The Pennsylvania State University
The Graduate School
Intercollege Graduate Program in Genetics
GENETIC ANALYSIS OF SYNAPTIC TRANSMISSION
A Dissertation in
Genetics
by
Janani Iyer
2012 Janani Iyer
Submitted in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
August 2012
The dissertation of Janani Iyer was reviewed and approved* by the following:
Zhi-Chun Lai Professor of Biology, Biochemistry and Molecular Biology Chair of Committee
Wendy-Hanna Rose Associate Professor of Biochemistry and Molecular Biology
Melissa Rolls Assistant Professor Biochemistry and Molecular Biology
Richard Ordway Professor of Molecular Neuroscience and Genetics Dissertation Co-Advisor
Fumiko Kawasaki Assistant Professor of Biology Dissertation Co-Advisor
Robert Paulson Professor of Veterinary and Biomedical Sciences Genetics Graduate Program Chair
*Signatures are on file in the Graduate School.
iii
ABSTRACT
The transmission of electrical impulses at chemical synapses is fundamental to neural
function. The elucidation of in vivo molecular mechanisms involved in synaptic transmission has
been a major research objective in neuroscience. One important approach to achieve this goal is
genetic analysis in Drosophila melanogaster. The synaptic mechanisms of Drosophila are similar
to those in vertebrates and, with the advantage of performing in vivo functional analysis of native
synapses using genetic, molecular, biochemical, electrophysiological and ultrastructural
approaches, Drosophila is a powerful model system. Analysis of temperature sensitive (TS)
paralytic mutants of Drosophila has played an important role in elucidating the in vivo molecular
mechanisms of synaptic transmission. TS paralytic mutants may allow normal development and
function at permissive temperatures and reveal the physiological role of a specific gene product
following its acute perturbation at restrictive temperatures. Forward genetic screens for TS
paralytic mutants and screens for modifiers of existing TS mutants have revealed novel
mechanisms in synaptic transmission. The present work discusses the functional and molecular
characterization of two proteins participating in synaptic transmission, DISABLED and
COMPLEXIN, in the Drosophila model system.
Members of the DISABLED (DAB) family of proteins are known to play a conserved
role in endocytic trafficking of cell surface receptors by functioning as monomeric CLATHRIN
Associated Sorting Proteins (CLASPs) which recruit cargo proteins into endocytic vesicles. The present study reports analysis of a Drosophila disabled mutant, dabEC1 (enhancer of cac1) (Lisa
Posey, Masters Thesis, Penn State University, 2008), which was recovered as an enhancer of the presynaptic TS calcium channel mutant cacTS2. Genetic and functional characterization of dabEC1
has revealed a novel role for DAB proteins in chemical synaptic transmission. dabEC1 exhibits impaired synaptic function including a rapid, activity-dependent reduction in neurotransmitter
iv
release and disruption of synaptic vesicle endocytosis. In presynaptic boutons, Drosophila DAB
(dDAB) and CLATHRIN were highly co-localized within two distinct classes of puncta,
including relatively dim puncta which were located at active zones (AZs) which may reflect
endocytic mechanisms operating at neurotransmitter release sites. Finally, broader analysis of
endocytic proteins including DYNAMIN supported a novel role for CLATHRIN-mediated
endocytic mechanisms in rapid clearance of neurotransmitter release sites for subsequent vesicle
priming and refilling of the release-ready vesicle pool.
The current study of neurotransmitter release mechanisms also provides new insights into
the role of COMPLEXIN (CPX) proteins in synaptic transmission. CPX proteins are known to
interact with the Soluble NSF Attachment Protein REceptors (SNARE) proteins at the core of the
synaptic vesicle fusion machinery and regulate neurotransmitter release. Several studies,
including cpx knockout and knockdown studies in mouse, genetic deletion of cpx in Drosophila
and in vitro fusion assays, have demonstrated a role for CPX in supporting evoked
neurotransmitter release and an inhibitory or clamping role in suppressing spontaneous synaptic
vesicle fusion. However, the in vivo mechanisms of CPX function in neurotransmitter release
remain controversial. To further investigate these mechanisms, we have conducted a genetic
screen to obtain new mutants within the Drosophila cpx gene. Here we report recovery and
analysis of a new cpx mutant in Drosophila, cpx1257, revealing spatially defined and separable pools of CPX which make distinct contributions to its activation and clamping functions.
The mutation in cpx1257 deletes only the last C-terminal amino acid of CPX within the
well conserved CAAX motif for prenylation, a post-translational lipid modification implicated in
membrane targeting of CPX and other cytosolic proteins. Our immunocytochemical studies have
revealed that CPX is highly enriched at active zone (AZ) regions of the presynaptic plasma
membrane containing neurotransmitter release sites and is also detected in presynaptic plasma
membrane compartments outside of the AZ, including synaptic vesicles. Biochemical studies
v
confirmed membrane association of CPX and robust interactions of CPX with all the three
SNARE proteins. In contrast, at cpx1257 mutant synapses, AZ localization of CPX persists but there is selective loss of broader presynaptic membrane localization of CPX and, surprisingly, the
bulk of the CPX-SNARE protein interactions are also abolished in this mutant.
Functional analysis of cpx1257 began with electrophysiological studies at adult neuromuscular synapses in a previously reported cpx null mutant as a basis for comparison. These experiments demonstrated severe reduction in the activation of evoked neurotransmitter release, as indicated by a marked reduction in the amplitude of the Excitatory PostSynaptic Current
(EPSC). This phenotype is accompanied by a marked increase in the frequency of spontaneous fusion events, reflecting a loss of the CPX clamping function. Further, the null mutant exhibited an altered EPSC waveform, observed as a slowing of the EPSC rise and decay times with respect to wild-type. Interestingly, this is a presynaptic phenotype and thus indicates a role for CPX in determining the kinetics of neurotransmitter release. In contrast to the cpx null, the cpx1257 mutant exhibited a wild-type EPSC amplitude and waveform, demonstrating that cpx1257 retains the CPX
activation function in evoked neurotransmitter release. However, cpx1257 exhibited a selective loss
of clamping function as indicated by a marked increase in the frequency of spontaneous
neurotransmitter release. Together with the preceding immunocytochemical and biochemical
analysis, these findings indicate that spatially distinct and separable interactions of CPX with
presynaptic membranes and SNARE proteins appear to mediate separable activation and
clamping functions of CPX in neurotransmitter release.
We have used Drosophila melanogaster as a model system to investigate the in vivo
mechanisms of synaptic transmission. Our analysis of DISABLED and COMPLEXIN function in
neurotransmitter release advances our understanding of synaptic function and illustrates the
power of forward genetic analysis in defining the molecular basis of physiological processes.
vi
TABLE OF CONTENTS
LIST OF FIGURES ...... viii
LIST OF TABLES ...... xi
ACKNOWLEDGEMENTS ...... xii
Chapter 1 Introduction ...... 1
1.1 Chemical Synaptic Transmission ...... 1 1.1.1 Synapses and Synaptic Transmission ...... 1 1.1.2 Synaptic Vesicle Membrane Trafficking Cycle...... 6 1.1.3 Synaptic Vesicle Pools ...... 9 1.1.4 Drosophila melanogaster: A genetic model system for the study of synaptic transmission ...... 12 1.2 Molecular Mechanisms of Synaptic Vesicle Exocytosis ...... 14 1.2.1 Membrane fusion ...... 14 1.2.2 The Soluble N-ethylmaleimide Sensitive Factor (NSF) Attachment proteins Receptors (SNAREs) ...... 16 1.2.3 NSF and Soluble NSF Attachment Protein (SNAP) ...... 22 1.2.4 Presynaptic voltage gated calcium channel ...... 23 1.2.5 Regulators of Synaptic Vesicle Exocytosis ...... 24 1.2.5.1 Synaptotagmin ...... 24 1.2.5.2 Complexin ...... 25 1.3 Synaptic Vesicle Endocytosis ...... 39 1.3.1 Clathrin Mediated Endocytosis...... 40 1.3.2 DISABLED (DAB) ...... 46
Chapter 2 Materials and Methods ...... 49
2.1 Drosophila Stocks ...... 49 2.2 Behavioral Analysis ...... 50 2.3 Molecular Biology ...... 50 2.3.1 Generation of DNA constructs carrying UAS-EGFP-cpx and UAS-EGFP- cpx1257 ...... 50 2.3.2 Generation of DNA construct carrying dab-EGFP...... 51 2.3.3 Generation of transgenic strains ...... 57 2.3.4 Identification of molecular lesions in dap160 mutants ...... 59 2.3.5 Identification of the molecular lesions in syt1 mutant ...... 59 2.4 Biochemistry ...... 61 2.4.1 Antibodies ...... 61 2.4.2 Western Blotting ...... 62 2.4.3 Co-Immunoprecipitation ...... 65 2.4.4 Immunocytochemistry ...... 66 2.4.5 Imaging using a Confocal Microscope ...... 67
vii
2.4.6 Phase Partitioning ...... 67
Chapter 3 DISABLED Functions in CLATHRIN-Mediated Synaptic Vesicle Endocytosis and Exo-Endocytic Coupling at the Active Zone ...... 70
3.1 Abstract ...... 70 3.2 Introduction ...... 71 3.3 Results ...... 72 3.4 Discussion ...... 79 3.5 Figures ...... 86 3.5 Supplementary Information ...... 93
Chapter 4 A new Drosophila Mutant Permits Separation and Molecular Dissection of the COMPLEXIN Fusion Clamp Function in Synaptic Exocytosis ...... 104
4.1 Abstract ...... 105 4.2 Introduction ...... 106 4.3 Materials and Methods ...... 107 4.4 Results ...... 112 4.5 Discussion ...... 118 4.6 Figures ...... 122 4.7 Supplementary Figures...... 130
Chapter 5 Novel dap160 mutant alleles in Drosophila melanogaster ...... 104
5.1 Introduction ...... 135 5.2 Results ...... 136 5.3 Discussion ...... 137 5.4 Figures ...... 139
Chapter 6 Discussion ...... 104
6.1 DISABLED (DAB) in Synaptic Transmission ...... 142 6.2 COMPLEXIN in Synaptic Transmission ...... 147
Appendix ...... 154
Bibliography ...... 161
viii
LIST OF FIGURES
Figure 1-1: Illustration of Motor neuron and Chemical synapse...... 4
Figure 1-2: Two modes of neurotransmitter release...... 5
Figure 1-3: Mechanisms of Synaptic Vesicle Trafficking ...... 7
Figure 1-4: Core proteins involved in Synaptic Vesicle trafficking...... 10
Figure 1-5: Synaptic Vesicle Pools ...... 11
Figure 1-6: Model for Membrane fusion...... 15
Figure 1-7: Domain structure of the three neuronal SNARE proteins involved in synaptic vesicle fusion...... 19
Figure 1-8: Model of the neuronal SNAREs assembled into the SNARE core complex...... 20
Figure 1-9: Structure of the SNARE core complex...... 21
Figure 1-10: Complexin Protein family ...... 27
Figure 1-11: Schematic of the protein farnesylation mechanism...... 30
Figure 1-12: 12 The Complexin-SNARE complex...... 32
Figure 1-13: Complexin domain structure...... 37
Figure 1-14: Proposed model for CPX function ...... 38
Figure 1-15: Endocytic Adaptor Protein Interactions...... 42
Figure 1-16: Stages of clathrin mediated endocytosis (CME)...... 45
Figure 2-1: dab ORF...... 55
Figure 2-2: Schematic for establishing transgenic lines (TG) ...... 58
Figure 2-3: Overview of Western Blotting...... 64
Figure 2-4: Phase partitioning...... 69
Figure 3-1: A Drosophila disabled mutant...... 86
ix
Figure 3-2: dDAB functions in synaptic transmission and exhibits a close spatial and functional relationship with CLATHRIN...... 87
Figure 3-3: Ultrastructural analysis reveals disruption of synaptic vesicle endocytosis and persistence of active zone-associated and docked vesicles at dab mutant and CLATHRIN RNAi synapses...... 89
Figure 3-4: A rapid functional requirement for CLATHRIN-mediated endocytic mechanisms may reflect impaired refilling of the release-ready vesicle pool...... 90
Figure 3-5: A working model for CLATHRIN-mediated endocytic mechanisms in rapid clearance of neurotransmitter release sites...... 92
Figure 3-S1: A new dab mutation recovered as an enhancer of the cacTS2 temperature- sensitive paralytic phenotype...... 98
Figure 3-S2: Characteristics of the dab synaptic phenotype...... 99
Figure 3-S3: A newly generated rabbit anti-dDAB antibody recognizes dDAB...... 99
Figure 3-S4: Presynaptic distribution of endogenous dDAB and CHC-EGFP...... 100
Figure 3-S5: Presynaptic distribution of endocytic proteins including DYNAMIN, DAP160 and α-ADAPTIN...... 101
Figure 3-S6: The presynaptic distribution of dSYT1 and other synaptic vesicle proteins appears to be preserved at dab mutant synapses...... 102
Figure 3-S7: Recovery in PPD at DLM neuromuscular synapses from dab, CHC-RNAi and shiTS1 ...... 103
Figure 3-S8: A dap160 mutant exhibits an activity-dependent decrease in excitatory postsynaptic currents (EPSCs) resembling that of other endocytic mutants...... 103
Figure 3-S9: Non-CLATHRIN-coated membrane invaginations (omega structures) occur at the AZ...... 104
Figure 4-1: A New cpx mutant, cpx1257...... 122
Figure 4-2: Electrophysiological analysis of cpx mutant phenotypes at DLM neuromuscular synapses...... 123
Figure 4-3: Presynaptic localization of wild-type CPX and mutant CPX1257 at DLM neuromuscular synapses...... 125
Figure 4-4: CPX membrane association is lost in the CPX1257 mutant which lacks the CaaX farnesylation motif...... 127
Figure 4-5: Interaction of CPX with SNARE proteins...... 128
x
Figure 4-6: Selective loss of CPX from presynaptic membrane compartments in the cpx1257 mutant suggests that membrane association promotes CPX function in clamping spontaneous release ...... 129
Figure 4-S1: A genetic screen to isolate new complexin mutants...... 130
Figure 4-S2: CPX isoforms contain distinct C-terminal domains...... 131
Figure 4-S3: Electrophysiogical analysis of cpx mutant phenotypes at larval neuromuscular synapses ...... 132
Figure 4-S4: Endogenous CPX membrane association is lost in the CPX1257 mutant lacking the CaaX farnesylation motif...... 133
Figure 4-S5: Presynaptic localization of CPX isoform E at DLM neuromuscular synapses...... 134
Figure 5-1: Genetic screen identifies two new dap160 mutants...... 139
Figure 5-2: New dap160 mutants...... 140
Figure 5-3: Western Blot analysis of dap160 mutants...... 141
Figure A-1: Genetic screen identifies a new syt1 mutant ...... 158
Figure A-2: New syt1 mutant ...... 159
Figure A-3: Western Blot analysis of syt1 mutant ...... 160
xi
LIST OF TABLES
Table 2-1: Primers for generation of EGFP tagged cpx DNA constructs ...... 54
Table 2-2: Primers for generation of dab-EGFP DNA construct...... 55
Table 2-3: Primers for sequencing of dab ...... 56
Table 2-4: Primers for identification of dap160 mutations ...... 60
Table 2-5: Primers for identification of the syt1 mutation...... 60
Table 3-S1: Summary of Ultrastructural Analysis...... 104
xii
ACKNOWLEDGEMENTS
With great pleasure I reminisce the time spent in my graduate life and the countless people who have helped in my research all through these years. Though I owe a lot to everyone,
here I put to record a few of them.
First and foremost, I would like to thank Dr. Rick Ordway, my advisor, whom I owe the
biggest debt of gratitude for all his support and encouragement. Rick, I want to thank you for
allowing me to be part of your lab at a time when I was going through a very rough patch. You
took a chance with me and I am very grateful for that. You have been instrumental in my
development and growth as a graduate student. Your valuable guidance, constant encouragement,
constructive criticism and support has made my Ph.D experience productive and stimulating.
I would like to thank Dr. Fumiko Kawasaki, my co-advisor. Fumi, I have always admired and tried to emulate your great attention to detail and ability to diligently execute experiments. I would like to thank you for being so patient with me as I’ve progressed, learning from my mistakes along the way. The joy and enthusiasm that you have for research is contagious and has been motivational for me, especially during the tough periods. I am thankful for the excellent example you have provided me as a successful woman scientist.
I am indebted to both Rick and Fumi for their tremendous impact on my graduate education. Their guidance and friendship have been invaluable on both an academic and a personal level, for which I am extremely grateful. This thesis would not have been possible without their help, support, optimism and patience.
xiii
The members of our lab, past and present have contributed immensely to my personal
and academic growth at Penn State. The group has been a source of friendships as well as good
advice. I would like to acknowledge Huaru Yan, a great friend and an ideal teacher. She helped
me settle into the lab, taught me fly genetics and was very generous in her time. I have learnt
from Yan how to be optimistic about research even when things don’t go your way. Another great
friend is Rie, who taught me to do adult dissections, by far the hardest thing I’ve had to do in lab.
I have tormented her so many times with questions about dissection and sought her help with bad
dissections, which she always magically fixed. She’s also helped me with several other things, be
it collecting 400 fly heads or helping me with Western Blotting or other experiments. Apart from
that she’s always patiently offered an ear to my frustrations and given me good advice. Thank
you Rie for being a great friend. I have also developed great friendship with Alex, an
undergraduate student. Alex is my go to person when I need help with English grammar and
pronunciations. Her theatrical skills have helped me in overcoming some of my public speaking
fright.
I would to take this opportunity in thanking all the past and current members of Ordway lab. Several other people have contributed directly or indirectly to this project. Lisa Posey, a former graduate student, was involved in the isolation and initial analysis of dab mutant. I took over the DAB project from Lisa and she was very generous with her time to explain me the details. Andrew Lutas, a former undergraduate student was also involved in the genetic screen in the DAB project. Emily and Sam, former undergraduate students did the ultrastructural analysis in the DAB project. Andrew, Emily and I became good friends and I have enjoyed working with them. Another former graduate student Wenhua taught me to do larval dissections. Chris, Giselle and Kerry worked very hard on the genetic screen in the COMPLEXIN project. It would not have been possible to complete this large screen without them. I would also like to acknowledge
xiv
Yunzhen for so graciously making fly food and for good conversations especially during the
weekends. I have also enjoyed working with and getting to know Rolfy, another excellent
undergraduate student.
I have had the privilege of working with Alex and Ximena on a number of molecular
biology projects. I couldn’t have asked for better people to work with. Ximena is always so
cheerful and very skilled in executing experiments. Alex is very smart, incredibly talented and
very efficient with her time. Alex has been my partner in crime for all the cupcake adventures. I
have spent wonderful time both working in lab and having fun outside lab with Alex, Kerry and
Ximena. They have enlightened me about the undergraduate experience at Penn State.
My time at Penn State was made enjoyable in large part due to the many friends that have
become a part of my life. I am grateful for our memorable trips, our get togethers and in general
the wonderful time spent with Kris, Meenakshi, Nitin, Megha, Sujana, Kuang, Kiran and
Daipayan.
Lastly, I would like to thank my family for all their love, support and encouragement. For
my mother who raised me with a love of science and supported me in all my pursuits. Thank you
amma for everything. For my dad who believed in my abilities and supported me no matter what.
For my brother, Jay, whose innate curiosity and interest in my research has always made me think
out the box. He has been a constant source of motivation, optimism and encouragement. I would also like to thank my parents-in-law for their support. And finally for my loving, encouraging and patient husband Ashu, whose support during the final stages of my Ph.D is greatly appreciated. I would not have been able to complete my thesis without his constant encouragement and support.
1 Chapter 1
Introduction
1.1. Chemical Synaptic Transmission
1.1.1. Synapses and Synaptic transmission
Neurons are specialized cells in the nervous system that are responsible for rapid
information processing. Humans have billions of neurons in the nervous system; the brain alone
has more than 100 billion. Neurons process information related to motor, sensory and cognitive
functions and transmit this information to other neurons and cells. They form neural circuits and
networks that account for complex processes such as learning, memory, motor functions, emotions etc. The ability of the neuron to transmit information stems from its unique and specialized structure.
A neuron is comprised of three major compartments: the dendrites, soma and axon
(Figure 1.1.A). Dendrites branch out and receive electrical signals from other neurons. The soma
or cell body, which houses common cellular organelles such as the nucleus and rough
endoplasmic reticulum, is the center of cell metabolism and integration of electrical signals. The
signals received by the dendrites are integrated in the soma and transmitted to other cells via the
axons. Transmission occurs at synapses, which are points of contact between the axon and the
target cell. Synaptic transmission is key to neuronal function and is regulated in different types of
synaptic plasticity. Synapses can be of two kinds: electrical and chemical. At electrical synapses,
the pre- and postsynaptic neurons are electrically connected through gap junctions. Electrical
neurotransmission is mediated by the passive flow of current through gap junction pores that
directly connect the cytoplasm of both neurons. In contrast, chemical synapses have no physical
cytoplasmic continuity, rather the surface membrane of the axon (presynaptic membrane) and the
surface membrane of the target cell (postsynaptic membrane) are separated by a small gap called
2 the synaptic cleft. At chemical synapses, electrical signals are transmitted via presynaptic release
of chemical neurotransmitters into the synaptic cleft. Neurotransmitters are stored within the
presynaptic axon terminal in small secretory vesicles (synaptic vesicles). Synaptic vesicles (SVs)
release neurotransmitters following their exocytic fusion with specialized regions of the pre-
synaptic membrane called Active Zones (AZ). By TEM (Transmission Electron Microscopy) AZs
maybe recognized as electron dense regions of the pre-synaptic plasma membrane directly
opposite to post-synaptic densities occurring at the postsynaptic membrane. AZs contain several proteins that are specialized for synaptic vesicle exocytosis, such as voltage gated calcium channels, vesicle fusion machinery (SNARE proteins) and scaffolding proteins (Sudhof 2004).
During chemical synaptic transmission, an action potential elicits neurotransmitter release at the AZs. These neurotransmitters then diffuse across the synaptic cleft and bind to receptors in the postsynaptic membrane (Figure 1.1.B). The postsynaptic receptors may either be ligand gated ion channels or metabotropic receptors linked to ion channels via a second messenger pathway.
The binding of neurotransmitter to its receptor activates the ion channels and results in excitation or inhibition of the post-synaptic membrane.
In addition to the evoked neurotransmission, in which multiple synaptic vesicles fuse synchronously with the presynaptic plasma membrane upon arrival of an action potential at the nerve terminal (Figure 1.2.A), there is another mode of neurotransmitter release called spontaneous neurotransmission. During spontaneous release there is no need for an action potential and each event corresponds to spontaneous fusion of a single synaptic vesicle with the presynaptic plasma membrane (Figure 1.2.B). These release events have smaller post synaptic current amplitudes and are referred to as mini EPSC (mEPSC). Spontaneous release was first observed in the 1950s at the frog neuromuscular junction (Fatt 1952) after which Katz and coworkers proposed the quantal hypothesis. According to the quantal hypothesis, neurotransmitter release occurs in discrete packages called “quanta”, which produce the minis
3 (Del Castillo 1954). Since their discovery by Katz and coworkers, the functional significance of minis has been debatable; for long they have been thought to represent background noise.
However, there is growing evidence that such spontaneous release events regulate maturation and stability of synaptic networks and play a role in neuronal signaling (Kavalali, Chung et al. 2011).
Furthermore, analyses of minis are useful in understanding the functional properties (both pre-
and post-synaptic) of individual synapses (Otsu and Murphy 2003; Kavalali, Chung et al. 2011).
Synaptic vesicles that store neurotransmitters are key to synaptic transmission.
Communication ensues between neurons by the release of neurotransmitters from vesicles
through a series of sequential steps, as described in the next section.
4
Figure 1.1 Illustration of Motor neuron and Chemical synapse. (A) Illustration of a motor neuron (Modified from Kandel et al., 2000). (B) Illustration of a chemical synapse (Modified from (Ordway 2004)).
5
Figure 1.2. Two modes of neurotransmitter release [Modified from (Kavalali, Chung et al. 2011)]. (A) Evoked neurotransmission, in which multiple synaptic vesicles fuse synchronously in response to action potential, to release the neurotransmitter. (B) Spontaneous neurotransmission, where each spontaneous release event involves spontaneous fusion of a single synaptic vesicle.
6 1.1.2. Synaptic Vesicle (SV) Membrane Trafficking Cycle
Extensive research on neurotransmitter release mechanisms has defined 4 functional steps in SV trafficking: docking, priming, fusion/exocytosis and recycling/endocytosis. SVs filled with neurotransmitter dock at the plasma membrane of the pre-synaptic terminal specifically at the AZ,
where they are primed for rapid fusion and neurotransmitter release (Figure 1.3) (Lin and Scheller
2000). The priming process involves a series of biochemical steps which prepare the docked
vesicles for rapid fusion with the pre-synaptic membrane upon calcium influx (Banerjee, Barry et
al. 1996). The arrival of an action potential at the pre-synaptic terminal leads to membrane
depolarization, which activates and opens the voltage-gated calcium channels (VGCC) located at
the AZ. Subsequent influx of calcium causes primed vesicles to fuse with the presynaptic plasma
membrane. Vesicle fusion begins with the opening of a fusion pore and after release of
neurotransmitter into the synaptic cleft results in full collapse of the vesicle into the presynaptic
plasma membrane. Apart from this full fusion model, there is another model of vesicle fusion,
called “kiss-and-run”, in which the vesicle fuses transiently with the plasma membrane and is
retrieved simply by closure of the fusion pore. Although there is evidence of kiss-and-run
endocytosis for dense core vesicle exocytosis occurring in many neuroendocrine cells (Harata,
Aravanis et al. 2006), whether it plays a significant role in synaptic vesicle recycling in nerve
terminals under physiological conditions remains controversial. After complete fusion, new
synaptic vesicles are formed by membrane internalization through endocytosis, refilled with
neurotransmitters and recycled for another round of vesicle fusion (Sudhof 2004). Each of these
steps involves several proteins acting in concert to ensure vesicle trafficking (Figure 1.4).
7
Figure 1.3. Mechanisms of Synaptic Vesicle Trafficking. At the presynaptic terminal, synaptic vesicles charged with neurotransmitter are mobilized from a reserve pool to close apposition to the plasma membrane in the docking process. The docked vesicles are then primed for calcium- triggered fusion in a readily releasable pool. Calcium influx following activation of voltage gated calcium channels by membrane depolarization, triggers primed vesicles to fuse and release the neurotransmitters. After fusion, new vesicles are formed via endocytosis and neurotransmitters are loaded into newly formed vesicles. These newly formed vesicles are recycled either locally in the releasable pool or through the reserve pool for subsequent rounds of fusion.
8 Synaptic vesicle fusion is mediated by soluble N-ethylmaleimide sensitive factor (NSF) attachment protein receptors (SNAREs). The SNARE proteins involved in SV trafficking are: vesicle-SNARE (v-SNARE) neuronal SYNAPTOBREVIN (n-SYB), and two target membrane
SNARE proteins SYNTAXIN (SYX) and SNAP25. In the presynaptic terminal, vesicles from reserve pool physically contact the plasma membrane in a process called docking. The docked vesicles undergo a series of biochemical interactions in preparation for rapid fusion in response to a presynaptic action potential, in the priming process. SNARE complexes are formed by the assembly of one SNARE helix from the v-SNARE and three from the two t-SNAREs (Figure
1.4). According to current models, vesicle priming involves initial assembly at the N-terminal end of the SNARE 4 helix bundle to form a loose trans-SNARE complex. Subsequent zippering of SNARE complexes in response to calcium influx may drive vesicle fusion (Chen and Scheller
2001; Sorensen, Wiederhold et al. 2006; Kawasaki and Ordway 2009).
SNAREs are key players in vesicle fusion, but there are other proteins that are required to ensure the speed and accuracy of the neurotransmitter release. For example, COMPLEXINS, function as regulators of SNAREs in a post-priming step just before fusion and
SYNAPTOTAGMIN, a transmembrane protein of the synaptic vesicles functions as a calcium sensor for neurotransmitter release (Sudhof 2004). After integration of the vesicle into the plasma membrane, a very stable cis-SNARE complex is formed. These cis-SNARE complexes may be disassembled by N-ethylmaleimide sensitive factor (NSF) and soluble NSF attachment protein
(SNAP). The disassembly of SNARE complex by NSF releases the t-SNAREs for subsequent vesicle priming. This has been confirmed by biochemical analyses (both in vitro and in vivo)
(Sollner, Whiteheart et al. 1993; Tolar and Pallanck 1998). NSF binds to the SNARE complex via
SNAP and together they form a 20S complex. The 20S complex is dissociated in an ATP- dependent fashion (Sollner, Bennett et al. 1993). Vesicle recycling via clathrin-mediated
9 endocytosis follows. Vesicles at each stage of the synaptic vesicle cycle form distinct vesicle pools as are described in the next section.
1.1.3. Synaptic Vesicle Pools
Decades ago studies in cat sympathetic ganglia demonstrated that some synaptic vesicles are more easily released than others, suggesting that all synaptic vesicles are not functionally equivalent (Birks 1961). Research in other synaptic preparations such as Drosophila NMJ, frog
NMJ, the mammalian calyx of Held and cultured hippocampal synapses have confirmed this observation and proposed the existence of three separate synaptic pools: a readily releasable pool
(RRP), a recycling pool and a reserve pool (Rizzoli and Betz 2005). The RRP consists of vesicles that are docked and primed at the AZ and released immediately in response to a stimulus (Figure
1.5). The recycling pool consists of synaptic vesicles that are released during moderate frequency stimulation and are rapidly recycled (Figure 1.5). It contains approximately 10 to 20% of all vesicles at the synapse. The reserve pool vesicles, which comprise about 80 to 90% of the total synaptic vesicle population, are reluctant to release and are mobilized only during high frequency stimulation when other vesicle pools have been depleted (Denker and Rizzoli 2010).
This synaptic vesicle pool paradigm was originally developed based on experiments utilizing high-frequency stimulation, which differs from the low frequency stimuli observed under physiological conditions. Recent evidence from physiological stimulation experiments indicates that the reserve and recycling pools are not spatially distinct as was previously thought and that these two populations are intermixed. With strong stimulation, release is too fast to allow intermixing of the two vesicle pools. On the contrary, the release of vesicles under physiological stimulation for longer periods is slow and hence differentiation between the two pools is difficult to observe (Denker and Rizzoli 2010).
10
Figure 1.4. Core proteins involved in Synaptic Vesicle trafficking. In the presynaptic terminal, docked vesicles are primed for fusion. During priming, v-SNARE, synaptobrevin forms a loose trans-SNARE complex with the two t-SNAREs, syntaxin and SNAP-25. Arrival of an action potential at the presynaptic terminal leads to membrane depolarization that activates voltage- gated calcium channels (VGCC). Influx of calcium through VGCC triggers fusion of primed synaptic vesicles and neurotransmitter release. After fusion, N-ethylmaleimide sensitive factor (NSF) disassembles cis-SNARE complexes and releases syntaxin and SNAP 25 for subsequent vesicle priming. Vesicles are recycled via endocytosis mediated by dynamin and other proteins.
11
Figure 1.5 Synaptic Vesicle Pools (Modified from (Denker and Rizzoli 2010)). The readily releasable pool (RRP; depicted in red) consists of the vesicles docked at the active zone and primed for release. Under moderate stimulation conditions, after depletion of the RRP, the recycling pool vesicles are recruited to the active zone (left arrow). The reserve pool vesicles (blue) are recruited only after the depletion of the recycling pool (right arrow), which occurs under very high stimulation conditions.
12 Analysis of the molecular mechanisms involved in SV trafficking and neurotransmitter
release has identified a number of proteins and this progress has begun to address their specific roles. Sections 1.2 and 1.3 describe the functions and interactions of the core components of the neurotransmitter release apparatus.
1.1.4. Drosophila melanogaster: A genetic model System for the study of synaptic
transmission
Molecular mechanisms of synaptic transmission remain a topic of intensive investigation
in cellular and molecular neuroscience. In order to understand the in vivo molecular mechanisms of synaptic transmission, genetic analysis is extremely important. The Drosophila model system has played a key role in this through the use of classical genetics, electrophysiology, imaging, biochemical analysis and a wide variety of methods available for gene manipulation in transgenic animals. Drosophila melanogaster is thus an extremely powerful and sophisticated model system for identifying and analyzing complex biological processes. Similarity of synaptic mechanisms in
Drosophila and vertebrates further render Drosophila a good model system for the study of synaptic transmission.
Classical genetic approaches in Drosophila may be used to introduce mutations in single genes and analyze their in vivo functions. Among these, the study of temperature sensitive (TS) paralytic mutants has been quite significant. TS mutants have been effective tools for in vivo analysis of specific proteins in neural function (Suzuki, Grigliatti et al. 1971; Grigliatti, Hall et al.
1973). TS mutants develop and function normally at permissive (e.g ambient) temperatures but exhibit paralysis or motor defects at elevated temperatures, indicating acute perturbation of proteins function. Thus TS paralytic mutants affecting synaptic transmission can provide insights into the molecular mechanisms of synaptic function. Forward genetic screens for TS paralytic mutants and screens for modifiers of existing TS mutants have led to discovery of novel synaptic mechanisms. For example, TS alleles of shibire and comatose were originally identified in a
13 classical genetic screen for mutants exhibiting paralysis at elevated temperatures (Suzuki,
Grigliatti et al. 1971; Siddiqi and Benzer 1976). Analysis of shibire mutants demonstrated that
DYNAMIN (the shibire gene product) is essential in SV endocytosis (discussed later) (Takei,
Mundigl et al. 1996; Kawasaki, Hazen et al. 2000; Murthy and De Camilli 2003). In comtST17, a
TS paralytic NSF mutant, electrophysiological and ultrastructural analysis demonstrated an in vivo role for NSF in priming of synaptic vesicles (Ordway, Pallanck et al. 1994; Pallanck,
Ordway et al. 1995; Pallanck, Ordway et al. 1995; Kawasaki, Mattiuz et al. 1998). The comatose
gene product, dNSF1, was later shown to function in the disassembly of the cis-SNARE
complexes following vesicle fusion (Tolar and Pallanck 1998). Recent work from our lab has
provided evidence supporting the role of dNSF in replenishing active zone t-SNAREs for
subsequent vesicle priming (Kawasaki and Ordway 2009).
A TS paralytic allele of the cacophony (cac) gene, cacTS2 was recovered in a genetic
screen for modifiers of comatose (Dellinger, Felling et al. 2000). The cacophony gene encodes
the α1 subunit of primary pre-synaptic voltage-gated calcium channel (Kawasaki, Felling et al.
2000; Kawasaki, Collins et al. 2002). cacTS2 was recovered as an enhancer of comatose. However,
when isolated from comatose, cacTS2 exhibits rapid paralysis at 38°C. To further our understanding of synaptic transmission, former members of our lab performed a forward genetic screen for modifiers of cacTS2. A new TS mutation, dabEC1 was recovered as an enhancer of cacTS2
from this screen and this mutation was mapped to the disabled (dab) gene. The characterization
of dabEC1 and functional analysis of DAB is discussed in Chapter 3.
14 1.2. Molecular Mechanisms of Synaptic Vesicle Exocytosis
1.2.1. Membrane Fusion
Membrane fusion is essential for many cellular functions, including vesicle trafficking in eukaryotic cells. Some of the well characterized examples of membrane fusion are synaptic vesicle exocytosis, vacuole fusion and trafficking with the Golgi apparatus in yeast. All types of membrane fusion involve the merging of two separate hydrophobic lipid rich membranes into one continuous membrane with an aqueous environment. This is an energetically expensive process which requires removal of water and dehydration of the polar phospholipid headgroups.
According to one popular model of membrane fusion, the stalk hypothesis (Jahn, Lang et al.
2003), the two membranes first come within close contact (Figure 1.6.A) and a few lipid headgroups in the proximal layers merge to form a stalk (Figure 1.6.B). This is followed by expansion of the stalk leading to fusion of the distal layers to form a hemifusion intermediate
(Figure 1.6.C). This hemifusion intermediate then initiates the formation of a lipidic fusion pore
(Figure 1.6.D). Finally expansion of the fusion pore allows the completion of membrane fusion
(Figure 1.6.E) (Jahn, Lang et al. 2003; Jackson and Chapman 2006).
Membranes do not fuse spontaneously owing to high repulsion between the opposing bilayers that generate a huge energy barrier. These repulsive forces are overcome in biological membranes with the help of fusion proteins (Jahn and Sudhof 1999). As shown in Figure 1.6.A, the fusion proteins (SYX, SYB and SNAP25) act as scaffolds by pulling the two membranes, plasma membrane and vesicular membrane to close proximity.
15
Figure 1.6 Model for Membrane fusion [Modified from (Jackson and Chapman 2006)]. Proteins draw the two lipid bilayers together (A), then a few lipid headgroups in the proximal layers merge to form a stalk (B), followed by stalk expansion leading to a hemifusion intermediate (C). Then a lipidic fusion pore is formed (D) and expansion of this fusion pore allows the completion of membrane fusion (E).
16 1.2.2. The Soluble N–ethylmaleimide Sensitive Factor (NSF) Attachment protein Receptors
(SNAREs)
The mechanisms governing membrane fusion have been studied extensively and are
conserved from yeast to mammals. Despite the superficial differences between secretion in yeast
(constitutive) and secretion of neurotransmitters from synaptic vesicles (regulated), the basic
mechanism and core proteins involved are similar. SNARE (soluble N-ethylmaleimide-sensitive
factor attachment protein receptor) proteins were first characterized in late 1980s and, since then,
progress in research has established that SNAREs are essential for membrane fusion. A series of
studies showing cleavage of the neuronal SNAREs by tetanus and botulinum toxins (peptidases
that block neurotransmitter release), have demonstrated that SNAREs are essential for synaptic
vesicle fusion (Schiavo, Benfenati et al. 1992; Blasi, Chapman et al. 1993; O'Connor, Heuss et al.
1997). These proteins form a complex which acts as a receptor for two proteins, NSF and its
binding partner alpha-SNAP (Soluble NSF Adaptor Protein) which have been identified as
essential regulators of vesicular transport. They were named SNAREs to indicate their binding to
SNAP (Sollner, Whiteheart et al. 1993).
