan, R., Schroeter, E. H., Weintraub, H., and Nye, J. S. (1996). Signal transduc- tion by activated mNotch: importance of proteolytic processing and its regulation by the extracellular domain. Proc Natl Acad Sci U S A, 93(4):1683–1688. Kosik, K. S., Joachim, C. L., and Selkoe, D. J. (1986). Microtubule-associated protein tau (tau) is a major antigenic component of paired helical filaments in Alzheimer disease. Proc Natl Acad Sci U S A, 83(11):4044–4048. Kovacs, D. M., Fausett, H. J., Page, K. J., Kim, T. W., Moir, R. D., Merriam, D. E., Hollister, R. D., Hallmark, O. G., Mancini, R., Felsenstein, K. M., Hyman, B. T., Tanzi, R. E., and Wasco, W. (1996). Alzheimer-associated presenilins 1 and 2: neuronal expression in brain and localization to intracellular membranes in mammalian cells. Nat Med, 2(2):224–229. Kuge, O., Hara-Kuge, S., Orci, L., Ravazzola, M., Amherdt, M., Tanigawa, G., Wieland, F. T., and Rothman, J. E. (1993). zeta-COP, a subunit of coatomer, is required for COP-coated vesicle assembly. J Cell Biol, 123(6 Pt 2):1727–1734. Lammich, S., Okochi, M., Takeda, M., Kaether, C., Capell, A., Zimmer, A.-K., Edbauer, D., Walter, J., Steiner, H., and Haass, C. (2002). Presenilin-dependent intramembrane proteolysis of CD44 leads to the liberation of its intracellular domain and the secretion of an Abeta-like peptide. J Biol Chem, 277(47):44754– 44759. LaVoie, M. J. and Selkoe, D. J. (2003). The Notch ligands, Jagged and Delta, are sequentially processed by alpha-secretase and presenilin/gamma-secretase and release signaling fragments. J Biol Chem, 278(36):34427–34437. Leem, J. Y., Vijayan, S., Han, P., Cai, D., Machura, M., Lopes, K. O., Veselits, M. L., Xu, H., and Thinakaran, G. (2002). Presenilin 1 is required for maturation and cell surface accumulation of nicastrin. J Biol Chem, 277(21):19236–19240. Lefranc-Jullien, S., Sunyach, C., and Checler, F. (2006). APPepsilon, the epsilon- secretase-derived N-terminal product of the beta-amyloid precursor protein, be- haves as a type I protein and undergoes alpha-, beta-, and gamma-secretase cleav- ages. J Neurochem, 97(3):807–817. Lemos-Chiarandini, C. D., Ivessa, N. E., Black, V. H., Tsao, Y. S., Gumper, I., and Kreibich, G. (1992). A Golgi-related structure remains after the brefeldin A-induced formation of an ER-Golgi hybrid compartment. Eur J Cell Biol, 58(2):187–201. 149 Bibliography Letourneur, F., Gaynor, E. C., Hennecke, S., Démollière, C., Duden, R., Emr, S. D., Riezman, H., and Cosson, P. (1994). Coatomer is essential for retrieval of dilysine- tagged proteins to the endoplasmic reticulum. Cell, 79(7):1199–1207. Levitan, D., Doyle, T. G., Brousseau, D., Lee, M. K., Thinakaran, G., Slunt, H. H., Sisodia, S. S., and Greenwald, I. (1996). Assessment of normal and mutant hu- man presenilin function in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 93(25):14940–14944. Levy-Lahad, E., Wasco, W., Poorkaj, P., Romano, D. M., Oshima, J., Pettingell, W. H., Yu, C. E., Jondro, P. D., Schmidt, S. D., and Wang, K. (1995a). Can- didate gene for the chromosome 1 familial Alzheimer’s disease locus. Science, 269(5226):973–977. Levy-Lahad, E., Wijsman, E. M., Nemens, E., Anderson, L., Goddard, K. A., We- ber, J. L., Bird, T. D., and Schellenberg, G. D. (1995b). A familial Alzheimer’s disease locus on chromosome 1. Science, 269(5226):970–973. Lewis, M. J. and Pelham, H. R. (1992). Ligand-induced redistribution of a hu- man KDEL receptor from the Golgi complex to the endoplasmic reticulum. Cell, 68(2):353–364. Li, J., Ma, J., and Potter, H. (1995). Identification and expression analysis of a potential familial Alzheimer disease gene on chromosome 1 related to AD3. Proc Natl Acad Sci U S A, 92(26):12180–12184. Li, Y., Zhou, W., Tong, Y., He, G., and Song, W. (2006). Control of APP process- ing and Abeta generation level by BACE1 enzymatic activity and transcription. FASEB J, 20(2):285–292. Li, Y. M., Lai, M. T., Xu, M., Huang, Q., DiMuzio-Mower, J., Sardana, M. K., Shi, X. P., Yin, K. C., Shafer, J. A., and Gardell, S. J. (2000a). Presenilin 1 is linked with gamma-secretase activity in the detergent solubilized state. Proc Natl Acad Sci U S A, 97(11):6138–6143. Li, Y. M., Xu, M., Lai, M. T., Huang, Q., Castro, J. L., DiMuzio-Mower, J., Harrison, T., Lellis, C., Nadin, A., Neduvelil, J. G., Register, R. B., Sardana, M. K., Shearman, M. S., Smith, A. L., Shi, X. P., Yin, K. C., Shafer, J. A., and Gardell, S. J. (2000b). Photoactivated gamma-secretase inhibitors directed to the active site covalently label presenilin 1. Nature, 405(6787):689–694. Lin, X., Koelsch, G., Wu, S., Downs, D., Dashti, A., and Tang, J. (2000). Hu- man aspartic protease memapsin 2 cleaves the beta-secretase site of beta-amyloid precursor protein. Proc Natl Acad Sci U S A, 97(4):1456–1460. 150 Bibliography Lindner, M. D., Plone, M. A., Schallert, T., and Emerich, D. F. (1997). Blind rats are not profoundly impaired in the reference memory Morris water maze and cannot be clearly discriminated from rats with cognitive deficits in the cued platform task. Brain Res Cogn Brain Res, 5(4):329–333. Linstedt, A. D. and Hauri, H. P. (1993). Giantin, a novel conserved Golgi membrane protein containing a cytoplasmic domain of at least 350 kDa. Mol Biol Cell, 4(7):679–693. Lippincott-Schwartz, J., Donaldson, J. G., Schweizer, A., Berger, E. G., Hauri, H. P., Yuan, L. C., and Klausner, R. D. (1990). Microtubule-dependent retrograde transport of proteins into the ER in the presence of brefeldin A suggests an ER recycling pathway. Cell, 60(5):821–836. Lippincott-Schwartz, J., Yuan, L. C., Bonifacino, J. S., and Klausner, R. D. (1989). Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell, 56(5):801–813. Liu, F., Dowling, M., Yang, X.