University of Massachusetts Medical School eScholarship@UMMS
GSBS Dissertations and Theses Graduate School of Biomedical Sciences
2014-04-01
Role of Astrocytes in Sculpting Neuronal Circuits in the Drosophila CNS: A Dissertation
Ozge E. Tasdemir-Yilmaz University of Massachusetts Medical School
Let us know how access to this document benefits ou.y Follow this and additional works at: https://escholarship.umassmed.edu/gsbs_diss
Part of the Developmental Neuroscience Commons, and the Molecular and Cellular Neuroscience Commons
Repository Citation Tasdemir-Yilmaz OE. (2014). Role of Astrocytes in Sculpting Neuronal Circuits in the Drosophila CNS: A Dissertation. GSBS Dissertations and Theses. https://doi.org/10.13028/M23P4G. Retrieved from https://escholarship.umassmed.edu/gsbs_diss/729
This material is brought to you by eScholarship@UMMS. It has been accepted for inclusion in GSBS Dissertations and Theses by an authorized administrator of eScholarship@UMMS. For more information, please contact [email protected].
ROLE OF ASTROCYTES IN SCULPTING NEURONAL CIRCUITS IN THE
DROSOPHILA CNS
A Dissertation Presented
By
Ozge Evrim Tasdemir Yilmaz
Submitted to the Faculty of the
University of Massachusetts Graduate School of Biomedical Sciences, Worcester
in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
April 1st, 2014
Program in Neuroscience
ii
ROLE OF ASTROCYTES IN SCULPTING NEURONAL CIRCUITS IN THE DROSOPHILA CNS
A Dissertation Presented By Ozge Evrim Tasdemir Yilmaz
The signatures of the Dissertation Defense Committee signify completion and approval as to style and content of the Dissertation
Marc R. Freeman, Ph.D., Thesis Advisor
Eric Baehrecke, Ph.D., Member of Committee
Vivian Budnik, Ph.D., Member of Committee
Fen-Biao Gao, Ph.D., Member of Committee
Kim McCall, Ph.D., Member of Committee
The signature of the Chair of the Committee signifies that the written dissertation meets the requirements of the Dissertation Committee
Michael Francis, Ph.D., Chair of Committee
The signature of the Dean of the Graduate School of Biomedical Sciences signifies that the student has met all graduation requirements of the school.
Anthony Carruthers, Ph.D., Dean of the Graduate School of Biomedical Sciences
Program in Neuroscience April 1st, 2014
iii
This work is dedicated to my great husband Erdem and my amazing family, my parents Fatoş and Kanber and my sister Müge, for their love, support and encouragement.
iv
Acknowledgements
First of all, I would like to thank my thesis advisor Marc Freeman for all of his
guidance, patience and support through my graduate education. I am truly happy and
grateful to be one of his graduate students. He is one of the most brilliant, enthusiastic
and inspiring people that I have ever known. He has been such a great mentor and has
inspired me to be one. Marc, thank you for being very understanding during the time
when I had to take care of my mom in Turkey, letting me take some time off and also
being very welcoming when I wanted to come back to continue my studies. I greatly
appreciate that gesture and will never forget that.
Secondly, I would like to thank my committee members Michael Francis, Eric
Baehrecke, Vivian Budnik and Fen-Biao Gao for their advice, support and guidance throughout the years. I would like to thank my outside committee member Kim McCall for her great input that improved my thesis. I also would like to thank my former qualifier committee chair Mark Alkema for helping me getting through the qualifier exam.
I would like to thank all past and present Freeman lab members: Amy Sheehan,
Mary Logan, Johnna Doherty, Tim Rooney, Rebecca Bernardos, Jaeda Coutinho-Budd,
Allie Muthukumar, Lukas Neukomm, Jen MacDonald, Rachel Bradshaw, Jen Ziegenfuss,
Jeannette Osterloh, Edith Plada, Tsai-Yi Lu, Tobias Stork, Megan Corty, Jon Farley,
Owen Peters, Nicki Fox, Yuly Fuentes, Kim Kerr, Zhiguo Ma, Tom Burdett, Jemeen
Sreedharan, Sam Licciardo and Michelle Avery. You all made Freeman lab such a great environment to work in. I appreciate all your help with presentations and experiments, all v
the ideas, support and friendship during all these years. I would also like to thank Jaeda
Coutinho-Budd, Nicki Fox and Tim Rooney for editing this manuscript.
I would like to thank my husband, Erdem, for his love, constant help and encouragement during my good and bad times. I would like to thank my family: my mom
Fatoş, my dad Kanber and my sister Müge, for their love and support throughout these years and encouragement to pursue my dream. To my mom: wherever you are, I hope you are seeing/feeling that I have accomplished to finish my PhD! At last, I’d like to thank to all who gave joy to my life in graduate school: my sweet niece Gizem Defne, my dear brother-in-law Tolga, my cool brother-in-law Ercan, my dear in-laws Aliye and
Mehmet Yilmaz, and all my friends.
vi
Abstract
The nervous system is composed of neurons and glia. Glial cells have been neglected and thought to have only a supportive role in the nervous system, even though
~60% of the mammalian brain is composed of glia. Yet, in recent years, it has been shown that glial cells have several important functions during the development, maintenance and function of the nervous system. Glial cells regulate both pre and post mitotic neuronal survival during normal development and maintenance of the nervous system as well as after injury, are necessary for axon guidance, proper axon fasciculation, and myelination during development, promote synapse formation, regulate ion balance in the extracellular space, are required for normal synaptic function, and have immune functions in the brain. Although glia have crucial roles in nervous system development and function, there are still much unknown about the underlying molecular mechanisms in glial development, function and glial-neuronal communication.
Drosophila offers great opportunity to study glial biology, with its simple yet sophisticated and stereotypic nervous system. Glial cells in flies show great complexity similar to the mammalian nervous system, and many cellular and molecular functions are conserved between flies and mammals. In this study, I use Drosophila as a model organism to study the function of one subtype of glia: astrocytes. The role of astrocytes in synapse formation, function and maintenance has been a focus of study. However, their role in engulfment and clearance of neuronal debris during development remains unexplored. vii
I generated a driver line that enables the study of astrocytes in Drosophila. In chapter two of this thesis, I characterize astrocytes during metamorphosis, when extensive neuronal remodeling takes place. I found that astrocytes turn into phagocytes in a cell-autonomous, steroid-dependent manner, by upregulating the phagocytic receptor
Draper and forming acidic phagolysosomal structures. I show that astrocytes clear neuronal debris during nervous system remodeling and that this is a novel function for astrocytes during the development of nervous system. I analyzed two different neuronal populations: MB γ neurons that prune their neurites and vCrz+ neurons that undergo apoptosis. I discovered that MB γ axons are engulfed by astrocytes using the Draper and
Crk/Mbc/dCed-12 pathways in a partially redundant way. Interestingly, Draper is required for clearance of vCrz+ cell bodies, while Crk/Mbc/dCed-12, but not Draper, are required for clearance of vCrz+ neurites. Surprisingly, I also found that loss of Draper delayed vCrz+ neurite degeneration, suggesting that glia facilitate neurite destruction through engulfment signaling.
Taken together, my work identifies a novel function for astrocytes in the clearance of synaptic and neuronal debris during developmental remodeling of the nervous system. Additionally, I show that Crk/Mbc/dCed-12 act as a new glial signaling pathway required for pruning, and surprisingly, that glia use different engulfment pathways to clear neuronal debris generated by cell death versus local pruning.
viii
Table of Contents
Acknowledgements iv
Abstract vi
Table of Contents viii
List of Figures xi
List of Third Party Copyrighted Material xiv
List of Symbols, Abbreviations or Nomenclature xv
Preface xix
CHAPTER I: Introduction 1
Neural circuit assembly and circuit remodeling during development 2
Roles of glia in the nervous system 7
Glial subtypes in the mammalian brain 8
Drosophila as a model organism for studying glia 11
Glial subtypes in Drosophila 13
Neural remodeling during Drosophila metamorphosis 16
Molecular mechanisms of neural remodeling 19
Glial phagocytic functions during circuit remodeling and in response to injury 25
Molecular mechanisms of engulfment 30
CHAPTER II: Astrocytes engage unique molecular programs to engulf pruned neuronal debris from distinct subsets of neurons 34
Abstract 35
Introduction 36 ix
Results 39
Discussion 50
Materials and Methods 56
CHAPTER III: Discussion 91
Astrocytes’ role as phagocytes 93
Finding new molecules required for astrocytic engulfment 96
Astrocytes are the primary phagocytes in the neuropil 97
Do Drosophila astrocytes have any role in injury response? 99
What drives Drosophila astrocytes to turn into phagocytes? 99
Why are different phagocytic pathways engaged by glia during different contexts of
remodeling? 104
What are the eat-me cues? 106
Glia instruct neural circuit remodeling 111
Cellular and molecular similarities between neuronal remodeling and neuronal
degeneration after injury and disease 115
Conclusions 118
APPENDIX I: Investigation of the function of the alrm gene 120
alrm is expressed in longitudinal glia in the Drosophila embryo 121
What are longitudinal glial cells and how do they develop? 122
Alrm is a leucine-rich repeat secreted protein 123
Glial expressed Alrm::myc preferentially accumulates in the neuropil 125
alrm expression is regulated by repo 126 x
Generation of alrm mutants 127
alrm null mutants do not display gross defects in axon pathfinding 128
alrm is not required as a trophic factor for Eg+ neurons in Drosophila embryo 129
Behavioral testing of alrm nulls 130
Larval locomotion is normal in alrm nulls 130
Larval alrm nulls do not display olfaction defects towards attractant olfactory
stimulus 132
alrm mutants do not display drastic learning or memory defects 133
alrm mutants do not have drastic tolerance defects in hypoxic conditions 134
Morphological studies of alrm nulls 135
Loss of alrm has no drastic effect on NMJ formation 135
alrm is not required for MB γ neuron pruning 136
MB lobes are normal in the adult alrm nulls 137
General astrocyte morphology is normal in alrm mutants 138
Future directions to uncover a possible role of Alrm protein: 138
References 160
xi
List of Figures
Figure 2.1: Astrocytes transform into phagocytes at the initiation of neural circuit remodeling
Figure 2.2: EcR signaling cell-autonomously regulates transformation of astrocytes into phagocytes
Figure 2.3: Astrocytes engulf and clear synaptic material from the neuropil
Figure 2.4: Draper and Crk/Mbc/dCed-12 pathways are required for the formation of astrocytic vacuoles
Figure 2.5: EcR, Draper and Crk/Mbc/dCed-12 function in astrocytes to promote MB γ neuron clearance
Figure 2.6: EcR, Draper and Crk/Mbc/dCed-12 function in astrocytes to promote clearance of vCrz+ neuronal debris.
Figure 2.7: Loss of Draper delays neurite degeneration and clearance of vCrz+ neuronal cell bodies.
Figure 2.8: Astrocytes’ position relative to the neuropil in the CNS.
Figure 2.9: Ultrastructural analysis of control animal brains at L3.
Figure 2.10: Draper immunoreactivity does not overlap with glial cells other than astrocytes in the neuropil at 6hrs APF.
Figure 2.11: Astrocytic knockdown of Draper almost completely blocks Draper upregulation in astrocytes at 3hrs APF.
Figure 2.12: dCed-6 is not activated at pupariation, nor is it modulated by EcR signaling in astrocytes.
Figure 2.13: Astrocytes do not express EcRA isoform at pupariation.
Figure 2.14: Astrocytic Draper overexpression does not rescue the EcRDN blockade of astrocyte morphological transformation.
Figure 2.15: EcRDN blockade of Draper expression in astrocytes is not efficient at later stages of metamorphosis.
xii
Figure 2.16: Astrocytic phagolysosomes engulf nc82+ and FasII+ material in the neuropil.
Figure 2.17: Crk is required for the proper formation of vacuoles in astrocytes.
Figure 2.18: Draper and Crk/Mbc/dCed-12 function in astrocytes to promote clearance of synaptic material from the neuropil.
Figure 2.19: Crk is required in astrocytes for proper clearance of synaptic material in the neuropil.
Figure 2.20: Compared to other regions of the neuropil, MB lobes are infiltrated less by astrocytes.
Figure 2.21: Categories used to quantify MB γ pruning phenotypes.
Figure 2.22: Blocking endocytic activity of astrocytes leads to MB γ neuronal pruning defects.
Figure 2.23: Knockdown of Draper specifically in astrocytes leads to MB γ neuronal pruning defects.
Figure 2.24: Knockdown of Crk/Mbc/dCed-12 pathway components specifically in astrocytes leads to MB γ neuronal pruning defects.
Figure 2.25: Pan-glial knockdown of dCed-12 does not perturb clearance of vCrz+ cell bodies.
Figure 2.26: Draper function in non-astrocyte glia is required for proper clearance of vCrz+ cell bodies.
Figure 2.27: There are low to undetectable levels of Draper immunoreactivity in vCrz+ neurons.
Figure 2.28: Draper does not function in vCrz+ neurons for the clearance of vCrz+ debris.
Figure A.1: alrm is expressed in longitudinal glia (LG).
Figure A.2: alrm is colocalized with longitudinal glial marker F263.
Figure A.3: alrm expression is downregulated in gcm misexpression background.
Figure A.4: alrm is expressed in fewer glia in ectopic Gcm background.
xiii
Figure A.5: Alrm is secreted from S2 cells.
Figure A.6: Glial expressed Alrm::Myc preferentially accumulates in the neuropil.
Figure A.7: alrm expression is regulated by repo.
Figure A.8: Generation of alrm mutants.
Figure A.9: alrm mutants do not have axon pathfinding defects in the embryo.
Figure A.10: alrm nulls have normal number of Eg+ neurons.
Figure A.11: Larval locomotion is normal in alrm nulls.
Figure A.12: Larval alrm nulls do not display olfaction defects.
Figure A.13: alrm nulls do not display gross learning or memory defects.
Figure A.14: alrm mutants do not have drastic tolerance defects in hypoxic conditions.
Figure A.15: alrm nulls do not have a drastic effect on NMJ formation.
Figure A.16: alrm is not required for MB γ pruning.
Figure A.17: Morphology of MB lobes is normal in alrm null adults.
Figure A.18: Astrocyte morphology is normal in alrm nulls.
xiv
List of Third Party Copyrighted Material
The following figures were reproduced from a journal: No permission required
Figure Number Publisher
All Chapter II figures Cold Spring Harbor Laboratory Press
xv
List of Symbols, Abbreviations or Nomenclature
20-HE: 20-Hydroxyecdysone
alrm: astrocytic leucine-rich repeat molecule
Apaf-1: apoptotic protease activating factor-1
APF: after puparium formation
axo: axotactin
BAI1: Brain-specific angiogenesis inhibitor 1
BBB: blood-brain barrier
BDNF: brain-derived neurotrophic factor
bss: bang senseless
C4da: class IV da
ced: cell death abnormal
ChIP: Chromatin Immunoprecipitation
CNS: central nervous system
CR3: complement receptor 3
Crz: Corazonin
cyt c: cytochrome c
dLGN: dorsal lateral geniculate nucleus
DN: dominant-negative
DRG: dorsal root ganglia eas: easily shocked
EcR: Ecdysone receptor xvi
Eg: Eagle
GB: glioblast gcm: glial cells missing
GEF: guanine nucleotide exchange factor
GMC: ganglion mother cell
GP: glial precursor
Hid: Head involution defective
InR: Insulin-like receptor
JH: juvenile hormone
LG: longitudinal glia
LRR: leucine-rich repeat
MB: mushroom body
MTZ: myelination transition zone
Myo: Myoglianin nAChR: nicotinic acetylcholine receptor
NB: neuroblast
NGB: neuroglioblast
NGF: nerve growth factor
NMJ: neuromuscular junction
NPC: neural precursor cell
NT-3: neurotrophin-3
ORN: olfactory receptor neuron xvii
OT: optic tectum
PC: Purkinje cell
PCD: programmed cell death
PFA: paraformaldehyde
PG: perineurial glia
PNS: peripheral nervous system
Prtp: Pretaporter
PS: Phosphatidylserine
PTB: phosphotyrosine-binding
RGC: Retinal ganglion cell
RI: response index
Rpr: Reaper
SC: superior colluculus shits: temperature-sensitive shibire
SPG: subperineurial glia
SVZ: subventricular zone
Tai: Taiman
TEM: transmission electron microscopy
TRAP: translating ribosome affinity purification
TUNEL: Terminal deoxynucleotidyl transferase dUTP nick end labeling
Ub: ubiquitin
UPS: ubiquitin-proteasome system xviii
Usp: Ultraspiracles
VDRC: Vienna Drosophila RNAi Center
VNC: ventral nerve chord
WB: western blot
XIAP: X-linked inhibitor of apoptosis protein
xix
Preface
All work described in this thesis was performed at University of Massachusetts Medical
School in the lab of Marc Freeman except the learning and memory assays in Appendix I, which was performed in the lab of Scott Waddell at University of Massachusetts Medical
School. In Appendix I, Samuel Licciardo, MD performed alrm expression in LG and generation of alrm mutants; Kimberly Kerr, PhD performed Alrm secretion studies in S2 cells; Christopher Burke, PhD performed olfactory memory assays and Zhiguo Ma, PhD performed larval olfactory plate assays and the rest of the experiments were performed by me.
1
CHAPTER I: Introduction
2
Neural circuit assembly and circuit remodeling during development
As the nervous system develops, many cells and connections are made and lead to
remarkable growth; many neurons form and extend their dendrites and axons, find their
targets; glial cells are born and myelin sheaths form around the axons; blood vessels
branch around these cells and extracellular matrix fills around these cells. Amazingly, the human brain continues to get bigger from around 400 grams at birth to 1400 grams in adulthood even though most of the neurogenesis is complete (Dekaban, 1978).
Surprisingly, during this time of growth, many neurons and glia die in the brain via programmed cell death (PCD). It is still not clear why shortly after neurons are made they are committed to die. This was a surprise in the field since it was not predicted that neurons should die shortly after they were made.
Depending on the brain region, 20 to 80% of the differentiated cells/neurons die during development (Oppenheim, 1991). The survival of neurons depends on several factors from different sources: glial cells, postsynaptic neurons, presynaptic neurons, nearby neurons and the circulatory system. One reason for cell death is to match the size
(or needs) of the neuronal population to the target field. Studies in which target fields are completely removed, partially ablated or expanded artificially showed that the size of the
target field is critical in determining the number of projection neurons that survive
(HAMBURGER and LEVI-MONTALCINI, 1949; Hollyday and Hamburger, 1976).
However, neuronal cell death is not only a mechanism for random matching of cell
number to target field size, but also leads to selective elimination of neurons whose axons
make erroneous projections in a wrong target area or in a wrong region within the target 3
field and also elimination of functionally redundant neurons (Fawcett et al., 1984). What
factors in the target tissue are essential for driving neuronal competition for survival? The
first such factor found in an amazing series of experiments is nerve growth factor (NGF)
(LEVI-MONTALCINI and HAMBURGER, 1951). Later on, additional growth factors
have been identified, such as brain-derived neurotrophic factor (BDNF) and
neurotrophin-3 (NT-3) (Barde et al., 1982; Hohn et al., 1990; Maisonpierre et al., 1990;
Rosenthal et al., 1990). Another type of mechanism that regulates neuronal death is
carried out by circulating hormones. It is well documented in invertebrates like
Drosophila that ecdysone hormone regulates death of a number of neurons. In certain
songbirds, some forebrain structures critical for song development undergo marked cell
loss in females and that effect can be prevented by treatment with oestrogen hormone
(Konishi and Akutagawa, 1987). Finally, some neurons die in a genetically programmed
manner just after their generation, which is extensively studied in C. elegans (Ellis and
Horvitz, 1986). The molecular mechanisms underlying the PCD of cells in C. elegans
will be described later.
During late embryonic development in mice, neural progenitor cells that give rise
to neurons die extensively (Thomaidou et al., 1997). By injecting BrdU (to indicate if the
cells are proliferating) and examining for Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) labeling (to indicate if the cells are dying) it has been found that about 70% of the TUNEL-labeled cells in the subventricular zone (SVZ) were colabeled for BrdU, suggesting that just after proliferation, the cells entered a death phase
(Thomaidou et al., 1997). The reason for such extensive numbers of dying neuronal 4
progenitors is still not known in detail. There is an in vitro study in which quail neural crest cells are cultured on a nonadhesive substrate to prevent them from dispersing and the cells became positive for TUNEL labeling compared to the ones that could disperse in culture (Maynard et al., 2000).This suggests that maybe fate-restricted progenitors that fail to disperse and maintain more cell-cell contacts are fated to die in vivo.
Not only after their birth are neurons eliminated, but some also die after they have elaborated their fine processes and innervated their targets. For example: over 60% of differentiated frog spinal motor neurons are eliminated (Hughes, 1961); ~ 50% of embryonic chick ciliary ganglion neurons die after projecting to their targets and receiving afferent synaptic input (Landmesser and Pilar, 1972, 1974); ~ 40% of mouse
GABAergic cortical neurons get eliminated after migrating to their specific locations, acquiring inhibitory neuron characteristics, and forming synapses with their neighboring cells (Southwell et al., 2012).
Neural circuit refinement not only occurs by neuronal death; some neurons selectively remove specific projections without themselves dying. Selective elimination of neuronal processes and synapses is a common event in many biological processes and is critical for neural function. Precise neural connectivity is achieved by removing such exuberant processes. Optimization of circuits allows for growth, learning, and memory.
Finally, after injury or in disease situations, specific elimination or stabilization of affected circuits can offer advantages for repair. Uncovering the underlying molecular mechanisms of neural circuit refinement would allow an understanding of how precise connectivity is achieved as well as addressing the neurological diseases where neurons 5
undergo degeneration of their axons. The underlying mechanisms may be common in development as well as in neurological diseases or in response to injury.
Selective elimination of neuronal processes can be divided into two categories: small-scale events such as synaptic elimination and local pruning of axonal or dendritic arbors, and large-scale events such as elimination of major axon branch and major collaterals. One form of synaptic elimination occurs through retraction. A classical example of axonal retraction is shown in vivo at the mammalian neuromuscular junction
(NMJ). Initially muscle cells are innervated by multiple motorneurons, and during development the motorneurons eliminate individual branches by retracting that eventually results in innervation of muscle cells by single motorneurons (Walsh and
Lichtman, 2003). As axons retract, they shed many axosomes, which are membrane- bound remnants containing synaptic organelles. These axosomes are formed and eliminated by engulfing Schwann cells (Bishop et al., 2004). Axons compete for maintenance of their synapses and as the loser retracts, the winner gets strengthened and takes up the space that is emptied. Other than mammalian NMJ, climbing fiber inputs to cerebellar Purkinje cells also reduce from poly-neuronal to mono-neuronal connections
(Mariani and Changeux, 1981).
A classical example for local arbor pruning in mammalian nervous system is the segregation of retinogeniculate and geniculocorticular projections that bring the visual input to the brain. Retinal ganglion cells (RGCs) project from retina to dorsal lateral geniculate nucleus (dLGN) and dLGNs project to layer IV of the cortex. Initially in both the retinogeniculate and geniculocorticular projection systems, the two monocular inputs 6
are overlapped. Later, the visual inputs segregate into eye-specific patterns reducing from
binocular to monocular connections (Crair et al., 2001; LeVay et al., 1978; Shatz and
Stryker, 1978; Sretavan and Shatz, 1986; Sur et al., 1982). This segregation occurs
through local pruning of overlapping axonal arbors. As in NMJ remodeling, eye-specific
segregation is driven by neural activity (Feller, 2002). When one of the eyes of the animal is closed or retinal activity is blocked during a critical period, RGC inputs to the dLGN do not segregate into eye-specific layers (Penn et al., 1998; Stellwagen and Shatz,
2002; Sur et al., 1982) Monocular deprivation leads to weakening of the synapses in layer
IV of the cortex from projections originating from the deprived eye. The vast majority of the remaining cortical neurons only respond to the open eye (Wiesel and Hubel, 1963).
Blockade of spontaneous activity in one retina has been shown to alter subsequent eye- specific segregation in the dLGN, revealing that this process is competition driven (Penn et al., 1998).
Apart from small-scale eliminations, it has been demonstrated that some neurons undergo large-scale elimination. Examples of such events were shown to occur during the development of area-specific subcortical projections of layer V neurons of the neocortex
(O’Leary and Koester, 1993) and of retinotopic map in the optic tectum of chicks and superior colliculus of mice (Hindges et al., 2002). Large-scale elimination involves pruning of many millimeters of axons. For the development of layer V subcortical projections (Luo and O’Leary, 2005) and retinotopic mapping (Nakamura and O’Leary,
1989), axon degeneration rather than retraction is shown to be the mechanism of elimination. 7
Such neuronal circuit eliminations mechanisms (programmed cell death and local
pruning) are conserved during evolution and also occur in Drosophila and will be
described in a later section.
Roles of glia in the nervous system
The nervous system is comprised of two primary cell types: neurons and glia.
Neurons have the capability to fire action potentials but glial cells do not. Therefore,
although there have been many studies about neurons for years, glial cells have been
neglected. Actually, in the mammalian brain, around 60% of the brain is composed of
glia. For over a century glial cells were thought as supporting cells for neurons. In the
recent years, however, many important functions of glial cells have been found.
Mounting evidence shows that glia have crucial functions during the development,
maintenance and function of the nervous system (Barres, 2008).
Glial cells regulate both pre and post mitotic neuronal survival in Drosophila and in mammals either through direct contact or by providing factors to neurons during normal functioning of the nervous system as well as after injury (Booth et al., 2000; Bush et al., 1999; Dearborn and Kunes, 2004; Platel et al., 2010; Riethmacher et al., 1997; Tao et al., 2011; Woldeyesus et al., 1999; Xiong and Montell, 1995). Glial cells are necessary for axon guidance and proper axon fasciculation in Drosophila and in mammals (Auld,
1999; Bastiani and Goodman, 1986; Dearborn and Kunes, 2004; Gilmour et al., 2002;
Halter et al., 1995; Hidalgo and Booth, 2000; Hidalgo et al., 2001; Hosoya et al., 1995;
Jones et al., 1995; Shu and Richards, 2001; Woldeyesus et al., 1999; Xiong et al., 1994).
Glial cells promote synapse formation, regulate ion balance in the extracellular space, 8
regulate neurotransmitter recycling at the synapse, and are required for normal synaptic
function (Allen and Barres, 2005; Barres, 2008; Feng and Ko, 2008; Freeman and
Rowitch, 2013; Stevens, 2008; Ullian et al., 2001).
In vertebrates, myelination of axons is important for fast conduction of nerve
impulses, and myelin is made by glial cells (Freeman and Doherty, 2006; Nave, 2010;
Nickols et al., 2003). Lastly, glial cells have immune functions in the brain: they clear
debris from the environment so that the inflammatory and toxic effects of the remaining
debris will not harm uninjured neuronal and glial cells (Freeman, 2006; Freeman and
Doherty, 2006; Kurant, 2011; Mallat et al., 2005).
Glial subtypes in the mammalian brain
As we look at the evolution of animals, we see that the number of different kinds
of glial cells increase as we get closer to humans (Freeman and Rowitch, 2013). Subtypes of mammalian glial cells in the mammalian brain include astrocytes, microglia, oligodendrocytes and in the peripheral nervous system (PNS), Schwann cells. Each subtype of glial cells differ in their functions, morphology and molecular profile.
Oligodendrocytes in the central nervous system (CNS) and Schwann cells in the
PNS are glial cells that make myelin sheaths around axons. Myelin sheaths enable
saltatory conduction of action potentials (Freeman and Rowitch, 2013; Nave, 2010;
Nickols et al., 2003). Oligodendrocytes also provide trophic support to neurons by
secreting BDNF and NT-3 (Dai et al., 2003). Some Schwann cells are non-myelinating,
but still ensheath and support peripheral nerves. In contrast to the oligodendrocytes in the
CNS, Schwann cells can promote regeneration of axons after injury and then remyelinate 9
those axons (Bunge, 1994; Ide et al., 1983; Torigoe et al., 1996). Apart from these functions, Schwann cells also eliminate synapses: during synapse elimination in the developing NMJ, axons shed axosomes (membrane bound remnants containing synaptic organelles), which are engulfed and eliminated by Schwann cells (Bishop et al., 2004).
Mesodermal cells that are committed to the macrophage lineage infiltrate the developing CNS and then give rise to microglia in the brain. Microglia are the professional phagocytes that have immune functions in the CNS. They actively engulf neuronal cell corpses and cellular debris from the environment in the CNS (Garden and
Möller, 2006; Magnus et al., 2002; Mallat et al., 2005). Microglia respond to neuronal death or injury more efficiently than astrocytes and they also regulate inflammation, cytotoxicity, and antigen presentation, making microglia proper immune cells within the
CNS (Freeman and Rowitch, 2013; Parnaik et al., 2000). Apart from clearance of cell corpses and cellular debris, microglia also sculpt exuberant synaptic connections during developmental refinement of the retinogeniculate system (Schafer et al., 2012).
Astrocytes are the most numerous cell type in the mammalian brain. Astrocytes are a heterogeneous population, and based on morphology, antigenic phenotype and location there are two main classes of astrocytes. Protoplasmic astrocytes reside in the gray matter and their processes ensheath synapses and blood vessels. Fibrous astrocytes reside in white matter and contact nodes of Ranvier and blood vessels (Barres, 2008).
Astrocytes have functions in controlling the balance of ions such as K+ ions in the extracellular space and also clearance of neurotransmitters at the synapse. They also provide metabolic support to neurons, regulate blood flow, and maintain the blood-brain 10
barrier (BBB). Recent findings uncovered exciting new roles for astrocytes in the
modulation of synapses. According to the traditional view of the synapse, pre and
postsynaptic terminals are the only components that regulate synaptic information flow. It
has been found that astrocytes are closely associated with synapses and now the synaptic
connections are referred as tripartite synapse (Araque et al., 1999). Astrocytes also release factors that promote synapse formation and function (Barres, 2008;
Christopherson et al., 2005; Ullian et al., 2001). Schwann cells have similar functions in promoting formation and maintenance of the NMJ in the periphery (Feng and Ko, 2008).
