University of Nebraska Medical Center DigitalCommons@UNMC

Theses & Dissertations Graduate Studies

Fall 12-15-2017

A Role for δ-catenin in Synaptic Regulation

Li Yuan University of Nebraska Medical Center

Follow this and additional works at: https://digitalcommons.unmc.edu/etd

Part of the Developmental Neuroscience Commons

Recommended Citation Yuan, Li, "A Role for δ-catenin in Synaptic Regulation" (2017). Theses & Dissertations. 228. https://digitalcommons.unmc.edu/etd/228

This Dissertation is brought to you for free and open access by the Graduate Studies at DigitalCommons@UNMC. It has been accepted for inclusion in Theses & Dissertations by an authorized administrator of DigitalCommons@UNMC. For more information, please contact [email protected]. i

A role for δ-catenin in synaptic regulation

by

Li Yuan

A DISSERTATION

Presented to the Faculty of

the University of Nebraska Graduate College

in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

Pharmacology & Experimental Neuroscience Graduate Program

Under the Supervision of Professor Jyothi Arikkath

University of Nebraska Medical Center

Omaha, Nebraska

October, 2017

Supervisory Committee:

Wallace B. Thoreson, PhD. Stephen Bonasera, M.D., Ph.D.

Howard S. Fox, M.D., Ph.D. Shilpa Buch, Ph.D. i

Acknowledgement

First, a big thank you to my mentor Dr. Jyothi Arikkath. She has always been available and provided me with invaluable comments, suggestions and immense support on every aspect of my training, especially presentation and scientific writing. I am ex- tremely grateful for her mentorship and expectation. I would like to acknowledge and thank my committee members, Dr. Wallace Thoreson, Dr. Stephen Bonasera, Dr. How- ard Fox and Dr. Shilpa Buch, for their invaluable guidance and support. I very much ap- preciate their inputs and advices during my training. Additionally, I would like to thank

Dipika Singh and James Buescher for their guidance and help during my formative years in the lab. A special thank you to Dr. Eunju Seong for the enjoyable discussions, invalua- ble suggestion and caring friendship. I would like to thank all present and past members of Dr. Arikkath’s lab for their kindly help during my training.

Finally, I would like to thank my parents, who unconditionally support and encour- age me. I also want to thank other members from my family for the unconditionally sup- port for my parents when I was not able to be there. Last, I would like to thank dindin and gingin, as well as Anand Suresh, for enormous help and support.

ii

A role for δ-catenin in synaptic regulation

Li Yuan, Ph.D.

University of Nebraska, 2017

Supervisor: Jyothi Arikkath, Ph.D.

The cadherin-catenin complex regulates cell-cell adhesion and signal transduc- tion in epithelial cells. It is becoming increasingly evident that components of the com- plex regulate various aspects of neuronal architecture and function. δ-catenin is a cyto- solic component of the cadherin-catenin complex and is predominantly expressed in the central nervous system. Loss of CTNND2, which encodes δ-catenin, is associated with intellectual disability and mutations in CTNND2 have been identified in autism, suggest- ing that δ-catenin is a critical component of the molecular machinery underlying neural circuit function. We have previously demonstrated that δ-catenin regulates multiple as- pects of synaptic and dendritic development, including dendritic arborization, spine den- sity, architecture and function, consistent with the protein being a key player in regulating neuronal circuit formation and function in the developing brain.

We have identified novel forms of δ-catenin that participate in synaptic regulation.

Our data indicate that novel forms of δ-catenin that include either the N-terminal

(DcatNT) or C-terminal (DcatCT) regions are expressed in different regions of the brain.

These forms are generated via two mechanisms, one of which involves NMDA receptor

(NMDAR), Ca2+ and calpain-dependent cleavage while the other mechanism is NMDAR- independent. The DcatCT and NT forms generated in an NMDAR-independent manner are differentially dependent on the lysosome for their generation. Functionally, loss of the domain containing the predicted sites allowing for generation of DcatCT and NT per- iii

turbs the density of a subpopulation of dendritic protrusions. Thus, our data provide evi- dence for a key role for δ-catenin in regulating a subset of dendritic protrusions, and im- plicate critical functional roles for δ-catenin in neural circuit wiring in the developing hip- pocampus.

iv

Table of Contents

Title Page ...... i

Acknowledgement ...... i

Abstract ...... ii

Table of Contents ...... iv

List of Figures ...... vii

List of Tables ...... x

List of Abbreviations ...... xi

Introduction ...... 1

The cadherin-catenin cell adhesion complex in central neurons ...... 1

The p120ctn family of proteins ...... 1

δ-catenin ...... 2

The expression and localization of δ-catenin ...... 2

δ-catenin in regulation of synaptic density, architecture, plasticity and function ...... 3

Emerging roles for CTNND2 in neurodevelopmental and neurological disorders ...... 8

CHAPTER 1 Materials and methods ...... 10

Animals ...... 11

Cell culture and transfection ...... 11

Plasmids ...... 12

Pharmacological reagents ...... 14

Lysate preparation and immunoblotting ...... 15 v

In vitro cleavage assay ...... 17

Calculate molecular weight ...... 17

δ-catenin purification ...... 18

Immunocytochemistry and microscopy ...... 18

Statistics ...... 19

Immunoblot and morphology analysis ...... 19

Fraction and approximate density for each population ...... 20

Determine P1 density threshold ...... 21

Above threshold fraction comparison ...... 21

CHAPTER 2 DcatCT and NT are two novel forms of δ-catenin ...... 22

Two novel isoforms of δ-catenin in the brain ...... 23

DcatCT and NT are generated via a calpain-dependent mechanism ...... 27

CHAPTER 3 DcatCT and NT can be generated by two independent mechanisms ....33

DcatCT and NT can be generated in an NMDAR-dependent manner by calpain cleavage

...... 34

The basal generation of DcatCT and NT is NMDAR-independent ...... 38

CHAPTER 4 The functional role of δ-catenin cleavage in dendritic protrusion development 51

Mapping the calpain cleavage sites on δ-catenin ...... 52

δ-catenin affects dendritic protrusion density ...... 55

Discussion ...... 63 vi

Different forms of δ-catenin ...... 63

Differential regulation of DcatCT and NT levels by degradation ...... 65

Regulation of dendritic protrusions by proteolytic cleavage of δ-catenin ...... 65

Future directions ...... 68

Identify regulatory mechanisms for δ-catenin cleavage ...... 68

Identify the functional consequence for calpain-mediated cleavage of δ-catenin in

excitotoxicity ...... 68

Identify the mechanism for the basal generation of DcatCT and NT ...... 69

Identify additional functional roles for DcatCT and NT ...... 69

Implications for autism ...... 70

Reference ...... 71

vii

List of Figures

Introduction

Figure 1: Schematic of δ-catenin-ABP/GRIP-AMPAR complex.

CHAPTER 2 DcatCT and NT are two novel forms of δ-catenin

Figure 2: The expression of two novel forms of δ-catenin, DcatNT and DcatCT, in the brain.

Figure 3: The expression of DcatNT and CT in different brain regions.

Figure 4: The generation of DcatCT and N-terminal fragments can be recapitulated in ly- sates from δ-catenin-transfected HEK cells.

Figure 5: The generation of DcatCT and NT is chelator-dependent in an in vitro cleavage assay.

Figure 6: The generation of DcatCT and NT is mediated by calpain in an in vitro cleav- age assay.

CHAPTER 3 DcatCT and NT can be generated by two independent mechanisms viii

Figure 7: Glutamate induces NMDAR- and calpain-dependent generation of DcatCT and

NT.

Figure 8: NMDA induces Ca2+-influx and calpain-mediated cleavage of δ-catenin.

Figure 9: DcatCT consists of two forms, DcatCTI and II.

Figure 10: Inhibition of neural activity has no effect on the basal level of DcatCT and NT.

Figure 11: Inhibition of NMDAR has no effect on the basal level of DcatCT and NT.

Figure 12: Inhibition of L-type VGCC has no effect on the basal level of DcatCT and NT.

Figure 13: Disrupt the normal function of lysosome decreases the basal level of DcatNT, but not CT.

Figure 14: Inhibition of cathepsin B decreases the basal level of DcatNT, but not CT.

Figure 15: The basal and NMDAR-dependent forms of DcatCT/NT are generated in two independent pathways.

CHAPTER 4 The functional role of δ-catenin cleavage in dendritic protrusion de- velopment

Figure 16: Mapping the calpain cleavage sites on δ-catenin by in vitro cleavage assay.

Figure 17: Cleavage of δ-catenin regulates the dendritic protrusion density. ix

Figure 18: Cleavage of δ-catenin regulates the upper limit of the density of P1 protru- sions.

x

List of Tables

CHAPTER 1 Materials and methods

Table 1: Primers for truncated δ-catenin.

xi

List of Abbreviations

AMPAR α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor

Asco Ascomycine

ARM Armadillo

Baf Bafilomycin A1

CA CA-074 Me, CA-074 methyl ester

CDK5 Cyclin Dependent Kinase 5

DcatCT δ-catenin C-terminus

DcatNT δ-catenin N-terminus

Glut Glutamate

GRIP Glutamate receptor interacting protein

Het Heterozygous

Homo Homozygous

L-NAME Nω-Nitro-L-arginine methyl ester

Lact Lactacystin

LE Long exposure

LTP Long-term potentiation

MK (+)-MK 801 xii

MDL MDL28170

Nimo Nimodipine

NLS Nuclear localization signal

NMDA N-Methyl-D-aspartic acid

NMDAR NMDA receptor

NOS Nitric oxide synthase

P1/2/3 Population 1/2/3

PSD95 Postsynaptic density protein 95

SBDP Spectrin breakdown product

SE Short exposure shRNA Short hairpin RNA

Thap Thapsigargin

TTX Tetrodotoxin

VGCC Voltage-dependent calcium channel

WT Wild-type

1

Introduction*

The cadherin-catenin cell adhesion complex in central neurons

Cadherins are transmembrane proteins that mediate cell-cell adhesion in a variety of cell types. More recently, it has become clear that they have extensive functional roles in neurons and are associated with multiple neuronal functional roles including adhesion and synaptic plasticity. Their functional roles are of particular significance in multiple as- pects of neuronal morphology, synaptic structure, synaptic efficacy and molecular pro- cesses related to learning and memory [1–3]. Cadherins mediate adhesion across synap- ses via homophilic trans-interactions across the synaptic cleft [2]. The cytosolic intracel- lular terminal of cadherins is associated with a variety of proteins, collectively called caten- ins. Of these, β-catenin is perhaps the most well studied and several studies indicate crit- ical roles for this protein in a variety of neuronal phenotypes including pre- and postsyn- aptic development, memory and addiction [4–7]. A group of catenins associated with the classical cadherins also include members of the p120ctn family of proteins. The p120ctn family of proteins form a family distinct from the β-catenin and α-catenin families and bind to cadherin at a site distinct from the β-catenin binding site [8]. Over the last several years, it has become obvious that the functional roles of the p120ctn family of proteins are quite diverse and include both cadherin-dependent and -independent functional roles in various aspects of neuronal morphogenesis.

The p120ctn family of proteins

The p120ctn family includes 4 proteins encoded by four independent genes. These include p120ctn (encoded by CTNND1), δ-catenin (also called NPRAP or neurojungin [9], encoded by CTNND2), p0071 (plakophilin-4, encoded by PKP4) and ARVCF (encoded by

* Introduction is modified from Seminars in Cell & Developmental Biology, Vol 69, L Yuan, J Arikkath, Functional roles of p120ctn family of proteins in central neurons, 70- 82, Copyright (2017), with permission from Elsevier. 2

ARVCF). Evolutionarily, they are generated from an ancient ‘δ-catenin-like’ gene and di- vided into two major classes, one of which comprises p0071 and δ-catenin, while the other includes p120ctn and ARVCF [10]. p120ctn, ARVCF and p0071 are expressed in central neurons and widely distributed in non-neuronal tissues [11–14]. δ-catenin is unique in that it has a more restricted expression to central neurons [9,15].

δ-catenin

Similar to other members of the p120ctn family, δ-catenin includes an N-terminal domain, a series of 9 armadillo (ARM) repeats and C-terminal domain. The N-terminal domain of δ-catenin bears a coil-coil domain, while the end of the C-terminal domain con- tains a PDZ-binding motif.

So far, two splice variants for δ-catenin are known in mouse [9]. One of these includes a 25 aa insertion at the beginning of the first helix of ARM 7. Both isoforms are expressed in various brain regions, and in both neuron and glia [16]. While no functional differences between the two isoforms are known, the 25 aa cassette is localized spatially proximal to the insert region on the arch of the ARM domain [17], suggesting it may regu- late the structure and function of the insert region.

The expression and localization of δ-catenin

δ-catenin is widely expressed in different regions of the brain [9,16], both during early development and in the mature brain [18–21]. δ-catenin is also expressed in neural progenitor cells, although their functional roles in these cells remains incompletely char- acterized [22,23]. In addition, recent studies combining electrophysiological profiling with single-cell RNAseq analysis demonstrate the expression of δ-catenin is significantly lower 3

in burst-spiking pyramidal cells compared to regular-spiking pyramidal cells in the subicu- lum [20], suggesting neuron type specific expression patterns and functional roles. Further, analysis of the primary visual cortical region in mice indicated that the expression of δ- catenin is enriched in inhibitory somatosensory neurons [24,25].

In primary neurons, δ-catenin is localized to growth cones [26], dendritic segments and postsynaptic terminals in central neurons [27,28]. It is not entirely clear if δ-catenin is also localized at presynaptic terminals.

δ-catenin in regulation of synaptic density, architecture, plasticity and function

Given the key links between the cadherin-catenin cell adhesion complex and the actin cytoskeleton, it is not surprising that δ-catenin regulates dendritic development, syn- aptic density and architecture in neurons.

δ-catenin has been shown as a key regulator for dendritic development. Loss of δ- catenin during dendritic development reduces dendritic arbors [27]; in contrast, overex- pression of δ-catenin increases the branch number [29,30].

Compared to the dendritic phenotype, the effects of loss of δ-catenin on excitatory spine density remain somewhat confusing. Studies indicate that overexpression of δ- catenin in mature primary hippocampal neurons increases the density of synapses [31].