-J., and Kao, G. D. (2004). Caspase-mediated specific cleavage of human histone deacetylase 4. J Biol Chem, 279(33):34537–34546. Liu, S., Bromley-Brits, K., Xia, K., Mittelholtz, J., Wang, R., and Song, W. (2008). TMP21 degradation is mediated by the ubiquitin-proteasome pathway. Eur J Neurosci, 28(10):1980–1988. Liu, S., Zhang, S., Bromley-Brits, K., Cai, F., Zhou, W., Xia, K., Mittelholtz, J., and Song, W. (2011). Transcriptional Regulation of TMP21 by NFAT. Mol Neurodegener, 6(1):21. Logeat, F., Bessia, C., Brou, C., LeBail, O., Jarriault, S., Seidah, N. G., and Israël, A. (1998). The Notch1 receptor is cleaved constitutively by a furin-like convertase. Proc Natl Acad Sci U S A, 95(14):8108–8112. Logue, S. F., Paylor, R., and Wehner, J. M. (1997). Hippocampal lesions cause learning deficits in inbred mice in the Morris water maze and conditioned-fear task. Behav Neurosci, 111(1):104–113. López-Schier, H. and Johnston, D. S. (2002). Drosophila nicastrin is essential for the intramembranous cleavage of notch. Dev Cell, 2(1):79–89. Lowe, M. and Kreis, T. E. (1995). In vitro assembly and disassembly of coatomer. J Biol Chem, 270(52):31364–31371. Lowe, M. and Kreis, T. E. (1996). In vivo assembly of coatomer, the COP-I coat precursor. J Biol Chem, 271(48):30725–30730. 151 Bibliography Luo, W., Wang, H., Li, H., Kim, B. S., Shah, S., Lee, H.-J., Thinakaran, G., Kim, T.-W., Yu, G., and Xu, H. (2003). PEN-2 and APH-1 coordinately regulate proteolytic processing of presenilin 1. J Biol Chem, 278(10):7850–7854. Luo, Y., Bolon, B., Kahn, S., Bennett, B. D., Babu-Khan, S., Denis, P., Fan, W., Kha, H., Zhang, J., Gong, Y., Martin, L., Louis, J. C., Yan, Q., Richards, W. G., Citron, M., and Vassar, R. (2001). Mice deficient in BACE1, the Alzheimer’s beta-secretase, have normal phenotype and abolished beta-amyloid generation. Nat Neurosci, 4(3):231–232. Majoul, I., Straub, M., Hell, S. W., Duden, R., and Söling, H. D. (2001). KDEL- cargo regulates interactions between proteins involved in COPI vesicle traffic: measurements in living cells using FRET. Dev Cell, 1(1):139–153. Malhotra, V., Serafini, T., Orci, L., Shepherd, J. C., and Rothman, J. E. (1989). Purification of a novel class of coated vesicles mediating biosynthetic protein trans- port through the Golgi stack. Cell, 58(2):329–336. Manetto, V., Perry, G., Tabaton, M., Mulvihill, P., Fried, V. A., Smith, H. T., Gambetti, P., and Autilio-Gambetti, L. (1988). Ubiquitin is associated with ab- normal cytoplasmic filaments characteristic of neurodegenerative diseases. Proc Natl Acad Sci U S A, 85(12):4501–4505. Mansour, S. J., Skaug, J., Zhao, X. H., Giordano, J., Scherer, S. W., and Melançon, P. (1999). p200 ARF-GEP1: a Golgi-localized guanine nucleotide exchange pro- tein whose Sec7 domain is targeted by the drug brefeldin A. Proc Natl Acad Sci U S A, 96(14):7968–7973. Marambaud, P., Shioi, J., Serban, G., Georgakopoulos, A., Sarner, S., Nagy, V., Baki, L., Wen, P., Efthimiopoulos, S., Shao, Z., Wisniewski, T., and Robakis, N. K. (2002). A presenilin-1/gamma-secretase cleavage releases the E-cadherin intracellular domain and regulates disassembly of adherens junctions. EMBO J, 21(8):1948–1956. Marambaud, P., Wen, P. H., Dutt, A., Shioi, J., Takashima, A., Siman, R., and Robakis, N. K. (2003). A CBP binding transcriptional repressor produced by the PS1/epsilon-cleavage of N-cadherin is inhibited by PS1 FAD mutations. Cell, 114(5):635–645. Martin, B. L., Schrader-Fischer, G., Busciglio, J., Duke, M., Paganetti, P., and Yankner, B. A. (1995). Intracellular accumulation of beta-amyloid in cells express- ing the Swedish mutant amyloid precursor protein. J Biol Chem, 270(45):26727– 26730. Marzioch, M., Henthorn, D. C., Herrmann, J. M., Wilson, R., Thomas, D. Y., Bergeron, J. J., Solari, R. C., and Rowley, A. (1999). Erp1p and Erp2p, partners for Emp24p and Erv25p in a yeast p24 complex. Mol Biol Cell, 10(6):1923–1938. 152 Bibliography Masters, C. L., Simms, G., Weinman, N. A., Multhaup, G., McDonald, B. L., and Beyreuther, K. (1985). Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci U S A, 82(12):4245–4249. May, P., Reddy, Y. K., and Herz, J. (2002). Proteolytic processing of low density lipoprotein receptor-related protein mediates regulated release of its intracellular domain. J Biol Chem, 277(21):18736–18743. Melançon, P., Glick, B. S., Malhotra, V., Weidman, P. J., Serafini, T., Gleason, M. L., Orci, L., and Rothman, J. E. (1987). Involvement of GTP-binding “G” proteins in transport through the Golgi stack. Cell, 51(6):1053–1062. Meziane, H., Ouagazzal, A.-M., Aubert, L., Wietrzych, M., and Krezel, W. (2007). Estrous cycle effects on behavior of C57BL/6J and BALB/cByJ female mice: implications for phenotyping strategies. Genes Brain Behav, 6(2):192–200. Moelleken, J., Malsam, J., Betts, M. J., Movafeghi, A., Reckmann, I., Meissner, I., Hellwig, A., Russell, R. B., Sllner, T., Brügger, B., and Wieland, F. T. (2007). Differential localization of coatomer complex isoforms within the Golgi apparatus. Proc Natl Acad Sci U S A, 104(11):4425–4430. Mori, H., Kondo, J., and Ihara, Y. (1987). Ubiquitin is a component of paired helical filaments in Alzheimer’s disease. Science, 235(4796):1641–1644. Morinaga, N., Adamik, R., Moss, J., and Vaughan, M. (1999). Brefeldin A inhibited activity of the sec7 domain of p200, a mammalian guanine nucleotide-exchange protein for ADP-ribosylation factors. J Biol Chem, 274(25):17417–17423. Morinaga, N., Moss, J., and Vaughan, M. (1997). Cloning and expression of a cDNA encoding a bovine brain brefeldin A-sensitive guanine nucleotide-exchange protein for ADP-ribosylation factor. Proc Natl Acad Sci U S A, 94(24):12926–12931. Morinaga, N., Tsai, S. C., Moss, J., and Vaughan, M. (1996). Isolation of a brefeldin A-inhibited guanine nucleotide-exchange protein for ADP ribosylation factor (ARF) 1 and ARF3 that contains a Sec7-like domain. Proc Natl Acad Sci U S A, 93(23):12856–12860. Morris, R. (1981). Spatial localisation does not depend on the presence of local cues. Learning Motivation, 12:239–261. Morris, R. G., Garrud, P., Rawlins, J. N., and O’Keefe, J. (1982). Place navigation impaired in rats with hippocampal lesions. Nature, 297(5868):681–683. Mullan, M., Houlden, H., Windelspecht, M., Fidani, L., Lombardi, C., Diaz, P., Rossor, M., Crook, R., Hardy, J., and Duff, K. (1992). A locus for familial early- onset Alzheimer’s disease on the long arm of chromosome 14, proximal to the alpha 1-antichymotrypsin gene. Nat Genet, 2(4):340–342. 