Astrocytes are also important for synaptic plasticity: recently, astrocytes were shown to engulf synapses during developmental synaptic remodeling in retinogeniculate system in
vivo (Chung et al., 2013). In vitro, astrocytes can have phagocytic functions (Roldán et
al., 1997; Tansey and Cammer, 1998) and engulf apoptotic cells (Chang et al., 2000) or
amyloid-ß (Wyss-Coray et al., 2003). In vivo, astrocytes in the postlaminar optic nerve
head myelination transition zone (MTZ) express Mac-2, a molecule implicated in
phagocytic activity, and internalize axonal evulsions, and this event appears increased in
frequency in glaucoma models (Nguyen et al., 2011). Other than these specific
phagocytic functions, astrocytes have been implicated to be non-professional phagocytes
and are much less efficient than microglia in clearance in response to injury (Parnaik et
al., 2000). In response to injury, astrocytes become reactive: they undergo hypertrophy
and change their gene expression, they secrete molecules which either lead to neuronal
death or survival, they form glial scars, and they have critical roles in sealing the BBB 11
after brain injury (Bush et al., 1999; Chen and Swanson, 2003; Sofroniew, 2005;
Sofroniew and Vinters, 2010).
Drosophila as a model organism for studying glia
It is apparent that glial cells have crucial roles in many aspects of nervous system
development and normal function as well as after injury. However there are still many
unanswered questions about the underlying molecular mechanisms in glial development
and function. How do glial cells and neurons communicate? What are the underlying
molecular mechanisms of glial clearance of neural debris as neural circuits refine during
development? What other roles do glial cells have in the mature brain? Are glial cell
populations diverse? How do glial cells contribute to neurodegenerative diseases?
In mammalian systems critical glial developmental events occur in utero and that makes studying glial development challenging due to the difficulty in accessing the animals. Additionally, molecular tools to precisely label and manipulate specific glial subtypes in vivo are limited (Coutinho-Budd and Freeman, 2013). Drosophila melanogaster offers great advantages to study glial development as well as function: the generation time is fast compared to vertebrates and mammals, there are many genetic and molecular tools available to label and manipulate distinct glial and neuronal populations in vivo (including at the single-cell level). Genetic manipulations and mutant clones can be achieved with MARCM technique (Lee and Luo, 1999, 2001) and it is possible to do live-imaging especially in the earlier stages of Drosophila development.
Drosophila has a simpler nervous system than mammals yet it is sophisticated enough to study many aspects of nervous system development and function. Importantly, 12
many functions of mammalian glial cells and structural and molecular mechanisms are
conserved in fly glial cells. Drosophila glial and neuronal linages and subtypes are well
defined and the nervous system is very stereotyped.
Both Drosophila and mammalian macroglial cells (astrocytes, oligodendrocytes
and Schwann cells) derive from neural precursor cells (NPCs). Gliogenesis of lateral glia
(glia other than midline glia) in Drosophila is primarily controlled by a single gene, glial
cells missing (gcm). gcm encodes a transcription factor which is expressed in all glia
except midline glia (Akiyama et al., 1996; Hosoya et al., 1995; Jones et al., 1995;
Schreiber et al., 1997; Vincent et al., 1996). gcm loss in neuroglial progenitors leads to
transformation of presumptive glial cells into neurons, and conversely, gcm gain-of- function lead to transformation of presumptive neurons into glia (Hosoya et al., 1995;
Jones et al., 1995; Vincent et al., 1996). Although Gcm is conserved in mammals, its function in glial fate specification is not conserved in mammals (Akiyama et al., 1996;
Kim et al., 1998). However, a recent study reveals that in both systems the underlying epigenetic pathway is conserved. Glial differentiation from NPCs requires low levels of histone acetylation in Drosophila (Flici et al., 2011); similarly oligodendrocyte differentiation requires low levels of histone acetylation in the mammalian CNS (Shen et al., 2005). Although initial phases of glial differentiation occur by different molecular mechanisms, many morphological and functional properties of glial cells are conserved from flies to mammals (e.g. ensheathment of CNS axons, association with synapses, regulation of synapse formation, BBB formation/function, neurotransmitter recycling, trophic support of neurons, injury response and engulfment, PNS axon ensheathment, 13
interaction with neuronal cell bodies) (Awasaki et al., 2008; Banerjee et al., 2006;
Daneman and Barres, 2005; Doherty et al., 2009; Edenfeld et al., 2005; Freeman and
Doherty, 2006; Freeman and Rowitch, 2013; MacDonald et al., 2006; Rival et al., 2004;
Soustelle et al., 2002).
Glial subtypes in Drosophila
Drosophila glial cells are characterized most extensively in the embryonic stages of development. Types of embryonic glia include: neuropil glia, cortex glia, surface glia, and peripheral glia. Neuropil glia, as the name implies, are closely associated with the neuropil. Similar to oligodendrocytes, neuropil glia ensheath target axons and support proper fasciculation (Ito et al., 1995; Klämbt et al., 1991). Ensheathment of axons is thought to be important for isolation from the extracellular environment to enable for firing.
Longitudinal glia (LG) are a subclass of neuropil glia. LG are derived exclusively from a type of glioblast (GB), the glial precursor (GP). GP lies laterally. After the first division of GP, the progeny begin migrating medially and dorsally. LG then make contact and ensheath motorneurons and interneurons that are developing in the CNS (Schmidt et al., 1997). At late embryonic stages, LG flank the midline and reside dorsally at each side of the midline along the length of the CNS. LG regulate axon pathfinding, axon fasciculation and promote neuron survival (Booth et al., 2000; Hidalgo and Booth, 2000;
Jacobs et al., 1989).
Cortex glia have some physical similarities to mammalian astrocytes as they are closely associated with neuronal cell bodies. Cortex glia are embedded to the cortex, 14
extend their processes around neuronal cell bodies forming honeycomb shaped structures
(Pereanu et al., 2005). Cortex glia have close contact with the BBB and tracheal elements
(Ito et al., 1995; Pereanu et al., 2005). This suggests that cortex glia could supply gases and nutrients to the neurons that they ensheath; this role is similar to mammalian astrocytic functions. In the Drosophila embryo, a glial subtype having specific immune functions like microglia does not exist; all glia seem to have the capability of immune functions, such as engulfment of cell corpses (Freeman et al., 2003; Sonnenfeld and
Jacobs, 1995).
Similar to Schwann cells, peripheral glia, derived from CNS glia, ensheath and support peripheral nerves in the PNS (Auld et al., 1995; Leiserson et al., 2000). During larval stages, peripheral glia invade the NMJ, have dynamic transient interactions with the synaptic boutons of the NMJ and are required for synaptic growth. Similar to astrocytic synaptic clearance in the CNS, peripheral glia remove the presynaptic debris that are continuously shed as NMJ grows during larval development. This clearance is important for normal synaptic growth. When glial clearance of such synaptic debris is blocked, synaptic growth is inhibited (Fuentes-Medel et al., 2009).
There are two types of surface glia: perineurial (PG) and subperineurial glia
(SPG). Both of these glia form a sheath around the CNS, isolating the nervous system from the hemolymph. SPG, the inner ones, constitutes the BBB in embryonic as well as larval and adult stages (Schwabe et al., 2005; Stork et al., 2008). In contrast, the outer layer, PG, proliferate through postembryonic development and so may remain immature until adult stages and possibly helps make up the adult BBB (Awasaki et al., 2008). 15
Neuropil glia can be divided into two glial subtypes in later developmental stages: astrocytes and ensheathing glia. Drosophila astrocytes have not yet been well characterized since genetic tools that nicely label astrocytes were missing. I established a
Gal4 driver line, alrm-Gal4, which nicely labels astrocytes during several developmental stages (Doherty et al., 2009). The alrm-Gal4 driver became a very useful tool and enabled the study of many aspects of Drosophila astrocyte biology. In Chapter II, I characterized Drosophila astrocytes at larval and pupal stages. As shown in Chapter II, astrocytes reside at the edge of the neuropil, exhibit tufted morphology and infiltrate the neuropil region (Doherty et al., 2009). Several observations from our laboratory argue for
Drosophila astrocytes being similar to mammalian astrocytes: Firstly, they exhibit tufted morphology similar to mammalian astrocytes. Secondly, Drosophila astrocytes express the neurotransmitter transporters EAAT and GABAT similar to mammalian astrocytes
(Rival et al., 2004; Stacey et al., 2010) (Tobias Stork, unpublished observations). Thirdly, as shown in Chapter II, Drosophila astrocyte processes are closely associated with synapses (Tobias Stork, unpublished observations). Finally, as shown in Chapter II,
Drosophila astrocytes engulf and clear neuronal material during neural circuit remodeling using Draper signaling. Recently, mammalian astrocytes were shown to prune synapses during retinogeniculate remodeling using MEGF10, a Draper homolog, and MERTK
(Chung et al., 2013).
Ensheathing glia ensheath neuropil structures and are the primary phagocytes in the adult Drosophila brain. Upon injury, they respond by upregulating Draper, the phagocytic receptor necessary for the clearance of debris, and phagocytose axonal debris 16
(Doherty et al., 2009). In this manner, they resemble microglia in the mammalian CNS.
Initially, I made the observation that Drosophila astrocytes do not respond to injury in the same way as ensheathing glia do in the adult. When the olfactory receptor neurons
(ORNs) that project from the antenna to the olfactory lobe in the adult brain are severed, astrocytes do not upregulate Draper or engulf and clear the neuronal debris. I also did not observe a significant morphological change in astrocytes in response to injury (Doherty et al, 2009).
Neural remodeling during Drosophila metamorphosis
Drosophila is a holometabolous insect; the larval stages look very different than
the adult. Between larval and adult stages, the animal goes through metamorphosis, a
stage when the body is transformed extensively. Novel organs emerge from the imaginal
discs, such as wings, eyes, legs, genitalia and antenna; and some organs get destroyed,
such as salivary glands, muscles, midgut and hindgut (Fuchs and Steller, 2011). As the
body undergoes such changes, not surprisingly, nervous system also undergoes extensive
reorganization.
Some larval neurons are conserved through metamorphosis and others get
eliminated through PCD mechanisms. The ones that remain until adulthood undergo
remodeling of both their central and peripheral processes. Moreover, many new neurons
are added during metamorphosis. These adult-specific neurons are born from the
embryonic neuroblasts that had persisted through the larval stage, having arrested their
development after birth. They then mature into functional neurons during metamorphosis
(Truman, 1990). 17
The major steroid hormone in flies, 20-Hydroxyecdysone (20-HE), drives many developmental events throughout the life cycle of Drosophila. Ecdysone is secreted from the prothoracic gland, which is situated in the anterior end of the larva. Then ecdysone is converted to 20-HE (in short will be referred as ecdysone) by the peripheral tissues. This is the active form of the hormone. Before hatching and every molting event, there is a pulse of ecdysone secreted together with another hormone, juvenile hormone (JH).
Disappearance of JH early in the third instar larval stage and then secretion of ecdysone at late third instar stage leads to pupariation, which is the transformation of the larva into the prepupa. At this stage, the prepupa is white in color for an hour period, and is referred as 0hr APF (after puparium formation). It is a convenient marker used to time the subsequent stages. Later, a small titer of ecdysone is secreted around 10hr APF, which is followed by head eversion at 12hr APF and this is the start of the pupal stage. During pupal stages, a higher titer of ecdysone drives adult differentiation from pupa (Truman,
1990).
Ecdysone binds to a heterodimer receptor composed of Ecdysone receptor (EcR) and Ultraspiracles (Usp) (Koelle et al., 1991; Yao et al., 1993). The EcR/Usp complex in turn triggers transcriptional cascades of primary-response genes, BR-C, E74 and E75, and secondary-response genes. The mechanisms by which ecdysone initiates the transformation from larva to adult emerged from studies of giant polytene chromosomes in larval salivary glands. Transcribed genes are represented by puffs, and observations of puffing patterns led to the identification of gene networks regulated by ecdysone
(Ashburner et al., 1974; BECKER, 1959; Thummel, 1996). EcR has three isoforms; EcR- 18
B1, EcR-B2 and EcR-A (Talbot et al., 1993). These isoforms have common DNA- and
ecdysone-binding domains, and the difference is in their N-terminal regions.
Interestingly, tissues belonging to the same metamorphic class have the same expression
pattern of EcR isoforms. EcR-B1 is highly expressed in neurons that undergo pruning or
PCD and is required for these processes (Kuo et al., 2005; Lee et al., 2000; Schubiger et
al., 1998, 2003; Truman et al., 1994). EcR-A is highly expressed in neurons that mature
and grow adult-specific projections (Truman et al., 1994).
One subtype of neurons that die during metamorphosis by PCD is peptidergic
vCrz+ neurons. Corazonin neuropeptide is expressed in a few neurons in the brain and eight pairs in the ventral nerve chord (VNC). The ones in the VNC (vCrz+) die through apoptosis and by 6hr APF, almost all of their debris is cleared from the neuropil (Choi et al., 2006). The EcR-B1 isoform is expressed in these neurons and both EcR-B1 and –B2 isoforms are involved in the PCD of vCrz+ neurons.
Some neurons do not die but remodel their projections. For example, class IV da
(C4da) sensory neurons in the body wall remodel their dendrites during metamorphosis.
These neurons express EcR-B1 and ecdysone signaling is required for the initiation of dendrite degradation (Kuo et al., 2005). In the CNS, an extensively studied pruning event occurs in mushroom body (MB) γ neurons. MB neurons are an important component of learning and memory in Drosophila. There are 3 MB neuron subtypes; α/β, γ and α’/β’.
Dendrites and axonal projections of the γ subtype undergo local degeneration and by 18hr
APF, the debris are eliminated. These neurons also express the EcR-B1 isoform, and ecdysone signaling is required for the initiation of their remodeling (Lee et al., 2000; 19
Watts et al., 2003). Pruning is a multistep event that includes destabilization of the
cytoskeleton, fragmentation, and clearance of debris.
In this thesis, I studied the PCD of vCrz+ neurons and MB γ neuron pruning mechanism as well as synaptic removal during metamorphosis. Drosophila offers a nice system to study neuronal circuit refinement since the basic mechanisms and some molecular components are conserved through evolution. It is easy to stage the animals, the timing of metamorphic events are the same between animal to animal, the cell populations are stereotyped, and there are many genetic tools to manipulate specific neuronal and glial populations.
Although there are mechanisms of PCD and local pruning described in Drosophila, the underlying molecular mechanisms involved in each type of remodeling are still a mystery. Studies of synaptic removal in the Drosophila CNS and the molecular mechanisms involved for their clearance is lacking. Also, as will be discussed later, the
(glial) cell types that are in charge of clearing the debris from the environment and the underlying molecular mechanisms used to accomplish that is unknown.
Molecular mechanisms of neural remodeling
Studies in C. elegans led to the discovery of caspases, which are involved in PCD.
Ced-3 was the first caspase identified to function in PCD (Yuan and Horvitz, 1990). Ced-
3 is a cysteine protease, which cleaves proteins after an aspartate residue. Therefore the
name caspase comes from cysteine requiring aspartate protease. In mammals, in response
to various stimuli, mitochondria outer membrane permeability rises and that leads to the
release of cytochrome c (cyt c). In flies, cyt c may be released or remain on the surface of 20
the mitochondrial outer membrane (Wang and Youle, 2009). cyt c in the cytoplasm binds
to apoptosis protease activating factor-1 (Apaf-1), the mammalian homolog of ced-4, and that leads to procaspase-9 recruitment to this complex that is called an apoptosome.
Activated caspase-9 cleaves and activates pro-caspase 3, the mammalian homolog of ced-
3. Fly homolog of Caspase-9 is Dronc and of Caspase-3 are Drice and Dcp-1. Dronc is an initiator caspase and Dcp-1 and Drice are effector caspases. Inhibitor of apoptosis (IAP) proteins suppress apoptosis by directly inhibiting caspases both in flies and mammals
(Fuchs and Steller, 2011; Kumar, 2007; Riedl and Shi, 2004).
As mentioned, many elements in the death pathway are conserved in flies and mammals, although the pathways are more sophisticated in mammals. The activation of death by apoptotic proteins is regulated by some mechanisms to ensure the death of specific neurons; mitochondrion membrane permeability is controlled by large family of proteins so that cyt c does not leak out. Secondly, activation of pro-caspases are tightly regulated. The Bcl-2 family has members that promote survival, as well as those that promote cell death; e.g. Bax is pro-apoptotic and Bcl-x is anti-apoptotic. Disruption of the bcl-x gene leads to increased PCD in neurons (Motoyama et al., 1995). Conversely, in bax knockout mice, PCD is blocked in sympathetic ganglia and motor neurons and significantly reduced in many areas of the CNS (Fan et al., 2001; White et al., 1998).
Double knockout mouse deficient for both bcl-x and bax had PCD levels returned to normal in the spinal cord (Shindler et al., 1997).
In Drosophila, developmental cell death is determined by three pro-apoptotic proteins: Reaper (Rpr), Grim and Hid (Head involution defective). In cells that die during 21
metamorphosis, activation of EcR leads to activation of these death proteins, which in
turn removes IAP-mediated negative regulation of caspases and leads to their activation.
PCD of vCrz+ neurons also involves caspase activity. Overexpression of p35, a caspase inhibitor, suppress the death of vCrz+ neurons for at least 1 day (Choi et al., 2006), and I
found this to be true for at least 2 days (data not shown). Blockade of ecdysone signaling
in these neurons or analysis of EcR mutants showed survival of these neurons; therefore
ecdysone signaling initiates PCD in vCrz+ neurons (Choi et al., 2006).
Caspases may also have non-apoptotic functions. For example, local caspase
activity was shown to be confined to the dendrites of C4da neurons during pruning
(Williams et al., 2006). However, caspase activity is not involved in MB γ neuron axon
pruning. Caspase-3 immunoreactivity was not detected around MB γ neuron cell bodies
or axons. Additionally, expression of p35 or DIAP1, both of which inhibit caspases, does
not affect MB γ neuron remodeling (Awasaki et al., 2006). Such differences in the
requirement of local caspase activity between dendritic and axonal remodeling shows
differences in the molecular pathways used.
Ecdysone signaling and EcR/Usp action is cell-autonomously required for MB γ
neuron remodeling (Lee et al., 2000). As expected, only the γ neuron subtype expresses
EcR-B1, not the α’/β’ subtype (Lee et al., 1999). It has been found that Baboon, the
Drosophila TGF-β/activin type I receptor, regulates EcR-B1 expression and therefore
mediates MB γ neuronal remodeling (Zheng et al., 2003). As will be discussed later,
Myoglianin (Myo) secreted from glia is a TGF-β ligand and instructs the remodeling
(Awasaki et al., 2011). There is another molecule regulating EcR-B1. FTZ-F1 is a 22
nuclear receptor that regulates neuron remodeling by activating EcR-B1 and repressing
HR39. This pathway is independent of the TGF-β pathway, suggesting that EcR-B1 expression is regulated by at least two different parallel pathways (Boulanger et al.,
2011). The ubiquitin-proteasome system (UPS) is required for MB γ neuron remodeling to occur (Watts et al., 2003). In the UPS pathway, ubiquitin (Ub), 76-aa protein, is covalently linked to a substrate through activity of E1, E2 and E3 enzymes.
Polyubiquitinated proteins are then targeted for degradation by 26S proteasome
(Weissman, 2001). The UPS also regulates dendrite pruning of C4da neurons (Kuo et al.,
2006).
Similar to dendrite pruning but in contrast to MB γ neuron axon pruning, caspases are also utilized in neurite pruning during development in mice. Preventing DR6 function, a death receptor protein, delays pruning of sensory axons in vitro and of retinocollicular axons in vivo. Additionally, inhibition of Bax by chemicals and knockdown of Caspase-6 were shown to prevent sensory axon degeneration in Campenot chambers (Nikolaev et al., 2009). A later study showed that caspase-3 and caspase-6 knockout mice have a delay in developmental pruning of retinocollicular axonal projections of RGCs in vivo (Simon et al., 2012).
There has not been much progress in understanding the underlying molecular mechanisms regulating either the small-scale or large-scale pruning events in the mammalian nervous system. Layer V subcortical axon collaterals undergo large-scale pruning and only Otx1 has been identified as regulating their pruning (Weimann et al.,
1999). Another large-scale pruning event that has been studied occurs during topographic 23
mapping of the RGC projections in optic tectum (OT) of chick or its mammalian homolog, the superior colluculus (SC) of rodents. Ephs and ephrins have a graded expression pattern in the retina and SC, respectively. This graded pattern is crucial for axon guidance and governs the development of the initial retinotopic map so that sufficient RGC connections form at the proximity of the topographically correct location.
After initial rough map formation, retinal waves generate correlated patterns of RGC activity that promotes synaptic potentiation, as well as depression. Such correlated RGC activity is required during a brief critical period for establishment of a refined topographic map (O’Leary and McLaughlin, 2005). Correlated firing of RGCs leads to strengthening of connections and noncorrelated firing of RGCs drives competitive interactions between them, which leads to their segregation or elimination (Zhang and
Poo, 2001). Mice lacking nicotinic acetylcholine receptor (nAChR) β2 subunit selectively lack nAChR-mediated retinal waves during the first postnatal week before the onset of visually evoked patterned activity and RGC projections do not undergo topographic remodeling (Bansal et al., 2000; O’Leary and McLaughlin, 2005). As mentioned previously, activity-dependent remodeling also occurs in RGC projections in dLGN as well as at the NMJ.
Although in vertebrates, neural-activity dependent competition between axons is involved in both synaptic, local arbor and large-scale arbor pruning, in Drosophila, refinement of circuits seems to be regulated by genetic mechanisms and the brain circuits may to a large extent be hardwired prior to eclosion. For example, in the visual system, ultrastructural analysis of mutants with defects in the generation of electrical potentials 24
(norpA, trp;trpl), neurotransmitter release (hdc, syt), or vesicle endocytosis (synj) showed
that synaptic targeting, synapse formation or refinement of synapses in pupal
photoreceptors is independent of neural activity (Hiesinger et al., 2006). Also at the NMJ,
although during larval growth neural activity is involved in the growth of initial NMJs
(Fuentes-Medel et al., 2009), initial wiring of motorneurons is accomplished during late
embryogenesis and the muscles remain multiply innervated, suggesting that competition
doesn’t occur in Drosophila (Keshishian et al., 1993). Adding to that, in the olfactory
system, it has been shown that olfactory receptor neurons (ORNs) and their postsynaptic
targets, projection neurons, are independently pre-specified, adding to the hardwiring
view of circuits in Drosophila (Jefferis et al., 2001). In contrast to this view, some studies
suggest that neural activity has a function during reorganization of some circuits. Just
after eclosion of the adult, the optic lobe volume grows and this growth is sensitive to
visual experience (Barth et al., 1997). During metamorphosis, CSD interneurons,
serotonergic neurons that have dendritic projections in the antennal lobe, prune their
axonal and dendritic processes in an ecdysone-dependent manner. Expression of tetanus
toxin (to abolish total evoked and most spontaneous release), temperature-sensitive
shibire (shits) (to block synaptic vesicle recycling and neurotransmitter release) and
expression of an inwardly rectifying K+ channel (to inhibit action potential generation) in these neurons leads to defects in their patterning (Roy et al., 2007). However, it should be noted that CSD neurons do not elaborate excess projections to be eliminated later, like seen in RGC connections in the mammalian brain, and such manipulations did not alter pruning but prevented the inhibition of later phases of projection elaboration. 25
Glial phagocytic functions during circuit remodeling and in response to injury
As mentioned previously, during nervous system development, excess numbers of cells and connections are made and extensive cell death and connection refinement occurs. These changes lead to immense amounts of debris, which if not dealt with efficiently, can have toxic effects on other cells, as well as stimulate autoimmune and inflammatory responses (Albert et al., 1998; Ren and Savill, 1998; Sauter et al., 2000;
Savill et al., 2002). One of the important functions of glia is the phagocytosis and elimination of such excess axonal or dendritic projections, cell corpses and debris generated during development as well as in response to injury and disease. As programmed cell death is a conserved mechanism to get rid of harmful material from the body, phagocytosis is also conserved through evolution. In C. elegans, there are no professional phagocytes, and cells dying by PCD are engulfed by neighboring cells of many types. In fact, studies in C. elegans led to the discovery of the molecular pathways that are required for clearance of cell corpses (Reddien and Horvitz, 2004).
Microglia are the resident immune cells and are the professional phagocytes in the mammalian brain. They dynamically survey the brain and have been implicated in the removal of debris in neurodegenerative diseases and in response to injury (Garden and
Möller, 2006; Magnus et al., 2002; Mallat et al., 2005). Microglia have recently been implicated in synaptic removal during developmental retinogeniculate refinement
(Stevens et al., 2007). Astrocytes secrete factor(s) that lead to localization of C1q, the initiating protein in the complement cascade, at synapses. In the immune system, C1q 26
opsonizes the dead cells/debris and triggers a cascade that leads to downstream complement protein C3 deposition (Gasque, 2004). Microglia express complement receptor 3(CR3)/C3 which presumably responds to the C3 at the synapse and leads to synapse engulfment since both C3 and CR3 KO mice had decreased level of engulfment of synapses by microglia and refinement defects (Schafer et al., 2012). It would be interesting to find out if microglia are involved in synaptic pruning in other regions of the brain as well.
Astrocytes also have phagocytic functions but are generally implicated as non- professional phagocytes as they respond to injury less efficiently than microglia (Parnaik et al., 2000). In vitro, astrocytes can have phagocytic properties (Roldán et al., 1997;
Tansey and Cammer, 1998), engulf apoptotic cells (Chang et al., 2000) and amyloid-β
(Wyss-Coray et al., 2003). In vivo, MTZ astrocytes in the postlaminar optic nerve head express Mac-2, a molecule implicated in phagocytosis of myelin by microglia, and internalize axonal cytoplasm and organelles, and this event appears to be increased in frequency in glaucoma models (Nguyen et al., 2011). Until recently, a phagocytic role for astrocytes during developmental pruning has not been shown directly. Recently, astrocytes also have been shown to mediate synapse elimination during developmental retinogeniculate refinement, and this requires the MEGF10 (the mammalian homolog of
Draper) and MERTK phagocytic pathways (Chung et al., 2013). It will be interesting to find out the interplay between astrocytes, microglia and synaptic connections in the future. How are synapses tagged by glia? Which synapses are engulfed by astrocytes and microglia? Are microglia and astrocytes redundant in synaptic refinement at RGC 27
synapses or do they engulf different types of synapses? How do glia decide which
synapses to remove? Are astrocytes as efficient in synaptic removal as microglia?
Glia or phagocytes do not randomly engulf neurons or other cells; they recognize
‘eat-me’ or ‘do not eat-me’ cues on other cells and soluble ‘come-get-me’ signals. Eat-me
signals on apoptotic cells are markers recognized by phagocytes to engulf. The best
characterized eat-me cue is phosphatidylserine (PS); a phospholipid which gets
translocated from the inner to the outer leaflet of the lipid bilayer when the cell becomes
apoptotic (Fadok et al., 1992, 1998, 2001; Hoffmann et al., 2001; Schlegel and
Williamson, 2001). PS externalization is a conserved mechanism as it occurs in C.
elegans, Drosophila and mammals. Living cells present do not-eat-me signals to
phagocytes to prevent their engulfment. During apoptosis, such signals are somehow
disabled and this leads to engulfment (Brown et al., 2002). Come-get-me signals are
soluble attraction cues that enable phagocytes to migrate to the sites of apoptosis (Lauber
et al., 2003).
In Drosophila, cell death occurs in three phases during development: in mid-to- late embryonic stages, metamorphosis and after eclosion. During embryogenesis, almost half of the cells die in the nervous system (Abrams et al., 1993; Rogulja-Ortmann et al.,
2007). Glia, being the principal phagocytes, engulf and clear the cell corpses (Freeman et al., 2003; Kurant et al., 2008). During metamorphosis, both PCD and neuronal pruning take place. Yet, the cell type required to clear the debris remains elusive. As axons of MB
γ neurons undergo local degeneration, glial cells have been shown to engulf and clear their debris (Awasaki and Ito, 2004; Watts et al., 2004). However, the type of glia 28
responsible for clearance of MB γ neuronal debris has not been investigated. In my thesis,
I describe that astrocytes are the primary phagocytes removing MB γ neuronal debris.
vCrz+ neurons undergo apoptosis which is regulated by a steroid-dependent program.
Yet, the cell type responsible of clearing their debris has not been studied before. In my thesis, I describe that glia clear the debris of vCrz+ neurons. Interestingly, astrocytes clear their neurites whereas nonastrocytic glia clear their cell bodies.
Glial phagocytic activity is crucial during synapse remodeling. At the mammalian
NMJ, as axons compete with each other and synaptic elimination takes place, the loser axon retracts and sheds axosomes which are membrane bound remnants. These axosomes have been found in Schwann cells, suggesting glial engulfment of the retracting axon parts (Bishop et al., 2004). It remains to be determined if Schwann cells are active participants in deciding which axonal parts are removed, or if they play a more passive role and simply engulf already fragmented axons. During Drosophila larval NMJ development, presynaptic neurons continuously shed debris. Peripheral glia, which are closely associated with the NMJ, are responsible for clearing the debris shed at the synapse. Blocking glial engulfment prevents synaptic growth (Fuentes-Medel et al.,
2009).
In the Drosophila adult, our lab made significant contributions in the generation of an injury assay and in uncovering the glial responses to injury. Olfactory receptor neurons (ORNs) project from the antenna and the maxillary pulp to the antennal lobe in the adult brain and removal of antennae or maxillary palps leads to the degeneration of
ORNs. Upon such injury, glia respond by upregulating the phagocytic receptor Draper 29
and clear neuronal debris in the central brain (MacDonald et al., 2006). Subsequently, different subtypes of glia have been investigated for response to injury, and ensheathing glia have been shown to be the glial subtype responsible for clearance (Doherty et al.,
2009). I made the initial observation in the lab that astrocytes do not respond to injury or
clear neuronal debris like ensheathing glia do, and general astrocyte morphology looks
normal upon injury (Doherty et al., 2009). However, it should be noted that astrocytes
may have other functions than engulfment of debris. Through secretion or cell-cell
contact, they may signal to ensheathing glia to aid the clearance of debris.
Glia not only engulf neuronal corpses, but also promote the destruction of the
target cell. In the cerebellum, large number of Purkinje cells (PCs) die by apoptosis.
Microglia have been found to be in contact with or engulf some of the PCs, which
expressed Caspase-3, an early indicator of apoptosis. Ablation of microglia in slices led
to a significant reduction in the death of PCs. These results strongly argue that engulfing
microglia promote the death of Purkinje cells, possibly through superoxide ions (Marín-
Teva et al., 2004). However, in that study, the engulfment machinery isn’t implicated to
have a role in promoting PCD. The role of the engulfment machinery in promoting the
death of the target cell has been shown in C. elegans. Partial loss of function of a killer
gene leads to the survival of some cells that normally undergo apoptosis, and mutations
in engulfment genes (ced-1, -6, -7, -2, -5, -10, and -12) enhanced the frequency of the
survival of those cells. Also, mutations in the engulfment genes alone led to survival of
some cells that would normally die (Hoeppner et al., 2001; Reddien et al., 2001). As will
be described in Chapter II, during my studies I found that glia promote the death and 30
fragmentation of vCrz+ neurons during metamorphosis through Draper (Drosophila homolog of CED-1). In addition, draper null mutations were recently shown to delay the fragmentation of dendrites in C4da sensory neurons which undergo remodeling during metamorphosis (Han et al., 2014).