The neuronal SNARE proteins involved in synaptic vesicle trafficking are: n-SYB,
SYXand SNAP25 (synaptosomal-associated protein of 25kDa). While n-SYB is a transmembrane protein of the synaptic vesicles (Figure 1.7.C), SYX is a transmembrane protein of the pre-
synaptic plasma membrane (Figure 1.7.A). The other t-SNARE, SNAP-25 is a peripheral
membrane protein attached to the pre-synaptic plasma membrane by palmitoylation of several cysteine residues (Figure 1.7.B). All SNAREs have a characteristic SNARE motif: a conserved stretch of 60-70 amino acid residues arranged in heptad repeats. The SNARE motif is functionally important as it mediates the assembly of SNAREs into a SNARE complex. SYX and
n-SYB each have a single carboxy-terminal transmembrane domain that is connected to one
SNARE motif by a short linker. On the other hand SNAP-25, which lacks a transmembrane
17 domain, contains two different SNARE motifs located at its N- and C- termini. These motifs are joined by a flexible linker containing four palmitoylated cysteines that mediates membrane association (Hanson, Heuser et al. 1997; Brunger 2005; Jahn and Scheller 2006). Synaptobrevin also has a short unstructured N-terminal lumenal region.
The structure of the core complex has been elucidated using freeze-etch electron microscopy (Hanson et al., 1997), fluorescence resonance energy transfer (Lin and Scheller
1997), electron paramagnetic resonance spectroscopy (Poirier 1998) and later confirmed by the analysis of the X-ray crystal structure (Sutton, Fasshauer et al. 1998). These studies have shown that the coiled-coil domains of the SNARE proteins associate in a parallel alignment from N-to
C-termini. They form a four helix bundle, with one helix from SYX and n-SYB and two helices contributed by SNAP-25 (Figure 1.8).
Although the individual SNAREs proteins are unstructured, they spontaneously assemble into a highly stable ternary SNARE complex. This complex is resistant to SDS (Sodium Dodecyl
Sulfate) denaturation and cleavage by the botulinum and tetanus neurotoxins (Hayashi, McMahon et al. 1994). Structural analysis of the core complex has revealed that the four helix bundle can be divided into 16 layers denoted from -7 to +8, where conserved residues from each helix interact to make a hydrophobic core of the complex with the exception of the central layer (Figure 1.9).
The central layer 0 is composed of an Arginine (R) residue from n-SYB, one Glutamine (Q) from
SYX and two Glutamines (Q) from the two SNAP-25 helices.The ionic interactions between the positively charged guanidino groups of Arginine residue with the carboxyl groups from each of the three Glutamine residues contribute to the polar nature of the central layer (Sutton, Fasshauer et al. 1998).These ionic interactions are sealed by the flanking leucine-zipper layers, which stabilize the four-helix bundle by enhancing the electrostatic interactions within this ionic layer.
NSF and alpha-SNAP (discussed later) function in the disassembly of this complex, possibly by
18 breaking this seal exposing the ionic layer to the solvent and thereby weakening the interhelix
interactions (Sutton, Fasshauer et al. 1998).
SNAREs drive fusion: Zippering model
SNARE motifs in the SNAREs on opposite membranes (vesicle and plasma membrane) intially bind at their N-termini to form a loose trans-SNARE complex that brings the membranes close together. This assembly begins with the formation of a binary “acceptor” complex between
SYX and SNAP25 with 1:1 stoichiometry in the plasma membrane (Fasshauer and Margittai
2004), followed by the binding of n-SYB with its N-terminus to the N-termini of the SNAP25:
SYX (1:1) acceptor complex. The trans-complex is associated with the primed vesicle state which then partially zippers up from the N-to C-termini of the helices to maintain the vesicles in a hemifused state. Influx of calcium triggers the full zippering of the SNARE helices to form a stable four helix bundle. This zippering is thought to catalyze membrane fusion by forcing membranes together and overcoming the energy barrier for fusion. After fusion and collapse of the vesicle into the target membrane, the SNAREs are located in the same membrane in cis-
SNARE conformation. This complex is then disassembled into free SNAREs for further rounds of binding and fusion through a process mediated by NSF and alpha-SNAP (discussed in the next section).
Several studies provide evidence in support of the zipper model of SNARE function in synaptic vesicle fusion. Early evidence came from the observation that the alpha helices in the ternary SNARE complex are oriented in a parallel fashion as required for N- to C- termini zippering (Sutton, Fasshauer et al. 1998). In vitro studies on SNARE complex assembly using deletion mutants also support the N-to C-terminal assembly (Fasshauer and Margittai 2004).
Mutational analysis of N-terminal, middle and C-terminal regions of the SNAP25 SNARE motifs demonstrated that N-terminal assembly is implicated in vesicle priming, whereas C-terminal assembly drives fusion (Sorensen, Wiederhold et al. 2006).
19
Figure 1.7 Domain structure of the three neuronal SNARE proteins involved in synaptic vesicle fusion [Modified from (Lang and Jahn 2008)]. SYX 1A and SNAP-25 (with two SNARE motifs) are associated with the presynaptic membrane, whereas SYB 2 is synaptic vesicle associated.
20
Figure 1.8 Model of the neuronal SNAREs assembled into the SNARE core complex [Modified from (Rizo and Sudhof 2002)]. A trans-SNARE complex consists of four α helices, one from vesicle-SNARE, n-SYB (shown in red), one from a target-SNARE, SYX (in yellow), and two from another target-SNARE, SNAP25 (in green and blue).
21
Figure 1.9 Structure of the SNARE core complex [Modified from (Sutton, Fasshauer et al. 1998)]. A) 16 stacked layers of interacting side chains, from -7 to +8, mediate the interactions between the SNARE coiled-coil domains. B) At layer 0, SYX and SNAP-25, each contribute a glutamine residue and SYB contributes an arginine to form ionic bonding.
22 1.2.3 NSF and soluble NSF attachment protein (SNAP)
After vesicle fusion, all the SNAREs reside together in the plasma membrane in a cis-
configuration. Disassembly of these low energy cis-SNARE complexes is required to recycle the
SNAREs for function in further rounds of neurotransmitter release. This disassembly requires considerable energy that is provided by NSF (Jahn and Scheller 2006).
NSF, a homohexamer of 75kDa subunits, was identified as a cytosolic protein required for vesicular transport in a cell-free Golgi membrane fusion system (Block, Glick et al. 1988;
Malhotra, Orci et al. 1988). The yeast homolog of NSF, Sec18p was identified as cytosolic
protein necessary for all steps of intracellular vesicle trafficking (Novick, Field et al. 1980). NSF,
a member of the AAA+ protein family, forms a hexameric ring structure in which each NSF
subunit contains 3 domains. These include an amino terminal domain that undergoes
conformational changes during catalytic cycle and two ATP-binding domains D1 and D2. The
ATPase activity of the D1 domain is crucial for core-complex disassembly, whereas the D2 domain has little ATPase activity but is required for hexamerization (Rizo and Sudhof 2002; Jahn and Scheller 2006). NSF does not directly interact with the SNARE complex, but binds to the ternary complex via a cytosolic factor called α-SNAP (soluble NSF attachment protein) and together they form a 20S complex. The 20S complex is dissociated in an ATP-dependent fashion
(Sollner, Bennett et al. 1993).
Drosophila has a single homolog of a α-type SNAP (Ordway, Pallanck et al. 1994) and two homologs of NSF, dNSF1 and dNSF2 (Ordway, Pallanck et al. 1994; Boulianne and Trimble
1995; Pallanck, Ordway et al. 1995). The amino acid sequence of dNSF1 is 84.5% identical to dNSF2. dNSF1 is encoded by comatose and genetic analysis of temperature-sensitive paralytic mutations in comatose revealed an in vivo role of NSF in synaptic transmission (Kawasaki,
Mattiuz et al. 1998; Kawasaki, Hazen et al. 1999; Kawasaki and Ordway 1999) and SNARE complex disassembly (Tolar and Pallanck 1998). comtST17 harbors a mutation (G274Q) in the D1
23 domain of NSF that disrupts its ATPase activity, leading to accumulation of cis-SNARE
complexes (Tolar and Pallanck 1998). Subsequent genetic analysis has demonstrated that NSF is essential for maintaining neurotransmitter release during sustained stimulation and that NSF functions in replenishing active zone t-SNAREs for subsequent vesicle priming (Kawasaki and
Ordway 2009).
1.2.4 Presynaptic Voltage Gated Calcium Channels
Presynaptic voltage gated calcium channels (VGCC) play a vital role in chemical synaptic transmission by providing calcium influx for regulated neurotransmitter release. When an action potential arrives at the presynaptic plasma membrane, depolarization triggers the opening of VGCCs. The resulting calcium entry leads to fusion of the synaptic vesicles with the presynaptic plasma membrane and the release of neurotransmitter. The calcium ions thus act as a second messenger by coupling the membrane potential to neurotransmitter release. VGCCs have been classified into six different types based on their distinct physiological and pharmacological properties: L, N, P/Q, R and T-types. Of these N and P/Q-type channels are concentrated at the presynaptic nerve terminals and are involved in neurotransmitter release (Catterall 1998; Catterall
2000). These channels are composed of several subunits including a pore forming primary structural α1 subunit and the three auxiliary subunits β, α2δ and γ. The auxiliary subunits directly interact with the α1 subunit and have been found to modulate the trafficking and biophysical properties of the α1 subunit. The α2δ and β subunits are important determinants of the channel’s current amplitude as well as its activation and inactivation kinetics (Arikkath and Campbell
2003).
The α1 subunit of presynaptic calcium channels in Drosophila is encoded by the cacophony (cac) gene. Originally identified in mutants exhibiting courtship defects, subsequent studies showed that cac encodes a voltage-gated calcium channel α1 subunit homologous to mammalian presynaptic calcium channel α1 subunits (P/Q and N-type channels) (Smith, Wang et
24 al. 1996). Earlier studies in our lab isolated a temperature-sensitive (TS) paralytic mutant of cac,
TS2 cac and first demonstrated that cac- encodes a primary presynaptic calcium channel in
Drosophila, contributing to synaptic transmission (Dellinger, Felling et al. 2000; Kawasaki,
Felling et al. 2000). It was also shown that a GFP-tagged α1 subunit, CAC1-EGFP, is localized at active zones (Kawasaki, Zou et al. 2004) and this GFP tagged cac1 transgene has proven to be quite useful as an active zone marker for studies at native synapses.
1.2.5 Regulators of Synaptic Vesicle Exocytosis
1.2.5.1 Synaptotagmin
Synaptic transmission does not proceed until calcium enters the presynaptic terminal in response to the arrival of an action potential. This calcium dependence of neurotransmission was discovered by Katz and his colleagues in 1960s (Katz and Miledi 1968). Since then it has been proposed that a calcium sensor is present in the presynaptic nerve terminal that binds to calcium and triggers neurotransmitter release. SYNATOTAGMIN1 is the best characterized calcium sensor for neurotransmitter release (Fernandez-Chacon 2001; Koh 2003).
SYNAPTOTAGMINs (SYTs) constitute a large family of proteins with 17 isoforms in humans and 7 isoforms in flies. SYTs contain a single short N-terminal transmembrane region and two C-terminal calcium binding domains, C2A and C2B (Bai and Chapman 2004). SYT1 is an integral vesicle membrane protein that is anchored to the vesicle through its N-terminal transmembrane domain (Bai and Chapman 2004). The C2A and C2B domains adopt beta- sandwich structures with three flexible loops extending from one end and bind three and two
Ca2+ ions, respectively. Through these flexible loops, the C2 domains bind to phospholipids in the vesicle and plasma membranes in a calcium dependent manner [reviewed in (Rizo and
Rosenmund 2008)]. Evidence from mutational analysis has demonstrated that SYTI acts as a calcium sensor in neurotransmitter release (Fernandez-Chacon, Konigstorfer et al. 2001). Ca2+ binding to C2B domain is more important for neurotransmitter release as a larger decrease in
25 release is observed for mutations in C2B domain relative to C2A (Mackler, Drummond et al.
2002; Robinson, Ranjan et al. 2002). In addition, recent studies indicate that calcium dependent
C2B domain binding to the vesicle and plasma membranes can bring these membranes together
[reviewed in (Rizo and Rosenmund 2008)].
SYT1 interacts with individual SNAREs and SNARE complexes, but its association with the SNARE complex most likely triggers release. The exact mechanism of how this binding causes release is not clear, but one model suggests that SYT1 binds to SNARE complexes even in the absence of Ca2+ and upon Ca2+ entry, may bind phospholipids, destabilize the fusion intermediate and open the fusion pore (Sudhof 2004).
As discussed in the previous section, SNARE proteins and their assembly are key for all intracellular membrane fusion reactions. At synapses, a large number of regulatory proteins interact with SNAREs to mediate rapid and calcium triggered synaptic vesicle fusion.
1.2.5.2 COMPLEXIN
COMPLEXINS (CPXs) were originally identified as small hydrophilic proteins associated with the synaptic SNARE complex. They constitute a family of small helical neuronal proteins that bind with high affinity to the assembled SNARE complex in a 1:1 stoichiometry and regulate its function (Ishizuka, Saisu et al. 1995; McMahon, Missler et al. 1995; Takahashi,
Yamamoto et al. 1995; Chen, Tomchick et al. 2002; Reim, Wegmeyer et al. 2005). CPXs are highly conserved from invertebrates to vertebrates but are not found in yeast, suggesting that they have evolved as neuron-specific regulators of membrane fusion (Brose 2008). Recent studies have revealed an association of altered complexin levels with neurological disorders, such as schizophrenia, Huntington's disease, depression, bipolar disorder, Parkinson's disease,
Alzheimer's disease, traumatic brain injury, Wernicke's encephalopathy, and fetal alcohol syndrome (Brose 2008). Thus understanding the function of complexin and its mechanism of
26 action will elucidate regulation of neurotransmitter release and may also provide clues on the
etiology of neurological diseases.
Complexin protein family
In mammals, complexins form a family of four small proteins containing 134-160 amino
acid residues (Figure 1.10.A). They are localized to presynaptic terminals where they regulate
neurotransmitter release (Yamada 1999). Mammalian CPXI and II are 86% identical and CPXs
III and IV show high homology to each other (58%), but only limited homology to CPXs I and II
(24-28% identity) (Reim, Wegmeyer et al. 2005). In contrast to the mammals, invertebrates
typically have only one complexin gene, with the exception of C.elegans, which has two
complexin genes (Figure 1.10.B) (Reim, Wegmeyer et al. 2005). Complexins are highly conserved across species, with the invertebrate CPXs showing similarity to mammalian CPXI and
II, except for the conserved C-terminal CAAX motif (discussed later) (Reim, Wegmeyer et al.
2005).
Mammalian CPXI and CPXII are highly enriched in the brain (Yamada 1999), with some expression at retinal synapses (amacrine cells) (Reim, Wegmeyer et al. 2005). while CPXIII and
CPXIV are predominantly expressed in the retina, with limited CPXIII expression in the brain including hippocampus, olfactory bulb, cortex, inferior colliculus, thalamus, and striatum (Reim,
Wegmeyer et al. 2005). In cortex and cerebellum, CPXI has been found in axosomatic terminals, which are inhibitory GABAergic, and CPXII in axodendritic terminals, which are excitatory glutamatergic terminals (Takahashi, Yamamoto et al. 1995; Yamada 1999). These results led to the assumption that CPXI modulates inhibitory and CPXII modulates excitatory neurotransmitter release. However, functional studies in hippocampal neurons deficient in CPXI and II showed that the two isoforms function redundantly in glutamatergic and GABAergic synaptic transmission, indicating that CPXI and CPXII are not generally distributed into inhibitory and excitatory synapses (Reim, Mansour et al. 2001).
27
Figure 1.10 Complexin Protein family [Modified from (Brose 2008)]. (A) Primary amino acid sequences complexins from vertebrate and invertebrate species were aligned for maximal homology. Blue shading indicates residues that are identical in the majority of sequences and light green indicates similar residues. Orange bar: core alpha-helix mediating SNARE complex binding, Green bar: accessory alpha-helix and Purple bar: N-terminal sequence that facilitates fusion. CAAX-box farnesylation consensus motifs are shown in red boxes. B) Phylogenetic tree illustrating the homology between different complexins. ag, A. gambia; ce, C. elegans; ci, C. intestinalis; CPX, complexin; dm, D. melanogaster; h, human; hm, H. medicinalis; lp, L. pealei; m, mouse; nj, Narke japonica and xl, X. laevis.
28 mRNA distribution of CPXI and II has suggested that CPXI may be involved in circuits of motor
learning and sensory processing, while CPXII may be involved in circuits of cognition, emotional
behavior and control of voluntary movement (Freeman and Morton 2004). CPXIII and IV are
differentially expressed in the retina, with CPXIII enriched in cone photoreceptor ribbon synapses
and CPXIV in rod photoreceptor ribbon synapses. CPXs III and IV, unlike CPXs I and II contain
a C-terminal CAAX motif (Reim, Wegmeyer et al. 2005). This motif represents a conserved
sequence for prenylation and is also found in a majority of invertebrate CPX isoforms.
Drosophila has a single gene that encodes for the CPX protein (Tokumaru, Umayahara et
al. 2001). An earlier study in Drosophila reported that CPX, which is enriched in both CNS and
peripheral synapses, is expressed in presynaptic terminals (Huntwork and Littleton 2007). This
study generated a cpx null mutant. The cpx null mutants are semilethal, with most animals dying
before adult eclosion. The escaper adults show severe motor defects and disruption of synaptic
transmission. The cpx locus may encode up to six different protein sequences, by alternative splicing of its exon. The isoforms mainly differ in their C-termini, the predominant CPX in the nervous system (Supplementary Figure 4.S2.B) contains the conserved CAAX prenylation motif, which is also present in mammalian CPXs III and IV.
Prenylation:
Prenylation refers to the posttranslational modification of proteins, which involves covalent addition of an isoprenyl lipid moiety to the conserved cysteine residue (C) in the C- terminal CAAX motif. The CAAX motif consists of a cysteine (C), followed by two aliphatic amino acid residues (AA) and a terminal residue (X) that usually dictates the type of lipid group
(farnesyl or geranyl-geranyl) to be added. For example, farnesyl group is added if there is a methione, glutamine or serine at X and geranyl-geranyl group is added if there is leucine or isoleucine in the X position (Zhang and Casey 1996). Mammalian CPXs III and IV are known to
29 be farnesylated (Reim, Wegmeyer et al. 2005). Invertebrate CPXs have a CAAX motif, with
glutamine in the X position indicating that they also get farnesylated.
There are three enzymes involved in farnesylation: farnesyl transferase, a CAAX protease
and a carboxy methylase (Figure 1.11). As a first step, cysteine residue in the CAAX motif gets farnesylated by farnesyl transferase, in the cytosol. This farnesylated protein then travels to the cytosolic surface of the ER, where the three C-terminal residues, AAX, are cleaved by the CAAX protease. This proteolysis is followed by methylation of the carboxyl group in the farnesylated cysteine. Such a modification leads to the production of a mature protein with highly hydrophobic
C-terminus (Casey 1992; Winter-Vann and Casey 2005). Farnesylation of proteins promotes their association with cellular membranes; without farnesylation, the protein remains cytosolic and is unable to function properly (Gomes, Ali et al. 2003). In addition, farnesylation of proteins is
important for protein-protein interactions and protein stability (Zhang and Casey 1996; Winter-
Vann and Casey 2005). In case of mammalian CPXs, III and IV, containing the CAAX motifs,
these isoforms are farnesylated and their farnesylation mediates localization to synapses. It has
been postulated that targeting of CPXs III and IV to the membranes of the retinal ribbon
synapses, may contribute to the unique release properties at these synapses (Reim, Wegmeyer et
al. 2005). Some isoforms of Drosophila CPX have a CAAX motif at their C-terminal, however their farnesylation has not been shown.
30
Figure 1.11 Schematic of the protein farnesylation mechanism [Modified from(Schillo, Belusic et al. 2004)]. The cysteine is modified with a farnesyl residue by a CAAX farnesyltransferase. After the post-translational modification, the last three amino acids AAX are removed by a cellular protease, and the newly exposed carboxyl group of the farnesylated cysteine residue is methylated.
31 Structure of Complexin and SNARE binding
NMR analysis has revealed that CPX contains a conserved central a-helical domain
(residues 26-83 in rat) that mediates binding to assembled ternary SNARE complex (Pabst,
Hazzard et al. 2000; Pabst, Margittai et al. 2002). Initial biochemical studies also provided evidence for this interaction of CPXs through the central alpha helix (McMahon, Missler et al.
1995). It has also been shown that CPXs do not affect SNARE complex assembly or disassembly
(Pabst, Margittai et al. 2002; Weninger, Bowen et al. 2008). The N and C termini are largely unstructured and hence their interactions are not as well characterized as those of the central region. CPXs are thus composed of central helices (accessory and main) and N and C terminal regions. The SNARE binding domain is highly conserved among species and between the different CPX isoforms. All the four mammalian CPXs I, II, III and IV bind to the SNARE
complex, but CPXIV has lower SNARE-binding affinity (Reim, Wegmeyer et al. 2005). CPX has
been shown to bind to monomeric SYX (McMahon, Missler et al. 1995) and a recent study using
single molecule fluorescence has provided evidence for weak CPX binding to a binary complex
of SNAP25 and SYX (Weninger, Bowen et al. 2008).
The three dimensional structure of CPX-SNARE complex was elucidated by X-ray
crystallography and NMR spectroscopy studies of the CPX/SNARE complex (Bracher, Kadlec et
al. 2002; Chen, Tomchick et al. 2002)(Figure 1.12). These studies revealed that CPX central
alpha helix, comprised of residues from 48-70, binds in an antiparallel fashion to the groove
between SYX and n-SYB helices via coiled-coil interactions (Figure 1.12). Bracher et al., 2002
reported that CPX interacts with SYX and n-SYB at residues surrounding the ionic zero layer of
the SNARE complex. The N-terminal part, comprising of residues 30-47, forms a stable alpha-
helix. This helix designated as accessory alpha-helix does not participate directly in SNARE
binding. NMR and X-ray data show that only residues from 48-70 in CPX interact with the
SNARE complex (Chen, Tomchick et al. 2002).
32
Figure 1.12 The COMPLEXIN-SNARE complex [Modified from (Brose 2008)]. (A) Ribbon diagram and (B) space filling model of the mammalian COMPLEXIN-SNARE complex. SNAP25 is shown in blue, SYX-1 in yellow, n-SYB-2 in red and CPX-1 in orange. CPX-1 binds in an antiparallel fashion in the groove between SYX and n-SYB.
33 COMPLEXIN binding does not alter the ternary complex structure, but stabilizes the interface between SYX and n-SYB as shown by deuterium exchange experiments (Chen,
Tomchick et al. 2002). There are strong repulsive forces preventing the formation of this interface during SNARE complex assembly. On the basis of the COMPLEXIN/SNARE structural analysis, it was proposed that CPX may function as a “tape” that seals the interface and stabilizes the primed state of vesicles at the plasma membrane, thereby assisting SNARE complex assembly and subsequent vesicle fusion (Chen, Tomchick et al. 2002). In contrast, Bracher et al., 2002 suggested based on the squid CPX structure that CPX may function as part of a complex, along with other accessory factors such as the potential Ca2+ sensor SYT, to regulate membrane fusion by arresting synaptic vesicles in a primed and fusion ready state for calcium triggered exocytosis.
Complexin in Synaptic vesicle exocytosis: Function and mechanism of action
Synaptic vesicle exocytosis involves a highly specialized fusion reaction that is regulated
both spatially and temporally and CPX plays a key role in this regulation. Since its identification as a SNARE binding protein, a variety of genetic, biochemical, structural and electrophysiological studies have provided evidence supporting CPX function in synaptic vesicle
fusion. These studies have been performed in several in vitro and in vivo systems including
liposomes, cells with flipped SNARES, cultured neurons, adrenal chromaffin cells, PC12 cells
and sperm cells. However, depending on the cell type, model system or the experimental
approach used, these studies provide conflicting evidence regarding the specific role of CPX in
regulated exocytosis. Hence, the function of CPX remains controversial.
Presynaptic injection of recombinant CPXII into Aplysia buccal ganglia neurons
decreased neurotransmitter release, and injection of murine anti-CPXII antibody increased
release, thus suggesting an inhibitory effect of CPX (Ono, Baux et al. 1998). In support of this,
two independent studies on PC12 cells showed that overexpression of CPXI/II inhibits
acetylcholine and dopamine release (Itakura, Misawa et al. 1999; Liu, Guo et al. 2007). In studies
34 of the acrosomal exocytosis, addition of excess of CPX to the permeabilized human sperm stopped exocytosis at the hemifusion state, where the SNAREs are assembled in loose-SNARE complexes. This block can be released by addition of the SYNAPTOTAGMIN VI C2B domain and calcium (Roggero, De Blas et al. 2007). Also, overexpression of a CPXI-SYB2 fusion protein that selectively increases presynaptic levels of CPX results in a marked reduction of neurotransmitter release (Tang, Maximov et al. 2006). In vitro studies have also provided some insights to the role of CPXs and their interplay with other proteins in exocytosis. Fusion of cultured cells expressing flipped SNARE proteins facing extracellularly was blocked by both membrane anchored and soluble CPXI (Giraudo, Eng et al. 2006). Further reconstituted liposome fusion mediated by SNAREs was also blocked by CPXs (Schaub, Lu et al. 2006). Both these blocks in fusion were relieved upon addition of calcium and SYT. These in vitro results along with the observation that SYT competes with CPX for SNARE binding and displaces it from the
SNARE complex in a calcium dependent manner, suggests a molecular connection between CPX and calcium stimulation (Tang, Maximov et al. 2006).
The preceding studies have provided evidence suggesting a fusion clamp model for CPX function. According to this clamp model, the SNARE fusion machinery is constitutively active and vesicle fusion is arrested by a clamping mechansim in the absence of the trigger signal
(calcium influx). In this mechanism, CPX is postulated to bind assembled SNAREs and function as a clamp that arrests fusion prior to calcium influx. Consistent with this hypothesis, genetic deletion of CPX in Drosophila melanogaster greatly enhances spontaneous release but decreases evoked neurotransmitter release, providing in vivo support for the fusion clamp model (Huntwork and Littleton 2007). Also a recent CPX knockdown study of mouse cortical glutamatergic synapses reported increased spontaneous and decreased evoked neurotransmitter release
(Maximov, Tang et al. 2009). The clamping mechanism has been studied in the liposome fusion assay, where it was observed that CPX specifically blocked fusion at the hemifused state,
35 indicating that CPX stalls the SNARE complex assembly at the hemifused state and prevents the
full zippering of SNAREs (Schaub, Lu et al. 2006).
In addition to its clamping function, several studies have provided strong evidence that
CPXs play a positive role in regulated exocytosis. CPX knockdown studies in pancreatic insulin secreting cell lines and mast cells showed reduced regulated secretion (Abderrahmani,
Niederhauser et al. 2004; Tadokoro, Nakanishi et al. 2005). Proteoliposome fusion assays have demonstrated a facilitatory role of CPXs in stimulating fusion (Yoon, Lu et al. 2008; Malsam,
Seiler et al. 2009). Regulating synaptic transmission, hippocampal autaptic neurons of CPX I/II double knock-out and CPX I/II/III triple knock mice exhibit a marked reduction in both evoked and spontaneous neurotransmission, with no significant change in the readily releasable vesicle pool (RRP). Since the RRP represents the vesicles that are docked and primed, the normal RRP observed in CPX knock-out mice indicates that CPX may function in exocytosis after priming
(Reim, Mansour et al. 2001; Xue, Stradomska et al. 2008). Also, studies of adrenal chromaffin cells from CPX knockout mice have demonstrated a decrease in calcium triggered exocytosis
(Cai, Reim et al. 2008). These results support a stimulatory function for CPXs in regulated exocytosis, however a CPX knockdown study of mouse cortical glutamatergic synapses reported increased spontaneous and decreased evoked neurotransmitter release (Maximov, Tang et al.
2009).
Taken together, the data suggest the existence of inhibitory as well as facilitatory roles for CPX in SNARE mediated exocytosis (Figure 1.13). Recent evidence from structure function- analysis of CPXI in CPXI/II knockout mice indicates that different domains of CPX have distinct roles (Xue, Reim et al. 2007; Giraudo, Garcia-Diaz et al. 2008). SNARE complex binding by the central alpha-helix is essential for CPX function. The CPX N-terminus facilitates release whereas the accessory helix (between N-terminus and central alpha helix) performs the clamping function by inhibiting spontaneous release (Xue, Reim et al. 2007). Biophysical and physiological
36 analyses of CPX have now shown that interaction of the N-terminus of CPX with the C-terminus of the SNARE complex contributes to the facilitatory role of CPX. Recent studies have shown that the CPX C-terminus is required for priming and clamping synaptic exocytosis (Kaeser-Woo,
Yang et al. 2012).
CPX is thus required to activate SNAREs for subsequent Ca2+ triggering by synaptotagmin, clamping of SNAREs to prevent spontaneous fusion and priming of vesicles for fusion. One of the proposed models for CPX function is shown in Figure 1.15 (Xue, Craig et al.
2010). It postulates that the central alpha-helix of CPX binds to the partially assembled SNARE complex and helps it assemble further (Figure 1.14.b). Meanwhile, the CPX accessory alpha-helix replaces the C-terminus of n-SYB SNARE motif from the four helix bundle and thus prevents C-
terminal assembly (Figure 1.14.b). Upon Ca2+ influx and/or SYT interaction, the CPX N-terminus
releases the accessory-alpha helix (Figure 1.14.c), thereby stabilizing the C-terminus of the
SNARE complex and allowing full zippering of the SNARE complex (Figure 1.14.d).
37
Figure 1.13 Complexin domain structure [Modified from (Maximov, Tang et al. 2009)]. Schematic representation of the structural organization and domains in Complexin. In some homologs, including mouse CPXIII and IV as well as the single Drosophila CPX, there is an evolutionarily conserved “CaaX motif” at the C-terminus that represents a consensus sequence for posttranslational prenylation.
38
Figure 1.14 Proposed model for CPX function [Modified from (Xue, Craig et al. 2010)]. (a) Partial assembly of SNARE complex at the N-terminus during priming. ‘N’ and ‘C’ indicate N and C termini respectively. VM, vesicle membrane; PM, Plasma membrane. (b) CPX central helix binds to the SNARE complex and helps it assemble, while CPX accessory helix replaces the C-terminus of n-SYB-2 SNARE motif in the SNARE four helix bundle, preventing C-terminal assembly. (c) CPX accessory helix is released from the SNARE complex by CPX N-terminus, SYT/Ca2+ or both allowing full assembly of the SNARE complex C-terminus. (d) SNARE complex is fully assembled.
39 1.3 Synaptic Vesicle Endocytosis
After the synaptic vesicle fuses with the presynaptic plasma membrane and releases the
neurotransmitter, the exocytosed vesicle membrane proteins and lipids are recycled via
endocytosis to restore functional synaptic pools for reuse and to ensure long-term functionality of
the synapse (Dittman and Ryan 2009; Haucke, Neher et al. 2011). There are three main endocytic
pathways that are considered to participate in SV recycling: kiss-and-run exocytosis and
endocytosis, classical clathrin-mediated endocytosis and bulk endocytosis. In the case of kiss and
run exo- and endocytosis, vesicles fuse only transiently at the plasma membrane and after the release of neurotransmitter, the vesicle may be retrieved without loss of its identity (Ceccarelli,
Hurlbut et al. 1973; Gandhi and Stevens 2003). There have been many experiments using FM dyes, bromophenol blue and pH-sensitive probes to study vesicle membrane retrieval, and there are results both in favor (Klingauf, Kavalali et al. 1998; Gandhi and Stevens 2003; Harata, Choi et al. 2006) and against (Richards, Guatimosim et al. 2000; Zenisek, Steyer et al. 2002;
Fernandez-Alfonso and Ryan 2004; Granseth, Odermatt et al. 2006) the kiss and run mechanism.
Thus, its role in the synaptic vesicle cycle is currently unresolved. Clathrin-mediated endocytosis represents the main pathway for synaptic vesicle recycling (Murthy and De Camilli 2003;
Granseth, Odermatt et al. 2006; Dittman and Ryan 2009) and is described in the next section.
While it has been difficult to identify proteins functioning in kiss-and-run, molecular details of clathrin-mediated endocytosis are well characterized (Brodin, Low et al. 2000). Bulk endocytosis occurs under strong stimulation conditions. In such conditions, there is accumulation of synaptic vesicle proteins at the presynaptic nerve terminal that may trigger bulk uptake of the presynaptic plasma membrane into vacuolar invaginations as evident at retinal bipolar cell terminals (Holt,
Cooke et al. 2003) and in cultured neurons (Clayton, Evans et al. 2008). These invaginations may form endosome like compartments, which are then formed into synaptic vesicles after proper sorting of lipids and proteins (Geumann, Schafer et al. 2010).
40 1.3.1 Clathrin Mediated Endocytosis
According to the classical model, collapse of the synaptic vesicle into the plasma membrane is followed by retrieval of the vesicle membrane through clathrin mediated endocytosis (CME). This process can occur either directly by invagination of clathrin coated pits
from the plasma membrane or by vesicle budding from an endosomal structure after bulk
endocytosis. CME entails the proper sorting of proteins into recycled vesicles. A large number of
proteins are involved in this process, including CLATHRIN, the AP-2 complex, AP180, Eps15,
DYNAMIN, DAP160 (DYNAMIN Associated Protein 160), ENDOPHILIN, SYNAPTOJANIN and SYNAPTOTAGMIN (Jarousse and Kelly 2001). This is contrary to the earlier belief that
CLATHRIN and AP-2 are sufficient to explain how cargo laden clathrin coated pits are generated
(Jarousse and Kelly 2001). However, CLATHRIN and AP-2 are still key players and the hub for organization of the endocytic machinery, with several accessory proteins assisting them. CME of synaptic vesicles is thought to occur at specialized regions surrounding the AZ, called the periactive zone (PAZ) (Roos and Kelly 1998). Studies of nerve terminals in Drosophila have indicated that several endocytic proteins are enriched at the PAZ, including
INTERSECTIN/DAP160, EPS15, DYNAMIN and ENDOPHILIN (Estes, Roos et al. 1996;
Gonzalez-Gaitan and Jackle 1997; Roos and Kelly 1998; Roos and Kelly 1999; Haucke, Neher et al. 2011). The PAZ also contains cytoskeletal elements, including ACTIN and perhaps SEPTIN filaments (Haucke, Neher et al. 2011).
The basic unit for assembly of the clathrin coat is the triskelion, which is formed by the trimerization of CLATHRIN Heavy Chain (CHC) and CLATHRIN Light Chain (CLC). The
CHCs are linked via a C-terminal trimerization domain and CLC interacts with the CHC trimerization domain (Pishvaee, Munn et al. 1997). CLATHRIN is a self-polymerizing scaffold and organizer of CCV formation. Adaptors are proteins that link cargo into the clathrin-coated pit and accessory proteins assist in the formation of clathrin coat vesicles. Classical adaptor
41 complexes are AP1, AP2, AP3 and AP4. In synaptic vesicle trafficking, CME is centered on the
AP2 adaptor complex. The AP2 adaptor is a heterotetrameric complex comprising of 4 subunits:
α, β2, σ2 and μ2, that are tightly bound to each other (Figure 1.15). The α and β2 subunits, each comprise of a large trunk domain and an appendage domain, which are connected by a flexible linker. The AP2 “core” domain comprises the large trunk domains of α and β2 subunits, and σ2 and μ2. The core domain of AP2 binds to proteins containing endocytic motifs and to membrane proteins, including synaptotagmin. The α subunit, also known as α-adaptin, binds to phophatidylinositol 4,5-bisphosphate (PIP2) through its N-terminus and positions AP2 on the membrane.The C-terminus region of α-adaptin is called the appendage or ear domain, which binds to other adaptor proteins and endocytic accessory proteins. The β2 subunit binds to the terminal domain of clathrin heavy chain promoting lattice assembly (Owen, Vallis et al. 2000) and the μ2 subunit binds to a YXXΦ motif, which mediates internalization of surface membrane receptors (Jing, Spencer et al. 1990; Kirchhausen, Bonifacino et al. 1997). The σ2 subunit seems to stabilize the AP2 complex (Reider and Wendland 2011).There are other alternative adaptors, also referred to as CLASPs (for clathrin-associated sorting proteins) like epsins, AP180, auxilin and DAB2 (discussed later), which are single proteins that recruit cargo proteins into clathrin- coated pits (Figure.1.15) (Traub, Downs et al. 1999; Morris and Cooper 2001). These alternative adaptors can function either independently or in conjunction with the AP2 adaptor complex.
42
Figure 1.15 Endocytic Adaptor Protein Interactions [Modified from (Traub 2003)]. A schematic of possible interactions between clathrin, AP-2 and monomeric adaptors (CLASPs). Sorting signals recognized by the different adaptors are represented in black boxes. PIP2 binding sites are indicated by the spherical grey attachments.
43 Apart from the core components of CME, there are several accessory proteins that assist endocytosis. These include CLASPs that are involved in cargo selection and coat formation
(discussed above), membrane bending accessory proteins (AMPHIPHYSINS, ENDOPHILIN,
EPSINS), proteins with scaffolding functions (EPS15, DAP160), vesicle scission protein
(DYNAMIN) and uncoating proteins (AUXILIN, SYNAPTOJANIN, HSC70 (Jung and Haucke
2007). Drosophila DAP160 contains four SH3 (Src Homology) domains each of which binds to the proline rich domain of different endocytic proteins like DYNAMIN (Roos and Kelly 1998).
DAP160 is also required for the localization of several endocytic proteins and hence may play a key role as a scaffold, which anchors proteins to the site of endocytosis (Koh, Verstreken et al.
2004; Marie, Sweeney et al. 2004). A recent forward genetic screen for synaptic transmission mutants led to the isolation of two new alleles of dap160. Isolation and analysis of these mutants is described in Chapter 5.
CME can be divided into four major steps: (i) clathrin coat nucleation and assembly, (ii) clathrin coated pit (CCP) maturation, (iii) clathrin coated vesicle, (CCV) fission, and (iv) uncoating. In the first step (Figure 1.16 Step1), the adaptor protein complex AP2 and other
adaptors are recruited to PtdIns(4,5)P2 (PIP2) and cargo-enriched membrane sites. CLATHRIN
then binds to the adaptor complex and is thus recruited to the membrane where it starts
polymerizing into a lattice (Jung and Haucke 2007). The adaptors bind to cargo proteins and
sequester into the forming clathrin bud. This clathrin lattice assembly is accompanied by
membrane bending (Figure 1.16 Step 2), which is aided by accessory proteins, and results in the
formation of a CCV. In the fission step, the GTPase DYNAMIN then forms a ring-like structure
around the neck of an invaginating clathrin coated vesicle (Figure 1.16 Step 3) (Hinshaw and
Schmid 1995) and mediates a GTP dependent fission reaction that leads to the internalization of
the CCV. In the last step, after internalization of the vesicle, the clathrin coat is removed by a
44 SYNAPTOJANIN-ENDOPHILIN protein complex (Figure 1.16 Step 4) (Gad, Ringstad et al.
2000; Verstreken, Koh et al. 2003).