153 Bibliography Mumm, J. S., Schroeter, E. H., Saxena, M. T., Griesemer, A., Tian, X., Pan, D. J., Ray, W. J., and Kopan, R. (2000). A ligand-induced extracellular cleav- age regulates gamma-secretase-like proteolytic activation of Notch1. Mol Cell, 5(2):197–206. Nakamura, N., Lowe, M., Levine, T. P., Rabouille, C., and Warren, G. (1997). The vesicle docking protein p115 binds GM130, a cis-Golgi matrix protein, in a mitotically regulated manner. Cell, 89(3):445–455. Nakamura, N., Rabouille, C., Watson, R., Nilsson, T., Hui, N., Slusarewicz, P., Kreis, T. E., and Warren, G. (1995). Characterization of a cis-Golgi matrix protein, GM130. J Cell Biol, 131(6 Pt 2):1715–1726. Narula, N., McMorrow, I., Plopper, G., Doherty, J., Matlin, K. S., Burke, B., and Stow, J. L. (1992). Identification of a 200-kD, brefeldin-sensitive protein on Golgi membranes. J Cell Biol, 117(1):27–38. Narula, N. and Stow, J. L. (1995). Distinct coated vesicles labeled for p200 bud from trans-Golgi network membranes. Proc Natl Acad Sci U S A, 92(7):2874–2878. Naruse, S., Thinakaran, G., Luo, J. J., Kusiak, J. W., Tomita, T., Iwatsubo, T., Qian, X., Ginty, D. D., Price, D. L., Borchelt, D. R., Wong, P. C., and Sisodia, S. S. (1998). Effects of PS1 deficiency on membrane protein trafficking in neurons. Neuron, 21(5):1213–1221. Näslund, J., Schierhorn, A., Hellman, U., Lannfelt, L., Roses, A. D., Tjernberg, L. O., Silberring, J., Gandy, S. E., Winblad, B., and Greengard, P. (1994). Rela- tive abundance of Alzheimer A beta amyloid peptide variants in Alzheimer disease and normal aging. Proc Natl Acad Sci U S A, 91(18):8378–8382. Ni, C. Y., Murphy, M. P., Golde, T. E., and Carpenter, G. (2001). gamma - Secretase cleavage and nuclear localization of ErbB-4 receptor tyrosine kinase. Science, 294(5549):2179–2181. Nickel, W., Sohn, K., Bünning, C., and Wieland, F. T. (1997). p23, a major COPI- vesicle membrane protein, constitutively cycles through the early secretory path- way. Proc Natl Acad Sci U S A, 94(21):11393–11398. Nilsson, T., Jackson, M., and Peterson, P. A. (1989). Short cytoplasmic sequences serve as retention signals for transmembrane proteins in the endoplasmic reticu- lum. Cell, 58(4):707–718. Niu, T.-K., Pfeifer, A. C., Lippincott-Schwartz, J., and Jackson, C. L. (2005). Dy- namics of GBF1, a Brefeldin A-sensitive Arf1 exchange factor at the Golgi. Mol Biol Cell, 16(3):1213–1222. 154 Bibliography Nordstedt, C., Caporaso, G. L., Thyberg, J., Gandy, S. E., and Greengard, P. (1993). Identification of the Alzheimer beta/A4 amyloid precursor protein in clathrin-coated vesicles purified from PC12 cells. J Biol Chem, 268(1):608–612. Nunan, J., Shearman, M. S., Checler, F., Cappai, R., Evin, G., Beyreuther, K., Mas- ters, C. L., and Small, D. H. (2001). The C-terminal fragment of the Alzheimer’s disease amyloid protein precursor is degraded by a proteasome-dependent mech- anism distinct from gamma-secretase. Eur J Biochem, 268(20):5329–5336. Ohkuma, S. and Poole, B. (1978). Fluorescence probe measurement of the intralyso- somal pH in living cells and the perturbation of pH by various agents. Proc Natl Acad Sci U S A, 75(7):3327–3331. Oltersdorf, T., Ward, P. J., Henriksson, T., Beattie, E. C., Neve, R., Lieberburg, I., and Fritz, L. C. (1990). The Alzheimer amyloid precursor protein. Identification of a stable intermediate in the biosynthetic/degradative pathway. J Biol Chem, 265(8):4492–4497. Omura, S., Fujimoto, T., Otoguro, K., Matsuzaki, K., Moriguchi, R., Tanaka, H., and Sasaki, Y. (1991). Lactacystin, a novel microbial metabolite, induces neuri- togenesis of neuroblastoma cells. J Antibiot (Tokyo), 44(1):113–116. Ooi, C. E., Dell’Angelica, E. C., and Bonifacino, J. S. (1998). ADP-Ribosylation fac- tor 1 (ARF1) regulates recruitment of the AP-3 adaptor complex to membranes. J Cell Biol, 142(2):391–402. Orci, L., Glick, B. S., and Rothman, J. E. (1986). A new type of coated vesicular carrier that appears not to contain clathrin: its possible role in protein transport within the Golgi stack. Cell, 46(2):171–184. Orci, L., Palmer, D. J., Amherdt, M., and Rothman, J. E. (1993a). Coated vesicle assembly in the Golgi requires only coatomer and ARF proteins from the cytosol. Nature, 364(6439):732–734. Orci, L., Palmer, D. J., Ravazzola, M., Perrelet, A., Amherdt, M., and Rothman, J. E. (1993b). Budding from Golgi membranes requires the coatomer complex of non-clathrin coat proteins. Nature, 362(6421):648–652. Orci, L., Ravazzola, M., Volchuk, A., Engel, T., Gmachl, M., Amherdt, M., Perrelet, A., Sollner, T. H., and Rothman, J. E. (2000). Anterograde flow of cargo across the golgi stack potentially mediated via bidirectional ”percolating” COPI vesicles. Proc Natl Acad Sci U S A, 97(19):10400–10405. Orci, L., Stamnes, M., Ravazzola, M., Amherdt, M., Perrelet, A., Söllner, T. H., and Rothman, J. E. (1997). Bidirectional transport by distinct populations of COPI-coated vesicles. Cell, 90(2):335–349. 