Other than their phagocytic functions, glia also instruct the remodeling of neurons. Glia (cortex glia and astrocytes) were shown to secrete Myo, which activates the
TGFβ receptor pathway in MB γ neurons, which induces the expression of EcR-B1. This is critical for MB γ neuron remodeling since block of Myo in glia disrupts pruning of MB
γ neurons (Awasaki et al., 2011). This shows that glia-to-neuron signaling is critical for proper circuit remodeling in this system.
Molecular mechanisms of engulfment
The molecular pathways crucial for the engulfment of cell corpses have been
initially discovered from screens done in C. elegans. A number of cell death abnormal
(ced) mutants have been identified that disrupt engulfment of cell corpses (Ellis et al.,
1991; Hedgecock et al., 1983; Reddien and Horvitz, 2004). Two partially redundant
pathways have been identified. In one pathway, CED-2/CED-5/CED-12 forms a complex and acts as a guanine nucleotide exchange factor (GEF) that activates Rac1 (Brugnera et al., 2002). Activated Rac1 modulates cytoskeletal arrangements during cell corpse engulfment (Zhou et al., 2001a). The second pathway is composed of the transmembrane receptor CED-1 and its adaptor protein CED-6. CED-1 acts in the phagocytic cell as a phagocytic receptor to recognize an eat-me cue on corpses (Zhou et al., 2001b), and activates downstream pathways through the phosphotyrosine-binding (PTB) domain- 31
containing CED-6 (Liu and Hengartner, 1998) to engulf and degrade the corpses (Yu et al., 2008). Loss of either pathway results in partial suppression of cell corpse engulfment, and inactivation of both pathways enhanced this phenotype (Ellis et al., 1991; Gumienny et al., 2001; Kinchen et al., 2005; Zhou et al., 2001b). Studies showed that the orthologs of both of these pathways have conserved functions in elimination of cell corpses in
Drosophila and vertebrates (Etchegaray et al., 2012; Van Goethem et al., 2012; Van Ham et al., 2012; Manaka et al., 2004; Smits et al., 1999), as well as by glia in the nervous system (Freeman et al., 2003; Wu et al., 2009).
Draper, the Drosophila homolog of CED-1, has been shown to be required for clearance of pruned MB γ axonal debris during metamorphosis (Awasaki et al., 2006), pruned dendrites of C4da neurons during metamorphosis (Williams et al., 2006), presynaptic debris shed during larval NMJ growth (Fuentes-Medel et al., 2009), and axonal debris generated by the injury-induced degeneration of ORNs in the adult fly
(MacDonald et al., 2006). The Crk/Mbc/dCed-12 complex, the Drosophila homolog of
CED-2/CED-5/CED-12, has been shown to promote phagocytosis during clearance of axonal debris generated by injury (Ziegenfuss et al., 2012). The function of this complex in the pruning of neuronal circuits has not been investigated in Drosophila. Although glia have been implicated as responsible for neuronal pruning of MB γ neurons in Drosophila
(Watts et al., 2004), precisely the subtypes of glia responsible in pruning remains unclear.
Jedi-1 and MEGF10, mammalian homologs of Draper, are expressed in satellite precursor cells and have roles in phagocytosis of apoptotic cell corpses in developing dorsal root ganglia (DRG) (Wu et al., 2009). Jedi-1 and MEGF10 mediate engulfment 32
through the tyrosine kinase Syk (Scheib et al., 2012), the mammalian homolog of Shark, which has been shown to have the same role in axonal debris clearance in Drosophila
(Ziegenfuss et al., 2008). MEGF10, along with MERTK, has recently been shown to have a role in astrocytic synaptic clearance during developmental retinogeniculate refinement
(Chung et al., 2013). MERTK has also been shown to regulate CrkII/Dock180/Rac1 modules, which are the mammalian homologs of Crk/Mbc/Rac1, for controlling cytoskeletal rearrangements during phagocytosis in vitro (Wu et al., 2005). Brain-specific angiogenesis inhibitor 1 (BAI1) has been implicated as an engulfment receptor upstream of ELMO/Dock180/Rac (homolog of dCed-12/Mbc/Rac1) and was shown to bind PS by its thrombospondin type 1 repeats within its extracellular region (Park et al., 2007). BAI1 is expressed in neurons as well as astrocytes in the mammalian brain and in culture.
BAI1-expressing astrocytes were shown to engulf apoptotic debris with BAI1 accumulating within the phagocytic cup (Sokolowski et al., 2011). It would be interesting to find out if astrocytes use the BAI1 pathway, as well as MEGF10 and MERTK, in the removal of debris during development in vivo.
The structures shared by CED-1, Draper and MEGF10 include EGF-like repeats in the extracellular domain, a single transmembrane domain, and two common motifs,
NPXY and YXXL in the intracellular domain. NPXY motif is a binding site for PTB domain proteins and YXXL motif is a binding site for SH2 domain-containing proteins.
In C. elegans, CED-1 promotes engulfment by activating downstream signal transduction through both of these motifs (Zhou et al., 2001b). CED-6, a PTB domain adaptor protein, binds to the NPXY motif of CED-1 and is important for cell corpse removal (Liu and 33
Hengartner, 1998; Su et al., 2002). The non-receptor tyrosine kinase Shark (the
Drosophila homolog of Syk), binds to Draper through YXXL immunoreceptor tyrosine- based activation motif (ITAM) and is essential for phagocytosis of axonal debris and neuronal cell corpses (Ziegenfuss et al., 2008). The ITAM motif of Draper is a key domain found in mammalian immunoreceptors such as Fc, T-cell and B-cell receptors.
These data indicate that Draper and the Crk/Mbc/dCed-12 pathways are evolutionarily conserved by structure as well as function, and studying phagocytosis in Drosophila would give essential molecular insight into the cellular and molecular mechanisms of engulfment.
In Chapter II, I describe a phagocytic role for astrocytes in Drosophila. I investigate the function of the Draper and Crk/Mbc/dCed-12 pathways in astrocytes during the removal of pruned neuronal debris, apoptotic neuronal debris, and synaptic clearance during metamorphosis. I identified Crk/Mbc/dCed-12 as a new signaling pathway required for pruning of neurites during metamorphosis. I found that these two pathways are utilized differentially by glia in the clearance of different subsets of neurons that remodel by different mechanisms.
34
CHAPTER II: Astrocytes engage unique molecular programs to engulf pruned neuronal debris from distinct subsets of neurons
The authors wish to thank all members of the Freeman laboratory for helpful advice and discussions, and Tobi Stork, Lukas Neukomm, Megan Corty, Tsai-Yi Lu,
Jaeda Coutinho-Budd and A. Nicole Fox for critical reading of the manuscript. We thank
Amy Sheehan for production of the alrm-Gal80 construct, and Takeshi Awasaki, Tobi
Stork, and Jae Park for generously providing reagents. nc82 (developed by E. Buchner), anti-Repo (8D12; developed by C. Goodman) and 1D4 (developed by C. Goodman) were obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biology, Iowa
City, IA 52242. The EM portion of the project was supported by Award Number
S10RR027897 from the National Center For Research Resources. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health. This work was supported by NIH grant NS053538 (to MRF) and MRF is an Investigator of the
Howard Hughes Medical Institute.
The following work is reprinted from the Genes and Development article of the same name published as:
Ozge E. Tasdemir-Yilmaz and Marc R. Freeman. Genes and Development, January 1,
2014; 28(1):20-33. 35
Abstract
Precise neural circuit assembly is achieved by initial over-production of neurons and
synapses, followed by refinement through elimination of exuberant neurons and
synapses. Glial cells are the primary cells responsible for clearing neuronal debris, but the
cellular and molecular basis of glial pruning is poorly defined. Here we show that
Drosophila larval astrocytes transform into phagocytes through activation of a cell-
autonomous, steroid-dependent program at the initiation of metamorphosis and are the primary phagocytic cell type in the pupal neuropil. We examine the developmental elimination of two neuron populations—mushroom body (MB) γ neurons and vCrz+
neurons—where only neurites are pruned, or where entire cells are eliminated,
respectively. We find that MB γ axons are engulfed by astrocytes using the Draper and
Crk/Mbc/dCed-12 signaling pathways in a partially redundant manner. In contrast, while
elimination of vCrz+ cell bodies requires Draper, elimination of vCrz+ neurites is mediated by Crk/Mbc/dCed-12, but not Draper. Intriguingly, we also found elimination of Draper delayed vCrz+ neurite degeneration, suggesting that glia promote neurite
destruction through engulfment signaling. This work identifies a novel role for astrocytes
in clearance of synaptic and neuronal debris, Crk/Mbc/dCed-12 as a new glial pathway
mediating pruning, and reveals, unexpectedly, that the engulfment signaling pathways
engaged by glia depend upon whether neuronal debris was generated through cell death
or local pruning.
36
Introduction
In complex metazoans neurons neurites, and synapses are produced in excess
early in development, and this is followed by a later phase of selective elimination of
subsets of neurons, or parts of neurons (termed pruning). The overproduction of neurons
or formation of excessive connections can be remarkable—~80% of neurons in
mesencephalic nucleus of the trigeminal nerve of chick embryos undergo cell death (von
Bartheld and Bothwell, 1993; Cowan et al., 1984); and in the mammalian
retinogeniculate system thalamic neurons prune ~10 synaptic inputs for every successful
input retained (Chen and Regehr, 2000). Such early exuberant brain construction is
thought to impart increased developmental plasticity through which trophic and activity-
dependent mechanisms can be utilized to fine-tune neuronal numbers and connectivity for optimum circuit performance.
Neural circuit refinement can proceed by the wholesale elimination of a neuron and its projections, or by the selective destruction of specific axons, dendrites or synapses. For instance, ~50% of chick ciliary ganglion cells undergo cell death to remove entire cells after they have received afferent synaptic input and elaborated axonal projections (Landmesser and Pilar, 1972, 1974). Similarly, ~40% of mouse GABAergic inhibitory cortical neurons undergo cell death after they have migrated from their place of birth, acquired morphological features of cortical inhibitory neurons, and formed synaptic contacts with surrounding cells (Southwell et al., 2012). In contrast, other mammalian neurons selectively prune only specific parts of cells. For example motorneurons eliminate individual branches when neuromuscular junction (NMJs) resolve states of 37
poly-innervation to mono-innervation (Bishop et al., 2004; Walsh and Lichtman, 2003); and branch-specific elimination occurs during eye-specific segregation of retinal ganglion cell inputs into the dorsal lateral geniculate nucleus (Chen and Regehr, 2000; Hooks and
Chen, 2006).
Neural circuit refinement leads to the production of large amounts of neuronal debris in the form of neuronal cell corpses, fragmented axons and dendrites, and pruned synapses requiring disposal. Impressively, essentially all developmentally produced neuronal debris is rapidly eliminated from the nervous system through the activity of phagocytic glia. In the mammalian CNS microglia are primarily responsible for clearing neural debris (Cuadros and Navascués, 1998; Marín-Teva et al., 1999; Parnaik et al.,
2000), and in the PNS Schwann cells appear to play a key role in the elimination of exuberant motor neuron inputs (Bishop et al., 2004). The timely clearance of CNS debris is essential to avoid tissue inflammation and autoimmune responses (Savill et al., 2002).
In addition, the nervous system appears especially sensitive to the presence of neural debris during synapse formation. For instance, suppressing the engulfing activity of glia and/or muscle at the Drosophila NMJ leads to accumulation of neural debris at the NMJ, and in turn inhibits synaptic growth (Fuentes-Medel et al., 2009).
Molecular insights into the removal of cell corpses came first from work in C. elegans where a number of cell death abnormal (ced) mutants were found to delay or block the engulfment of dead cells (reviewed in Reddien and Horvitz 2004). The CED-
2/CED-5/CED-12 complex, which acts as a guanine nucleotide exchange factor (GEF)
that activates Rac1 (Brugnera et al., 2002), was found to modulate cytoskeletal events 38
during cell corpse internalization (Zhou et al., 2001a). A second pathway composed of
CED-1 and CED-6 was also found to act genetically in parallel to the CED-2/CED-
5/CED-12 pathway— loss of either pathway lead to a partial suppression of cell corpse engulfment, while simultaneous inactivation of both pathways suppressed clearance much more strongly (Ellis et al., 1991; Gumienny et al., 2001; Kinchen et al., 2005; Zhou et al., 2001b). CED-1 is a transmembrane receptor that acts in the engulfing cell and is thought to recognize cell corpses (Zhou et al., 2001b), and degrade them (Yu et al., 2008) through the PTB-domain containing protein CED-6 (Liu and Hengartner, 1998). Similar roles for Drosophila and vertebrate orthologs of these genes have revealed conserved functions for these pathways in the elimination of cell corpses (Etchegaray et al., 2012;
Van Goethem et al., 2012; Van Ham et al., 2012; Manaka et al., 2004; Smits et al., 1999), including by glia in the nervous system (Freeman et al., 2003; Wu et al., 2009).
Draper, the Drosophila homolog of CED-1, is also required in glia for the removal of pruned MB γ axonal debris during metamorphosis (Awasaki et al., 2006), sensory neuron dendrites (Williams et al., 2006), presynaptic debris shed during larval
NMJ growth (Fuentes-Medel et al., 2009), and axons that have undergone injury-induced degeneration in the adult (MacDonald et al., 2006). Likewise we recently discovered a role for Crk/Mbc/dCed-12 (Drosophila CED-2/CED-5/CED12) in the clearance of axonal debris after axotomy (Ziegenfuss et al., 2012). However precisely which subtypes of Drosophila glia are responsible for neuronal pruning in the CNS remains unclear.
Mammalian microglia have been ascribed a role in synaptic pruning (Paolicelli et al.,
2011; Schafer et al., 2012; Stevens et al., 2007), but additional subtypes of glia may also 39
sculpt neuronal circuits, and molecular pathways mediating glial pruning of neurons
remain poorly defined. In this study we describe a key role for astrocytes in clearance of
neural debris during neural circuit reorganization. We identify the Crk/Mbc/dCed-12 complex as a new signaling pathway required for glial pruning of neurites. Surprisingly, we find the precise molecular pathways engaged by glia to internalize neurite debris are linked with the type of intrinsic molecular programs used to drive neurite degeneration.
Results
Astrocytes acquire phagocytic properties during neural circuit remodeling
We wished to define the cellular and functional changes that occur in larval astrocytes
during the larval to adult transition. We therefore examined the morphology of astrocytes
in the Drosophila brain and ventral nerve cord (VNC) during larval and pupal stages
using the alrm-Gal4 driver to express membrane-tethered GFP (alrm>GFP) (Doherty et
al., 2009) specifically in astrocytes. At third instar larval (L3) and at white puparium
stages (i.e. 0hr after puparium formation, APF), astrocyte cell bodies resided at the edge
of the neuropil (Figure 2.1A, Figure 2.8A, B). Astrocyte processes extended exclusively
into the neuropil, exhibited a tufted morphology, with very fine processes infiltrating the
entire neuropil and associating closely with synapses (Figure 2.1A, 3A; Figure 2.8A, B;
T. Stork and M. Freeman, unpublished observations). Unexpectedly, astrocyte processes underwent a dramatic transformation at puparium formation such that at 1hr APF fine processes became slightly enlarged and took on a vacuolated appearance (Figure 2.1A).
Vacuolization of astrocyte membranes became more extreme by 6hrs APF, later astrocyte 40
processes became progressively sparser in the CNS, and by 48hrs APF astrocyte
membranes were absent from the neuropil (Figure 2.1A). Throughout metamorphosis
astrocyte cell bodies remained at the periphery of the neuropil, and their membranes
appeared to re-infiltrate the neuropil (by 96hrs APF) prior to eclosion as adult animals
(Figure 2.1A).
We performed transmission electron microscopy (TEM) to define the cellular
basis of this astrocyte transformation in greater detail. We compared the morphology of
astrocytes in the VNC of control animals at L3 (Figure 2.9) and 6hrs APF (Figure 2.1B).
We defined astrocytes according to the following criteria: (1) the position of their cell
bodies adjacent to the neuropil, (2) their electron-dense cytoplasm compared to surrounding cells, and (3) the fibrous morphology infiltrating to the neuropil. Consistent with our light microscopic observations, we found that L3 astrocyte membranes were fibrous and extended profusely into the synaptic neuropil (Figure 2.9). By 6hrs APF however, astrocyte morphology had changed dramatically: astrocyte membranes, when compared to those of L3 animals, were less fibrous and their cytoplasm was filled with many large vacuoles that appeared to be filled with cellular debris, suggesting that astrocytes were engulfing neuronal material.
We examined expression of Draper, a marker for engulfing cells in Drosophila
(Doherty et al., 2009; MacDonald et al., 2006; Ziegenfuss et al., 2008, 2012), using
Draper antibody and found that astrocytes at L3 expressed low to undetectable levels of
Draper. However at 1hr APF, Draper became detectable in astrocytes, and by 6hrs APF astrocytic Draper levels were significantly elevated (Figure 2.1C) and specific to 41
astrocytes in the neuropil (Figure 2.10, 2.11). dCed-6, which is required for Draper signaling, was expressed in astrocytes at L3 and 6hrs APF, although in contrast to Draper total dCed-6 levels did not change noticeably between L3 and 6hrs APF (Figure 2.12).
Finally, lysosomal activity (detected by Lysotracker) was not obvious in the neuropil at
L3, however robust lysosomal activity was observed in the neuropil by 6hrs APF and all
Lysotracker+ staining was found within astrocytic vacuoles (Figure 2.1D). Together these data argue that astrocytes take on a phagocytic phenotype at pupariation, and suggest that astrocytes may be the primary phagocytic cell type in the pupal neuropil.
EcR cell-autonomously regulates transformation of astrocytes into phagocytes
The steroid hormone 20-Hydroxyecdysone (ecdysone) is a major regulator of neuronal remodeling and cell death during metamorphosis (Jiang et al., 1997; Robinow et al.,
1993; Schubiger et al., 1998; Truman et al., 1994; White et al., 1997). Given that astrocytes transform into phagocyte-like cells at this time of intense steroid-mediated developmental signaling, we wished to determine whether ecdysone signaling modulated astrocyte phagocytic phenotypes.
Ecdysone binds to a heterodimer receptor composed of the Ecdysone receptor
(EcR) and Ultraspiracles (Usp) (Koelle et al., 1991; Yao et al., 1993). There are multiple isoforms of EcR termed A, B1, and B2 (Talbot et al., 1993). EcR-B1 is expressed at high levels in neurons as they initiate pruning of larval projections (Truman et al., 1994) and is required for pruning (Kuo et al., 2005; Lee et al., 2000; Schubiger et al., 1998, 2003), while EcR-A appears to be up-regulated as neurons initiate outgrowth of adult-specific 42
branches (Truman et al., 1994). Using antibodies specific for EcR-B1 and EcR-A we
found that astrocytes express EcR-B1 at 0hr APF (Figure 2.2A) but not EcR-A (Figure
2.13).
We next assayed for EcR function in astrocytes by driving expression of two
different dominant negative versions of EcR: UAS-EcRDN F645A and UAS-EcRDN W650A
(Cherbas et al., 2003). In animals expressing EcRDN constructs in astrocytes
(alrm>EcRDN) we found astrocyte morphology appeared normal at L3 (Figure 2.12), suggesting that EcR signaling is not critical before L3 for normal astrocyte development.
However, astrocytic EcRDN expression completely blocked their morphological
transformation into the highly vacuolated morphology apparent in controls at 6hrs APF
(Figure 2.2B). Suppression of astrocyte morphological transformation was even
accomplished when EcRDN was driven in 1-2 astrocytes rather than the entire population
(Figure 2.2C). Consistent with full suppression of the transformation of astrocytes into
phagocytes, EcRDN expression in astrocytes also blocked the Draper upregulation that normally occurs from 0-6hr APF (Figure 2.2D), as well as induction of phagolysosomal activity (Figure 2.2E).
Since Draper is a potent regulator of phagocytic status of glial cells in Drosophila, we next overexpressed Draper in astrocytes in a EcRDN background to determine whether
Draper activation was sufficient to make astrocytes phagocytic. Despite high level expression of Draper in EcRDN animals, astrocytes failed to transform into phagocytes at
puparium formation (Figure 2.14), indicating that additional factors downstream of EcR
are necessary for activation of the phagocyte program in astrocytes. We attempted to 43
determine the developmental consequences of complete blockade of astrocyte phagocytic
function throughout metamorphosis, however we found that the effects of EcRDN were only transient: despite continued expression of EcRDN in astrocytes, Draper activation was
obvious by 48hrs APF (Figure 2.15). Not all molecular components of the engulfment
machinery are activated in astrocytes at pupariation. For instance, astrocytic dCed-6
levels at L3 are similar to those at 6hrs APF, and we find no obvious change in dCed-6 in
astrocytes expressing EcRDN (Figure 2.12). These results demonstrate that loss of EcR signaling is sufficient to cell autonomously suppress the transformation of astrocytes into phagocytes at pupariation.
Astrocytes engulf synaptic material from the neuropil
Neuronal remodeling during metamorphosis results in the loss of nearly all synapses in the neuropil by 48hrs APF and adult-specific synapses are subsequently generated (A.
Muthukumar and M. Freeman, unpublished observations). We used antibodies to the presynaptic active zone marker Bruchpilot (nc82) to mark synapses and examine their fate during the first 48hrs of metamorphosis. At L3 nc82 labeled synapses throughout the neuropil, by 6hrs APF we found that many nc82+ puncta were located within the astrocytic vacuoles (Figure 2.3A, B, Figure 2.16A).
We next sought to determine whether astrocyte transformation into phagocytes, or other events downstream of EcR in astrocytes, were required for the elimination of CNS synapses. Based on nc82+ staining it appeared that most synapses were eliminated by 6hrs
APF, and were nearly gone by 12-18hrs APF (Figure 2.3D). Impressively, blockade of 44
EcR signaling in astrocytes suppressed the clearance of nc82+ synapses from multiple brain regions at 18hrs APF (Figure 2.3E). Blockade of astrocytic endocytic function during early pupal stages by expression of Shibirets (Shits) resulted in a similar suppression of synaptic clearance from the neuropil (Figure 2.3F). Finally, when we examined the ultrastructure of astrocytes at 6hrs APF we found the large vacuoles filled with cellular debris normally present in control astrocytes (Figure 2.1B) were absent from astrocytes expressing EcRDN (Figure 2.3C). These data argue that some portion of internalized debris in astrocytes at 6hrs APF is synaptic material.
Draper and Crk/Mbc/dCed-12 act in a partially redundant fashion to promote astrocyte transformation into phagocytes and synaptic clearance
The Draper signaling pathway and the Crk/Mbc/dCed-12 complex are essential for glial clearance of axonal debris after axotomy in Drosophila (MacDonald et al., 2006;
Ziegenfuss et al., 2012). Loss of either pathway leads to a near complete blockade of glial engulfment of degenerating olfactory receptor neurons in the adult (Ziegenfuss et al.,
2012). Draper has also previously been shown to play a role in neuronal pruning in
Drosophila (Awasaki et al., 2006), but in which subtype of glia Draper functions to engulf axonal debris remains unknown, and roles for Crk/Mbc/dCed-12 in synaptic, axonal, or dendritic pruning remain unexplored.
We assayed for requirements for Draper in the transformation of astrocytes into phagocytes. In the posterior abdominal region of the VNC at 6hrs APF, drprΔ5 null mutants exhibited similar numbers of astrocytic vacuoles compared to controls, although 45
vacuole size was significantly reduced (by ~50% in drprΔ5 null mutants compared to controls) (Figure 2.4A, B, C). In contrast, knockdown of dCed-12 in astrocytes (alrm> dCed-12RNAi) led to a very strong decrease in astrocytic vacuole number and size: vacuole areas were reduced by 90%, and vacuole numbers were reduced by 70% compared to controls (Figure 2.4A, B, C). Knockdown of Crk in astrocytes led to vacuole areas reduced by ~20% compared to controls (Figure 2.17A, B, C), further supporting a requirement for the Crk/Mbc/dCed-12 complex in the induction of the astrocytic phagocyte program. To determine whether loss of both signaling pathways enhanced the observed changes, we knocked down dCed-12 in a drprΔ5 null background and found a near complete loss of astrocytic vacuole formation (Figure 2.4A, B, C). A similar phenotype was seen when we knocked down Crk in astrocytes in drprΔ5 null background, but to a lesser extent (Figure 2.17A, B, C).
We next assayed for clearance of synaptic structures at 18hrs APF in drprΔ5 mutants and astrocytic dCed-12RNAi animals. We found that drprΔ5 nulls had a significant amount of nc82+ debris in the abdominal region of VNC (Figure 2.4D, 2.18). Astrocytic knockdown of dCed-12 resulted in a significant accumulation of nc82+ debris in both the
SOG and abdominal region of the VNC. Astrocytic knockdown of Crk also gave a similar phenotype but to a lesser effect than dCed-12RNAi (Figure 2.19A, B). When we
depleted both Draper and Crk/Mbc/dCed-12 signaling by astrocytic knockdown of dCed-
12 in drprΔ5 null background, we observed significantly elevated levels of nc82+ debris compared to knockdown of either pathway alone, or controls (Figure 2.4D, 2.18). These data indicate that Draper and Crk/Mbc/dCed-12 act in a partially redundant fashion to 46
transform astrocytes into phagocytes and promote engulfment of synaptic material during
metamorphosis.
EcR, Draper and Crk/Mbc/dCed-12 function in astrocytes to promote MB γ neuron clearance
In draper null mutants pruned MB γ neuron axon clearance is delayed by approximately
2 days, but by adult stages axonal material is ultimately pruned (Awasaki et al., 2006).
Based on our analysis above, astrocytes are ideal candidates for clearing pruned axons in addition to synaptic material since the timing of astrocyte transformation is coincident with MB γ neuron pruning and their processes are close to the MB before clearance
(Figure 2.19). In addition, Crk/Mbc/dCed-12 is a potential new signaling pathway functioning with Draper to clear pruned MB γ neuron axons and could account for the complete (though delayed) clearance in the absence of Draper.
We first assayed MB γ neuron pruning in animals where transformation into phagocytes was blocked by astrocytic EcRDN expression (Figure 2.5A). FasII antibody labels MB α/β and γ neurons which can be easily discriminated by position, morphology, and pruning: at L3 FasII labels the entire MB, but by 18hrs APF MB γ neurons are pruned and FasII staining is only seen in thin medial and dorsal processes, which are the developing adult specific α/β lobes (Awasaki and Ito, 2004; Awasaki et al., 2006)
(Figure 2.5A, Figure 2.21A, B, C). We found that while the dorsal and medial lobes of
MB γ neurons were largely pruned by 18hrs APF in controls, pruning of these axonal 47
branches was strongly inhibited by astrocytic EcRDN expression (Figure 2.5A). Blockade
of astrocytic endocytic function using Shits also resulted in a suppression of MB γ axon clearance, although to a lesser extent than EcRDN (Figure 2.22A, B). To determine if
astrocytes phagocytose pruning MB γ neurons, we examined astrocytic phagolysosomes
at early puparium stage. At 6hrs APF, we found FasII+ MB axonal material located within
the astrocytic phagolysosomes and increased infiltration of the MB lobes by astrocyte
processes (Figure 2.5C, Figure 2.16B). These data indicate that transformation of
astrocytes into phagocytes through EcR signaling is required for proper MB γ neuron
pruning.
We next examined MB γ neuron pruning in controls, draper mutants, dCed-12RNAi
animals, and animals in which both Draper and dCed-12 signaling had been depleted. We
scored pruning phenotypes at multiple time points including 18 and 48hrs APF, and in
newly eclosed adults (Figure 2.5D-G). Consistent with previous observations, in drprΔ5
null mutants, we found a strong suppression of MB γ neuron pruning at 18hrs APF and a
mild phenotype in adults (Figure 2.5D-G). This defect in clearance represented astrocytic
Draper function as expression of draperRNAi in astrocytes resulted in a similar phenotype
(Figure 2.23A, B). Astrocytic knockdown of dCed-12 also resulted in a strong
suppression of MB γ neuron pruning at 18hrs APF and a mild phenotype at 48hrs APF,
although pruning in adults was similar to that in controls (Figure 2.5D-G). Similar results
were found with astrocytic knockdown of Crk or mbc, although the observed phenotypes
were slightly weaker than those observed with dCed-12RNAi (Figure 2.24A, B). These 48
observations identify the Crk/Mbc/dCed-12 complex as a new glial signaling pathway
required for MB axonal pruning.
Interestingly, astrocytic knockdown of dCed-12 in drprΔ5 null mutant background
resulted in an additive phenotype, with dCed-12RNAi, drprΔ5 animals exhibiting a stronger suppression of MB γ axon pruning at 48hrs APF and into adult stages (Figure 2.5D-G).
Similar results were found with astrocytic knockdown of Crk in drprΔ5 null mutant
background (Figure 2.24C, D). We conclude that Draper and Crk/Mbc/dCed-12 act in a
partially redundant fashion during neuronal pruning events.
Loss of Draper and Crk/Mbc/dCed-12 delay clearance of vCrz+ neurons
Astrocyte membranes are present throughout the larval neuropil (except MB lobes as stated previously) at late larval stages, and all appear to transform into phagocytes. This observation suggests that astrocytes have widespread roles in neuronal pruning. To explore this possibility we examined the peptidergic neurons expressing the Corazonin
(Crz) neuropeptide in the VNC (vCrz) that undergo apoptosis during early metamorphosis between 0-6hrs APF, and whose neurites are eliminated (Choi et al.,
2006; Lee et al., 2011).
In controls at L3, α-Crz antibody labels 8 pairs of vCrz neurons and a few neurons in the brain, and by 6hrs APF, almost all of the vCrz+ neuronal debris was
cleared from the neuropil (Choi et al., 2006). We blocked astrocyte phagocytic function
by expressing EcRDN in astrocytes and found many vCrz+ cell bodies and neurite debris
remained at 6hrs APF, and neurite debris lingered until 18hrs APF (Figure 2.6A). We 49
next examined animals depleted for Draper, dCed12, or both, and quantified the
clearance of vCrz+ cell bodies and neurites. We found a striking difference in the
requirements for Draper and dCed12 in the clearance of vCrz+ cell bodies versus neurites.