DYNAMIN is a large homomeric GTPase which plays a key role in the fission of clathrin coated vesicles and other endocytic vesicles. DYNAMIN is encoded by the shibire gene in
Drosophila and analysis of temperature sensitive shibire mutants provided the first evidence for the role of DYNAMIN function in fission in CME. Each of the DYNAMIN homomeric oligomer subunit is comprised of five conserved domains, including an N-terminal GTPase domain, a middle domain involved in tetramerization and higher-order self-assembly (Ramachandran, Surka et al. 2007) and intracellular targeting (Liu, Surka et al. 2008), a pleckstrin-homology (PH) domain that binds to phosphatidylinositol lipids (Yarar, Surka et al. 2008), a GTP effector domain
(GED) that mediates self-assembly and assembly-enhanced GTPase activity (Sever, Muhlberg et al. 1999) and a C-terminal proline/arginine rich domain (PRD) that binds to SH3 domain (Src homology 3 domain)-containing proteins (Schmid, McNiven et al. 1998).
45
Figure 1.16 Stages of clathrin mediated endocytosis (CME) [Modified from (Jung and Haucke 2007) ]. Stage 1: Recruitment of AP2 and other adaptors to PIP2 and cargo-enriched membrane sites. Stage 2: Clathrin lattice assembly and formation of CCP. Stage 3: Formation of a CCV and fission by dynamin. Stage 4: Uncoating
46 1.3.2 DISABLED (DAB)
Early studies identified the disabled (dab) mutant as a dominant modifier of an Ableson tyrosine kinase (Abl) mutant (Gertler, Hill et al. 1993). dab mutations were mistakenly mapped to
a gene designated dab but were later mapped to the closely linked neurotactin (nrt) gene (Liebl,
Rowe et al. 2003). Thus, until recently no actual mutational analysis of dab gene functions had been carried out.
The DISABLED (DAB) protein encoded by dab has been implicated in several signaling pathways such as Sevenless signaling (Le and Simon 1998), Abl signaling (Gertler, Hill et al.
1993; Liebl, Rowe et al. 2003) and Notch signaling (Giniger 1998; Le Gall, De Mattei et al.
2008). However DAB’s function in these signaling pathways has remained questionable owing to the lack of a dab mutant in Drosophila. In parallel, genetic analysis of DAB protein function has proceeded in mouse and C.elegans. The two mammalian homologs of DAB, DAB1 and DAB2 share a highly conserved N-terminal PTB domain that is characteristic of endocytic adaptor proteins (Xu, Yang et al. 1995; Howell, Gertler et al. 1997). PTB domains were first identified in adaptor proteins belonging to the Shc family (Kavanaugh and Williams 1994). These domains of
Shc were found to bind to the NPXY motif sequence of receptors such as Trk, EGFR, and insulin receptor when the NPXY tyrosine is phosphorylated (Batzer, Blaikie et al. 1995; Bork and
Margolis 1995; Van der Geer, Wiley et al. 1995). The PTB domains of both DAB1 and DAB2 bind to the NPXY sequence in members of the Low Density Lipoprotein Receptor (LDLR) family, and in amyloid precursor protein (APP), preferentially when the NPXY tyrosine is not phosphorylated (Howell, Lanier et al. 1999; Morris and Cooper 2001).
Adaptor proteins mediate the formation of clathrin-coated vesicles through their binding to CLATHRIN, specific cargo proteins, phospholipids and possibly other adaptors (Sorkin 2004).
AP2 is the major clathrin adaptor involved in CME at the plasma membrane and was earlier thought to be a universal adaptor. However, studies have now suggested the presence of other
47 cargo specific adaptors called CLASPs (Clathrin Associated Sorting Proteins), which function
either in conjunction with or independent of AP2 (Conner and Schmid 2003). DAB2 is included
as a member of CLASPs, which participate in clathrin mediated endocytosis of proteins including
receptors, transporters and other membrane proteins. CLASPs include AP180, EPSIN, NUMB, β-
ARRESTIN and ARH (Figure 1.15) (Traub 2003; Robinson 2004).
DAB2 co-localizes with AP2, LDLR and clathrin in clathrin-coated pits (Morris and
Cooper 2001). DAB2 binds the α-ADAPTIN subunit of AP2, as well as non-phosphorylated
NPXY motifs, CLATHRIN, phosphoinositides and MYOSIN VI (Mishra, Keyel et al. 2002;
Morris, Arden et al. 2002). DAB2 has been shown to function in constitutive endocytosis of cell
surface receptors (Teckchandani, Toida et al. 2009), but a role at synapses has not yet been
explored. DAB1, the other mammalian homolog of DAB is also an adaptor protein. It is part of
the Reelin signaling pathway which controls neuroblast migration during brain development
(Herz and Chen 2006). Along with its role in Reelin signaling, DAB1 may regulate the
internalization of the Reelin receptors VLDLR and ApoER2 (Morimura, Hattori et al. 2005).
Studies of the single C.elegans DAB homolog, Ce-DAB1 have shown a conserved role
for the protein in Clathrin Mediated Endocytosis. In vitro studies indicate that Ce- DAB1 shares
properties with mouse DAB2 including binding to the lipoprotein receptors LRP-1 and LRP-2
(Kamikura and Cooper 2003), CLATHRIN and the C.elegans homolog of α-ADAPTIN
(Kamikura and Cooper 2006). Ce-DAB-1 localizes in AP2 containing vesicles and this
localization is disrupted by the depletion of CLATHRIN or AP2 through RNAi knockdown. The
in vivo studies using a null mutant reveal a role for Ce-DAB-1 in endocytosis of yolk proteins in
oocytes by the lipoprotein receptor RME-2 in oocytes as well as a role in solute endocytosis by
macrophage-like coelomocytes (Holmes, Flett et al. 2007).
Recent research has shown genetic interaction between dab-1 (Ce- DAB-1) and both ehs-
1 (the C.elegans homolog of eps-15) and itsn-1 (the C.elegans homolog of dap160) (Wang,
48 Bouhours et al. 2008). Eps-15 and dap160 (Intersectin) are both scaffold proteins functioning in synaptic vesicle endocytosis. dab-1 null mutants, exhibit both uncoordinated locomotion and mild
defects in growth and egg laying. Although both the itsn-1 and ehs-1 null mutants have normal
development and locomotion, the itsn-1; dab-1 and ehs-1; dab-1 double mutants exhibit severe
and progressive paralysis and subsequent lethality in the larval stage of development. These
results suggest that Ce-DAB1, Intersection and Eps15 may function in the same pathway.
In our studies, a forward genetic screen for modifiers of the TS paralytic phenotype of
cacTS2 led to the identification of a recessive mutation in Drosophila dab. Chapter 3 summarizes
the results from characterization and analysis of this mutant.
49 Chapter 2
Materials and Methods
2.1. Drosophila Fly stocks
cpx1257, dap160EC1, dap160130and syt1425 were generated in our laboratory (details are provided in
the result section). The stocks Iso2, Iso3, dabEC1, cacTS2, shiTS1, and Appl-GAL4 were all from
our laboratory stock collection. cpxSH1 and sytAD4 were kindly provided by Troy Littleton
(MIT, Cambridge, MA) and Tom Schwarz (Harvard University), respectively. SNAP-25TS and the dap160 alleles, dap160Δ1 and dap160Δ6 were gifts from David Deitcher (Cornell University,
Ithaca, NY) and Hugo Bellen (Baylor College of Medicine, Houston, TX), respectively. UAS-
Chc-EGFP was provided by Simon Bullock (University of Cambridge, Cambridge, United
Kingdom). For use as endosomal markers, UAS-GFP-Rab5, UAS-GFP-Rab7, and UAS-GFP-
myc-2xFYVE were obtained from Marcos González-Gaitán (University of Geneva, Geneva,
Switzerland) and UAS-Rab11-GFP was provided by the Bloomington Stock Center. UAS-
Dap160 was provided by Graeme Davis (UCSF, San Francisco, CA). UAS-CPX 3B was
generously provided by Troy Littleton. In addition, the following chromosomal aberrations and
transgenic lines were obtained from the Bloomington Stock Center: Df(3L)Exel6130,
P{EPgy2}DabEY10190, UAS-dab, UAS-EGFP-Clc, Df(3R)Exel6140, Df(2L)Exel6277 and
Df(2L)Exel6047. CHC-RNAi, cpx-RNAi transgenic lines were from the Vienna Drosophila RNAi
Center. WT flies used were Canton-S (CS).
As indicated in Figure 2.2, the following strains were used to establish transgenic lines:
1118 1 Xa 1 1 1 (1) w , (2) w ; SM5; TM3 /ap , (3) w , Df (1) sd / FM7C, (4) w ; Sco/CyO and (5) w ;;
TM3/TM6B.
All the stocks and crosses were cultured on a conventional cornmeal-molasses-yeast medium at
20°C. The ingredients for fly medium: 1000ml Molasses, 108g Agar, 1000ml Cornmeal, 400ml
50 Brewers Yeast, 40ml Propionic Acid, 90ml 20% Tegosept (p-methoxy benzoic acid) in 95%
ethanol and 9L H O. 2
2.2. Behavioral analysis
Behavioral analysis was performed as described previously (Dellinger, Felling et al. 2000;
Kawasaki, Felling et al. 2000; Kawasaki, Collins et al. 2002; Brooks, Felling et al. 2003). Two-
day-old flies reared at 20ºC were tested in groups of six for TS behavioral phenotypes. The test
was conducted in empty transparent plastic vials sealed by rayon plugs. These vials were
submerged in a water bath and heated to the testing temperature. Once the desired temperature
was reached, flies were transferred into the pre-heated vials and their behavior was observed for
motor defects/paralysis. The time point at which three flies failed to stand was recorded as the
time for 50% paralysis. Five groups of flies (n=5) were tested and their mean was recorded as
average 50% paralysis time. Behavior of each group was observed for up to 20 minutes and 3ml
of water was added to the rayon plug after 5 minutes to prevent the flies from dehydrating.
2.3 Molecular Biology
2.3.1 Generation of DNA constructs carrying UAS-EGFP-cpx and UAS-EGFP-cpx1257
Drosophila cpx cDNA in pOT2 vector (clone ID GH27718, GenBank Accession number:
AY121629) was obtained from the Drosophila Genomics Research Center. The cpx ORF was
PCR amplified from this cDNA using pfu polymerase, for both EGFP-CPX and EGFP-CPX1257
DNA constructs. To make EGFP-CPX construct, the primer set used to amplify the cpx ORF are listed in Table 2.1. After PCR amplification, the cpx PCR products were sequentially digested with BglII and KpnI and the digested fragments were gel purified. The digested cpx PCR product was ligated with a pBluescriptII SK- vector fragment (also digested with BglII and KpnI) containing EGFP ORF without stop codon. The ligation products were transformed into DH5α competent cells by heat shock method. Single colonies were cultured in 2XYT liquid medium
(16g tryptone, 10g yeast extract and 5g NaCl per liter) containing Ampicillin at 37ºC with
51 constant shaking for 15 hours. Plasmid DNA was extracted from the cultures using a standard
alkaline lysis protocol. Diagnostic restriction digests was performed to verify insertion of the
correct PCR products. After verifying the correct clone, single colonies of that clone were
cultured in 2XYT liquid medium containing Ampicillin at 37ºC overnight and the next day,
plasmid DNA was prepared using kit (Stratagene). Diagnostic digestion was performed to verify
the clone and quantify it. The plasmid DNA prepared from the kit was sent to the Sequencing
center at University Park Campus, Penn State University. After sequence confirmation, the
EGFP-CPX fragment was shuttled into the P-element based transformation vector, pUAST, using
NotI and KpnI sites. The EGFP-CPX clone in pUAST was cultured in 100 ml 2XYT liquid medium at 37ºC for 16 hours. Plasmid DNA was prepared for generation of transgenic flies using a midiprep kit (Qiagen) and was quantitated by performing a series of restriction enzyme digests.
A similar cloning strategy was followed to make EGFP-CPX1257construct, except that the initial
PCR amplification was performed using Cpx_F_BglII as the forward primer and CpxM_R_KpnI as the reverse primer (Table 2.1). This reverse primer was generated to introduce a stop codon in
place of the last amino acid codon in the cpx ORF.
2.3.2 Generation of DNA constructs carrying UAS-dab-EGFP
dDAB contains a conserved N-terminal PTB domain, thus we decided to tag dab with
EGFP at the C-terminus. The dab-EGFP DNA construct was generated by the assembly of five dab ORF fragments (Figure 2.1) cloned in multiple steps. Each of the dab ORF fragments, generated with primer sets as listed in Table 2.2, were cloned separately into the pBluescript SK- vector. The templates to amplify the five fragments were: dabA- cDNA from Iso3; dabB1 and B2- genomic prep from Iso3; dabC and D- genomic prep from UAS-Dab. Genomic DNA was prepared using the protocol from Dr. Zhi-Chun Lai’s lab that is routinely used in Dr. Ordway’s lab. cDNA was prepared from mRNA as described below.
52 Extraction of messenger RNA (mRNA) from fly heads
mRNA was obtained to clone the dab sequence. The mRNA was prepared in a
conventional way. For each sample, 50 to 100 fly heads were collected and transferred into a 1.5
ml eppendorf tube on ice. They were homogenized in homogenization buffer (3M LiCl and 6M
Urea) (100 μl for ~50 heads) using an RNAase-free plastic pestle. Heads were homogenized for
thirty seconds and returned to ice for one minute. This process was repeated for a few times until
the heads were well homogenized. Homogenates were incubated on ice for thirty minutes and
then at -20°C for at least two hours. The samples were centrifuged at 14,000 rpm for fifteen
minutes at 4°C and the pellet was resuspended in resuspension buffer (10mM Tris·HCl, pH 7.6
and 0.5% SDS) (200 μl resuspension buffer for 50 heads). In order to remove the remaining
proteins, resuspended samples were mixed thoroughly with 200 μl of phenol: chloroform:
isoamyl alcohol (25:24:1) and centrifuged at 14,000 rpm for fifteen minutes at 4°C. The top
aqueous layer of each sample was extracted with 200μl of chloroform: isoamyl alcohol (24:1) and spun at 14,000 rpm for fifteen minutes at 4°C. The RNA was then precipitated by mixing the aqueous layer with 1/10 volume of 4M NaCl and three volumes of 100% ethanol, at -20°C. After overnight incubation at -20°C, RNA was spun down at 14,000 rpm for fifteen minutes at 4°C and washed with 200 μl 70% ethanol and air dried. The RNA pellet was then resuspended in 13.75μl of DEPC H2O (3ml DEPC in 1.5L distilled H2O.
Synthesis of the first strand complementary DNA (cDNA) from mRNA
The first strand cDNA was synthesized with BRL MuLV reverse transcriptase. For each sample, 13.75μl RNA (from 50 heads or from 100 heads with DNAase treatment) was mixed with 1.25μl of RNAsin (Promega, 40U/μl), 5μl of 10μM random primers (or 3’ specific primers),
10μl of 5mM dNTPs, 10μl of 5X MuLV buffer, 5μl of 0.1M DTT and 5 μl of BRL MuLV reverse transcriptase. This mixture was incubated at 37 °C for an hour. After incubation, 1: 20 dilution was used as a template in PCR.
53 After cloning the dab ORF fragments separately into the pBluescript SK- vector, sequencing was carried out to confirm the clones. The primers used for sequencing the dab ORF are listed in Table 2.3. These dab ORF fragments were then assembled into a single clone containing a complete dab ORF without the stop codon in pBluescript SK- vector carrying the
EGFP ORF. This clone was subsequently shuttled into the KpnI and NotI sites of the transformation vector, pUAST (Brand and Perrimon 1993).
54 Table 2.1: Primers for generation of EGFP tagged cpx DNA constructs
DNA Primer Sequence Construct Cpx_F_BglII 5' GAAGATCTATGGCGGCCTTCATAGCTAAG EGFP-cpx 3' Cpx_R_KpnI 5' GGGGTACCTCACTGCATGACACATTTTC 3'
Cpx_F_BglII 5' GAAGATCTATGGCGGCCTTCATAGCTAAG EGFP-cpx1257 3' CpxM_R_KpnI 5' GGGGTACCTCACATGACACATTTTCCCT 3'
55
Figure 2.1 dab ORF. Five dab ORF fragments assembled into a clone containing the entire dab ORF without the stop codon. This ORF was tagged with EGFP at the N-terminus.
Table 2.2: Primers for generation of dab-EGFP DNA construct
dab Primer Sequence Fragment DabA Fwd 5’ GCGGCCGCATGGTCAAGTCCCTGGTGGC 3’ dabA DabC Lisa 5’ GGAGCGGGCAGCGGATCGGG 3’
DabB Fwd 5’ CGAACACCATTATTGCACAG 3’ dabB1 Dab3178 5’ GAGTTGGTGCATCGTTAAAG 3’
Dab3073 5’ AGCAAACTGAGCACCATGAC 3’ dabB2 Dab4740 Rev 5’ GTGATGATACCAGCTGGGGG 3’
DabC Fwd 5’ CGTGTCCCAACTCATCGATAC 3’ dabC DabC Rev 5’ CCACTCATCATCAGGCCACC 3’
DabD Fwd 5’ GCTAGGCGGTGATGTGGTGCTG 3’ dabD DabD Revnew 5’ GGAATTCCTCACATGTTTTCATCGAACTTTG 3’
56
Table 2.3: Primers for sequencing of dab
dab Primer Sequence Fragment dabA Dab701 5’ TTAGCACCACGAATGGAACG 3’
T3 5’ AATTAACCCTCACTAAAGGG 3’
dabB1 Dab2334 5’ TGCCTCTGTTTCACTCAACG 3’
Dab2370 5’ TGATGTGATCTCCAGTATAAG 3’
dabB2 T3 5’ AATTAACCCTCACTAAAGGG 3’
Dab6606 5’ CAGAGGGCTTGGAGGTGAAC 3’
Dab7365 5’ TCAGCACGTCGCCGATTCCC 3’
dabC DabF 5’ GTCCAAGCTACGATTTCGAC 3’
DabG 5’ ATCGCGACTATCTAGGCTGC 3’
dabD Dab6484 5’ ACATTATCGTCAAATCGCAG 3’
57 2.3.3 Generation of transgenic strains
w1118 flies (~ 300 flies) were transferred to a cage placed on an egg laying plate (4mm
high, 90ml unsuphured molasses, 22g select agar, 250μl Tegosept stock and 556ml distilled H2O).
Embryos (younger than 30-minutes) were collected from the egg laying plate, rinsed with water in a small sieve and aligned in a row of 20-25 embryos on the surface of a 2% agar cube
(20X13X 5mm). The embryos were then transferred to the surface of a micro cover slip (18 x 18
mm) with a glue strip on one side, by pressing the glue side facedown onto the embryos. The
micro cover slip was then placed on a glass slide (25 x 75 mm) and the embryos were covered
with oil (1:19 ratio of Halocarbon 27 oil to Halocarbon 700 oil, Sigma). 10X injection buffer
(50mM KCl, 1mM NaH2PO4 and 1mM Na2HPO4 pH 6.8) was added to the plasmid DNA prepared by Qiagen midiprep kit to a final concentration of 1X. The injection solution contained
0.4 to 0.8μg/μl plasmid DNA, 0.2μg/μl helper plasmid carrying a P-element transposase gene and
1:10 diluted green food color (McCormick). Injection needles were made by pulling glass capillary tubes (Sutter, Novato, CA) with an electrode puller.
Prior to injection, needles were each filled with ~1 μl of the injection solution. A glass side containing the glued embryos was placed on an upright microscope (Zeiss, Germany). A micro-manipulator was used to control the movement of the needle. The injection solution containing the DNA was then injected into the posterior tip of each embryo before cellularization.
Cellularized embryos were destroyed by piercing with the needle. After injection, micro cover slips were placed (embryo side facing up) on a fly food plate (100mm diameter and 13mm high) with upto 8 cover slips in a plate. Food plates were kept at room temperature for embryos to hatch. After two days, F0 larvae (F0 in Figure 2.2) were transferred into fresh fly food vials and kept at room temperature. Subsequently, the F0 adult flies were crossed individually according to the scheme in Figure 2.2. Transgenic lines were selected and established as stocks.
58
59
2.3.4 Identification of the molecular lesions in dap160 mutants
Genomic DNA was prepared for sequencing of dap160 mutations from homozygous mutant flies and Iso2 flies (parent strain). The genomic DNA was PCR amplified using the forward primer, dap160A, and the reverse primer, dap160B (Table 2.4). The PCR products were gel purified, quantified and sent for sequencing using primers dap160A, dap160B and dap160C
(Table 2.4). All sequencing was carried out at the sequencing facility, Penn State University,
University Park, PA. Sequences from the mutants were compared with those from Iso2 and
molecular lesions in dap160EC1 and dap160130 were identified.
2.3.4 Identification of the molecular lesion in syt1 mutant
Genomic DNA was prepared from homozygous mutant flies (syt1425/syt1425) and Iso2
flies (parent strain). The genomic DNA was PCR amplified using the forward primer, SytUT1_
Fwd, and the reverse primer, SytUT2_ Rev (Table 2.5). The PCR products were gel purified,
quantified and sent for sequencing using the primers in Table 2.5. Sequences from the mutants
were compared with those from Iso2 and molecular lesion in syt1425 was identified.
60
Table 2.4: Primers for identification of dap160 mutations
Primer Sequence dap160A (Fwd) 5’ GTGCGAGGATCTGTACAAGG 3’ dap160B (Rev) 5’ GCAATGACCTTGTCTGCATG 3’ dap160C (Fwd) 5’ CGGTAGAATCTCTTGATCAG 3’
Table 2.5: Primers for identification of the syt1 mutation
Primer Sequence
SytUT1_ Fwd 5’ TCATGTGAGACTAGCGAGCC 3’
SytUT2_ Rev 5’ TCTGACGGTTCGCGTACATG 3’
Syt400 5’ TAATCTTCTTCTGTGTGCGG 3’
61 2.4 Biochemistry
2.4.1 Antibodies
Immunoprecipitation: Immunoprecipitation of EGFP tagged CPX was carried out using 2-5μg
polyclonal anti-GFP antibody (Invitrogen).
Western blotting: These studies used the following primary antibodies: rabbit polyclonal anti-
dDAB antibody (1:500), generously provided by Eric Liebl (Denison University, Granville, OH);
a second rabbit anti-dDAB antibody (1:10,000), generated and affinity purified in collaboration
with Linton Traub (University of Pittsburgh, Pittsburgh, PA); rabbit polyclonal anti-CPX
antibody (1:10,000), kindly provided by Troy Littleton (MIT, Cambridge, MA); mouse
monoclonal anti-GFP, JL-8 (1:500), from Clontech, CA; mouse monoclonal anti-SYX antibody
(crude serum), 8C3 (1:10), from DSHB Hybridoma, University of Iowa, IO; rabbit anti-NSYB
antibody (1:30), developed in Dr. Richard Ordway’s lab (Penn State University, State College,
PA); rabbit anti-SNAP-25 antibody (1:2,000), a gift from Dr. David Deitcher (Cornell University,
Ithaca, NY); mouse monoclonal anti-ACTIN antibody (crude serum), JLA20 (1:10 ), from DSHB
Hybridoma, University of Iowa, IO. Tubulin served as a loading control and was detected using a monoclonal anti-acetylated α TUBULIN (Sigma, MO) at a dilution of 1:2,000,000.
As secondary antibody incubation, horseradish peroxidase (HRP)-conjugated anti-mouse
(1:5,000) and anti-rabbit antibodies (1:10,000) (Amersham biosciences, Arlington Heights, IL)
were used.
Immunocytochemistry: These studies used the following primary antibodies: rabbit anti-dDAB
(1:2,500) generated and affinity-purified in collaboration with Linton Traub (University of
Pittsburgh, Pittsburgh, PA), rabbit anti-SYT Dsyt CL1(1:5,000) (Noreen Reist, Colorado State
University, Fort Collins, CO), mAb nc82 anti-BRUCHPILOT (1:50) (Erich Buchner, University of Wurzburg, Wurzburg, Germany), rabbit anti-DYNAMIN (1:2,000) (Mani Ramaswami,
University of Dublin, Dublin, Ireland), rabbit anti-DAP160 (1:1,000) (Graeme Davis, University
62 of California, San Francisco, CA), rabbit anti–α-ADAPTIN (1:2,500) (Marcos Gonzalez-Gaitan,
University of Geneva, Geneva, Switzerland), rabbit anti-neuronal SYNAPTOBREVIN (1:10),
and rabbit anti-vesicular glutamate transporter (1:2,500) (Hermann Aberle, Max Planck Institute,
Tubingen, Germany), rabbit anti-CPX (1:10,000) (Troy Littleton, MIT, Cambridge, MA), Cy5-
conjugated rabbit anti-HRP (1:200) was obtained from Jackson Immunoresearch Laboratories.
Secondary antibodies were obtained from Invitrogen and included Alexa Fluor 488 or
568-conjugated anti-mouse IgG (1:200), Alexa Fluor 488 or 568-conjugated anti-rabbit IgG
(1:200) and Alexa Fluor 647-conjugated anti-mouse IgG (1:200).
2.4.2 Western Blotting
Western Blot analysis was performed as described previously (Kawasaki, Zou et al. 2004).
1) Sample preparation: Flies were collected in eppendorf tubes and rapidly frozen in liquid nitrogen. Fly heads were then separated from the body by repetitive vortexing and freezing in liquid nitrogen. Fly heads were then collected under a Leica MS5 microscope and homogenized in 1X SDS sample buffer (125 mM Tris-HCl/SDS pH 6.8, 10% glycerol, 2% SDS, 1% mercaptolethanol and 0.5% bromophenol blue) with a Teflon pestle. These homogenized heads were boiled for three minutes followed by centrifugation at 14,000 rpm for 30 seconds.
2) Separation of proteins by size using SDS-PAGE (Figure 2.3.A): The supernatant of the homogenate was loaded to a SDS-PAGE (Sodium Dodecyl Sulfate Polyacrylamide Gel
Electrophoresis) gel. To observe the dDAB protein, homogenate equivalent to one head was loaded onto a 9% SDS-PAGE gel. For observing CPX, 0.5 heads were loaded onto a 12% SDS-
PAGE gel. A low range prestained SDS-PAGE size standard (Bio-Rad Laboratories, CA) was loaded as well, to allow estimation of protein size. The gel was run in 1X SDS electrophoresis buffer (25 mM Tris, 190 mM glycine and 1% SDS) at a current of 15 mA per gel for about an hour.
63 3) Transfer of the separated proteins to a solid support like nitrocellulose/PVDF membrane
(Figure 2.3.B): After the proteins were separated on the gel based on their molecular mass, they were transferred to nitrocellulose membranes (Pall) in transfer buffer (25mM Tris, 190mM glycine and 20% methanol) with a constant voltage of 14 volts at room temperature overnight
(14-16 hrs).
4) Blocking and incubation of the membrane with a target protein specific primary antibody:
After transfer, the membranes were blocked with 5% non-fat milk in 1XPBS (Brown’s lab,
171mM NaCl, 4mM Na2HPO4 3.4mM KCl, and 1.84mM KH2PO4 pH to 7.5) at room temperature
for one hour. Blocking is followed by incubation with specific primary antibody solution at 4ºC
with constant agitation overnight.
5) Addition of HRP-conjugated secondary antibody to the membrane: The following day the
membranes were washed three times with 1XPBST (0.1% Tween-20 in 1X PBS) for ten minutes each and incubated with HRP-conjugated secondary antibody, which is specific for the primary antibody, for two hours with constant agitation at room temperature.
6) Detection of the target protein by chemiluminescence reaction (Figure 2.3.C): We develop our Western Blots by detecting the HRP-conjugated secondary antibodies using an enhanced version of chemiluminescence, ECL kit (Amersham Bioscience, IL). The membranes, after the secondary antibody incubation, were washed three times with 1XPBST for ten minutes each.
Solution A and solution B provided in the ECL kit were mixed at the ratio of 40:1. Membranes were allowed to react with the mixture for five minutes in the dark. After the reaction, membranes were either scanned with a phosphor image scanner (Storm 860, GE Molecular Dynamics) or exposed to X-ray films (Amersham Bioscience, IL). Proteins sizes were estimated by comparison to dual size standards or low molecular weight size standards (Bio-rad Laboratories, CA). A α-
Tubulin antibody (Sigma, MO) was used as loading control.
64
Figure 2.3 Overview of Western Blotting (Image courtesy: http://static.abdserotec.com/Lit- pdfs/Brochures1/westernblotbook.pdf) (A) Separation of proteins by electrophoresis. (B) Transfer of proteins from gel to a blotting membrane. (C) Detection of target protein.
65 2.4.3 Co-immunoprecipitation (Co-IP):
Co-immunoprecipitation was performed to investigate other proteins interacting with
CPX, as described previously (Zou, Yan et al. 2008; Yu, Kawasaki et al. 2011). Wild-type
(Canton-S) flies and flies expressing the EGFP-CPX transgene were transferred into 2ml cryo tubes, eight for each genotype with each tube containing ~ 50 flies. The flies were frozen by immersing the tubes in liquid nitrogen. The flies were decapitated and fly heads were collected as previously described in the sample preparation in Section.2.4.2. Three hundred and sixty heads for each genotype were transferred into separate 1.5ml eppendorf tubes on ice and homogenized with a Teflon pestle in 600μl fresh working lysis buffer [(1% CHAPS from Sigma, 20mM Tris·Cl pH7.5, 10mM EDTA, 120mM NaCl, 50mM KCl, 2mM DTT and 1:100 protease inhibitor cocktail from Sigma)]. The homogenization was performed on ice to prevent protein degradation.
This was followed by centrifugation of the homogenized samples at 14,000 rpm for 10 minutes at
4ºC and the pellet was ground one final time with the pestle to ensure thorough homogenization.
The homogenate was then incubated for thirty minutes with constant rotation at 4ºC followed by centrifugation at 14,000 rpm for 10 minutes at 4ºC. 550μl of the supernatant, which is the final head lysate, was transferred into a fresh 1.5ml tube, precoated with stock lysis buffer, and precleared with 50 μl of 50% slurry of protein A-sepharose beads for one hour with constant agitation at 4ºC. After preclearing, the beads were pelleted at 5000 rpm for 10 minutes at 4ºC.
20μl of the precleared lysate was saved as the pre-immunoprecipitation (pre-IP) input sample.
500μl of the remaining lysate (~300 fly heads) was incubated with 2.5μl of anti-GFP polyclonal antibody (Invitrogen) and 50 μl of 50% slurry protein A-sepharose beads precoated with BSA at
4ºC with constant agitation for two hours. After this step, the beads were pelleted by centrifugation at 5000 rpm for 1 minute at 4ºC and washed by working lysis buffer five times.
During each wash, beads were incubated with 1ml of ice cold working lysis buffer for five minutes with constant agitation at 4ºC and recovered by centrifuging at 5,000 rpm for 1 minute at
66 4ºC. After washing, 40μl of 1X SDS sample buffer was added to the beads and proteins were
eluted from the beads by boiling for three minutes. 20μl of 2X SDS sample buffer was added to
the pre-IP input samples and boiled for three minutes. After boiling, the pre-IP input sample was
diluted to 0.05H/μl by addition of 200μl of 1X SDS sample buffer. To examine IP of the EGFP-
CPX with anti-GFP, 10μl of the pre-IP input sample (~ 0.5 heads) and 4μl (~30 heads) out of the
40μl IP sample were loaded to a 12% SDS-PAGE gel. To examine co-IP of the SNARE proteins
with EGFP-CPX, 10μl of the pre-IP input sample and 12μl (~90 heads) out of the 40μl IP sample were loaded into a 12% SDS-PAGE gel. Later, western blotting was performed as described previously. Similar procedure was carried out to compare SNARE protein binding of EGFP-CPX and EGFP-CPX1257.
For co-IP experiments investigating interactions between SNARE proteins and EGFP-
CPX in the wild-type or comtST17 background, flies were transferred into separate 15 ml centrifugation tubes and completely immersed into a 38 °C water bath. After a 10-minute heat shock at 38 °C, the flies were immediately frozen in liquid nitrogen. The rest of the procedure was performed as described above.
2.4.4 Immunocytochemistry
Adult Drosophila flies were dissected in 1.8mM Ca2+ and 4mM Mg2+ saline solution
(128mM NaCl, 2mM KCl, 1.8mM Ca2+, 4mM Mg2+, 5mM Hepes and 36mM sucrose, pH 7.0) as previously described (Kawasaki, Mattiuz et al. 1998). The lateral surface of dorsal longitudinal
flight muscle (DLM) were exposed by dissection and fixed in the saline solution containing 4%
paraformaldehyde for 30 minutes at room temperature. The preparations were then washed with
saline solution, PBS and PBT (0.2% Triton X-100 in PBS), for 10 minutes each. The washed
preparations were incubated with blocking buffer (5% normal goat serum in 1XPBT) for one hour
at constant agitation, and then incubated with primary antibodies diluted in blocking buffer for
either two hours at room temperature, or overnight at 4°C. This was followed by washing of the
67 preparations three times with PBT, and twice with PBS, for six minutes each. Washed
preparations were incubated with fluorophore-conjugated secondary antibodies and plasma membrane marker, Cy5-conjugated anti-HRP, diluted in 1XPBS containing 5% normal goat serum for two hours. After the secondary antibody incubation, the preparations were washed five times with PBS for six minutes each and mounted in a 1:1 mixture of PBS and glycerol between two cover slips for imaging.
2.4.5 Imaging using a Confocal Microscope
Confocal imaging of mounted DLM preparations was performed using an Olympus
FV1000 confocal microscope (Tokyo, Japan). All images were collected with a PlanApo 60x 1.4 numerical aperture oil objective (Olympus, Japan), gain of 1, offset of 0, zoom of 5 and at and a z-step size of 0.2μm. Acquisition and processing of images were performed with the Fluoview software (Olympus, Japan). Maximum projections of two or three optical sections were generated for display.
2.4.6 Phase Partitioning
We used the phase partitioning method to determine the membrane association of WT
CPX and CPX1257 (Figure 2.4). The protocol used was adapted from Bordier et al., with several modifications (Bordier 1981; Wang and Coppel 2002; Mathias, Chen et al. 2011). Triton X-114
(Sigma) was first preconditioned by dissolving 1 ml of Triton X-114 in 100 ml of ice cold 1X
PBS, with continuous mixing at 4°C for 1 h. This solution was then warmed to 37°C for 6 h until the aqueous and detergent phases were separated. The upper aqueous phase (~90 ml) was discarded and replaced with an equal volume of ice-cold 1X PBS. This condensation procedure was repeated twice to get the final detergent-enriched phase containing ~10% (w/v) detergent.
This was used as a stock (stored at 4°C) to produce all subsequent solutions with PBS.
Flies (~60) were collected in eppendorf tubes and rapidly frozen in liquid nitrogen. Fly heads were then separated from the body by repetitive vortexing and freezing in liquid nitrogen.
68 50 fly heads were then collected under a microscope and homogenized in 200µl of PBS with 1%
TritonX-114 containing protease inhibitors [100µl ice cold precondensed PBS with TritonX-114
10% (w/v) +900 µl ice cold PBS + 10µl protease inhibitor cocktail (Sigma)]. The homogenization
was performed on ice to prevent protein degradation. This was followed by centrifugation of the
homogenized samples at 13,000 rpm for 5 minutes at 4ºC, and the pellet was ground one final
time with the pestle to ensure thorough homogenization. The homogenate was then incubated for
one hour on ice, followed by centrifugation at 14,000 rpm for 30 minutes at 4ºC to remove
insoluble debris. For phase separation, the supernatant was transferred into a fresh tube and
incubated at 37ºC for 15mins (The solution will become cloudy at this stage). This was followed
by centrifugation at 5000g for 15mins at room temperature. The aqueous phase (supernatant) was
then transferred to a fresh tube (~150µl) and stock Triton X-114 solution [~10% (w/v)] was
added to achieve a final detergent concentration of ~1%. The detergent phase, an oily droplet at
the bottom of the tube, was resuspended in 175µl of ice-cold PBS with 0.05% TritonX-114. The separated aqueous and detergent phases were kept on ice for 10 minutes (leave the detergent phase on ice until it clears of detergent micelles). Then the aqueous and detergent phase solutions were warmed up to 37°C for 15 minutes, followed by centrifugation for 5 minutes. The supernatant from the first aqueous phase was transferred to a new tube, while the supernatant from the original detergent phase was discarded. This phase separation procedure was repeated twice for each of the phases to deplete any hydrophilic proteins from the detergent phase and to avoid any membrane protein contamination in the aqueous phase.
After the final phase separation, the proteins were precipitate by adding TCA (100%) to a final concentration of 15% and incubated on ice for 30mins. Then the pellet was recovered by
centrifuging at 14,000rpm for 20 min at 4ºC. The pellet was washed in ice-cold acetone (~300µl)
at 14,000rpm for 10mins at 4ºC. The supernatant was discarded and the pellet was air dried. Both the aqueous and the detergent pellets were resuspended in 50µl of 1X SDS sample buffer and
69 boiled for 3mins. The proteins enriched in the aqueous and detergent phases were then separated
by SDS-PAGE. Later, western blotting was performed as described previously.
Figure 2.4 Phase partitioning protocol.
70
Chapter 3
The following content was published in the Proceedings of the National Academy of Sciences
108(25):E222-9 (2011)
DISABLED Functions in CLATHRIN-Mediated Synaptic Vesicle Endocytosis and Exo-Endocytic Coupling at the Active Zone.
Fumiko Kawasaki, Janani Iyer, Lisa L. Posey, Chichun E. Sun, Samantha E. Mammen, Huaru Yan and Richard W. Ordway Department of Biology and Center for Molecular and Cellular Neuroscience, The Pennsylvania State University, University Park, PA 16802
Author Contributions: Several members of our group have made important contributions to this work. Electrophysiology was performed by Fumiko Kawasaki. Electron Microscopy was done by Chichun Sun and Samantha Mammen. Lisa Posey played a major role in recovery of dabEC1 mutant and its molecular characterization. Huaru Yan provided excellent technical support for several aspects of the project.