These abilities of δ-catenin to enhance spine density could be inhibited by the use of a dominant negative Rac construct [31], indicating the ability of δ-catenin to influence spine and synapse density is regulated indirectly through small G proteins, which in turn regulate the actin cytoskeleton. Conversely, the authors demonstrated that loss of δ-catenin using siRNA-mediated knockdown leads to a decrease in the density of spines and synapses

[31]. Importantly, these reduced densities could be rescued by the introduction of an 4

siRNA-resistant construct of δ-catenin, confirming specificity [31]. These studies are also supported by other published studies which indicate that overexpression of δ-catenin pro- motes an increase in the density of spines and synapses in primary hippocampal neurons

[32]. However, these studies are in direct contrast to another study in primary hippocam- pal neurons in which loss of δ-catenin leads to an increase in the density and function of excitatory synapses both with specific shRNA knockdown and in primary neurons from δ- catenin N-term mice, a mouse model that includes an N-terminal truncated form of δ- catenin [33,34]. Further, these abilities of δ-catenin were dependent on its ability to coop- erate with PDZ domain containing regulators via its PDZ-binding motif, but like p120ctn, it is also independent of cadherin binding [33,34]. It is unclear how these contrasting results can be reconciled. One likely possibility is that since δ-catenin is a component of the cad- herin-catenin cell adhesion complex, the ability of δ-catenin to influence spine and syn- apse density is related to the density of neurons in the culture. Other possibilities include splice variant specific effects or the levels of δ-catenin or neuron subtype specific effects.

While it is clear that δ-catenin is a key regulator of excitatory synapse density and function, further studies are necessary to resolve the exact nature of its abilities.

Loss of δ-catenin also impairs both spine headwidth and length in hippocampal neurons [28]. The ability of δ-catenin to modulate spine architecture is mediated by its ability to link to cadherins and PDZ domain effectors. Knockdown of δ-catenin also impairs structural plasticity in hippocampal neurons. More specifically, knockdown of δ-catenin impairs the ability of primary hippocampal neurons to alter spine headwidth in response to treatment with tetrodotoxin [33]. However, it is still unclear whether overexpression of

δ-catenin affects spine architecture.

The PDZ-binding motif of δ-catenin confers the protein the ability to link to and functionally influence a variety of synaptic scaffolding proteins. AMPA receptors (AMPAR) 5

interact with δ-catenin via scaffolding protein ABP/GRIP to localize to plasma membrane

(Fig. 1) [35]. Overexpression of δ-catenin increases the level of GRIP, surface GluR2 and ratio of AMPA/NMDA [36]. In contrast, knockdown of δ-catenin leads to decrease of GRIP and surface GluR2, with no change of the AMPA/NMDA ratio, which may due to the dif- ference of experimental systems [36]. Unlike the effects on the AMPAR, there is no alter- ation in NMDA receptor response with loss or overexpression of δ-catenin [36]. The ability of δ-catenin to regulate synaptic density is influenced by its ability to be phosphorylated by cyclin dependent kinase 5 (CDK5). δ-catenin can be directly phosphorylated at resi- dues S300 and S357 by CDK5 [37]. This phosphorylation promotes its membrane asso- ciation [37]. A mutant form of δ-catenin expressing S to A mutations at residues 300 and

357, representing constitutively activated forms, enhances the density of dendritic protru- sions [37]. Loss of CDK5-mediated phosphorylation of δ-catenin enhances the associa- tion between δ-catenin and N-cadherin, GRIP (Glutamate receptor interacting protein) and

PSD95 (postsynaptic density protein 95), resulting in an increase in the surface AMPAR and ratio of AMPA/NMDA compared to the wild-type δ-catenin [37].

6

Figure 1: Schematic of δ-catenin-ABP/GRIP-AMPAR complex. 7

δ-catenin also binds to the postsynaptic scaffolding proteins PSD95 and SSCAM

[35,38]. Interestingly, PSD95 and SSCAM also regulate localization of AMPAR to the plasma membrane [39,40]. Since δ-catenin has a single PDZ-binding motif, it is unclear what the stoichiometry of PDZ proteins associated with the cadherin-δ-catenin complex is and how these proteins cooperate to regulate the trafficking and stabilization of AMPAR.

Knockdown of δ-catenin in primary neurons impairs the ability of neurons to pro- mote an increase in the density and length of protrusions in response to stimulation by

DHPG, which promotes activation of the Group 1 mGluR receptors [31]. DHPG stimulation promotes phosphorylation of δ-catenin at multiple residues, dissociation from PSD95 and cadherin and increased association with cortactin, suggesting that the abilities of δ-catenin to regulate spine density in a DHPG-dependent manner may be at least in part due to its ability to alter links with regulators of the actin cytoskeleton, including cortactin, and cad- herin in a phosphorylation-dependent manner [31].

Members of the p120ctn family of proteins are also being recognized as key play- ers in the molecular machinery underlying learning. Chemical long-term potentiation (LTP) promotes the palmitoylation of δ-catenin by DHHC5 at residue C960-961 located within the 3rd helix of ARM 8 and this promotes interaction of δ-catenin with cadherin at synapses, activity-induced stabilization of cadherin at synapses and spine enlargement and insertion of GluA1 and GluA2 receptors [34]. Further context-dependent fear conditioning in mice promotes δ-catenin palmitoylation and increased δ-catenin-cadherin associations at hip- pocampal synapses [34]. Thus, δ-catenin is a critical component of the machinery that allows coordination of synaptic alterations involved in memory formation.

δ-catenin N-term mice have behavioral deficits, including deficient spatial learning in water maze and associate learning in fear conditioning [41]. Consistently, these δ- catenin mutant mice have abnormal LTP [41]. 8

Emerging roles for CTNND2 in neurodevelopmental and neurological disorders

Mutations and variations in CTNND2, encoding δ-catenin, have been associated with several neurological and neurodevelopmental disorders. The cri du chat syndrome is associated with a cat-like cry, dysmorphic facial features, microcephaly, and intellectual disability [42]. The syndrome is associated with variable deletions of chromosome 5p re- sulting in loss of one or more genes from the locus. CTNND2 is localized at 5p15.2 and the loss of CTNND2 in these individuals correlates with the intellectual disability phenotype

[43]. Additional studies indicate a strong correlation with hemizygosity of CTNND2 and intellectual disability. An out of frame deletion spanning exons 4-7 of CTNND2 in a human has been associated with altered facial features and borderline intellectual disability [44].

In addition, two patients with intronic or exonic deletions of CTNND2 have been described to have intellectual disability [45]. Consistent with the critical role of the gene in intellectual disability, an individual with the cri du chat syndrome with an additional complex genetic rearrangement resulting in a partial deletion and partial duplication of CTNND2 has a mild cognitive phenotype [46,47].

More recent studies suggest a critical role for CTNND2 in the pathology of autism.

Significant excess of missense and copy number variations have been identified in indi- viduals with autism [32]. Interestingly, these variations span the entire protein and are not restricted to a single domain, suggesting that likely several key functional features of δ- catenin are required for its ability to participate in neural circuit wiring crucial to higher order functions.

Copy number variations in CTNND2 have more recently been identified in cerebral palsy [48], although the correlation between these variants and the pathology of the dis- order remains incompletely characterized. Copy number variations associated with gene 9

duplication and disruption of the terminal portion of CTNND2 have been identified in schiz- ophrenia [49]. Recent studies have also implicated CTNND2 in anxiety disorders [50], although these links need further exploration and validation.

Although δ-catenin has been found in the retina [51], it is still lack of evidence that

δ-catenin also expresses in the lens. Interestingly, genetic variations in CTNND2 have also been implicated in the development of cortical lens opacities associated with cataract and cognitive change that precede the development of Alzheimer’s disease [52]. This var- iant leading to a single amino acid substitution, G810R, in δ-catenin, promotes an increase in the secretion of Aβ1-42, suggesting a pathological mechanism for a link between this variation in CTNND2 and the pathology of cataracts and Alzheimer’s disease. It would be interesting to examine variations in CTNND2 in larger cohorts of Alzheimer’s disease pa- tients to define the extent of contribution of variations in CTNND2 to the pathology of Alz- heimer’s disease. 10

CHAPTER 1

Materials and methods

11

Animals

All experiments were approved by the University of Nebraska Medical Center In- stitutional Animal Care and Use Committee. Animals were housed in 12/12 dark-light cy- cle with free access to water and food. δ-catenin N-term mouse was a generous gift from

Dr. Xin Liu (University of California, Los Angeles) [41]. Brain tissue from different regions and in vitro assay were collected from C57BL/6 mice. For all experiments excepting the in vitro assay, only male mice were used.

Cell culture and transfection

All cells cultures were incubated at 37 °C with 5% CO2. HEK 293 cell was a gen- erous gift from Dr. Shilpa Buch (University of Nebraska Medical Center). HEK cells were maintained in DMEM (Thermo Fisher, 11965-092/SH30022.01) supplement with 10% heat-inactivated FBS (ATLABS) and pen-strep (Thermo Fisher, SV30010). Primary hip- pocampal and cortical cultures were prepared from E18–19 Sprague-dawley rat (Charles

River Laboratories) as described previously [53] with plating density 18,750 cells/cm2 and 50,000 cells/cm2 respectively. After 4-6 hr after plating, cells were switched to maintenance medium, containing Neurobasal Medium (Thermo Fisher, 21103049), B-27

Supplement (Thermo Fisher, 17504044) and pen-strep (Thermo Fisher, 15070-063). For in vitro assay (Fig. 5C), hippocampal culture was treated with mitotic inhibitor 5 µM Ara-

C (Sigma, C1768) at DIV 2-4 to minimize the number of glia cells to specifically test the cleavage on neurons. Otherwise, gilia were not excluded to mimic the physiological con- dition in vivo. Half of the media were changed every 3-5 and 2-4 days for hippocampal and cortical culture respectively. 12

Lipofectamine 2000 (Thermo Fisher) was used for transfection according to man- ufacturer’s protocol. For neurons, plasmids were prepared by EndoFree Plasmid Kit (QI-

AGEN) and Plasmid Plus Kit (QIAGEN) to minimize endotoxin level. For HEK cells, plas- mids prepared by HiSpeed Plasmid Kits (QIAGEN) were also included. All the media were changed 1 day after transfection for HEK cells and half of the media were changed

4-8 hr after transfection for neurons.

Plasmids

The cDNA construct with the long isoform of mouse δ-catenin (Uniprot identifier:

O35927-1) was a generous gift from Dr. Werner Franke (German Cancer Research

Center, Germany) [9]. The EGFP-δ-catenin constructs carries the short isoform of mouse δ-catenin (Uniprot identifier: O35927-2) and was a generous gift from Dr.

Shernaz Bamji (University of British Columbia) [34]. The δ-cat shRNA and the shRNA- resistant δ-catenin construct basing on the mouse long isoform were described previ- ously [33]. The pSuper GFP vector was used as a control for the δ-cat shRNA. TAP-δ- catenin was generated by the Mouse Genome Engineering core facility of University of

Nebraska Medical Center. It carries the long isoform of mouse δ-catenin with its N-termi- nus fused with a TAP tag: Myc-TEV-linker-His-tag, cloned into the pcDNA3.1 vector.

All the truncated δ-catenin were generated by GeneArt site-directed mutagenesis system (Thermo Fisher, A13282) with AccuPrime Pfx DNA Polymerase (Thermo Fisher,

12344040) and corresponding primers (Table 1) according to manufacturer’s protocol using shRNA-resistant δ-catenin cDNA construct as template. It is worth mentioning that 13

the DNA sequence encoding ~200-300 aa in δ-catenin has very high GC-content, there- fore ≥ 4 % DMSO was included during PCR. All the truncation sites were confirmed by

Sanger sequencing.

Table 1: Primers for truncated δ-catenin.

Construct name Primers

5’-ATCGTCGTGTCCTCGCCTCTGCCGCCAGCA-3’ DI321440 5’- TGCTGGCGGCAGAGGCGAGGACACGACGAT-3’

5’-ATCGTCGTGTCCTCGTATGAAGACCGTGTC-3’ DI321420 5’-GACACGGTCTTCATACGAGGACACGACGAT-3’

5’- ATCGTCGTGTCCTCGGGCACCTACGCCACC-3’ DI321350 5’-GGTGGCGTAGGTGCCCGAGGACACGACGAT-3’

5’-CTGAGCTCCACCATCGGATCCCGAGCCTCG-3’ DI351390 5’-CGAGGCTCGGGATCCGATGGTGGAGCTCAG-3’

5’-GGCAGCCTGGCAGCTCAGAGCCAGGGGGAT-3’ DI391435 5’-ATCCCCCTGGCTCTGAGCTGCCAGGCTGCC-3’

5’-ACGAGCACAGCCCCGGGCAGCCAACACGGG-3’ DI457468 5’-CCCGTGTTGGCTGCCCGGGGCTGTGCTCGT-3’

5’-TCCTCCCCTGGTGTCGTCCCCTTGCAGCGC-3’ DI462463 5’- GCGCTGCAAGGGGACGACACCAGGGGAGGA-3’

5’- CAGAGCCAGGGGGAT GCCACCTTCCAGAGG-3’ DI441480 5’- CCTCTGGAAGGTGGCATCCCCCTGGCTCTG-3’

5’- GCACATACCGGCACCCCGTCCTCCCCTGGT-3’ DI450455 5’- ACCAGGGGAGGACGGGGTGCCGGTATGTGC-3’

DI421430 5’-CACATAGACCCCATCATGAGGAGTCTCAGC-3’ 14

5’-GCTGAGACTCCTCATGATGGGGTCTATGTG-3’

5’-TATCAGAAGCCCCCTCGCACGAGCACAGCC-3’ DI431450 5’-GGCTGTGCTCGTGCGAGGGGGCTTCTGATA-3’

5’-CTGAGCTCCACCATCAAGCATTCGCAGGAG-3’ DI351370 5’- CTCCTGCGAATGCTTGATGGTGGAGCTCAG-3’

5’-TCTGAGCAGTACAGCGGATCCCGAGCCTCG-3’ DI371390 5’-CGAGGCTCGGGATCCGCTGTACTGCTCAGA-3’

5’-GGCAGCCTGGCAGCTTATGAAGACCGTGTC-3’ DI391420 5’- GACACGGTCTTCATAAGCTGCCAGGCTGCC-3’

Pharmacological reagents

Pharmacological reagents for neuron treatments were listed below: D-AP5

(Tocris, 0106), (+)-MK 801 (Tocris, 0924), CNQX (Tocris, 1045), ALLN (Sigma, A6185),

MDL28170 (Sigma, M6690), MG132 (Sigma, M7449), tetrodotoxin (Tocris, 1069), lacta- cystin (Calbiochem, 426100), bafilomycin A1 (Tocris, 1334) and L-NAME (Sigma,

N5751). CA-074 Me was a generous gift from Dr. Huangui Xiong (University of Nebraska

Medical Center). Thapsigargin was a generous gift from Dr. Keshore Bidasee (University of Nebraska Medical Center). Cortical neurons were treated between DIV10-28. Before treatment, the volume of the medium was measured to ensure no variability due to evap- oration. Control samples were treated with same volume of H2O or DMSO accordingly.