In drprΔ5 null mutants at 18hrs APF, nearly all vCrz+ neurites were eliminated from the
neuropil while the majority of vCrz+ cell bodies remained (Figure 2.6B, C, D). In
contrast, astrocytic knockdown of dCed12 resulted in a lack of vCrz+ cell bodies, but
neurite debris was significantly retained (Figure 2.6B, C, D). Pan-glial knockdown of
dCed12 also resulted in a lack of vCrz+ cell bodies and a partial suppression of clearance of neurites (Figure 2.25A, B, C). Preservation of vCrz+ neurite debris at 18hrs APF
(rather than complete loss of Crz+ signal) indicates that the Crz peptide is stable for at
least 18hrs APF and a useful marker for unengulfed neurite debris.
Glial clearance of cell bodies and neurites appears to be a separable and additive
function for Draper and dCed-12. Elimination of both pathways did not increase the
number of cell bodies or amount of neurite debris when compared to elimination of each
pathway alone (Figure 2.6B, C, D). In addition, these pathways appear to be acting in
different subsets of glial cells. Draper appears to function in non-astrocyte glia (e.g. ensheathing or cortex glia) to clear vCrz+ cell bodies since astrocytic knockdown of
Draper had no effect on cell body clearance (Figure 2.6B, C, D) and non-astrocyte glial
depletion of draper led to similar levels of remaining vCrz+ cell bodies compared to depletion from all glia (Figure 2.26A, B). Clearance is likely not mediated by neurons themselves since vCrz+ neurons did not express Draper at 6hrs APF (Figure 2.27) and 50
vCrz+ neuronal depletion of draper had no effect on neurite or cell body clearance
(Figure 2.28A, B, C).
Engulfing cells such as microglia can promote apoptotic death of target cells, but this has generally been examined before neurite elaboration has occurred (Cunningham et al., 2013; Marín-Teva et al., 2004). Can engulfing activity of glia promote neurite destruction? To answer this question we labeled vCrz+ neuronal membranes with membrane-tethered GFP (Crz-Gal4; UAS-mCD8::GFP) and assayed clearance in controls and draper mutants. By 6hrs APF the majority of cell bodies and neurites were cleared in controls (Figure 2.7A-E). However, removing only a single copy of draper
(drprΔ5/+) resulted in dominant suppression of clearance of both vCrz+ cell bodies and neurites at 6hrs APF (Figure 2.7A-E), which was enhanced in drprΔ5 homozygous
mutants (Figure 2.7A-E). This observation argues the fragmentation of vCrz+ neurons is highly sensitive to draper gene dosage and suggests engulfing glia can actively promote the destruction of targets during clearance.
Discussion
Widespread cell death and pruning occur in developing nervous systems of complex
metazoans. Roles for astrocytes in the clearance of neuronal debris remain undefined. In
this study we show that Drosophila larval astrocytes transform from supportive cells into
phagocytes that engulf significant amounts of pruned synapses and neural debris.
Astrocytic engulfment of pruned synapses and MB γ neural debris requires the Draper 51
signaling pathway, which acts in a partially redundant fashion with the Crk/Mbc/dCed-12
complex, a pathway we implicate for the first time in neural pruning events. Detailed
genetic analysis of these pathways revealed unexpected, context-dependent usage of
these signaling pathways to clear subcompartments of neurons during nervous system
reorganization.
Drosophila larval astrocytes transform into phagocytes at the initiation of
metamorphosis
Astrocytes serve as regulators of synapse formation and function and are generally
supportive of neural circuits. Based on several lines of evidence we show astrocytes
transform morphologically and functionally into phagocytes at pupariation and engulf
significant amounts of neural debris. Phagocytic astrocytes take on a highly vacuolated
appearance, up-regulate the engulfment molecule Draper, contain cytoplasmic vacuoles
filled with debris that stains for synaptic or axonal markers, and exhibit high levels of
lysosomal activity. Activation of this phagocytic program depends upon cell-autonomous
signaling through the EcR receptor since blockade of EcR in even single astrocytes
suppressed their transformation by all morphological and molecular criteria. Blockade of
astrocyte phagocytic function by multiple methods (e.g. EcRDN or Shits) suppressed the clearance of synapses throughout the CNS, axons of MB γ neurons in the central brain and the neurites of vCrz+ cells in the VNC. This latter observation defines vCrz+ neurons
as a new system to explore astrocyte-neuron interactions during neuronal apoptosis and
neurite or synapse elimination in the CNS. 52
While roles for Drosophila glia in the engulfment of pruned MB γ axons have been previously described (Awasaki and Ito, 2004; Awasaki et al., 2006, 2011; Watts et al., 2004), it was unexpected that astrocytes would be the phagocytic cell type. In the adult Drosophila brain a second type of neuropil glial cells, ensheathing glia, are responsible for engulfing injured axonal debris while astrocytes fail to respond in a detectable way morphologically or molecularly (e.g. by Draper upregulation) (Doherty et al., 2009). The stark difference in glial subtypes executing phagocytic function in the pupa versus the adult could result from differences in glial genetic programs during development versus in mature glia, or it could indicate a key difference in the molecular nature of pruned neurites compared to those undergoing injury-induced axon degeneration.
The extent to which mammalian astrocytes are phagocytes in vivo has remained unclear until recently. In culture, astrocytes can be phagocytic (Roldán et al., 1997;
Tansey and Cammer, 1998) and engulf apoptotic cells (Chang et al., 2000) or amyloid-ß
(Wyss-Coray et al., 2003). In vivo the postlaminar optic nerve head myelination transition zone (MTZ) astrocytes express Mac-2, a molecule implicated in phagocytic activity, and internalize axonal evulsions, and this event appears increased in frequency in glaucoma models (Nguyen et al., 2011). Most impressively, Chung et al. (2013) recently directly demonstrated that astrocytes engulf synaptic material in vivo, and that this event is mediated by MEGF10 (mouse Draper) and the engulfment receptor MERTK. Thus
Draper/MEGF10-dependent engulfment of pruned synapses appears to be an ancient astrocytic mechanism, whether the same is true for the Crk/Mbc/dCed-12 pathway 53
remains to be determined. Transcriptome data from purified mouse forebrain astrocytes
support this notion since molecular pathways for engulfment are highly enriched,
including in addition to MEGF10, Gulp1 (dCed-6) and Crk, Dock1 (Mbc) and Elmo
(dCed-12) (Cahoy et al., 2008). It will be important to explore whether these additional
pathways have similar roles in astrocytes for the pruning of mammalian neural circuits
during development, and whether astrocyte engulfment activity is modified in
neurological diseases involving axonal, dendritic, or synaptic loss.
Activation of astrocyte phagocytic function requires Draper and the Crk/Mbc/dCed-12
signaling pathway
We have examined two types of neurons that undergo different types of developmental
reorganization. MB γ neurons prune their medial and dorsal axon branches and dendrites,
their cell bodies remain viable, and at midpupal stages they re-extend medial axon branches to establish adult-specific connectivity (Lee et al., 2000; Watts et al., 2004). In contrast, vCrz+ neurons exhibit complete neurite degeneration and cell bodies undergo
apoptotic death and eventually are completely eliminated (Choi et al., 2006). We find that
in addition to there being remarkable differences in the patterns of fragmentation
exhibited by these subsets of neurons, there are also critical differences in the engulfment
signaling pathways used to promote their initial destruction and clearance.
Previous work has revealed a role for Draper in MB γ neuron pruning, with
draper null mutants exhibiting a delay of approximately 2 days in the clearance of pruned
MB γ neuron axonal debris (Awasaki et al., 2006). We provide evidence supporting a key 54
role for the Crk/Mbc/dCed-12 complex in the clearance of MB γ neuron axonal debris,
and demonstrate that Crk/Mbc/dCed-12 knockdown and draper mutants have additive
phenotypes, indicating these signaling pathways act in a partially redundant fashion in
astrocytes to promote clearance of pruned MB γ neuron axons. While these phenotypes are additive, we note that draper mutants exhibit a stronger delay in MB γ neuron pruning, suggesting that Draper signaling plays a more prominent role in this brain region than signaling through the Crk/Mbc/dCed-12 complex. However we cannot exclude the possibility that a Crk, mbc, or dCed-12 mutant might have a stronger phenotype than the
RNAi lines used in this study.
vCrz+ neuronal clearance appears to involve both Draper signaling and the
Crk/Mbc/dCed-12 complex, with the former primarily promoting clearance of vCrz+ cell
bodies, and the latter driving clearance of degenerating vCrz+ neurites. In draper null mutants at 18hrs APF the majority of neuronal cell bodies remained at the edge of the neuropil while neurite debris was largely cleared. Reciprocally, astrocytic knockdown of dCed-12 suppressed neurite clearance, while vCrz+ cell bodies were promptly eliminated.
The lack of additivity of the phenotypes for either neurite or cell body clearance in
draper null mutants with astrocytic dCed-12RNAi suggests that elimination of cell bodies is
primarily driven by the Draper pathway, while neurite clearance is largely accomplished
by signaling through Crk/Mbc/dCed-12 complex.
Our work reveals that the genetic pathways engaged by glia to engulf pruned
neuronal material versus apoptotic neurons are context dependent, and correlate with the
type of destructive program. Interestingly, the molecular pathways that mediate axonal 55
degeneration during axon pruning versus apoptosis are also distinct in mammalian
cultured neurons. Local withdrawal of NGF in the axon activates Caspase 6-dependent
axon degeneration, which is not sensitive to XIAP1 inhibition. In contrast, whole cell
NGF withdrawal leads to Caspase 6-independent apoptotic cell death and degeneration of
axons, which is then sensitive to XIAP1 (Cusack et al., 2013). These observations argue
that context matters when neurites and cell bodies are being destroyed. We speculate that
neuron-glia signaling during engulfment events might also be compartmentalized, with
neurites and cell bodies generating different types of “eat me” cues for clearance by glia.
Do engulfing glial cells actively promote the destruction of target neurons? We
made the intriguing observation that loss of a single copy of draper is sufficient to
dominantly suppress the elimination of vCrz+ cell bodies and neurites at 6hrs APF.
Moreover loss of two copies results in the retention of nearly all vCrz+ cell bodies and
significant parts of the vCrz+ scaffold, and many regions of the scaffold appeared morphologically intact, suggesting delay of neurite fragmentation. Previous work has shown that expression of the anti-apoptotic molecule P35 is sufficient to suppress the pruning of vCrz+ neurites for at least 1 day (Choi et al., 2006); we have extended this
observation and found this to be true for at least 2 days (O.E.T-Y. and M.R.F, data not
shown). Thus blocking apoptosis in vCrz+ is sufficient to significantly delay neurite degeneration. This observation, coupled with the known role for CED-1 in actively promoting the death of engulfment targets (Hoeppner et al., 2001; Reddien et al., 2001), suggests that engulfing non-astrocyte glia may promote neuronal apoptosis through
Draper signaling. This in turn would promote degeneration of the neurite scaffold after 56
cell body death. If this underlies the delay we observe in neurite degeneration in draper
mutants it would argue that glia actively sculpt neural networks by promoting the
destruction of selected subsets of neurons. Subsequent work exploring the survival of cell
bodies preserved in draper nulls will be essential to explore this exciting possibility.
Materials and Methods
Drosophila Strains
alrm-Gal4 (Doherty et al., 2009), pUAST-mCD8::GFP (Lee and Luo, 2001), draper∆5
(MacDonald et al., 2006), UAS-shibirets, w1118, UAS-EcR.B1-ΔC655.F645A and UAS-
EcR.B1-ΔC655.W650A (Cherbas et al., 2003), UAS-CrkRNAi (VDRC transformant 19061),
UAS-dCed12RNAi (VDRC transformant 107590), UAS-mbcRNAi (VDRC transformant
10644), Crz-Gal4 (Choi et al., 2006), repo-flp6-2 (a gift from T. Stork), pWiz-drprRNAi7b
(MacDonald et al., 2006), repo-Gal4 (Sepp et al., 2001), tub-Gal80, UAS-draperI (Logan et al., 2012). The alrm-Gal80 construct was generated by amplifying a 5kb region of the alrm promoter using the following primers: 5’-
GATCAACCGCGGACTACGCACAGATG-TGGTCATCTGAATAGGTTTC-3’ and
5’-AAATTTGTCTAGATAGTGGCGATCC-TTTCGCTCGGGAGCC-3’. The resulting
fragment was cloned into the Cas40-GAL80 vector and transgenic flies were generated
using standard methods by BestGene, Inc (Chino Hills, CA).
Immunohistochemistry 57
Brains were dissected in 1X PBS, fixed in PTX (1X PBS/0.1% Triton X-100)/4% formaldehyde for 20min at room temperature, washed in PTX and incubated overnight at
4°C with primary antibodies in PTX. Next day brains were washed in PTX, secondary antibodies were applied for 3 hours at room temperature, brains were washed in PTX at room temperature and mounted in Vectashield antifade reagent (Vector Labs).
Antibodies were used at the following dilutions: 1:500 anti-Draper (Freeman et al.,
2003); 1:5 mouse anti-Repo (Developmental Studies Hybridoma Bank); 1:200 mouse anti-GFP (Invitrogen); 1:500 rabbit anti-GFP (Invitrogen); 1:25 mouse anti-nc82
(Developmental Studies Hybridoma Bank); 1:5 mouse anti-EcRB1 and anti-EcRA
(Developmental Studies Hybridoma Bank); 1:5 mouse anti-FasII (Developmental Studies
Hybridoma Bank); 1:500 rabbit anti-Crz (Choi et al., 2006); 1:500 rat anti-dCed-6 (gift from T. Awasaki, Janelia Farm, Ashburn, VA). Anti IgG secondary antibodies were
FITC, Cy3 or Cy5 conjugated and used at 1:200 and Cy3 conjugated anti-HRP was used at 1:300 dilution (all Jackson Immunoresearch).
Shibirets temperature shift experiments
UAS-shibirets flies were crossed to alrm-Gal4. Flies were raised at 18°C, shifted to 30°C
at 0hr APF, and maintained at 30°C until the brains were dissected for
immunohistochemistry.
Lysotracker stains
Brains were dissected in 1XPBS and incubated with LysoTracker Red DND-99 58
(Invitrogen) at a dilution of 1:5000 in 1XPBS for 15 min in the dark. Brains were then washed 2 times for a total of 10min with PBS, fixed for 10 min with PTX
(1XPBS/0.1%Triton-X100)/4% formaldehyde and washed 5 times for a total of 30min with PTX in the dark. Brains were mounted in Vectashield antifade reagent (Vector Labs) and imaged immediately.
Mosaic analysis with repressible cell marker clone production
Genotypes of flies for MARCM clones were: (1) for wild-type astrocyte clones: alrm-
Gal4, UAS-mCD8::GFP, repo-flp6-2/+ ; FRT2A/FRT2A, tub-Gal80 (2) for EcRDN expressing astrocyte clones: alrm-Gal4, UAS-mCD8::GFP, repo-flp6-2/UAS-EcRDN F645A ;
FRT2A/FRT2A, tub-Gal80, respectively.
Transmission Electron Microscopy
2 brains for each genotype were prepared and sectioned for electron microscopy. Brains
were quickly dissected in the fixation buffer (2.5% gluteraldehyde in 0.1 M Sodium
Cacodylate buffer pH 7.2) and incubated in fixation buffer for an hour at 4ºC. Then
fixation buffer was changed with fresh buffer and the brains were incubated overnight at
4ºC. Samples were then processed and analyzed at the University of Massachusetts
Medical School Electron Microscopy core facility according to standard procedures.
Briefly, fixed samples were moved into fresh 2.5% gluteraldehyde in 0.1 M sodium
cacodylate buffer and left overnight at 4oC. The samples were then rinsed twice in
fixation buffer and post-fixed with 1% osmium tetroxide for 1h at room temperature. 59
Samples were then washed twice with dH2O for 5 minutes and dehydrated through a graded ethanol series of 20% increments, before two changes in 100% ethanol. Samples were then infiltrated first with two changes of 100% propylene oxide and then with a
50%/50% propylene oxide / SPI-Pon 812 resin mixture. The following day three changes of fresh 100% SPI-Pon 812 resin were done before the samples were polymerized at 68oC in plastic capsules. The samples were then reoriented and thin cross sections were taken at abdominal regions of the VNC of the brains. The sections were placed on copper support grids, and contrasted with lead citrate and uranyl acetate. Sections were examined using the FEI Tecani 12 BT with 80Kv accelerating voltage, and images were captured using a Gatan TEM CCD camera.
Confocal imaging and quantification
In experiments where nc82 staining was quantified, samples were fixed, stained and imaged side by side with the same settings. Quantification of nc82 intensity was performed using ImageJ software on the middle Z section of SOG, thorax and abdominal regions. For SOG, one measurement from each segment (total 2 measurements per brain), for thorax (T1 and T2), and for the abdominal region, 2 measurements per hemisegment
(total 4 measurements per brain) were taken. Identical areas of measurement were used in each experiment. Background was measured from the cortex region using the same area,and subtracted from each measured value. Mean values were plotted in all figures.
Astrocyte phagolysosome formation was quantified at the most posterior region of the abdominal segments of the VNC. 20 confocal stacks were taken over 9.5 µm 60
thickness in the middle of the VNC. The dimensions of the region quantified were 73 x
73 x 9.5 µm. A line was drawn through the middle of the neuropil along the
anteroposterior axis in abdominal segments of each side of VNC and the total number of
vacuoles that made contact with the line was measured and their area calculated. The
LSM 5 Pascal (Zeiss) software was used for the calculation of the area.
MB γ neuron pruning phenotypes were quantified by categorizing them according
to severity. For 18hrs APF images, “none” had no extra debris; “mild” had small amounts
of debris around the tip of the dorsal lobe; “intermediate” had debris at the tips of the
dorsal and medial lobes, and sometimes the edge of the dorsal branch; and “strong” had
nearly intact dorsal and medial lobes. For 48hrs APF and newly-eclosed adults, three
regions had extra debris: either at the tip of the dorsal or medial lobes, or at the heel of
the MB. “None” had no extra debris; “mild” had extra debris in one region;
“intermediate” had extra debris at two regions; and “strong” had extra debris at all three
regions of the MB.
Quantification of vCrz+ debris was performed using Slidebook (Intelligent
Imaging Innovations) software. Brains were fixed and stained side-by-side, and imaged the same day and with identical microscope settings. Background was corrected by creating a threshold for minimal signal intensity, and regions of quantification were limited by creating a second mask limited to the neuropil area. Identical parameters were used for all samples in each experiment. Sum intensity has been plotted in all graphs.
For quantification of neurites in experiments with Crz-Gal4, UAS-mCD8::GFP, neurites were divided into segments (medial, lateral, or horizontal) and each segment was 61
examined for continuity. If the segment was continuous, it was counted as 1, and if not
continuous, counted as 0. Lateral and medial portions of antero-posterior connectives, and horizontal portions of the scaffold were quantified.
Statistical analysis
GraphPad Prism6 was used to perform one-way ANOVA followed by Tukey post hoc tests.
62
Figure 2.1: Astrocytes transform into phagocytes at the initiation of neural circuit remodeling (A) Astrocytes in controls were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP, green) in (A), (C), (D), neuropil with HRP-Cy3 (red) and glial nuclei with anti-Repo (blue). Time points are as indicated (APF). Thoracic and brain regions are shown. Inset, high mag view of boxed region. Single z-sections are shown for (A), (C), and (D). Scale bars, 20 µm. Inset, 10 µm. (B) TEM image from a cross-section of the abdominal VNC of a control animal. An astrocyte (outlined yellow) at the edge of the neuropil, exhibiting cytoplasmic vacuoles (magenta). Genotype: alrm-Gal4, UAS-mCD8::GFP/+. Scale bar, 2 µm. (C) Draper was labeled in red. Time points as indicated (APF). Insets, high mag view of boxed regions. Genotype: alrm-Gal4, UAS-mCD8::GFP. Scale bars, 20 µm. Inset, 10 µm. (D) Lysotracker was labeled in red. Note that all Lysotracker+ puncta in the neuropil were surrounded by GFP+ membranes of astrocytes. Inset, high mag view of vacuole filled with Lysotracker. Genotype: alrm-Gal4, UAS-mCD8::GFP. Scale bars, 20 µm. Inset, 5 µm.
63
Figure 2.2: EcR signaling cell-autonomously regulates transformation of astrocytes into phagocytes (A) Astrocytes in controls were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green) in (A-E), EcR-B1 (blue) and glial nuclei with anti-Repo (red). Insets, high mag view of an astrocyte cell body. All confocal images (A-E) are single z confocal slices. Scale bars, 20 µm. Inset, 10 µm. (B) The neuropil was labeled with HRP-Cy3 (red) and in (B), (D) glial nuclei with anti-Repo (blue). Time points are as indicated (APF). Note lack of transformation of astrocytes at 6hrs APF. In (B), (D) and (E), genotypes: control, alrm-Gal4,UAS-mCD8::GFP/+ and alrm>EcRDN is alrm-Gal4,UAS-mCD8::GFP/UAS- EcRDN. Scale bars, 20 µm. (C) Astrocyte MARCM clones at L3 and 6hrs in controls (alrm-Gal4,UAS-mCD8::GFP, repoflp6-2;FRT2A/FRT2A,tub-Gal80), or astrocytes expressing EcRDN (alrm-Gal4,UAS-mCD8::GFP, repoflp6-2/UAS-EcRDN;FRT2A/FRT2A,tub- Gal80) at 6hrs APF. Scale bars, 20 µm. (D) Draper was labeled in red. Time points are as indicated (APF). Note strong knockdown of Draper in astrocytes in alrm>EcRDN animals compared to controls at 6hrs APF. Scale bars, 20 µm. (E) Lysotracker was labeled in red. Time points are as indicated (APF). Lysotracker+ puncta in astrocytes was blocked at 6hrs APF in alrm>EcRDN animals. Scale bars, 20 µm.
64
Figure 2.3: Astrocytes engulf and clear synaptic material from the neuropil (A) In (A) and (D), confocal images are single z confocal slices. Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green), glial nuclei with anti-Repo (blue), in (A), (D) active zones with nc82 (antibody for Brp, red). Time points: L3, third instar larva; APF, hrs after puparium formation. Top inset, high mag view of boxed region. Insets below are other examples of vacuoles. Scale bars, 20 µm. Inset, 1 µm. (B) Graph shows fluorescence intensity per pixel in each channel for inset in (A) of the astrocyte membrane and nc82+ puncta along line drawn through a vacuole. (C) TEM image from a cross-section of the abdominal VNC of an alrm>EcRDN animal. An astrocyte (outlined yellow) at the edge of the neuropil. Time point is as indicated (APF). Genotype: alrm>EcRDN is alrm-Gal4,UAS-mCD8::GFP/UAS-EcRDN. Scale bars, 2 µm. (D) Time points are as indicated (APF). Genotypes: control, alrm-Gal4/+ and alrm>EcRDN is alrm- Gal4/UAS-EcRDN. Scale bars, 20 µm. (E-F) Quantification of pixel intensity of nc82+ puncta at L3 and 18hrs APF in SOG, thorax and abdominal regions of VNC. (E) Genotypes: EcRDN/+ is UAS-EcRDN/+ and alrm>EcRDN is alrm-Gal4/UAS-EcRDN. Error bars represent ± s.e.m. N values: alrm-Gal4/+ N ≥ 20, UAS-EcRDN/+ N ≥ 16, alrm- Gal4/UAS-EcRDN N ≥ 20 (N denotes number of measurements), *P< 0.05, **P< 0.01, ****P< 0.0001. (F) Genotypes: shits/+ is UAS-shits/+ and alrm> shits is alrm-Gal4/UAS- shits. Flies were raised at 18°C, then shifted to the restrictive temperature (30°C) at 0hr APF for 18hr. Error bars represent ± s.e.m.; N≥12, ****P< 0.0001. 65
Figure 2.4: Draper and Crk/Mbc/dCed-12 pathways are required for the formation of astrocytic vacuoles (A) Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green). Images zoomed at abdominal tip of the VNC at 6hrs APF. Insets show one representative vacuole. Images are single z confocal slices. Scale bars, 20 µm. Inset, 2 µm. Genotypes: control = alrm-Gal4,UAS-mCD8::GFP/+ 5 5 drprΔ /+ = alrm-Gal4,UAS-mCD8::GFP/+; drprΔ /+ 5 5 drprΔ = alrm-Gal4,UAS-mCD8::GFP/+; drprΔ alrm>dCed-12RNAi = alrm-Gal4, UAS-mCD8::GFP/UAS-dCed-12RNAi RNAi 5 RNAi 5 alrm>dCed-12 + drprΔ = alrm-Gal4, UAS-mCD8::GFP/UAS-dCed-12 /+; drprΔ (B) Quantification of number of vacuoles of figures in (A). Error bars represent ± s.e.m.; number of brains ≥ 5, two hemisegments measured in each brain, ***P< 0.001, ****P< 0.0001. (C) Quantification of cross-section area of vacuoles of figures in (A). Error bars represent ± s.e.m.; number of brains ≥ 5, two hemisegments measured in each brain, *P< 0.05, **P< 0.01, ****P< 0.0001. (D) Quantification of pixel intensity of nc82+ puncta in SOG, thorax and abdominal regions of VNC at L3 and 18hrs APF from Figure 2.18. Genotypes: alrm>dCed-12RNAi = alrm-Gal4/UAS-Ced-12RNAi RNAi 5 RNAi 5 alrm>dCed-12 + drprΔ = alrm-Gal4/UAS-dCed-12 ; drprΔ Error bars represent ± s.e.m.; N values for alrm-Gal4/+ N≥18, alrm>dCed-12RNAi N≥20, 5 RNAi 5 alrm-Gal4/+; drprΔ N≥ 16, alrm>dCed-12 + drprΔ N≥18, *P< 0.05, **P< 0.01,***P< 0.001, ****P< 0.0001.
66
Figure 2.5: EcR, Draper and Crk/Mbc/dCed-12 function in astrocytes to promote MB γ neuron clearance (A) Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green) in (A), (C), glial nuclei with anti-Repo (blue), MB lobes with anti-FasII (red) in (A), (C), (D). The adult specific dorsal lobe is outlined (white line). Time point is as indicated (APF). Genotypes: wt, alrm-Gal4,UAS-mCD8::GFP/+ and alrm>EcRDN, alrm-Gal4,UAS- mCD8::GFP/UAS-EcRDN. All confocal images (A-D) are z-projections. Scale bars, 10 µm. (B) Quantification of MB γ neuron pruning phenotype from (A) by using the categories shown in Figure 2.21 (see Materials and Methods). wt, N=24, alrm>EcRDN, N=20 hemisegments quantified. (C) Image of magnified dorsal MB lobe. Note FasII+ debris inside astrocytic vacuoles. Time point is as indicated (APF). Genotype: alrm-Gal4, UAS-mCD8::GFP/+. Scale bars, 10 µm. (D) MB debris was scored at the time points indicated (APF). Arrows indicate extra MB γ debris. Scale bars, 10 µm. Genotypes: wt = alrm-Gal4/+; alrm>dCed-12RNAi = alrm-Gal4/UAS-dCed-12RNAi; drpr∆5 = alrm-Gal4/+; drpr∆5; alrm>dCed-12RNAi+ drpr∆5 = alrm-Gal4/UAS-dCed-12RNAi; drpr∆5 (E-G) Quantification of MB γ neuron pruning phenotype from (D). N values: wt, N=36; alrm>dCed-12RNAi, N=22; 67
drpr∆5, N=32; alrm>dCed-12RNAi+ drpr∆5, N=23 hemisegments quantified at 18hrs APF. wt, N=58; alrm>dCed-12RNAi, N=30; drpr∆5, N=26; alrm>dCed-12RNAi+ drpr∆5, N=22 hemisegments quantified at 48hrs APF. N values: wt, N=38; alrm>dCed-12RNAi, N=20; drpr∆5, N=18; alrm>dCed-12RNAi+ drpr∆5, N=32 hemisegments quantified at newly- eclosed adults.
68
Figure 2.6: EcR, Draper and Crk/Mbc/dCed-12 function in astrocytes to promote clearance of vCrz+ neuronal debris. (A) vCrz+ neurons were labeled with anti-Crz (green) in (A-B). Time points: L3, third instar larva. Genotypes: control is alrm-Gal4, UAS-mCD8::GFP/+ and alrm>EcRDN is alrm-Gal4, UAS-mCD8::GFP/UAS-EcRDN. Confocal images in (A-B) are z-projection images. Scale bars, 20 µm. (B) Time points as indicated in (A). Scale bars, 20 µm. Genotypes: control = alrm-Gal4/+ alrm>drprRNAi = alrm-Gal4/drprRNAi drpr∆5 = alrm-Gal4/+; drpr∆5 alrm>dCed-12RNAi = alrm-Gal4/dCed-12RNAi alrm>dCed-12RNAi + drpr∆5 = alrm-Gal4/dCed-12RNAi; drpr∆5 (C) Quantification of pixel intensity of vCrz+ debris from (B). N values for (C-D): alrm- Gal4/+, N=32; alrm-Gal4/drprRNAi, N=11; alrm-Gal4/+; drpr∆5, N=11; alrm-Gal4/dCed- 12RNAi, N=14; alrm-Gal4/dCed-12RNAi; drpr∆5, N=12 brains quantified. Error bars represent ± s.e.m. ****P< 0.0001. (D) Quantification of the percentage of remaining vCrz+ cell bodies from (B). Error bars represent ± s.e.m. ****P< 0.0001.
69
Figure 2.7: Loss of Draper delays neurite degeneration and clearance of vCrz+ neuronal cell bodies. (A) vCrz+ neurons were labeled with GFP (Crz-Gal4, UAS-mCD8::GFP, green). Time points are as indicated (APF). Scale bars, 20 µm. Genotypes: control = Crz-Gal4, UAS-mCD8::GFP; + 5 5 drprΔ /+ = Crz-Gal4, UAS-mCD8::GFP; drprΔ /+ 5 5 drprΔ = Crz-Gal4, UAS-mCD8::GFP; drprΔ (B-D) Quantification of % intact lateral (B), medial (C), and horizontal (D) axons of + 5 vCrz neurons per hemisegment from (A). For (B-E) control, N=30; drprΔ /+, N=30 and drpr∆5, N=32 hemisegments quantified. Error bars represent ± s.e.m. *P< 0.05, ***P< 0.001, ****P< 0.0001 (E) Quantification of % cell bodies of vCrz+ neurons per hemisegment from (A).