3.1 ABSTRACT
Members of the DISABLED (DAB) family of proteins are known to play a conserved role in endocytic trafficking of cell surface receptors by functioning as monomeric CLATHRIN
Associated Sorting Proteins (CLASPs) which recruit cargo proteins into endocytic vesicles. Here we report a Drosophila disabled mutant revealing a novel role for DAB proteins in chemical
synaptic transmission. This mutant exhibits impaired synaptic function including a rapid, activity-
dependent reduction in neurotransmitter release and disruption of synaptic vesicle endocytosis. In
presynaptic boutons, Drosophila DAB (dDAB) and CLATHRIN were highly co-localized within two distinct classes of puncta, including relatively dim puncta which were located at active zones
(AZs) and may reflect endocytic mechanisms operating at neurotransmitter release sites. Finally, broader analysis of endocytic proteins including DYNAMIN supported a general role for
71 CLATHRIN-mediated endocytic mechanisms in rapid clearance of neurotransmitter release sites
for subsequent vesicle priming and refilling of the release-ready vesicle pool.
3.2 INTRODUCTION
Intensive study of chemical synaptic transmission has led to detailed molecular models of synaptic vesicle endocytosis (Jung and Haucke 2007; Smith, Renden et al. 2008; Dittman and
Ryan 2009; Shupliakov and Brodin 2010). Ongoing efforts seek to fully define the underlying molecular components and interactions and their spatial organization with respect to neurotransmitter release sites. The present study reveals novel molecular mechanisms and interactions in synaptic vesicle endocytosis involving a member of the DISABLED family of
Phosphotyrosine Binding (PTB) domain proteins. The Drosophila disabled (dab) gene, the founding member of the disabled gene family, was first identified in a screen for genetic modifiers of abelson (Gertler, Bennett et al. 1989). Although putative dab mutations were subsequently mapped to a different gene (Liebl, Rowe et al. 2003), recent studies have established functional interactions of dDAB and ABELSON (Song, Kannan et al. 2010). Proteins closely related to dDAB, including mouse DAB-2 (mDAB-2) and the single C. elegans family member, ceDAB-1, function as CLATHRIN-associated adaptors in endocytic trafficking of cell surface receptors (Morris and Cooper 2001; Mishra, Keyel et al. 2002; Kamikura and Cooper
2006). These DABs have been shown to interact with CLATHRIN, the AP-2 adaptor complex and other components of the endocytic machinery through conserved binding motifs [(Morris and
Cooper 2001; Mishra, Keyel et al. 2002; Kamikura and Cooper 2006) and see Figure 3.1a].
Notably, mDAB-2, ceDAB-1 and dDAB share a PTB/DAB Homology (DH) domain near the N- terminus (Figure 3.1b) which interacts with specific vesicle cargo proteins and phosphatidylinositol-4,5-diphosphate (PtdIns-4,5-P2) (Yun, Keshvara et al. 2003; Traub 2009).
Neither DAB proteins nor PTB domain interactions have been previously implicated in synaptic vesicle trafficking.
72 The present study reveals a role for DAB proteins in synaptic vesicle endocytosis and extends previous work on interactions of endocytic and exocytic mechanisms in neurotransmitter release. Here we employ a glutamatergic neuromuscular synapse of the Drosophila adult as a model for genetic analysis of molecular mechanisms determining conserved properties of glutamatergic synapse function(Kawasaki and Ordway 2009; Danjo, Kawasaki et al. 2011).
Previous analysis in this model examined the temperature sensitive (TS) DYNAMIN mutant, shibire, and revealed a rapid role for endocytic mechanisms in maintaining neurotransmitter release during synaptic activity (Kawasaki, Hazen et al. 2000; Neher 2010). Subsequent work has
extended these findings and showed that inhibition of DYNAMIN or AP-2 disrupts fast refilling
of the release-ready synaptic vesicle pool (Hosoi, Holt et al. 2009; Wu, McNeil et al. 2009;
Haucke, Neher et al. 2011). The results reported here demonstrate colocalization of CLATHRIN
and DAB at the AZ and provide functional and ultrastructural evidence supporting their
participation in rapid CLATHRIN-dependent endocytic mechanisms which may clear
neurotransmitter release sites for subsequent synaptic vesicle priming and refilling of the release-
ready vesicle pool.
3.3 RESULTS
Genetic and Molecular Characterization of a New dab Mutant. The dabEC1 mutation was
recovered in a classic ethane methyl sulfonate (EMS) mutagenesis screen for modifiers of the
TS presynaptic voltage-gated calcium channel α1 subunit mutant, cacTS2 (Kawasaki, Felling et al.
2000; Kawasaki, Collins et al. 2002; Brooks, Felling et al. 2003) that had been recovered
previously as a modifier of comatose (Dellinger, Felling et al. 2000). dabEC1 is an enhancer of cacTS2 (Figure 3.S1) but also exhibited phenotypes in a cac+ genetic background. The isolated dabEC1 mutation produced TS impairment of motor behavior and, as described in the following section, a clear synaptic phenotype. Through genetic mapping, a small pool of candidate genes was selected and subjected to sequence analysis. A molecular lesion was identified in the dab
73 gene, which introduced an early stop codon within the conserved N-terminal PTB/DH domain
(Figure 3.1b) and Western analysis confirmed that dDAB expression was not detected in this mutant (Figure 3.1c). Furthermore, an independently generated mutant allele, dabEY10190, failed to
complement dabEC1. Finally, using the GAL4-UAS system (Brand and Perrimon 1993), neural expression of an available UAS-dab transgene (Gertler, Hill et al. 1993) was found to rescue the dabEC1 behavioral phenotype. These findings define a new dab mutant and reveal a role for dDAB
in neural function.
dDAB Functions in Synaptic Transmission. Excitatory Postsynaptic Currents (EPSCs) were
recorded at dorsal longitudinal muscle (DLM) neuromuscular synapses of the Drosophila adult
(Kawasaki and Ordway 2009). At wild-type synapses, 1Hz train stimulation produces modest
short-term synaptic depression at both 20 and 33 °C (Figure 3.2a and b and Figure 3.S2a). In
dabEC1 at the restrictive temperature of 33 °C, the first stimulus produced a wild-type EPSC amplitude and waveform, indicating that basic properties of synaptic function are preserved in this mutant (Figure 3.2a). In contrast, subsequent 1 Hz stimulation produced a clear activity- dependent reduction in EPSC amplitude with respect to wild-type (Figure 3.2a and b).
Furthermore, the dabEC1 synaptic phenotype was rescued by neural (presynaptic) expression of a
UAS-dab-EGFP transgene (Figure 3.2b), demonstrating that dab mutant synapses exhibit an
activity-dependent reduction in neurotransmitter release. The dab synaptic phenotype was not
strictly TS as it was also observed at 20 °C (Figure 3.S2a and b), consistent with the severe nature
of the dabEC1 molecular lesion. Moreover, during prolonged stimulation, dab mutant synapses exhibit strong depression but sustain a reduced steady-state level of neurotransmitter release
(Figure 3.S2c and Discussion). The preceding results demonstrate an important role for dDAB in
neurotransmitter release.
dDAB and CLATHRIN are colocalized at AZs and play similar functional roles in synaptic
transmission. The established role for DABs in CLATHRIN-mediated endocytosis, together with
74 the activity-dependent reduction in neurotransmitter release observed at dab mutant synapses,
suggested a novel function for DAB proteins in synaptic vesicle endocytosis. Accordingly, efforts were made to examine the distribution of dDAB within presynaptic boutons. However, immunocytochemical analysis was ineffective using the previously reported anti-dDAB antibody
(Gertler, Hill et al. 1993) despite the fact that Western blotting with the same antibody clearly detected endogenous dDAB in the nervous system (Figure 3.1c). Fortunately, a new anti-dDAB
was antibody generated in collaboration with Dr. Linton Traub (University of Pittsburgh School of Medicine). This antibody recognized the dDAB protein (Figure 3.S3) and was suitable for immunocytochemical analysis of endogenous or transgenically expressed dDAB at DLM neuromuscular synapses. These studies revealed a striking distribution of dDAB within presynaptic boutons including two classes of puncta with different spatial relationships to active zones.
One class of dDAB puncta, referred to as non-Active Zone (non-AZ), included relatively bright signals which were either adjacent to or spatially separated from the AZ [Figure 3.2c-g
(arrows in d) and w]. In addition, closer inspection revealed a weaker class of dDAB puncta which were colocalized with the AZ (co-AZ) [Figure 3.2c-g (arrowheads in d) and w] and thus raised the possibility that dDAB participates in endocytic processes at neurotransmitter release sites. Note that non-AZ puncta did not surround AZs as might be expected for a periactive zone
(PAZ) marker but rather formed distinct puncta that were often separated from AZs. This pattern of dDAB localization was further established by imaging native EGFP fluorescence of dDAB-
EGFP (Figure 3.2h-l). Importantly, the distribution of CLATHRIN LIGHT CHAIN (CLC) within
presynaptic boutons exhibited very similar classes of non-AZ and co-AZ puncta [Figure 3.2m-q
(arrows and arrowheads in n) and x), which were highly colocalized with dDAB (Figure 3.2r-v).
A similar distribution was also observed for CLATHRIN HEAVY CHAIN (CHC), the other component of the CLATHRIN lattice (Figure 3.S4b). Strong colocalization suggests a close
75 interaction of DAB and CLATHRIN in synaptic vesicle endocytosis and appears to be unusual
among endocytic proteins including DYNAMIN and DAP160 (INTERSECTIN) which are
distributed more broadly (Figure 3.S5a-n). The distribution of the CLATHRIN- and DAB- associated adaptor protein AP-2 overlapped with that of CLATHRIN and included co-AZ and non-AZ puncta (Figure 3.S5o-u). The subcellular domain or structure corresponding to non-AZ puncta has not been identified, despite double labeling studies with several markers for intracellular membrane compartments (Figure 3.S5v-z and Methods), and thus this remains a very interesting issue for further investigation (Discussion). The observed colocalization of dDAB and
CLATHRIN at the AZ supports a role for CLATHRIN-mediated endocytic mechanisms at neurotransmitter release sites. Although the co-AZ puncta were relatively weak, they represent true AZ localization as confirmed by direct comparison of co-AZ signals in dDAB-EGFP and
CLC-EGFP preparations with controls lacking EGFP expression (Figure 3.S4c and d).
The relationship of dDAB and CLATHRIN was further explored through functional analysis of DLM neuromuscular synapses following knockdown of CHC through neural expression of a CHC RNAi trangene. These synapses exhibited an activity-dependent reduction in EPSC amplitude resembling that of the dabEC1 mutant in response to 1 Hz (Figure 3.2y) or 5 Hz
(Figure 3.2z) stimulation. Taken together, the preceding observations suggest physical and
functional interactions of dDAB and CLATHRIN in neurotransmitter release as well as a role for
these mechanisms at the AZ.
Ultrastructural analysis reveals disruption of synaptic vesicle endocytosis and persistence of
active zone-associated and docked vesicles at dab mutant and CLATHRIN RNAi synapses.
To complement the preceding findings, ultrastructural analysis of synaptic vesicle trafficking was
carried out at adult DLM neuromuscular synapses of dab mutant or CHC-RNAi preparations. To
facilitate detection of phenotypes, these studies were performed at synapses exposed to a
restrictive temperature of 33 °C and stimulated at 20 Hz. For comparison, unstimulated
76 preparations were examined as well. The basic ultrastructure of dab and CHC RNAi synapses,
including the presynaptic dense body (t-bar), synaptic cleft and postsynaptic density, was similar
to that of wild type. However, synaptic vesicle endocytosis was clearly disrupted at stimulated
dab and CHC-RNAi synapses (Figure 3.3a) as indicated by ultrastructural phenotypes
characteristic of those previously reported for defects in CLATHRIN-mediated synaptic vesicle
endocytosis [c.f. (Koenig, Kosaka et al. 1989; Takei, Mundigl et al. 1996; Shupliakov, Low et al.
1997; Koh, Korolchuk et al. 2007; Heerssen, Fetter et al. 2008; Kasprowicz, Kuenen et al. 2008)].
First, synaptic vesicle size was increased in both mutants with respect to wild-type (Figure 3.3a and b), consistent with a role for dDAB (along with CLATHRIN) in synaptic vesicle formation.
Furthermore, dab boutons exhibited large cisternae-like membrane structures rarely observed in wild type and these were even more prevalent in CHC-RNAi preparations (Figure 3.3a and c).
Finally, and importantly, the number of docked synaptic vesicles at either dab or CHC-RNAi synapses persisted or was increased with respect to wild-type (Figure 3.3d), indicating that synaptic vesicles are present at release sites but unable to fuse efficiently with the plasma membrane. Further analysis determined the distributions of synaptic vesicles at progressively increasing distance from the AZ. Three distinct areas were defined by concentric shells centered at the base of the presynaptic dense body (Areas 1, 2 and 3; Figure 3.3e). Notably, the density of
synaptic vesicles close to the AZ (Area1) at dab mutant and CHC-RNAi synapses was similar to or higher than that of wild-type (Figure 3.3f). At a greater distance from the AZ (Area 3), dab
mutant and CHC-RNAi synapses exhibited a reduced density of vesicles relative to wild type.
These findings suggest that enhanced synaptic depression following loss of endocytic function
does not result from reduction in available synaptic vesicles at release sites. Rather, it appears that
endocytic mechanisms are required at the AZ for efficient release of neurotransmitter from
available vesicles (Kawasaki, Hazen et al. 2000; Hosoi, Holt et al. 2009; Wu, McNeil et al. 2009).
77 With regard to synaptic activity, stimulated and unstimulated wild-type synapses exhibited very similar ultrastructure with the notable exception of a significant reduction in docked vesicles at stimulated synapses [1.93 ± 0.17 (n=40) and 3.92 ± 0.20 (n=64) vesicles/AZ at stimulated and unstimulated synapses, respectively (p < 0.01)]. Although synaptic stimulation generally produced a more severe ultrastructural phenotype at mutant synapses, unstimulated preparations also exhibited significant changes with respect to wild type, for example in synaptic vesicle size and the prevalence of membrane cisternae. A summary table of ultrastructural data from stimulated and unstimulated preparations is provided in Table 3.S1.
A rapid functional requirement for CLATHRIN-mediated endocytic mechanisms may reflect impaired refilling of the release-ready vesicle pool. Previous studies of the TS
DYNAMIN mutant, shibire, indicated a rapid role for DYNAMIN in sustaining neurotransmitter release during synaptic activity (Kawasaki, Hazen et al. 2000). In the present study, disruption of dDAB or CLATHRIN function was not conditional. However, as observed in the shibire
(DYNAMIN) TS mutant, each produced activity-dependent reduction in EPSC amplitude with rapid onset during train stimulation (Figure 3.4a). Such an activity-dependent phenotype following disruption of endocytic mechanisms might reflect depletion of synaptic vesicles which limits refilling of the release-ready vesicle pool, however this was not the case in shibire
(DYNAMIN) (Kawasaki, Hazen et al. 2000) or at dab or CHC-RNAi synapses (Figure 3.3).
Rather, the ability of available vesicles to release neurotransmitter at active zones appears to be impaired in all three mutants and thus additional analysis of synaptic ultrastructure in shibire
(DYNAMIN) was carried out to further examine its relationship to the dab and CHC-RNAi phenotypes.
As in wild type, dab and CHC-RNAi, ultrastructural analysis of shibire (DYNAMIN) synapses was performed at a restrictive temperature of 33 ˚C following 20-Hz stimulation. For comparison, unstimulated preparations were examined as well. The phenotype of stimulated
78 shibire (DYNAMIN) synapses exhibited clear similarities and differences with respect to those of
dab and CHC-RNAi. One interesting difference was that vesicle size was not increased in shibire
(DYNAMIN) (Figure 3.4b, i and ii and c), probably because new vesicles are not formed at the restrictive temperature in this mutant. Furthermore, CLATHRIN-coated intermediates were observed only at shibire (DYNAMIN) synapses (Figure 3.4b, i-iii). As reported previously, these
structures formed from both internal membrane cisternae (Figure 3.4b, i and ii) and the
presynaptic plasma membrane (Figure 3.4b, iii) and reflect arrest of DYNAMIN-dependent
vesicle fission in CLATHRIN-mediated endocytosis (Koenig and Ikeda 1989; Takei, Mundigl et
al. 1996). CLATHRIN-coated intermediates at the plasma membrane were occasionally found
immediately adjacent to but not within the AZ (Figure 3.4b, iii). Striking similarities of the
shibire (DYNAMIN) synaptic phenotype to those of dab and CHC-RNAi were observed as well.
The prevalence of internal membrane cisternae was greatly elevated in shibire (DYNAMIN)
(Figure 3.4b, i and ii and d) and revealed the apparent formation of cisternae as large
invaginations (bulk endocytosis) of the plasma membrane (Figure 3.4b, iv). Finally, and
importantly, shibire (DYNAMIN) synapses exhibited a similar change in the distribution of
synaptic vesicles, including a marked increase in the number of docked vesicles (Figure 3.4e) and
AZ-associated vesicles (Figure 3.4f) with respect to wild type.
The preceding findings indicate a role for CLATHRIN-mediated endocytic mechanisms
at the AZ in sustaining neurotransmitter release during synaptic activity. This requirement was
further investigated by employing paired pulse paradigms to examine a role for endocytic
mechanisms following a single stimulus and their contributions to recovery in paired-pulse
depression (PPD). Our previous work at DLM neuromuscular synapses established that a TS
mutation in the t-SNARE protein, SNAP-25, disrupts synaptic vesicle priming and consequently
slows both fast and slow recovery in PPD with respect to wild type (Kawasaki and Ordway
2009). In the present study, PPD was examined in three mutants disrupting synaptic vesicle
79 endocytosis: shibire (DYNAMIN), dab, and CHC-RNAi. In each case, the initial EPSC amplitude was similar to that of wild-type (Figures. 3.2 and 3.4a), suggesting a normal release- ready vesicle pool, and both fast and slow recovery components were slowed as observed in the
SNAP-25 TS mutant (Figure 3.4g and h). These observations indicate that a single stimulus is sufficient to impair subsequent synaptic function in each mutant and suggest a rapid role for endocytic mechanisms in sustaining neurotransmitter release at the AZ. The possibility that similar SNAP-25 and endocytic mutant phenotypes in PPD result from a common defect in synaptic vesicle priming, and thus refilling of the release-ready vesicle pool, was further explored in a double TS mutant carrying both the shibire (DYNAMIN) and SNAP-25 TS mutations. While each of these mutations alone produced a similar short-term depression phenotype in response to train stimulation at 33 ˚C, the double mutant was quite similar to each single mutant alone (Figure
3.4i). Thus impairment of synaptic vesicle priming in the SNAP-25 TS mutant occludes further enhancement of synaptic depression following disruption of endocytic function. Taken together with the observed persistence or accumulation of docked vesicles in endocytic mutants, these findings suggest that rapid CLATHRIN-mediated endocytic mechanisms are required at the active zone for efficient synaptic vesicle priming and refilling of the release-ready pool.
3.4 DISCUSSION
The results reported here reveal a function for the DISABLED family of CLASPs in synaptic vesicle endocytosis and further define the molecular basis for a rapid role of endocytic mechanisms in sustaining neurotransmitter release during synaptic activity.
DAB function in synaptic vesicle endocytosis. By revealing a function for DAB proteins in synaptic vesicle endocytosis, the present study has implicated a novel molecular component as well as an established set of DAB protein interactions in this process. dDAB function in synaptic vesicle endocytosis appears to involve interactions with CLATHRIN (Figure 3.2), and possibly
AP-2 (Figure 3.S5o-u), which are likely to be mediated by conserved binding motifs (Figure 3.1).
80 In addition, the PTB/DH domain of DAB proteins binds phosphoinositides (Mishra, Keyel et al.
2002; Yun, Keshvara et al. 2003) which are known to play an important role in synaptic vesicle
trafficking (Di Paolo and De Camilli 2006). Finally, and importantly, the CLASP function of
DAB proteins involves PTB/DH domain binding to a sorting motif (NPxY) in the cytosolic domain of cargo proteins (Yun, Keshvara et al. 2003; L.M. 2009; Traub 2009). An initial survey of Drosophila synaptic vesicle proteins revealed an NPxY motif in the cytoplasmic domain of
SYNAPTOTAGMIN I (dSYT1). SYT1 proteins play an important role in synaptic transmission by serving as a calcium sensor for neurotransmitter release (reviewed in (Chapman 2008)) and also function in synaptic vesicle endocytosis (Poskanzer, Marek et al. 2003; Dittman and Ryan
2009). The dSYT1 NPxY motif (residues 387-390; NPYY), which is identical in mammalian
SYT1 proteins and conserved in other SYTs, is located within the C2B domain near a basic region previously implicated in SYT1-AP-2 interactions [see Fig. 2 in (Grass I, Thiel S et al.
2004)]. This motif is of great potential interest because no classical endocytosis signals have been
identified previously in SYTs (Maritzen T, J et al. 2010). Finally, it is of interest to consider the roles of co-AZ and non-AZ populations of dDAB (and CLATHRIN). The rapid onset synaptic phenotypes observed are likely to reflect localization and function of dDAB and CLATHRIN at the AZ. Further study is required to address whether non-AZ domains may mark an endosomal compartment from which CLATHRIN-mediated vesicle formation may occur (e.g. Figure 3.4b, i and ii). Thus far, immunocytochemistry with markers for several endosomal membrane compartments did not show strong colocalization with dDAB and CLATHRIN, as shown in
Figure 3.S5 for the FYVE domain marker for endosomes containing phosphatidylinositol–3– phosphate (Gillooly DJ, Morrow IC et al. 2000).
The preceding considerations raise interesting questions about how the loss of specific dDAB molecular interactions may contribute to the resulting synaptic phenotype. It seems unlikely that the dab synaptic phenotype reflects simple mis-sorting of a synaptic vesicle protein
81 given the similar phenotypes associated with loss of function for several different endocytic
proteins (Figures 3.3 and 3.4). Rather, as described in the following section, it appears that common features of these phenotypes, including a rapid activity-dependent reduction in
neurotransmitter release, slowed recovery in paired-pulse depression, enlarged membrane
cisternae and persistence or accumulation of AZ-associated and docked synaptic vesicles, appear
to reflect a general loss of endocytic function. Consistent with this interpretation, dSYT1
exhibited a wild-type distribution at dab mutant synapses (Figure 3.S6a-j). Thus either dDAB does not participate in dSYT1 sorting or sufficient redundancy is provided by interactions of
SYT1 with at least two other CLATHRIN-associated adaptor proteins, AP-2 and Stonins
(Dittman and Ryan 2009; Maritzen T, J et al. 2010). Furthermore, the distributions of other synaptic vesicle proteins at dab mutant synapses, including neuronal SYNAPTOBREVIN [n-
SYB; (DiAntonio, Burgess et al. 1993)] and the vesicular glutamate transporter [dVGLUT;
(Daniels, Collins et al. 2004)], were similar to those of wild-type (Figure 3.S6k-t and u-d’, respectively). Thus, synaptic vesicle composition appears to be preserved in dab. These findings are consistent with the wild-type EPSC observed in response to the first stimulus (Figure 3.2a) and complete recovery in PPD (Figure 3.S7), as expected for the presence and recovery of a fully
functional release-ready vesicle pool.
Finally, dab mutant synapses exhibit strong depression during prolonged stimulation but
sustain a reduced steady-state level of neurotransmitter release. This is in contrast to the
DYNAMIN mutant, shiTS1, in which EPSC amplitudes progressively decline to zero (Figure
3.S2c). In light of the severe nature of the molecular lesion in dabEC1, these findings suggest persistence of residual synaptic vesicle endocytosis in the absence of dDAB. This likely reflects redundancy in the mechanisms of synaptic vesicle endocytosis as shown for several other endocytic proteins (Grass I, Thiel S et al. 2004; Dittman and Ryan 2009; Kim SH and TA 2009).
82 Rapid Endocytic Mechanisms Maintaining Neurotransmitter Release During Synaptic
Activity. Our previous analysis in the shibire (DYNAMIN) mutant demonstrated a rapid,
activity-dependent reduction in neurotransmitter release which could not be explained simply by the classical role of DYNAMIN in recycling synaptic vesicles (Kawasaki, Hazen et al. 2000).
Rather, it was suggested that accumulation of endocytic intermediates at release sites may occlude fast refilling of the release-ready vesicle pool and that their rapid clearance contributes to maintenance of neurotransmitter release during synaptic activity. Recent studies at the Calyx of
Held have further established a rapid role for DYNAMIN and AP-2 in maintaining neurotransmitter release which preceded formation of endocytic vesicles and directly demonstrated its requirement for fast refilling of the release-ready vesicle pool (Hosoi, Holt et al.
2009; Wu, McNeil et al. 2009). The present study provides several new insights into the mechanisms by which endocytic processes regulate exocytosis. First, localization of both dDAB and CLATHRIN at the active zone (Figure 3.2) suggests a local role for endocytic mechanisms near release sites. Second, persistence or accumulation of the docked synaptic vesicle pool at
AZs (Figures 3.3 and 3.4) indicates a post-docking role for endocytic mechanisms in synaptic vesicle fusion. Third, the similar electrophysiological and ultrastructural phenotypes observed following loss of DYNAMIN, dDAB or CLATHRIN function strongly support a general role for
CLATHRIN-mediated endocytic mechanisms in this process. Finally, comparing and combining endocytic loss of function with a SNAP-25 TS mutant suggests that rapid endocytic mechanisms are required for t-SNARE-mediated synaptic vesicle priming (Figure 3.4g-i).
Together with previous work, the results reported here support a working model in which components of the CLATHRIN-mediated endocytic machinery first interact at the AZ to clear neurotransmitter release sites and subsequently mediate vesicle formation in the PAZ (Figure 3.5).
Key features of this model are discussed in the following text, including the spatial distribution of
83 endocytic proteins and CLATHRIN-coated vesicle intermediates with respect to neurotransmitter release sites.
Our observation of dDAB, CLATHRIN and AP-2 localization to the AZ was greatly
facilitated by the ability to examine isolated AZs at DLM neuromuscular synapses and by the
restricted spatial distributions of these proteins. Note that these features are distinct from those of
larval neuromuscular synapses [c.f.(Gonzalez-Gaitan and Jackle 1997; Heerssen, Fetter et al.
2008; Kasprowicz, Kuenen et al. 2008; Li X., Kuromi H. et al. 2010) and that our previous studies suggest no rapid role for DYNAMIN in exo-endocytic coupling in this preparation (Wu,
Kawasaki et al. 2005). At adult neuromuscular synapses, the rapid functional roles of DYNAMIN and DAP160 (Figure 3.S8) suggest these proteins are also present at the AZ and participate in
early stages of CLATHRIN-mediated endocytosis. However, their broader distribution within
boutons makes it more difficult to confirm localization to the AZ (Figure 3.S5a-u). With respect
to DYNAMIN, which is known to complete vesicle formation through membrane fission
(Mettlen M., Pucadyil T. et al. 2009), previous studies have shown it is also present and functionally important at early stages of CLATHRIN-mediated endocytosis (Evergren E.,
Tomilin N. et al. 2004; Loerke D., Mettlen M. et al. 2009). AZ localization of endocytic proteins
might suggest a mechanism of release site clearance involving rapid formation of CLATHRIN-
coated vesicles directly from the AZ, however this mechanism is not favored primarily because
ultrastructural studies indicate that CLATHRIN-coated vesicles form at PAZ, rather than AZ,
regions of the plasma membrane [(Miller T.M. and Heuser J.E. 1984) (Ringstad N., Gad H. et al.
1999) and Figure 3.4b, iii]. Although elongated membrane invaginations occur at the AZ in the
shibire (DYNAMIN) mutant during recovery from massive synaptic vesicle depletion, these do
not appear to have CLATHRIN coats (Koenig and Ikeda 1996). In the present study, smaller
membrane invaginations (omega structures) observed at the AZ were not CLATHRIN-coated but
often exhibited filamentous connections with the presynaptic dense body (T-Bar) (Figures 3.3a, iv
84 and 3.4b, ii and Figure 3.S9). These structures occurred at low frequencies which were not significantly different among all genotypes. It remains unclear whether or not they are related to rapid synaptic vesicle endocytosis (Wu X.S. and L.G. 2009) and how they may contribute to neurotransmitter release.
Regarding the spatial distribution of endocytic intermediates relative to AZs, remarkable ultrastructural analysis has defined a sequence of events in synaptic vesicle exocytosis (Heuser and Reese 1981) and endocytosis (Miller T.M. and Heuser J.E. 1984; Miller and Heuser 1984)
with high time resolution. Vesicle fusion was observed within several msec after a stimulus and
appeared to deposit clusters of large particles into the AZ region of the plasma membrane.
Particle clusters were maximally abundant at 20 msec after stimulation and disappeared rapidly
over the following 200 msec (Miller T.M. and Heuser J.E. 1984), consistent with a role for rapid
clearance of release sites in synaptic vesicle priming and short-term depression [(Kawasaki,
Hazen et al. 2000; Neher and Sakaba 2008; Hosoi, Holt et al. 2009) and Figure 3.4]. Particle
clusters were thought to dissipate after vesicle fusion (Heuser and Reese 1981) and further
analysis suggested they may reassemble within CLATHRIN-coated pits at the PAZ (Miller T.M.
and Heuser J.E. 1984). In contrast, recent super-resolution light microscopy studies showed that native synaptic vesicle proteins deposited in the plasma membrane from different vesicles do not mix prior to endocytosis but rather remain in distinct clusters and are retrieved separately (Opazo
F., Punge A. et al. 2010). Moreover, plasma membrane-resident synaptic vesicle proteins may tend to be excluded from the AZ as observed in conventional confocal imaging of the native synaptic vesicle protein, SYNAPTOBREVIN (Kawasaki and Ordway 2009). In our current working model (Figure 3.5), endocytic mechanisms facilitate rapid clearance of synaptic vesicle
proteins from release sites either within larger assemblies corresponding to single vesicles or
smaller protein complexes which subsequently assemble in the PAZ as CLATHRIN-coated pits.
A parallel process was described in our recent work at DLM neuromuscular synapses
85 demonstrating activity-dependent redistribution of t-SNARE proteins from AZ to PAZ regions, which likely reflects their participation in plasma membrane cis-SNARE complexes (Kawasaki and Ordway 2009). In comparing these studies, it is of interest to consider the relative contributions of release site clearance and t-SNARE availability to synaptic vesicle priming
(Figure 3.5) and whether AZs maintain their distinctive protein composition, as well as
functionality of neurotransmitter release sites, through sorting mechanisms which distinguish
among different functional or biochemical states of a protein.
METHODS. The following detailed methods are provided in Supporting Materials: Drosophila
Strains, Mutagenesis and Screening, Sequence Analysis of the dabEC1 Mutant, Generation of
Transgenic Lines, Generation of a Polyclonal anti-dDAB Antibody, Western Analysis,
Electrophysiology, Immunocytochemistry and Confocal Microscopy, Transmission Electron
Microscopy, TEM Image Analysis, and Analysis of Numerical Data.
ACKNOWLEDGMENTS
We thank numerous colleagues for kindly providing reagents used in this study. We are especially grateful to Dr. Linton Traub (University of Pittsburgh School of Medicine) for his generosity in collaborative efforts to generate a new anti-dDAB antibody. Dr. Eric Liebl (Denison
University) generously provided original dDAB antibodies which were instrumental in this project. Finally, we acknowledge the Penn State University Electron Microscopy Facility for training and helpful suggestions regarding ultrastructural studies. This work was supported by the
National Institutes of Health, National Institute of Neurological Diseases and Stroke (R01
NS065983) and a seed grant from the Penn State University Huck Institutes of the Life Sciences.
86
3.5 FIGURES
Figure 3.1. A Drosophila disabled mutant. a. Schematic of the Drosophila DAB protein and interaction domains in comparison to mouse DAB-2 [adapted from (Maurer and Cooper 2005)]. Binding partners are indicated in parentheses. b. The dabEC1 mutation. Alignment of PTB/DH domain sequences from mouse DAB-2, Drosophila DAB and C. elegans DAB-1. Amino acid identities (black) and similarities (gray) are shaded. A nonsense mutation in dabEC1 (indicated in red) truncates the protein within the conserved N-terminal PTB domain. Protein sequence accession numbers: Mouse DAB-2 (NP_075607), Drosophila (AAF49424) and C. elegans DAB- 1 (NP_495731). c. Western analysis indicates that full-length dDAB protein is absent in dabEC1 homozygotes and hemizygotes. Flies heterozygous for either dabEC1 or Df(3L)Exel6130, which removes the dab locus, exhibit a reduced level of dDAB relative to wild-type flies.
87
Figure 3.2. dDAB functions in synaptic transmission and exhibits a close spatial and functional relationship with CLATHRIN. (a) dabEC1 exhibits an activity-dependent reduction
88 in neurotransmitter release. Excitatory postsynaptic currents (EPSCs) were recorded from adult DLM neuromuscular synapses at 33°C. Recordings were obtained from wild-type (WT) and dabEC1/Df(3L)Exel6130 (dabEC1/Df) synapses during 1 Hz train stimulation (100 seconds). All 100 EPSCs are superimposed. Initial EPSC amplitudes in WT and dabEC1/Df were 2.05 ± 0.08 µA (n=19) and 2.15 ± 0.14 µA (n=12), respectively, and not significantly different. (b) Peak EPSC amplitudes normalized to the initial amplitude are plotted as a function of stimulus number (n = 5). Here and in subsequent figures, data points represent the mean ± SEM. At 33°C, marked enhancement of steady-state depression was observed in dabEC1/Df relative to wild type. This phenotype was rescued (dab rescue) by neural (presynaptic) expression of a UAS transgene encoding wild-type dDAB fused with EGFP at its C-terminus (UAS-dab-EGFP). (c-v) dDAB and CLATHRIN are colocalized at AZs. Confocal immunofluorescence and native EGFP fluorescence images of DLM neuromuscular synapses expressing UAS-dab (c-v), UAS-dab- EGFP (h-l), UAS-EGFP-clc (CLC; Clathrin Light Chain) (m-q) or both the UAS-dab and UAS- EGFP-clc transgenes (r-v). Anti-HRP and anti-BRP label the neuronal plasma membrane and presynaptic active zones, respectively, and native EGFP fluorescence is denoted as "GFP”. dDAB and CLATHRIN show strong colocalization. Both proteins are distributed in relatively bright puncta adjacent to active zones (non-AZ; arrows in Panels d, i, n and s-v) as well as dim puncta located at active zones (co-AZ; arrowheads in Panels d, i, n and s-v). A similar distribution of co- AZ and non-AZ puncta was observed for endogenous dDAB (Fig. 3.S4a). Comparisons of pixel intensity profiles for dDAB (w) or CLC (x) to those for the active zone marker, BRP, BRUCHPILOT. (Insets) Images in Panels w and x correspond to Panels f and p, respectively. A white line shown in each inset image designates a line of pixels whose intensities were normalized to the maximum pixel intensity and plotted as a function of distance. Similar activity- dependent synaptic phenotypes observed at dab mutant and CLATHRIN Heavy Chain (CHC)- RNAi synapses in response to 1 Hz (y) or 5 Hz (z) stimulation (n = 4-5). The Initial EPSC amplitude at CHC-RNAi synapses was 2.12 ± 0.13 µA (n=11) and not significantly different from WT.
89
Figure 3.3. Ultrastructural analysis reveals disruption of synaptic vesicle endocytosis and persistence of active zone-associated and docked vesicles at dab mutant and CLATHRIN RNAi synapses. (a) Transmission electron microscope (TEM) images of DLM neuromuscular synapses from wild-type (WT), dabEC1/ Df(3L)Exel6130 and CHC-RNAi preparations. Synapses were fixed after 20 Hz stimulation (1200 pulses) at 33 °C. With respect to wild type, dab mutant and CHC-RNAi exhibited an increase in synaptic vesicle size (b) and an increase in the prevalence of large cisternae-like membrane structures (c,d) Both dab mutant and CHC-RNAi synapses exhibited persistence or possibly accumulation of the docked synaptic vesicle pool with respect to wild type. (e,f) The distribution of synaptic vesicles in relation to distance from the AZ. Synaptic vesicle densities within areas 1-3 (e) are plotted in Panel f. In both mutants with respect to wild type, AZ-associated synaptic vesicles (area 1) persisted or possibly increased whereas vesicle density further from the AZ (area 3) was reduced.
90
Figure 3.4. A rapid functional requirement for CLATHRIN-mediated endocytic mechanisms may reflect impaired refilling of the release-ready vesicle pool. (a) Rapid onset of activity-dependent synaptic phenotypes elicited by 5 Hz train stimulation in WT, shiTS1, dabEC1 and CHC-RNAi preparations (n=4-8). The Initial EPSC amplitude at shiTS1 synapses was 2.07 ± 0.16 µA (n=8) and not significantly different from WT (Fig. 3.2). (b) TEM images of shiTS1 DLM neuromuscular synapses exposed to 33 °C and stimulated at 20 Hz for 1 minute. CLATHRIN-coated intermediates (arrows) were observed forming from internal cisternae (b i,ii)
91 and from the plasma membrane (b iii). shiTS1 exhibited wild-type synaptic vesicle size (b,c). However, both the prevalence of membrane cisternae (b,d) and the number of docked vesicles (e) were markedly increased with respect to wild type. (f) The distribution of synaptic vesicles in relation to distance from the AZ at wild-type and shiTS1 synapses. In shiTS1, AZ-associated synaptic vesicles (area 1) were markedly increased with respect to wild type whereas vesicle density further from the AZ (area 3) was greatly reduced. (g,h) Paired-pulse depression (PPD) and analysis of recovery in PPD at wild-type and mutant DLM neuromuscular synapses. The time course of recovery in PPD was fit with two exponential components (h) The amplitude of each component is shown in brackets. Slowing of recovery in PPD was observed in three endocytic mutants and was comparable to that reported previously in SNAP-25TS. Corresponding plots of recovery in PPD are provided in Fig 3.S7. Wild-type (WT) and SNAP-25TS data are those published in a previous study (Kawasaki and Ordway 2009). (i) Relationship of SNAP-25TS and shiTS1 synaptic phenotypes. Activity-dependent synaptic phenotypes elicited by 5 Hz train stimulation at 33 °C in shiTS1 and SNAP-25TS and double mutants (n = 4-8). The double mutant phenotype was similar to that of either single mutant alone.