155 Bibliography Ostermann, J., Orci, L., Tani, K., Amherdt, M., Ravazzola, M., Elazar, Z., and Rothman, J. E. (1993). Stepwise assembly of functionally active transport vesicles. Cell, 75(5):1015–1025. Palade, G. (1975). Intracellular aspects of the process of protein synthesis. Science, 189(4200):347–358. Palmer, D. J., Helms, J. B., Beckers, C. J., Orci, L., and Rothman, J. E. (1993). Binding of coatomer to Golgi membranes requires ADP-ribosylation factor. J Biol Chem, 268(16):12083–12089. Pardossi-Piquard, R., Böhm, C., Chen, F., Kanemoto, S., Checler, F., Schmitt- Ulms, G., George-Hyslop, P. S., and Fraser, P. E. (2009). TMP21 transmembrane domain regulates gamma-secretase cleavage. J Biol Chem, 284(42):28634–28641. Pavel, J., Harter, C., and Wieland, F. T. (1998). Reversible dissociation of coatomer: functional characterization of a beta/delta-coat protein subcomplex. Proc Natl Acad Sci U S A, 95(5):2140–2145. Paylor, R., Spencer, C. M., Yuva-Paylor, L. A., and Pieke-Dahl, S. (2006). The use of behavioral test batteries, II: effect of test interval. Physiol Behav, 87(1):95–102. Pepperkok, R., Scheel, J., Horstmann, H., Hauri, H. P., Griffiths, G., and Kreis, T. E. (1993). Beta-COP is essential for biosynthetic membrane transport from the endoplasmic reticulum to the Golgi complex in vivo. Cell, 74(1):71–82. Perez, R. G., Soriano, S., Hayes, J. D., Ostaszewski, B., Xia, W., Selkoe, D. J., Chen, X., Stokin, G. B., and Koo, E. H. (1999). Mutagenesis identifies new signals for beta-amyloid precursor protein endocytosis, turnover, and the generation of secreted fragments, including Abeta42. J Biol Chem, 274(27):18851–18856. Perez, R. G., Squazzo, S. L., and Koo, E. H. (1996). Enhanced release of amyloid beta-protein from codon 670/671 ”Swedish” mutant beta-amyloid precursor pro- tein occurs in both secretory and endocytic pathways. J Biol Chem, 271(15):9100– 9107. Perry, G., Friedman, R., Shaw, G., and Chau, V. (1987). Ubiquitin is detected in neurofibrillary tangles and senile plaque neurites of Alzheimer disease brains. Proc Natl Acad Sci U S A, 84(9):3033–3036. Peter, F., Plutner, H., Zhu, H., Kreis, T. E., and Balch, W. E. (1993). Beta-COP is essential for transport of protein from the endoplasmic reticulum to the Golgi in vitro. J Cell Biol, 122(6):1155–1167. Petit, A., Bihel, F., Alvès da Costa, C., Pourquié, O., Checler, F., and Kraus, J. L. (2001). New protease inhibitors prevent gamma-secretase-mediated production of Abeta40/42 without affecting Notch cleavage. Nat Cell Biol, 3(5):507–511. 156 Bibliography Phillips, R. G. and LeDoux, J. E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci, 106(2):274–285. Ponder, C. A., Kliethermes, C. L., Drew, M. R., Muller, J., Das, K., Risbrough, V. B., Crabbe, J. C., Gilliam, T. C., and Palmer, A. A. (2007). Selection for contextual fear conditioning affects anxiety-like behaviors and gene expression. Genes Brain Behav, 6(8):736–749. Ponte, P., Gonzalez-DeWhitt, P., Schilling, J., Miller, J., Hsu, D., Greenberg, B., Davis, K., Wallace, W., Lieberburg, I., and Fuller, F. (1988). A new A4 amyloid mRNA contains a domain homologous to serine proteinase inhibitors. Nature, 331(6156):525–527. Qing, H., Zhou, W., Christensen, M. A., Sun, X., Tong, Y., and Song, W. (2004). Degradation of BACE by the ubiquitin-proteasome pathway. FASEB J, 18(13):1571–1573. Rand, M. D., Grimm, L. M., Artavanis-Tsakonas, S., Patriub, V., Blacklow, S. C., Sklar, J., and Aster, J. C. (2000). Calcium depletion dissociates and activates heterodimeric notch receptors. Mol Cell Biol, 20(5):1825–1835. Ratovitski, T., Slunt, H. H., Thinakaran, G., Price, D. L., Sisodia, S. S., and Borchelt, D. R. (1997). Endoproteolytic processing and stabilization of wild-type and mutant presenilin. J Biol Chem, 272(39):24536–24541. Réchards, M., Xia, W., Oorschot, V. M. J., Selkoe, D. J., and Klumperman, J. (2003). Presenilin-1 exists in both pre- and post-Golgi compartments and recycles via COPI-coated membranes. Traffic, 4(8):553–565. Reinhard, C., Harter, C., Bremser, M., Brügger, B., Sohn, K., Helms, J. B., and Wieland, F. (1999). Receptor-induced polymerization of coatomer. Proc Natl Acad Sci U S A, 96(4):1224–1228. Reisel, D., Bannerman, D. M., Schmitt, W. B., Deacon, R. M. J., Flint, J., Bor- chardt, T., Seeburg, P. H., and Rawlins, J. N. P. (2002). Spatial memory dissoci- ations in mice lacking GluR1. Nat Neurosci, 5(9):868–873. Roberds, S. L., Anderson, J., Basi, G., Bienkowski, M. J., Branstetter, D. G., Chen, K. S., Freedman, S. B., Frigon, N. L., Games, D., Hu, K., Johnson-Wood, K., Kappenman, K. E., Kawabe, T. T., Kola, I., Kuehn, R., Lee, M., Liu, W., Motter, R., Nichols, N. F., Power, M., Robertson, D. W., Schenk, D., Schoor, M., Shopp, G. M., Shuck, M. E., Sinha, S., Svensson, K. A., Tatsuno, G., Tintrup, H., Wijsman, J., Wright, S., and McConlogue, L. (2001). BACE knockout mice are healthy despite lacking the primary beta-secretase activity in brain: implications for Alzheimer’s disease therapeutics. Hum Mol Genet, 10(12):1317–1324. 157 Bibliography Rock, K. L., Gramm, C., Rothstein, L., Clark, K., Stein, R., Dick, L., Hwang, D., and Goldberg, A. L. (1994). Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class I molecules. Cell, 78(5):761–771. Rogaev, E. I., Sherrington, R., Rogaeva, E. A., Levesque, G., Ikeda, M., Liang, Y., Chi, H., Lin, C., Holman, K., and Tsuda, T. (1995). Familial Alzheimer’s disease in kindreds with missense mutations in a gene on chromosome 1 related to the Alzheimer’s disease type 3 gene. Nature, 376(6543):775–778. Rojo, M., Emery, G., Marjomäki, V., McDowall, A. W., Parton, R. G., and Gruen- berg, J. (2000). The transmembrane protein p23 contributes to the organization of the Golgi apparatus. J Cell Sci, 113 ( Pt 6):1043–1057. Rojo, M., Pepperkok, R., Emery, G., Kellner, R., Stang, E., Parton, R. G., and Gru- enberg, J. (1997). Involvement of the transmembrane protein p23 in biosynthetic protein transport. J Cell Biol, 139(5):1119–1135. Schellenberg, G. D., Bird, T. D., Wijsman, E. M., Orr, H. T., Anderson, L., Ne- mens, E., White, J. A., Bonnycastle, L., Weber, J. L., and Alonso, M. E. (1992). Genetic linkage evidence for a familial Alzheimer’s disease locus on chromosome 14. Science, 258(5082):668–671. Scheuner, D., Eckman, C., Jensen, M., Song, X., Citron, M., Suzuki, N., Bird, T. D., Hardy, J., Hutton, M., Kukull, W., Larson, E., Levy-Lahad, E., Viitanen, M., Peskind, E., Poorkaj, P., Schellenberg, G., Tanzi, R., Wasco, W., Lannfelt, L., Selkoe, D., and Younkin, S. (1996). Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nat Med, 2(8):864–870. Schleifer, L. S., Kahn, R. A., Hanski, E., Northup, J. K., Sternweis, P. C., and Gilman, A. G. (1982). Requirements for cholera toxin-dependent ADP- ribosylation of the purified regulatory component of adenylate cyclase. J Biol Chem, 257(1):20–23. Schoenfeld, A. R., Davidowitz, E. J., and Burk, R. D. (2000). Elongin BC complex prevents degradation of von Hippel-Lindau tumor suppressor gene products. Proc Natl Acad Sci U S A, 97(15):8507–8512. Schroeter, E. H., Kisslinger, J. A., and Kopan, R. (1998). Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain. Nature, 393(6683):382– 386. Seeger, M., Nordstedt, C., Petanceska, S., Kovacs, D. M., Gouras, G. K., Hahne, S., Fraser, P., Levesque, L., Czernik, A. J., George-Hyslop, P. S., Sisodia, S. S., 158 Bibliography Thinakaran, G., Tanzi, R. E., Greengard, P., and Gandy, S. (1997). Evidence for phosphorylation and oligomeric assembly of presenilin 1. Proc Natl Acad Sci U S A, 94(10):5090–5094. Seemann, J., Jokitalo, E., Pypaert, M., and Warren, G. (2000). Matrix pro- teins can generate the higher order architecture of the Golgi apparatus. Nature, 407(6807):1022–1026. Selkoe, D. and Kopan, R. (2003). Notch and Presenilin: regulated intramembrane proteolysis links development and degeneration. Annu Rev Neurosci, 26:565–597. Selkoe, D. J., Podlisny, M. B., Joachim, C. L., Vickers, E. A., Lee, G., Fritz, L. C., and Oltersdorf, T. (1988). Beta-amyloid precursor protein of Alzheimer disease occurs as 110- to 135-kilodalton membrane-associated proteins in neural and non- neural tissues. Proc Natl Acad Sci U S A, 85(19):7341–7345. Serafini, T., Orci, L., Amherdt, M., Brunner, M., Kahn, R. A., and Rothman, J. E. (1991a). ADP-ribosylation factor is a subunit of the coat of Golgi-derived COP-coated vesicles: a novel role for a GTP-binding protein. Cell, 67(2):239–253. Serafini, T., Stenbeck, G., Brecht, A., Lottspeich, F., Orci, L., Rothman, J. E., and Wieland, F. T. (1991b). A coat subunit of Golgi-derived non-clathrin-coated vesicles with homology to the clathrin-coated vesicle coat protein beta-adaptin. Nature, 349(6306):215–220. Shah, S., Lee, S.-F., Tabuchi, K., Hao, Y.-H., Yu, C., LaPlant, Q., Ball, H., Dann, C. E., Südhof, T., and Yu, G. (2005). Nicastrin functions as a gamma-secretase- substrate receptor. Cell, 122(3):435–447. Shaw, G. and Chau, V. (1988). Ubiquitin and microtubule-associated protein tau immunoreactivity each define distinct structures with differing distributions and solubility properties in Alzheimer brain. Proc Natl Acad Sci U S A, 85(8):2854– 2858. Shen, J., Bronson, R. T., Chen, D. F., Xia, W., Selkoe, D. J., and Tonegawa, S. (1997). Skeletal and CNS defects in Presenilin-1-deficient mice. Cell, 89(4):629– 639. Sherrington, R., Rogaev, E. I., Liang, Y., Rogaeva, E. A., Levesque, G., Ikeda, M., Chi, H., Lin, C., Li, G., and Holman, K. (1995). Cloning of a gene bearing missense mutations in early-onset familial Alzheimer’s disease. Nature, 375(6534):754–760. Shinotsuka, C., Yoshida, Y., Kawamoto, K., Takatsu, H., and Nakayama, K. (2002). Overexpression of an ADP-ribosylation factor-guanine nucleotide exchange fac- tor, BIG2, uncouples brefeldin A-induced adaptor protein-1 coat dissociation and membrane tubulation. J Biol Chem, 277(11):9468–9473. 159 Bibliography Shorter, J., Beard, M. B., Seemann, J., Dirac-Svejstrup, A. B., and Warren, G. (2002). Sequential tethering of Golgins and catalysis of SNAREpin assembly by the vesicle-tethering protein p115. J Cell Biol, 157(1):45–62. Shorter, J. and Warren, G. (1999). A role for the vesicle tethering protein, p115, in the post-mitotic stacking of reassembling Golgi cisternae in a cell-free system. J Cell Biol, 146(1):57–70. Sinha, S., Anderson, J. P., Barbour, R., Basi, G. S., Caccavello, R., Davis, D., Doan, M., Dovey, H. F., Frigon, N., Hong, J., Jacobson-Croak, K., Jewett, N., Keim, P., Knops, J., Lieberburg, I., Power, M., Tan, H., Tatsuno, G., Tung, J., Schenk, D., Seubert, P., Suomensaari, S. M., Wang, S., Walker, D., Zhao, J., McConlogue, L., and John, V. (1999). Purification and cloning of amyloid precursor protein beta-secretase from human brain. Nature, 402(6761):537–540. Sisodia, S. S., Koo, E. H., Beyreuther, K., Unterbeck, A., and Price, D. L. (1990). Evidence that beta-amyloid protein in Alzheimer’s disease is not derived by nor- mal processing. Science, 248(4954):492–495. Six, E., Ndiaye, D., Laabi, Y., Brou, C., Gupta-Rossi, N., Israel, A., and Logeat, F. (2003). The Notch ligand Delta1 is sequentially cleaved by an ADAM protease and gamma-secretase. Proc Natl Acad Sci U S A, 100(13):7638–7643. Sohn, K., Orci, L., Ravazzola, M., Amherdt, M., Bremser, M., Lottspeich, F., Fiedler, K., Helms, J. B., and Wieland, F. T. (1996). A major transmembrane protein of Golgi-derived COPI-coated vesicles involved in coatomer binding. J Cell Biol, 135(5):1239–1248. Söllner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E. (1993). SNAP receptors implicated in vesicle targeting and fusion. Nature, 362(6418):318–324. Song, W., Nadeau, P., Yuan, M., Yang, X., Shen, J., and Yankner, B. A. (1999). Proteolytic release and nuclear translocation of Notch-1 are induced by presenilin- 1 and impaired by pathogenic presenilin-1 mutations. Proc Natl Acad Sci U S A, 96(12):6959–6963. Sönnichsen, B., Lowe, M., Levine, T., Jämsä, E., Dirac-Svejstrup, B., and Warren, G. (1998). A role for giantin in docking COPI vesicles to Golgi membranes. J Cell Biol, 140(5):1013–1021. Springer, S., Chen, E., Duden, R., Marzioch, M., Rowley, A., Hamamoto, S., Mer- chant, S., and Schekman, R. (2000). The p24 proteins are not essential for vesicu- lar transport in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A, 97(8):4034– 4039. 