Ca2+-free medium was prepared from Ca2+- and Mg2+-free HBSS (Thermo Fisher,

14170112) supplement with 0.814 mM MgCl2, 10 mM HEPES and extra 24.4 mM D-glu-

2+ cose, then equilibrant with 5% CO2 at 37 °C. Neurons were incubated in Ca -free me- dium for 10-15 min before treatment. CaCl2 (1.8 mM final concentration) were mixed with 15

NMDA/vehicle immediately before adding to the cells. The final concentrations of Mg2+ and Ca2+ were identical to the maintenance medium. For KCl treatment, extra 55 mM

KCl was added to the maintenance medium for a final concentration of 60.33 mM.

Lysate preparation and immunoblotting

For brain tissue collection, animal was anaesthetized by isoflurane and the brain was dissected in cold 1XPBS on ice. Brain tissues were quick frozen on dry ice and sto- ried at -80 °C. Since δ-catenin in the brain tissue is susceptible to cleavage during the freeze-thaw cycle (data not shown), cold lysis buffer was added before thawing the tis- sue on the ice. 100-1000 µL cold lysis buffer was used depending on the size of the tis- sue. Lysis buffer includes 1:100 proteinase inhibitor (Sigma, P8340), 1:100 phosphatase inhibitor (Sigma, P0044) in the RIPA buffer (50 mM Tris pH7.4, 107 mM NaCl, 1% Triton

X-100, 0.1% SDS). Besides the in vitro assay, all lysis buffer include 5 mM EDTA and an extra 5 mM EGTA for neuron or tissue. Brain tissues were homogenized for 2 s on ice by sonicator (22-25% amplitude) for 3-8 times depending on the size of the tissue, with min- imally 60 s interval between rounds.

Before neuron lysis, cells were washed with cold 1XPBS. HEK cells were lysed without wash since they tend to detach. 70-100 µL cold lysis buffer was added per well in a 6 well plate. After being scraped off by a cold cell scraper, cells were first pipetted up and down for about 30 times, then sheered with a cold 26/27G needle for about 30 times. All the procedures were done on ice.

All the lysates were centrifuged 15,000 rpm for 15 min at 4 °C, then quick frozen on dry ice and storied at -80 °C. Supernatants were mixed with laemmli buffer and loaded on the gel (30-50 µg/well). For DcatCTI/II detection in neural lysate, 50 µg/well 16

was optimal. Both hand-casted and precasted gels (BioRad, 456-1084 and 456-1024) were used for SDS-PAGE. To separate DcatCTI/II, 7.5-8% gels were used. For separat- ing DcatCTI and II from neural lysate, we had to run the gels longer making β-tubulin un- usable as a house keeping marker. In such case, we used Na+/K+-ATPase as a loading control.

Dilution of antibodies for immunoblot were listed below: 1:900 TL (BD Biosci- ences, 611537), 1:50 J19 (Progen, 651152), 1:100-1:500 β-tubulin (DHSB, E7), 1:200 actin (DHSB, JLA20), 1:400 Na+/K+-ATPase (DHSB, a6F), 1: 5000 N-cadherin (BD Bio- sciences, 610920); 1: 1000 PSD95 (Millipore, MAB1598); 1:2000 CaMKIIɒ (Millipore,

6G905-532); 1:1000 GAPDH (Thermo, PA1-988); 1: 4000 ɒII-spectrin (clone AA6, Enzo,

BML-FG6090), 1:16000 HRP-conjugated anti-rabbit IgG (Jackson ImmunoResearch,

711-035-152) and 1:16000 HRP-conjugated anti-mouse IgG (Jackson Immu- noResearch, 711-035-151). 587 is a custom-made polyclonal rabbit antibody from

Thermo Fisher. It targets SELNYETSHYPASPDSWV of δ-catenin and was purified with

GST fusion protein. Myc antibody (clone 9E10) was produced by the core facility of Uni- versity of Nebraska Medical Center. ARVCF antibody was a generous gift from Dr. Keith

Johnson (University of Nebraska Medical Center).

Blots were detected by SuperSignal West Dura chemiluminescent substrate

(Thermo Fisher, 34076) and imaged by FluorChem HD2 (Cell Biosciences). All the blots were quantified by AlphaView (ProteinSimple) or ImageJ. Sum signal of each band/iso- form was subtracted by the sum background signal of the similar sized selection area.

All the protein quantifications were normalized to a housekeeping protein shown in the blot.

17

In vitro cleavage assay

As described earlier, the in vitro assay is identical to the lysate preparation ex- cept exclusion of chelator in the lysis buffer. For inhibitor testing (Fig. 6C), inhibitor or vesicle was mixed with lysis buffer immediately before adding to the cells. However, since CA-074 Me needs to be hydrolyzed by intracellular enzyme for irreversible inhibi- tion of cathepsin B, cells were pretreated with CA-074 Me instead of directly mixing the inhibitor with lysis buffer.

Calculate molecular weight

The molecular weight measurement was done in AlphaView and refers to protein ladders from BioRad (1610394/1610393). Only lanes right next to the protein ladder were used for measuring molecular weight to ensure accuracy. Molecular weight for

DcatNT and DcatCT were measured on blots probing with J19 and 587 respectively.

Due to the band of DcatNTII being thick, the molecular weight of DcatNTII was repre- sented by the higher boarder of band.

The molecular weight of DcatCTII from both NMDA- and glutamate-treated sam- ples were pooled to represent DcatCTII in neuron. Moreover, since the sample beside the ladder was always control, often having only DcatI (DcatICtrl), we thus calculated

DcatII as below:

DcatII = DcatCTICtrl − (DcatCTINMDA/Glut − DcatCTIINMDA/Glut)

DcatCTINMDA/Glut and DcatCTIINMDA/Glut are the DcatCTI and CTII from the NMDA- or gluta- mate-treated samples respectively. 18

δ-catenin purification

TAP- δ-catenin in HEK cells were purified by PureProteom magnetic protein A/G beads (Millipore, LSKMAGA02/LSKMAGG02) according to the manufacturer’s protocol with some modifications. The beads were cross-linked to Myc antibody then blocked with

5% BSA. The fresh lysate was incubated with the beads on a rotator at 4 °C overnight.

After washing with lysis buffer (a different protease inhibitor was used: Thermo Fisher

88266), the TAP- δ-catenin were eluted with glycine (pH 2.5). Elution was immediately neutralized by Tris buffer (pH 8.0) and then stored at -80 °C. Independent batch of puri- fied δ-catenin was used for each independent experiment.

For in vitro cleavage using purified δ-catenin, fresh lysate was prepared from non-transfected HEK cells and mixed with purified δ-catenin at 4 °C for 1 hr. The reac- tion was stopped by adding laemmli buffer to the samples. For heat inactivation, lysate was incubated at 80 °C for 10 min before the reaction.

Immunocytochemistry and microscopy

Briefly, coverslips were fixed in warm (37 °C) fixation buffer with 4% PFA and 4% sucrose in 1XPBS for 20 min. 10 min of 0.1% Triton X-100 was used for permeabiliza- tion. Images were taken by LSM 700 confocal microscope (Zeiss) with 2048 × 2048 dpi,

8 bit, Zoom 2, speed 6, 9 stacks with 0.47 µm interval. The expression of full-length or calpain-resistant δ-catenin in knockdown neurons were confirmed by immunostaining with δ-catenin. Dendritic protrusions from the thickest dendrite from each neuron was measured. All the protrusions located more than 10 μm from the soma were manually 19

analyzed in Z-stack in ImageJ. All the analyses were done blind to the experimental groups.

Statistics

Statistics tests were performed in Prism 7 (GraphPad) and MATLAB. D'Agostino and Pearson normality test were performed for all the data and parametric/nonparamet- ric test was performed accordingly. Data with too few number for the normality test was treated as normal distributed. n represents the number of independent experiments. CI was calculated as 95% confidence interval. Two-tailed p-value was calculated accord- ingly and significance level was 0.05. 0.05 ≤ p < 0.01: indicated by number in each panel; p ≤ 0.01: **; p ≤ 0.001: ***; p ≤ 0.0001: ****.

Immunoblot and morphology analysis

Quantification for different brain region (Fig. 2B) were first normalized to β- tubulin, then normalized to the average DcatCT or NT level of five brain regions in the same mouse to reduce the data to independent observations [54].

For analysis of pharmacological treatments, the treatment groups were paired to the control sample of the experiment. Moreover, the effect of treatment was measured by ratio instead of difference, compared to the control; thus, the results were log trans- formed before statistical analysis.

For analysis of MK801 and CNQX treatment (Fig. 11B-D), due to the null hypoth- esis is ‘the treatment affects the level of DcatCT/NT/SBDP, thus, correction for false pos- itive is not applicable in this case. For repeated measure two-way ANOVA, data were 20

presented by mean ± CI to accurately represent the relationship of data across groups in different tests.

For morphology analysis, all the data points were normalized to the geometric mean/median of the vector group of the experiment so that unpaired tested could be per- formed. Since the data were nonparametric distributed and rank-based test would be performed, thus they were not transformed to logarithmic scale.

Fraction and approximate density for each population

Because not all the protrusions were measured for width and length due to some of the protrusions overlaping with each other, fraction for each population in each den- drite was calculated from the total number of Pn protrusion measured:

Number of Pn Pn fraction = Number of (P1 + P2 + P3) n = 1, 2, 3.

Approximate density was used to represent the density for the population.

Approximate Pn density = Pn fraction × Protrusion density

Due to a few dendrites in the vector group having no protrusion from certain pop- ulation, Pn fraction or approximate density was normalized to the median of the vector group in each experiment.

21

Determine P1 density threshold

Probability frequency was estimated by Kernel smoothing function. To better vis- ualize the distribution of the data, estimation was calculated by bootstraping with re- placement for 500 times. The threshold was drawn according to the distribution differ- ence basing on the criteria mentioned in the text.

Above threshold fraction comparison

Above threshold fraction for each group was calculated from bootstrap with re- placement for 5,000 times. The difference between each group and the knockdown sam- ple, (each group - δ-cat shRNA), in each resampling was calculated and the p-value was equal to the number of (difference ≥ 0) divided by 5,000. The stack of five p-value were then corrected by Bonferroni-Dunn correction.

22

CHAPTER 2

DcatCT and NT are two novel forms of δ-catenin

23

Two novel isoforms of δ-catenin in the brain

We generated a rabbit polyclonal antibody to δ-catenin (587) using an epitope targeting the C-terminal (Fig. 2A). We validated the specificity of this antibody and two other commercial antibodies (TL and J19) using striatal lysates from wild-type, homozy- gous and heterozygous δ-catenin N-term mice [41]. As previously described [41], δ- catenin N-term mice include only the N-terminal region of δ-catenin up to 461 aa (Fig.

2A). Both the TL and J19 antibodies detected the truncated form of δ-catenin predicted in the δ-catenin N-term mice (Fig. 2B), while the C-terminal 587 antibody did not detect either full-length or N-term protein in the δ-catenin N-term mice consistent with the lack of the C-terminal region in these mice. These data confirm the specificity of the antibod- ies.

In addition to the full-length δ-catenin forms, we observed two bands with molec- ular weights of 47.67 ± 0.5774 and 40.67± 0.5774 kDa (n = 3, mean ± SD) detected by the TL and J19 antibodies, but not the 587 antibody (Fig. 2B), indicating that they include the N-terminal region of the protein, but not the C-terminal. These proteins are collec- tively referred to as DcatNT for δ-catenin N-terminus. Similarly, we observed a ~90 kDa band that could be detected by the C-terminal 587 antibody, but not the N-terminal TL and J19 antibodies (Fig. 2B). This band is referred to as DcatCT for δ-catenin C-termi- nus.

24

A

B

*

Figure 2: The expression of two novel forms of δ-catenin, DcatNT and DcatCT, in the brain.

(A) Schematic of protein structures of δ-catenin and the truncated δ-catenin in the δ- catenin N-term mouse model. Epitopes of δ-catenin antibodies: TL, J19 and 587, are indicated by blacklines. (B) Immunoblot for δ-catenin in striatum from wild-type (WT), heterozygous (Het) and homozygous (Homo) mice. Lysates were probed with δ- catenin antibodies, 587 (left), TL (middle) and J19 (right) as indicated. The top two sections in each panel are from the same blot. The first section is either a long expo- sure (LE) or dark adjusted (Dark) version for better visualize the protein with low ex- pression. The second section is either a short exposure (SE) or light adjusted (Light) version for better visualized the protein with high expression. * Non-specific band. For each genotype, n ≥ 3 animals were tested for each antibody.

25

We further characterized the expression pattern of DcatNT and DcatCT in differ- ent brain regions. Expression of DcatNT was relatively similar in the olfactory bulb, hip- pocampus, cortex, cerebellum and striatum as detected by both TL and J19 antibodies

(Fig. 3A-B). In contrast, expression of DcatCT varied in different regions with lower lev- els in the olfactory bulb and cerebellum compared to the hippocampus, cortex and stria- tum (Fig. 3A-B). These data indicate that multiple forms of δ-catenin that include either the N- or C-terminus are differentially expressed in different regions of the brain.

26

A

B

Figure 3: The expression of DcatNT and CT in different brain regions.

(A) Immunoblot for δ-catenin and (B) quantification of DcatCT and NT in different brain regions in 1-2 months old mice (n = 3, mean ± SD).

27

DcatCT and NT are generated via a calpain-dependent mechanism

We sought to examine the mechanisms by which DcatCT/NT are generated. We examined if the two forms of δ-catenin could be generated in transiently transfected HEK cells by transfecting full-length δ-catenin cDNA and immunoblot analysis (Fig.

4A). These lysates were prepared in EDTA-free buffers as indicated in the methods.

Both the DcatCT and NT fragments could be detected in these lysates. Similar results were observed with a shorter isoform of δ-catenin (Fig. 4B). Thus, the machinery for generating DcatCT and NT exists in HEK cells. We took advantage of this system to ex- amine the requirements for generation of DcatCT and NT. Since several proteases can cleave their substrates in a divalent ion-dependent manner that can be blocked by metal chelators, we examined if the generation of DcatCT in HEKs was dependent on chelator.

HEK cells were transfected with cDNA for δ-catenin and lysed in buffers containing dif- ferent levels of EDTA and the level of DcatCT was examined by immunoblot analysis

(Fig. 5A). With increasing concentration of EDTA, the levels of DcatCT were reduced, in- dicating the sensitivity of DcatCT generation on a divalent ion. To establish the identity of the divalent ion, we performed similar experiments in lysates substituted with Ca2+, Mg2+ or Zn2+. As shown in Fig. 5A, the inclusion of Ca2+ was sufficient to restore generation of

DcatCT. These data indicate that the generation of DcatCT is dependent on Ca2+.

28

A

B

Figure 4: The generation of DcatCT and N-terminal fragments can be recapitu- lated in lysates from δ-catenin-transfected HEK cells.