92
Figure 3.5. A working model for CLATHRIN-mediated endocytic mechanisms in rapid clearance of neurotransmitter release sites. In this model, synaptic vesicle priming and refilling of the release-ready vesicle pool depends on rapid endocytic mechanisms operating at the AZ (see also text). For simplicity, synaptic vesicle proteins to be cleared from the release site are not shown and the endocytic apparatus is represented by CLATHRIN alone. cis-SNARE complex at the PAZ reflects previous studies demonstrating redistribution of t-SNAREs from AZ to PAZ regions, where they likely reside in cis-SNARE complexes that are disassembled by NSF and SNAP (Kawasaki and Ordway 2009). Because it appears that endocytic mechanisms, but not NSF, are required for fast recovery in PPD (as is SNAP-25-dependent vesicle priming) (Kawasaki and Ordway 2009), release site clearance rather than availability of t-SNARES may limit synaptic vesicle priming after a single stimulus. SYB; the v-SNARE protein, SYNAPTOBREVIN. SYX; the t-SNARE protein, SYNTAXIN. CAC; presynaptic voltage-gated calcium channel. Synaptic protein representations were modeled after those reported previously (Rizo and Sudhof 2002).
93 3.5 SUPPORTING INFORMATION
SI Methods
Drosophila Strains dabEC1 (enhancer of cacTS2 1) was generated in the present study (see Mutagenesis and
Screening). cacTS2, shiTS1, Appl-GAL4 were from our laboratory stock collection. Previously generated stocks were generously provided by colleagues. SNAP-25TS and the dap160 alleles, dap1601 and dap1606, were gifts from Dr. David Deitcher (Cornell University, Ithaca, NY) and
Dr. Hugo Bellen (Baylor College of Medicine, Houston, TX), respectively. UAS-Chc-EGFP was
provided by Dr. Simon Bullock (University of Cambridge, Cambridge, UK). For use as
endosomal markers, UAS-GFP-Rab5, UAS-GFP-Rab7 and UAS-GFP–myc-2xFYVE were
obtained from Dr. Marcos González-Gaitán (University of Geneva, Geneva, Switzerland) and
UAS-Rab11-GFP was provided by the Bloomington Stock Center. In addition, the following chromosomal aberrations and transgenic lines were obtained from the Bloomington Stock Center.
Df(3)Exel6130; P{EPgy2}DabEY10190; UAS-dab; UAS-EGFP-Clc. Clathrin heavy chain RNAi transgenic lines were from the Vienna Drosophila RNAi Center. Wild-type flies were Canton-S.
Stocks and crosses were cultured on a conventional cornmeal-molasses-yeast medium at 20 °C.
Mutagenesis and Screening
To isolate third chromosome modifiers of cacTS2, cacTS2 males with an isogenized third chromosome were exposed for 24 hours to 25 mM ethyl methanesulfonate (EMS) (Dellinger,
Felling et al. 2000). Mutagenized males were mated with cacTS2 females carrying the visible third chromosome phenotypic marker, Lyra (Ly), in trans to a TM6c balancer chromosome. The F1 male flies were backcrossed to F0 females. After mating F2 heterozygous siblings, F3 flies homozygous for a mutagenized third chromosome in a cacTS2 genetic background were screened for altered cacTS2 behavior at 36°C. TS paralytic behavior was examined as described previously
(Brooks, Felling et al. 2003). Briefly, flies were placed in a vial preheated to 36˚C by immersion
94 in a water bath. 2-3 day-old flies were tested in groups of six. Five groups were examined for
each genotype (n = 5). Time for 50% paralysis represents the time at which three flies were no
longer able to stand. In all tests exceeding 5 min in duration, the cotton plug sealing the vial was
wet with water to prevent dehydration. Behavioral tests were truncated at 50 minutes if 50%
paralysis had not occurred.
Sequence Analysis of the dabEC1 Mutant.
Sequence analysis to identify the molecular lesion in dabEC1 was carried out essentially as
described previously (Kawasaki, Collins et al. 2002). Briefly, head cDNA preparations were
prepared from the dabEC1 mutant and used as template for PCR. A previously reported dab
sequence (Gertler, Hill et al. 1993) (accession number L08845) was used to design primers for
RT-PCR, PCR, and Sequencing. Gel purified PCR products were sequenced at the Penn State
University Nucleic Acids Facility. Sequences from the dabEC1 mutant were compared to those from the parent third chromosome used in the mutagenesis.
Transgenic Lines
A UAS-dab-EGFP transgenic line was generated to express wild-type dDAB fused with EGFP at its C-terminus. PCR amplification of the dab open reading frame (ORF) was carried out using
head cDNA preparations from Canton-S flies isogenized for the third chromosome as template.
The final fusion construct was incorporated into the transformation vector, pUAST (Brand and
Perrimon 1993) and transgenic flies were generated as described previously (Kawasaki, Collins et
al. 2002).Neural expression of UAS-transgenes was achieved using the Appl-GAL4 driver
(Torroja, Chu et al. 1999).
Generation of a Polyclonal anti-dDAB Antibody
A rabbit polyclonal antiserum, anti-dDAB, was generated in collaboration with Dr. Linton Traub of the University of Pittsburgh School of Medicine. The antiserum was raised against a GST fusion protein corresponding to amino acids 1-553 of dDAB, which includes the entire PTB
95 domain. Generation of the GST fusion protein and antisera was carried out in the Traub
laboratory. Specificity of the antibody for dDAB was confirmed by Western analysis of head
homogenates from wild-type and dabEC1 mutant flies.
Western Analysis
Western analysis of head homogenates was performed using conventional methods. The equivalent of one fly head was loaded per lane. The primary antibody, a rabbit polyclonal anti- dDAB antibody (Gertler, Hill et al. 1993), was generously provided by Dr. Eric Liebl (Denison
University, Granville, OH) and used at a dilution of 1:500. A second rabbit anti-dDAB antibody
generated and affinity-purified in collaboration with Dr. Linton Traub (University of Pittsburgh
School of Medicine) was used for Western analysis at a dilution of 1:1000. Detection was
performed with horseradish peroxidase (HRP)-conjugated secondary antibodies (Amersham
Biosciences, Arlington Heights, IL) and enhanced chemiluminescence (Amersham Biosciences
ECL Plus Western Blotting Detection System). Tubulin served as loading control and was detected using a monoclonal anti-acetylated α-tubulin antibody (Sigma, St. Louis, MO) at a dilution of 1:2,000,000.
Electrophysiology
Excitatory postsynaptic currents (EPSCs) were recorded at dorsal longitudinal flight muscle
(DLM) neuromuscular synapses of 3 to 5 day-old adults reared at 20°C. These experiments were performed as described previously (Kawasaki and Ordway 2009).
Immunocytochemistry and Confocal Microscopy
Immunocytochemistry was performed essentially as described previously (Kawasaki, Zou et al.
2004; Kawasaki and Ordway 2009). These studies employed the following primary antibodies
(many were generously provided by colleagues): A rabbit anti-dDAB antibody generated and affinity-purified in collaboration with Dr. Linton Traub (University of Pittsburgh School of
Medicine) (1:2500); rabbit anti-SYT Dsyt CL1(1:5000) [Dr. Noreen Reist (Colorado State
96 University, Fort Collins, CO)]; mAb nc82 anti-BRP (BRUCHPILOT) (1:50) [Dr. Erich Buchner
(Universitaet Wuerzburg, Germany)]; Rabbit anti-DYNAMIN (1:2000) [Dr. Mani Ramaswami
(University of Dublin, Ireland)]; Rabbit anti-DAP160 (1:1000) [Dr. Graeme Davis (University of
California, San Francisco)]; Rabbit anti-α-ADAPTIN (1:2500) [Dr. Marcos Gonzalez-Gaitan
(University of Geneva, Switzerland)]; Rabbit anti-neuronal SYNAPTOBREVIN (1:10); Rabbit anti-dVGLUT (1:2500) [Dr. Hermann Aberle (Max-Planck Institute, Tubingen, Germany)]. Cy5- conjugated rabbit anti-HRP (1:200) (Jackson Immunoresearch Laboratories, West Grove, PA).
Secondary antibodies included Alexa Fluor 568-conjugated anti-rabbit IgG (1:200) and Alexa
Fluor 647-conjugated anti-mouse IgG (1:200) (Invitrogen, Carlsbad, CA).
To observe the distribution of Drosophila SYNAPTOTAGMIN 1 (dSYT1), neuronal
SYNAPTOBREVIN and the Vesicular Glutamate Transporter after synaptic stimulation, DLM preparations were exposed to 33 °C for 7 minutes and then stimulated for 1 minute at 20 Hz prior to fixation with 33 °C chemical fixative. Samples were fixed for an additional 30 minutes at room temperature and further processing for immunocytochemistry was carried out as described.
DLM neuromuscular synapse preparations were imaged using an Olympus FV1000 confocal microscope (Olympus Optical, Tokyo, Japan) with a PlanApo 60x 1.4 numerical aperture oil objective (Olympus Optical) and a z-step size of 0.2 µm. Images were obtained and processed with Fluoview software (Olympus Optical).
Analysis of confocal images to quantify AZ localization of dDAB and CLATHRIN was carried out as follows. In single images of synapses expressing either dDAB-EGFP or EGFP-
CLC, regions of interest including only the AZ were defined using the AZ marker, BRP. For each AZ, an equivalent region of interest was placed adjacent to the AZ for measurement of background fluorescence. AZs which exhibited no overlap with non-AZ dDAB-EGFP or EGFP-
CLC puncta were analyzed. Quantitation of the AZ signal was based on the ratio of the integrated fluorescence intensity at the AZ (FAZ) and the corresponding background (FBG). To confirm the
97 validity of AZ fluorescence signals, control preparations lacking expression of either EGFP
fusion protein were analyzed in the same fashion (see Fig. 3.S4, c and d).
Transmission Electron Microscopy
TEM was performed on a JOEL JEM 1200 EXII microscope housed at the Penn State University
Electron Microscopy Facility. These studies employed conventional methods, essentially as
described (Kawasaki, Mattiuz et al. 1998; Kawasaki, Hazen et al. 2000).
TEM Image Analysis
Data were obtained from images including active zone profiles in which the “t-bar” morphology of the presynaptic dense body was visible. Counts of docked vesicles included those which were located within 10 nm of the presynaptic membrane and not more than 200 nm from the base of the t-bar. For measurements of synaptic vesicle size and density, a set of three concentric shells of radius 100, 200 and 300 nm defined three areas (Area 1, 2 and 3) of progressively increasing distance from the AZ (Fig. 3.3e). Each area was equally subdivided into 4 sectors to produce a total of 12 sectors. Sectors in which at least 90% of the area was occupied by the presynaptic bouton were included in the analysis of vesicle size and density. Cisternae were defined as vesicular intracellular membrane compartments for which any axis of the cross-sectional profile exceeded 100 nm in length. These were not counted as synaptic vesicles and not included in analysis of vesicle number or diameter. Counts of membrane cisternae included those which were located within 300 nm of the t-bar base.
Data Analysis
Microsoft (Seattle, WA) Excel was utilized to analyze numerical data and generate graphs. All data values are presented as mean ± SEM. Statistical significance was determined using the two- tailed Student's t test and significance was assigned to comparisons for which p 0.05.
98 Supplementary Figures
Figure 3.S1. A new dab mutation recovered as an enhancer of the cacTS2 temperature-sensitive paralytic phenotype. Although cacTS2 is paralyzed at 38 °C, it is resistant to paralysis at 36 °C (Brooks, Felling et al. 2003). In contrast, cacTS2;;dabEC1 double mutants exhibited rapid paralysis at 36 °C. Like cacTS2, the isolated dabEC1 mutant (in a cac+ genetic background) did not exhibit TS paralysis at 36 °C. Thus the dabEC1 mutation is defined as an enhancer of the cacTS2 paralytic phenotype.
99
Figure 3.S2. Characteristics of the dab synaptic phenotype. (a,b) dab exhibits an activity- dependent decrease in excitatory postsynaptic currents (EPSCs) at 20 °C. Recordings were obtained from wild-type (WT) and dabEC1/Df(3L)Exel6130 (dabEC1/Df) DLM neuromuscular synapses maintained at 20 °C and stimulated at 1 Hz (a) (n=4) or 5 Hz (b) (n=4-5). (c) dab mutant synapses exhibit strong depression during prolonged stimulation but sustain a reduced steady-state level of neurotransmitter release. Comparison of synaptic depression at WT, dabEC1/Df and shiTS1 synapses during prolonged stimulation (20 Hz for 1 minute) at the restrictive temperature of 33 °C (n=2 in each case).
Figure 3.S3. A newly generated rabbit anti-dDAB antibody recognizes dDAB. Western analysis of head homogenates detected full-length dDAB protein in WT. dDAB was absent in dabEC1 hemizygotes (dabEC1/Df) and reduced in flies carrying one wild-type copy of dab (+/Df). Lower molecular mass bands of approximately 215 and 190 kD do not appear to be derived from dDAB.
100
Figure 3.S4. (a) Presynaptic distribution of endogenous dDAB. Anti-HRP and anti-BRP label the neuronal plasma membrane and presynaptic active zones at DLM neuromuscular synapses. An anti-dDAB antibody was used to label endogenous dDAB. Endogenous dDAB is distributed in bright puncta adjacent to active zones (arrows) as well as dim puncta located at active zones (arrowheads). (b) The presynaptic distribution of CLATHRIN HEAVY CHAIN is similar to that of CLATHRIN LIGHT CHAIN. Confocal immunofluorescence and native EGFP fluorescence images of DLM neuromuscular synapses expressing a UAS-chc-EGFP (CHC; Clathrin Heavy Chain) transgene as shown for CLC in Fig. 3.2m-q. Arrow and arrowheads mark non-AZ and co- AZ puncta, respectively, as described in Fig. 3.2.c and d. Confirmation of dDAB and CLATHRIN localization at the AZ. To confirm the validity of co-AZ fluorescence signals for dDAB-EGFP (Fig. 3.2 h-l) and EGFP-CLC (Fig. 3.2 m-q), images were obtained from control preparations expressing neither dDAB-EGFP or EGFP-CLC. (c) Analysis of control images is
101 compared with those from dDAB-EGFP or EGFP-CLC expressing synapses. (d) The AZ signal was expressed by determining the ratio of the integrated fluorescence intensity at the AZ (FAZ) and the corresponding background (FBG) and subtracting 1 such that the absence of an AZ signal is represented as zero. The [FAZ /FBG - 1] values for control, dDAB-EGFP and EGFP-CLC were - 0.01 ± 0.03 (n=7), 0.21 ± 0.03 (n=6) and 0.59 ± 0.09 (n=6), respectively. Asterisks denote statistical significance with respect to control experiments.
Figure 3.S5. (a-v) Presynaptic distribution of endocytic proteins including DYNAMIN, DAP160 and α-ADAPTIN. The distribution of each endogenous protein was examined by immunocytchemistry at DLM neuromuscular synapses and compared with that of CLATHRIN (neural expression of CLC-EGFP as shown in Fig. 2r-v). Arrow and arrowheads mark non-AZ and co-AZ puncta, respectively, as described in Fig. 2. (w-z) Double labeling of DLM neuromuscular synapses for dDAB and an endosomal marker, EGFP-FYVE. Confocal immunofluorescence and native EGFP fluorescence images of synapses exhibiting neural expression of UAS-dab and UAS-EGFP-FYVE as shown in Fig. 2. No clear pattern of colocalization was observed for EGFP-FYVE and dDAB.
102
Figure 3.S6. The presynaptic distribution of dSYT1 and other synaptic vesicle proteins appears to be preserved at dab mutant synapses. Wild type and dab mutant DLM neuromuscular synapses were exposed to 33 °C and stimulated at 20 Hz for 1 minute (see Methods). Markers are as described in Fig. 2. SYT; dSYNAPTOTAGMIN1, SYB; neuronal SYNAPTOBREVIN, VGT; Vesicular Glutamate Transporter. All three synaptic vesicle proteins appeared to retain a wild- type distribution in the dab mutant.
103
Figure 3.S7. Recovery in PPD at DLM neuromuscular synapses from dab, CHC-RNAi and shiTS1. Plots of recovery time course in PPD as described previously (Kawasaki and Ordway 2009). The time course of recovery in PPD (n = 4-6) was fit with two exponential components (solid lines)
corresponding to the τfast and τslow values presented in Fig. 4h.
Figure 3.S8. A dap160 mutant exhibits an activity-dependent decrease in excitatory postsynaptic currents (EPSCs) resembling that of other endocytic mutants. Recordings were obtained from wild-type (WT), dabEC1/Df(3L)Exel6130 (dabEC1/Df) and dap1601/dap1606 DLM neuromuscular synapses exposed to 33 °C and stimulated at 1 Hz. Mean EPSC amplitudes as a percentage of initial amplitude are plotted as a function of stimulus number (n = 4-5).
104
Figure 3.S9. Non-CLATHRIN-coated membrane invaginations (omega structures) occur at the AZ. A TEM AZ profile from an example wild-type unstimulated preparation showing a membrane invagination at the AZ (arrow). These structures were defined as invaginations of the presynaptic plasma membrane occurring within 200 nm of the base of the presynaptic dense body (t-bar). This example is representative of 64 synaptic profiles in which 11 omega structures were observed.
Table 3.S1. Summary of Ultrastructural Analysis
105 Chapter 4
The following content is part of a manuscript that is in preparation to be published.
A New Drosophila Mutant Permits Molecular Dissection of the COMPLEXIN Fusion Clamp Function in Synaptic Exocytosis
Janani Iyer, Christopher J. Wahlmark, Giselle A. Kuser-Ahnert and Fumiko Kawasaki Department of Biology and Center for Molecular Investigation of Neurological Disorders Penn State University, University Park, PA 16802.
Author Contributions: Electrophysiology was performed by Fumiko Kawasaki. Christopher Wahlmark and Giselle A. Kuser-Ahnert made important contributions to the genetic screen, recovery of cpx1257 mutant and its molecular characterization.
4.1 ABSTRACT
COMPLEXINs (CPXs) are a conserved family of proteins which play an important role in synaptic vesicle fusion with the presynaptic plasma membrane during chemical neurotransmitter release. CPX binds to and regulates Soluble NSF Attachment protein REceptor
(SNARE) protein complexes, which are comprised of two target membrane (t)-SNAREs and one vesicle membrane (v)-SNARE and function as a core component of the neurotransmitter release machinery. Several studies have demonstrated that CPX plays multiple functional roles, including activation of neurotransmitter release evoked by a presynaptic action potential and an inhibitory or clamping function which suppresses spontaneous synaptic vesicle fusion. However, the in vivo
molecular mechanisms mediating these functions remain a matter of debate and further
investigation. Here we report recovery and analysis of a new cpx mutant in Drosophila, cpx1257,
revealing spatially defined and separable pools of CPX which make distinct contributions to the
activation and clamping functions. cpx1257 lacks only the last C-terminal amino acid of CPX,
which disrupts a C-terminal recognition sequence for prenylation, a post-translational lipid
modification implicated in membrane targeting of CPX and other cytosolic proteins.
Immunocytochemical analysis established localization of wild-type CPX to both active zone (AZ)
106 regions of the presynaptic plasma membrane containing neurotransmitter release sites and
broader presynaptic membrane compartments outside of the AZ, including synaptic vesicles.
Parallel biochemical studies confirmed CPX membrane association and demonstrated robust
binding interactions of CPX with all three SNAREs. This is in contrast to the CPX1257 mutant, in which AZ localization of CPX persists but general membrane localization and, surprisingly, the bulk of CPX-SNARE protein interactions are abolished. In electrophysiological analysis at adult
neuromuscular synapses, a previously reported cpx null mutant exhibited severely reduced
activation of evoked neurotransmitter release, as indicated by a marked reduction in the amplitude
of the Excitatory PostSynaptic Current (EPSC). This was accompanied by loss of the CPX
clamping function observed as a marked increase in the frequency of spontaneous fusion events.
Furthermore, the null mutant exhibited altered neurotransmitter release kinetics characterized by
marked slowing of the excitatory postsynaptic current (EPSC) rise and decay times with respect
to wild-type. Notably, the cpx1257 mutant exhibited a wild-type EPSC (amplitude and kinetics)
but selectively failed to clamp spontaneous fusion. These results indicate that spatially distinct
and separable interactions of CPX with presynaptic membranes and SNARE proteins appear to
mediate separable activation and clamping functions of CPX in neurotransmitter release.
4.2 INTRODUCTION
It is widely accepted that SNARE proteins function at the core of the neurotransmitter
release apparatus, where they promote exocytotic fusion of neurotransmitter-filled synaptic
vesicles with the presynaptic plasma membrane [reviewed in (Rizo and Sudhof 2002; Sudhof
2004)]. However, defining the mechanisms which provide precise and rapid regulation of
synaptic vesicle fusion remains among the foremost problems in cellular and molecular
neuroscience. The identification of CPX as a SNARE complex binding protein which can both
promote SV fusion evoked by a presynaptic action potential and suppress or “clamp” spontaneous
vesicle fusion has advanced our understanding of the molecular interactions regulating SNARE
107 function [reviewed in (Brose 2008)]. Recent models suggest that specific domains of CPX
(Figure 4.1A) contribute to different aspects of synaptic vesicle fusion. Whereas a “central helix” thought to participate in binding interactions with SNARE complexes is absolutely required for
CPX function, other domains appear to mediate specific aspects of CPX activity. The N-terminus is required for evoked vesicle fusion but does not regulate spontaneous release, whereas an adjacent “accessory helix” is necessary for clamping spontaneous vesicle fusion (Xue, Craig et al.
2010). Finally, recent studies have shown that the CPX C-terminus is specifically required for the clamping function (Martin, Hu et al. 2011; Kaeser-Woo, Yang et al. 2012). Of particular relevance to the present study is a specific CaaX motif found at the extreme C-terminus of several mammalian and Drosophila CPX isoforms. This motif has been shown to mediate CPX prenylation [a form of lipid modification; (Reim, Wegmeyer et al. 2005)] and has been implicated in both targeting CPX to membranes (Reim, Wegmeyer et al. 2005) and the CPX clamping function (Cho, Song et al. 2010). The form of prenylation demonstrated for mammalian CPX isoforms (CPX3 and 4) is farnesylation (Reim, Wegmeyer et al. 2005), consistent with previous studies indicating that specific residues in the X position of the CaaX motif (A,C,M,Q,S) selectively mediates farnesylation (Schillo, Belusic et al. 2004).
This progress is extended by new insights gained from the present study, in which the isolation and characterization of a new CPX mutant further defines the in vivo molecular basis of
CPX functions and interactions within the neurotransmitter release apparatus. This study provides the first demonstration of a specific subcellular distribution for CPX within the presynaptic terminal and a role for C-terminal farnesylation in mediating both association of CPX with presynaptic membranes and CPX clamping of spontaneous synaptic vesicle fusion. In addition, our parallel analysis of a previously reported cpx null mutant (Huntwork and Littleton 2007) revealed an important contribution of CPX in controlling the kinetics of neurotransmitter release.
108 4.3 MATERIALS AND METHODS
Drosophila Strains. Appl-GAL4 and w;;Ly/TM6c were from our laboratory stock collection. The cpxSH1 null mutant and the UAS-cpx transgenic line were generously provided by Troy Littleton
(MIT, Cambridge, MA). Deficiency lines, Df(3L)GN34 and Df(3R)Exel6140, were obtained from the Bloomington Stock Center. UAS-EGFP-cpx and UAS-EGFP-cpx1257 transgenic lines were generated in the current study (see “Generation of Transgenic lines”). Stocks and crosses were cultured on a conventional cornmeal-molasses-yeast medium at 20°C for electrophysiological analysis or at room temperature. Wild-type flies were Canton-S.
Mutagenesis and Screening. To recover new hypomorphic and conditional alleles of cpx, a genetic screen was carried out as summarized in Figure S1. Briefly, Iso3 males carrying an isogenized third chromosome were exposed to 25mM ethyl methanesulfonate (EMS) for 24 hours
(Dellinger, Felling et al. 2000). F2 flies heterozygous for a mutagenized third chromosome in trans to the third chromosome carrying deficiencies, Df(3L)GN34 and Df(3R)Exel6140, were screened for motor defects at 38 °C. The latter deficiency removes the cpx locus.
Molecular Characterization of the cpx1257 Mutant. Sequence analysis of cpx open reading frame
(ORF) was carried out essentially as described previously (Lutas, Wahlmark et al. 2012). Briefly,
genomic DNA was prepared from the cpx mutant and used as template for PCR. Gel-purified
PCR products were sequenced at the Penn State University Nucleic Acids Facility. Sequences
from the cpx mutant were compared to those from the parent third chromosome used in the
mutagenesis. The location of the cpx1257 mutation within an alternative exon is defined further in the following section.
Relevant information about cpx alternative exons and splice variants. As described in the
Results, the cpx1257 mutation was found to be isoform-specific in that it maps to an alternative exon. As indicated in flybase (www.flybase.org), the position of this exon within the genomic
DNA sequence begins at 3R:123210 and ends at 3R:124307. Splice variants including this exon
109 are exemplified here by transcript cpx-RU. The single known splice variant lacking this exon is transcript cpx-RE. The cpx coding sequence of the UAS-EGFP-cpx and UAS-EGFP-cpx1257
transgenes corresponds to that of transcript cpx-RU. The cpx coding sequence of the UAS-cpx
corresponds to that of transcript cpx-RE.
Generation of Transgenic Lines. Transformation construct for UAS-EGFP-cpx was generated
by inserting the cpx ORF with EGFP fused to its N-terminus into the P element transformation
vector, pUAST (Brand and Perrimon 1993). The cpx ORF was amplified from a cDNA clone
(clone ID: GH27718; GenBank accession number: AY121629; corresponds to transcript cpx-RU) obtained from the Drosophila Genomics Research Center. In the case of UAS-EGFP-cpx1257, the
codon for the last amino acid of the cpx ORF was removed. Generation of transgenic lines was achieved as described previously (Kawasaki, Zou et al. 2004). Neural expression of UAS transgenes was achieved using the Appl-GAL4 driver.
Immunocytochemistry and Confocal Microscopy. Immunocytochemistry and Confocal
Microscopy were performed essentially as described previously (Kawasaki, Iyer et al. 2011;
Lutas, Wahlmark et al. 2012). These studies employed the following primary antibodies: rabbit polyclonal anti-CPX antibody (1:1,000) [Troy Littleton (MIT, Cambridge, MA)]; rabbit anti-SYT
Dsyt CL1(1:5,000) [Noreen Reist (Colorado State University, Fort Collins, CO)]; mAb nc82 anti-
BRP (BRUCHPILOT) (1:50) (Developmental Studies Hybridoma Bank, University of Iowa,
Iowa City, IA, U.S.A.); Cy5-conjugated rabbit anti-HRP (1:200) (Jackson Immunoresearch
Laboratories, West Grove, PA). Secondary antibodies included Alexa Fluor 488-conjugated anti- mouse IgG (1:200), Alexa Fluor 568-conjugated anti-mouse IgG (1:200) and Alexa Fluor 568- conjugated anti-rabbit IgG (1:200) (Invitrogen, Carlsbad, CA). Adult DLM neuromuscular synapse preparations were imaged using an Olympus FV1000 confocal microscope (Olympus
Optical, Tokyo, Japan) with a PlanApo 60x 1.4 numerical aperture oil objective (Olympus
Optical) and a z-step size of 0.2 µm. Images were obtained and processed with Fluoview software
110 (Olympus Optical). All images shown in figures are maximum projections of two consecutive
optical z-sections.
Western Analysis. Western analysis of head homogenates was performed using conventional methods as described previously (Kawasaki, Zou et al. 2004; Zou, Yan et al. 2008). To examine
CPX, the equivalent of 0.5 fly heads was loaded per lane on a 12% SDS-PAGE gel. The primary antibody, a rabbit polyclonal anti-CPX antibody (Huntwork and Littleton 2007) was used at a dilution of 1:10,000. Detection was performed with horseradish peroxidase (HRP)-conjugated secondary antibodies (Amersham Biosciences, Arlington Heights, IL) and enhanced chemiluminescence (ECL Plus Western Blotting Detection System: Amersham Biosciences,
Arlington Heights, IL). TUBULIN served as a loading control and was detected using a monoclonal anti-acetylated α-TUBULIN antibody (Sigma, St. Louis, MO) at a dilution of
1:2,000,000. Analysis following Triton X-114 phase partitioning and co-Immunoprecipitation used the following primary antibodies: mAb JL-8 anti-GFP (1:500) (Clontech, Mountain View,
CA), mAb JLA20 anti-ACTIN (1:10) (Developmental Studies Hybridoma Bank), mAb 8C3 anti-
SYNTAXIN (1:10) (Developmental Studies Hybridoma Bank), rabbit anti-SNAP-25 antibody
(1:2,000) [David Deitcher (Cornell University, Ithaca, NY)], rabbit anti-NSYB antibody (1:30),
[Richard Ordway (Penn State University, State College, PA)]
Triton X-114 phase partitioning. Triton X-114 extractions were performed essentially as described previously (Bordier 1981; Wang and Coppel 2002; Mathias, Chen et al. 2011).
Precondensation of Triton X-114 (100%, Sigma) was performed as described to remove the more hydrophilic components from the reagent (Wang and Coppel 2002). The resulting detergent- enriched phase containing ~10% detergent in PBS (171 mM NaCl, 4 mM Na2HPO4, 3.4 mM KCl and 1.84 mM KH2PO4, pH 7.5) was used as a stock solution to prepare all other detergent solutions by dilution in PBS. Head homogenates from flies expressing EGFP-CPX, EGFP-
CPX1257 or soluble EGFP alone as well as Canton-S and cpx1257 were prepared by homogenizing
111
50 heads in 200µl of 1% Triton X-114 containing a 1:100 protease inhibitor cocktail (Sigma).
Following 1 hr incubation on ice and centrifugation for 10 min at 10,000 g at 4 ºC, 150µl of the final head lysate (supernatant) was subjected to phase partitioning. Phase partitioning was achieved by incubating the supernatants at 37 ºC for 15 min, followed by centrifugation at 5000 g for 15 min at room temperature. The upper aqueous phase was collected and the stock Triton X-
114 solution was added to achieve a final volume of 200µl containing a detergent concentration of ~1%. The detergent phase, an oily droplet at the bottom of the tube, was resuspended in 175µl
of 0.05 % TritonX-114 to achieve a final volume of 200 µl containing a detergent concentration of ~1%. These aqueous and detergent samples were further purified by repeating the phase partitioning process, essentially as described above, three times. After the final partitioning, the proteins were precipitated by adding 100 % TCA to a final concentration of 15% and incubated on ice for 30 min. Then the pellet was recovered by centrifuging at 10,000 g for 20 min at 4ºC, washed in ice-cold acetone at 10,000 g for 10 min at 4ºC and air dried. Both the aqueous and the detergent pellets were resuspended in 50µl of SDS sample buffer [62.5 mM Tris∙Cl/SDS, pH 6.8,
2% SDS, 1% (v/v) 2-mercaptoethanol, 10% (v/v) glycerol and 0.5% Bromophenol Blue] and boiled for 3min. For Western blot analysis, 10μl of the final samples was loaded to a 12% SDS-
PAGE gel.
Co-Immunoprecipitation (Co-IP) analysis. Co-Immunoprecipitation analysis was carried out essentially as described previously (Zou, Yan et al. 2008; Yu, Kawasaki et al. 2011). Fly head lysate was prepared from transgenic flies expressing EGFP-CPX or EGFP-CPX1257 as well as
Canton S controls. Five hundred microliters of lysate equivalent to 300 heads and 5 μg of a rabbit polyclonal anti-GFP antibody (Invitrogen, Carlsbad, CA) were incubated with 50 μl of a
50% slurry of ProteinA–Sepharose beads (Amersham Biosciences). After washing, beads were
pelleted, resuspended in 40μl of SDS sample buffer and boiled for 3 min to elute proteins. For
112 detection of EGFP-CPX or EGFP-CPX1257 (IP), the equivalent of 0.5 fly heads for pre-IP input and 4μl (~30 heads) of the IP sample were loaded to a 12% SDS-PAGE gel. To examine co-IP of the SNARE proteins, the equivalent of 0.5 fly heads for pre-IP input and 12μl (~90 heads) of the
IP sample were loaded into a 12% SDS-PAGE gel. After electrophoresis, gels were processed for
Western analysis.
Electrophysiology. Two-electrode voltage-clamp recordings of EPSCs and mEPSCs were recorded at 20°C from dorsal longitudinal flight muscle (DLM) neuromuscular synapses of 3 to 5 day-old adults and larval neuromuscular synapses of ventral longitudinal muscle 6 in abdominal segment A3. Saline solution was consisted of (in mM): 129 NaCl, 2 KCl, 4.0 MgCl2, 1.8 CaCl2,
5 HEPES, and 36 sucrose. The pH was adjusted to 7.0 using NaOH. These experiments were
performed as described previously (Wu, Kawasaki et al. 2005; Kawasaki and Ordway 2009) with
the following exceptions. mEPSC recordings were performed using an AxoClamp-2B amplifier
(Axon Instruments, Foster City, CA) and, in the case of recordings at DLM neuromuscular
synpases, the membrane potential was held at -80mV to improve the signal to noise ratio.
Data Analysis. Microsoft (Seattle, WA) Excel was utilized to analyze numerical data and
generate graphs. All data values are presented as mean ± SEM. Statistical significance was
determined using the two-tailed Student's t test and significance was assigned to comparisons for
which p 0.05.
4.4 RESULTS
Genetic and Molecular Characterization of a New cpx Mutant. Genetic analysis to examine the in vivo molecular mechanisms of CPX function was pursued through a forward genetic screen for new mutant alleles of the single Drosophila cpx gene. To complement a previously reported cpx null mutant (Huntwork and Littleton 2007), this screen was intended to recover hypomorphic and conditional alleles that may further define the in vivo molecular determinants of CPX function. A screen was performed using chemical mutagenesis and subsequent screening for cpx mutants in
113 F2 progeny carrying a mutagenized second chromosome in trans to a deficiency (deletion) which
removes cpx (see Methods and Figure 4.S1). The screen consisted of examining motor behavior at the elevated temperature to detect hypomorphic or temperature-sensitive (TS) phenotypes. One new mutant, initially referred to 1257, was recovered on the basis of its severe lack of motor co- ordination at 38 °C.
On the basis of genetic complementation testing with the cpx null mutant, 1257 was confirmed to be a new allele of cpx and named cpx1257. Sequence analysis of the cpx ORF revealed the molecular lesion in cpx1257. Remarkably, this mutation removes only the last C- terminal amino acid of CPX, which is a Q in the X position of the C-terminal CaaX sequence
(Figure 4.1A). This residue is required for farnesylation of certain mammalian CPX isoforms
(CPX3 and 4) and has been implicated in their membrane targeting (Reim, Wegmeyer et al.
2005). Western analysis of the cpx1257 mutant (Figure 4.1B) demonstrated wild-type CPX protein levels in cpx1257 (comparing the hemizygous conditions, cpx1257/Df(3R)Exel6140 and
+/Df(3R)Exel6140). Thus the cpx1257 phenotype appears to reflect the properties of the mutant protein rather than its expression level. Note that CPX1257 migrates at a slightly higher relative molecular mass in comparison to wild-type CPX (Figure 4.1B), most likely because of altered post-translational processing (see Discussion). Finally, the cpx1257 mutation is isoform-specific because it occurs in an alternative exon which is included in most isoforms expressed from the single Drosophila cpx gene (see Figure 4.S2A and Materials and Methods). Incidentally, the altered migration of CPX1257 in Western analysis also permits the conclusion that isoforms containing this alternative exon, and thus those affected by the cpx1257 mutation, are predominant in the nervous system (Figure 4.S2B). Lastly, final confirmation that cpx1257 is an allele of cpx is provided by transformation rescue experiments described in the following section.
Electrophysiological Analysis Reveals a Distinct Synaptic Phenotype in cpx1257. To examine the impact of cpx1257 on synaptic function, voltage-clamp analysis of synaptic currents was
114 carried out at adult Dorsal Longitudinal Muscle (DLM) neuromuscular synapses as described
previously (Kawasaki and Ordway 2009). This glutamatergic synapse exhibits detailed functional properties similar to those of mammalian synapses (Kawasaki and Ordway 2009) and has been fruitful in molecular analysis of synaptic function (Kawasaki and Ordway 2009; Kawasaki, Iyer
et al. 2011). In part as a basis for comparison, our electrophysiological studies of cpx began with the cpxSH1 null mutant, which has been studied previously in the larval neuromuscular synapse preparation using current clamp methods (Huntwork and Littleton 2007).
Recordings from DLM neuromuscular synapses of the cpxSH1 null mutant revealed both
anticipated and novel aspects of the mutant phenotype. Consistent with previous studies of mouse
(Reim, Mansour et al. 2001; Xue, Stradomska et al. 2008) and Drosophila (Huntwork and
Littleton 2007) null mutants, cpxSH1 exhibited a drastic reduction in evoked neurotransmitter release, reflected in an approximately 77% decrease in the peak Excitatory Postsynaptic Current
(EPSC) amplitude with respect to wild type (Figure 4.2A,B,E). Note that this amplitude reduction is substantially more severe than that reported previously for this mutant at larval neuromuscular synapses [(Huntwork and Littleton 2007)and see Figure 4.S3]. Also as anticipated from previous work, a marked increase in the frequency of miniature EPSCs (mEPSCs) was observed in cpxSH1 (Figure 4.2H,I), consistent with loss of CPX function in clamping spontaneous
synaptic vesicle fusion. Notably, in contrast to previous studies of larval neuromuscular synapses
utilizing current clamp methods (Huntwork and Littleton 2007), another interesting and novel
aspect of the cpxSH1 null mutant phenotype was observed. cpxSH1 exhibited an altered EPSC waveform, characterized by a slowing of the EPSC rise and decay times with respect to wild type
(Figure 4.2C,F,G). These synaptic phenotypes are rescued by neuronal (presynaptic) expression of a wild-type CPX (Figure 4.2D), indicating that each reflects loss of presynaptic CPX function.
Finally, to address any apparent differences between adult and larval neuromuscular synapses, voltage clamp analysis of larval neuromuscular synapses was also carried out in cpxSH1 and
115 demonstrated similar mutant phenotypes to those of the adult. These included an approximately
52% reduction in EPSC amplitude and a slowing of EPSC kinetics characterized by increased
EPSC rise and decay times (Figure 4.S3). Taken together, the preceding observations confirm the
importance of previously described CPX functions at Drosophila adult DLM neuromuscular
synapses and provide further support for previously suggested roles of CPX in determining the
kinetics of regulated exocytosis and neurotransmitter release (An, Grabner et al. 2010; Yang,
Kaeser-Woo et al. 2010).