160 Bibliography Stamnes, M. A., Craighead, M. W., Hoe, M. H., Lampen, N., Geromanos, S., Tempst, P., and Rothman, J. E. (1995). An integral membrane component of coatomer-coated transport vesicles defines a family of proteins involved in bud- ding. Proc Natl Acad Sci U S A, 92(17):8011–8015. Stamnes, M. A. and Rothman, J. E. (1993). The binding of AP-1 clathrin adaptor particles to Golgi membranes requires ADP-ribosylation factor, a small GTP- binding protein. Cell, 73(5):999–1005. Steiner, H., Capell, A., Pesold, B., Citron, M., Kloetzel, P. M., Selkoe, D. J., Romig, H., Mendla, K., and Haass, C. (1998). Expression of Alzheimer’s disease- associated presenilin-1 is controlled by proteolytic degradation and complex for- mation. J Biol Chem, 273(48):32322–32331. Steiner, H., Duff, K., Capell, A., Romig, H., Grim, M. G., Lincoln, S., Hardy, J., Yu, X., Picciano, M., Fechteler, K., Citron, M., Kopan, R., Pesold, B., Keck, S., Baader, M., Tomita, T., Iwatsubo, T., Baumeister, R., and Haass, C. (1999). A loss of function mutation of presenilin-2 interferes with amyloid beta-peptide production and notch signaling. J Biol Chem, 274(40):28669–28673. Steiner, H., Winkler, E., Edbauer, D., Prokop, S., Basset, G., Yamasaki, A., Kostka, M., and Haass, C. (2002). PEN-2 is an integral component of the gamma-secretase complex required for coordinated expression of presenilin and nicastrin. J Biol Chem, 277(42):39062–39065. Stenbeck, G., Harter, C., Brecht, A., Herrmann, D., Lottspeich, F., Orci, L., and Wieland, F. T. (1993). beta’-COP, a novel subunit of coatomer. EMBO J, 12(7):2841–2845. Stephens, D. J. and Austen, B. M. (1996). Metabolites of the beta-amyloid precursor protein generated by beta-secretase localise to the trans-Golgi network and late endosome in 293 cells. J Neurosci Res, 46(2):211–225. Strous, G. J., Berger, E. G., van Kerkhof, P., Bosshart, H., Berger, B., and Geuze, H. J. (1991). Brefeldin A induces a microtubule-dependent fusion of galactosyltransferase-containing vesicles with the rough endoplasmic reticulum. Biol Cell, 71(1-2):25–31. Sturchler-Pierrat, C., Abramowski, D., Duke, M., Wiederhold, K. H., Mistl, C., Rothacher, S., Ledermann, B., Bürki, K., Frey, P., Paganetti, P. A., Waridel, C., Calhoun, M. E., Jucker, M., Probst, A., Staufenbiel, M., and Sommer, B. (1997). Two amyloid precursor protein transgenic mouse models with Alzheimer disease-like pathology. Proc Natl Acad Sci U S A, 94(24):13287–13292. 161 Bibliography Sun, X., He, G., Qing, H., Zhou, W., Dobie, F., Cai, F., Staufenbiel, M., Huang, L. E., and Song, W. (2006a). Hypoxia facilitates Alzheimer’s disease patho- genesis by up-regulating BACE1 gene expression. Proc Natl Acad Sci U S A, 103(49):18727–18732. Sun, X., He, G., and Song, W. (2006b). BACE2, as a novel APP theta-secretase, is not responsible for the pathogenesis of Alzheimer’s disease in Down syndrome. FASEB J, 20(9):1369–1376. Sun, X., Tong, Y., Qing, H., Chen, C.-H., and Song, W. (2006c). Increased BACE1 maturation contributes to the pathogenesis of Alzheimer’s disease in Down syn- drome. FASEB J, 20(9):1361–1368. Sun, X., Wang, Y., Qing, H., Christensen, M. A., Liu, Y., Zhou, W., Tong, Y., Xiao, C., Huang, Y., Zhang, S., Liu, X., and Song, W. (2005). Distinct transcriptional regulation and function of the human BACE2 and BACE1 genes. FASEB J, 19(7):739–749. Sun, Z., Anderl, F., Fröhlich, K., Zhao, L., Hanke, S., Brügger, B., Wieland, F., and Béthune, J. (2007). Multiple and stepwise interactions between coatomer and ADP-ribosylation factor-1 (Arf1)-GTP. Traffic, 8(5):582–593. Sztul, E., Colombo, M., Stahl, P., and Samanta, R. (1993). Control of protein traffic between distinct plasma membrane domains. Requirement for a novel 108,000 protein in the fusion of transcytotic vesicles with the apical plasma membrane. J Biol Chem, 268(3):1876–1885. Sztul, E., Kaplin, A., Saucan, L., and Palade, G. (1991). Protein traffic between distinct plasma membrane domains: isolation and characterization of vesicular carriers involved in transcytosis. Cell, 64(1):81–89. Szul, T., Garcia-Mata, R., Brandon, E., Shestopal, S., Alvarez, C., and Sztul, E. (2005). Dissection of membrane dynamics of the ARF-guanine nucleotide ex- change factor GBF1. Traffic, 6(5):374–385. Takahashi, E., Niimi, K., and Itakura, C. (2009). Motor coordination impairment in aged heterozygous rolling Nagoya, Cav2.1 mutant mice. Brain Res, 1279:50–57. Takasugi, N., Tomita, T., Hayashi, I., Tsuruoka, M., Niimura, M., Takahashi, Y., Thinakaran, G., and Iwatsubo, T. (2003). The role of presenilin cofactors in the gamma-secretase complex. Nature, 422(6930):438–441. Tanzi, R. E., McClatchey, A. I., Lamperti, E. D., Villa-Komaroff, L., Gusella, J. F., and Neve, R. L. (1988). Protease inhibitor domain encoded by an amyloid protein precursor mRNA associated with Alzheimer’s disease. Nature, 331(6156):528–530. 162 Bibliography Tartakoff, A. M. (1983). Perturbation of vesicular traffic with the carboxylic ionophore monensin. Cell, 32(4):1026–1028. Teranishi, Y., Hur, J.-Y., Welander, H., Fr̊anberg, J., Aoki, M., Winblad, B., Fryk- man, S., and Tjernberg, L. O. (2009). Affinity pulldown of gamma-secretase and associated proteins from human and rat brain. J Cell Mol Med. Thinakaran, G., Harris, C. L., Ratovitski, T., Davenport, F., Slunt, H. H., Price, D. L., Borchelt, D. R., and Sisodia, S. S. (1997). Evidence that levels of presenilins (PS1 and PS2) are coordinately regulated by competition for limiting cellular factors. J Biol Chem, 272(45):28415–28422. Thinakaran, G., Teplow, D. B., Siman, R., Greenberg, B., and Sisodia, S. S. (1996). Metabolism of the ”Swedish” amyloid precursor protein variant in neuro2a (N2a) cells. Evidence that cleavage at the ”beta-secretase” site occurs in the golgi ap- paratus. J Biol Chem, 271(16):9390–9397. Togawa, A., Morinaga, N., Ogasawara, M., Moss, J., and Vaughan, M. (1999). Purification and cloning of a brefeldin A-inhibited guanine nucleotide-exchange protein for ADP-ribosylation factors. J Biol Chem, 274(18):12308–12315. Tong, Y., Zhou, W., Fung, V., Christensen, M. A., Qing, H., Sun, X., and Song, W. (2005). Oxidative stress potentiates BACE1 gene expression and Abeta genera- tion. J Neural Transm, 112(3):455–469. Touitou, R., Richardson, J., Bose, S., Nakanishi, M., Rivett, J., and Allday, M. J. (2001). A degradation signal located in the C-terminus of p21WAF1/CIP1 is a binding site for the C8 alpha-subunit of the 20S proteasome. EMBO J, 20(10):2367–2375. Traub, L. M., Ostrom, J. A., and Kornfeld, S. (1993). Biochemical dissection