Immunoblot for DcatCT and N-terminal fragments in an in vitro cleavage assay on

HEK cells transfected with (A) δ-catenin cDNA plus vector or δ-cat shRNA (n = 4), (B)

EGFP-tagged δ-catenin short form (n = 3). 29

To confirm that these data are relevant to neurons, we examined the chelator-de- pendence for generation of DcatCT and NT in neurons with and without chelator in the lysis buffer. Hippocampal tissue or primary rat cortical were lysis in a buffer with or with- out chelator and immunoblotted to detect DcatCT or NT. Similar to the results observed in HEK cells, the levels of DcatCT and NT were dependent on the chelator (Fig. 5B-C).

Since several proteases are Ca2+-dependent, we examined if DcatCT and NT could be generated by cleavage by a Ca2+-dependent protease. We purified TAP-tagged

δ-catenin from transiently transfected HEK cells and examined the ability of this protein to be cleaved by lysate from un-transfected HEK cells. Purified δ-catenin was incubated with increasing amounts of lysates from un-transfected HEK cells and the generation of

DcatCT examined by immunoblots (Fig. 6A). The generation of DcatCT showed a dose- dependence on the amount of lysate from un-transfected HEK cells. The generation of

DcatCT were abolished when the purified δ-catenin was incubated with heat-inactivated

HEK lysates (Fig. 6B). These results suggest that δ-catenin is cleaved to generate

DcatCT by a Ca2+-dependent protease that is ubiquitously expressed.

One candidate protease for Ca2+-dependent cleavage is calpain, a Ca2+-depend- ent cysteine protease with known roles in proteolysis in stroke, Alzheimer disease and

Parkinson disease [55]. We took advantage of known inhibitors of calpain to examine if

DcatCT and NT were generated by calpain-mediated cleavage of δ-catenin. Primary neurons were lysed in lysis buffer that included chelators and either ALLN, MDL28170 and MG132 and immunoblotted to detect DcatCT. As a positive control we also included lysates without chelators (Fig. 6C). In lysates with either of the inhibitors, the levels of

DcatCT were significantly decreased in comparison to lysates without chelators, sug- gesting that DcatCT may be generated in a calpain-dependent manner in neurons. Since 30

ALLN and MG132 are also proteasome inhibitors [56], we confirmed that these effects were not mediated by the proteasome by using a proteasome inhibitor lactacystin in the same assay (Fig. 6C). Since all three inhibitors also inhibit lysosomal protease cathepsin

B [56], we also examined if a cathepsin B inhibitor would have similar effects (Fig. 6D).

The cathepsin B inhibitor, CA-074 Me, did not alter the levels of DcatCT and NT. These results indicate that DcatCT and NT are generated in a calpain-dependent manner in

HEKs and neurons.

31

B A

C

Figure 5: The generation of DcatCT and NT is chelator-dependent in an in vitro cleavage assay.

(A) Immunoblot for DcatCT in lysates of δ-catenin-transfected HEK cells in indicated buffers with indicated dose of chelator and chelator supplemented with divalent ions

(n = 3-4). Immunoblot for DcatCT and NT in lysates of (B) P1-3 hippocampal tissue (n

= 4) and (C) primary hippocampal culture (DIV 20-21, n = 3) with/without chelators. 32

A B

C D

Figure 6: The generation of DcatCT and NT is mediated by calpain in an in vitro cleavage assay.

Immunoblot for (A) DcatCT in purified full length δ-catenin (TAP-δ-catenin) mixed with indicated dose of HEK cell lysate (n = 3); (B) DcatCT from purified full length δ- catenin mixed with control or heat-inactivated non-transfected HEK cell lysate (n = 3);

(C) DcatCT in an in vitro cleavage assay on cortical neurons supplemented with cal- pain and proteasome inhibitors (ALLN: 50 μM; MDL28170: 100 μM; MG132: 50 μM; lactacystin: 20 μM; n = 3-4); (D) DcatCT and NT in an in vitro cleavage assay on corti- cal neurons pretreated with 5 μM CA-074 Me for 15 min before lysis (n = 3). 33

CHAPTER 3

DcatCT and NT can be generated by two independent mechanisms

34

DcatCT and NT can be generated in an NMDAR-dependent manner by calpain cleavage

δ-catenin is localized at synapses and regulates synaptic structure and function

[57] and calpain can be activated via glutamate receptor activation [55]. We examined if the ability of δ-catenin to be cleaved by calpain can be induced by glutamate receptor activation. Primary cortical neurons in culture were treated with 20 μM glutamate for 5 min. After treatment, neurons were lysed in a buffer containing chelators to avoid con- founds from calpain-mediated cleavage occurring during the processing of the neurons and the levels of DcatCT and NT were examined by immunoblot analysis (Fig. 7A). The levels of DcatCT and NT increased with glutamate treatment indicating that they were regulated by glutamate receptor activation. To examine if these effects were mediated by NMDA receptors (NMDAR), similar experiments were performed in the presence of two NMDAR inhibitors, D-AP5 and MK801. Treatment with these inhibitors significantly reduced the levels of DcatCT and NT induced by glutamate treatment (Fig. 7A), indicat- ing that NMDAR activation-mediated cleavage of δ-catenin. To further confirm that this cleavage was calpain-dependent, we performed similar experiments in the presence of two calpain inhibitors, ALLN and MDL28170. The presence of the calpain inhibitors sig- nificantly blocked the cleavage of δ-catenin mediated by glutamate (Fig. 7A). Since

ALLN and MDL28170 also inhibit cathepsin B, to confirm that these effects were indeed mediated by calpain, we performed similar experiments in the presence of the cathepsin

B inhibitor CA-074 Me (CA). This inhibitor was unable to block glutamate induced cleav- age of δ-catenin (Fig. 7B). These results indicate that δ-catenin can be cleaved in an

NMDAR-dependent manner by calpain to generate DcatCT and NT.

35

A

B

Figure 7: Glutamate induces NMDAR- and calpain-dependent generation of

DcatCT and NT.

Immunoblot for DcatCT and NT in cortical neurons treated with 20 μM glutamate for 5 min with 10 min pretreatments with (A) NMDAR or calpain inhibitors (ED- Ctrl: control sample lysed without chelator; D-AP5: 100 µM, MK 801: 10 µM, ALLN: 50 µM and

MDL28170: 100 µM, n = 3), or (B) cathepsin B inhibitor (CA-074 Me: 5 μM, n = 3). 36

We further characterized the glutamate receptor dependence of δ-catenin cleav- age by calpain. Using a similar strategy as above, we confirmed that the cleavage of δ- catenin is mediated by NMDAR, by using NMDA instead of glutamate (Fig. 8A-B). Simi- lar to the results obtained for the glutamate-induced cleavage studies described above, our data indicate that the cleavage of δ-catenin is dependent on NMDAR and can be in- hibited by calpain inhibitors (Fig. 8C-D). The NMDAR and calpain-mediated cleavage of

δ-catenin was unaffected by tetrodotoxin (TTX), a voltage-dependent sodium channel blocker, Nimodipine, a L-type voltage-dependent calcium channel (VGCC) blocker, as- comycin, an inhibitor for the Ca2+-dependent phosphatase calcineurin, and lactacystin.

To further examine the requirement for Ca2+-influx in the ability of δ-catenin to be cleaved, primary neurons were treated with NMDA in Ca2+-free extracellular media and processed as above. The levels of DcatCT and NT were significantly reduced in the samples treated with NMDA in Ca2+-free extracellular media in comparison to media containing Ca2+ (Fig. 8E). These studies indicate that the calpain-mediated cleavage of

δ-catenin is mediated by NMDAR and requires Ca2+-influx through the NMDAR.

37

A B

C D

E

Figure 8: NMDA induces Ca2+-influx and calpain-mediated cleavage of δ- catenin.

Immunoblot for (A) DcatCT in cortical neurons treated with indicated dose of

NMDA/glutamate for 5 min (n = 3); (B) DcatCT in cortical neurons treated with 100

µM NMDA for indicated lengths of time (n = 3); (C) DcatCT and NT in cortical neurons treated with 100 µM NMDA for 5 min with pretreatments with NMDAR inhibitors or (D) calpain inhibitors (10 min pretreatment: TTX: 1 µM, lactacystin: 20 µM, nimodipine

(Nimo): 10 µM, ascomycine (Asco): 2 µM, n = 3-4); (E) DcatCT and NT in cortical neurons treated with 100 µM NMDA for 5 min in extracellular medium with/without

Ca2+ (n = 3). 38

The basal generation of DcatCT and NT is NMDAR-independent

On closer examination, we observed that the DcatCT included two separate bands (I and II), appearing as a duplex. We further characterized these forms using pri- mary neurons and immunoblot analysis (Fig. 9A). DcatCTI and II are very close in mo- lecular weight, 97.13 ± 0.8345 and 89.63 ± 1.302 kDa respectively (n = 8, mean ± SD).

As shown in Fig. 2B, there are two forms of DcatNT that differ in molecular weight of about 7 kDa. These data suggest that there are two calpain sites within δ-catenin that are in close proximity to each other. Consistently, Pearson correlation test showed that

DcatCTI/II and DcatNTII/I were correlated (Fig. 9B). Further, while DcatCTI is the pre- dominant form under basal conditions, the generation of DcatCTII is more sensitive to activity (Fig. 9A). There is a significant decrease in the ratio of DcatCTI to II (DcatCTI/II) and DcatNTII to I (DcatNTII/I) with glutamate treatment (Fig. 9C-D). Taken together, these data indicate that there are likely two sites of calpain cleavage on δ-catenin that differ in sensitivity to activity mediated by NMDAR activation.

39

A

B C D

Figure 9: DcatCT consists of two forms, DcatCTI and II.

(A) Immunoblot for two forms of DcatCT and NT, DcatCTI, CTII, NTI and NTII, in cor- tical neurons treated with NMDA/glutamate for 5 min (NMDA: 50 µM, glutamate: 20

µM). (B) Pearson correlation of the ratio of DcatCTI/II and DcatNTII/I. These data are pooled from replicates of the data shown in (A) and Fig. 7B, control (n = 10), NMDA

(n = 7) and glutamate (n = 6). (C-D) Quantification of DcatCTI/II and DcatNTII/I in the glutamate treated samples in (B) (paired t-test with Bonferroni-Dunn correction; mean

± SD). 40

Our data demonstrate that DcatCT is sensitive to cleavage upon NMDA stimula- tion (Fig. 8), suggesting that the basal level of DcatCTI is generated by basal neural ac- tivity and inhibition of NMDAR would block the generation of DcatCTI. However, blocking neural activity by TTX did not affect the levels of DcatCT and NT (Fig. 10A-C). Con- sistent with a lack of calpain activation under these conditions, there was no difference in the levels of SBDP (spectrin breakdown product) with TTX (Fig. 10D). SBDP is a com- monly used marker whose cleavage is regulated by calpain [55]. These results are sur- prising, considering that the NMDAR is expected to be suppressed under these condi- tions. One possibility is that the mEPSCs are sufficient to induce basal cleavage of δ- catenin. To evaluate this, primary neurons were treated with MK801. Under these condi- tions, there were no changes in the levels of DcatCT and NT (Fig. 11A-C). Similarly, treatment of neurons with the AMPAR inhibitor CNQX, or even CNQX and MK801 co- treatment, had no effect on the levels of DcatCT, NT and SBDP (Fig. 11A-D).

The basal levels of DcatCT and NT could also be affected by degradation of the protein in an NMDAR-dependent manner. To examine if the degradation of δ-catenin mediated by a proteasome-dependent pathway may explain these results, we similarly evaluated the effects of lactacystin-mediated inhibition of proteasome-mediated degra- dation on the levels of DcatCT and NT. Lactacystin treatment significantly increased the level of DcatNT and SBDP, but not CT (Fig. 11F-H), indicating that DcatNT and SBDP are degraded by proteasome (Fig. 11E). Co-treatment of lactacystin and MK801 does not significantly alter the levels of DcatNT and SBDP compared to lactacystin control

(Fig. 11E-H), indicating the basal δ-catenin cleavage is NMDAR-independent. Consist- ently no differences in the levels of DcatCT or NT were observed with D-AP5 treatment

(D-AP5: 50 µM for 3 hr; paired t-test, DcatCT: p = 0.5767, DcatNT: p = 0.2190; SBDP: p

= 0.0912; n = 4). 41

A

B C D

Figure 10: Inhibition of neural activity has no effect on the basal level of DcatCT and NT.

(A) Immunoblot and (B-D) quantification of DcatCT, NT and SBDP in cortical neurons treated with TTX for 3 hr (TTX: 1 µM; paired t-test; DcatCT: p = 0.0982, n = 6;

DcatNT: p = 0.8696, n=7; SBDP: p = 0.1991, n = 5; mean ± SD). 42

A

B C D

E F G H

Figure 11: Inhibition of NMDAR has no effect on the basal level of DcatCT and

NT.

(A) Immunoblot for DcatCT, NT and SBDP in cortical neurons treated with 3 hr inhibi- tors of NMDAR, AMPAR and proteasome (MK801: 10 µM, CNQX: 10 µM, Lact: 10

µM). (C-D) Quantification from samples with NMDAR and AMPAR inhibitors (paired t- test, n = 4, mean ± SD). (E) Diagram of predicted changes on the levels of

DcatCT/NT/SBDP based on generation and degradation. (F-H) Quantification from samples with proteasome inhibitor (paired t-test: Ctrl vs. Lact, n = 4; Lact vs. Lact+

MK, n = 3). 43

Besides NMDAR, another major source for intracellular Ca2+ in neuron is VGCC.

We evaluated the sensitivity of the basal pool of DcatCT and NT to the L-type VGCC blocker, nimodipine. While nimodipine treatment significantly increase the level of SBDP, indicating that under these conditions calpain was activated, this treatment had no effect on the basal levels of DcatCT or NT (Fig. 12).

These data indicate that there is a basal pool of DcatCT and NT that are very similar in size to the CT and NT forms generated by NMDAR-mediated calpain cleavage of δ-catenin. This pool is resistant to alterations by spontaneous neural activity mediated by NMDAR or AMPAR. Further, this pool is also unaltered by Ca2+-influx via L-type

VGCC. However, DcatNT from this pool is degraded in a proteasome-dependent man- ner. Taken together, these data indicate that (1) there is a basal pool of DcatCT and NT that is likely generated in an NMDAR-independent manner, (2) DcatCT and NT from this pool is likely differentially regulated since only the NT, but not the CT is degraded via a proteasome-dependent mechanism.

44

A

B C D

Figure 12: Inhibition of L-type VGCC has no effect on the basal level of DcatCT and NT.