With regard to the cpx1257, electrophysiological analysis at DLM neuromuscular synapses
revealed a synaptic phenotype that was distinct from the cpxSH1 null mutant. Notably, cpx1257
exhibits a selective increase in mEPSC frequency (Figure 4.2P-R) with no effect on either the
amplitude or waveform of the EPSC (Figure 4.2J-O). These observations indicate that the cpx1257
mutation does not alter the roles of CPX in calcium-triggered neurotransmitter release and control
of neurotransmitter release kinetics, but selectively disrupts clamping of spontaneous vesicle
fusion. The consequences of the cpx1257 mutation with regard to farnesylation of CPX and its clamping function are addressed in the following sections.
The presynaptic distribution of CPX at native synapses: Apparent loss of CPX membrane association in the cpx1257 mutant. The selective effect of the cpx1257 mutation on the clamping function of CPX reflects the underlying molecular lesion, which removes only the last C-terminal amino acid of CPX (the X in the C-terminal CaaX motif; Figure 4.1). The CaaX motif is required for farnesylation and implicated in membrane targeting of some mammalian CPX isoforms. Thus we considered that altered protein localization may contribute to the cpx1257 synaptic phenotype.
This possibility was examined by immunocytochemical studies at DLM neuromuscular synapses.
The favorable spatial resolution in this experimental model (Kawasaki and Ordway 2009) has allowed us to first define the subcellular distribution of CPX within the presynaptic terminal.
CPX is highly enriched at the AZ and also detected in a broader pattern which is colocalized with
116 synaptic vesicle markers (Figure 4.3A-O). Because synaptic vesicle clusters include a high
density of membranes, this distribution of CPX may reflect a more general presynaptic membrane
association, including the plasma membrane, rather than specific localization to SVs. Similar
distributions may be observed for synaptic vesicle proteins which are also found on the plasma
membrane (Kawasaki and Ordway 2009; Kawasaki, Iyer et al. 2011).
Immunocytochemical studies in the cpx1257 mutant, which is predicted to lack C-terminal farnesylation, revealed an altered CPX distribution. Interestingly, strong AZ localization of CPX was preserved in this mutant whereas the more diffuse distribution associated with synaptic vesicles/membranes appeared to be lost (Figure 4.3P-Y). Thus, a mutation which preserves AZ localization of CPX and disrupts association of CPX with synaptic vesicles/membranes produced a synaptic phenotype characterized by an increased mini frequency and no effect on the EPSC amplitude or waveform.
CPX Membrane Association is lost in the CPX1257 Mutant Lacking the CaaX Farnesylation
Motif. Although the presence of a C-terminal farnesylation motif in Drosophila CPX suggests that farnesylation is conserved with respect to mammalian forms, this has not been demonstrated.
Given the established role of farnesylation in promoting protein association with membranes, and the preceding immunocytochemical analysis indicating that membrane-targeting of CPX is disrupted in the cpx1257 mutant, we anticipated that biochemical studies of cpx1257 may reveal
differences in CPX association with membranes and possibly its protein binding partners.
Previous work demonstrated membrane association of prenylated proteins by partitioning tissue
homogenates into aqueous and detergent phases containing cytosolic and membrane-associated
proteins, respectively (Mathias, Chen et al. 2011). In the present study, head homogenates from
flies expressing EGFP-CPX, EGFP-CPX1257 or soluble EGFP alone were prepared with Triton X-
114 and subjected to phase separation as described previously (Bordier 1981; Wang and Coppel
2002; Mathias, Chen et al. 2011). Aqueous and detergent phases were examined by Western
117 analysis for the presence of CPX (Figure 4.4). While soluble EGFP partitioned into the aqueous
phase as expected, EGFP-CPX was found exclusively in the detergent phase as anticipated for
membrane-associated CPX. In contrast, EGFP-CPX1257 partitioned solely into the aqueous phase.
These results indicate increased membrane association of EGFP-CPX with respect to soluble
EGFP and that the mechanism mediating membrane association is abolished in EGFP-CPX1257
lacking the farnesylation motif. Similar results were obtained for endogenous CPX and CPX1257
(Figure 4.S4). Together with our immunocytochemical studies, these observations strongly
support the conclusion that wild-type CPX associates with presynaptic membrane compartments
through its farnesylation. Notably, in the cpx1257 mutant, CPX localization at the AZ and its
function in evoked synaptic vesicle fusion are preserved and thus CPX interactions at the AZ may
not require its association with presynaptic membranes.
Interaction of CPX with SNARE Proteins. The binding interactions of CPX with ternary
SNARE complexes have been well characterized for mammalian forms (McMahon, Missler et al.
1995; Chen, Tomchick et al. 2002; Reim, Wegmeyer et al. 2005). To examine whether these interactions are conserved in Drosophila and possibly altered in the cpx1257 mutant, co- immunoprecipitation studies of Drosophila CPX and neuronal SNAREs were carried out.
Analysis of tissue homogenates was performed using flies expressing EGFP-CPX or EGFP-
CPX1257 in the nervous system. This is advantageous in that efficient immunoprecipitation (IP)
may be achieved using an anti-GFP antibody followed by Western analysis of co-
imumunoprecipated proteins (Zou, Yan et al. 2008). IP of EGFP-CPX produced robust co-IP of
the three endogenous SNARE proteins, SYNTAXIN (SYX), SNAP-25 and neuronal-
SYNAPTOBREVIN (nSYB) from fly head homogenates (Figure 4.5A,B). The specificity of
these interactions was validated by co-IPs from control samples lacking the EGFP-CPX
expression (Figure 4.5A,B). Similar experiments carried out with homogenates prepared from
flies expressing EGFP-CPX1257 revealed interesting, and somewhat surprising, results (Figure
118 4.5C,D). For all three SNARE proteins, co-IP with EGFP-CPX1257 was drastically reduced with respect to EGFP-CPX.
Given that presynaptic membrane association of CPX was disrupted by the CPX1257
mutation (Figure 4.3U-Y and Figure 4.4), these IP results indicate that most of the observed interactions of wild-type CPX with SNARE proteins involve its association with presynaptic membranes. It is noteworthy that CPX interactions with t-SNAREs, which are essentially absent from SVs (Tolar and Pallanck 1998), are drastically reduced in the cpx1257 mutant despite
persistence of CPX localization to the AZ. Taken together, our results indicate that CPX
farnesylation promotes its interaction with the presynaptic plasma membrane, synaptic vesicles
and t- and v-SNAREs (see Discussion). Conversely, note that some persistent interactions
between CPX1257 and SNAREs were also evident in the co-IPs (Figure 4.5D). It is of interest to
consider that these may correspond to the preserved CPX1257 localization at the AZ (Figure 4.3P-
Y) where t-SNAREs are localized (Kawasaki and Ordway 2009) and thought to engage in trans-
SNARE complexes.
4.5 DISCUSSION
Through characterization of a new Drosophila CPX mutant, the present study advances our understanding of the molecular mechanisms mediating CPX function in neurotransmitter release.
Our findings provide new information about the subcellular distribution of CPX with respect to presynaptic membranes, as well as its molecular basis, and lead to a revised model implicating
CPX membrane association as a critical element in its clamping function and in vivo interactions with SNARE proteins (Figure 4.6).
As shown in Figure 6, our working model suggests that membrane association of CPX promotes its interactions with SNARE proteins and is required for the CPX clamping function. At wild-type synapses (Figure 6A), CPX is associated with presynaptic membranes, including clear localization to SVs as well as the AZ where neurotransmitter release sites reside (Figure 4.3). In
119 contrast, the cpx1257 mutant (Figure 4.6B) exhibits an altered CPX distribution reflecting loss of
membrane-associated CPX in general but retention of CPX localization at the AZ (Figure 4.3).
This altered distribution is also associated with loss of CPX membrane targeting in partitioning
assays (Figure 4.4), a drastic reduction in CPX-SNARE binding interactions (Figure 5) and loss
of the CPX clamping function (Figure 4.2) in the cpx1257 mutant. Notably, the activation role
for CPX in evoked neurotransmitter release is retained in this mutant (Figure 4.2), which likely
reflects retention of CPX localization to AZs and its interactions at neurotransmitter release sites,
which may account for residual CPX-SNARE binding (Figure 4.5). Thus our working model suggests that AZ-associated and non-AZ membrane-associated pools of CPX are localized by
different mechanisms, participate in different molecular interactions and make distinct functional
contributions to spontaneous and evoked release.
The preceding model, which is discussed further in the following text, extends and
modifies the conclusions of previous studies. Although the present work first defines the
subcellular distribution of CPX within the presynaptic terminal and the presence of two separable
CPX pools, previous work has demonstrated that farnesylation of CPX is required for localization
to synaptic regions of the axon and association with membranes in biochemical fractionation
studies (Reim, Wegmeyer et al. 2005). Other related studies have mapped the CPX clamping
function to its C terminus (Martin, Hu et al. 2011; Kaeser-Woo, Yang et al. 2012) and the Caax motif (Xue, Lin et al. 2009; Cho, Song et al. 2010). The following discussion of the present findings will focus on the relationship of CPX membrane association to its binding interactions with SNAREs, as well as the molecular mechanisms of the CPX clamping function.
The relationship of CPX membrane targeting to its interactions with SNARE proteins. The present study demonstrates for the first time that CPX is localized to the AZ and complements our previous work demonstrating AZ localization of the t-SNARE proteins, SYX and SNAP-25
(Kawasaki and Ordway 2009). These findings are consistent with models suggesting that CPX
120 interacts with trans-SNARE complexes, which are thought to occur between primed synaptic vesicles and the AZ region of the presynaptic plasma membrane, as well as with t-SNAREs in the absence of v-SNAREs (McMahon, Missler et al. 1995). Furthermore, CPX is also localized more generally to presynaptic membranes, including synaptic vesicles, presumably in the absence of trans-SNARE complexes. Surprisingly, loss of the non-AZ membrane localization in the
CPX1257 mutant (Figures 4.3 and 4.4) produced a drastic reduction in CPX interactions with both v- and t-SNAREs (Figure 4.5) despite the persistence of CPX localization at the AZ (Figure
3). These results indicate that the preponderance of CPX-SNARE interactions are not restricted to the AZ but rather occur more generally with plasma membrane t-SNAREs and synaptic vesicle v-SNAREs. It is of interest to note the farnesylation dependence of both CPX localization to SVs and CPX interaction with the v-SNARE, NSYB. However, because NSYB is also present on the plasma membrane (Kawasaki and Ordway 2009), it remains unclear whether the farnesylation- dependent interactions of CPX with NSYB occur on SVs. As discussed in the following section, such binding of CPX with free v-SNAREs on SVs would represent a novel in vivo interaction
which might contribute to the CPX clamping mechanism.
The molecular mechanisms of the CPX clamping function. The relationship of the preceding
molecular interactions to CPX function may be observed in parallel electrophysiological analysis
of the cpx1257 mutant (Figure 4.2). The finding that evoked neurotransmitter release is normal in
cpx1257 suggests that normal AZ localization of CPX in this mutant preserves a minor
component of the total CPX-SNARE interactions (see Figure 4.5), which occurs at the AZ and is
sufficient to support evoked release. In contrast, the loss of CPX clamping in this mutant suggests
that broader membrane localization of CPX, which is associated with the preponderance of CPX-
SNARE interactions occurring outside of the AZ, is critical to clamping. These observations raise
the possibility that pre-association of CPX with SNARE proteins outside of the AZ, either in the synaptic vesicle or plasma membrane, may contribute to the clamping mechanism. How the role
121 of CPX membrane association may relate to the mechanism of accessory helix function in CPX
clamping remains an interesting question.
It is of interest to consider the relevance of these mechanisms to CPX isoforms which are not farnesylated, such as mammalian CPX1 and 2 (Reim, Wegmeyer et al. 2005). Recent studies indicate that non-farnesylated isoforms of mammalian CPX also exhibit membrane binding activity through a C-terminal amphipathic helix (Seiler. F., Malsam et al. 2009) which is necessary for the CPX clamping function (Kaeser-Woo, Yang et al. 2012). In Drosophila, a non- farnesylated isoform of CPX was shown to be functional in CPX activation and clamping
(Huntwork and Littleton 2007). While the present work focused on farnesylated isoforms of
Drosophila CPX, which are predominant in the nervous system (Figure S2), this non-farnesylated isoform (DmCPX E; Figure 4.S2) exhibited a similar subcellular distribution within the presynaptic terminal (Figure 4.S5). The preceding observations suggest that the clamping function of CPX may depend upon its membrane targeting in general, rather than interactions specific to farnesylated forms.
ACKNOWLEDGMENTS
We thank Troy Littleton (MIT, Cambridge, MA) for kindly providing a cpx null mutant stock and an anti-CPX antibody. Noreen Reist (Colorado State University, Fort Collins, CO) and David
Deitcher (Cornell University, Ithaca, NY) generously provided anti-syt and anti-SNAP25 antibodies, respectively. We thank Richard Ordway (Penn State University) for his continuous encouragement and invaluable discussion throughout this work. This study was supported by
National Institutes of Health Grant R21MH085199-02.
122
4.6 FIGURES
Figure 4.1 A New cpx mutant, cpx1257. (A) Schematic of Drosophila CPX (isoform DmCPX-U) and the nonsense mutation in cpx1257. The cpx1257 mutation removes the last C-terminal amino acid of CPX, corresponding to the X in the C-CaaX motif required for farnesylation. This mutation is isoform-specific in that it occurs in an alternative exon encoding the C-terminus of CPX (highlighted in gray; see also Figure 4.S2). The domain organization shown was adapted from a previous study (Xue, Reim et al. 2007). The accession number for DmCPX-U is AY121629. (B) Western analysis of cpx1257. Western blot of fly head homogenates prepared from wild type (WT) and cpx hemizygotes heterozygous for Df(3R)Exel6140 (Df) which removes the cpx locus, as well as cpx1257/Df and cpxSH1/Df. Although CPX1257 migrates at a slightly higher relative molecular mass in comparison to wild-type CPX, the levels in the cpx1257/Df and +/Df samples are similar, indicating normal CPX expression in cpx1257. CPX is absent in cpxSH1/Df hemizygotes. Tubulin (TUB) was used as an internal loading control.
123
Figure 4.2 Electrophysiological analysis of cpx mutant phenotypes at DLM neuromuscular synapses. (A-I) The cpx null mutant, cpxSH1, exhibits reduced evoked neurotransmitter release, altered release kinetics and loss of the CPX clamping function. (A, B) Representative excitatory
124 postsynaptic current (EPSC) recordings from dorsal longitudinal flight muscle (DLM) neuromuscular synapses of wild type (WT) (A) and cpxSH1/Df(3R)Exel6140 (cpxSH1/Df) (B). cpx null mutant (cpxSH1/Df) synapses exhibit a marked reduction in the EPSC amplitude. (C) The cpxSH1 null mutant exhibits a slowing of the rise and decay times of the EPSC waveform. The EPSC in WT (black) was superimposed with a scaled version of the EPSC in cpxSH1/Df (gray). (D) The synaptic phenotypes in cpxSH1 were rescued by neural (presynaptic) expression of the wild-type CPX protein. Rescue of cpxSH1 was carried out in Appl-GAL4;;UAS-EGFP-cpx cpxSH1/Df(3R)Exel6140 flies. With respect to WT, the cpxSH1 mutant exhibited an decrease in EPSC amplitude (E) as well as an increase in rise (time-to-peak) (F) and in decay (t1/2) (G) kinetics. The EPSC amplitude in WT, cpxSH1/Df and cpxSH1 rescued was 1.97 ± 0.16 μA (n=4), 0.46 ± 0.08 μA (n=4) and 1.57 ± 0.03 μA (n=3), respectively. The EPSC rise time in WT, cpxSH1/Df and cpxSH1 rescued was 0.66 ± 0.02 msec (n=4), 1.27 ± 0.01 msec (n=4) and 0.65 ± 0.07 msec (n=3), respectively. The EPSC decay time in WT, cpxSH1/Df and cpxSH1 rescued was 0.90 ± 0.05 msec (n=4), 1.94 ± 0.10 msec (n=4) and 0.96 ± 0.02 msec (n=3), respectively. (H, I) Representative miniature EPSC (mEPSC) recordings from DLM neuromuscular synapses of WT (H) and cpxSH1/Df (I). The mEPSC frequency in cpxSH1/Df was severely elevated. (J-R) cpx1257 exhibits selective loss of the CPX clamping function. (J, K and L) Representative EPSC recordings from DLM neuromuscular synapses of WT (J), cpx hemizygotes [+/Df(3R)Exel6140 (+/Df)] (K) and cpx1257/Df(3R)Exel6140 (cpx1257/Df) (L). (M, N and O) No change in the EPSC amplitude or waveform was observed in the cpx1257 mutant. The data for WT is the same as shown in E, F and G. The EPSC amplitude in +/Df and cpx1257/Df was 1.95 ± 0.16 μA (n=4) and 1.85 ± 0.08 μA (n=5), respectively. The EPSC rise time in +/Df and cpx1257/Df was 0.65 ± 0.02 msec (n=4) and 0.74 ± 0.04 msec (n=5), respectively. The EPSC decay time in +/Df and cpx1257/Df was 0.99 ± 0.03 msec (n=4) and 1.00 ± 0.05 msec (n=5), respectively. (P, Q and R) Representative miniature EPSC (mEPSC) recordings from DLM neuromuscular synapses of WT (P), +/Df (Q) and cpx1257/Df (R). The mEPSC frequency in cpx1257/Df was severely elevated. Error bars indicate SEM and asterisks denote statistical significance at p 0.05.
125
Figure 4.3 Presynaptic localization of wild-type CPX and mutant CPX1257 at DLM neuromuscular synapses. (A-O) CPX is highly localized at the AZ and also exhibits a broader pattern which is colocalized with synaptic vesicles. Confocal immunofluorescence and native
126 EGFP fluorescence images of DLM neuromuscular synapses showing expression of endogenous CPX (A-E), neuronal (presynaptic) expression of EGFP-CPX (F-O) and either the active zone (AZ) marker, BRP (A-J) or the synaptic vesicle marker, SYT (K-O). The distribution of endogenous CPX (A-E) and EGFP-CPX (F-J) was highly localized at AZs and also detected in a more diffuse pattern which is colocalized with the synaptic vesicle marker, SYT (K-O). Anti- HRP, anti-BRUCHPILOT (BRP) and anti-SYNAPTOTAGMIN (SYT) label the neuronal plasma membrane, presynaptic AZs and synaptic vesicles, respectively. Native EGFP fluorescence is denoted as GFP. (P-Y) Confocal immunofluorescence and native EGFP fluorescence images of DLM neuromuscular synapses showing neuronal expression of EGFP-CPX1257. The distribution of EGFP-CPX1257 was highly localized at AZs (P-T) as observed for wild-type EGFP-CPX, however it lacked the broader distribution pattern which is colocalized with the synaptic vesicle marker, SYT (U-Y).
127
Figure 4.4 CPX membrane association is lost in the CPX1257 mutant which lacks the CaaX farnesylation motif. Head homogenates from flies expressing soluble EGFP alone (a), EGFP- CPX (b) or EGFP-CPX1257 (c) in the nervous system were subjected to phase partitioning in Triton X-114. Aqueous (A) and detergent (B) phases were then analyzed by immunoblotting with an anti-GFP antibody. While soluble EGFP partitioned into the aqueous phase as expected, EGFP-CPX was found exclusively in the detergent phase. In contrast, EGFP-CPX1257 partitioned solely into the aqueous phase. A control cytosolic protein, ACTIN, was detected in only the aqueous phase.
128
Figure 4.5 Interaction of CPX with SNARE proteins. (A, B) Coimmunoprecipitation of EGFP-CPX and SNARE proteins from fly head lysate. An anti-GFP antibody was used to precipitate EGFP-CPX from head lysate of flies expressing EGFP-CPX in the nervous system (b) or control flies (WT) lacking the transgene to confirm specificity of the IP (a). (B) IP of EGFP- CPX and co-IP of endogenous endogenous SNARE proteins, SYNTAXIN (SYX) and neuronal SYNAPTOBREVIN (NSYB). Specific co-IP of endogenous SNARE proteins was observed only in samples expressing EGFP-CPX. Pre-IP INPUT samples are shown in Panel A. (C, D) Comparison of SNARE binding interactions with CPX and the CPX1257 mutant. Coimmunoprecipitation of EGFP-CPX and EGFP-CPX1257 with SNARE proteins from fly head lysate. An anti-GFP antibody was used to precipitate EGFP-CPX or EGFP-CPX1257 from head lysate of flies expressing the respective fusion proteins in the nervous system (b, c). In comparison to wild-type EGFP-CPX, co-IP of all three SNARE proteins with EGFP-CPX1257 was drastically reduced (D). However, note that, for all SNAREs, residual co-IP with EGFP-CPX1257 was detected. Pre-IP INPUT samples (C) show similar levels of EGFP-CPX and EGFP-CPX1257.
129
Figure 4.6 Selective loss of CPX from presynaptic membrane compartments in the cpx1257 mutant suggests that membrane association promotes CPX function in clamping spontaneous release. (A) At wild-type synapses, CPX is associated with presynaptic membranes, including clear localization to SVs as well as the AZ where neurotransmitter release sites reside. Probable plasma membrane association of CPX (PM CPX?) is inferred from extensive farnesylation-dependent CPX binding to t-SNAREs which, given retention of AZ- localized CPX in this mutant, is likely to occur in non-AZ regions of the plasma membrane. (B) In contrast, the cpx1257 mutant exhibits an altered CPX distribution reflecting loss of broadly membrane-associated CPX but retention of CPX localization at the AZ. This altered distribution is associated with loss of CPX membrane targeting in partitioning assays, a drastic reduction in CPX-SNARE binding interactions and loss of the CPX clamping function in the cpx1257 mutant. Notably, the activation role for CPX in evoked neurotransmitter release is retained in this mutant and this is likely to be supported by CPX-SNARE binding at the AZ.
130
4.6 SUPPLEMENTARY FIGURES
Figure 4.S1 A genetic screen to isolate new complexin mutants. Male flies with an isogenized third chromosome (Iso3) were exposed to the mutagen, ethylmethane sulphonate (EMS). Mutagenized males were mated with females carrying the visible second chromosome marker, Lyra (Ly), in trans to a balancer chromosome, TM6c. The F1 male progeny carrying a mutagenized third chromosome (3*) in trans to TM6c were crossed to females carrying the third chromosome deficiencies, Df(3L)GN34 and Df(3R)Exel6140. The latter deficiency removes the cpx locus. The resulting F2 flies heterozygous for a mutagenized third chromosome in trans to the deficiency chromosome were screened for motor defects at 38 °C.
131
Figure 4.S2 CPX isoforms contain distinct C-terminal domains. (A) Two Drosophila CPX (DmCPX) isoforms generated by alternative splicing from the single cpx gene. A main difference between them is that the C-terminus (shown in different shades of gray) is encoded by two different alternative exons. DmCPX isoforms U and E, respectively, represent isoforms predicted to be farnesylated (CaaX containing) or not farnesylated (CaaX lacking). The exon containing the Caax motif is present in most isoforms (www.flybase, see also MATERIALS AND METHODS Section- Relevant information about cpx alternative exons and splice variants). (B) Predominant brain isoforms of Drosophila CPX contain the CaaX farnesylation motif. Western analysis of fly head homogenates prepared from WT, cpx hemizygotes heterozygotes for Df(3R)Exel6140 (+/Df), cpx1257/Df, cpxSH1/Df and cpxSH1/Df rescued by neural expression of the wild-type CPX isoform E. Rescue of cpxSH1 was carried out in Appl-GAL4;;UAS-cpx cpxSH1/Df(3R)Exel6140 flies. The predominance of CaaX-containing CPX isoforms in the Drosophila brain is inferred from the following observations. First, CPX-E migrates at a slightly lower relative molecular mass in comparison to wild-type CPX from +/Df and WT flies and does not appear to be detected
132 in head homogenates containing endogenous CPX (B). Second, the altered migration of CPX1257 appears to shift all of the detectable endogenous CPX signal (no normally migrating CPX is observed). Tubulin (TUB) was used as an internal loading control.
Figure 4.S3 Electrophysiogical analysis of cpx mutant phenotypes at larval neuromuscular synapses. The cpx null mutant, cpxSH1, exhibits reduced evoked neurotransmitter release, altered release kinetics and loss of the CPX clamping function. Representative EPSC recordings from ventral longitudinal muscles of wild type (WT) (A) and cpxSH1/Df(3R)Exel6140 (cpxSH1/Df) (B). cpx null mutant (cpxSH1/Df) synapses exhibit a marked reduction in the EPSC amplitude and a slowing of the rise and decay times of the EPSC waveform. (C) The EPSC in WT (black) was superimposed with a scaled version of the EPSC in cpxSH1/Df (red). These synaptic phenotypes are similar to those observed at adult DLM neuromuscular synapses in this mutant.
133
Figure 4.S4 Endogenous CPX membrane association is lost in the CPX1257 mutant lacking the CaaX farnesylation motif. Head homogenates from heterozygotes for Df(3R)Exel6140 (+/Df) and cpx1257/Df flies were subjected to phase partitioning as described in Figure 4. Aqueous (A) or detergent (B) phases were then analyzed by immunoblotting with an anti-CPX antibody. While wild-type CPX partitioned into the aqueous phase, CPX1257 was found exclusively in the detergent phase. As expected, control cytosolic (ACTIN) and membrane (NSYB) proteins were partitioned into the aqueous and detergent phases, respectively. These findings are similar to those described for the EGFP-CPX and EGFP-CPX1257 fusion proteins.
134
Figure 4.S5 Presynaptic localization of CPX isoform E at DLM neuromuscular synapses. Confocal immunofluorescence images of DLM neuromuscular synapses showing expression of CPX isoform E (CPX-E). To examine the presynaptic localization of CPX E, neural expression of CPX E in a cpx null mutant background was carried out in Appl-GAL4;;UAS-cpx cpxSH1/Df(3R)Exel6140 flies. CPX-E was visualized using the anti-CPX antibody. CPX-E was highly enriched at the AZs and also exhibited a broader distribution resembling that observed for endogenous CPX in WT and neurally-expressed EGFP-CPX (CPX isoform U). Anti-HRP and anti-BRUCHPILOT (BRP) label the neuronal plasma membrane and presynaptic AZs, respectively.
135 Chapter 5
Novel dap160 mutant alleles in Drosophila melanogaster
5.1 INTRODUCTION
In the process of chemical synaptic transmission, chemical neurotransmitter is released from the presynaptic axon and produces electrical excitation or inhibition of the postsynaptic cell.
Neurotransmitter release occurs by exocytosis when neurotransmitter-filled synaptic vesicles within the presynaptic terminal fuse with the presynaptic plasma membrane. Following synaptic vesicle fusion, vesicle membranes are retrieved from presynaptic plasma membrane through a process called synaptic vesicle endocytosis. This involves a complex of proteins including
CLATHRIN, AP-2 complex, AP180, Eps15, DYNAMIN, ENDOPHILIN, SYNAPTOJANIN,
SYNAPTOTAGMIN and the protein of interest in the current study, DAP160 (Jarousse and Kelly
2001).
DAP160 (Dynamin-Associated Protein 160kDa), a Drosophila homolog of Intersectin, is
a multidomain adaptor protein involved in synaptic vesicle endocytosis. It was identified in
Drosophila by virtue of its ability to bind DYNAMIN (Roos and Kelly 1998). It contains two N
terminal EH (Eps15 Homology) domains, a central coiled coil domain and four SH3 (Src
Homology) domains. The SH3 domains, which are of special significance in the current study,
bind to the proline rich domains (PRD) of different endocytic proteins like DYNAMIN (Roos and
Kelly 1998). Genetic studies provide evidence that DAP160 is required for the localization of
several endocytic proteins including DYNAMIN, and hence may play a key role as a scaffold
protein which anchors proteins to the site of endocytosis (Koh, Verstreken et al. 2004; Marie,
Sweeney et al. 2004). Here we report isolation of two new recessive mutations in the Drosophila
dap160 gene which may provide new insights into the structural basis of DAP160 function.
136 5.2 RESULTS
Identification of two new dap160 mutants. Two new mutant alleles of dap160 were isolated in forward genetic screens recently conducted in our laboratory. These screens involved chemical mutagenesis with ethane methylsulfonate (EMS) followed by screening for mutants exhibiting altered behavior. One allele, dap160130, was recovered in a screen for new temperature-sensitive
(TS) mutants exhibiting rapid paralysis or motor impairment at elevated temperature. The second
allele, dap160EC1 (enhancer of cacTS2), was isolated in a screen for modifiers of the TS paralytic presynaptic calcium channel mutant, cacTS2(Kawasaki, Felling et al. 2000; Kawasaki, Collins et
al. 2002; Brooks, Felling et al. 2003). The cacTS2 mutant was recovered previously as an enhancer
of the TS paralytic mutant, comatose (Dellinger, Felling et al. 2000). The two new dap160
mutants produced rapid TS paralysis (Figure 5.1). As shown, the dap160EC1 mutant, which was recovered as an enhancer of cacTS2, also exhibits a TS paralytic phenotype in a cac+ genetic
background. These two mutants were first identified as dap160 alleles by non-complementation
of TS paralysis with previously characterized dap160 mutants, dap160Δ , dap160Δ2 and dap160Δ6
(Koh, Verstreken et al. 2004; Marie, Sweeney et al. 2004).
Molecular characterization of new dap160 mutants. PCR amplification of the dap160 ORF from genomic DNA and subsequent sequence analysis led to the identification of the molecular lesions in these two mutant alleles. Sequences from the mutants were compared with those of the parent chromosome used in the mutagenesis. The mutation in dap160130 is a C to T transition leading to a premature stop codon (Figure 5.2A) which truncates the protein after SH3B (Figure
5.2B). In dap160EC1, the molecular lesion is a G to A transition at the 3’ splice site preceding
Exon 10 (Figure 5.2A). This mutation would most likely affect the normal splicing of the dap160
pre-mRNA by skipping exon10 or use of a cryptic splice site. Splicing out exon10 may create a
frame shift and lead to truncation (Figure 5.2B). Thus, this 3’ splice-site mutation would lead to
deletion of the SH3C and SH3D domains (Figure 5.2B). The consequence of this lesion is similar
137 to that of dap160130, in which truncation occurs before the third SH3 domain and thereby deletes the SH3C and SH3D domains (Figure 5.2B).
Based on the nature of their molecular lesions, the mutant proteins are expected to be of lower molecular weight than the wild-type DAP160 protein. The predicted molecular mass of wild-type DAP160 is 120kDa, yet the apparent molecular mass by SDS-PAGE is higher than that
(~ 180 kDa) (Figure 5.3), possibly due to the abundance of negatively charged amino acids in the
SH3 domains (Roos and Kelly 1998). The predicted sizes of DAP160130 and DAP160EC1 are
93kDa and 107kDa, respectively. Consistent with this prediction, the western analysis indicated
that the DAP160 protein in flies homozygous for either dap160130 or dap160EC1 migrates at
~140kDa and 160kDa, respectively (Figure 5.3).
5.3 DISCUSSION
Previous studies of dap160 null mutant flies show severe defects in the formation of neuromuscular junctions (NMJs) and in synaptic vesicle recycling (Koh, Verstreken et al. 2004;
Marie, Sweeney et al. 2004). Thus, DAP160 plays an important role in sustaining neurotransmitter release at synapses. Furthermore, all the available dap160 mutations are homozygous lethal, suggesting that dap160 is an essential gene (Koh, Verstreken et al. 2004;
Marie, Sweeney et al. 2004). The dap160 mutants described in the present study are TS paralytic
mutants, but their phenotypes are not strictly TS because they exhibit motor defects 20°C as well.
These TS mutants are unlike classical TS mutants, like shibire, in which a single amino acid
substitution leads to malfunctioning of the mutant proteins at high temperatures. The molecular
lesion in these two dap160 mutant alleles deletes two of the SH3 domains in the DAP160 protein
that are necessary for interactions with other endocytic proteins (Described below). The severe
nature of these mutants may contribute to their phenotypes at permissive temperatures. These TS
dap160 mutants can be used to analyze the in vivo physiological functions of DAP160 at mature synapses.
138 DAP160 and its homolog INTERSECTIN, have multiple SH3 domains in tandem (5 in vertebrates and C.elegans and 4 in D.melanogaster). DAP160 binds to proteins containing a PRD
via its SH3 domains. Each of the four SH3 domains has different binding partners and
biochemical characterization has identified some of the binding partners for the different SH3
domains, including DYNAMIN, SYNAPSIN and Drosophila SYNAPTOJANIN (Roos and Kelly
1998). These studies were done in vitro and thus systematic in vivo characterization of the SH3
domain function may help to define the precise roles of these domains and their binding
interactions in synaptic transmission. Thus the specific loss of SH3 domains, C and D, in the mutants described here (Figure 5.2B) may prove useful in the in vivo structure-function analysis of DAP160.
ACKNOWLEDMENTS This work was supported by the National Institutes of Health, National
Institute of Neurological Diseases and Stroke (Grant R01 NS065983). We thank Dr. Graeme
Davis (University of California, San Francisco, CA) for generously providing the anti-DAP160 antibody. We are grateful to Dr Hugo Bellen (Baylor College of Medicine, Houston, TX) for providing us with the dap160 alleles, dap160Δ1, dap160Δ2 and dap160Δ6. We would also like to
acknowledge Bloomington Stock Center for the deficiency stock, Df(2L)Exel6047, used in this
study. DNA sequencing was performed in the Pennsylvania State University Sequencing Facility,
University Park, PA with excellent support from the center staff.
139 5.4 FIGURES
Figure 5.1. Genetic screen identifies two new dap160 mutants. dap160EC1 and dap160130 both exhibit rapid paralysis at 38°C. The asterisk denotes statistical significance with respect to WT. WT flies are Canton-S. Behavioral analysis was performed as described previously (Brooks, Felling et al. 2003). Briefly, flies were placed in a vial preheated to 38°C by immersion in a water bath. Two-day-old flies reared at 20ºC were tested in groups of six for TS behavioral phenotypes. Five groups of flies (n=5) were tested for each genotype. Time for 50% paralysis represents the time at which 3 flies were no longer able to stand. Behavior test for wild-type (WT) flies was truncated after 20 minutes. In tests exceeding 5 minute duration, 3ml of water was added to the rayon plug after 5 minutes to prevent dehydration.
140
Figure 5.2. New dap160 mutants. (A) Schematic of dap160 gene structure showing the two new mutations. The numbered boxes represent exons and the lines joining them are the introns. The cross hatch areas are the 5’ and 3’ UTRs (Untranslated regions). For identifying the molecular lesions in the new mutants, genomic DNA from the homozygous mutant flies Iso2 flies (parent strain) were to PCR amplify the dap160 gene using the forward primer, dap160A (5’GTGCGAG GATCTGTACAAGG 3’), and the reverse primer, dap160B (5’GCAATGACCTTGTCTGCATG 3’). The PCR products were gel purified, quantified and sent for sequencing using primers dap160A, dap160B and dap160C (5’CGGTAGAATCTCTTGATCAG 3’). Sequences from the mutants were compared with those from Iso2 and molecular lesion in dap160EC1 and dap160130
were identified. (B) Domain structure of DAP160. DAP160 contain two amino-terminal Eps15 homology domains (EH, blue box), a coiled-coil domain (CC, yellow box), and four Src Q24 homology 3 domains (SH3, orange boxes). The locations of two existing, dap160 and dap160EMS , and two new dap160 lesions, dap160EC1 and dap160130 , are indicated.
141
Figure 5.3. Western Blot analysis of dap160 mutants. (A) Western Blot analysis of fly head homogenates prepared from WT, dap160 mutants (dap160EC1 and dap160130) and flies heterozygous for Df(2L)Exel6047 which removes the dap160 locus. Tubulin (TUB) was used as an internal loading control.
142 Chapter 6
Discussion
The present work has focused on understanding the molecular mechanisms of synaptic transmission. Genetic screens have led to the isolation of two new mutants and the present work discusses the functional characterization of the corresponding gene products in synaptic transmission, one in exocytosis and the other in endocytosis. The first part of our work utilized genetic, biochemical, electrophysiological and ultrastructural approaches to identify a novel role for DAB proteins in synaptic vesicle endocytosis. Furthermore, our studies of the spatial organization of DAB proteins and CLATHRIN as well as analysis of endocytic proteins such as
DYNAMIN indicate a novel role for CLATHRIN-mediated endocytic mechanisms in rapid clearance of neurotransmitter release sites for subsequent vesicle priming and refilling of the release-ready vesicle pool. In the second primary project, we have further explored the in vivo function of CPX at native synapses in adult DLM neuromuscular synapses. Isolation of a new cpx mutant has led to new insights into the molecular basis of CPX clamping function. Moreover, our studies indicate an in vivo role for CPX in regulating the kinetics of neurotransmitter release.
6.1 DISABLED (DAB) in Synaptic Transmission
Electrophysiological analysis of dabEC1 mutant synapses showed an activity-dependent
reduction in neurotransmitter release leading to strong short term synaptic depression. Such an
activity-dependent reduction of neurotransmitter release is observed in several endocytic mutants
such as shibire and dap160 as well as in a knockdown of CHC by RNAi. The established role of
DAB proteins in CLATHRIN-mediated endocytosis, along with the phenotype observed in synaptic electrophysiology suggested a novel function for DAB proteins in synaptic vesicle endocytosis.
Mouse DAB2 (mDAB2) and Ce-DAB1 have been shown to bind to clathrin (Mishra,
Keyel et al. 2002; Kamikura and Cooper 2006) via conserved clathrin-boxes that are found in
143 many endocytic proteins including Drosophila DAB (Shiina, Arai et al. 2001). Mutational studies of the clathrin-box in mDAB2 and Ce-DAB-1 have shown that these domains are important for clathrin binding of DABs (Mishra, Keyel et al. 2002; Kamikura and Cooper 2006). The conservation of the clathrin-boxes in DABs and the evidence that they are functional in C. elegans and mammals suggests that Drosophila Dab may be a clathrin-binding protein as well.
Consistent with this prediction, immunocytochemical analysis at DLM neuromuscular synapses show that DAB and CLATHRIN are highly co-localized. Thus dDAB functions in SV endocytosis appear to involve interactions with CLATHRIN. However two populations of dDAB and CLATHRIN puncta were observed with the presynaptic terminal, one which localizes with
AZs (co-AZ) and another population that did not (non-AZ) (Figure.3.2c-v in Chapter.3). The rapid onset synaptic phenotypes observed are likely to reflect localization and function of dDAB and CLATHRIN at the AZ, which is quite unusual among proteins involved in SV endocytosis.