of AP-1 recruitment onto Golgi membranes. J Cell Biol, 123(3):561–573. Tsuchiya, M., Price, S. R., Tsai, S. C., Moss, J., and Vaughan, M. (1991). Molecular identification of ADP-ribosylation factor mRNAs and their expression in mam- malian cells. J Biol Chem, 266(5):2772–2777. Võikar, V., Kõks, S., Vasar, E., and Rauvala, H. (2001). Strain and gender differ- ences in the behavior of mouse lines commonly used in transgenic studies. Physiol Behav, 72(1-2):271–281. van Gaalen, M. M. and Steckler, T. (2000). Behavioural analysis of four mouse strains in an anxiety test battery. Behav Brain Res, 115(1):95–106. Vassar, R., Bennett, B. D., Babu-Khan, S., Kahn, S., Mendiaz, E. A., Denis, P., Teplow, D. B., Ross, S., Amarante, P., Loeloff, R., Luo, Y., Fisher, S., Fuller, 163 Bibliography J., Edenson, S., Lile, J., Jarosinski, M. A., Biere, A. L., Curran, E., Burgess, T., Louis, J. C., Collins, F., Treanor, J., Rogers, G., and Citron, M. (1999). Beta- secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science, 286(5440):735–741. Vetrivel, K. S., Gong, P., Bowen, J. W., Cheng, H., Chen, Y., Carter, M., Nguyen, P. D., Placanica, L., Wieland, F. T., Li, Y.-M., Kounnas, M. Z., and Thinakaran, G. (2007). Dual roles of the transmembrane protein p23/TMP21 in the modula- tion of amyloid precursor protein metabolism. Mol Neurodegener, 2:4. Vetrivel, K. S., Kodam, A., Gong, P., Chen, Y., Parent, A. T., Kar, S., and Thi- nakaran, G. (2008a). Localization and regional distribution of p23/TMP21 in the brain. Neurobiol Dis, 32(1):37–49. Vetrivel, K. S., Zhang, X., Meckler, X., Cheng, H., Lee, S., Gong, P., Lopes, K. O., Chen, Y., Iwata, N., Yin, K.-J., Lee, J.-M., Parent, A. T., Saido, T. C., Li, Y.-M., Sisodia, S. S., and Thinakaran, G. (2008b). Evidence that CD147 modulation of beta-amyloid (Abeta) levels is mediated by extracellular degradation of secreted Abeta. J Biol Chem, 283(28):19489–19498. Ward, C. L., Omura, S., and Kopito, R. R. (1995). Degradation of CFTR by the ubiquitin-proteasome pathway. Cell, 83(1):121–127. Waters, M. G., Clary, D. O., and Rothman, J. E. (1992). A novel 115-kD peripheral membrane protein is required for intercisternal transport in the Golgi stack. J Cell Biol, 118(5):1015–1026. Waters, M. G., Serafini, T., and Rothman, J. E. (1991). ’Coatomer’: a cytosolic pro- tein complex containing subunits of non-clathrin-coated Golgi transport vesicles. Nature, 349(6306):248–251. Watson, P. J., Frigerio, G., Collins, B. M., Duden, R., and Owen, D. J. (2004). Gamma-COP appendage domain - structure and function. Traffic, 5(2):79–88. Wegmann, D., Hess, P., Baier, C., Wieland, F. T., and Reinhard, C. (2004). Novel isotypic gamma/zeta subunits reveal three coatomer complexes in mammals. Mol Cell Biol, 24(3):1070–1080. Weidemann, A., König, G., Bunke, D., Fischer, P., Salbaum, J. M., Masters, C. L., and Beyreuther, K. (1989). Identification, biogenesis, and localization of precur- sors of Alzheimer’s disease A4 amyloid protein. Cell, 57(1):115–126. Weidler, M., Reinhard, C., Friedrich, G., Wieland, F. T., and Rösch, P. (2000). Structure of the cytoplasmic domain of p23 in solution: implications for the for- mation of COPI vesicles. Biochem Biophys Res Commun, 271(2):401–408. 164 Bibliography Weisstaub, N. V., Zhou, M., Lira, A., Lambe, E., González-Maeso, J., Hornung, J.-P., Sibille, E., Underwood, M., Itohara, S., Dauer, W. T., Ansorge, M. S., Morelli, E., Mann, J. J., Toth, M., Aghajanian, G., Sealfon, S. C., Hen, R., and Gingrich, J. A. (2006). Cortical 5-HT2A receptor signaling modulates anxiety-like behaviors in mice. Science, 313(5786):536–540. Wen, C., Metzstein, M. M., and Greenwald, I. (1997). SUP-17, a Caenorhabditis elegans ADAM protein related to Drosophila KUZBANIAN, and its role in LIN- 12/NOTCH signalling. Development, 124(23):4759–4767. Wertkin, A. M., Turner, R. S., Pleasure, S. J., Golde, T. E., Younkin, S. G., Tro- janowski, J. Q., and Lee, V. M. (1993). Human neurons derived from a terato- carcinoma cell line express solely the 695-amino acid amyloid precursor protein and produce intracellular beta-amyloid or A4 peptides. Proc Natl Acad Sci U S A, 90(20):9513–9517. Whishaw, I. Q. and Tomie, J. (1996). Of mice and mazes: similarities between mice and rats on dry land but not water mazes. Physiol Behav, 60(5):1191–1197. Winkler, E., Hobson, S., Fukumori, A., Du?mpelfeld, B., Luebbers, T., Baumann, K., Haass, C., Hopf, C., and Steiner, H. (2009). Purification, Pharmacological Modulation, and Biochemical Characterization of Interactors of Endogenous Hu- man gamma-Secretase (dagger). Biochemistry. Wolfe, M. S., Xia, W., Ostaszewski, B. L., Diehl, T. S., Kimberly, W. T., and Selkoe, D. J. (1999). Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and gamma-secretase activity. Nature, 398(6727):513–517. Wong, P. C., Zheng, H., Chen, H., Becher, M. W., Sirinathsinghji, D. J., Trumbauer, M. E., Chen, H. Y., Price, D. L., der Ploeg, L. H. V., and Sisodia, S. S. (1997). Presenilin 1 is required for Notch1 and DII1 expression in the paraxial mesoderm. Nature, 387(6630):288–292. Wood, J. G., Mirra, S. S., Pollock, N. J., and Binder, L. I. (1986). Neurofib- rillary tangles of Alzheimer disease share antigenic determinants with the axonal microtubule-associated protein tau (tau). Proc Natl Acad Sci U S A, 83(11):4040– 4043. Yamaji, R., Adamik, R., Takeda, K., Togawa, A., Pacheco-Rodriguez, G., Ferrans, V. J., Moss, J., and Vaughan, M. (2000). Identification and localization of two brefeldin A-inhibited guanine nucleotide-exchange proteins for ADP-ribosylation factors in a macromolecular complex. Proc Natl Acad Sci U S A, 97(6):2567–2572. Yamasaki, K., Hayashida, S., Miura, K., Masuzaki, H., Ishimaru, T., Niikawa, N., and Kishino, T. (2000). The novel gene, gamma2-COP (COPG2), in the 7q32 imprinted domain escapes genomic imprinting. Genomics, 68(3):330–335. 