(A) Immunoblot and (B-D) quantification of DcatCT, NT and SBDP in cortical neurons treated with 3 hr L-type VGCC inhibitor (Nimo: 10 µM; paired t-test; DcatCT: p =

0.1410; DcatNT: p = 0.1408; n = 7; mean ± SD). 45

Lysosomes constitute another major protein degradation pathway in neurons. To examine if DcatCT is degraded via a lysosome-dependent pathway, we took advantage of a V-ATPase inhibitor, bafiloymince A1 (Baf). To test this, primary neurons were incu- bated with Baf for 3 hr. Under these conditions, it is expected that the Baf will alter the lysosomal pH thus inhibiting its function [58]. Baf treatment did not affect the levels of

DcatCT, but significantly reduced the levels of NT (Fig. 13A-C). These results could be explained by a model in which both cleavage of full-length δ-catenin and degradation of

DcatCT are both regulated by the lysosome (Fig. 11E).

To confirm that Baf treatment effects on δ-catenin are not a secondary conse- quence of cell stress and cell death, we examined the levels of SBDP (Fig. 13D) and ob- served no significant differences in the levels of this product in control or Baf-treated neurons. To further confirm that these results are not a secondary consequence of ER stress in neurons under Baf-treated conditions, we examined the levels of DcatCT and

NT in neurons treated with thapsigargin (Thap) for 3 hrs. Under these conditions, Thap is expected to induce ER stress [59] and DcatCT or NT as well as SBDP were not signifi- cantly altered (Fig. 13A-D). These data indicate that generation and degradation of

DcatCT under basal conditions is regulated by the lysosome while only the generation but not degradation of DcatNT is regulated by the lysosome.

Further, treatment of primary unstimulated neurons with MDL28170 altered the levels of DcatCT and NT in a manner similar to the results obtained with Baf, in which the levels of DcatCT was unaltered by treatment but the levels of NT was significantly re- duced (Fig. 13A-C). MDL28170 inhibits the classic calpains and cathepsin B but there is less support to indicate that it effective inhibits atypical calpains [60,61]. Under these conditions, SBDP was also not altered in levels (Fig. 13D), indicating that under basal 46

conditions, the typical calpains are not activated and it is uncertain whether the basal generation SBDP is mediated by atypical calpain. The decreased levels of DcatNT under these conditions in consistent with a model in which MDL28170 inhibits cathepsin B, which in turn disrupts lysosome function thus decreasing δ-catenin cleavage. To confirm these results, primary neurons were treated with CA-074 Me. This treatment caused a significant decrease of DcatNT but not change of CT or SBDP (Fig. 14). Thus, the inhibi- tion of cathepsin B is sufficient to inhibit the basal generation of DcatCT and NT. These results, taken together with the data above, suggest that DcatCT and NT exist in multiple forms. A fraction of the DcatCT and NT are generated in an NMDAR-dependent manner by typical calpain-mediated cleavage. Another fraction is likely not cleaved by the typical calpains, but is generated in a lysosome-dependent manner.

47

A

B C D

Figure 13: Disrupt the normal function of lysosome decreases the basal level of

DcatNT, but not CT.

(A) Immunoblot and (B-D) quantification of DcatCT, NT and SBDP in cortical neurons treated with 3 hr bafilomycin A1 (Baf), thapsigargin (Thap) and MDL28170 (Thap: 200 nM, Baf: nM, MDL: µM; paired t-test with Bonferroni-Dunn’s correction, n = 4-5, mean

± SD). 48

A

B C D

Figure 14: Inhibition of cathepsin B decreases the basal level of DcatNT, but not

CT.

(A) Immunoblot and (B-D) quantification of DcatCT, NT and SBDP in cortical neurons treated with 3 hr cathepsin B inhibitor (CA-074 Me: 1 µM; paired t-test, n = 4, mean ±

SD). 49

To assess if the basal cleavage and NMDAR-dependent cleavage of δ-catenin are correlated, we examined the levels of DcatCT and NT in neurons treated with

NMDA/KCl after 3 hr Baf treatment (Fig.14A). Baf had no effect on NMDAR-induced cleavage of δ-catenin and ɒII-spectrin (Fig. 15B-D), indicating that the basal and

NMDAR-mediated cleavage of δ-catenin are two independent pathways. Interestingly, with depletion of synaptic vesicle by Baf, KCl still managed to induce δ-catenin cleavage, but not able to reduce DcatI/II and NTII/I (Fig. 15B-C, 15E-F). It is unclear whether the

KCl-induced δ-catenin cleavage is generated by VGCC activation, or NMDAR activation by synaptic vesicles belong to a reserve/inactive pool. Moreover, NMDA induced de- crease of DcatCTI/II and NTII/I were significantly lower compared to KCl (paired t-test, both p-value < 0.01, n = 4). These results are consistent with the previous observation that NMDAR activation preferably cleaves δ-catenin at the second cleavage site (Fig.

9C-D).

Taken together, these data indicate that (1) there are multiple forms of DcatCT and NT; (2) some of the DcatCT and NT forms are generated in an NMDAR-dependent manner by cleavage of δ-catenin by calpain; (3) some of the DcatCT and NT forms are generated in a lysosome-dependent and NMDA-independent manner; (4) DcatCT and

NT are degraded by independent mechanisms.

50

A

B C D

E F

Figure 15: The basal and NMDAR-dependent forms of DcatCT/NT are generated in two independent pathways.

(A) Immunoblot and (B-D) quantification of DcatCT, NT and SBDP, (E-F) DcatCTI/II and DcatNTII/I, in cortical neurons treated with 5 min NMDA/KCl after 3 hr Baf pre- treatment (Baf: 50 nM, NMDA: 50 µM for 5 min, KCl: extra 55 mM for 5 min; repeated measure two-way ANOVA, Dunnett's multiple comparisons test to the control sam- ples of both groups, n = 4, mean ± CI). 51

CHAPTER 4

The functional role of δ-catenin cleavage in dendritic protrusion development

52

Mapping the calpain cleavage sites on δ-catenin

As described above, the calpain cleavage in δ-catenin can be recapitulated in

HEK cells, since the molecular weights of the DcatCT generated in HEK cells are similar in size to those generated in neurons (Fig. 16A-B). Thus, we decided to take advantage of the cleavage assay in HEK cells to map the sites of calpain cleavage in δ-catenin.

Based on antibody epitope recognition site of the J19 antibody (392-309 aa) and its abil- ity to recognize the DcatNT (Fig. 2B), we reasoned that the cleavage site of calpain in δ- catenin was located in the region C-terminal to this epitope region. Further, the δ-catenin

N-term mice express a fragment of δ-catenin up to 461 aa. Our data indicate that this form can be cleaved by calpain (Fig. 2B). This led us to hypothesize that the site of cleavage is located within 310-460 aa. We generated several constructs of δ-catenin that included various deletions in this region, as shown in Fig. 16C. These constructs were transiently transfected in HEK cells, lysates prepared and the expression of

DcatCT were detected by immunoblot analysis (Fig. 16D). For these studies, lysates were prepared in buffers lacking chelator to allow for sufficient calpain activation.

Using this assay, our data (Fig. 16D) indicate that a construct that includes a de- letion in δ-catenin spanning amino acids 321-440 (DI321440) is significantly resistant to cleavage. A construct that includes a deletion in aa 321-420 (DI321420) allows genera- tion of DcatCTII, but not DcatCTI. Further, deletion of aa 351-390, but not aa 321-350, significantly blocks generation of DcatCTI. Taken together, these results indicate that the cleavage site that allows for generation of DcatCTI is within the region that includes aa

351-390. 53

We also note that while constructs that include deletions in aa 351-370

(DI351370) and aa 371-390 (DI371390) decrease the levels of DcatCTI that are gener- ated in comparison to the full-length δ-catenin, the constructs are not completely re- sistant to cleavage. Similar results were observed in case of DcatCTII with constructs that included deletions of aa 420-430 (DI420430) and aa 431-450 (DI431450). If the sec- ond calpain cleavage site exists within the region of aa 420-440, one would predict that construct that included deletions in this region (Dl391435, aa 391-435) would be re- sistant to cleavage that would allow generation of DcatCTII. However, we do observe some cleavage in constructs expressing these deletions. The deletion generated in the constructs would lead to a shorter form of the protein. Since the deleted form still has the cleavage site for DcatCTI, we may be unable to specifically identify DcatCTII in these ly- sates based on size. However, the 391-435 construct is more resistant to cleavage than full-length δ-catenin. Taken together, these data indicate that the regions between aa

351-390 and 390-440 are likely cleaved by calpain to generate DcatCTI and II respec- tively.

54

A B C

D

Figure 16: Mapping the calpain cleavage sites on δ-catenin by in vitro cleavage assay.

(A) The immunoblot represents an overlay of single blot adjusted to two different in- tensities to illustrate molecular weights of DcatCTI/II from in vitro cleavage assay and glutamate treatment on cortical neurons (red: light adjustment, black: dark adjust- ment, n = 4). (B) Quantification of molecular weights of DcatCTI and II generated from in vitro cleavage assay on δ-catenin-transfected HEK cell and glutamate/NMDA treated cortical neurons (two way-ANOVA with repeated measures between DcatCTI and II, Sidak’s multiple comparison; neuron: n = 8; HEK: n = 18; mean ± SD). (C) schematic of individual constructs of δ-catenin with indicated deletions between 301-

400 aa (the numbers of the construct name reflect amino acid numbers for the dele- tion on mouse δ-catenin). The thin line between black bars indicates the deletion area for each construct. The highlighted region and red dot line indicate 351-440 aa and

390-391 aa of mouse δ-catenin (D) Immunoblot for in vitro generation of DcatCT in

HEK cells transfected with indicated δ-catenin deletion constructs (N ≥ 3). 55

δ-catenin affects dendritic protrusion density

We sought to examine the functional role of calpain- and NMDAR-mediated cleavage of δ-catenin. To this end, we took advantage of the constructs of δ-catenin de- scribed in Fig. 16C that are resistant to calpain-mediated cleavage, particularly the 3 constructs DI321420, DI321440 and DI391435. We have previously demonstrated that knockdown of δ-catenin leads to an increase in excitatory spine density that is accompa- nied by an increase of mEPSC frequency [33]. Similar results are observed in primary neurons from the Dcat N-term mice [33].

Primary rat neurons in culture (DIV 12) were transfected with vector, shRNA to δ- catenin (δ-cat shRNA) or shRNA with either wild-type δ-catenin or the three calpain-re- sistant forms of δ-catenin. Neurons were fixed at DIV 18 and dendritic protrusion densi- ties were examined by confocal microscopy (Fig. 17A). In neurons expressing δ-cat shRNA, there was a significant increase in the density of dendritic protrusions that could be rescued by the introduction of full-length δ-catenin. We performed a similar analysis in neurons expressing shRNA and either of the three calpain-resistant constructs of δ- catenin. Our data demonstrate that the DI321420 can significantly rescue the shRNA knockdown phenotype (Fig. 17B). However, the DI391435 construct demonstrates bor- derline rescue and the DI321440 does not rescue, however, the p-values are very close to being significant (p = 0.0504). This led us to believe that there might be population dif- ferences within the groups and not taking these different populations in account might be obscuring significant data. Dendritic protrusions have various width and length partly re- flecting synapses at different levels of maturity or development [62,63]. To assess the distribution of various groups of dendritic protrusions, we examined the protrusion popu- 56

lation in vector expressing neurons and classified them into three groups based on pro- trusion length and width (Fig. 17C). These groups were classified as population 1 (P1), width ≤ 0.5 µm and length ≤ 1 µm; population 2 (P2): width ≤ 0.5 µm and length > 1 µm; population 3 (P3): width > 0.5 µm. According to the previous publications [62], these three populations, P1, P2 and P3, include stubby and short-thin spines or filopodia, long thin-spines or filopodia and mushroom spines respectively. We examined the distribution of dendritic protrusions in these three populations in neurons from vector, δ-cat shRNA or δ-catenin groups (Fig. 17D). In neurons expressing δ-cat shRNA, the density of den- dritic protrusions classified into either of the three categories was increased with δ- catenin knockdown. The effects of the knockdown on P1 an P2 could be significantly rescued by the wild-type δ-catenin. We similarly examined the ability of the three cal- pain-resistant constructs of δ-catenin to rescue the increase in protrusion density in- duced by δ-catenin knockdown in the P1 and P2 groups of protrusions (Fig. 17D). The increase in protrusion density mediated by δ-cat shRNA was rescued by the DI321420 and DI391435, which still can be partially cleaved; but not DI321440, which inhibits cleavage on both sites, in the P1 protrusions while the DI321420 and DI321440 but not

DI391435 rescued the phenotype of δ-catenin knockdown in the P2 protrusions. These results indicate that the calpain cleavage is necessary for the ability of δ-catenin to regu- late the density of the P1 protrusions. The phenotype of the P2 protrusions may be slightly more complex and may reflect incomplete or partial calpain cleavage and needs further investigation.

57

A C

D

B

Figure 17: Cleavage of δ-catenin regulates the dendritic protrusion density.

(A) Representative images of dendritic segments from hippocampal neurons trans- fected with vector, δ-catenin shRNA (δ-cat shRNA), DI321420, DI321440, DI391435 and δ-catenin. (B) Quantification of the density of dendritic protrusion for each group.

The densities were normalized to the geometric mean of the vector group in each ex- periment (Kruskal-Wallis test, Dunn’s multiple comparison test: vector, δ-catenin shRNA and δ-catenin (grey); δ-catenin shRNA, DI321420, DI321440 and DI391435

(black). 59-80 neurons per group from 4 independent experiments; median ± 25/75%

(box) and min/max (whisker)). 58

(C) Criteria for separation of dendritic protrusion into three groups. Individual dendritic protrusions from the vector group represented by a dot and plotted according to the width (x-axis) and length (y-axis) in the top-middle panel. The frequency distribution of the width of all the protrusions are shown in the bottom-middle panel and a threshold was drawn (red line) to separate the protrusions into two populations, P1+P2 and P3, according to the difference in distribution. The frequency distribution of the length of

P1+P2 is shown in the left panel and a threshold was drawn (blue line) to separate the population into two, P1 and P2, according to the difference in distribution. The dis- tribution of population in each of the P1-3 population is shown on the right panel.

3443 protrusions from 4 independent experiments. (D) Quantification of the approxi- mate density of each population of dendritic protrusions in groups indicated. The den- sities were normalized to the median of the vector group in each experiment (Kruskal-

Wallis test, Dunn’s multiple comparison test: vector, δ-catenin shRNA and δ-catenin for all populations (grey); δ-catenin shRNA, DI321420 and DI321440, DI391435 for

P1 and P2 (black). 50-72 neurons per group from 4 independent experiments; median

± 25/75% (box) and min/max(whisker)). 59

We further characterized the effects of the calpain-resistant form of δ-catenin on the P1 protrusions. The density of the P1 protrusions can be defined as the product of the percentage of P1 protrusions in the whole population and the dendritic protrusion density (see methods). Thus, the increase in the density of P1 protrusions with knock- down of δ-catenin could be a function of increase of one or both of these parameters. To examine this relationship, we plotted the dendritic protrusion density and P1 percentage for individual neurons from each group (Fig. 18A). The heatmap indicates the density of the P1 protrusions in each group (Fig. 18A). In neurons expressing vector, the percent- age of the P1 protrusions decreased with an increase in the density of protrusions. One possible explanation of this observation is that there is an upper limit to the density of P1 protrusions in individual dendrite. If such a limit exists, the prediction would be that (1) a threshold can be drawn with the density of P1 protrusions and the majority of dendrite should have a density of P1 protrusions that fall below the threshold; (2) frequency distri- bution of density of P1 protrusions from all the dendrite would demonstrate a non-para- metric distribution and a small population of dendrite would pass the threshold.