70
Figure 2.8: Astrocytes’ position relative to the neuropil in the CNS. Astrocyte cell bodies reside at the edge of the neuropil and their processes infiltrate the neuropil. (A) Astrocytes were labeled with GFP (green), neuropil with nc82 (red) and glial nuclei with anti-Repo (blue). Time point: L3, third instar larva. Image is a single z confocal slice showing x,y, x,z and y,z planes of the posterior abdominal region of the VNC. Genotype: alrm>GFP is alrm-Gal4, UAS-mCD8::GFP/+. Scale bar, 20 µm. (B) Astrocytes and glial nuclei were labeled as in (A), neuropil with HRP-Cy3 (red). Image is a single z confocal slice of the VNC. Time point: L3, third instar larva. Arrow shows an astrocyte MARCM clone. Tracing indicates the edge of the neuropil. Genotype: alrm-Gal4, UAS-mCD8::GFP, repoflp6-2; FRT2A/FRT2A, tub-Gal80. Scale bar, 20 µm.
71
Figure 2.9: Ultrastructural analysis of control animal brains at L3. TEM image from a cross-section of abdominal VNC showing an astrocyte at the edge of the neuropil traced in yellow. Genotype is alrm-Gal4, UAS-mCD8::GFP/+. Scale bar, 2 µm.
72
Figure 2.10: Draper immunoreactivity does not overlap with glial cells other than astrocytes in the neuropil at 6hrs APF. Membranes of glial cells except astrocytes were labeled with GFP (UAS- mCD8::GFP/+;repo-Gal4, alrm-Gal80/+, green), glial nuclei with anti-Repo (blue), anti- Draper (red). Time point is as indicated (APF). Image is a single z confocal slice. Note Draper immunoreactivity does not co-localize with non-astrocyte glial subtypes in the neuropil. Scale bar, 20 µm.
73
Figure 2.11: Astrocytic knockdown of Draper almost completely blocks Draper upregulation in astrocytes at 3hrs APF. Astrocytes were labeled with GFP (green), glial nuclei with anti-Repo (blue), anti-Draper (red). Time point is as indicated (APF). Image is a single z confocal slice showing SOG and part of thoracic regions of the VNC. Note Draper immunoreactivity in astrocyte cell bodies and membranes was almost completely gone whereas Draper immunoreactivity in the cortex remains. Genotype: alrm>drprRNAi is alrm-Gal4, UAS-mCD8::GFP/UAS- drprRNAi. Scale bar, 20 µm.
74
Figure 2.12: dCed-6 is not activated at pupariation, nor is it modulated by EcR signaling in astrocytes. Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green) and anti- dCed-6 in red. Time points are as indicated (APF). The antibody stain and imaging are done on the same day and under the same conditions. The images are single z confocal slices. The levels of dCed-6 is similar between control and alrm>EcRDN at L3. At 6hr APF, the levels of dCed-6 in control is similar to the levels in control and alrm>EcRDN at L3. Insets, high magnification view of boxed regions. Genotypes: control is alrm- Gal4,UAS-mCD8::GFP/+ and alrm>EcRDN is alrm-Gal4,UAS-mCD8::GFP/EcRDN. Scale bars, 20 µm. Insets, 10 µm.
75
Figure 2.13: Astrocytes do not express EcRA isoform at pupariation. Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP, green) and anti-EcRA in red. Time point is as indicated (APF). Inset, high magnification view of boxed region. The astrocyte cell body is devoid of EcRA immunostain. Genotype: control is alrm- Gal4,UAS-mCD8::GFP. Scale bars, 20 µm. Insets, 10 µm.
76
Figure 2.14: Astrocytic Draper overexpression does not rescue the EcRDN blockade of astrocyte morphological transformation. Astrocytes were labeled with GFP (green), glial nuclei with anti-Repo (blue), anti-Draper (red). Time point is as indicated (APF). Image is a single z confocal slice showing SOG and part of thoracic regions of the VNC. Note Draper immunoreactivity in astrocytes was significantly higher compared to alrm>EcRDN in Figure 2.2D indicating overexpression of Draper in astrocytes. Genotype: alrm>EcRDN + drprI is UAS-EcRDN/UAS-drprI; alrm- Gal4, UAS-mCD8::GFP/+. Scale bar, 20 µm.
77
Figure 2.15: EcRDN blockade of Draper expression in astrocytes is not efficient at later stages of metamorphosis. Astrocytes were labeled with GFP (green), glial nuclei with anti-Repo (blue), anti-Draper (red). Time point is as indicated (APF). Image is a single z confocal slice showing brain and SOG regions. Note presence of Draper immunoreactivity in astrocyte cell bodies (one shown by the arrow) as well as membranes. Genotype: alrm>EcRDN is alrm-Gal4, UAS-mCD8::GFP/UAS-EcRDN. Scale bar, 20 µm.
78
Figure 2.16: Astrocytic phagolysosomes engulf nc82+ and FasII+ material in the neuropil. (A) Astrocytes were labeled with GFP (green), presynaptic active zones with nc82 (red) and glial nuclei with anti-Repo (blue). Time point is as indicated (APF) in (A) and (B). Images are single z confocal slices zoomed in the neuropil of VNC. Two examples of astrocytic phagolysosomes having nc82+ puncta inside them are shown in x,y, x,z and y,z planes. Genotype in (A) and (B): alrm>GFP is alrm-Gal4, UAS-mCD8::GFP/+. Scale bar, 10 µm. (B) Astrocytes were labeled as in (A), and MB lobes with anti-FasII (red). Image is a single z confocal slice and high mag view of dorsal MB lobe in figure 2.5C. Arrow points to one of the astrocyte phagolysosomes engulfing FasII+ material in x,y, x,z and y,z planes. Scale bar, 10 µm.
79
Figure 2.17: Crk is required for the proper formation of vacuoles in astrocytes. (A) Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+, green). Images are high mag view of the very posterior of the abdominal region of the VNC neuropil. The images are single z confocal slices. Time point is as indicated (APF). Scale bars, 20 µm. Genotypes in (A-C): control = alrm-Gal4,UAS-mCD8::GFP/+ drpr∆5/+ = alrm-Gal4,UAS-mCD8::GFP/+; drpr∆5/+ drpr∆5 = alrm-Gal4,UAS-mCD8::GFP/+; drpr∆5 alrm>CrkRNAi = alrm-Gal4,UAS-mCD8::GFP/CrkRNAi alrm>CrkRNAi + drpr∆5 = alrm-Gal4,UAS-mCD8::GFP/CrkRNAi; drpr∆5 (B) Quantification of the cross-section area of the vacuoles in (A). In (B) and (C), error bars represent mean ± s.e.m, *P< 0.05, **P<0.01, ***P<0.001, ****P< 0.0001, number of brains quantified ≥ 5, two hemisegments quantified in each brain. (C) Quantification of the number of the vacuole-like structures as in (A).
80
Figure 2.18: Draper and Crk/Mbc/dCed-12 function in astrocytes to promote clearance of synaptic material from the neuropil. Images show abdominal segments of VNC of brains stained for nc82 (red). Time point is as indicated (APF). Genotypes: alrm>dCed-12RNAi = alrm-Gal4/dCed-12RNAi alrm>dCed-12RNAi + drpr∆5 = alrm-Gal4/dCed-12RNAi; drpr∆5 Scale bar, 20 µm.
81
Figure 2.19: Crk is required in astrocytes for proper clearance of synaptic material in the neuropil. (A) Images show abdominal segments of VNC of brains stained for nc82 (red). Time point is as indicated (APF). Genotype: alrm>CrkRNAi is alrm-Gal4/CrkRNAi. Scale bars, 20 µm. (B) Quantification of pixel intensity of nc82+ puncta of the brains shown in (A). The values of 18hr APF are normalized to the values of the same genotype at L3. Error bars represent mean ± s.e.m. ****P<0.0001. N≥28 measurements.
82
Figure 2.20: Compared to other regions of the neuropil, MB lobes are infiltrated less by astrocytes. Astrocytes were labeled with GFP (green) and MB lobes with anti-FasII (red). Time point: L3, third instar larva. Image is a single z confocal slice showing x,y, x,z and y,z planes. Arrow shows an astrocyte cell body at the top of the MB dorsal lobe. Some of the processes of astrocytes were seen infiltrating the dorsal and medial lobes, but fewer than other regions of the neuropil. Genotype: alrm>GFP is alrm-Gal4, UAS-mCD8::GFP/+. Scale bar, 10 µm.
83
Figure 2.21: Categories used to quantify MB γ pruning phenotypes. Brains were immunostained with anti-FasII to visualize MB γ neurons. Phenotypes in the MB lobes were classified into four types of categories. (See Materials and Methods for detailed explanation of the categories). These categories were used for quantification seen in Figures 2.5, 2.22, 2.23, 2.24. Images are confocal z-stack projections. Time points are as indicated (APF). (A) Categories for 18hr APF. Traced regions are adult-specific dorsal α/β MB lobes. In (A-C), arrows show regions where extra debris was seen. Scale bars for (A-C), 10 µm. (B) Categories for 48hr APF. (C) Categories for adult.
84
Figure 2.22: Blocking endocytic activity of astrocytes leads to MB γ neuronal pruning defects. (A) MB γ neurons were labeled with anti-FasII (red). Images are confocal z-stack projections. Time point is as indicated (APF). Traced regions are adult-specific dorsal α/β MB lobes. Arrows show regions where extra debris was seen. Endocytic activity in astrocytes is blocked by using alrm-Gal4 driver to express UAS-shits. The flies were raised at 18°C, then shifted to 30°C from prepupa (0hr APF) for 18hr. Scale bars, 10 µm. Genotypes are: shits /+ = UAS -shits/+ alrm>shits = alrm-Gal4/UAS-shits (B) Quantification of MB γ neuronal pruning phenotype of images in (A). N values: alrm- Gal4/+ N=19, shits/+ N=11, alrm>shits N=22 hemisegments.
85
Figure 2.23: Knockdown of Draper specifically in astrocytes leads to MB γ neuronal pruning defects. (A) MB γ neurons were labeled with anti-FasII (red). Images are confocal z-stack projections. Time point is as indicated (APF). Traced regions are adult-specific dorsal α/β MB lobes. Arrows show regions where extra debris was seen. Scale bars, 10 µm. Genotypes are: drprRNAi/+ = UAS-drprRNAi/+ alrm>drprRNAi = alrm-Gal4/UAS-drprRNAi. N values: alrm-Gal4/+ N=31, drprRNAi/+ N=22, alrm>drprRNAi N=14 hemisegments. (B) Quantification of MB γ neuronal pruning phenotype of images in (A).
86
Figure 2.24: Knockdown of Crk/Mbc/dCed-12 pathway components specifically in astrocytes leads to MB γ neuronal pruning defects. (A) MB γ neurons were labeled with anti-FasII (red). Time points are as indicated (APF). Images are confocal z-stack projections. In (A) and (C), traced regions are adult-specific dorsal α/β MB lobes and arrows show regions where extra debris was seen. Scale bars, 10 µm. Genotypes in (A) and (B): control = alrm-Gal4/+ alrm>CrkRNAi = alrm-Gal4/UAS-CrkRNAi alrm>mbcRNAi = alrm-Gal4/+;+/UAS-mbcRNAi. (B) Quantification of MB γ neuronal pruning phenotype of images in (A). N values: control N=10, alrm>CrkRNAi N=12, alrm>mbcRNAi N=20 hemisegments. (C) MB γ neurons were labeled as in (A). Time points are as indicated (APF). Images are confocal z-stack projections. Arrows show regions where extra debris was seen. Scale bars, 10 µm. Genotypes in (C) and (D) are: CrkRNAi/+ = UAS-CrkRNAi/+ control = alrm-Gal4/+ alrm >CrkRNAi = alrm-Gal4/UAS-CrkRNAi alrm>CrkRNAi + drpr∆5 = alrm-Gal4/CrkRNAi; drpr∆5 (D) Quantification of MB γ neuronal pruning phenotype of images in (C) as in (B). N values: CrkRNAi/+ N=12, control N=58, alrm >CrkRNAi N=16, alrm>CrkRNAi + drpr∆5 N=12 hemisegments.
87
Figure 2.25: Pan-glial knockdown of dCed-12 does not perturb clearance of vCrz+ cell bodies. (A) Crz+ neurons were labeled with anti-Crz (green). Time point is as indicated (APF). Confocal images are z-projection images. Genotype is: repo>dCed-12RNAi = dCed-12RNAi/+; repo-Gal4/+. Scale bars, 20 µm. (B) Quantification of pixel intensity of vCrz+ debris of figures in (A). N values for (B) and (C): N=15 brains quantified for both repo-Gal4/+ and repo>dCed-12RNAi. The analysis for (B) and (C) (see Statistical analysis in Materials and Methods) was done along with the samples in figure 2.6B-D since the experiment was done together using the same conditions. Error bars represent ± s.e.m. (C) Quantification of the percentage of remaining vCrz+ cell bodies of figures in (A). Error bars represent ± s.e.m.
88
Figure 2.26: Draper function in non-astrocyte glia is required for proper clearance of vCrz+ cell bodies. (A) Crz+ neurons were labeled with anti-Crz (green). Time point is as indicated (APF). Confocal images are z-projection images. Genotypes: repo-Gal4, alrm-Gal80>drprRNAi = drprRNAi/+; repo-Gal4, alrm-Gal80/+. repo-Gal4>drprRNAi = drprRNAi/+; repo-Gal4/+. Scale bars, 20 µm. (B) Quantification of the percentage of remaining vCrz+ cell bodies of figures in (A). N values: drprRNAi/+, N=13; repo-Gal4, alrm-Gal80> drprRNAi, N=11; repo-Gal4> drprRNAi, N=11 brains quantified. Error bars represent ± s.e.m. ****P< 0.0001.
89
Figure 2.27: There are low to undetectable levels of Draper immunoreactivity in vCrz+ neurons. vCrz+ neurons were labeled with GFP (green), anti-Draper (red). Time point is as indicated (APF). Image is a single z confocal slice. Arrows show a vCrz+ neuronal cell body at the cortex in x,y, x,z and y,z planes. Asterisk shows an astrocyte cell body labeled with anti-Draper at the edge of the neuropil. Draper immunoreactivity is seen in the cortex, astrocyte cell bodies and astrocyte phagolysosomal structures in the neuropil. The two panels on the right are higher magnification views of the boxed region. Note Draper immunoreactivity does not co-localize with the vCrz+ neuronal cell body in the CNS. Genotype is crz>GFP is crz-Gal4, UAS-mCD8::GFP/+. Scale bar, 10 µm.
90
Figure 2.28: Draper does not function in vCrz+ neurons for the clearance of vCrz+ debris. (A) Crz+ neurons were labeled with anti-Crz (green). Time point is as indicated (APF). Confocal images are z-projection images. Scale bars, 20 µm. (B) Quantification of pixel intensity of vCrz+ debris of figures in (A). N values: drprRNAi/+, N=8; crz-Gal4/+, N=11; crz-Gal4/drprRNAi, N=11 brains quantified. Error bars represent ± s.e.m. (C) Quantification of the percentage of remaining vCrz+ cell bodies of figures in (A). N values: drprRNAi/+, N=13; crz-Gal4/+, N=11; crz-Gal4/drprRNAi, N=11 brains quantified.
91
CHAPTER III: Discussion
92
In this dissertation, I investigated the biology of astrocytes as neuronal refinement takes
place in Drosophila, specifically during metamorphosis. After making a Gal4 driver line
that enabled the study of astrocytes, I found that larval astrocytes reside at the edge of the
neuropil and their processes have a tufted morphology, infiltrate the neuropil region and
closely associate with synapses. Furthermore, these astrocytes transform into phagocytes,
upregulate engulfment receptor, Draper, and form acidic phagolysosomes, through
activation of a cell-autonomous ecdysone-dependent program at the initiation of metamorphosis. Astrocytes are the primary phagocytic cell type in the pupal neuropil, and engulf synaptic and neurite material. I examined the developmental removal of two
neuron populations- MB γ neurons and vCrz+ neurons- to study the elimination of either neurite processes or entire neurons, respectively. I found that Draper and Crk/Mbc/dCed-
12 signaling pathways are both required and act redundantly for the elimination of MB γ axons by astrocytes. In contrast, elimination of vCrz+ cell bodies requires Draper but not
Crk/Mbc/dCed-12, whereas elimination of vCrz+ neurites requires Crk/Mbc/dCed-12 but not Draper. Interestingly, I also found that loss of Draper delayed vCrz+ neurite
degeneration, suggesting that glia promote neurite destruction via engulfment signaling.
My work led to the generation of a genetic tool that enabled research on astrocytes in
Drosophila, identification of a novel role for astrocytes in the clearance of synaptic and
neuronal debris, Crk/Mbc/dCed-12 as a new glial engulfment pathway facilitating
pruning, and unpredictably, that different engulfment signaling pathways are engaged by
glia when neuronal debris was generated through cell death or local pruning. 93
Astrocytes’ role as phagocytes
As mentioned previously, astrocytes in culture can show phagocytic properties
(Roldán et al., 1997; Tansey and Cammer, 1998), and engulf apoptotic cells (Chang et al.,
2000) or amyloid-β (Wyss-Coray et al., 2003). However, astrocytes are generally
implicated as non-professional phagocytes, and respond to injury less efficiently than
microglia (Parnaik et al., 2000). In vivo data has shown that MTZ astrocytes in the
postlaminar optic nerve head express Mac-2, a molecule implicated in phagocytosis of
myelin by microglia, and internalize axonal cytoplasm and organelles; this event appears
to increase in frequency in glaucoma models (Nguyen et al., 2011). Until recently, a
phagocytic role for astrocytes in developmental pruning has not been directly shown. As
our manuscript was being prepared, Chung and colleagues showed astrocytic synapse
elimination during mammalian developmental retinogeniculate refinement that requires
MEGF10 (mammalian homolog of Draper) and MERTK phagocytic pathways (Chung et
al., 2013). That study, in combination with ours, shows that astrocytic engulfment,
synapse elimination, and the Draper/MEGF10 pathway are conserved cellular and
molecular mechanisms from flies to mammals. Therefore, using flies to investigate this
phenomenon and uncover the cellular and molecular mechanisms during development
will provide insight into the neuronal refinement in mammals.
MERTK has been implicated to regulate CrkII/Dock180/Rac1, the mammalian homolog of Crk/Mbc/Rac1, to control cytoskeletal rearrangements during phagocytosis in
vitro (Wu et al., 2005). Since MERTK is shown to act in astrocytes for phagocytosis of
synapses, and in my study the Crk/Mbc/dCed-12 complex is required for efficient 94
synaptic clearance, it is possible that Crk/CrkII and Mbc/Dock180 are also conserved for
this function. It would be interesting to see if that complex functions in mammalian
astrocytes for synaptic removal.
In Drosophila, no receptor has been implicated acting upstream of
Crk/Mbc/dCed-12 complex for phagocytosis. One possible receptor that is predicted from
the mammalian studies is MERTK, as it is acting in astrocytes for synaptic removal, as
well as acting upstream of CrkII/Dock180/Rac1. Protein BLAST search of MERTK to
Drosophila protein database did not reveal a high homology to any protein (data not
shown). Insulin-like receptor (InR) came up as a low possibility hit; however, it would be
interesting if InR has a role in astrocytic neuronal pruning as the upstream receptor of
Crk/Mbc/dCed-12. BAI1 is another possibility; since it is expressed in mammalian
astrocytes in vivo, accumulates around the phagocytic cup as astrocytes engulf apoptotic
debris in vitro (Sokolowski et al., 2011), binds to PS, and acts as an engulfment receptor
upstream of ELMO/Dock180/Rac (homolog of dCed-12/Mbc/Rac1) (Park et al., 2007).
Protein BLAST search of BAI1 in Drosophila protein database also failed to reveal a
high homology to any Drosophila protein (data not shown). Nevertheless, structurally
similar proteins to MERTK or BAI1 may exist in Drosophila that would act upstream of
the Crk/Mbc/dCed-12 complex.
Integrin can also be acting as a receptor upstream of Crk/Mbc/dCed-12 during
phagocytosis. Integrin is a heterodimer of α and β subunits. In mammals, integrin αvβ3
and αvβ5 are implicated as receptors for MFG-E8 that binds to PS on apoptotic cells and
MFG-E8 triggers DOCK180-dependent Rac1 activation (Akakura et al., 2004; Hanayama 95
et al., 2002). C. elegans integrin α, INA-1, has been shown to be required for efficient
removal of cell corpses and acts upstream of CED-2/-5/-12 (Hsu and Wu, 2010). In
Drosophila, integrin αPS3/βv serves as a receptor for phagocytosis of apoptotic cells in vitro (Nagaosa et al., 2011; Nonaka et al., 2013). It has not been shown in Drosophila whether integrin αPS3/βv acts upstream of Crk/Mbc/dCed-12 or acts in glia for neuronal debris removal. RNAi lines for integrin αPS3 and βv can be requested and expressed in astrocytes to test for clearance defect of axonal debris. There are five α and two β subunits for Drosophila integrin. RNAi lines for the other α subunits (mew, if, ItgαPS4,
ItgαPS5) and βPS (mys) can also be expressed in astrocytes to test for any clearance defect of axonal debris. Integrins may also act in a separate pathway from the
Crk/Mbc/dCed-12 complex, so in addition to astrocytes, Integrins can be knocked down in all glia and vCrz+ cell body clearance can be assayed.
PVR, Drosophila homolog of the PDGF/VEGF receptor, was shown to act
upstream of Crk, Mbc and dCed-12 during thorax closure (Ishimaru et al., 2004).
Although a phagocytic role has not been implicated for PVR, it is possible that PVR may
act in astrocytes upstream of Crk/Mbc/dCed-12 for phagocytosis of axonal debris. PvrRNAi can be requested from Ishimaru and colleagues as well as VDRC stock center and expressed in astrocytes to look for a phenotype in MB γ and vCrz+ axonal debris
clearance. PVR may also act in a separate pathway from the Crk/Mbc/dCed-12 complex,
so in addition to astrocytes, PVR can be knocked down in all glia and vCrz+ cell body clearance can be assayed. 96
Other than the possible receptors discussed above, there are other receptors
implicated for phagocytosis in other tissues of Drosophila such as: Croquemort, Eater,
Nimrod, Peste, and Sr-CI (Franc et al., 1999; Kocks et al., 2005; Kurucz et al., 2007;
Philips et al., 2005; Rämet et al., 2001). These receptors may also act in glia during removal of neuronal debris either upstream of Crk/Mbc/dCed-12 or not. RNAi lines for those genes can be requested and expressed in glia to test for clearance defect of neuronal debris.
Finding new molecules required for astrocytic engulfment
In order to uncover underlying molecules in the astrocytic transformation of
astrocytes and pruning of neuronal debris--one of which could be the receptor upstream
of Crk/Mbc/dCed-12 complex--translating ribosome affinity purification (TRAP) can be
used to reveal the specific genetic profile of astrocytes during these events. TRAP
enables capturing and profiling of actively translated mRNAs from specific subtypes of
cells without disturbing the cellular integrity (Heiman et al., 2008). This method has been
used in mouse to profile different subsets of neurons, and helped to identify known and
new markers (Doyle et al., 2008). Furthermore, TRAP has been adapted for use in
Drosophila with GAL4/UAS system (Thomas et al., 2012). UAS-GFP::L10a (GFP-
tagged ribosomal protein L10a) can be expressed in astrocytes using alrm-Gal4 driver to
get mRNAs specifically expressed in astrocytes. By the use of anti-GFP-coated beads, the
ribosomes that are translating the newly synthesized mRNAs will then be
immunoprecipitated. The mRNAs will then be recovered, isolated, and sequenced. The
genotypes and stages for astrocyte profiling would be: control (alrm-Gal4 > UAS- 97
GFP::L10a) before metamorphosis (stage L3), control just after pupariation (e.g. 3hr
APF) and EcRDN (alrm-Gal4 > UAS-GFP::L10a + UAS-EcRDN) just after pupariation
(e.g. 3hr APF). Comparison of controls at 3hr APF and stage L3 would reveal genes that are upregulated as well as downregulated in astrocytes at metamorphosis. Comparison of
EcRDN at 3hr APF to control at 3hr APF would show expression or repression of what genes are regulated by ecdysone signaling. Not all the genes that come up from the sequencing data will be expected to be real candidates. While comparing sequence data, draper expression level can be used to assess the validity of the results, since Draper protein level is significantly upregulated in astrocytes during metamorphosis and downregulated when EcRDN is expressed in astrocytes (see Chapter II). alrm and repo genes are expected to be found in control L3 sample, and elav and other neuronal-specific genes are not expected to be found (negative control). Transmembrane proteins would be likely candidates for upstream receptors. Proteins that are known to function in macrophages in immune response would also be likely candidates. After the selection of candidates, availability for antibodies against those genes, any mutants, dominant- negative (DN) or RNAi lines would be investigated. Antibodies would be used in order to find if the candidate genes are expressed, upregulated, or downregulated in astrocytes, and whether they are regulated by ecdysone signaling. Mutant, DN and RNAi lines of the candidate genes would be used to test for clearance defects.
Astrocytes are the primary phagocytes in the neuropil
I found that at 6hr APF, all Draper immunoreactivity in the neuropil region
belongs to astrocytes. Non-astrocytic glia do not colocalize with Draper staining in the 98
neuropil at this stage. Additionally, all of the Lysotracker+ structures were surrounded by astrocyte membranes. Adding to that, draper knockdown in astrocytes gave a strong MB
γ pruning phenotype similar to draper nulls (see Chapter II). These data show that astrocytes are the primary phagocytes in the neuropil. Blocking EcR signaling in astrocytes led to strong phenotypes in clearance of neuronal and synaptic material; however, debris got cleared later during metamorphosis, since alrm > EcRDN adults did not have debris of MB γ neurons or vCrz+ neurons (data not shown). This is most likely due to the second metamorphic high titer of ecdysone hormone that activates the wildtype
EcR in astrocytes, and the EcRDN is not sufficient to block all ecdysone signaling.
Evidence supporting that hypothesis is the upregulation of Draper in astrocytes at 48hr
APF (see Chapter II). In the future, EcR mutant astrocyte clones can be made to see if
Draper upregulation and morphological transformation are blocked in astrocytes. Do
other glia compensate for clearance when astrocytes are prevented by phagocytic
transformation by EcRDN? alrm > EcRDN pupa did not show immunoreactivity for Draper
in the neuropil at 6hr APF; however, other glia may react slowly, and later stages can be
checked for Draper upregulation in the neuropil, along with non-astrocytic glial markers.
On the other hand, Draper may not be the only receptor that other glia use for
engulfment. In the future, one can block EcR signaling in astrocytes with alrm-QF driver
(with the available QUAS-EcRDN), and simultaneously look at non-astrocytic glia with
repo-Gal4, alrm-Gal80 driving a membrane tethered fluorescent reporter. If other glia
compensate, the same experiment can be done to dissect out which glial subtype(s) are 99
responsible with the use of Gal4 drivers for specific glial subtypes. It would also be interesting to find out the underlying mechanism leading to such compensation.
Do Drosophila astrocytes have any role in injury response?
As mentioned previously, I made an initial observation that astrocytes do not respond to injury in the Drosophila adult CNS (Doherty et al., 2009). I have also injured larval nerves that come out of the CNS and extend to the body wall, and found that after injury, Draper immunoreactivity occurs in the neuropil region but did not colocalize with the astrocyte membranes (data not shown). We should note that astrocytes might secrete some unidentified molecules after injury, and they may be important regulators of injury signaling to other glia, presumably to further attract ensheathing glia to the injury site.
Also, there may be slight morphological differences that occur after injury. It would be insightful to live-image the astrocytic processes in a clonal fashion, and analyze if astrocytes have dynamic changes in their morphology after injury. There is a possibility that they may retract to give way for ensheathing glia to reach to the site of injury and clear the debris. Also, there may be signaling events that go on between astrocytes and ensheathing glia; astrocytes may relay an injury signal to ensheathing glia, and coordinate the injury response. As mammalian astrocytes communicate through Ca2+ signals, astrocytes may also use Ca2+ signaling to communicate with each other, as well as to ensheathing glia.
What drives Drosophila astrocytes to turn into phagocytes?
Although astrocytes do not respond to injury in a phagocytic manner, they are 100
capable of responding to neuronal debris generated during metamorphosis. What makes astrocytes adopt phagocytic properties during metamorphosis? As shown in Chapter II, cell-autonomous ecdysone signaling instructs the abrupt changes in the properties of astrocytes. Ecdysone signaling regulates formation of Lysotracker-positive acidic phagolysosomes, upregulation of Draper in astrocytes, and elimination of neuronal material from the CNS neuropil. This is no surprise, since ecdysone signaling regulates many events including neuronal death, as well as remodeling in the nervous system during metamorphosis (Lee et al., 2000; Robinow et al., 1993). The reason why astrocytes do not engulf neuronal debris after injury, but do engulf neuronal debris generated during metamorphosis could be that different combination of signals may be created in the neurons in these different contexts. Additionally, astrocytes may not have the molecular machinery in the adult or larval stages needed to respond to injury.
Ecdysone-signaling may be required to change the expressional profile of the astrocytes, making them capable of acting as phagocytes. In fact, as discussed previously, the expression profiling of astrocytes may be performed at the prepupal stages to understand the underlying molecular mechanisms of astrocytic phagocytosis, and find new molecules that might act in those pathways.
Is ecdysone signaling the only mechanism that activates astrocytes? Do neuronal cues have any role, or are they needed in the transformation of astrocytes into phagocytes? Can premature activation of ecdysone signaling in astrocytes turn them into phagocytes prematurely and make them eat neuronal material? If ecdysone signaling were activated in astrocytes later in adult stages, would astrocytes be equipped to respond 101
to injury in this case, or become overactivated and chew up neuronal material? There is
no transgenic fly line that has a constitutively active form of EcR; however, premature
EcR signaling in astrocytes can be triggered by expressing UAS-EcRB1, along with
UAS-Usp (Ghbeish et al., 2001), which can be checked by staining with anti-EcR and anti-Usp antibodies (Sutherland et al., 1995). A constitutively active form of Taiman
(Tai), which had been shown to act as EcR-coactivator and trigger premature ecdysone signaling in border cell migration in Drosophila (Jang et al., 2009), can be expressed along with Usp and EcR. At the same time, ecdysone must be injected, since hormone binding is required in order to activate the cascade and trigger activation by Tai. Tai expression can be checked by staining with anti-Tai antibody (Jang et al., 2009). To see if this premature EcR activation turns astrocytes into phagocytes, Draper upregulation and astrocyte morphology would be assayed. If the astrocytes then upregulate Draper and form phagolysosomes at third instar larva (L3), that would suggest that ecdysone signaling is sufficient for the astrocyte to phagocyte transformation, and neuronal cues are not required. To see if astrocytes engulf any neuronal material in those phagolysosomes, or just randomly form phagolysosomal structures, staining for neuronal markers and astrocyte membranes must be done and imaged. Formation of phagolysosomes without any neuronal cue(s) would be very surprising. It would be interesting if astrocytes would become overactivated, and actively destruct neuronal structures. If activated EcR signaling still will not activate astrocytes prematurely, one possibility is that neuronal eat-me cues are also required. Another possibility is that other necessary non-autonomous signals from surrounding cells (neurons or glia) are not 102
expressed until metamorphosis. We have to keep in mind that the environment of
astrocytes (other cells and extracellular matrix components) may prevent premature
activation of astrocytes. Alternatively, activation of ecdysone signaling in astrocytes can
be attempted in the adult stages to see if astrocytes become hyperactive and chew up
neuronal material. Also, antennal lobe and maxillary pulp injury assays can be performed
(Doherty et al., 2009) to see if astrocytes in this situation would engulf degenerated
axonal material. Interestingly, Draper signaling is required in ensheathing glia to clear
injury-induced neuronal debris (Doherty et al., 2009), and in astrocytes for clearing pruned debris (see Chapter II). That suggests that some signals during these different events must be shared since the engulfment pathway is shared. It is also possible that there may be additional cues generated after injury that are not generated during apoptosis or pruning. Even though astrocytes may be forced molecularly to adopt pupal characteristics, they still may not respond to injury. It would give great insight to the field when such cues will be uncovered.