165 Bibliography Yan, R., Bienkowski, M. J., Shuck, M. E., Miao, H., Tory, M. C., Pauley, A. M., Brashier, J. R., Stratman, N. C., Mathews, W. R., Buhl, A. E., Carter, D. B., Tomasselli, A. G., Parodi, L. A., Heinrikson, R. L., and Gurney, M. E. (1999). Membrane-anchored aspartyl protease with Alzheimer’s disease beta-secretase ac- tivity. Nature, 402(6761):533–537. Yang, D.-S., Tandon, A., Chen, F., Yu, G., Yu, H., Arawaka, S., Hasegawa, H., Duthie, M., Schmidt, S. D., Ramabhadran, T. V., Nixon, R. A., Mathews, P. M., Gandy, S. E., Mount, H. T. J., George-Hyslop, P. S., and Fraser, P. E. (2002). Ma- ture glycosylation and trafficking of nicastrin modulate its binding to presenilins. J Biol Chem, 277(31):28135–28142. Yu, G., Chen, F., Nishimura, M., Steiner, H., Tandon, A., Kawarai, T., Arawaka, S., Supala, A., Song, Y. Q., Rogaeva, E., Holmes, E., Zhang, D. M., Milman, P., Fraser, P. E., Haass, C., and George-Hyslop, P. S. (2000a). Mutation of conserved aspartates affects maturation of both aspartate mutant and endogenous presenilin 1 and presenilin 2 complexes. J Biol Chem, 275(35):27348–27353. Yu, G., Nishimura, M., Arawaka, S., Levitan, D., Zhang, L., Tandon, A., Song, Y. Q., Rogaeva, E., Chen, F., Kawarai, T., Supala, A., Levesque, L., Yu, H., Yang, D. S., Holmes, E., Milman, P., Liang, Y., Zhang, D. M., Xu, D. H., Sato, C., Rogaev, E., Smith, M., Janus, C., Zhang, Y., Aebersold, R., Farrer, L. S., Sorbi, S., Bruni, A., Fraser, P., and George-Hyslop, P. S. (2000b). Nicastrin mod- ulates presenilin-mediated notch/glp-1 signal transduction and betaAPP process- ing. Nature, 407(6800):48–54. Zhang, Z., Nadeau, P., Song, W., Donoviel, D., Yuan, M., Bernstein, A., and Yankner, B. (2000). Presenilins are required for gamma-secretase cleavage of beta-APP and transmembrane cleavage of Notch-1. Nat Cell Biol, 2:463–465. Zhao, L., Helms, J. B., Brügger, B., Harter, C., Martoglio, B., Graf, R., Brunner, J., and Wieland, F. T. (1997). Direct and GTP-dependent interaction of ADP ribosylation factor 1 with coatomer subunit beta. Proc Natl Acad Sci U S A, 94(9):4418–4423. Zhao, L., Helms, J. B., Brunner, J., and Wieland, F. T. (1999). GTP-dependent binding of ADP-ribosylation factor to coatomer in close proximity to the binding site for dilysine retrieval motifs and p23. J Biol Chem, 274(20):14198–14203. Zhao, X., Lasell, T. K. R., and Melançon, P. (2002). Localization of large ADP- ribosylation factor-guanine nucleotide exchange factors to different Golgi compart- ments: evidence for distinct functions in protein traffic. Mol Biol Cell, 13(1):119– 133. 166 Zhou, S., Zhou, H., Walian, P. J., and Jap, B. K. (2005). CD147 is a regulatory subunit of the gamma-secretase complex in Alzheimer’s disease amyloid beta- peptide production. Proc Natl Acad Sci U S A, 102(21):7499–7504. Zhou, W., Qing, H., Tong, Y., and Song, W. (2004). BACE1 gene expression and protein degradation. Ann N Y Acad Sci, 1035:49–67. Zhou, W. and Song, W. (2006). Leaky scanning and reinitiation regulate BACE1 gene expression. Mol Cell Biol, 26(9):3353–3364. 167 Appendix N values for behaviour data Growth Data (Age) 3m 6m 12m Female Control 10 12 12 S2P23 13 22 23 Male Control 9 19 20 S2P23 22 26 29 TOTAL Control 19 31 32 S2P23 35 48 52 Table A.1: N values for growth curve data. Number of female and male Con- trol (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate growth curve data. Hanging Wire (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 10 7 5 8 7 S2P23 13 10 12 13 11 Male Control 9 10 9 11 11 S2P23 22 4 5 22 24 TOTAL Control 19 17 14 19 18 S2P23 35 14 17 35 35 Table A.2: N values for hanging wire data. Number of female and male Con- trol (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate hanging wire data. 168 N values for behaviour data Rotarod (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 10 7 5 8 7 S2P23 13 10 12 13 11 Male Control 9 10 9 11 11 S2P23 22 4 5 22 24 TOTAL Control 19 17 14 19 18 S2P23 35 14 17 35 35 Table A.3: N values for rotarod data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate ro- tarod data. Open Field (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 8 7 5 8 7 S2P23 13 10 12 13 11 Male Control 11 10 9 11 11 S2P23 22 4 5 22 22 TOTAL Control 19 17 14 19 18 S2P23 35 14 17 35 33 Table A.4: N values for open field data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate open field data. 169 N values for behaviour data Light-Dark box (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 7 7 5 8 7 S2P23 12 10 12 11 9 Male Control 11 7 10 9 8 S2P23 16 3 5 15 15 TOTAL Control 18 14 15 17 15 S2P23 28 13 17 26 24 Table A.5: N values for Light-Dark box data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to gen- erate Light-Dark box data. Y-maze (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 8 7 5 8 7 S2P23 13 11 12 13 8 Male Control 11 10 10 12 8 S2P23 22 4 5 22 17 TOTAL Control 19 17 15 20 15 S2P23 35 15 17 35 25 Table A.6: N values for Y-maze data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate Y- maze data. 170 N values for behaviour data Water Maze (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 8 4 5 7 4 S2P23 12 9 12 12 9 Male Control 11 8 9 11 8 S2P23 20 3 5 18 18 TOTAL Control 19 12 14 18 12 S2P23 32 12 17 30 27 Table A.7: N values for water maze data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate water maze data. Contextual Fear (Age) Näıve Experienced 3m 6m 12m 6m 12m Female Control 8 6 5 8 6 S2P23 13 9 11 10 10 Male Control 11 10 9 11 10 S2P23 22 4 4 21 21 TOTAL Control 19 16 14 19 16 S2P23 35 13 15 31 31 Table A.8: N values for contextual fear conditioning data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate contextual fear conditioning data. 171 N values for behaviour data Cued Fear Näıve (Age) 3m 6m 12m Female Control 6 0 4 S2P23 6 1 7 Male Control 3 5 4 S2P23 5 8 8 TOTAL Control 9 5 8 S2P23 11 9 15 Table A.9: N values for cued fear conditioning data. Number of female and male Control (C57BL/6) and S2P23 (TMP21 heterozygous knockout) mice used to generate cued fear conditioning data. 172

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0103326/manifest

Comment

Related Items