According to these criteria, we defined a threshold by examining the frequency distribution of the density of P1 protrusions in the vector group by bootstrapping analysis

(Fig. 18B). This threshold defines the predicted upper limit of the density of P1 protru- sions in the vector group. Thus, we define two populations of P1 density, one above and one below the threshold. Once we define this threshold, we sought to examine if the fraction of P1 density above the threshold differ in neurons expressing δ-cat shRNA or the different constructs described above by bootstrapping (Fig. 18C). In neurons ex- pressing δ-cat shRNA, the fraction of P1 protrusions that were above the threshold were significantly increased compared to vector. This knockdown phenotype could be rescued by all the constructs, including wild-type δ-catenin, but not the DI321440. These results 60

are consistent with the model in which δ-catenin regulates the upper limit of density of

P1 protrusions in each neuron and mechanically requires calpain-mediated cleavage to perform this functional role.

61

A

B C

Figure 18: Cleavage of δ-catenin regulates the upper limit of the density of P1 protrusions.

(A) Relationship between the percentage of P1 protrusion and dendritic protrusion density in each group. Individual neurons from each group are represented by a dot and plotted according to the normalized dendritic protrusion density (x-axis) and nor- malized P1 fraction (y-axis). The normalized approximate P1 density of each neuron is indicated by heatmap. The P1 fraction were normalized to the median of the vector group in each experiment (50-72 neurons per group from 4 independent experi- ments). The P1 density threshold is indicated by the blackline. 62

(B) Estimation of the P1 density threshold. Probability density estimates were calcu- lated from the normalized approximate P1 density from resampling the vector group

(black line; 72 neurons from 4 independent experiments) and a threshold was esti- mated and is indicated by a red dot line. (C) Bootstrap comparison of the fraction of neurons with above threshold P1 density in groups indicated (p-value are calculated by bootstrap paired t-test with δ-cat shRNA, with Bonferroni-Dunn correction, 50-72 neurons per group from 4 independent experiments, median ± CIs).

63

Discussion

In this study, we have made several important observations: (1) multiple novel forms of δ-catenin that include only the N-terminal region (DcatNT) or C-terminal region (DcatCT) are differentially expressed in different regions of the brain; (2) some forms of DcatCT and NT are generated by an NMDAR, Ca2+ and calpain-dependent mechanism; (3) some forms of DcatCT and NT are generated by an NMDAR-independ- ent, but lysosome-dependent mechanism; (4) the DcatCT and NT forms can be de- graded by independent mechanisms; (5) δ-catenin regulates the density of a subpopula- tion of dendritic protrusions and this ability of δ-catenin is dependent on the domain of δ- catenin that includes the predicted calpain cleavage site. These studies thus identify en- dogenous novel forms of δ-catenin and establish a functional role associated with these forms.

Different forms of δ-catenin

Our data demonstrate that novel forms of δ-catenin, DcatCT and NT, are ex- pressed differentially in different regions of the brain. While our studies have focused on the cortex and the hippocampus, the expression of these forms in different regions of the brain implies that they may have additional functional roles that may or may not be re- lated to synaptic regulation. Our data clearly demonstrate that some forms of DcatCT and NT are generated by calpain-mediated cleavage (Fig. 7-8). However, some forms of

DcatCT and NT may or may not be dependent on typical calpain for generation (Fig. 13).

The lysosome is an organelle that allows for protein degradation in an acidic environ- ment. Our data, in this regard is surprising, show lysosomal function regulates both gen- eration and degradation of basal forms of DcatCT and the generation of basal forms of 64

DcatNT. Certainly, other mechanisms that might allow the generation of these forms ex- ist, including translational or proteolytic mechanisms.

We note that there are multiple forms of DcatCT and NT that differ slightly in size.

Based on the differences in the molecular weights, we predict that these forms differ from each other by about 50-70 aa. While these differences are small, they are suffi- cient, in theory, to allow differential functions for the different forms, especially via mech- anisms that involve post-translational modifications or protein-protein interactions.

One possible proteolytic mechanism that may allow cleavage of δ-catenin to gen- erate the basal forms of DcatCT and NT may be via an atypical calpain-dependent mechanism. Much less is known about the atypical calpains in comparison to the typical calpains. There are six members of atypical calpain [56]. In the brain, calpain 5, 7, 10 and 15 are expressed in different regions, including hippocampus and cortex [24,25,64–

66]. It is unclear if the typical calpain inhibitors, including MDL28170, are effective in blocking the activity of atypical calpains similar to typical calpain [60,61]. Our data demonstrate that MDL28170 is unable to inhibit the basal generation of SBDP, a known cleavage product of the typical calpains [67]. This suggests that either the typical cal- pains are negligibly activated in unstimulated cortical neurons or basal generation of

SBDP is mediated by proteases that are similar in substrate-specificity to calpain. One possibility is that this protease includes members of the atypical calpain family. Further experimental verification is necessary to address these possibilities.

65

Differential regulation of DcatCT and NT levels by degradation

Our data indicate that the two forms of δ-catenin, DcatCT and NT, can be differ- entially degraded. Our data also demonstrate that inability to cleave δ-catenin has detri- mental consequences on the density of a subset of dendritic protrusions in hippocampal neurons. However, our studies do not exclude the possibility that the two forms DcatCT and NT have additional functional roles that are unique to each form. The C-terminal of

δ-catenin includes regions that are necessary for cadherin binding and PDZ-interactions and a predicted NLS motif [57]. This region also includes the palmitoylation site on δ- catenin [34] that is key for the ability of δ-catenin to regulate synaptic plasticity. The C- terminal half of the protein likely also contains interaction sites for other currently uniden- tified proteins. Similarly, the N-terminal region of δ-catenin includes the interaction sites for DIPA family of proteins [68]. Further, the cleaved DcatCT and NT are likely structur- ally distinct from the full-length protein thus allowing for differential interactions with other proteins. Thus, it is certainly possible that the cleaved forms of δ-catenin have additional functional roles that are distinct from the full-length forms. The use of cleaved forms of proteins to perform functional roles that are different from the full-length protein is widely used by different cell types [69–72]. Such functional roles for δ-catenin remain to be identified and are likely to be of great interest to neural circuit formation and function in the developing brain.

Regulation of dendritic protrusions by proteolytic cleavage of δ-catenin

Our data indicate that knockdown of δ-catenin predominantly increases the popu- lations of two groups of dendritic protrusions, the P1 and P2, represent the stubby and short thin-spines or filopodia, and long thin-spines or filopodia respectively. The increase in the percentage of the P1 protrusions with knockdown of δ-catenin cannot be rescued 66

by constructs that lack the region that includes the calpain cleavage site, suggesting that this region is necessary for the ability of δ-catenin to regulate the density of P1 protru- sions. The increase in the density of P2 observed with knockdown of δ-catenin can be rescued by constructs that lack the calpain cleavage site and the full-length δ-catenin protein, indicating that the full-length protein regulates the density of P2 independent of the ability of δ-catenin to be cleaved while the functional effects of δ-catenin on P1 pro- trusions require it to be proteolytically processed. Thus, δ-catenin can differentially regu- late the density of two subgroups of dendritic protrusions in a manner dependent on its ability to be subjected to proteolytic regulation. This is a novel mechanism for regulating densities of different protrusion populations by regulation of a single molecule.

The P1 population includes stubby spines and short thin-spines and short filopo- dia. The short filopodia likely represent an intermediate stage along the process of pro- trusion formation or elimination. While the stubby and short-thin spines are expected to have active synapses [62,73], the stability of both these populations requires contact-de- pendent mechanisms [62,63]. Our data indicate that the cleavage of δ-catenin is directly linked to the upper limit of the density of protrusions as defined by the P1 population.

This implies that neurons have an inherent mechanism to regulate the density of this class of protrusions. To do this, it is essential that the neuron has the ability to compute the density of protrusions and evaluate if the density of P1 protrusions is close to the up- per limit and initiate a cascade of events that maintains the density of these protrusions.

How δ-catenin participates in this computation and contributes to regulating the density of P1 protrusions is not entirely clear. However, it is tempting to speculate that synaptic inputs from NMDAR contributes to this evaluation process. While our data do not provide direct evidence for a direct functional role for cadherin-mediated regulation of these pro- cesses in the ability of δ-catenin to regulate protrusion subpopulations, it is tempting to 67

speculate that proteolytic regulation of δ-catenin modulates its ability to interact with cad- herin and influence the adhesive properties of cadherin thereby indirectly influencing synaptic contacts. Further, since the cadherin-catenin complex is linked to the actin cyto- skeleton [2], the alterations in the interactions between δ-catenin and cadherin could, in theory, promote signaling to the actin cytoskeleton thereby influencing protrusion density or stabilization. Similarly, since δ-catenin associated with cadherin binds to several PDZ domain-containing synaptic proteins [57], it is possible that cleavage of δ-catenin influ- ences its association with PDZ domain scaffolding proteins thus impacting synapse as- sembly or disassembly.

In summary, we have identified novel forms of δ-catenin that are differentially regulated both in terms of their generation and degradation by protease- and lysosome- dependent mechanisms. Further, we have identified a functional role for these forms in regulating the density of a subgroup of dendritic protrusions. Taken together with previ- ous data [33], these data suggest that δ-catenin is a critical regulator of excitatory synap- ses in central neurons and thus a critical regulator of neural circuit formation in the de- veloping brain. More recent data indicate that loss of δ-catenin correlates with intellec- tual disability and mutations in CTNND2 encoding δ-catenin have been identified in au- tism [32]. We propose that loss of δ-catenin in intellectual disability or autism mutations perturb its ability to regulate excitatory synapses thus perturbing neural circuit formation and contributing to the behavioral deficits observed in these disorders. Identifying how autism mutations in CTNND2 perturb neural circuit formation and function will likely have a great impact on our ability to pharmacologically intervene for therapy in this and re- lated forms of autism and related disorders.

68

Future directions

Identify regulatory mechanisms for δ-catenin cleavage

δ-catenin can be divided into multiple pools including those associated with cad- herin and a nuclear pool [57]. Further, the protein has multiple sites that can be post- translational modified [57]. Since calpain likely recognizes its substrates by recognizing a 3D structure and not a sequence of amino acids [56], the 3D structure of δ-catenin is likely to be critical for its ability to serve as a substrate for calpain. Association with cellu- lar molecules in different cellular compartments and post-translational modifications could alter the 3D structure of δ-catenin thus likely altering its susceptibility to calpain- mediated cleavage and downstream functional roles. While our data identify a pool of δ- catenin that is cleaved and regulates dendritic protrusions, it does not exclude the possi- bility that other pools of δ-catenin are similarly cleaved. It would be interesting to identify the pools of δ-catenin that are in a conformation that make it vulnerable to calpain cleav- age and identify the functional roles of this cleavage in diverse cellular compartments and brain regions.

Identify the functional consequence for calpain-mediated cleavage of δ-catenin in exci- totoxicity

Extra glutamate is released into synaptic cleft and induce excitotoxicity in differ- ent disease models, for example epilepsy, traumatic brain injury and stroke [74,75]. Ex- tensive studies have been done on targeting calpain-mediated cleavage for neural pro- tection under these disease conditions [55,56]. Besides preventing neural death, it is also crucial to prevent synaptic damage and preserve the normal neural function and connectivity under excitotoxicity conditions. δ-catenin is known to be a structural compo- nent of dendritic protrusions [57], it is unclear whether the increase δ-catenin cleavage 69

mediated by calpain is a consequence or a protective mechanism for excitotoxicity in neurons. By inducing excitotoxicity on neurons with calpain-resistant δ-catenin, we will get a better understanding on the functional consequence of δ-catenin cleavage in neu- rotoxicity and how to prevent synaptic damage by regulating δ-catenin cleavage by cal- pain.

Identify the mechanism for the basal generation of DcatCT and NT

Our data demonstrate the basal and NMDAR-dependent generation of DcatCT and NT are mediated via two independent pathways. Although we demonstrate that the basal generation is regulated by lysosome function, the exact mechanism is entirely un- clear. Since there is currently no evidence of lysosomal localization of δ-catenin, the ba- sal generation of DcatCT and NT could be localized in other subcellular compartments and regulated by signal modulators from lysosome. Since lysosome is an important or- ganelle for cellular processes, it would not be a surprise that it plays a critical role in reg- ulating dendritic protrusion development. However, little is known about this process

[76,77]. Thus, by understanding the mechanism for basal generation of DcatCT and NT, we can provide a new link between lysosome and dendritic protrusion development.

Moreover, by understanding the downstream regulators of lysosome, diagnostic tools and therapeutic targets could be developed.

Identify additional functional roles for DcatCT and NT

Our data indicate that DcatCT and NT are differentially expressed and regulated in different regions of the brain. This raises the possibility that in addition to the func- tional roles we have described, the DcatCT and NT forms may have additional functional 70

roles that are unique to each form. To address these questions, it is necessary to identify the subcellular localization and distribution of these forms in neurons and establish mouse models that express or lack these forms in a conditional manner. The availability of this data and tools would allow the dissection of additional functional roles that might be performed by these variants.

Implications for autism

Mutations in CTNND2 have been identified in autism. These mutations are point mutations and span the entire length of the protein and do not show preference for a sin- gle domain of the protein suggesting that they may vary in their inability to participate in neuronal functions. It would be interesting to examine the ability of these mutations to be cleaved by calpain and establish how these mutations affect the ability of δ-catenin to participate in synaptic regulation. Interestingly, the R454H mutation is in close proximity to the predicted region of calpain cleavage. It is tempting to speculate that the presence of this mutation will alter the 3D structure of δ-catenin and thus alter its ability to serve as a substrate for calpain and regulate synapses. Such studies would highly benefit from the availability of mouse models that harbor this mutation in CTNND2.

71

Reference

[1] S.C. Suzuki, M. Takeichi, Cadherins in neuronal morphogenesis and function,

Dev. Growth Differ. 50 (2008) S119–S130. doi:10.1111/j.1440-

169X.2008.01002.x.

[2] E. Seong, L. Yuan, J. Arikkath, Cadherins and catenins in dendrite and synapse

morphogenesis, Cell Adh. Migr. 9 (2015) 202–213.

doi:10.4161/19336918.2014.994919 [doi].