Expression of DrprI in astrocytes while EcR signaling was blocked was not sufficient to transform astrocytes to phagocytes (see Chapter II), suggesting that additional pathways (other than Draper downstream of EcR signaling) are needed to make them phagocytic. I have shown that both Crk/Mbc/dCed-12 and Draper pathways are required for phagolysosome formation (see Chapter II), suggesting that
Crk/Mbc/dCed-12 pathway may be working together with Draper pathway to make astrocytes phagocytic. In the future, transgenic lines having UAS-Crk, UAS-Mbc and
UAS-dCed-12 can be generated and expressed along with UAS-DrprI in the EcRDN 103
background, and test if that would be sufficient for the formation of phagolysosomes.
If we block neuronal programming, remodeling and death, during metamorphosis,
would astrocytes still become phagocytic and engulf neuronal material, or would they be
prevented from that? Assuming that ecdysone regulates programming of all neurons, pan-
neuronal driver elav-Gal4 can be used to block EcR signaling, while labeling neuronal
processes with GFP to check if death or remodeling is blocked (elav-Gal4 ; UAS-
mCD8GFP / UAS-EcRDN). I previously attempted to do this experiment, however the progeny of this cross die. Therefore, tub-Gal80ts should be used to restrict EcRDN expression to later stages, and allow normal development to occur at previous stages. To do that, larvae can be shifted to permissive temperature to activate elav-Gal4, and therefore EcRDN expression, during metamorphosis. If astrocytes become phagocytic, upregulate Draper, and form phagolysosomes, that suggests that they may be actively engulfing neuronal material. If not, that suggests that ecdysone signaling in astrocytes is not sufficient to trigger astrocyte engulfment, and neuronal signaling to astrocytes is required. The second possibility would be expected since it is previously shown that blocking EcR signaling in MB γ neurons leads to severe suppression of glial infiltration in the MB lobes (Awasaki and Ito, 2004); however, only Draper staining around MB lobes was shown in that study, whereas using astrocyte membrane marker would reveal higher resolution.
I have made the observation that UAS-p35 expression in vCrz neurons led to block of apoptosis for at least 2 days (see Chapter II). Astrocytes activate EcR signaling normally, so if astrocytes would actively engulf neuronal material, we would expect to 104
see degenerated vCrz+ neurites. This suggests that astrocytes require a signal to promote engulfment in addition to the ecdysone signaling.
Why are different phagocytic pathways engaged by glia during different
contexts of remodeling?
As shown in Chapter II, I found that different molecular pathways were engaged
by glia to clear axonal debris generated by pruning versus apoptosis. For MB γ neuronal
pruning, both Draper and Crk/Mbc/dCed-12 pathways are required and act redundantly
for clearance of axonal debris. In contrast, for clearance of vCrz+ neuronal debris generated during apoptosis, Crk/Mbc/dCed-12 pathway is required for neurite clearance, whereas the Draper pathway is dispensable. Interestingly, for vCrz+ cell body clearance,
Draper pathway, but not the Crk/Mbc/dCed-12 pathway, is required for glial clearance.
What cue(s) activate(s) the Draper pathway but not the Crk/Mbc/dCed-12 pathway for
cell body clearance? What cue(s) activate(s) the Crk/Mbc/dCed-12 pathway but not the
Draper pathway for neurite clearance? Interestingly, different compartments of the same
neuron activate different engulfment pathways, suggesting that different eat-me cue(s)
could be generated in different compartments. Why is the Draper pathway not required
for neurite clearance during apoptosis but is required during pruning? It is possible that
different eat-me cue(s) were presented by the neurites in the case of apoptosis versus
pruning. These observations could be due to differences in the molecular program in each
remodeling event. 105
One molecular difference between apoptosis and pruning is the involvement of caspases. As vCrz+ neurons undergo apoptosis, caspases are activated. The overexpression of p35, a caspase inhibitor, suppressed their death for at least a day (Choi et al., 2006), and I found this to be true for at least 2 days (data not shown). However for
MB γ neuronal remodeling, caspases are not involved. Caspase-3 was not detected around MB γ neuron cell bodies or in axons, and expression of p35 or DIAP1, another caspase inhibitor, did not affect MB γ neuron remodeling (Awasaki et al., 2006). Such difference in the pathways could lead to different eat-me cue representation on the neurites.
A recent study showed that distinct pathways are involved in axon degeneration during apoptosis and axon pruning (Cusack et al., 2013). In that study, microfluidic chambers were used to culture neurons so that neuronal cell bodies reside at one side of the chamber and axons reside at the other side. NGF withdrawal from both compartments led to apoptosis, whereas withdrawal from just the axonal compartment led to a phenomenon similar to axonal pruning. Using this system, the group found that, the molecular pathways are distinct in two different events. Although Caspase-9 and -3 are required in both events, local axon pruning requires Caspase-6 but not Apaf-1, whereas axon degeneration during apoptosis requires Apaf-1 and not Caspase-6. Also, during axon-selective degeneration induced by local NGF withdrawal, proteasome activity and
X-linked inhibitor of apoptosis protein (XIAP) protect the cell bodies, but not the axons, from caspase activity. According to these observations that different signaling pathways occur in separate neuronal compartments, we can speculate that neurites and cell bodies 106
could generate different types of eat-me cues for clearance by glia, and that could lead to
glia engaging different engulfment pathways.
Another possibility that we should consider while discussing the Draper pathway
being required in the clearance of MB axonal debris, but not for the vCrz+ apoptotic
debris, is that these neurons may have different molecular programs--not just because of
their differential remodeling, but also from being different subset of neurons. These
neurons may have different intrinsic programs. Still, it would be interesting to find the
different molecular programs between these two neuronal populations as they go through
their remodeling in order to find the signals that activate the unique engulfment
pathways.
What are the eat-me cues?
How do astrocytes distinguish between degenerating versus non-degenerating
neurites? Is there an eat-me cue on the degenerating neurites for astrocytes to recognize,
and if so, what is the eat-me cue that astrocytes are recognizing? There are some
candidate molecules implicated in previous studies, but in many contexts eat-me cues are still a mystery. PS is one of the best-characterized recognition molecules on apoptotic cells. Normally, PS, a membrane phospholipid, resides in the inner leaflet of the plasma membrane; after activation of apoptosis, PS gets flipped to outer leaflet on the cell surface (Fadok et al., 1992, 1998; Schlegel and Williamson, 2001). The exposure of PS on apoptotic cells is conserved in C. elegans, Drosophila and mammals (van den Eijnde et al., 1998; Nagata et al., 2010). The exposure of PS is caspase-dependent (Martin et al., 107
1996), but the underlying mechanism is elusive. PS may bind to a phagocytic receptor directly, or through a bridging molecule that binds to PS (Nagata et al., 2010).
Previously, a receptor for PS was identified and named as PSR (Fadok et al.,
2000). Some other groups also reported that PSR deficiency disrupts engulfment of apoptotic cells, resulting in embryonic lethality in mouse (Kunisaki et al., 2004; Li et al.,
2003), and delays engulfment of cell corpses in C. elegans (Wang et al., 2003). Another group that independently established PSR (Ptdsr) knockout mice showed that PSR-null macrophages do not have a defect in engulfment of apoptotic cells (Böse et al., 2004).
PSR was later shown to be a chromatin-remodeling factor present in the nucleus (Chang et al., 2007). Tim-4, Integrin, Stabilin-2 and BAI1 are proposed receptors for PS (Nagata et al., 2010). As stated previously, BAI1 has been shown to be expressed in astrocytes and acts upstream of ELMO/Dock180/Rac (homologs of Drosophila dCed-12/Mbc/Rac1)
(Park et al., 2007; Sokolowski et al., 2011). Astrocytes were shown to phagocytose apoptotic glioma cells in a PS-dependent way in vitro (Chang et al., 2000). These reports suggest that PS could be recognized by Drosophila astrocytes. Interestingly, mammalian
PS receptor, Stabilin-2, was shown to interact with GULP (mammalian dCed-6) via its
NPXY motif (Park et al., 2008), similar to the interaction of Draper with dCed-6 via its
NPXY motif. It is still a mystery if Draper recognizes PS in vivo for phagocytosis. CED-
1 has been suggested to recognize PS on apoptotic cells in C. elegans (Venegas and
Zhou, 2007), while phagocytosis of apoptotic S2 cells mediated by Draper is independent of PS (Manaka et al., 2004). A recent study showed that Draper extracellular domains,
EMI and NIM, are required for PS-binding in vitro (Tung et al., 2013), suggesting that 108
Draper may function as a receptor recognizing PS in vivo. This has to be tested in vivo by
making a transgenic fly line with constructs in which Draper extracellular EMI and NIM
domains are missing, inserted into a drpr-null background, and testing for apoptotic
clearance of vCrz+ neurons by glia. It has not yet been shown during metamorphosis if PS
is exposed in apoptotic neurons, nor if its recognized by any receptor, so it would be a
novel observation.
PS was generally visualized by injection of fluorescently conjugated Annexin V, a
PS binding protein, into the Drosophila embryo or bath application in vitro. Injection of
fluorescently conjugated Annexin V into the pupal case can be performed to visualize PS
exposure on vCrz+ neurites and cell bodies, unless the development is perturbed by the injection. That would show PS exposure but not a signal for phagocytosis. A transgenic fly line having secreted Annexin V-GFP (or RFP) under UAS element (UAS-Annexin V-
GFP) can be made and then crossed to alrm-Gal4, to visualize if axonal debris displays
PS before being engulfed by astrocytes. Such a construct has been made and used in zebrafish to live-image apoptotic cells (van Ham et al., 2010); therefore, it may work in
Drosophila as well. Before metamorphosis, Annexin V-GFP would not be expected to accumulate on neuronal processes, but will be expressed in astrocytes. Starting with 0-1hr
APF and later, stages could be visualized to see if Annexin V-GFP secreted from astrocytes accumulate on neuronal processes and cell bodies. At the same time, astrocytic membranes can be labeled with another fluorescent reporter (e.g. RFP, mcherry) to see if astrocytes touch the membranes as Annexin V-GFP accumulates on neuronal material.
Also, the timing of the Annexin V-GFP accumulation and fragmentation can be 109
visualized. Although PS exposure is not expected since caspase activity is not involved, the same transgenic line can be used to see if PS exposure occurs during MB γ neuronal remodeling. Mouse MFG-E8, a secreted protein, was shown to be a specific marker for
PS detection in C. elegans (Venegas and Zhou, 2007). Ectopic expression of MFG-
E8::GFP was shown to result in accumulation of persistent cell corpses in C. elegans
(Venegas and Zhou, 2007) and reduced phagocytosis in Drosophila (Tung et al., 2013), so accumulation of neuronal debris may occur in Drosophila neuropil as well. That, in fact, would be an indicator of MFG-E8::GFP masking PS or another cue on the apoptotic neuronal material from recognition and phagocytosis by glia. UAS-MFG-E8::GFP transgenic fly line used in Tung et al, 2013 can be requested and expressed in glia to test if MFG-E8::GFP from glia accumulates on dying vCrz+ neurons. Moreover, accumulation of neuronal debris with ectopic expression of MFG-E8::GFP can be assayed in drpr null background, and compared to MFG-E8::GFP alone and drpr null alone. If there is no further increase of debris that would suggest that PS or another cue masked by MFG-E8 is recognized by Draper.
Recently, the ER protein, Pretaporter (Prtp), has been reported to relocate to the surface of the membrane in apoptotic cells in Drosophila embryo and is a ligand of
Draper in S2 cell culture (Kuraishi et al., 2009). In that study, prtp nulls had normal MB
γ neuronal pruning, suggesting that Draper and Crk/Mbc/dCed-12 pathways have a different ligand in that context. Prtp expression in neurons can be tested with anti-Prtp antibody, although staining should be done without membrane permeabilization to visualize extracellular Prtp. To test if Prtp is a ligand required for vCrz+ neuronal 110
clearance, prtp nulls can be assayed. If defects in neurite or cell body clearance were
observed, that would suggest that Prtp might serve as a ligand for vCrz+ apoptotic clearance. It would be interesting to find if Prtp or another ligand is specifically required for either neurite or cell body clearance, supporting the idea that different eat-me cues are represented in different compartments as well as in different contexts of remodeling.
There are many open questions regarding synapse elimination by astrocytes during metamorphosis. Do all synapses dismantle during metamorphosis, or do specific ones undergo elimination? When I did nc82 stains to label active zones during metamorphosis, I observed that most nc82 staining in the VNC is gone. That suggests that most synapses are dismantled. Only a region close to midline showed nc82 staining at early stages (data not shown), but whether those synapses are gone at later stages is not known. Maybe the dynamics of the dismantling of those synapses is slower than the rest.
In all of the TEM sections of abdominal regions I looked at, I could not find any synapses; however, I did not look at sections of the whole VNC, therefore I cannot definitively say that all synapses are gone. To definitively observe if all synapses are destructed, serial EM must be performed for VNC. It would be interesting if certain synapses are retained and others are eliminated. If there were such a preference, it would be interesting to know the underlying mechanism.
How do astrocytes clear synapses? Do astrocytes clear specific synapses? Are there specific cues from synapses that signal to astrocytes? During mammalian retinogeniculate refinement, it’s been shown that synapses are tagged with complement molecules for elimination. Signal(s) produced by immature astrocytes lead to localization 111
of C1q, an initiating protein in the complement cascade, at the synapses. Accumulation of
C1q presumably leads to localization of C3 at the synapses. Microglia express CR3/C3,
which presumably responds to the C3 at the synapse and leads to synapse engulfment.
Both C3 and CR3 KO mice, as well as C1q KO mice, had decreased level of engulfment
of synapses by microglia and retinogeniculate refinement defects (Schafer et al., 2012;
Stevens et al., 2007). It would be interesting if synapses were tagged for elimination
during metamorphosis as during retinogeniculate refinement. There are complement-like
molecules in Drosophila, and if antibodies are available, staining for those molecules
along with the active zone antibody (nc82) can be performed. Also mutants, DN and
RNAi lines for such molecules can be tested to see if loss of complement-like molecules
disrupts synapse elimination.
Glia instruct neural circuit remodeling
I found that loss of one copy of drpr dominantly suppresses elimination of vCrz+ cell bodies and neurites. Moreover, loss of two copies of drpr (drpr nulls) leads to retention of nearly all vCrz+ cell bodies and higher level of morphologically intact neurite scaffold, suggesting delayed fragmentation of neurites (see Chapter II). The delay of fragmentation of neurite bundles is observed with the resolution obtained with confocal microscopy; for a more definitive answer, serial EM must be performed on vCrz+
neurons.
Caspases are involved in the death of vCrz+ neurons, and expression of p35 blocks their death as well as neurite degeneration for at least 2 days (see Chapter II). Therefore, blocking apoptosis is sufficient to significantly delay neurite degeneration. Previously, 112
the C. elegans homolog of Draper, CED-1, was shown to promote the death of
engulfment targets (Hoeppner et al., 2001; Reddien et al., 2001). This, along with our
data, suggests that engulfing glia promote the execution of apoptosis of vCrz+ neurons via
Draper signaling. Promoting the apoptosis of the cell body would in turn promote degeneration of the neurites. This argues for glia actively sculpting neural networks by promoting the death of neurons. Do glia selectively destruct specific subset of neurons, and how do glia decide which neurons to instruct? Maybe during apoptosis, a very early cue from neurons signal to glia to guide glia to promote destruction of those neurons.
Is apoptosis delayed in drpr nulls? I performed TUNEL assay in order to see if apoptosis is delayed in drpr nulls. vCrz+ neurons become positive for TUNEL during
apoptosis (Choi et al., 2006), which I have observed as well; however, the TUNEL assay
did not label all of the vCrz+ neurons at the same time (Choi et al., 2006), and this could be due to the problem of penetration to the VNC. Therefore, the TUNEL assay was not pursued to compare apoptotic condition of vCrz+ neurons in drpr nulls.
Does the Draper signaling affect caspase activity in vCrz+ neurons, thereby delay
neuronal apoptosis? In order to test if loss of Draper affects caspase activity during vCrz+ apoptosis, I expressed Apoliner, a caspase reporter, in vCrz+ neurons in control and drpr nulls. This construct has a cleavage site for effector caspases Drice and Dcp-1 in between membrane-tagged RFP and nuclear GFP coding sites. Before caspase activity, the cell membranes are labeled with RFP and GFP. If effector caspases are active in the cell, that leads to separation of nuclear GFP after cleavage. The cleaved GFP translocates to nucleus, while RFP stays at the membrane (Bardet et al., 2008). The percentage of vCrz+ 113
cell bodies with nuclear GFP was quantified at 1hr, 2hr and 3hr APF. At 1hr and 2hr
APF, percent of vCrz+ cells with nuclear GFP in drpr null was not significantly different from controls; however, at 3hr APF, the percentage of vCrz+ cells in drpr nulls was
significantly lower than control (2way ANOVA, with Bonferroni’s multiple comparison
test). On average, 81% of vCrz+ cells had nuclear GFP in control, while only 61% of those had nuclear GFP in drpr nulls, resulting in 25% fewer cell bodies with nuclear GFP
(data not shown). Thus, there is slightly lower caspase activity in vCrz+ neurons in drpr nulls compared to control at 3hr APF. Such slight dampening of caspase activity may be critical for the timely execution of apoptosis. This may be the reason for the delay of vCrz+ neuronal destruction, although it may not be the only effect since there is still a significant amount of caspase activity. Another possibility is that the Draper pathway may affect the apoptotic cascade downstream of caspase activation. We must note that drpr nulls do not completely prevent vCrz+ neuronal destruction, so there could also be
additional engulfment pathway(s) that function together with the Draper pathway in
promoting vCrz+ neuronal death. It is still elusive how engulfment pathways could
promote death of target cells; in the context of neuronal-glial signaling, it has not been
studied extensively. Just recently, Han et al showed that epidermal cells facilitate the
fragmentation of sensory neuron dendrites via Draper during their developmental pruning
in Drosophila (Han et al., 2014). Moreover, in the Drosophila ovary, overexpression of
drpr in follicle cells induces death of the nurse cells (Etchegaray et al., 2012). These are
parallel with our observations and, along with the C. elegans study, suggest that
promoting the target cell degeneration could be a conserved function of the Draper 114
pathway. Subsequent work searching for the interaction of the glial Draper pathway
(together with other additional engulfment pathway(s)) and the neuronal apoptotic
pathway would shed light to the elusive neuronal-glial signaling mechanisms underlying
glial control of neuronal death. Future work exploring the survival of the vCrz+ cell
bodies in drpr nulls will be crucial.
Which glial subtype promotes vCrz+ neuronal death? Draper knockdown in astrocytes does not lead to retention of vCrz+ cell bodies at 18hr APF (see Chapter II).
This suggests that a subtype of glia other than astrocytes are responsible for the clearance
of vCrz+ cell bodies. The position of cortex glia makes them suitable to engulf vCrz+ cell
bodies, as well as promote their death, since cortex glial processes surround the neuronal
cell bodies (see Chapter II). Additionally, cortex glia express Draper (see Chapter II);
therefore, we suggest that cortex glia use the Draper pathway to promote vCrz+ neuronal death, as well as remove their cell bodies. In the future, a cortex glial driver (NP2222-
Gal4) can be used to knockdown Draper expression (UAS-drprRNAi) specifically in cortex glia, and by staining for Corazonin, the number of vCrz+ cell bodies can be quantified and compared to drpr nulls. Efficiency of drpr knockdown can be assayed by staining for
Draper. Cortex glia do not remove neuronal debris upon injury (Doherty et al., 2009). In fact, cortex glial engulfment of neuronal corpses has not been implicated before, hence understanding cortex glial engulfment mechanism would give new insight into the biology of this glial subtype.
Is Draper signaling the only mechanism of glial control of neuronal remodeling?
Previously, it has been shown that glia instruct MB γ neuron pruning by secreting Myo, a 115
TGF-β ligand, which activates TGF-β signaling in MB γ neurons to upregulate EcR-B1 expression (Awasaki et al., 2011). In that study, both cortex glia and astrocytes were contributing Myo to neurons, though blocking myo in cortex glia led to a significantly enhanced phenotype compared to astrocytic knockdown. Since we propose that cortex glia instruct vCrz+ neuronal death, it would be interesting to find if Myo from cortex glia also acts upstream of EcR signaling, thereby initiating vCrz+ neuronal apoptosis. If so, it would be interesting to examine the communication between the Myo/TGF-β and Draper pathways in instructing vCrz+ neuronal death.
Cellular and molecular similarities between neuronal remodeling and
neuronal degeneration after injury and disease
One of the goals of research on developmental pruning, and elimination of
processes after injury and in disease, is to understand the underlying processes; from that,
the ultimate goal is to design treatments to regenerate the complex organization of neural
circuitry in the adult nervous system that is damaged by chronic disease or injury.
Research on developmental remodeling and clearance mechanisms in Drosophila can
give insights for that ultimate goal, since there are similarities in the cellular and
molecular mechanisms underlying such events when compared to higher organisms.
Adding to that, Drosophila has fast generation time, powerful genetic tools to manipulate
specific types of cell populations in temporal manner, and a stereotyped nervous system.
Studies on developmental pruning and elimination mechanisms are essential to
provide information on how developmental plasticity occurs, and to understand the 116
mechanisms of how neuronal degeneration takes place in response to injury and in diseases. There are several similarities between axon degeneration in disease and injury, and axonal degeneration during pruning, suggesting that the mechanisms in developmental axonal pruning may be reused during axonal degeneration after injury and in neurodegenerative diseases.
After an axon is cut, the segment of the axon distal to the injury site degenerates over a period of days by a process named as Wallerian degeneration (Waller, 1850). In
Wlds mice, Wallerian degeneration is blocked, and neurons are protected. During developmental pruning, axons undergo degeneration similar to Wallerian degeneration.
MB γ neuronal pruning in Drosophila shares many similarities with Wallerian degeneration in mammals: 1) both processes are regulated by cell-autonomous action of a genetic program. As mentioned previously, MB γ neuronal pruning is regulated cell- autonomously by ecdysone signaling. Transplantation experiments to Wlds mice showed that severed Wlds axons surrounded by wild type cells behave similarly to those in Wlds mice, showing that the Wlds protective effect is autonomous to neurons (Glass et al.,
1993). 2) Degradation of the microtubule cytoskeleton is identified as the earliest sign of
MB γ axon pruning (Watts et al., 2003), and in an in vitro model of Wallerian degeneration (Zhai et al., 2003). 3) Inhibition of caspase activity had no effect on axon pruning in Drosophila MB γ neurons (Watts et al., 2003), and did not prevent or delay
Wallerian degeneration (Finn et al., 2000). 4) UPS is cell-autonomously required for MB
γ axon pruning, and may be involved in the initiation of the axon degeneration program
(Watts et al., 2003), similar to the early involvement of UPS in Wallerian degeneration 117
(Zhai et al., 2003). These data support the notion that similar underlying mechanisms are
involved in injury-induced Wallerian degeneration and developmental axon degeneration.
Contrary to that, Wlds protein has no effect on developmental axon degeneration in
Drosophila or mice (Hoopfer et al., 2006). Such a difference could be due to distinctions
in the triggering mechanism. It is proposed that Wlds protein likely functions at earlier steps in Wallerian degeneration, and the later steps in the pathway are similar to developmental degeneration (Hoopfer et al., 2006). Interestingly, although Wlds has no
effect on MB γ neuronal pruning, it delays developmental dendrite remodeling of sensory
C4da neurons in Drosophila (Schoenmann et al., 2010). In that study, it was shown that
enzymatic acitivity of Nmnat1, a portion of Wlds, is required for normal dendritic
pruning. The fact that caspase activity is involved in dendrite pruning, but not in
Wallerian degeneration, adds complexity to that. Interestingly, Wlds protein has no effect
on dendrite pruning of MB γ neurons (Hoopfer et al., 2006), but does on dendrite pruning
of sensory C4da neurons. Although ecdysone signaling regulates both events, it is
obvious that there are differences in the pathways. These findings point to shared
mechanisms, as well as complexities in Wallerian degeneration of axons and
developmental pruning of Drosophila MB γ neurons and sensory dendrites. There are
probably points of convergence and divergence in their pathways. We can use
Drosophila to uncover the machineries of developmental pruning events to shed light on
not only the developmental pruning mechanisms in mammalian nervous system, but also
on Wallerian degeneration. 118
During many chronic neurological diseases, axon segments undergo a process similar to Wallerian degeneration. Introduction of the Wlds protein in pmn mice, which normally display motor axon degeneration in a “dying back” mode (Schmalbruch et al.,
1991), delayed axon degeneration and rescued motor neuron death (Ferri et al., 2003) and in P0 mutants, which normally display myelin-related axon loss, reduced axon loss and increased motor neuron survival (Samsam et al., 2003). Since there are similarities between developmental pruning and Wallerian degeneration, studying developmental pruning may also give insight into some aspects of the degeneration mechanisms in neurological diseases.
Conclusions
Drosophila offers great opportunity to do forward and reverse genetics, has a simple yet sophisticated and stereotypic nervous system, a fast generation time, and powerful genetic tools to simultaneously label and manipulate specific subtypes of cells in the same organism. As discussed, many functions of glia are conserved in flies and mammals. Adding to that, many cellular mechanisms are conserved as well, including the engulfment pathways used by glia to get rid of debris generated during development, as well as in response to injury and in disease. The underlying mechanisms of neuronal remodeling and clearance of the debris generated during neuronal remodeling are still elusive. Studies of glial function in Drosophila will give new insights into the neuron- glial signaling mechanisms involved in the engagement of phagocytic pathways during clearance of neuronal subcompartments in different contexts of developmental remodeling, as well as during neurodegenerative events after injury and in diseases. My 119
work showed that astrocytes are an important player in clearance of neuronal debris, a novel function for astrocytes during development of nervous system. That opened a new direction, as well as questions regarding astrocyte biology, which we can address with the tools available Drosophila. Additionally, I found that, depending on the context, glia engage different phagocytic pathways for clearance--a very underappreciated topic.
Exploring the mechanisms of neuronal-glial signaling during different remodeling events can give us insight into the complexity of the nervous system as it assembles, and provide us new treatments and therapeutics for neurodegeneration after injury and in diseases.
120
APPENDIX I: Investigation of the function of the alrm gene
The work presented in this appendix, except olfactory memory assays, was performed in
the lab of Dr Marc R. Freeman. All experiments were performed by me, except those
noted below. I would like to thank Yuly Fuentes-Medel, PhD for teaching and helping for the NMJ experiments. Olfactory memory assays were done in Dr Scott Waddell’s lab in
University of Massachusetts Medical School.
Many thanks for experimental help goes to:
Samuel Licciardo, MD: alrm expression in LG and generation of alrm mutants
Kimberly Kerr, PhD: Alrm secretion studies in S2 cells
Christopher Burke, PhD: Olfactory memory assays
Zhiguo Ma, PhD: Larval olfactory plate assays
121
alrm is expressed in longitudinal glia in the Drosophila embryo
Gliogenesis of lateral glia in the Drosophila embryo is controlled by a single gene, gcm.
gcm encodes a transcription factor that is expressed very early in the specification of all
glia, with the exception of midline glia (Akiyama et al., 1996; Hosoya et al., 1995; Jones
et al., 1995; Schreiber et al., 1997; Vincent et al., 1996). gcm loss in neuroglial
progenitors lead to transformation of presumptive glial cells into neurons and conversely,
gcm overexpression can lead to the transformation of presumptive neurons into glia
(Hosoya et al., 1995; Jones et al., 1995; Vincent et al., 1996). In a screen to identify glial
genes, Freeman et al. (2003) used a combination of different approaches – computational
Gcm-binding site search algorithm, cDNA microarrays and Drosophila database searches- and identified over 40 new glial-expressed genes. Two phenotypes were used to help demonstrate glial expression and regulation by gcm: (1) genes were assayed for loss of expression in gcm mutants, and (2) genes were assayed for upregulation when gcm was ectopically expressed.
When gcm was overexpressed in neuroblasts by using sca-Gal4 driver, the majority of the glial genes were upregulated (Freeman et al., 2003). For example, repo expression is activated in the vast majority of neuronal lineages at high levels in a sca-Gal4/UAS-Gcm background. Curiously, the expression of one of the genes from the screen, CG11910, which we named later as astrocytic leucine-rich repeat molecule (alrm), was downregulated when gcm was overexpressed. In control embryos, alrm is expressed specifically in the longitudinal glia (Figure A.1, A.2) and its expression is eliminated in gcm mutants, arguing for glial specificity. However in the gcm-overexpression 122
background, alrm expression, rather than increase with the production of additional glial cells, is decreased and is observed in fewer glia than in control animals (Figure A.3, A.4).
A similar observation, that alrm does not show an increase, was also noted later by another group (Altenhein et al., 2006). This observation makes alrm unique in that its expression goes down when one forces the specification of increased glial in the embryonic CNS. One attractive explanation might be that while alrm expression is restricted to the LG, its activation might require contact between glia and neurons. Hence the elimination of most neurons (by transforming them into glia) might explain the decreased expression of alrm in these backgrounds. This suggests that understanding how alrm is regulated might reveal important mechanistic insights into how neuron!glia signaling regulates glial gene expression. In addition, since Alrm is predicted to be a secreted protein, defining the function of this gene might give insight into the molecular mechanisms of glia!neuron signaling important for development.
In this Appendix I describe my work, and that of others, aimed at defining the mechanism by which alrm expression is activated, and potential functional roles for Alrm in nervous system development and function.
What are longitudinal glial cells and how do they develop?