Other endocytic proteins such as DYNAMIN and DAP160 are more broadly distributed and often enriched in regions surrounding the AZ (the peri-AZ). The non-AZ population of DAB puncta are either adjacent to or spatially separated from the AZ, but do not surround the AZ in the peri-AZ
(PAZ) as observed for other endocytic proteins. CLATHRIN not only mediates vesicle formation at the plasma membrane, but also participates in CCV-mediated protein sorting at the trans-Golgi network and endosomal compartments (Puertollano 2004). According to one of the possible pathways of synaptic vesicle recycling, the synaptic vesicle proteins may completely disperse in the plasma membrane following fusion, removing all identity of the original vesicle membrane.
These synaptic vesicle proteins may then get retrieved from the plasma membrane by clathrin mediated endocytic machinery, but need to be sorted in endosomes to form new synaptic vesicles
(Rizzoli and Jahn 2007). Bulk endocytosis, the major alternative pathway for SV recycling when
CME is limiting (Glyvuk, Tsytsyura et al. 2010) proceeds via endosomal sorting (Geumann,
Schafer et al. 2010). In studies investigating the dynamics of EGFR-AP (epidermal growth factor
144 receptor-adaptor protein) interactions and sub-cellular localization of clathrin, it was reported that
adaptor protein AP1, CLATHRIN and accessory factor EPS15 are present in the EGFR-
containing endosomes (Sorkina, Bild et al. 1999). Further, a more recent study performed at
mouse hippocampal synapses provides evidence for synaptic vesicle reformation by AP-1 and clathrin mediated mechanisms from endosomes (Glyvuk, Tsytsyura et al. 2010). These results suggest that clathrin mediated endocytic mechanisms operate at endosomes as well. Thus, the non-AZ CLATHRIN puncta may reflect the presence of CLATHRIN in endosomal compartments, and DAB may function along with CLATHRIN in endocytic mechanisms operating in these compartments. Intensive efforts to identify the sub-cellular domain corresponding to the non-AZ DAB puncta using markers, like RAB5 (early endosome marker),
RAB7 (late endosome marker), RAB11 (recycling endosome marker) and LAMP (lysosome marker) for intracellular membrane compartments have been unsuccessful. So far, none of the endosomal compartment markers have shown co-localization with dDAB and CLATHRIN, as shown for the FVYE domain marker (Gillooly DJ, Morrow IC et al. 2000) for endosomes containing phosphatidylinositol-3-phosphate (Figure3.S5 in Chapter 3). This remains a very interesting issue requiring further investigation. One approach to address this may be immuno-
Electron Microscopy which uses antibody labeled with gold particles to determine protein localization to subcellular organelles with nanometer resolution (Faulk and Taylor 1971). This approach may provide clues to the sub-cellular localization of DAB and CLATHRIN. We are currently optimizing this approach in our laboratory.
Our results reveal that DAB, together with CLATHRIN, functions in SV endocytosis, possibly as an adaptor protein. DAB contains a PTB (Phosphotyrosine binding) domain, which has been shown to bind phosphoinositides and NPXY motifs in certain receptor proteins. NPXY is a sorting motif found in the cytosolic domain of cargo proteins that is recognized by the PTB containing adaptor proteins. In an effort to identify potential cargo proteins for dDAB, an initial
145 survey of Drosophila synaptic vesicle proteins revealed an NPXY motif in the cytoplasmic
domain of SYNAPTOTAGMIN I (dSYT1). SYT1 proteins play an important role in synaptic
transmission by serving as a calcium sensor for neurotransmitter release and also function in
synaptic vesicle endocytosis. The dSYT1 NPXY motif (residues 387-390; NPYY), which is identical in mammalian SYT1 proteins and conserved in other SYTs, is located within the C2B domain near a basic region previously implicated in SYT1-AP-2 interactions. This motif is of great potential interest because no classical endocytosis signals have been identified previously in
SYTs and sorting of dSYT1 could be mediated by dDAB as well as AP2. However, immunocytochemical studies indicate WT distribution of dSYT1 in dab mutant synapses.
Furthermore, the loss of dDAB protein shows phenotypes similar to loss of other endocytic proteins, including a rapid activity dependent reduction in neurotransmitter release, slowed recovery in PPD (paired pulse paradigm), enlarged membrane cisternae, and persistence or accumulation of AZ associated and docked synaptic vesicles. These phenotypes reflect a general loss of endocytic function and not just mis-sorting of a synaptic vesicle protein. This suggests that either dDAB does not participate in dSYT1 sorting or sufficient redundancy is provided by interactions of SYT1 with at least two other CLATHRIN-associated adaptor proteins, AP-2 and stonins. Despite the severe nature of the molecular lesion in dabEC1, electrophysiological data during prolonged stimulation suggests the persistence of residual SV endocytosis and reflects redundancy in the mechanisms of SV endocytosis.
After exocytosis, synaptic vesicle membrane proteins and lipids need to be recycled at the periactive zone (PAZ) to restore functional synaptic vesicle pools (Haucke, Neher et al. 2011).
Also, complete fusion of synaptic vesicle membranes results in the accumulation of synaptic vesicle proteins at the release site, and may cause a functional block of the previously used active zone slot (Haucke, Neher et al. 2011). Studies at the Calyx of Held synapses indicate that the number of release sites per active zone may be fixed (three at this synapse) and that each release
146 site must be used several times per second to support the neurotransmitter release during high
frequency stimulation (Neher 2010). Further, during such high-frequency stimulation, the demand for clearing the release sites surpasses the need to recycle synaptic vesicles, which outnumber release sites by several orders of magnitude (Haucke, Neher et al. 2011). Thus clearance of
release sites may be the rate-limiting step for neurotransmitter release and delayed clearance may
lead to short term depression (Neher 2010). Our lab’s previous analysis in the shibire
(DYNAMIN) mutant demonstrated rapid synaptic fatigue within 20 ms of stimulation that could not be explained by synaptic vesicle depletion in this mutant (Kawasaki, Hazen et al. 2000).
Rather it was suggested that DYNAMIN may function in rapid clearance of the endocytic intermediates from the release sites to maintain neurotransmitter release during synaptic activity.
Recent studies at the Calyx of Held have further established a rapid role for DYNAMIN and AP2 in maintaining neurotransmitter release, which occurs prior to the formation of endocytic vesicles, and directly demonstrated its requirement for fast refilling of the release-ready vesicle pool
(Hosoi, Holt et al. 2009; Wu X.S. and L.G. 2009). These results indicate that the processes of exocytosis and endocytosis are linked and that clearance of the release sites is essential for maintenance of neurotransmitter release. At present this coupling between exocytosis and endocytosis is poorly understood, with the mechanisms of synaptic vesicle protein/endocytic intemediate removal from the release site unknown. The present study provides insights into the mechanisms of exocytic-endocytic coupling. The ultrastructural analysis of the dab mutant, shi mutant and CHC knockdown synapses revealed persistence or accumulation of docked synaptic vesicles at the AZ (Figure.3 and 4B-F in Chapter 3). This result, along with the localization of dDAB and CLATHRIN proteins at the AZ (Figure 3.2c-v in Chapter.3), indicates a post-docking role for clathrin mediated endocytic mechanisms at AZ in sustaining neurotransmitter release.
Also, slowed recovery in PPD in these three endocytic mutants, which is similar to that observed in the SNAP25-TS mutant (Figure 3.4g and h), further suggests a rapid role for endocytic
147 mechanisms in sustaining neurotransmitter release. The localization of dDAB and CLATHRIN at
the AZ, along with their rapid role in sustaining neurotransmitter release suggest that
CLATHRIN-mediated endocytic machinery first operates at the AZ to clear neurotransmitter
release sites and subsequently mediates vesicle formation at the PAZ. In our previous work, we
have demonstrated an activity dependent redistribution of t-SNARE proteins from AZ to PAZ
regions, which reflects their participation in plasma membrane cis-SNARE complexes (Kawasaki
and Ordway 2009).
The results reported here, taken together with our previous work (Kawasaki and Ordway
2009) support a working model (Figure 3.5 in Chapter.3) where CLATHRIN-mediated endocytic
mechanisms facilitate rapid clearance of synaptic vesicle proteins from neurotransmitter release
sites. This is necessary for synaptic vesicle priming and refilling of the readily releasable vesicle
pool.
6.2 COMPLEXIN in Synaptic Transmission
Synaptic vesicle exocytosis is a tightly regulated process that is specialized to occur
within microseconds of calcium influx. Although SNARE proteins are essential, they require
other proteins to regulate the speed and precision of synaptic vesicle fusion. Research in different
experimental systems over the past decade has revealed a role for CPX in regulating SNARE
function in synaptic vesicle exocytosis. CPXs perform dual functions in promoting evoked
neurotransmitter release and in clamping spontaneous release. Furthermore, structure-function
analyses have demonstrated that different domains in CPX contribute to their distinct functions.
In vitro liposome fusion assays have predicted a fusion clamp model that is supported by
evidence from CPX knockdown studies in cultured mouse neurons demonstrating that clamping
of spontaneous fusion is abrogated (Giraudo, Eng et al. 2006; Schaub, Lu et al. 2006; Maximov,
Tang et al. 2009). While analysis of CPX knockdown in mice shows decreased evoked release
along with elevated spontaneous release (Maximov, Tang et al. 2009), CPX knockout mice
148 exhibit only decreased evoked release with no change in spontaneous release suggesting only a
facilitatory role for CPXs in neurotransmitter release (Reim, Mansour et al. 2001; Xue,
Stradomska et al. 2008). In these knockout mice three (CPXI,II and III) out of the four CPX isoforms have been deleted, although the fourth isoform, CPXIV, has not been detected in the brain so far, there is a possibility of compensation by this isoform (Xue, Stradomska et al. 2008).
Also, these knockout mice die within few hours after birth and so all recordings from the brain
slices have been carried out at immature synapses. To circumvent this problem, most of the
analysis was done in autaptic hippocampal neurons (Reim, Mansour et al. 2001). An autapse is a
self-synapse, where there neuron forms a connection with itself. Although autaptic cultures allow the study of synaptic transmission in a single cell, providing a huge advantage in the analysis of synaptic function, autapses also exhibit some disadvantages. The synaptic responses widely vary from one autaptic culture to another. Also the autapses are not physiologically identical to regular synapses from between a pre- and postsynaptic neuron (Maximov, Pang et al. 2007). In one case,
significant differences were observed in analyses of synaptic transmission in synaptotagmin (syt)
deficient neurons performed in autaptic cultures and in Drosophila larval neuromuscular
synapses. Electrophysiological analysis at Drosophila larval neuromuscular synapses in syt null
mutants exhibited substantial increase in spontaneous neurotransmitter release (elevated mini
frequency) (Littleton, Stern et al. 1993; Broadie, Bellen et al. 1994; DiAntonio and Schwarz
1994). In contrast, autaptic hippocampal cultures from mouse syt null mutants exhibited normal
mini frequency (Geppert, Goda et al. 1994). Subsequent work in cortical neurons from SYT
knockout mice demonstrated elevated mini frequency(Pang, Sun et al. 2006), similar to data from
Drosophila neuromuscular synapses. Thus the absence of clamping function of CPX as observed
in these autaptic cultures may not be representative of CPX function at native synapses. In
addition, in mice CPX is fully expressed and localized to synapses at approximately 20 days after
birth (Reim, Wegmeyer et al. 2005), and the recordings from the hippocampal autaptic cultures
149 from CPX triple knockout mice were done after 10-16 days of growth (Xue, Stradomska et al.
2008). At this stage CPX may not be localized to synapses and so these studies may not reveal all
the aspects of CPX function at synapse.
In contrast to mammals, Drosophila has only one cpx gene and a null mutant generated in
a previous study (Huntwork and Littleton 2007) represents a true knockout of CPX. Previous
analysis of this cpx null mutant indicates an in vivo functional role for CPX as a fusion clamp
(Huntwork and Littleton 2007). Electrophysiological recordings in this study were carried out at the larval neuromuscular synapse, in contrast to our work at adult neuromuscular synapses. The latter model has several attractive properties for genetic and molecular analysis of synaptic function. Adult DLM neuromuscular synapses exhibit a high degree of similarity to mammalian synapses (cerebellar CF-PC synapses) with regard to the properties of synaptic function. In addition, DLM synapse morphology, including highly branched axons with small boutons containing only one or two active zones, provides improved spatial resolution and is thus advantageous for the study of protein distribution within presynaptic terminals. In the previous
work from our lab, we used this synapse model to examine the distribution of t-SNAREs at AZs and their activity-dependent redistribution in the comatose (NSF) mutant (Kawasaki and Ordway
2009). In the present study these approaches were applied to AZ localization of endocytic proteins.
Despite decades of research, the precise function of CPX in regulation synaptic vesicle fusion is still controversial. Our current work on CPX function, initiated by the isolation of a new cpx mutant, provides new insights into the molecular basis of CPX functions and interactions within the neurotransmitter release apparatus.
CPX has an important role in neurotransmitter release kinetics:
The altered EPSC waveform in the cpx null mutant (Figure 4.2C) suggests a novel presynaptic role for CPX in regulating neurotransmitter release kinetics. There have been no
150 reports of altered EPSC waveform in mouse cpx mutants, although a possible role for CPX in
clamping asynchronous release has been reported in one study (Yang, Kaeser-Woo et al. 2010).
Recent work in PC12 cells using sophisticated imaging methods have implicated a role for CPX
in controlling the properties of fusion pore formation between secretory granule and surface
membrane (An, Grabner et al. 2010). These and our preliminary observations from
electrophysiological analysis suggest two alternative hypotheses:
(1) CPX contributes to the operation of the fusion pore.
(2) CPX contributes to the synchronization of normal synaptic vesicle fusion events
in response to an action potential.
The primary approach to distinguish between these possible roles of CPX will involve
comparison of the EPSC and mEPSC waveforms. If the altered EPSC waveform in the cpx null
mutant reflects a contribution of CPX to the operation of fusion pore, then the mEPSC waveform
would be similarly altered. Alternatively, asynchronous but otherwise normal synaptic vesicle
fusion events would produce a normal mEPSC waveform but an altered EPSC.
Fusion Clamp Model:
The in vivo electrophysiological analysis of the two complexin mutants, cpxSH1 and
cpx1257 have further defined the dual functions of CPX in supporting evoked neurotransmitter
release and in clamping spontaneous release. Our findings provide evidence to support the dual
functions of CPX (Figure 4.2B, E,I and R) and are consistent with the previous studies in cpx null
mutant in Drosophila (Huntwork and Littleton 2007) and CPX knockdown in mouse cortical
neurons (Maximov, Tang et al. 2009).
The presence of elevated minis together with normal evoked release in the cpx1257 mutant
(Figure.4.2L, M and R), where only the last amino acid is deleted (Figure 4.1A), shows that the
two functions are separable in vivo. The cpx1257 mutant has a mutation in the C-terminal CAAX motif. This motif is a conserved signal for farnesylation and has been implicated in membrane
151 targeting of mammalian CPXs III and IV (Reim, Wegmeyer et al. 2005). We have demonstrated
by sub-cellular fractionation (Phase Partitioning) in flies expressing EGFP-CPX and EGFP-
CPX1257 that WT CPX is membrane associated (Figure 4.4), possibly mediated by the farnesylation of the cysteine in the CAAX motif at the C-terminal. This membrane association is lost in EGFP-CPX1257 as predicted for this presumably farnesylation deficient mutant. Our
findings from immunocytochemical studies at native synapses have defined the sub-cellular
localization of CPX for the first time. The sub-cellular localization of CPX in WT and cpx1257
mutant synapses (Figure 4.3) and electrophysiology data at cpx1257 mutant synapses, lead us to
hypothesize that farnesylation is required for association of CPX with the SVs/presynaptic
membranes but not for CPX localization to the AZ and that these two populations make distinct
functional contributions to spontaneous and evoked release.
Studies in C.elegans and cultured mouse neurons have also shown that the CPX C-
terminal is necessary for clamping spontaneous release (Martin, Hu et al. 2011; Kaeser-Woo,
Yang et al. 2012). However, the C-terminal domain swapping studies between CPXI and CPXIII in cultured mouse neurons, show that C-terminal domain in CPXIII, which has the CAAX motif, does not rescue the increased mini frequency observed in the knockdown studies (Kaeser-Woo,
Yang et al. 2012). These findings suggested that the membrane anchoring of CPXIII via its C- terminal farnesylation blocks its clamping function. In contrast, recent work in Drosophila has shown that mCPXIII and IV can rescue the elevated mini frequency in the Drosophila cpx null mutant, with the rescue by mCPXIV being almost comparable to rescue by Drosophila CPX
(Cho, Song et al. 2010). This study also investigated the role of the CPX C-terminal CAAX motif in Drosophila, by generating chimeras of DmCPX and mCPXI, in which the last 4 residues were swapped, thereby generating DmCPXΔCAAX/+MFKK and mCPXIΔMFKK/+CAAX. Their results
demonstrate that DmCPXΔCAAX/+MFKK does not rescue the elevated mini phenotype observed in
the cpx null mutant (Cho, Song et al. 2010). In another study, it has been suggested that C-
152 terminal CAAX motif of Drosophila CPX is also required to clamp spontaneous release in mouse
autaptic neurons, as indicated by elevated mini frequency in CPX TKO mice expressing
Drosophila CPX with the mutated cysteine in CAAX motif (Xue, Lin et al. 2009). These results
suggest that CPX farnesylation plays a key role in clamping spontaneous neurotransmitter
release. However, there are mammalian CPX isoforms that are not farnesylated, such as CPX I
and II (Reim, Wegmeyer et al. 2005). But recent studies indicate that these non-farnesylated
isoforms of mammalian CPX also exhibit membrane association through a C-terminal
amphipathic helix (Seiler. F., Malsam et al. 2009) , which is necessary for clamping function. In
Drosophila, although the farnesylated forms of CPX are predominant in the nervous system
(Figure 4.S2), a non-farnesylated isform (DmCPX E) was shown to be functional in CPX activation and clamping (Huntwork and Littleton 2007) and we found that this isoform, DmCPX
E, exhibits similar sub-cellular localization to the farnesylated isform in the presynaptic terminal
(Figure 4.S5). Taken together with our findings, these results suggest that the clamping function of CPX may depend on its membrane association in general, rather than interactions specific to the farnesylated forms.
Interactions of CPX and CPX1257 with other synaptic proteins:
Strong co-immunoprecipitation (co-IP) of all SNAREs with EGFP-CPX has confirmed the robust SNARE binding of CPX in Drosophila (Figure 4.5A and B). In vitro assays indicate that CPX binds primarily to trans-SNARE complexes rather than free SNAREs (Pabst, Hazzard et al. 2000), but several studies report binding of CPX to SYNTAXIN and/or binary t-SNARE complexes (McMahon, Missler et al. 1995; Weninger, Bowen et al. 2008). Thus, these co-IP results may reflect the interactions of CPX with free SNAREs, binary t-SNARE complexes
(SYNTAXIN: SNAP25 complex) and/or trans-SNARE complexes. Although t-SNAREs are well distributed throughout the presynaptic plasma membrane, they are highly localized at the AZ
(Kawasaki and Ordway 2009). Hence our observation of AZ localization of CPX (Figure 4.3, A-
153 J) is consistent with the models suggesting CPX interaction with trans-SNARE complexes and t-
SNAREs.
Surprisingly, in the CPX1257 mutant, the loss of non-AZ association/ SV association resulted in a marked reduction of CPX interactions with both the v- and t-SNAREs, as indicated by the co-IP results (Figure 4.5, C and D). In this mutant, we see reduction of CPX-SNARE interactions despite the persistence of CPX localization at the AZ (Figure 4.3). These results indicate that CPX-SNARE interactions are not restricted to the AZ but rather occur more generally with the plasma membrane t-SNAREs and SV v-SNAREs. Further, normal evoked neurotransmitter release and normal AZ localization of CPX in cpx1257 mutant suggests that a minor component of the total CPX-SNARE interactions is preserved in this mutant (Figure 4.5 D) and is sufficient to support evoked release. The lack of association of non-farnesylated CPX1257
with the SVs and the reduced co-IP of the v-SNARE, n-SYB, suggests a novel interaction of CPX
with n-SYB. The membrane association of farnesylated CPX may facilitate an interaction with n-
SYB (and other vesicle proteins such as VGLUT). This pre-association of CPX with n-SYB on the SV may be required for clamping the spontaneous events. In general, these studies suggest that membrane association of CPX is very important for its interactions with other proteins.
In summary, Drosophila CPX undergoes post-translational modification and gets farnesylated at the cysteine residue in the C-terminal CAAX motif. This farnesylation mediates membrane association and targets CPX to membranes. CPX is highly concentrated at the AZ
(AZ-pool) and is also associated with the SV (non-AZ membrane associated pool). The AZ population seems to be important in supporting evoked fusion, whereas the pre-association of non-AZ pool of CPX with SNARE proteins outside the AZ (in synaptic vesicle or plasma membrane) may contribute to the clamping mechanism.
154 APPENDIX
Novel syt1 mutant allele in Drosophila melanogaster
Synaptic transmission is fundamental to neural function. At chemical synapses, neurons
communicate with one another by transmitting electrical signals through the release of
neurotransmitters (Del Castillo 1954). During chemical synaptic transmission, arrival of action
potential triggers the influx of Ca2+ via voltage-gated calcium channels (VGCC). These neurotransmitters then diffuse across the synaptic cleft and bind to receptors in the postsynaptic membrane, which results in activation or inhibition of the post-synaptic membrane. The influx of
Ca2+ is essential for evoked neurotransmitter release (Mulkey and Zucker 1991).
SYNAPTOTAGMIN1 (SYT1), a transmembrane protein of the synaptic vesicles is thought to be
the calcium sensor, which triggers vesicle fusion [reviewed in (Sudhof 2004)].
SYTs contain a single short N-terminal transmembrane region and two C-terminal calcium binding domains, C2A and C2B (Bai and Chapman 2004). SYT1 is an integral vesicle membrane protein that is anchored to the vesicle through its N-terminal transmembrane domain
(Bai and Chapman 2004). The C2A and C2B domains adopt beta-sandwich structures with three
flexible loops extending from one end and bind three and two Ca2+ ions, respectively. Through
these flexible loops, the C2 domains bind to phospholipids in the vesicle and plasma membranes
in a calcium dependent manner [reviewed in (Rizo and Rosenmund 2008)]. Evidence from mutational analysis has demonstrated that SYTI acts as a calcium sensor in neurotransmitter release (Fernandez-Chacon, Konigstorfer et al. 2001). Ca2+ binding to C2B domain is more important for neurotransmitter release as a larger decrease in release is observed for mutations in
C2B domain relative to C2A (Mackler, Drummond et al. 2002; Robinson, Ranjan et al. 2002). In addition, recent studies indicate that calcium dependent C2B domain binding to the vesicle and
155 plasma membranes can bring these membranes together [reviewed in (Rizo and Rosenmund
2008)].
SYT1 interacts with individual SNAREs and SNARE complexes, but their association with the SNARE complex most likely triggers the release. The exact mechanism of how this binding results in release is not clear, but one of the models is that SYT binds to the SNARE complexes even in the absence of Ca2+, but as soon as Ca2+ enters the cell, it may switch and bind phospholipids, destabilize the fusion intermediate and open the fusion pore [reviewed in
(Sudhof 2004)]. Here we report the isolation of a new recessive mutation in the Drosophila syt1
gene which may provide new insights to SYT1 function.
Results
Identification of syt1 mutant. A new mutant allele of syt1 was isolated in a forward genetic
screen recently conducted in our laboratory. The new allele, syt1425, was recovered in an EMS
(ethane methylsulfonate) screen for new temperature-sensitive (TS) mutants exhibiting rapid
paralysis or motor impairment at elevated temperature. This allele produced rapid TS paralysis
(Figure 1). This mutant was first identified as a syt1 allele by non-complementation of TS
paralysis with the previously characterized syt1 mutant, sytAD4 (DiAntonio, Parfitt et al. 1993).
Molecular characterization of the new syt1425 mutant. PCR amplification of the syt1 ORF from
genomic DNA and subsequent sequence analysis led to the identification of the molecular lesion,
which is a deletion of nucleotide 734 in the coding sequence. This deletion results in a frameshift
that leads to a premature stop codon after the 250th amino acid. This truncation is in the calcium
binding C2A domain that deletes half of the C2A domain and the full C2B domain (Figure 2).
Based on the nature of the molecular lesion, the mutant protein is predicted to be of half
the molecular weight of the wild-type SYT1 protein. The predicted molecular mass of wild-type
SYT1 is 53kDa, but the apparent molecular mass by SDS-PAGE is a little higher than that (~
65kDa). The Western Blot analysis of fly head homogenates from flies homozygous for syt1425and
156 other genotypes is shown in Figure 3. The predicted size of SYT11425 is 28kDa. However, in flies
homozygous for syt1425 there is reduced level of protein migrating at the molecular mass as the wild-type SYT1 (65kDa). Similar reduction was seen in syt1425/Df flies (Figure 3.) and sytAD4/Df
flies (Data not shown). These indicate that the band seen in these mutant flies is probably from
another isoform. Since sytAD4 is a null mutant in which synaptic activity is severely impaired,
this syt1 isoform detected here may not be functional at the synapses.
Discussion
Previous studies of syt1 null mutant flies show reduction in calcium induced neurotransmitter release (Littleton, Stern et al. 1993). Thus, SYT1 plays an important role in sustaining
neurotransmitter release at synapses. Both the C2 domains of SYT1 bind to Ca2+ and this
binding induces simultaneous binding of SYT1 to both the fusing phospholipid membrane and
the assembling SNARE complex (Pang, Sudhof 2010). While, blocking of Ca2+ binding to the
C2B domain blocks synchronous exocytosis (Mackler et al., 2002), blocking Ca2+ binding to the
C2A domain leads to drastic reduction in exocytosis and additionally reduces the apparent Ca2+
co-operativity of exocytosis (Shin, Xu et al., 2009). These results indicate that the two C2
domains co-operate with each other and contribute to triggering exocytosis. Three syt1 alleles
(AD1, AD3, and AD4) have been biochemically characterized (DiAntonio and Schwarz 1994;
Fukuda, Kabayama et al. 2000; Littleton, Bai et al. 2001). As indicated in Figure A.2, sytAD4 is a null allele caused by a premature stop codon that deletes the transmembrane and cytoplasmic domains of the protein and disrupts all of SYT's known interactions (DiAntonio and Schwarz
1994). sytAD1 has a premature stop codon that deletes the C2B domain and reduces Ca2+-
dependent binding of SYT to SNAREs and Ca2+-dependent oligomerization. C2A domain is intact in this mutant and contributes to normal phospholipid binding (Davis, Bai et al. 1999;
Littleton, Bai et al. 2001). sytAD3 has an altered amino acid in C2B domain (Y364N) that disrupts
Ca2+-dependent conformational changes in C2B that are required for oligomerization of SYT. The
157 mutation in this sytAD3 does not abolish SNARE or phospholipid binding (Fukuda, Kabayama et
al. 2000; Littleton, Bai et al. 2001). Each of these mutants affects synaptic vesicle exocytosis differently, thereby allowing the different functions of SYT to be separated and analyzed. The syt1425 described in the present study can be useful in analyzing SYT1 function. The deletion of
almost half of the C2A domain in syt1425 (Figure A.2) may prove useful in the in vivo structure- function analysis of SYT1 and may reveal a new role of SYT1.
Acknowledgements. This This study was supported by National Institutes of Health Grant
R21MH085199-02. We thank Dr. Noreen Reist (Colorado State University, Fort Collins, CO) for
generously providing the anti-SYT Dsyt CL1 antibody. We are grateful to Dr Tom Schwarz
(Children's Hospital, Boston, MA) for providing us with the syt null allele, sytAD4. We would also
like to acknowledge Bloomington Stock Center for the deficiency stock, Df(2L)Exel6277, used in
this study. DNA sequencing was performed in the Pennsylvania State University Sequencing
Facility, University Park, PA with excellent support from the center staff.
158 FIGURES
Figure A.1 Genetic screen identifies a new syt1 mutant. syt1425 exhibits rapid paralysis at 38°C.Behavior test for wild-type (WT) flies was truncated after 20 mins. The asterisk denotes statistical significance with respect to WT. WT flies are Canton-S. Behavioral analysis was performed as described previously (Brooks, Felling et al. 2003). Briefly, flies were placed in a vial preheated to 38°C by immersion in a water bath. Two-day-old flies reared at 20ºC were tested in groups of six for TS behavioral phenotypes. Five groups of flies (n=5) were tested for each genotype. Time for 50% paralysis represents the time at which 3 flies were no longer able to stand. Behavior test for wild-type (WT) flies was truncated after 20 minutes. In tests exceeding 5 minute duration, 3ml of water was added to the rayon plug after 5 minutes to prevent dehydration.
159
Figure A.2 New syt1 mutant. Domain structure of SYT1 [Modified from (Littleton, Stern et al. 1994)] The locations of the existing and the new syt1 alleles are indicated. TM represents the transmembrane domain and C2A and C2B are the two Ca2+ binding domains. N and C represent the N and C termini. For identifying the molecular lesions in the new syt1 mutant, genomic DNA from the homozygous mutant flies Iso2 flies (parent strain) were used to PCR amplify the syt1 gene using the forward primer, SytUT1_ Fwd (5’ TCATGTGAGACTAGCGAGCC 3’), and the reverse primer, SytUT2_ Rev (5’ TCTGACGGTTCGCGTACATG 3’). The PCR products were gel purified, quantified and sent for sequencing using primers SytUT1_ Fwd, SytUT2_ Rev and Syt400 (5’ TAATCTTCTTCTGTGTGCGG 3’). Sequences from the mutants were compared with those from Iso2 and molecular lesion in syt1425 was identified.
160
Figure A.3 Western Blot analysis of syt1 mutant. Western Blot analysis of fly head homogenates prepared from WT, homozygous syt1425 mutant, heterozygous syt1425 mutant, transheterozygotes (syt1425/Df) mutant and flies heterozygous for Df(2L)Exel6277 which removes syt1 locus. Flies heterozygous for Df(2L)Exel6277 exhibited a reduced level of SYT1 compared to the wild-type. Tubulin (TUB) was used as an internal loading control.
161 Bibliography
Abderrahmani, A., G. Niederhauser, et al. (2004). "Complexin I regulates glucose-induced
secretion in pancreatic beta-cells." Journal of Cell Science 117(11): 2239-2247.
An, S. J., C. P. Grabner, et al. (2010). "Real-time visualization of complexin during single
exocytic events." Nature Neuroscience 13(5): 577-583.
Arikkath, J. and K. P. Campbell (2003). "Auxiliary subunits: essential components of the voltage-
gated calcium channel complex." Current Opinion in Neurobiology 13: 298-307.
Bai, J. and E. R. Chapman (2004). "The C2 domains of synaptotagmin – partners in exocytosis."
Trends in Biochemical Sciences 29: 143-151.
Banerjee, A., V. A. Barry, et al. (1996). "N-ethylmaleimide-sensitive factor acts at a prefusion
ATP-dependent step in Ca2+-activated exocytosis." The Journal of Biological Chemistry
271: 20223-20226.
Batzer, A. G., P. Blaikie, et al. (1995). "The phosphotyrosine interaction domain of Shc binds an
LXNPXY motif on the epidermal growth factor receptor." Molecular and cellular biology
15: 4403–4409.
Birks, R. I., Macintosh, F.C. (1961). "Acetylcholine metabolism of asympathetic ganglion."
Canadian Journal of Biochemistry and Physiology 39: 787-827.
Blasi, J., E. R. Chapman, et al. (1993). "Botulinum neurotoxin C1 blocks neurotransmitter release
by means of cleaving HPC-1/syntaxin." EMBO J 12(12): 4821-4828.
Block, M. R., B. S. Glick, et al. (1988). "Purification of an N-ethylmaleimide-sensitive protein
catalyzing vesicular transport." Proc Natl Acad Sci U S A 85(21): 7852-6.
Bordier, C. (1981). "Phase separation of integral membrane proteins in Triton X-114 solution."
Journal of Biological Chemistry 256(4): 1604-1607.
Bork, P. and B. Margolis (1995). "A phosphotyrosine interaction domain." Cell 80: 693–694.
162 Boulianne, G. and W. S. Trimble (1995). "Identification of a second homolog of N-
ethylmaleimide-sensitive fusion protein that is expressed in the nervous system and
secretory tissues of Drosophila." Proceedings of the National Academy of Sciences,
U.S.A. 92: 7095-7099.
Bracher, A., J. Kadlec, et al. (2002). "X-ray structure of a neuronal complexin-SNARE complex
from squid." Journal of Biological Chemistry 277(29): 26517-26523.
Brand, A. H. and N. Perrimon (1993). "Targeted gene expression as a means of altering cell fates
and generating dominant phenotypes." Development 118: 401-415.
Broadie, K., H. J. Bellen, et al. (1994). "Absence of synaptotagmin disrupts excitation-secretion
coupling during synaptic transmission." Proceedings of the National Academy of
Sciences of the United States of America 91(22): 10727-10731.
Brodin, L., P. Low, et al. (2000). "Sequential steps in clathrin-mediated synaptic vesicle
endocytosis." Current Opinion in Neurobiology 10(3): 312-320.
Brooks, I. M., R. Felling, et al. (2003). "Genetic analysis of a synaptic calcium channel in
Drosophila: intragenic modifiers of a temperature-sensitive paralytic mutant of
cacophony." Genetics 164: 163-171.
Brose, N. (2008). "For better or for worse: complexins regulate SNARE function and vesicle
fusion." Traffic 9(9): 1403-1043.
Brunger, A. T. (2005). "Structure and function of SNARE and SNARE-interacting proteins."
Quarterly Reviews of Biophysics 38(1): 1-47.
Cai, H., K. Reim, et al. (2008). "Complexin II plays a positive role in Ca2+-triggered exocytosis
by facilitating vesicle priming." Proceedings of the National Academy of Sciences of the
United States of America 105(49): 19538-43.
Casey, P. J. (1992). "Biochemistry of protein prenylation." Journal of Lipid Research 33(12):
1731-1740.
163 Catterall, W. A. (1998). "Structure and function of neuronal Ca2+ channels and their role in
neurotransmitter release." Cell Calcium 24: 307-323.
Catterall, W. A. (2000). "Structure and regulation of voltage-gated Ca2+ channels." Annual
Review of Cell and Developmental Biology 16: 521-55.
Ceccarelli, B., W. P. Hurlbut, et al. (1973). "Turnover of transmitter and synaptic vesicles at the
frog neuromuscular junction." Journal of Cell Biology 57(2): 499-524.
Chapman, E. R. (2008). "How does synaptotagmin trigger neurotransmitter release?" Annual
Review of Biochemistry 77: 615-641.
Chen, X., D. R. Tomchick, et al. (2002). "Three-dimensional structure of the complexin/SNARE
complex." Neuron 33(3): 397-409.
Chen, Y. A. and R. H. Scheller (2001). "SNARE-mediated membrane fusion." Nature Reviews
Molecular Cell Biology 2: 98-106.
Cho, R. W., Y. Song, et al. (2010). "Comparative analysis of Drosophila and mammalian
complexins as fusion clamps and facilitators of neurotransmitter release." Molecular &
Cellular Neuroscience 45(4): 389-397.
Clayton, E. L., G. J. Evans, et al. (2008). "Bulk synaptic vesicle endocytosis is rapidly triggered
during strong stimulation." Journal of Neuroscience 28(26): 6627-6632.
Conner, S. D. and S. L. Schmid (2003). "Differential requirements of AP2 in clathrin-mediated
endocytosis." The Journal of Cell Biology 162: 773-779.
Daniels, R. W., C. A. Collins, et al. (2004). "Increased expression of the Drosophila vesicular
glutamate transporter leads to excess glutamate release and a compensatory decrease in
quantal content." 24: 10466-10474.
Danjo, R., F. Kawasaki, et al. (2011). "A tripartite synapse model in Drosophila." PLoS One 6:
e17131.
164 Davis, A. F., J. Bai, et al. (1999). "Kinetics of synaptotagmin responses to Ca2+ and assembly
with the core SNARE complex onto membranes." Neuron 24: 363-376.
Del Castillo, J., Katz, B. (1954). "Quantal components of the end plate potential." Journal of
Physiololgy 124(3): 560-573.
Dellinger, B. B., R. Felling, et al. (2000). "Genetic modifiers of the Drosophila NSF mutant,
comatose, include a temperature-sensitive paralytic allele of the calcium channel α1
subunit gene, cacophony." Genetics 155: 203-211.
Denker, A. and S. O. Rizzoli (2010). "Synaptic vesicle pools: an update." Frontiers in Synaptic
Neuroscience 2(135): 1-12.
Di Paolo, G. and P. De Camilli (2006). "Phosphoinositides in cell regulation and membrane
dynamics." Nature 443: 651-657.
DiAntonio, A., R. W. Burgess, et al. (1993). "Identification and characterization of Drosophila
genes for synaptic vesicle proteins." The Journal of Neuroscience 13: 4924-4935.
DiAntonio, A., K. D. Parfitt, et al. (1993). "Synaptic transmission persists in synaptotagmin
mutants of Drosophila." Cell 73: 1281-1290.
DiAntonio, A. and T. L. Schwarz (1994). "The effect on synaptic physiology of synaptotagmin
mutations in Drosophila." Neuron 12: 909-920.
Dittman, J. and T. A. Ryan (2009). "Molecular circuitry of endocytosis at nerve terminals."
Annual Review of Cell and Developmental Biology 25: 133-160.
Estes, P. S., J. Roos, et al. (1996). "Traffic of dynamin within individual Drosophila synaptic
boutons relative to compartment-specific markers." J Neurosci 16(17): 5443-56.
Evergren E., Tomilin N., et al. (2004). "A pre-embedding immunogold approach for detection of
synaptic endocytic proteins in situ." Journal of Neuroscience Methods 135: 169-174.
165 Fasshauer, D. and M. Margittai (2004). "A transient N-terminal interaction of SNAP-25 and
syntaxin nucleates SNARE assembly." Journal of Biological Chemistry 279(9): 7613-
7621.
Fatt, P., Katz, B. (1952). "Spontaneous subthreshold activity at motor nerve endings." Journal of
Physiololgy 117(1): 109 -128.
Faulk, W. P. and G. M. Taylor (1971). "An immunocolloid method for the electron microscope."
Immunochemistry 8(11): 1081-1083.
Fernandez-Alfonso, T. and T. A. Ryan (2004). "The kinetics of synaptic vesicle pool depletion at
CNS synaptic terminals." Neuron 41(6): 943-953.