[3] J. Arikkath, Molecular mechanisms of dendrite morphogenesis., Front. Cell.

Neurosci. 6 (2012) 61. doi:10.3389/fncel.2012.00061.

[4] F. Mills, A.K. Globa, S. Liu, C.M. Cowan, M. Mobasser, A.G. Phillips, S.L.

Borgland, S.X. Bamji, Cadherins mediate cocaine-induced synaptic plasticity and

behavioral conditioning, Nat. Neurosci. (2017). doi:10.1038/nn.4503.

[5] K.A. Maguschak, K.J. Ressler, β-catenin is required for memory consolidation,

Nat. Neurosci. 11 (2008) 1319–1326. doi:10.1038/nn.2198.

[6] Y. Sun, S.X. Bamji, β-Pix Modulates Actin-Mediated Recruitment of Synaptic

Vesicles to Synapses, J. Neurosci. 31 (2011) 17123–17133.

doi:10.1523/JNEUROSCI.2359-11.2011.

[7] S.X. Bamji, K. Shimazu, N. Kimes, J. Huelsken, W. Birchmeier, B. Lu, L.F.

Reichardt, Role of β-catenin in synaptic vesicle localization and presynaptic

assembly., Neuron. 40 (2003) 719–31.

http://www.ncbi.nlm.nih.gov/pubmed/14622577 (accessed March 19, 2017).

[8] A.S. Yap, C.M. Niessen, B.M. Gumbiner, The juxtamembrane region of the

cadherin cytoplasmic tail supports lateral clustering, adhesive strengthening, and

interaction with p120ctn, J. Cell Biol. 141 (1998) 779–789. 72

[9] R. Paffenholz, W.W. Franke, Identification and localization of a neurally expressed

member of the plakoglobin/armadillo multigene family, Differentiation. 61 (1997)

293–304. doi:S0301-4681(09)60682-4 [pii].

[10] R.H. Carnahan, A. Rokas, E.A. Gaucher, A.B. Reynolds, The molecular evolution

of the p120-catenin subfamily and its functional associations, PLoS One. 5 (2010)

e15747. doi:10.1371/journal.pone.0015747 [doi].

[11] A. Keirsebilck, S. Bonne, K. Staes, J. van Hengel, F. Nollet, A. Reynolds, F. van

Roy, Molecular cloning of the human p120ctn catenin gene (CTNND1):

expression of multiple alternatively spliced isoforms, Genomics. 50 (1998) 129–

146. doi:S0888-7543(98)95325-3 [pii].

[12] H. Sirotkin, B. Morrow, B. Saint-Jore, A. Puech, R. Das Gupta, S.R. Patanjali, A.

Skoultchi, S.M. Weissman, R. Kucherlapati, Identification, characterization, and

precise mapping of a human gene encoding a novel membrane-spanning protein

from the 22q11 region deleted in velo-cardio-facial syndrome, Genomics. 42

(1997) 245–251. doi:S0888754397947340 [pii].

[13] A.F. Paulson, E. Mooney, X. Fang, H. Ji, P.D. McCrea, Xarvcf, Xenopus member

of the p120 catenin subfamily associating with cadherin juxtamembrane region, J.

Biol. Chem. 275 (2000) 30124–30131. doi:10.1074/jbc.M003048200 [doi].

[14] M. Hatzfeld, C. Nachtsheim, Cloning and characterization of a new armadillo

family member, p0071, associated with the junctional plaque: evidence for a

subfamily of closely related proteins, J. Cell Sci. 109 ( Pt 1 (1996) 2767–2778.

[15] Q. Lu, B.J. Aguilar, M. Li, Y. Jiang, Y.-H. Chen, Genetic alterations of δ-

catenin/NPRAP/Neurojungin (CTNND2): functional implications in complex human

diseases, Hum. Genet. 135 (2016) 1107–1116. doi:10.1007/s00439-016-1705-3. 73

[16] Y. Kawamura, Q.W. Fan, H. Hayashi, M. Michikawa, K. Yanagisawa, H. Komano,

Expression of the mRNA for two isoforms of neural plakophilin-related arm-repeat

protein/delta-catenin in rodent neurons and glial cells, Neurosci. Lett. 277 (1999)

185–188. doi:S0304394099008757 [pii].

[17] N. Ishiyama, S.H. Lee, S. Liu, G.Y. Li, M.J. Smith, L.F. Reichardt, M. Ikura,

Dynamic and static interactions between p120 catenin and E-cadherin regulate

the stability of cell-cell adhesion, Cell. 141 (2010) 117–128.

doi:10.1016/j.cell.2010.01.017 [doi].

[18] N. Chauvet, M. Prieto, C. Fabre, N.K. Noren, A. Privat, Distribution of p120

catenin during rat brain development: potential role in regulation of cadherin-

mediated adhesion and actin cytoskeleton organization, Mol. Cell. Neurosci. 22

(2003) 467–486. doi:S1044743103000307 [pii].

[19] R.-H. Duparc, D. Boutemmine, M.-P. Champagne, N. Tétreault, G. Bernier, Pax6

is required for delta-catenin/neurojugin expression during retinal, cerebellar and

cortical development in mice, Dev. Biol. 300 (2006) 647–655.

doi:10.1016/j.ydbio.2006.07.045.

[20] C. Foldy, S. Darmanis, J. Aoto, R.C. Malenka, S.R. Quake, T.C. Sudhof, Single-

cell RNAseq reveals cell adhesion molecule profiles in electrophysiologically

defined neurons, Proc. Natl. Acad. Sci. U. S. A. 113 (2016) E5222-31.

doi:10.1073/pnas.1610155113 [doi].

[21] R.P. Munton, R. Tweedie-Cullen, M. Livingstone-Zatchej, F. Weinandy, M.

Waidelich, D. Longo, P. Gehrig, F. Potthast, D. Rutishauser, B. Gerrits, C. Panse,

R. Schlapbach, I.M. Mansuy, Qualitative and quantitative analyses of protein

phosphorylation in naive and stimulated mouse synaptosomal preparations, Mol.

Cell. Proteomics. 6 (2007) 283–293. doi:M600046-MCP200 [pii]. 74

[22] M. Lee, H. Ji, Y. Furuta, J.I. Park, P.D. McCrea, p120-catenin regulates REST

and CoREST, and modulates mouse embryonic stem cell differentiation, J. Cell

Sci. 127 (2014) 4037–4051. doi:10.1242/jcs.151944 [doi].

[23] K. Toyo-oka, T. Wachi, R.F. Hunt, S.C. Baraban, S. Taya, H. Ramshaw, K.

Kaibuchi, Q.P. Schwarz, A.F. Lopez, A. Wynshaw-Boris, 14-3-3epsilon and Zeta

Regulate Neurogenesis and Differentiation of Neuronal Progenitor Cells in the

Developing Brain, J. Neurosci. 34 (2014) 12168–12181.

doi:10.1523/JNEUROSCI.2513-13.2014 [doi].

[24] Allen Institute for Brain Science, Allen Brain Atlas Data Portal: Cell Taxonomy,

(n.d.). http://casestudies.brain-map.org/celltax (accessed September 1, 2017).

[25] B. Tasic, V. Menon, T.N.T. Nguyen, T.T.K. Kim, T. Jarsky, Z. Yao, B.B. Levi, L.T.

Gray, S.A. Sorensen, T. Dolbeare, D. Bertagnolli, J. Goldy, N. Shapovalova, S.

Parry, C.C. Lee, K. Smith, A. Bernard, L. Madisen, S.M. Sunkin, M. Hawrylycz, C.

Koch, H. Zeng, Z. Yao, C.C. Lee, N. Shapovalova, S. Parry, L. Madisen, S.M.

Sunkin, M. Hawrylycz, C. Koch, H. Zeng, Adult mouse cortical cell taxonomy

revealed by single cell transcriptomics, Nat. Neurosci. 19 (2016) 335–346.

doi:10.1038/nn.4216.

[26] M.C. Martinez, T. Ochiishi, M. Majewski, K.S. Kosik, Dual regulation of neuronal

morphogenesis by a delta-catenin-cortactin complex and Rho, J. Cell Biol. 162

(2003) 99–111. doi:10.1083/jcb.200211025 [doi].

[27] J. Arikkath, I. Israely, Y. Tao, L. Mei, X. Liu, L.F. Reichardt, Erbin controls

dendritic morphogenesis by regulating localization of delta-catenin, J. Neurosci.

28 (2008) 7047–7056. doi:10.1523/JNEUROSCI.0451-08.2008 [doi].

[28] L. Yuan, E. Seong, J.L. Beuscher, J. Arikkath, delta-Catenin Regulates Spine 75

Architecture via Cadherin and PDZ-dependent Interactions, J. Biol. Chem. 290

(2015) 10947–10957. doi:10.1074/jbc.M114.632679 [doi].

[29] H. Kim, J.R. Han, J. Park, M. Oh, S.E. James, S. Chang, Q. Lu, K.Y. Lee, H. Ki,

W.J. Song, K. Kim, Delta-catenin-induced dendritic morphogenesis. An essential

role of p190RhoGEF interaction through Akt1-mediated phosphorylation, J. Biol.

Chem. 283 (2008) 977–987. doi:M707158200 [pii].

[30] D. Edbauer, D. Cheng, M.N. Batterton, C.F. Wang, D.M. Duong, M.B. Yaffe, J.

Peng, M. Sheng, Identification and characterization of neuronal mitogen-activated

protein kinase substrates using a specific phosphomotif antibody, Mol. Cell.

Proteomics. 8 (2009) 681–695. doi:10.1074/mcp.M800233-MCP200 [doi].

[31] K. Abu-Elneel, T. Ochiishi, M. Medina, M. Remedi, L. Gastaldi, A. Caceres, K.S.

Kosik, A delta-catenin signaling pathway leading to dendritic protrusions, J. Biol.

Chem. 283 (2008) 32781–32791. doi:10.1074/jbc.M804688200 [doi].

[32] T.N. Turner, K. Sharma, E.C. Oh, Y.P. Liu, R.L. Collins, M.X. Sosa, D.R. Auer, H.

Brand, S.J. Sanders, D. Moreno-De-Luca, V. Pihur, T. Plona, K. Pike, D.R.

Soppet, M.W. Smith, S.W. Cheung, C.L. Martin, M.W. State, M.E. Talkowski, E.

Cook, R. Huganir, N. Katsanis, A. Chakravarti, Loss of delta-catenin function in

severe autism, Nature. 520 (2015) 51–56. doi:10.1038/nature14186 [doi].

[33] J. Arikkath, I.F. Peng, Y.G. Ng, I. Israely, X. Liu, E.M. Ullian, L.F. Reichardt, Delta-

catenin regulates spine and synapse morphogenesis and function in hippocampal

neurons during development, J. Neurosci. 29 (2009) 5435–5442.

doi:10.1523/JNEUROSCI.0835-09.2009 [doi].

[34] G.S. Brigidi, Y. Sun, D. Beccano-Kelly, K. Pitman, M. Mobasser, S.L. Borgland,

A.J. Milnerwood, S.X. Bamji, Palmitoylation of delta-catenin by DHHC5 mediates 76

activity-induced synapse plasticity, Nat. Neurosci. 17 (2014) 522–532.

doi:10.1038/nn.3657 [doi].

[35] J.B. Silverman, S. Restituito, W. Lu, L. Lee-Edwards, L. Khatri, E.B. Ziff, Synaptic

anchorage of AMPA receptors by cadherins through neural plakophilin-related

arm protein AMPA receptor-binding protein complexes, J. Neurosci. 27 (2007)

8505–8516. doi:27/32/8505 [pii].

[36] T. Ochiishi, K. Futai, K. Okamoto, K. Kameyama, K.S. Kosik, Regulation of AMPA

receptor trafficking by δ-catenin, Mol. Cell. Neurosci. 39 (2008) 499–507.

doi:10.1016/j.mcn.2008.06.002.

[37] C.P. Poore, J.R. Sundaram, T.K. Pareek, A. Fu, N. Amin, N.E. Mohamed, Y.L.

Zheng, A.X. Goh, M.K. Lai, N.Y. Ip, H.C. Pant, S. Kesavapany, Cdk5-mediated

phosphorylation of delta-catenin regulates its localization and GluR2-mediated

synaptic activity, J. Neurosci. 30 (2010) 8457–8467.

doi:10.1523/JNEUROSCI.6062-09.2010 [doi].

[38] N. Ide, Y. Hata, M. Deguchi, K. Hirao, I. Yao, Y. Takai, Interaction of S-SCAM with

Neural Plakophilin-RelatedArmadillo-Repeat Protein/δ-Catenin, Biochem.

Biophys. Res. Commun. 256 (1999) 456–461. doi:10.1006/bbrc.1999.0364.

[39] G.A. Yudowski, O. Olsen, H. Adesnik, K.W. Marek, D.S. Bredt, Acute Inactivation

of PSD-95 Destabilizes AMPA Receptors at Hippocampal Synapses, PLoS One.

8 (2013) e53965. doi:10.1371/journal.pone.0053965.

[40] E. Danielson, N. Zhang, J. Metallo, K. Kaleka, S.M. Shin, N. Gerges, S.H. Lee, S-

SCAM/MAGI-2 Is an Essential Synaptic Scaffolding Molecule for the GluA2-

Containing Maintenance Pool of AMPA Receptors, J. Neurosci. 32 (2012) 6967–

6980. doi:10.1523/JNEUROSCI.0025-12.2012. 77

[41] I. Israely, R.M. Costa, C.W. Xie, A.J. Silva, K.S. Kosik, X. Liu, Deletion of the

neuron-specific protein delta-catenin leads to severe cognitive and synaptic

dysfunction, Curr. Biol. 14 (2004) 1657–1663. doi:10.1016/j.cub.2004.08.065 [doi].

[42] P. Cerruti Mainardi, Cri du Chat syndrome., Orphanet J. Rare Dis. 1 (2006) 33.

doi:10.1186/1750-1172-1-33.

[43] M. Medina, R.C. Marinescu, J. Overhauser, K.S. Kosik, Hemizygosity of delta-

catenin (CTNND2) is associated with severe mental retardation in cri-du-chat

syndrome, Genomics. 63 (2000) 157–164. doi:10.1006/geno.1999.6090 [doi].

[44] R. Asadollahi, B. Oneda, P. Joset, S. Azzarello-Burri, D. Bartholdi, K. Steindl, M.

Vincent, J. Cobilanschi, H. Sticht, R. Baldinger, R. Reissmann, I. Sudholt, C.T.

Thiel, A.B. Ekici, A. Reis, E.K. Bijlsma, J. Andrieux, A. Dieux, D. FitzPatrick, S.