In the Drosophila CNS, neural progenitors delaminate from the ectoderm, and neural progenitors undergo repeated asymmetric divisions to produce daughter cells termed ganglion mother cells (GMCs) and regenerate themselves. Each GMC then divides to produce neurons and/or glia. There are three types of neural progenitors: neuroblasts (NBs), which give rise to only neurons; neuroglioblasts (NGBs), which give 123
rise to neurons and glia; and GBs, which give rise to only glial cells (Bossing et al., 1996;
Schmidt et al., 1997).
Longitudinal glial cells (LG) are derived from the glial precursor (GP). The GP is
born at late stage 10 at the lateral edge of the neuroepithelium. After the first division at
late stage 10, the GP progeny begin to migrating medially and dorsally. Between stages
11-12, more progeny (the LG) are made and migrate medially to make contact with the
longitudinal commissures which houses motorneurons and interneurons that are
developing in the CNS. Ultimately, the GP gives rise to seven to nine LG per
hemisegment (Schmidt et al., 1997). At late embryonic stages, the LG flank the midline
and reside dorsally at each side of the midline along the length of the CNS. Early in their
development the LG express Gcm, are Repo+, and a very specific late marker for LG is
the enhancer trap line F263 (Jacobs et al., 1989). To determine whether the alrm
transcript is expressed in LG, we used the F263 line to label LG and performed a
fluorescent RNA in situ to the alrm gene. We found a perfect overlap between the
cytoplasmic β-gal from the F236 transgene and the alrm mRNA signal. (Figure A.2).
Thus alrm appears to be specifically in the LG at mid to late embryonic stages.
Alrm is a leucine-rich repeat secreted protein
The alrm gene is located on the right arm of the 3rd chromosome at the cytogenetic location 96D2-3. The predicted Alrm protein sequence has multiple leucine-rich repeats
(LRR), which maybe involved in protein-protein interactions. Nearly all LRR proteins are transmembrane domain proteins, GPI-anchored to the outer membrane leaflet, or soluble secreted molecules (de Wit et al., 2011). Based on the absence of an obvious 124
transmembrane domain or GPT anchor site, we sought to determine whether Alrm was a
secreted protein, and if LG could release Alrm in vivo. As a first step we expressed Myc-
tagged versions of alrm in Drosophila S2 cells and assayed for Alrm-Myc protein in the
culture media.
Insect Schneider S2 cells were seeded with a concentration of 1 x 106 cells/ml for
2ml. The constructs in pAc vector were transfected into S2 cells with Effectene from
Qiagen using the manufacturer’s protocol. An empty pAc vector served as the control.
Media was removed 48 hrs after transfection to avoid contamination by cells that may have lysed during or immediately after transfection. Cells were resuspended in 0.5ml of media and incubated for 24 hours. Supernatant and cells were then collected. The supernatant was centrifuged at 8000 rpm with increments of 1min for total time 12:30 minutes to concentrate from 500µl to ~50µl using centricon tube (YM-30). Loading buffer was added and samples were boiled for 10 minutes and froze at -20°C. Cells were harvested by washing 1ml 1X PBS, and centrifugation at 4°C at 3500rpm for 10 minutes.
Extraction buffer, including DTT and protease inhibitor, was added, cells were centrifuged at 4°C for 10 minutes. Supernatant was removed, and centrifuged again, LB was added, boiled for 5 minutes and froze for storage. All samples were analyzed by western blot and the antibody used was anti-myc. As a control for cell lysis and the presence of cytosolic proteins in the supernatant, anti-actin levels were determined in the supernatant and lysed cell samples. The dilutions for the antibodies were: 1:500 for anti- myc and 1:1600 for anti-actin. 125
When cells were transfected with both UAS-alrm-myc and Gal4pAc, we observed high levels of Alrm in samples from both lysed cells and the supernatant (Figure A.5). In contrast, while abundant actin was present in the lysed cells, we did not detect actin in the supernatant, arguing for minimal cell lysis and that Alrm-Myc was efficiently secreted into the supernatant. Alrm-Myc was not detected in negative controls: in cells transfected with UAS-alrm-myc with pAc since Gal4 element to drive expression of Alrm is missing; or in cells transfected with UAS-myc with Gal4pAc since alrm sequence is missing in the construct. Except untransfected cells, all cells were also positive for GFP, which was cotransfected as a marker for transfection rates. These data indicate that Alrm-Myc is efficiently secreted in vitro and suggests that Alrm could be secreted in glia in vivo.
Curiously, we note that the molecular weight of Alrm-Myc is approximately 2-fold larger than the predicted molecular weight, in a complex that is resistant to detergents and DTT.
Thus Alrm-Myc may form a very stable homo-dimer, or heterodimer with another yet to be identified molecule.
Glial expressed Alrm::myc preferentially accumulates in the neuropil
We next wished to explore the possibility that Alrm was secreted in the intact nervous system by LG. We therefore overexpressed Alrm-Myc in glia by using the pan- glial repo-Gal4 driver to express UAS-alrm::myc and then assayed for anti-Myc. We found that Alrm-Myc accumulated at high levels in glial cell bodies, but also that Alrm-
Myc preferentially accumulated in the neuropil as compared to the cell cortex region of the brain (Figure A.6). That over-expressed Alrm-Myc did not accumulate throughout the
CNS suggests some specificity to its localization to the synapse-rich synaptic neuropil. 126
alrm expression is regulated by repo
There are several genes that have been identified as targets of gcm, the master
transcriptional regulator of glial cell differentiation. repo encodes a homeodomain
transcription factor that is expressed in all lateral glial cells, all glia except midline glia
(Campbell et al., 1994; Halter et al., 1995; Xiong et al., 1994). Repo binds the ATT
sequence in the CAATTA homeodomain binding motif to activate downstream genes
(Halter et al., 1995; Yuasa, 2003). When we analyzed the 1kb upstream sequence of alrm
locus, two CAATTA motifs are detected upstream of alrm. One site is ~ 350bp upstream
and the other ~ 828bp upstream of the translational start site of the alrm gene (Figure
A.7). This observation suggested that Repo might bind and activate alrm and we
therefore investigated whether repo mutants exhibited defects in alrm expression during
embryogenesis.
repo3702 is an enhancer trap line identified from a lethal P-element insertion collection as a lethal, loss of function allele of repo (Halter et al., 1995; Xiong et al.,
1994). In the repo3702 mutant background we performed RNA in situ hybridizations to
detect the alrm transcript. To positively identify homozygous loss of function mutants,
we also co-stained for the Repo protein and used the absence of Repo as an indicator. As
described above, we found that alrm expression is activated around stage 13 and remains
high through late embryonic stages. 100 embryos between stages 13-16 were scored for
the absence of Repo and alrm. 75% of the embryos were both positive for Repo and alrm.
25% of the population however, lacked Repo immunoreactivity and all of those embryos
were also negative for alrm expression (Figure A.7). Thus loss of repo function is 127
sufficient to block activation of alrm expression in the LG, arguing that alrm is a direct or
indirect transcriptional target of Repo. Based on the presence of the Repo binding sites in
the alrm promoter we speculate that this regulation is direct. In the future, Chromatin
Immunoprecipitation (ChIP) assays can be performed with Repo to identify if Repo binds
to regulatory region of alrm. Such assay has been done recently to identify potential Repo
targets and the same experimental procedure can be employed (Kerr et al., 2014). To
show that the predicted CAATTA motifs upstream of alrm are binding sites for Repo,
gel-shift assays could be performed using oligonucleotide probes containing the predicted
motifs and probes that don’t (negative control). Additionally, DNA pull-down assays
using DNA probes labeled with high affinity tag, like biotin, to purify the DNA-protein
complex and later detect the presence of Repo by western blot (WB).
Generation of alrm mutants
alrm mutants had previously made by Samuel Licciardo in the lab (Licciardo,
2006). Ends-out homologous recombination was used to generate alrm knock-outs (Gong and Golic, 2004), whereby the alrm locus was replaced by mini-white gene (Figure A.8).
We first verified the absence of alrm locus by PCR amplification of alrm-specific sequences (Figure A.8). Adult alrm homozygous mutants were homogenized in genomic
DNA extraction buffer (2% Triton X-100, 100mM NaCl, 10mM Tris-HCl-pH 8.0, 1mM
EDTA-pH 8.0, 1% SDS), vortexed with phenol:chloroform (1:1), and centrifuged (14K x
5min). A 462bp fragment of the 1kb ORF was amplified from control and putative alrm null alleles. The primers used were 5’-
CAGTGGTACCTTGGCCTTTGCCAAGGCAGTCCG-3’ and 5’- 128
CAGTGGTACCGTGCGTATCAATTGAAGCG-3’. We amplified for 35 cycles, with
30s each at 95°C, 55°C, and 72°C. PCR products were analyzed on 0.8% agarose gel.
Several potential knockout alleles were identified, and elimination of the alrm locus was confirmed by RNA in situ hybridization with alrm RNA probes to control and alrm null embryos (Figure A.8).
alrm null mutants do not display gross defects in axon pathfinding
Slit, a LRR molecule, is secreted by midline glial cells and regulates axon guidance through interaction with another LRR protein, Robo, during Drosophila embryogenesis (Battye et al., 1999). slit and robo mutants have axon misrouting defects; growth cones that normally do not cross the midline do so in the mutants (Battye et al.,
1999; Kidd et al., 1999; Seeger et al., 1993). It is known that LG are required for orienting projections of some of the pioneer growth cones, fasciculation and defasciculation events as axonal trajectories form and to keep longitudinal axons away from the midline (Hidalgo et al., 2001). As mentioned previously, Alrm is a LRR molecule presumably secreted from LG. According to these findings, Alrm may act as a cue from LG to neurons and regulate axon pathfinding during embryogenesis.
Previously, Sam Licciardo argued in his thesis that alrm mutants might have mild defects in the organization and fasciculation of fascicles in the embryo (Licciardo, 2006).
To explore this possibility further, I stained for FasII, a cell-adhesion molecule, that enables labeling of three fascicles of the longitudinally projecting axon bundles in the 129
embryo and assayed alrm mutants for any axon pathfinding defects, breaks or fusion of
fascicles.
Prior to this analysis, alrm mutants were outcrossed for four generations to the control
w(CS) line to remove any background mutations. alrm mutants showed no detectable
defects in fusion or break of FasII+ fascicles or collapse of fascicles in the midline (Figure
A.9). We conclude that loss of alrm does not lead to major defects in morphogenesis of the axon scaffold.
This experiment shows that either Alrm is not required for proper axon pathfinding during embryogenesis or there are some compensatory mechanisms. The reason why alrm mutants did not show a phenotype may be due to redundancy with other
LRR molecules. It has been shown that LG transiently express robo during embryogenesis (Kinrade et al., 2001). As robo mutants have axon pathfinding defects and
Robo is also a LRR molecule, a future analysis of robo; alrm double mutants could be done to see if there is an enhancement of axon pathfinding phenotype.
alrm is not required as a trophic factor for Eg+ neurons in Drosophila embryo
Drosophila glia regulate neuronal survival during embryogenesis. Targeted genetic
ablation of glia or transforming glial fate to neuronal fate by mutation of gcm lead to non-
autonomous death of the follower neurons (Booth et al., 2000). In addition, contact
between LG and axons appears necessary to maintain neuronal survival; ablation of LG
alone after establishment of longitudinal axons is sufficient to induce neuronal death 130
(Booth et al., 2000). Recently, it has been shown that DmMANF, Drosophila homolog of
mammalian MANF and CDNF, is expressed in glia, secreted in vivo and essential for
maintenance of neurites of dopaminergic neurons in the embryo (Palgi et al., 2009). It has
been speculated that glial cells provide trophic support to neurons during development,
however underlying molecular players still remain elusive. We suspected that Alrm may
be required as a trophic factor for the survival of a group of neurons in the embryo.
Eagle+ (Eg+) neurons have a stereotypical position in the embryo and are small in
number; therefore it is easy to score these neurons in each hemisegment. There are 4 Eg+ neurons close to the midline, 1 GW and 3 EW neurons per hemisegment (Higashijima et al., 1996). Control and alrm mutant embryos were stained as described above, and the rabbit anti-Eg antibody was used at 1:5 dilution. We found that alrm nulls have number of Eg+ neurons that are indistinguishable from control animals (Figure A.10). This shows
that either Alrm does not have a role as trophic support of subset of Eg+ neurons or there are some compensatory mechanisms. In the future, other groups of neurons may be analyzed in alrm nulls to see if Alrm has a role as trophic factor for a specific group of neurons.
Behavioral testing of alrm nulls
Larval locomotion is normal in alrm nulls
LG, similar to oligodendrocytes in mammals, wrap around the neuropil, ensheath
and insulate axons in the CNS. Such insulation is essential for isolating axons from the
environment and so enables proper neuronal firing. Drosophila axotactin (axo), encoding 131
a member of the neurexin protein superfamily, is secreted by a subset of LG and localizes
to axonal tracts. axo nulls cause loss of conduction at elevated temperatures (Yuan and
Ganetzky, 1999). As shown previously, Alrm is presumably secreted from LG and
misexpressed Alrm-myc localizes to axonal tracts, as does Axotactin. If alrm nulls have
defects in axonal conduction, that could lead to defects in locomotion. A study recently
showed that Eaat1, expressed by LG and a few other types of glia, is essential for larval
locomotion (Stacey et al., 2010). As mentioned in Chapter II, Alrm+ LG adopt a tufted morphology later in larval stages, become astrocytes and infiltrate the neuropil region.
Astrocytes’ processes are closely associated with synapses, and so we asked if Alrm is acting as a signaling molecule from astrocytes to neuronal circuits important for larval locomotion. alrm mutants along with their controls (w(CS), w(CS); alrmC1A and w(CS); alrmB4AB) were scored for larval crawling. The larvae were put on an agar plate in a dark room so there would be no preference to light. The plate was put on a grid surface illuminated from below. Larval locomotion was scored as number of times a larva crossed a grid line for 5 minutes. As seen from Figure A.11, there was no significant difference between alrm nulls and controls. This shows that either Alrm is not essential for larval locomotion or there are some compensatory mechanisms. However, it is possible that, as seen for axo mutants, alrm nulls may have defects in axonal conduction in normal or elevated temperatures. Therefore, w(CS); alrmC1A and w(CS); alrmB4AB third
instar larvae were heat-shocked at 37°C for 5 min to determine if they would paralyze as
axo mutants do after a heat shock. However, larvae of neither alrm null line were
paralyzed after heat-shock. It is still possible that a slight change in conduction could be 132
missed in this assay. In the future, electrophysiological recordings in alrm nulls could
show if there is any problem in action potentials.
Larval alrm nulls do not display olfaction defects towards attractant olfactory
stimulus
Astrocytes infiltrate their processes into the neuropil region, as well as in the
antennal lobes (data not shown). Astrocytes are closely associated with synapses in the
neuropil region (see Chapter II). Astrocytes may use Alrm as an important signaling
molecule to regulate the olfactory circuitry. Factors from any glial subtype, including
astrocytes that regulate olfaction remain elusive. We asked if alrm null larvae have any
olfaction defects and tested that by larval olfactory plate assay.
The larval olfactory assay was adapted from (Lilly and Carlson, 1990). Larvae were placed on a Petri plate containing solidified agarose. Two filter discs (placed in cut- off 1.5 mL Eppendorf tube lids) were placed at opposite edges of the plate. After placing
the larvae in the center of the plate with a brush, an aliquot of 25 µl of odorant (isoamyl
acetate, an attractant, diluted 1:1000 in mineral oil) was placed on one disc and an equal
volume of the diluent (mineral oil, M.O.) was added to the other, which is a negative
control. Larvae were allowed to crawl for 5 min in dark, after which the number of
animals on each half of the plate was counted. A response index (RI) was calculated by
subtracting the number of animals on the control half of the plate (C) from the number on
the stimulus half (S) and dividing by the total: RI = (S - C)/(S + C). RI thus varies from 1
(complete attraction) to 0 (no response).
w(CS), w(CS); alrmC1A and w(CS); alrmB4AB lines were tested for olfaction. The 133
alrm mutants did not show significantly different RI compared to control (Figure A.12).
This result suggests that alrm is not required for larval olfactory response to attractant
odorant stimulus.
alrm mutants do not display drastic learning or memory defects
The role of glial cells in olfactory learning and memory has not yet been
established. No factors have been uncovered that signal from glia to MBs to elicit
learning and memory. Astrocytes infiltrate the olfactory lobes (Doherty et al., 2009) as
well as MB lobes in the adult (data not shown). We asked if astrocytes use Alrm as a
signaling molecule to regulate olfactory learning and memory. We tested if alrm null adults have learning and memory defects. Appetitive (sugar) learning was performed as previously described (Krashes and Waddell, 2011a), and shock learning was performed according to a previously published protocol (Krashes and Waddell, 2011b). The performance index (PI) for each learning assay was calculated as: Half score = ((# flies with correct choice) – (# flies with wrong choice))/ total # flies. Half score was determined in the reciprocal experiment in which the other odorant was paired with the electric shock. Then the two half-scores were averaged to get the PI. PI of 1 would represent perfect performance with all of the flies making the right choice. PI of 0 would represent equal distribution of flies.
The two alrm nulls, alrmC1A and alrmB4AB, were tested. In sucrose learning,
neither of the nulls was significantly different than control (Figure A.13). In shock
learning, alrmC1A flies were, unpredictably, slightly better than control, however alrmB4AB flies were similar to control (Figure A.13). When tested for 24hr memory, alrmC1A flies 134
performed slightly worse than control (mean PI for control is 0.2 and alrmC1A flies had a mean of 0.05), however, alrmB4AB flies performed similar to control (Figure A.13).
Results for alrmC1A flies are contradictory; for appetitive learning they are similar to control, for shock learning even better than control, however for 24hr memory slightly worse than control. Contrary to alrmC1A, alrmB4AB flies performed similar to control in all
assays. Since both lines are missing the alrm gene, we would expect similar phenotype
for both lines. Therefore, according to these results, we concluded that alrm nulls do not
have a drastic adverse effect on olfactory learning and memory. It is also possible that
there may be redundant molecular pathway(s) that compensate with the loss of alrm.
alrm mutants do not have drastic tolerance defects in hypoxic conditions
Gas exchange in Drosophila is mediated by trachea, made up of a branched
network of air-filled tubes that penetrate most tissues including the CNS. Most major
trachea of the Drosophila brain grow along glial cells and glia restrict their growth and
branching (Pereanu et al., 2007). Tracheal branches are close to astrocyte membranes in
the neuropil region (Pereanu et al., 2007). While analyzing TEM images for astrocyte
morphology in the neuropil region of third instar larvae, we spotted tracheal elements in
close proximity to astrocyte membranes (data not shown). The close association between
astrocyte processes and trachea could allow for gas exchange between neurons and their
environment (Freeman and Rowitch, 2013).
We asked if Alrm is a necessary factor provided from astrocytes to trachea for
normal functioning of the tracheal system. If there were a problem with gas exchange, the
alrm mutants would not be able to survive under hypoxic conditions. We tested alrm 135
mutants along with controls (w(CS), w(CS); alrmC1A and w(CS); alrmB4AB) in hypoxic
conditions. After 23 hours of egg laying, the adults were removed from the vials and then
half of the vials (2 vials) were put in hypoxia chamber (5% oxygen), while the other half
were kept in normoxia conditions. After 22 days, both sets of vials were essayed for
eclosion rates. Eclosion rates were calculated as the ratio of eclosed pupa to the total
number of pupa (eclosed and non-eclosed) and plotted as percentage. In hypoxia
condition, alrmC1A had slightly lower eclosion rate than control and alrmC1A normoxia
levels, but alrmB4AB had similar rate to its normoxia levels (Figure A.14). Though, we have to keep in mind that eclosion rate of alrmB4AB in normoxia is lower than that of control and alrmC1A. This experiment was done once and total number of pupa that have essayed is between 153-326. This experiment could be repeated to determine if alrmB4AB
eclosion rate in hypoxia would come up lower than control levels in normoxia. Even so,
the eclosion rate of alrm nulls in hypoxia is 22% lower than control, not a very drastic
effect. Future experiments could be done to see if there is a statistically significant lower
eclosion rate in alrm nulls under hypoxia condition.
Morphological studies of alrm nulls
Loss of alrm has no drastic effect on NMJ formation
As mentioned previously, LG wrap around the neuropil, ensheath and insulate
axons in the CNS. As shown previously, Alrm is presumably secreted from LG and
misexpressed Alrm-myc localizes to axonal tracts. Dendrites of motor neurons and the
axonal endings of the presynaptic neurons that contact them reside in the neuropil. Alrm 136
localized to CNS neuropil may be required for normal circuit functioning and elimination
of the Alrm signal from glia to neurons may lead to problems with circuit
formation/function. We asked if alrm nulls have any defects in NMJ formation.
The lines tested were CS, w(CS), w(CS); alrmC1A and w(CS); alrmB4AB. The NMJ preparations, immunolabeling and confocal microscopy were done as previously described (Kerr et al., 2014). The samples were fixed with 4% paraformaldehyde (PFA) for 15 minutes. The dilutions for the antibodies were: 1:20000 for anti-DLG and 1:100 for HRP-Cy3. The experiment was repeated on two different days and samples from both days were analyzed. Neither alrm null showed drastic overall NMJ morphological phenotypes (Figure A.15). Out of 10 NMJs examined in alrmC1A nulls, two had a smaller
overall NMJ field and one had boutons larger, but overall less arborized than control
boutons. Out of 13 NMJs examined in alrmB4AB nulls, one had a smaller overall NMJ
field and two had boutons larger, but overall less arborized than control boutons.
Although general morphology looked grossly normal, follow-up experiments must be
done, including quantification of the size and the number of the type Ib (big) and Is
(small) boutons. It is possible that subtle phenotypes may have been missed. In addition,
the postsynaptic clustering of GluRIIA can be analyzed by immunolabeling for GluRIIA
and quantifying mean intensity and volume. Further, electrophysiological recordings of
the NMJ can be performed to see if the synaptic transmission is altered in alrm nulls.
alrm is not required for MB γ neuron pruning
As shown in Chapter II, astrocytes phagocytose and remove pruned debris during
MB γ neuron pruning. Myo secreted from cortex glia and astrocytes has been shown to be 137
required for the neuronal activation of EcR, which in turn triggers MB γ neuronal
remodeling (Awasaki et al., 2011). We asked if Alrm has a similar role in MB γ neuron
pruning. alrm null and control pupal brains were dissected at 18hr APF, stained with anti-
FasII antibody and then quantified for MB γ neuron pruning phenotypes. The protocol for
the staining and imaging and the categories used to quantify phenotypes can be seen in
Chapter II. As shown in Figure A.16, alrm nulls do not show more severe pruning
phenotype compared to control. This experiment shows that Alrm protein is not required
for proper MB γ neuron pruning.
MB lobes are normal in the adult alrm nulls
As stated previously, alrm null embryos have a normal number of Eg+ neurons (3
EW and 1 GW). As mentioned in Chapter II, alrm is expressed in astrocytes at larval, pupal and adult stages. Their processes infiltrate the neuropil. One neuropil region that astrocytes are closely associated with is the MB lobes. MB neurons are critical for the associative learning and memory of olfactory stimuli (Davis, 1993; Heisenberg et al.,
1985). MBs consist of three types of neurons, γ, α’/β’ and α/β. They have stereotypical morphology and there are antibodies available to label them. As stated previously, pruned
MB γ neurons were properly cleared at 18hr APF during metamorphosis in alrm nulls.
However MB neurons undergo changes during later stages of metamorphosis; γ neurons re-extend axonal projections medially to form the adult γ lobe and α/β neurons are born and their axons form the adult-specific α/β lobes (Lee et al., 1999). Alrm may have a role in the formation, axon guidance and/or maintenance of the adult-specific MB lobes. We asked if alrm nulls have any morphological defects in MB lobes in the adult brain. The 138
anti-Trio antibody labels γ, α’ and β’ lobes and the anti-FasII antibody labels α and β lobes. When used together, these two antibodies stain all of the MB lobes. When we stained with these antibodies, the MB lobes in alrm nulls have normal morphology compared to controls (Figure A.17). This shows that either Alrm is not required for development and/or maintenance of MB neurons or there are some compensatory mechanisms.
General astrocyte morphology is normal in alrm mutants
We asked if loss of alrm would lead to cell-autonomous morphological defects in astrocytes. Alrm signal to neurons could, in turn, signal to astrocytes and potentially regulate their morphology. Another possibility is that Alrm could act as a cell-adhesion molecule, as it is a LRR molecule and loss of it could lead to defects of astrocyte morphology. In addition astrocytes may secrete Alrm which could regulate astrocytes cell-autonomously. We made a line that labels astrocyte membranes in an alrm null background; alrm-Gal4, UAS-mCD8::GFP; alrmCIA. In adult brains, astrocytes infiltrate the neuropil similar to controls (Figure A.18). Therefore, overall astrocyte morphology was normal in alrm null adult brains.
Future directions to uncover a possible role of Alrm protein:
1. Three extracellular LRR protein families have been implicated in vertebrate
synapse formation (de Wit et al., 2011). After extensive synapse elimination
during prepupal stages at metamorphosis, synapse formation occurs in the
Drosophila neuropil (Allie Muthukumar, unpublished observations). 139
Ultrastructural analysis of the late pupal/adult brains after eclosion can be done to
look at the number and structure of the synapses in alrm nulls.
2. It has been shown that LG transiently express robo during embryogenesis
(Kinrade et al., 2001). As robo mutants have axon pathfinding defects and Robo
is also a LRR molecule, a future analysis of robo; alrm double mutants could be
done to see if there is an enhancement of axon pathfinding phenotypes. Also, as
Slit is secreted from glia and a LRR molecule, slit; alrm double mutants could be
assayed for enhancement of axon pathfinding defects. Quantification of triple
mutants of slit, robo and alrm could further be done.
3. Subsets of neurons other than Eg+ neurons in the embryo and MB neurons in the
adult could be analyzed in alrm nulls to see if Alrm is required for trophic
support, maintenance or proper fasciculation of specific groups of neurons.
4. Electrophysiological recordings in alrm nulls could be carried out to show if there
is any problem in action potentials. Recordings could be carried out at room
temperature as well as elevated temperatures, as performed for axo mutants.
However, when we heat-shocked alrm nulls for 5 min, they did not undergo
paralysis, so a major defect in axonal conduction in elevated temperature is
unlikely.
5. alrm null larvae could be tested for gustatory preference defects.
6. alrm null larvae could be tested for visual preference defects.
7. From the work in our lab (Tobias Stork unpublished observations), we know that
Drosophila astrocytes tile with each other, similar to mammalian astrocytes. 140
Analysis of alrm nulls for tiling defects could be performed. It is possible that
Alrm secreted from astrocytes may be required for astrocyte-astrocyte
communication to cover neuropil with non-overlapping domains.
8. As mentioned previously, follow-up experiments of the NMJ phenotype could be
performed such as quantification of the size and the number of the type Ib (big)
and Is (small) boutons, quantification of GluRIIA mean intensity and volume to
analyze GluRIIA distribution and clustering.
9. Astrocyte cytoplasmic inclusions have been observed in some cases of pediatric
epilepsy (Hazrati et al., 2008; Visanji et al., 2012). Also, in adult model systems
astrocyte dysfunction is involved in abnormal neuronal excitability (Gómez-
Gonzalo et al., 2010). There are mutants in Drosophila that cause seizures, one of
them being easily shocked (eas) and the other bang senseless (bss). eas and bss
flies exhibit normal behavior under normal conditions. When long mechanical
stimuli are used, such as vortexing, such mutants become paralyzed, while wild-
type flies are unaffected (Parker et al., 2011; Pavlidis et al., 1994). To test for
paralysis and recovery time, flies are vortexed in a vial and the number of flies
recovered is noted after some time. alrm nulls can be tested the same way along
with controls to see if they have paralysis phenotype. If they do, that would
suggest that Alrm is required for normal electrical activity in target axons. Adult
giant fiber system neural circuit has been used to look for the electrophysiological
basis in paralysis in bang-sensitive mutants eas and bss (Parker et al., 2011;
Pavlidis et al., 1994), so that system can be used to see if electrophysiological 141
properties are affected in alrm mutant adults. On the contrary, alrm mutants may have a defect like maleless-no action potential-temperature sensitive (mlenapts) mutant, which suppresses bang-sensitive defect of eas (Ganetzky and Wu, 1982).
It could be tested if alrm mutants suppress bang-sensitive defect of eas by testing double mutants for eas;; alrm.
142
Figure A.1: alrm is expressed in longitudinal glia (LG) Ventral and lateral view of RNA in situ hybridization for alrm in control embryo. (Figure adapted from (Licciardo, 2006))
143
Figure A.2: alrm is colocalized with longitudinal glial marker F263 F263 line was labeled by staining for cytoplasmic β-gal (red) and alrm mRNA was labeled by fluorescent RNA in situ to against alrm gene (green). Both signals perfectly overlap. (Figure adapted from (Licciardo, 2006))
144
Figure A.3: alrm expression is downregulated in gcm misexpression background repo mRNA was labeled by RNA in situ hybridization against repo gene and alrm mRNA was labeled by RNA in situ hybridization against alrm gene. The genotype is: ectopic Gcm, sca-Gal4/UAS-gcm. (Figure adapted from (Licciardo, 2006))
145
Figure A.4: alrm is expressed in fewer glia in ectopic Gcm background. (A) Control and ectopic Gcm embryos stained for alrm mRNA by RNA in situ hybridization. The genotype is: ectopic Gcm, sca-Gal4/UAS-gcm. (B) Quantification of number of alrm+ glia per hemisegments. There is ~30% reduction in alrm+ glia in ectopic Gcm background. Error bars represent s.e.m., n>8. (Figure adapted from (Licciardo, 2006))
146
Figure A.5: Alrm is secreted from S2 cells. Cells were transfected with the constructs shown above the lanes. Cells and supernatants after transfection were analyzed by anti-myc WB to see if Alrm is secreted from S2 cells to the supernatant. Anti-Actin WB was performed to confirm transfection in cells and as a control for cell lysis and the presence of cytosolic proteins in the supernatant.
147
Figure A.6: Glial expressed Alrm::Myc preferentially accumulates in the neuropil. A ventral and cross-sectional view of a late stage embryo stained for Myc to label Alrm::Myc (red) and GFP to label glia (green). The genotype is: UAS-GFP; repo-Gal4/ UAS-alrm::myc.
148
Figure A.7: alrm expression is regulated by repo. (A) Scheme showing CAATTA motifs 1kb upstream of alrm locus. Two CAATTA motifs were detected upstream of alrm. One site is ~ 350bp upstream and the other ~ 828bp upstream of the translational start site of the alrm gene. (B) RNA in situ hybridization for alrm in control (repo3702/TM6,Tb) and repo3702 null embryos. repo3702 nulls lack alrm staining.