Fernandez-Chacon, R., A. Konigstorfer, et al. (2001). "Synaptotagmin I functions as a calcium
regulator of release probability." Nature 410: 41-49.
Fernandez-Chacon, R., Konigstorfer, A., Gerber, S.H., García, J., Matos, M.F., Stevens, C.F.,
Brose, N., Rizo, J., Rosenmund, C., Südhof, T.C. (2001). "Synaptotagmin I functions as a
calcium regulator of release probability." Nature 410(6824): 41-49.
Freeman, W. and A. J. Morton (2004). "Differential messenger RNA expression of complexins in
mouse brain." Brain Research Bulletin 63(1): 33-44.
Fukuda, M., H. Kabayama, et al. (2000). "Drosophila AD3 mutation of synaptotagmin impairs
calcium-dependent self-oligomerization activity." FEBS Letters 482: 269-272.
Gad, H., N. Ringstad, et al. (2000). "Fission and uncoating of synaptic clathrin-coated vesicles are
perturbed by disruption of interactions with the SH3 domain of endophilin." Neuron 27:
301–312.
Gandhi, S. P. and C. F. Stevens (2003). "Three modes of synaptic vesicular recycling revealed by
single-vesicle imaging." Nature 423: 607-613.
Geppert, M., Y. Goda, et al. (1994). "Synaptotagmin I: a major Ca2+ sensor for transmitter
release at a central synapse." Cell 74(4): 717-727.
166 Gertler, F. B., R. L. Bennett, et al. (1989). "Drosophila abl tyrosine kinase in embryonic CNS
axons: a role in axonogenesis is revealed through dosage-sensitive interactions with
disabled." Cell 58: 103-113.
Gertler, F. B., K. K. Hill, et al. (1993). "Dosage-sensitive modifiers of Drosophila abl tyrosine
kinase function: prospero, a regulator of axonal outgrowth, and disabled, a novel tyrosine
kinase substrate." Genes & Development 7: 441-453.
Geumann, U., C. Schafer, et al. (2010). "Synaptic membrane proteins form stable microdomains
in early endosomes." Microscopy Research and Technique 73(6): 606-617.
Gillooly DJ, Morrow IC, et al. (2000). " Localization of phosphatidylinositol 3-phosphate in yeast
and mammalian cells." EMBO J 13: 4924-4935.
Giniger, E. (1998). "A role for Abl in Notch signaling." Neuron 20(4): 667-681.
Giraudo, C. G., W. S. Eng, et al. (2006). "A clamping mechanism involved in SNARE-dependent
exocytosis." Science 313(5787): 676-680.
Giraudo, C. G., A. Garcia-Diaz, et al. (2008). "Distinct domains of complexins bind SNARE
complexes and clamp fusion in vitro." Journal of Biological Chemistry 283(30): 21211-
21219.
Glyvuk, N., Y. Tsytsyura, et al. (2010). "AP-1/sigma1B-adaptin mediates endosomal synaptic
vesicle recycling, learning and memory." EMBO J. 29(8): 1318-1330.
Gomes, A. Q., B. R. Ali, et al. (2003). "Membrane targeting of Rab GTPases is influenced by the
prenylation motif." Molecular Biology of the Cell 14(5): 1882-1899.
Gonzalez-Gaitan, M. and H. Jackle (1997). "Role of Drosophila alpha-adaptin in presynaptic
vesicle recycling." Cell 21: 767-776.
Granseth, B., B. Odermatt, et al. (2006). "Clathrin-mediated endocytosis is the dominant
mechanism of vesicle retrieval at hippocampal synapses." Neuron 51(6): 773-786.
167 Grass I, Thiel S, et al. (2004). "Recognition of a basic AP-2 binding motif within the C2B domain
of synaptotagmin is dependent on multimerization." Journal of Biological Chemistry 279:
54872-54880.
Grigliatti, T. A., L. Hall, et al. (1973). "Temperature-sensitive mutations in Drosophila
melanogaster XIV. A selection of immobile adults." Molecular and General Genetics 120:
107-114.
Hanson, P. I., J. E. Heuser, et al. (1997). "Neurotransmitter release - four years of SNARE
complexes." Current Opinion in Cell Biology 7: 310-315.
Harata, N. C., A. M. Aravanis, et al. (2006). "Kiss-and-run and full-collapse fusion as modes of
exo-endocytosis in neurosecretion." Journal of Neurochemistry 97(6): 1546-1570.
Harata, N. C., S. Choi, et al. (2006). "Frequency-dependent kinetics and prevalence of kiss-and-
run and reuse at hippocampal synapses studied with novel quenching methods." Neuron
49(2): 243-256.
Haucke, V., E. Neher, et al. (2011). "Protein scaffolds in the coupling of synaptic exocytosis and
endocytosis." Nature Reviews Neuroscience 12(3): 127-138.
Hayashi, T., H. McMahon, et al. (1994). "Synaptic vesicle membrane fusion complex: action of
clostridial neurotoxins on assembly." EMBOJ 13: 5051-5061.
Heerssen, H., R. D. Fetter, et al. (2008). "Clathrin dependence of synaptic-vesicle formation at
the Drosophila neuromuscular junction." Current Biology 18: 401-409.
Herz, J. and Y. Chen (2006). "Reelin, lipoprotein receptors and synaptic plasticity." Nature
Reviews Neuroscience 7: 850-859.
Heuser, J. E. and T. S. Reese (1981). "Structural changes after transmitter release at the frog
neuromuscular junction." The Journal of Cell Biology 88: 564-580.
Hinshaw, J. and S. L. Schmid (1995). "Dynamin Self-assembles into Rings Suggesting a
Mechanism for Coated Vesicle Budding." Nature 374: 190-192.
168 Holmes, A., A. Flett, et al. (2007). "C. elegans Disabled is required for cell-type specific
endocytosis and is essential in animals lacking the AP-3 adaptor complex." Journal of
Cell Science 120: 2741-2751.
Holt, M., A. Cooke, et al. (2003). "Bulk membrane retrieval in the synaptic terminal of retinal
bipolar cells." Journal of Neuroscience 23(4): 1329-1339.
Hosoi, N., M. Holt, et al. (2009). "Calcium dependence of exo- and endocytotic coupling at a
glutamatergic synapse." Neuron 63: 216-229.
Howell, B. W., F. B. Gertler, et al. (1997). "Mouse disabled (mDab1): a Src binding protein
implicated in neuronal development." The EMBO Journal 16: 121-132.
Howell, B. W., L. M. Lanier, et al. (1999). "The disabled 1 phosphotyrosine-binding domain
binds to the internalization signals of transmembrane glycoproteins and to
phospholipids." Molecular and cellular biology 19: 5179–5188.
Huntwork, S. and J. T. Littleton (2007). "A complexin fusion clamp regulates spontaneous
neurotransmitter release and synaptic growth." Nature Neuroscience 10(10): 1235-1237.
Ishizuka, T., H. Saisu, et al. (1995). "Synaphin: a protein associated with the docking/fusion
complex in presynaptic terminals." Biochemical and Biophysical Research
Communications 213(3): 1107-1114.
Itakura, M., H. Misawa, et al. (1999). "Transfection analysis of functional roles of complexin I
and II in the exocytosis of two different types of secretory vesicles." Biochemical and
Biophysical Research Communications 265(3): 691-696.
Jackson, M. B. and E. R. Chapman (2006). "Fusion pores and fusion machines in Ca2+-triggered
exocytosis." Annual Review of Biophysics and Biomolecular Structure 35: 135-160.
Jahn, R., T. Lang, et al. (2003). "Membrane fusion." Cell 112: 519-533.
Jahn, R. and R. H. Scheller (2006). "SNAREs--engines for membrane fusion." Nature Reviews
Molecular Cell Biology 7(9): 631-43.
169 Jahn, R. and T. C. Sudhof (1999). "Membrane fusion and exocytosis." Annual Review of
Biochemistry 68: 863-911.
Jarousse, N. and R. B. Kelly (2001). "Endocytotic mechanisms in synapses." Current Opinion in
Cell Biology 13: 461-469.
Jing, S. Q., T. Spencer, et al. (1990). "Role of the human transferrin receptor cytoplasmic domain
in endocytosis: localization of a specific signal sequence for internalization." The Journal
of Cell Biology 110: 283-294.
Jung, N. and V. Haucke (2007). "Clathrin-mediated endocytosis at synapses." Traffic 8: 1129–
1136.
Kaeser-Woo, Y. J., X. Yang, et al. (2012). "C-terminal complexin sequence is selectively
required for clamping and priming but not for Ca2+ triggering of synaptic exocytosis."
Journal of Neuroscience 32(8): 2877-2885.
Kamikura, D. M. and J. A. Cooper (2003). "Lipoprotein receptors and a disabled family
cytoplasmic adaptor protein regulate EGL-17/FGF export in C. elegans." Genes &
Development 17: 2798-2811.
Kamikura, D. M. and J. A. Cooper (2006). "Clathrin interaction and subcellular localization of
Ce-DAB-1, an adaptor for protein secretion in Caenorhabditis elegans." Traffic 7: 324-
336.
Kasprowicz, J., S. Kuenen, et al. (2008). "Inactivation of clathrin heavy chain inhibits synaptic
recycling but allows bulk membrane uptake." The Journal of Cell Biology 182: 1007-
1016.
Katz, B. and R. Miledi (1968). "The role of calcium in neuromuscular facilitation." Journal of
Physiololgy 195(2): 481-492.
Kavalali, E. T., C. Chung, et al. (2011). "Spontaneous neurotransmission: an independent
pathway for neuronal signaling?" Physiology (Bethesda) 26(1): 45-53.
170 Kavanaugh, W. M. and L. T. Williams (1994). "An alternative to SH2 domains for binding
tyrosine-phosphorylated proteins." Science 266: 1862-1865.
Kawasaki, F., S. C. Collins, et al. (2002). "Synaptic calcium channel function in Drosophila:
Analysis and transformation rescue of temperature-sensitive paralytic and lethal
mutations of cacophony." The Journal of Neuroscience 22: 5856-5864.
Kawasaki, F., R. Felling, et al. (2000). "A temperature-sensitive paralytic mutant defines a
primary synaptic calcium channel in Drosophila." The Journal of Neuroscience 20: 4885-
4889.
Kawasaki, F., M. Hazen, et al. (1999). "A strictly presynaptic function for the Drosophila NSF
protein, dNSF1, includes a role in fast recycling of synaptic vesicle membranes."
Abstract. The 1999 Cold Spring Harbor meeting on Neurobiology of Drosophila.
Kawasaki, F., M. Hazen, et al. (2000). "Fast synaptic fatigue in shibire mutants reveals a rapid
requirement for dynamin in synaptic vesicle membrane trafficking." Nature Neuroscience
3: 859-860.
Kawasaki, F., J. Iyer, et al. (2011). "The DISABLED protein functions in CLATHRIN-mediated
synaptic vesicle endocytosis and exoendocytic coupling at the active zone." Proceedings
of the National Academy of Sciences of the United States of America 108(25): E222-229.
Kawasaki, F., A. M. Mattiuz, et al. (1998). "Synaptic physiology and ultrastructure in comatose
mutants define an in vivo role for NSF in neurotransmitter release." The Journal of
Neuroscience 18: 10241-10249.
Kawasaki, F. and R. W. Ordway (1999). "A new model synapse preparation allows direct
demonstration of a strictly presynaptic role for the Drosophila NSF protein, dNSF1."
Abstract. The 40th Annual Drosophila Research Conference.
Kawasaki, F. and R. W. Ordway (2009). "Molecular mechanisms determining conserved
properties of short-term synaptic depression: roles of NSF and SNAP-25 defined in
171 Drosophila." Proceedings of the National Academy of Sciences of the United States of
America 106: 14658-14663.
Kawasaki, F., B. Zou, et al. (2004). "Active zone localization of presynaptic calcium channels
encoded by the cacophony." The Journal of Neuroscience 24: 282-285.
Kim SH and R. TA (2009). "Synaptic vesicle recycling at CNS snapses without AP-2." Journal of
Neuroscience 29: 3865-3874.
Kirchhausen, T., J. S. Bonifacino, et al. (1997). "Linking cargo to vesicle formation: receptor tail
interactions with coat proteins." Current Opinion in Cell Biology 9: 488-495.
Klingauf, J., E. T. Kavalali, et al. (1998). "Kinetics and regulation of fast endocytosis at
hippocampal synapses." Nature 394(6693): 581-585.
Koenig, J. H. and K. Ikeda (1989). "Disappearance and reformation of synaptic vesicle membrane
upon transmitter release observed under reversible blockage of membrane retrieval." The
Journal of Neuroscience 9: 3844-3860.
Koenig, J. H. and K. Ikeda (1996). "Synaptic vesicles have two distinct recycling pathways." The
Journal of Cell Biology 135: 797-808.
Koenig, J. H., T. Kosaka, et al. (1989). "The relationship between the number of synaptic vesicles
and the amount of transmitter released." The Journal of Neuroscience 9: 1937-1942.
Koh, T. W., Bellen, H.J. (2003). "Synaptotagmin I, a Ca2+ sensor for neurotransmitter release."
Trends in Neurosciences 26(8): 413-422.
Koh, T. W., V. I. Korolchuk, et al. (2007). "Eps15 and Dap160 control synaptic vesicle
membrane retrieval and synapse development." The Journal of Cell Biology 178: 309-
322.
Koh, T. W., P. Verstreken, et al. (2004). "Dap160/intersectin acts as a stabilizing scaffold
required for synaptic development and vesicle endocytosis." Neuron 43: 193-205.
172 L.M., T. (2009). "Tickets to ride: selecting cargo for clathrin-regulated internalization." Nature
Reviews Molecular Cell Biology 10: 583-596.
Lang, T. and R. Jahn (2008). Core Proteins of the Secretory Machinery. Pharmacology of
Neurotransmitter Release (Handbook of experimental pharmacology). T. C. Südhof and
K. Starke, Springer-Verlag Berlin Heidelberg. 184: 107-127.
Le Gall, M., C. De Mattei, et al. (2008). "Molecular separation of two signaling pathways for the
receptor, Notch." Developmental Biology 313(2): 556-567.
Le, N. and M. A. Simon (1998). "Disabled is a putative adaptor protein that functions during
signaling by the sevenless receptor tyrosine kinase." Molecular and cellular biology 18:
4844-4854.
Li X., Kuromi H., et al. (2010). "Bicaudal-D binds clathrin heavy chain to promote its transport
and augments synaptic vesicle recycling." EMBO J 29: 992-1006.
Liebl, E. C., R. G. Rowe, et al. (2003). "Interactions between the secreted protein Amalgam, its
transmembrane receptor Neurotactin and the Abelson tyrosine kinase affect axon
pathfinding." Development 130: 3217-3226.
Lin, R. C. and R. H. Scheller (1997). "Structural organization of the synaptic exocytosis core
complex." Neuron 19(5): 1087-1094.
Lin, R. C. and R. H. Scheller (2000). "Mechanisms of synaptic vesicle exocytosis." Annual
Review of Cell and Developmental Biology 16: 19-49.
Littleton, J. T., J. Bai, et al. (2001). "Synaptotagmin mutants reveal essential functions for the
C2B-domain in Ca2+-triggered fusion and recycling of synaptic vesicles in vivo." Journal
of Neuroscience 21: 1421-1433.
Littleton, J. T., M. Stern, et al. (1994). "Calcium dependence of neurotransmitter release and rate
of spontaneous vesicle fusions are altered in Drosophila synaptotagmin mutants."
173 Proceedings of the National Academy of Sciences of the United States of America 91:
10888-10892.
Littleton, J. T., M. Stern, et al. (1993). "Mutational analysis of Drosophila synaptotagmin
demonstrates its essential role in Ca(2+)-activated neurotransmitter release." Cell 74(6):
1125-1134.
Liu, J., T. Guo, et al. (2007). "Overexpression of complexin in PC12 cells inhibits exocytosis by
preventing SNARE complex recycling." Biochemistry (Mosc) 72(4): 439-444.
Liu, Y.-W., M. C. Surka, et al. (2008). "Isoform and splice-variant specific functions of dynamin-
2 revealed by analysis of conditional knock-out cells." Molecular Biology of the Cell 19:
5347-5359.
Loerke D., Mettlen M., et al. (2009). "Cargo and dynamin regulate clathrin-coated pit
maturation." PLoS Biology 7.
Lutas, A., C. J. Wahlmark, et al. (2012). "Genetic analysis in Drosophila reveals a role for the
mitochondrial protein p32 in synaptic transmission." G3 (Bethesda) 2(1): Jan.
Mackler, J. M., J. A. Drummond, et al. (2002). "The C(2)B Ca(2+)-binding motif of
synaptotagmin is required for synaptic transmission in vivo." Nature 418: 340-344.
Malhotra, V., L. Orci, et al. (1988). "Role of an N-ethylmaleimide-sensitive transport component
in promoting fusion of transport vesicles with cisternae of the Golgi stack." Cell 54(2):
221-7.
Malsam, J., F. Seiler, et al. (2009). "The carboxy-terminal domain of complexin I stimulates
liposome fusion." Proceedings of the National Academy of Sciences of the United States
of America 106(6): 2001-2006.
Marie, B., S. T. Sweeney, et al. (2004). "Dap160/intersectin scaffolds the periactive zone to
achieve high-fidelity endocytosis and normal synaptic growth." Neuron 43.
174 Maritzen T, P. J, et al. (2010). "Stonins--specialized adaptors for synaptic vesicle recycling and
beyond?" Traffic 11(8-15).
Martin, J. A., Z. Hu, et al. (2011). "Complexin has opposite effects on two modes of synaptic
vesicle fusion." Current Biology 21(2): 97-105.
Mathias, R. A., Y. S. Chen, et al. (2011). "Triton X-114 phase separation in the isolation and
purification of mouse liver microsomal membrane proteins." Methods 54(4): 396-406.
Maurer, M. E. and J. A. Cooper (2005). "Endocytosis of megalin by visceral endoderm cells
requires the Dab2 adaptor protein." Journal of Cell Science 118: 5345-5355.
Maximov, A., Z. P. Pang, et al. (2007). "Monitoring synaptic transmission in primary neuronal
cultures using local extracellular stimulation." Journal of Neuroscience Methods 161(1):
75-87.
Maximov, A., J. Tang, et al. (2009). "Complexin controls the force transfer from SNARE
complexes to membranes in fusion." Science 323(5913): 516-521.
McMahon, H. T., M. Missler, et al. (1995). "Complexins: cytosolic proteins that regulate SNAP
receptor function." Cell 83(1): 111-119.
Mettlen M., Pucadyil T., et al. (2009). "Dissecting dynamin's role in clathrin-mediated
endocytosis." Biochemical Society Transactions 37: 1022-1026.
Miller T.M. and Heuser J.E. (1984). "Endocytosis of synaptic vesicle membrane at the frog
neuromuscular junction." Journal of Cell Biology 98: 685-698.
Miller, T. M. and J. E. Heuser (1984). "Endocytosis of synaptic vesicle membrane at the frog
neuromuscular junction." Journal of Cell Biology 98: 685-698.
Mishra, S. K., P. A. Keyel, et al. (2002). "Disabled-2 exhibits the properties of a cargo-selective
endocytic clathrin adaptor." The EMBO Journal 21: 4915-4926.
Morimura, T., M. Hattori, et al. (2005). "Disabled1 regulates the intracellular trafficking of reelin
receptors." The Journal of Biological Chemistry 280: 16901-16908.
175 Morris, S. M., S. D. Arden, et al. (2002). "Myosin VI binds to and localises with Dab2,
potentially linking receptor-mediated endocytosis and the actin cytoskeleton." Traffic 3:
331-341.
Morris, S. M. and J. A. Cooper (2001). "Disabled-2 colocalizes with the LDLR in clathrin-coated
pits and interacts with AP-2." Traffic 2: 111-123.
Mulkey, R. M. and R. S. Zucker (1991). "Action potentials must admit calcium to evoke
transmitter release." Nature 350(6314): 153-155.
Murthy, V. N. and P. De Camilli (2003). "Cell biology of the presynaptic terminal." Annual
Review of Neuroscience 26: 701-728.
Neher, E. (2010). "What is rate-limiting during sustained synaptic activity: vesicle supply or the
availability of release sites." Frontiers in Synaptic Neuroscience 2(144): 1-6.
Neher, E. and T. Sakaba (2008). "Multiple roles of calcium ions in the regulation of
neurotransmitter release." Neuron 59: 861-872.
Novick, P., C. Field, et al. (1980). "Identification of 23 complementation groups required for
post-translational events in the yeast secretory pathway." Cell 21(1): 205-15.
O'Connor, V., C. Heuss, et al. (1997). "Disruption of syntaxin-mediated protein interactions block
neurotransmitter secretion." Proceedings of the National Academy of Sciences, U.S.A. 94:
12186-12191.
Ono, S., G. Baux, et al. (1998). "Regulatory roles of complexins in neurotransmitter release from
mature presynaptic nerve terminals." European Journal of Neuroscience 10(6): 2143-
2152.
Opazo F., Punge A., et al. (2010). "Limited intermixing of synaptic vesicle components upon
vesicle recycling." Traffic 11: 800-812.
Ordway, R. W. (2004). "Quantal size fits central synaptic depression." Proceedings of the
National Academy of Sciences of the United States of America 101: 907-908.
176 Ordway, R. W., L. Pallanck, et al. (1994). "Neurally expressed Drosophila genes encoding
homologs of the NSF and SNAP secretory proteins." Proceedings of the National
Academy of Sciences of the United States of America 91: 5715-5719.
Otsu, Y. and T. H. Murphy (2003). "Miniature Transmitter Release: Accident of Nature or
Careful Design?" Sci. STKE 2003(211): 54.
Owen, D. J., Y. Vallis, et al. (2000). "The structure and function of the beta 2-adaptin appendage
domain." EMBO J. 19(16): 4216-4227.
Pabst, S., J. W. Hazzard, et al. (2000). "Selective interaction of complexin with the neuronal
SNARE complex. Determination of the binding regions." Journal of Biological
Chemistry 275(26): 19808-19818.
Pabst, S., M. Margittai, et al. (2002). "Rapid and selective binding to the synaptic SNARE
complex suggests a modulatory role of complexins in neuroexocytosis." Journal of
Biological Chemistry 277(10): 7838-7848.
Pallanck, L., R. W. Ordway, et al. (1995). "A Drosophila NSF mutant." Nature 376: 25.
Pallanck, L., R. W. Ordway, et al. (1995). "Distinct roles for N-ethylmaleimide-sensitive fusion
protein (NSF) suggested by the identification of a second Drosophila NSF homolog."
The Journal of Biological Chemistry 270: 18742-18744.
Pang, Z. P., J. Sun, et al. (2006). "Genetic analysis of synaptotagmin 2 in spontaneous and Ca2+-
triggered neurotransmitter release." EMBO J. 25(10): 2039-2050.
Pishvaee, B., A. Munn, et al. (1997). "A novel structural model for regulation of clathrin
function." The EMBO Journal 16: 2227-2239.
Poirier, M. A., Xiao, W., Macosko, J.C., Chan, C., Shin, Y.K., Bennett, M.K. (1998). "The
synaptic SNARE complex is a parallel four-stranded helical bundle." Nature Structural
Biology 5(9): 765-769.
177 Poskanzer, K. E., K. W. Marek, et al. (2003). "Synaptotagmin I is necessary for compensatory
synaptic vesicle endocytosis in vivo." Nature 426: 559-563.
Puertollano, R. (2004). "Clathrin-mediated transport: assembly required. Workshop on Molecular
Mechanisms of Vesicle Selectivity." EMBO Reports 5(10): 942-946.
Ramachandran, R., M. Surka, et al. (2007). "The dynamin middle domain is critical for
tetramerization and higher-order self-assembly." EMBO J. 26: 559-566.
Reider, A. and B. Wendland (2011). "Endocytic adaptors--social networking at the plasma
membrane." Journal of Cell Science 124(10): 1613-1622.
Reim, K., M. Mansour, et al. (2001). "Complexins regulate a late step in Ca2+-dependent
neurotransmitter release." Cell 104(1): 71-81.
Reim, K., H. Wegmeyer, et al. (2005). "Structurally and functionally unique complexins at retinal
ribbon synapses." Journal of Cell Biology 169(4): 669-680.
Richards, D. A., C. Guatimosim, et al. (2000). "Two endocytic recycling routes selectively fill
two vesicle pools in frog motor nerve terminals." Neuron 27(3): 551-559.
Ringstad N., Gad H., et al. (1999). "Endophilin/SH3p4 is required for the transition from early to
late stages in clathrin-mediated synaptic vesicle endocytosis." Neuron 24: 143-154.
Rizo, J. and C. Rosenmund (2008). "Synaptic vesicle fusion." Nature Structural and Molecular
Biology 15(7): 665-74.
Rizo, J. and T. C. Sudhof (2002). "SNAREs and Munc18 in synaptic vesicle fusion." Nature
Reviews Neuroscience 3: 641-653.
Rizzoli, S. O. and W. J. Betz (2005). "Synaptic vesicle pools." Nature Reviews Neuroscience 6:
57-69.
Rizzoli, S. O. and R. Jahn (2007). "Kiss-and-run, collapse and 'readily retrievable' vesicles."
Traffic 8(9): 1137-1144.
178 Robinson, I. M., R. Ranjan, et al. (2002). "Synaptotagmins I and IV promote transmitter release
independently of Ca(2+) binding in the C(2)A domain." Nature 418: 340-344.
Robinson, S. M. (2004). "Adaptable adaptors for coated vesicles." Trends in Cell Biology 14:
167-174.
Roggero, C. M., G. A. De Blas, et al. (2007). "Complexin/synaptotagmin interplay controls
acrosomal exocytosis." Journal of Biological Chemistry 282(36): 26335-26343.
Roos, J. and R. B. Kelly (1998). "Dap160, a neural-specific Eps15 homology and multiple SH3
domain-containing protein that interacts with Drosophila dynamin." The Journal of
Biological Chemistry 273: 19108-19119.
Roos, J. and R. B. Kelly (1999). "The endocytic machinery in nerve terminals surrounds sites of
exocytosis." Current Biology 9(23): 1411-1414.
Schaub, J. R., X. Lu, et al. (2006). "Hemifusion arrest by complexin is relieved by Ca2+-
synaptotagmin I." Nature Structural and Molecular Biology 13(8): 748-750.
Schiavo, G., F. Benfenati, et al. (1992). "Tetanus and botulinum-B neurotoxins block
neurotransmitter release by proteolytic cleavage of synaptobrevin." Nature 359(6398):
832-835.
Schillo, S., G. Belusic, et al. (2004). "Targeted mutagenesis of the farnesylation site of
Drosophila Ggammae disrupts membrane association of the G protein betagamma
complex and affects the light sensitivity of the visual system." Journal of Biological
Chemistry 279(35): 36309-36316.
Schmid, S. L., M. A. McNiven, et al. (1998). "Dynamin and its partners: a progress report."
Current Opinion in Cell Biology 10: 504-512.
Seiler. F., J. Malsam, et al. (2009). "A role of complexin-lipid interactions in membrane fusion."
FEBS Letters 583: 2343–2348.
179 Sever, S., A. B. Muhlberg, et al. (1999). "Impairment of dynamin's GAP domain stimulates
receptor-mediated endocytosis." Nature 398: 481-486.
Shiina, T., K. Arai, et al. (2001). "Clathrin box in G protein-coupled receptor kinase 2." Journal
of Biological Chemistry 276(35): 33019-33026.
Shupliakov, O. and L. Brodin (2010). "Recent insights into the building and cycling of synaptic
vesicles." Experimental Cell Research 316: 1344-1350.
Shupliakov, O., P. Low, et al. (1997). "Synaptic vesicle endocytosis impaired by disruption of
dynamin-SH3 domain interactions." Science 276: 259-263.
Siddiqi, O. and S. Benzer (1976). "Neurophysiological defects in temperature-sensitive paralytic
mutants of Drosophila melanogaster." Proceedings of the National Academy of Sciences
of the United States of America 73: 3253-3257.
Smith, L. A., X. Wang, et al. (1996). "A Drosophila calcium channel α1 subunit gene maps to a
genetic locus associated with behavioral and visual defects." The Journal of Neuroscience
16: 7868-7879.
Smith, S. M., R. Renden, et al. (2008). "Synaptic vesicle endocytosis: fast and slow modes of
membrane retrieval." Trends in Neurosciences 31: 559-568.
Sollner, T., M. K. Bennett, et al. (1993). "A protein assembly-disassembly pathway in vitro that
may correspond to sequential steps of synaptic vesicle docking, activation, and fusion."
Cell 75: 409-418.
Sollner, T., S. Whiteheart, et al. (1993). "SNAP receptors implicated in vesicle targeting and
fusion." Nature 362: 318-324.
Song, J. K., R. Kannan, et al. (2010). "Disabled is a bona fide component of the Abl signaling
network." Development 137: 3719-3727.
180 Sorensen, J. B., K. Wiederhold, et al. (2006). "Sequential N- to C-terminal SNARE complex
assembly drives priming and fusion of secretory vesicles." The EMBO Journal 25: 955-
966.
Sorkin, A. (2004). "Cargo recognition during clathrin-mediated endocytosis: a team effort."
Current Opinion in Cell Biology 16: 392–399.
Sorkina, T., A. Bild, et al. (1999). "Clathrin, adaptors and eps15 in endosomes containing
activated epidermal growth factor receptors." Journal of Cell Science 112(3): 317-327.
Sudhof, T. C. (2004). "The synaptic vesicle cycle." Annual Review of Neuroscience 27: 509-547.
Sutton, R. B., D. Fasshauer, et al. (1998). "Crystal structure of a SNARE complex involved in
synaptic exocytosis at 2.4 Å resolution." Nature 395: 347-353.
Suzuki, D. T., T. Grigliatti, et al. (1971). "Temperature-sensitive mutations in Drosophila
melanogaster, VII. A mutation (paraTS) causing reversible adult paralysis." Proceedings
of the National Academy of Sciences of the United States of America 68: 890-893.
Tadokoro, S., M. Nakanishi, et al. (2005). "Complexin II facilitates exocytotic release in mast
cells by enhancing Ca2+ sensitivity of the fusion process." Journal of Cell Science
118(10): 2239-2246.
Takahashi, S., H. Yamamoto, et al. (1995). "Identification of two highly homologous presynaptic
proteins distinctly localized at the dendritic and somatic synapses." FEBS Letters 368(3):
455-460.
Takei, K., O. Mundigl, et al. (1996). "The synaptic vesicle cycle: a single vesicle budding step
involving clathrin and dynamin." The Journal of Cell Biology 133: 1237-1250.
Tang, J., A. Maximov, et al. (2006). "A complexin/synaptotagmin 1 switch controls fast synaptic
vesicle exocytosis." Cell 126(6): 1175-1187.
Teckchandani, A., N. Toida, et al. (2009). "Quantitative proteomics identifies a Dab2/integrin
module regulating cell migration." Journal of Cell Biology 186(1): 99-111.
181 Tokumaru, H., K. Umayahara, et al. (2001). "SNARE complex oligomerization by
synaphin/complexin is essential for synaptic vesicle exocytosis." Cell 104(3): 421-432.
Tolar, L. A. and L. Pallanck (1998). "NSF function in neurotransmitter release involves
rearrangement of the SNARE complex downstream of synaptic vesicle docking." The
Journal of Neuroscience 18: 10250-10256.
Torroja, L., H. Chu, et al. (1999). "Neuronal overexpression of APPL, the Drosophila homologue
of the amyloid precursor protein (APP), disrupts axonal transport." Current Biology 9:
489-492.
Traub, L. M. (2003). "Sorting it out: AP-2 and alternate clathrin adaptors in endocytic cargo
selection." The Journal of Cell Biology 163: 203-208.
Traub, L. M. (2009). "Tickets to ride: selecting cargo for clathrin-regulated internalization."
Nature Reviews Molecular Cell Biology 10: 583-596.
Traub, L. M., M. A. Downs, et al. (1999). "Crystal structure of the alpha appendage of AP-2
reveals a recruitment platform for clathrin-coat assembly." Proceedings of the National
Academy of Sciences of the United States of America 96: 8907-8912.
Van der Geer, P., S. Wiley, et al. (1995). "A conserved amino-terminal Shc domain binds to
phosphotyrosine motifs in activated receptors and phosphopeptides." Current Biology 5:
404-412.
Verstreken, P., T. W. Koh, et al. (2003). "Synaptojanin is recruited by endophilin to promote
synaptic vesicle uncoating." Neuron 40: 733-748.
Wang, L. and R. L. Coppel (2002). "Triton X-114 phase partitioning for antigen
characterization." Methods in Molecular Methods 72: 581-585.
Wang, W., M. Bouhours, et al. (2008). "ITSN-1 controls vesicle recycling at the neuromuscular
junction and functions in parallel with DAB-1." Traffic 9: 742-754.
182 Weninger, K., M. E. Bowen, et al. (2008). "Accessory proteins stabilize the acceptor complex for
synaptobrevin, the 1:1 syntaxin/SNAP-25 complex." Structure 16(2): 308-320.
Winter-Vann, A. M. and P. J. Casey (2005). "Post-prenylation-processing enzymes as new targets
in oncogenesis." Nature Reviews Cancer 5(5): 402-412.
Wu X.S. and W. L.G. (2009). "Rapid endocytosis does not recycle vesicles within the readily
releasable pool." Journal of Neuroscience 29: 11038-11042.
Wu, X. S., B. D. McNeil, et al. (2009). "Ca2+ and calmodulin initiate all forms of endocytosis
during depolarization at a nerve terminal." Nature Neuroscience 12: 1003-1010.
Wu, Y., F. Kawasaki, et al. (2005). "Properties of short-term synaptic depression at larval
neuromuscular synapses in wild-type and temperature-sensitive paralytic mutants of
Drosophila." Journal of Neurophysiology 93: 2396–2405.
Xu, X. X., W. Yang, et al. (1995). "Cloning of a novel phosphoprotein regulated by colony-
stimulating factor 1 shares a domain with the Drosophila disabled gene product." The
Journal of Biological Chemistry 270: 14184–14191.
Xue, M., T. K. Craig, et al. (2010). "Binding of the complexin N terminus to the SNARE
complex potentiates synaptic-vesicle fusogenicity." Nature Structural and Molecular
Biology 17(5): 568-575.
Xue, M., Y. Q. Lin, et al. (2009). "Tilting the balance between facilitatory and inhibitory
functions of mammalian and Drosophila Complexins orchestrates synaptic vesicle
exocytosis." Neuron 64(3): 367-380.
Xue, M., K. Reim, et al. (2007). "Distinct domains of complexin I differentially regulate
neurotransmitter release." Nature Structural and Molecular Biology 14(10): 949-958.
Xue, M., A. Stradomska, et al. (2008). "Complexins facilitate neurotransmitter release at
excitatory and inhibitory synapses in mammalian central nervous system." Proceedings of
the National Academy of Sciences of the United States of America 105(22): 7875-7880.
183 Yamada, M., Saisu, H., Ishizuka, T., Takahashi, H., Abe, T. (1999). "Immunohistochemical
distribution of the two isoforms of synaphin/complexin involved in neurotransmitter
release: localization at the distinct central nervous system regions and synaptic types."
Neuroscience 93(1): 7-18.
Yang, X., Y. J. Kaeser-Woo, et al. (2010). "Complexin clamps asynchronous release by blocking
a secondary Ca(2+) sensor via its accessory alpha helix." Neuron 68(5): 907-920.
Yarar, D., M. C. Surka, et al. (2008). "Snx9 activities are regulated by multiple phosphoinositides
through both PX and BAR domains." Traffic 9: 133-146.
Yoon, T. Y., X. Lu, et al. (2008). "Complexin and Ca2+ stimulate SNARE-mediated membrane
fusion." Nature Structural and Molecular Biology 15(7): 707-713.
Yu, W., F. Kawasaki, et al. (2011). "Activity-dependent interactions of NSF and SNAP at living
synapses." Molecular & Cellular Neuroscience 47(1): 19-27.
Yun, M., L. Keshvara, et al. (2003). "Crystal structures of the Dab homology domains of mouse
disabled 1 and 2." The Journal of Biological Chemistry 278: 36572-36581.
Zenisek, D., J. A. Steyer, et al. (2002). "A membrane marker leaves synaptic vesicles in
milliseconds after exocytosis in retinal bipolar cells." Neuron 35(6): 1085-1097.
Zhang, F. L. and P. J. Casey (1996). "Protein prenylation: molecular mechanisms and functional
consequences." Annual Review of Biochemistry 65: 241-269.
Zou, B., H. Yan, et al. (2008). "MAP1 structural organization in Drosophila: In vivo analysis of
FUTSCH reveals heavy and light chain subunits generated by proteolytic processing at a
conserved cleavage site." The Biochemical Journal 414: 63-71.
VITA
Janani Iyer
EDUCATION August, 2012 Ph.D., Genetics, The Pennsylvania State University, UP May, 2006 M. Sc., Genomics, Madurai Kamaraj University, India May, 2004 B. Sc., Biochemistry, Delhi University, India
RESEARCH EXPERIENCE Aug ‘07-June’12 Graduate Research Assistant, The Pennsylvania State University Aug ‘05-May ’06 Masters Graduate Student, Madurai Kamaraj University, India May-July ‘05 Visiting Student’s Research Program (VSRP), Tata Institute Of Fundamental Research, Mumbai, India PUBLICATIONS Kawasaki F, Iyer J, Posey LL, Sun CE,Mammen SE, Yan H, Ordway RW (2011) The DISABLED protein functions in CLATHIN-mediated synaptic vesicle endocytosis and exoendocytic coupling at the active zone. PNAS 108(25): E222E2229
PRESENTATIONS Janani Iyer, Christopher J. Wahlmark, Giselle A. Kuser-Ahnert and Fumiko Kawasaki. Genetic and Molecular Analysis of Drosophila COMPLEXIN Function in Synaptic Transmission. Society for Neuroscience, Nov12-16, 2011.
Janani Iyer, Huaru Yan, Chi-Chun Sun, Fumiko Kawasaki and Richard W. Ordway. DISABLED Functions in CLATHRIN-Mediated Synaptic Vesicle Endocytosis and Exo- Endo Coupling at the Active Zone. Annual Genetics Symposium, May 2011.
Janani Iyer, Huaru Yan, Chi-Chun Sun, Fumiko Kawasaki and Richard W. Ordway. The CLATHRIN Associated Sorting Protein, DISABLED Functions in Synaptic Vesicle Endocytosis. Annual Genetics Symposium, May 2010.