Ritter, A. Baumer, B. Latal, B. Plecko, O.G. Jenni, A. Rauch, The clinical

significance of small copy number variants in neurodevelopmental disorders, J.

Med. Genet. 51 (2014) 677–688. doi:10.1136/jmedgenet-2014-102588.

[45] C. Belcaro, S. Dipresa, G. Morini, V. Pecile, A. Skabar, A. Fabretto, CTNND2

deletion and intellectual disability, Gene. 565 (2015) 146–149.

doi:10.1016/j.gene.2015.03.054.

[46] J.M. Sardina, A.R. Walters, K.E. Singh, R.X. Owen, V.E. Kimonis, Amelioration of

the typical cognitive phenotype in a patient with the 5pter deletion associated with

Cri-du-chat syndrome in addition to a partial duplication of CTNND2, Am. J. Med.

Genet. Part A. 164 (2014) 1761–1764. doi:10.1002/ajmg.a.36494.

[47] B. Zhang, M. Willing, D.K. Grange, M. Shinawi, L. Manwaring, M. Vineyard, S.

Kulkarni, C.E. Cottrell, Multigenerational autosomal dominant inheritance of 5p

chromosomal deletions., Am. J. Med. Genet. A. 170 (2016) 583–93. 78

doi:10.1002/ajmg.a.37445.

[48] G. McMichael, S. Girirajan, A. Moreno-De-Luca, J. Gecz, C. Shard, L.S. Nguyen,

J. Nicholl, C. Gibson, E. Haan, E. Eichler, C.L. Martin, A. MacLennan, Rare copy

number variation in cerebral palsy., Eur. J. Hum. Genet. 22 (2014) 40–5.

doi:10.1038/ejhg.2013.93.

[49] T. Vrijenhoek, J.E. Buizer-Voskamp, I. van der Stelt, E. Strengman, C. Genetic

Risk and Outcome in Psychosis (GROUP) Consortium, C. Sabatti, A. Geurts van

Kessel, H.G. Brunner, R.A. Ophoff, J.A. Veltman, Recurrent CNVs disrupt three

candidate genes in schizophrenia patients., Am. J. Hum. Genet. 83 (2008) 504–

10. doi:10.1016/j.ajhg.2008.09.011.

[50] M.G. Nivard, H. Mbarek, J.J. Hottenga, J.H. Smit, R. Jansen, B.W. Penninx, C.M.

Middeldorp, D.I. Boomsma, Further confirmation of the association between

anxiety and CTNND2: replication in humans., Genes. Brain. Behav. 13 (2014)

195–201. doi:10.1111/gbb.12095.

[51] R. Paffenholz, C. Kuhn, C. Grund, S. Stehr, W.W. Franke, The Arm-Repeat

Protein NPRAP (Neurojungin) Is a Constituent of the Plaques of the Outer

Limiting Zone in the Retina, Defining a Novel Type of Adhering Junction, Exp. Cell

Res. 250 (1999) 452–464. doi:10.1006/excr.1999.4534.

[52] G. Jun, J.A. Moncaster, C. Koutras, S. Seshadri, J. Buros, A.C. McKee, G.

Levesque, P.A. Wolf, P. St George-Hyslop, L.E. Goldstein, L.A. Farrer, δ-Catenin

is genetically and biologically associated with cortical cataract and future

Alzheimer-related structural and functional brain changes., PLoS One. 7 (2012)

e43728. doi:10.1371/journal.pone.0043728.

[53] G.M. Beaudoin 3rd, S.H. Lee, D. Singh, Y. Yuan, Y.G. Ng, L.F. Reichardt, J. 79

Arikkath, Culturing pyramidal neurons from the early postnatal mouse

hippocampus and cortex, Nat. Protoc. 7 (2012) 1741–1754.

doi:10.1038/nprot.2012.099; 10.1038/nprot.2012.099.

[54] M. Valcu, C.-M. Valcu, Data transformation practices in biomedical sciences, Nat.

Publ. Gr. 8 (2011). doi:10.1038/nmeth0211-104.

[55] P.S. Vosler, C.S. Brennan, J. Chen, Calpain-mediated signaling mechanisms in

neuronal injury and neurodegeneration., Mol. Neurobiol. 38 (2008) 78–100.

doi:10.1007/s12035-008-8036-x.

[56] Y. Ono, T.C. Saido, H. Sorimachi, Calpain research for drug discovery: challenges

and potential, Nat. Publ. Gr. 15 (2016). doi:10.1038/nrd.2016.212.

[57] L. Yuan, J. Arikkath, Functional roles of p120ctn family of proteins in central

neurons, Semin. Cell Dev. Biol. 69 (2017) 70–82.

doi:10.1016/j.semcdb.2017.05.027.

[58] P.M. Haggie, A.S. Verkman, Unimpaired Lysosomal Acidification in Respiratory

Epithelial Cells in Cystic Fibrosis, J. Biol. Chem. 284 (2009) 7681–7686.

doi:10.1074/JBC.M809161200.

[59] C.M. Oslowski, F. Urano, Measuring ER stress and the unfolded protein response

using mammalian tissue culture system., Methods Enzymol. 490 (2011) 71–92.

doi:10.1016/B978-0-12-385114-7.00004-0.

[60] M. Seremwe, R.G. Schnellmann, W.B. Bollag, Calpain-10 Activity Underlies

Angiotensin II-Induced Aldosterone Production in an Adrenal Glomerulosa Cell

Model., Endocrinology. 156 (2015) 2138–49. doi:10.1210/en.2014-1866.

[61] M.A. Smith, R.G. Schnellmann, Calpains, mitochondria, and apoptosis.,

Cardiovasc. Res. 96 (2012) 32–7. doi:10.1093/cvr/cvs163. 80

[62] H. Hering, M. Sheng, DENDRITIC SPINES: STRUCTURE, DYNAMICS AND

REGULATION, Nat. Rev. Neurosci. 2 (2001) 880–888. doi:10.1038/35104061.

[63] K.M. Harris, Structure, development, and plasticity of dendritic spines., Curr. Opin.

Neurobiol. 9 (1999) 343–8. http://www.ncbi.nlm.nih.gov/pubmed/10395574

(accessed September 15, 2017).

[64] E.S. Lein, M.J. Hawrylycz, N. Ao, M. Ayres, A. Bensinger, A. Bernard, A.F. Boe,

M.S. Boguski, K.S. Brockway, E.J. Byrnes, L. Chen, L. Chen, T.-M. Chen, M. Chi

Chin, J. Chong, B.E. Crook, A. Czaplinska, C.N. Dang, S. Datta, N.R. Dee, A.L.

Desaki, T. Desta, E. Diep, T.A. Dolbeare, M.J. Donelan, H.-W. Dong, J.G.

Dougherty, B.J. Duncan, A.J. Ebbert, G. Eichele, L.K. Estin, C. Faber, B.A. Facer,

R. Fields, S.R. Fischer, T.P. Fliss, C. Frensley, S.N. Gates, K.J. Glattfelder, K.R.

Halverson, M.R. Hart, J.G. Hohmann, M.P. Howell, D.P. Jeung, R.A. Johnson,

P.T. Karr, R. Kawal, J.M. Kidney, R.H. Knapik, C.L. Kuan, J.H. Lake, A.R.

Laramee, K.D. Larsen, C. Lau, T.A. Lemon, A.J. Liang, Y. Liu, L.T. Luong, J.

Michaels, J.J. Morgan, R.J. Morgan, M.T. Mortrud, N.F. Mosqueda, L.L. Ng, R.

Ng, G.J. Orta, C.C. Overly, T.H. Pak, S.E. Parry, S.D. Pathak, O.C. Pearson, R.B.

Puchalski, Z.L. Riley, H.R. Rockett, S.A. Rowland, J.J. Royall, M.J. Ruiz, N.R.

Sarno, K. Schaffnit, N. V. Shapovalova, T. Sivisay, C.R. Slaughterbeck, S.C.

Smith, K.A. Smith, B.I. Smith, A.J. Sodt, N.N. Stewart, K.-R. Stumpf, S.M. Sunkin,

M. Sutram, A. Tam, C.D. Teemer, C. Thaller, C.L. Thompson, L.R. Varnam, A.

Visel, R.M. Whitlock, P.E. Wohnoutka, C.K. Wolkey, V.Y. Wong, M. Wood, M.B.

Yaylaoglu, R.C. Young, B.L. Youngstrom, X. Feng Yuan, B. Zhang, T.A.

Zwingman, A.R. Jones, Genome-wide atlas of gene expression in the adult

mouse brain, Nature. 445 (2007) 168–176. doi:10.1038/nature05453.

[65] Allen Institute for Brain Science, Allen Brain Atlas Data Portal: Mouse Brain, 81

(n.d.). http://mouse.brain-map.org/ (accessed September 1, 2017).

[66] X. Zhao, E. Lein, A. He, S. Smith, C. Aston, F. Gage, H. Phatnani, P. Guarnieri, C.

Caneda, N. Ruderisch, S. Deng, S. Liddelow, C. Zhang, R. Daneman, T. Maniatis,

B. Barres, J. Wu, C. Dang, S. Datta, N. Dee, A. Desaki, T. Desta, E. Diep, T.

Dolbeare, M. Donelan, H. Dong, J. Dougherty, B. Duncan, A. Ebbert, G. Eichele,

L. Estin, C. Faber, B. Facer, R. Fields, S. Fischer, T. Fliss, C. Frensley, S. Gates,

K. Glattfelder, K. Halverson, M. Hart, J. Hohmann, M. Howell, D. Jeung, R.

Johnson, P. Karr, R. Kawal, J. Kidney, R. Knapik, C. Kuan, J. Lake, A. Laramee,

K. Larsen, C. Lau, T. Lemon, A. Liang, Y. Liu, L. Luong, J. Michaels, J. Morgan,

R. Morgan, M. Mortrud, N. Mosqueda, L. Ng, R. Ng, G. Orta, C. Overly, T. Pak, S.

Parry, S. Pathak, O. Pearson, R. Puchalski, Z. Riley, H. Rockett, S. Rowland, J.

Royall, M. Ruiz, N. Sarno, K. Schaffnit, N. Shapovalova, T. Sivisay, C.

Slaughterbeck, S. Smith, K. Smith, B. Smith, A. Sodt, N. Stewart, K. Stumpf, S.

Sunkin, M. Sutram, A. Tam, C. Teemer, C. Thaller, C. Thompson, L. Varnam, A.

Visel, R. Whitlock, P. Wohnoutka, C. Wolkey, V. Wong, M. Wood, M. Yaylaoglu,

R. Young, B. Youngstrom, X. Yuan, B. Zhang, T. Zwingman, A. Jones, Hipposeq:

a comprehensive RNA-seq database of gene expression in hippocampal principal

neurons, J. Comp. Neurol. 441 (2016) 187–196–196. doi:10.1002/cne.1406.

[67] R. Nath, K.J. Raser, D. Stafford, I. Hajimohammadreza, A. Posner, H. Allen, R. V

Talanian, P. Yuen, R.B. Gilbertsen, K.K. Wang, Non-erythroid alpha-spectrin

breakdown by calpain and interleukin 1 beta-converting-enzyme-like protease(s)

in apoptotic cells: contributory roles of both protease families in neuronal

apoptosis., Biochem. J. 319 ( Pt 3) (1996) 683–90.

http://www.ncbi.nlm.nih.gov/pubmed/8920967 (accessed September 14, 2017).

[68] N.O. Markham, C.A. Doll, M.R. Dohn, R.K. Miller, H. Yu, R.J. Coffey, P.D. 82

McCrea, J.T. Gamse, A.B. Reynolds, DIPA-family coiled-coils bind conserved

isoform-specific head domain of p120-catenin family: potential roles in

hydrocephalus and heterotopia, Mol. Biol. Cell. 25 (2014) 2592–2603.

doi:10.1091/mbc.E13-08-0492 [doi].

[69] E.C. Ferber, M. Kajita, A. Wadlow, L. Tobiansky, C. Niessen, H. Ariga, J. Daniel,

Y. Fujita, A role for the cleaved cytoplasmic domain of E-cadherin in the nucleus.,

J. Biol. Chem. 283 (2008) 12691–700. doi:10.1074/jbc.M708887200.

[70] U. Steinhusen, V. Badock, A. Bauer, J. Behrens, B. Wittman-Liebold, B. Dörken,

K. Bommert, Apoptosis-induced Cleavage of β-Catenin by Caspase-3 Results in

Proteolytic Fragments with Reduced Transactivation Potential, J. Biol. Chem. 275

(2000) 16345–16353. doi:10.1074/jbc.M001458200.

[71] D. Gu, N.K. Tonthat, M. Lee, H. Ji, K.P. Bhat, F. Hollingsworth, K.D. Aldape, M.A.

Schumacher, T.P. Zwaka, P.D. McCrea, Caspase-3 cleavage links delta-catenin

to the novel nuclear protein ZIFCAT., J. Biol. Chem. 286 (2011) 23178–88.

doi:10.1074/jbc.M110.167544.

[72] M. Lee, Y.T. Kwon, M. Li, J. Peng, R.M. Friedlander, L.-H. Tsai, Neurotoxicity

induces cleavage of p35 to p25 by calpain, Nature. 405 (2000) 360–364.

doi:10.1038/35012636.

[73] M. Papa, M. Bundman, V. Greenberger, M. Segal, Morphological analysis of

dendritic spine development in primary cultures of hippocampal neurons, J.

Neurosci. 15 (1995). http://www.jneurosci.org/content/15/1/1.short (accessed

September 15, 2017).

[74] M. Berg, T. Bruhn, F.F. Johansen, N.H. Diemer, Kainic acid-induced seizures and

brain damage in the rat: different effects of NMDA- and AMPA receptor 83

antagonists., Pharmacol. Toxicol. 73 (1993) 262–8.

http://www.ncbi.nlm.nih.gov/pubmed/8115308 (accessed September 14, 2017).

[75] Y. Wang, Z. Qin, Molecular and cellular mechanisms of excitotoxic neuronal

death, Apoptosis. 15 (2010) 1382–1402. doi:10.1007/s10495-010-0481-0.

[76] Z. Padamsey, L. McGuinness, S.J. Bardo, M. Reinhart, R. Tong, A. Hedegaard,

M.L. Hart, N.J. Emptage, Activity-Dependent Exocytosis of Lysosomes Regulates

the Structural Plasticity of Dendritic Spines., Neuron. 93 (2017) 132–146.

doi:10.1016/j.neuron.2016.11.013.

[77] M.S. Goo, L. Sancho, N. Slepak, D. Boassa, T.J. Deerinck, M.H. Ellisman, B.L.

Bloodgood, G.N. Patrick, Activity-dependent trafficking of lysosomes in dendrites

and dendritic spines, J. Cell Biol. (2017).

http://jcb.rupress.org/content/early/2017/06/16/jcb.201704068 (accessed

September 14, 2017).