149
Figure A.8: Generation of alrm mutants. (A) The scheme of generation of alrm mutants by homologous recombination. alrm locus was replaced by mini-white gene. (B) The absence of alrm locus was confirmed by PCR amplification of alrm-specific sequences from control and alrmC1A. (C) Late stage control and alrmC1A null embryos stained for alrm mRNA by RNA in situ hybridization. alrmC1A null embryo lacks alrm staining. (Figure adapted from (Licciardo, 2006)).
150
Figure A.9: alrm mutants do not have axon pathfinding defects in the embryo Late stage control and alrm null embryos were stained for FasII to label three axon fascicles. The number of hemisegments, breaks in fascicles and fusion of fascicles quantified were shown in the table.
151
Figure A.10: alrm nulls have normal number of Eg+ neurons. (A) Control and alrm null embryos were stained for Eg and neurons close to the midline (1 GW and 3 EW) were counted. (B) Quantification of the Eg+ neurons shown in (A) per hemisegment. Error bars represent ± SEM, N≥51, (*) P < 0.05. The statistical analysis used was one-way ANOVA with Tukey’s multiple comparison test.
152
Figure A.11: Larval locomotion is normal in alrm nulls. Quantification of gridbars crossed by control and alrm null third instar larvae for 5 minutes was plotted. Error bars represent ± SEM, N≥5. The statistical analysis used was one-way ANOVA with Tukey’s multiple comparison test.
153
Figure A.12: Larval alrm nulls do not display olfaction defects Response index of control and alrm null third instar larvae for isoamyl acetate was plotted. Error bars represent ± SEM, N≥5. The statistical analysis used was one-way ANOVA with Tukey’s multiple comparison test.
154
Figure A.13: alrm nulls do not display gross learning or memory defects. (A) Performance index for sucrose learning was plotted for control and alrm null adults. (B) Performance index for shock learning was plotted for control and alrm null adults. (C) Performance index for 24hr memory was plotted for control and alrm null adults.
155
Figure A.14: alrm mutants do not have drastic tolerance defects in hypoxic conditions. Eclosion rate for control and alrm nulls in normoxia and hypoxia conditions was plotted. N≥153.
156
Figure A.15: alrm nulls do not have a drastic effect on NMJ formation. Control and alrm null NMJ preparations at muscle 6/7 were labeled with the presynaptic marker anti-HRP (red) and the postsynaptic marker anti-DLG (green).
157
Figure A.16: alrm is not required for MB γ pruning. (A) MB lobes were labeled with anti-FasII (red) at 18hr APF. Confocal images are z- projections. Arrows indicate extra MB γ neuron debris. (B) Quantification of MB γ neuron pruning phenotype from (A) using the categories shown in Chapter II. N-values are as follows: w(CS), N = 22; w(CS); alrmC1A, N = 11 hemisegments quantified.
158
Figure A.17: Morphology of MB lobes is normal in alrm null adults. MB α and β lobes were labeled with anti-FasII (red) and γ, α’ and β’ lobes were labeled with anti-Trio (blue). Confocal images are z-projections. The genotypes are as indicated in the figure.
159
Figure A.18: Astrocyte morphology is normal in alrm nulls. Astrocytes were labeled with GFP (alrm-Gal4, UAS-mCD8::GFP/+; green). The genotypes are as indicated in the figure. Confocal images are single z-confocal slices. Images show astrocyte membranes infiltrating antennal lobes, Subesophageal ganglion (SOG) and antennal nerve.
160
References
Abrams, J.M., White, K., Fessler, L.I., and Steller, H. (1993). Programmed cell death during Drosophila embryogenesis. Dev. Cambridge Engl. 117, 29–43.
Akakura, S., Singh, S., Spataro, M., Akakura, R., Kim, J.-I., Albert, M.L., and Birge, R.B. (2004). The opsonin MFG-E8 is a ligand for the αvβ5 integrin and triggers DOCK180-dependent Rac1 activation for the phagocytosis of apoptotic cells. Exp. Cell Res. 292, 403–416.
Akiyama, Y., Hosoya, T., Poole, A.M., and Hotta, Y. (1996). The gcm-motif: a novel DNA-binding motif conserved in Drosophila and mammals. Proc. Natl. Acad. Sci. U. S. A. 93, 14912–14916.
Albert, M.L., Pearce, S.F.A., Francisco, L.M., Sauter, B., Roy, P., Silverstein, R.L., and Bhardwaj, N. (1998). Immature Dendritic Cells Phagocytose Apoptotic Cells via αvβ5 and CD36, and Cross-present Antigens to Cytotoxic T Lymphocytes. J. Exp. Med. 188, 1359–1368.
Allen, N.J., and Barres, B. a (2005). Signaling between glia and neurons: focus on synaptic plasticity. Curr. Opin. Neurobiol. 15, 542–548.
Altenhein, B., Becker, A., Busold, C., Beckmann, B., Hoheisel, J.D., and Technau, G.M. (2006). Expression profiling of glial genes during Drosophila embryogenesis. Dev. Biol. 296, 545–560.
Araque, a, Parpura, V., Sanzgiri, R.P., and Haydon, P.G. (1999). Tripartite synapses: glia, the unacknowledged partner. Trends Neurosci. 22, 208–215.
Ashburner, M., Chihara, C., Meltzer, P., and Richards, G. (1974). Temporal control of puffing activity in polytene chromosomes. Cold Spring Harb. Symp. Quant. Biol. 38, 655–662.
Auld, V. (1999). Glia as mediators of growth cone guidance: studies from insect nervous systems. Cell. Mol. Life Sci. 55, 1377–1385.
Auld, V.J., Fetter, R.D., Broadie, K., and Goodman, C.S. (1995). Gliotactin, a novel transmembrane protein on peripheral glia, is required to form the blood-nerve barrier in Drosophila. Cell 81, 757–767.
Awasaki, T., and Ito, K. (2004). Engulfing Action of Glial Cells Is Required for Programmed Axon Pruning during Drosophila Metamorphosis. Curr. Biol. 14, 668–677. 161
Awasaki, T., Tatsumi, R., Takahashi, K., Arai, K., Nakanishi, Y., Ueda, R., and Ito, K. (2006). Essential role of the apoptotic cell engulfment genes draper and ced-6 in programmed axon pruning during Drosophila metamorphosis. Neuron 50, 855–867.
Awasaki, T., Lai, S.-L., Ito, K., and Lee, T. (2008). Organization and postembryonic development of glial cells in the adult central brain of Drosophila. J. Neurosci. 28, 13742–13753.
Awasaki, T., Huang, Y., O’Connor, M.B., and Lee, T. (2011). Glia instruct developmental neuronal remodeling through TGF-β signaling. Nat. Neurosci. 14, 821– 823.
Banerjee, S., Pillai, A.M., Paik, R., Li, J., and Bhat, M.A. (2006). Axonal ensheathment and septate junction formation in the peripheral nervous system of Drosophila. J. Neurosci. 26, 3319–3329.
Bansal, A., Singer, J.H., Hwang, B.J., Xu, W., Beaudet, A., and Feller, M.B. (2000). Mice lacking specific nicotinic acetylcholine receptor subunits exhibit dramatically altered spontaneous activity patterns and reveal a limited role for retinal waves in forming ON and OFF circuits in the inner retina. J. Neurosci. 20, 7672–7681.
Barde, Y.A., Edgar, D., and Thoenen, H. (1982). Purification of a new neurotrophic factor from mammalian brain. EMBO J. 1, 549–553.
Bardet, P.-L., Kolahgar, G., Mynett, A., Miguel-Aliaga, I., Briscoe, J., Meier, P., and Vincent, J.-P. (2008). A fluorescent reporter of caspase activity for live imaging. Proc. Natl. Acad. Sci. U. S. A. 105, 13901–13905.
Barres, B. a (2008). The mystery and magic of glia: a perspective on their roles in health and disease. Neuron 60, 430–440.
Barth, M., Hirsch, H. V, Meinertzhagen, I. a, and Heisenberg, M. (1997). Experience- dependent developmental plasticity in the optic lobe of Drosophila melanogaster. J. Neurosci. 17, 1493–1504.
Von Bartheld, C.S., and Bothwell, M. (1993). Development of the mesencephalic nucleus of the trigeminal nerve in chick embryos: target innervation, neurotrophin receptors, and cell death. J. Comp. Neurol. 328, 185–202.
Bastiani, M.J., and Goodman, C.S. (1986). Guidance of neuronal growth cones in the grasshopper embryo. III. Recognition of specific glial pathways. J. Neurosci. 6, 3542– 3551.
162
Battye, R., Stevens, a, and Jacobs, J.R. (1999). Axon repulsion from the midline of the Drosophila CNS requires slit function. Development 126, 2475–2481.
BECKER, H.J. (1959). [The puffs of salivary gland chromosomes of Drosophilia melanogaster. Part 1. Observations on the behavior of a typical puff in the normal strain and in two mutants, giant and lethal giant larvae]. Chromosoma 10, 654–678.
Bishop, D.L., Misgeld, T., Walsh, M.K., Gan, W.-B., Lichtman, J.W., and Louis, S. (2004). Axon Branch Removal at Developing Synapses by Axosome Shedding. Movie 44, 651–661.
Booth, G.E., Kinrade, E.F., and Hidalgo, a (2000). Glia maintain follower neuron survival during Drosophila CNS development. Development 127, 237–244.
Böse, J., Gruber, A.D., Helming, L., Schiebe, S., Wegener, I., Hafner, M., Beales, M., Köntgen, F., and Lengeling, A. (2004). The phosphatidylserine receptor has essential functions during embryogenesis but not in apoptotic cell removal. J. Biol. 3, 15.
Bossing, T., Udolph, G., Doe, C.Q., and Technau, G.M. (1996). The embryonic central nervous system lineages of Drosophila melanogaster. I. Neuroblast lineages derived from the ventral half of the neuroectoderm. Dev. Biol. 179, 41–64.
Boulanger, A., Clouet-Redt, C., Farge, M., Flandre, A., Guignard, T., Fernando, C., Juge, F., and Dura, J.-M. (2011). ftz-f1 and Hr39 opposing roles on EcR expression during Drosophila mushroom body neuron remodeling. Nat. Neurosci. 14, 37–44.
Brown, S., Heinisch, I., Ross, E., Shaw, K., Buckley, C.D., and Savill, J. (2002). Apoptosis disables CD31-mediated cell detachment from phagocytes promoting binding and engulfment. Nature 418, 200–203.
Brugnera, E., Haney, L., Grimsley, C., Lu, M., Walk, S.F., Tosello-Trampont, A.-C., Macara, I.G., Madhani, H., Fink, G.R., and Ravichandran, K.S. (2002). Unconventional Rac-GEF activity is mediated through the Dock180-ELMO complex. Nat. Cell Biol. 4, 574–582.
Bunge, R.P. (1994). The role of the Schwann cell in trophic support and regeneration. J. Neurol. 242, S19–S21.
Bush, T.G., Puvanachandra, N., Horner, C.H., Polito, a, Ostenfeld, T., Svendsen, C.N., Mucke, L., Johnson, M.H., and Sofroniew, M. V (1999). Leukocyte infiltration, neuronal degeneration, and neurite outgrowth after ablation of scar-forming, reactive astrocytes in adult transgenic mice. Neuron 23, 297–308.
163
Cahoy, J.D., Emery, B., Kaushal, A., Foo, L.C., Zamanian, J.L., Christopherson, K.S., Xing, Y., Lubischer, J.L., Krieg, P.A., Krupenko, S.A., et al. (2008). A transcriptome database for astrocytes, neurons, and oligodendrocytes: a new resource for understanding brain development and function. J. Neurosci. 28, 264–278.
Campbell, G., Göring, H., Lin, T., Spana, E., Andersson, S., Doe, C.Q., and Tomlinson, a (1994). RK2, a glial-specific homeodomain protein required for embryonic nerve cord condensation and viability in Drosophila. Development 120, 2957–2966.
Chang, B., Chen, Y., Zhao, Y., and Bruick, R.K. (2007). JMJD6 is a histone arginine demethylase. Science 318, 444–447.
Chang, G.H., Barbaro, N.M., and Pieper, R.O. (2000). Phosphatidylserine-dependent phagocytosis of apoptotic glioma cells by normal human microglia, astrocytes, and glioma cells. Neuro. Oncol. 2, 174–183.
Chen, C., and Regehr, W.G. (2000). Developmental remodeling of the retinogeniculate synapse. Neuron 28, 955–966.
Chen, Y., and Swanson, R.A. (2003). Astrocytes and Brain Injury. 137–149.
Cherbas, L., Hu, X., Zhimulev, I., Belyaeva, E., and Cherbas, P. (2003). EcR isoforms in Drosophila: testing tissue-specific requirements by targeted blockade and rescue. Development 130, 271–284.
Choi, Y.-J., Lee, G., and Park, J.H. (2006). Programmed cell death mechanisms of identifiable peptidergic neurons in Drosophila melanogaster. Dev. Cambridge Engl. 133, 2223–2232.
Christopherson, K.S., Ullian, E.M., Stokes, C.C. a, Mullowney, C.E., Hell, J.W., Agah, A., Lawler, J., Mosher, D.F., Bornstein, P., and Barres, B. a (2005). Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120, 421–433.
Chung, W.-S., Clarke, L.E., Wang, G.X., Stafford, B.K., Sher, A., Chakraborty, C., Joung, J., Foo, L.C., Thompson, A., Chen, C., et al. (2013). Astrocytes mediate synapse elimination through MEGF10 and MERTK pathways. Nature 504, 394–400.
Coutinho-Budd, J., and Freeman, M.R. (2013). Probing the enigma: unraveling glial cell biology in invertebrates. Curr. Opin. Neurobiol. 23, 1073–1079.
Cowan, W.M., Fawcett, J.W., O’Leary, D.D., and Stanfield, B.B. (1984). Regressive events in neurogenesis. Science 225, 1258–1265. 164
Crair, M.C., Horton, J.C., Antonini, A., and Stryker, M.P. (2001). Emergence of ocular dominance columns in cat visual cortex by 2 weeks of age. J. Comp. Neurol. 430, 235– 249.
Cuadros, M. a, and Navascués, J. (1998). The origin and differentiation of microglial cells during development. Prog. Neurobiol. 56, 173–189.
Cunningham, C.L., Martínez-Cerdeño, V., and Noctor, S.C. (2013). Microglia regulate the number of neural precursor cells in the developing cerebral cortex. J. Neurosci. 33, 4216–4233.
Cusack, C.L., Swahari, V., Hampton Henley, W., Michael Ramsey, J., and Deshmukh, M. (2013). Distinct pathways mediate axon degeneration during apoptosis and axon- specific pruning. Nat. Commun. 4, 1876.
Dai, X., Lercher, L.D., Clinton, P.M., Du, Y., Livingston, D.L., Vieira, C., Yang, L., Shen, M.M., and Dreyfus, C.F. (2003). The trophic role of oligodendrocytes in the basal forebrain. J. Neurosci. 23, 5846–5853.
Daneman, R., and Barres, B.A. (2005). The blood-brain barrier--lessons from moody flies. Cell 123, 9–12.
Davis, R.L. (1993). Mushroom bodies and Drosophila learning. Neuron 11, 1–14.
Dearborn, R., and Kunes, S. (2004). An axon scaffold induced by retinal axons directs glia to destinations in the Drosophila optic lobe. Development 131, 2291–2303.
Dekaban, A.S. (1978). Changes in brain weights during the span of human life: relation of brain weights to body heights and body weights. Ann. Neurol. 4, 345–356.
Doherty, J., Logan, M.A., Taşdemir, O.E., and Freeman, M.R. (2009). Ensheathing glia function as phagocytes in the adult Drosophila brain. J. Neurosci. 29, 4768–4781.
Doyle, J.P., Dougherty, J.D., Heiman, M., Schmidt, E.F., Stevens, T.R., Ma, G., Bupp, S., Shrestha, P., Shah, R.D., Doughty, M.L., et al. (2008). Application of a translational profiling approach for the comparative analysis of CNS cell types. Cell 135, 749–762.
Edenfeld, G., Stork, T., and Klämbt, C. (2005). Neuron-glia interaction in the insect nervous system. Curr. Opin. Neurobiol. 15, 34–39.
Van den Eijnde, S.M., Boshart, L., Baehrecke, E.H., De Zeeuw, C.I., Reutelingsperger, C.P., and Vermeij-Keers, C. (1998). Cell surface exposure of phosphatidylserine during apoptosis is phylogenetically conserved. Apoptosis 3, 9–16. 165
Ellis, H.M., and Horvitz, H.R. (1986). Genetic control of programmed cell death in the nematode C. elegans. Cell 44, 817–829.
Ellis, R.E., Jacobson, D.M., and Horvitz, H.R. (1991). Genes required for the engulfment of cell corpses during programmed cell death in Caenorhabditis elegans. Genetics 129, 79–94.
Etchegaray, J.I., Timmons, A.K., Klein, A.P., Pritchett, T.L., Welch, E., Meehan, T.L., Li, C., and McCall, K. (2012). Draper acts through the JNK pathway to control synchronous engulfment of dying germline cells by follicular epithelial cells. Development 139, 4029–4039.
Fadok, V. a, Voelker, D.R., Campbell, P. a, Cohen, J.J., Bratton, D.L., and Henson, P.M. (1992). Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Immunol. 148, 2207–2216.
Fadok, V. a, Bratton, D.L., Frasch, S.C., Warner, M.L., and Henson, P.M. (1998). The role of phosphatidylserine in recognition of apoptotic cells by phagocytes. Cell Death Differ. 5, 551–562.
Fadok, V.A., Bratton, D.L., Rose, D.M., Pearson, A., Ezekewitz, R.A., and Henson, P.M. (2000). A receptor for phosphatidylserine-specific clearance of apoptotic cells. Nature 405, 85–90.
Fadok, V.A., de Cathelineau A, Daleke, D.L., Henson, P.M., and Bratton, D.L. (2001). Loss of phospholipid asymmetry and surface exposure of phosphatidylserine is required for phagocytosis of apoptotic cells by macrophages and fibroblasts. J. Biol. Chem. 276, 1071–1077.
Fan, H., Favero, M., and Vogel, M.W. (2001). Elimination of Bax expression in mice increases cerebellar purkinje cell numbers but not the number of granule cells. J. Comp. Neurol. 436, 82–91.
Fawcett, J.W., O’Leary, D.D., and Cowan, W.M. (1984). Activity and the control of ganglion cell death in the rat retina. Proc. Natl. Acad. Sci. U. S. A. 81, 5589–5593.
Feller, M.B. (2002). The role of nAChR-mediated spontaneous retinal activity in visual system development. J. Neurobiol. 53, 556–567.
Feng, Z., and Ko, C.-P. (2008). Schwann cells promote synaptogenesis at the neuromuscular junction via transforming growth factor-beta1. J. Neurosci. 28, 9599– 9609. 166
Ferri, A., Sanes, J.R., Coleman, M.P., Cunningham, J.M., and Kato, A.C. (2003). Inhibiting axon degeneration and synapse loss attenuates apoptosis and disease progression in a mouse model of motoneuron disease. Curr. Biol. 13, 669–673.
Finn, J.T., Weil, M., Archer, F., Siman, R., Srinivasan, A., and Raff, M.C. (2000). Evidence that Wallerian degeneration and localized axon degeneration induced by local neurotrophin deprivation do not involve caspases. J. Neurosci. 20, 1333–1341.
Flici, H., Erkosar, B., Komonyi, O., Karatas, O.F., Laneve, P., and Giangrande, A. (2011). Gcm/Glide-dependent conversion into glia depends on neural stem cell age, but not on division, triggering a chromatin signature that is conserved in vertebrate glia. Development 138, 4167–4178.
Franc, N.C., Heitzler, P., Ezekowitz, R.A., and White, K. (1999). Requirement for croquemort in phagocytosis of apoptotic cells in Drosophila. Science 284, 1991–1994.
Freeman, M.R. (2006). Sculpting the nervous system: glial control of neuronal development. Curr. Opin. Neurobiol. 16, 119–125.
Freeman, M.R., and Doherty, J. (2006). Glial cell biology in Drosophila and vertebrates. Trends Neurosci. 29, 82–90.
Freeman, M.R., and Rowitch, D.H. (2013). Evolving concepts of gliogenesis: a look way back and ahead to the next 25 years. Neuron 80, 613–623.
Freeman, M.R., Delrow, J., Kim, J., Johnson, E., and Doe, C.Q. (2003). Unwrapping glial biology: Gcm target genes regulating glial development, diversification, and function. Neuron 38, 567–580.
Fuchs, Y., and Steller, H. (2011). Programmed cell death in animal development and disease. Cell 147, 742–758.
Fuentes-Medel, Y., Logan, M. a, Ashley, J., Ataman, B., Budnik, V., and Freeman, M.R. (2009). Glia and muscle sculpt neuromuscular arbors by engulfing destabilized synaptic boutons and shed presynaptic debris. PLoS Biol. 7, e1000184.
Ganetzky, B., and Wu, C.F. (1982). Indirect Suppression Involving Behavioral Mutants with Altered Nerve Excitability in DROSOPHILA MELANOGASTER. Genetics 100, 597–614.
Garden, G.A., and Möller, T. (2006). Microglia biology in health and disease. J. Neuroimmune Pharmacol. 1, 127–137.
167
Gasque, P. (2004). Complement: A unique innate immune sensor for danger signals. Mol. Immunol. 41, 1089–1098.
Ghbeish, N., Tsai, C.C., Schubiger, M., Zhou, J.Y., Evans, R.M., and McKeown, M. (2001). The dual role of ultraspiracle, the Drosophila retinoid X receptor, in the ecdysone response. Proc. Natl. Acad. Sci. U. S. A. 98, 3867–3872.
Gilmour, D.T., Maischein, H.-M., and Nüsslein-Volhard, C. (2002). Migration and function of a glial subtype in the vertebrate peripheral nervous system. Neuron 34, 577– 588.
Glass, J.D., Brushart, T.M., George, E.B., and Griffin, J.W. (1993). Prolonged survival of transected nerve fibres in C57BL/Ola mice is an intrinsic characteristic of the axon. J. Neurocytol. 22, 311–321.
Van Goethem, E., Silva, E.A., Xiao, H., and Franc, N.C. (2012). The Drosophila TRPP cation channel, PKD2 and Dmel/Ced-12 act in genetically distinct pathways during apoptotic cell clearance. PLoS One 7, e31488.
Gómez-Gonzalo, M., Losi, G., Chiavegato, A., Zonta, M., Cammarota, M., Brondi, M., Vetri, F., Uva, L., Pozzan, T., de Curtis, M., et al. (2010). An excitatory loop with astrocytes contributes to drive neurons to seizure threshold. PLoS Biol. 8.
Gong, W.J., and Golic, K.G. (2004). Genomic deletions of the Drosophila melanogaster Hsp70 genes. Genetics 168, 1467–1476.
Gumienny, T.L., Brugnera, E., Tosello-Trampont, A.C., Kinchen, J.M., Haney, L.B., Nishiwaki, K., Walk, S.F., Nemergut, M.E., Macara, I.G., Francis, R., et al. (2001). CED- 12/ELMO, a novel member of the CrkII/Dock180/Rac pathway, is required for phagocytosis and cell migration. Cell 107, 27–41.
Halter, D. a, Urban, J., Rickert, C., Ner, S.S., Ito, K., Travers, a a, and Technau, G.M. (1995). The homeobox gene repo is required for the differentiation and maintenance of glia function in the embryonic nervous system of Drosophila melanogaster. Development 121, 317–332.
Van Ham, T.J., Mapes, J., Kokel, D., and Peterson, R.T. (2010). Live imaging of apoptotic cells in zebrafish. FASEB J. 24, 4336–4342.
Van Ham, T.J., Kokel, D., and Peterson, R.T. (2012). Apoptotic cells are cleared by directional migration and elmo1-dependent macrophage engulfment. Curr. Biol. 22, 830– 836. 168
HAMBURGER, V., and LEVI-MONTALCINI, R. (1949). Proliferation, differentiation and degeneration in the spinal ganglia of the chick embryo under normal and experimental conditions. J. Exp. Zool. 111, 457–501.
Han, C., Song, Y., Xiao, H., Wang, D., Franc, N.C., Jan, L.Y., and Jan, Y.-N. (2014). Epidermal Cells Are the Primary Phagocytes in the Fragmentation and Clearance of Degenerating Dendrites in Drosophila. Neuron 1–17.
Hanayama, R., Tanaka, M., Miwa, K., Shinohara, A., Iwamatsu, A., and Nagata, S. (2002). Identification of a factor that links apoptotic cells to phagocytes. Nature 417, 182–187.
Hazrati, L.-N., Kleinschmidt-DeMasters, B.K., Handler, M.H., Smith, M. Lou, Ochi, A., Otsubo, H., Rutka, J.T., Go, C., Weiss, S., and Hawkins, C.E. (2008). Astrocytic inclusions in epilepsy: expanding the spectrum of filaminopathies. J. Neuropathol. Exp. Neurol. 67, 669–676.
Hedgecock, E.M., Sulston, J.E., and Thomson, J.N. (1983). Mutations affecting programmed cell deaths in the nematode Caenorhabditis elegans. Science 220, 1277– 1279.
Heiman, M., Schaefer, A., Gong, S., Peterson, J.D., Day, M., Ramsey, K.E., Suárez- Fariñas, M., Schwarz, C., Stephan, D. a, Surmeier, D.J., et al. (2008). A translational profiling approach for the molecular characterization of CNS cell types. Cell 135, 738– 748.
Heisenberg, M., Borst, A., Wagner, S., and Byers, D. (1985). Drosophila mushroom body mutants are deficient in olfactory learning. J. Neurogenet. 2, 1–30.
Hidalgo, a, and Booth, G.E. (2000). Glia dictate pioneer axon trajectories in the Drosophila embryonic CNS. Development 127, 393–402.
Hidalgo, A., Kinrade, E.F., and Georgiou, M. (2001). The Drosophila neuregulin vein maintains glial survival during axon guidance in the CNS. Dev. Cell 1, 679–690.
Hiesinger, P.R., Zhai, R.G., Zhou, Y., Koh, T.-W., Mehta, S.Q., Schulze, K.L., Cao, Y., Verstreken, P., Clandinin, T.R., Fischbach, K.-F., et al. (2006). Activity-independent prespecification of synaptic partners in the visual map of Drosophila. Curr. Biol. 16, 1835–1843.
Higashijima, S., Shishido, E., Matsuzaki, M., and Saigo, K. (1996). eagle, a member of the steroid receptor gene superfamily, is expressed in a subset of neuroblasts and regulates the fate of their putative progeny in the Drosophila CNS. Development 122, 527–536. 169
Hindges, R., McLaughlin, T., Genoud, N., Henkemeyer, M., and O’Leary, D.D.M. (2002). EphB forward signaling controls directional branch extension and arborization required for dorsal-ventral retinotopic mapping. Neuron 35, 475–487.
Hoeppner, D.J., Hengartner, M.O., and Schnabel, R. (2001). Engulfment genes cooperate with ced-3 to promote cell death in Caenorhabditis elegans. Nature 412, 202–206.
Hoffmann, P.R., DeCathelineau, A.M., Ogden, C.A., Leverrier, Y., Bratton, D.L., Daleke, D.L., Ridley, A.J., Fadok, V.A., and Henson, P.M. (2001). Phosphatidylserine (PS) induces PS receptor-mediated macropinocytosis and promotes clearance of apoptotic cells. J. Cell Biol. 155, 649–659.
Hohn, A., Leibrock, J., Bailey, K., and Barde, Y.A. (1990). Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, 339–341.
Hollyday, M., and Hamburger, V. (1976). Reduction of the naturally occurring motor neuron loss by enlargement of the periphery. J. Comp. Neurol. 170, 311–320.
Hooks, B.M., and Chen, C. (2006). Distinct roles for spontaneous and visual activity in remodeling of the retinogeniculate synapse. Neuron 52, 281–291.
Hoopfer, E.D., McLaughlin, T., Watts, R.J., Schuldiner, O., O’Leary, D.D.M., and Luo, L. (2006). Wlds protection distinguishes axon degeneration following injury from naturally occurring developmental pruning. Neuron 50, 883–895.
Hosoya, T., Takizawa, K., Nitta, K., and Hotta, Y. (1995). glial cells missing: a binary switch between neuronal and glial determination in Drosophila. Cell 82, 1025–1036.
Hsu, T.-Y., and Wu, Y.-C. (2010). Engulfment of apoptotic cells in C. elegans is mediated by integrin alpha/SRC signaling. Curr. Biol. 20, 477–486.
Hughes, a (1961). Cell degeneration in the larval ventral horn of Xenopus laevis (Daudin). J. Embryol. Exp. Morphol. 9, 269–284.
Ide, C., Tohyama, K., Yokota, R., Nitatori, T., and Onodera, S. (1983). Schwann cell basal lamina and nerve regeneration. Brain Res. 288, 61–75.
Ishimaru, S., Ueda, R., Hinohara, Y., Ohtani, M., and Hanafusa, H. (2004). PVR plays a critical role via JNK activation in thorax closure during Drosophila metamorphosis. EMBO J. 23, 3984–3994.
170
Ito, K., Urban, J., and Technau, G.M. (1995). Distribution, classification, and development ofDrosophila glial cells in the late embryonic and early larval ventral nerve cord. Roux’s Arch. Dev. Biol. 204, 284–307.
Jacobs, J.R., Hiromi, Y., Patel, N.H., and Goodman, C.S. (1989). Lineage, migration, and morphogenesis of longitudinal glia in the Drosophila CNS as revealed by a molecular lineage marker. Neuron 2, 1625–1631.
Jang, A.C.-C., Chang, Y.-C., Bai, J., and Montell, D. (2009). Border-cell migration requires integration of spatial and temporal signals by the BTB protein Abrupt. Nat. Cell Biol. 11, 569–579.
Jefferis, G.S., Marin, E.C., Stocker, R.F., and Luo, L. (2001). Target neuron prespecification in the olfactory map of Drosophila. Nature 414, 204–208.
Jiang, C., Baehrecke, E.H., and Thummel, C.S. (1997). Steroid regulated programmed cell death during Drosophila metamorphosis. Development 124, 4673–4683.
Jones, B.W., Fetter, R.D., Tear, G., and Goodman, C.S. (1995). Glial Cells Missing: a Genetic Switch That Controls Glial Versus Neuronal Fate. Cell 82, 1013–1023.
Kerr, K.S., Fuentes-Medel, Y., Brewer, C., Barria, R., Ashley, J., Abruzzi, K.C., Sheehan, A., Tasdemir-Yilmaz, O.E., Freeman, M.R., and Budnik, V. (2014). Glial Wingless/Wnt Regulates Glutamate Receptor Clustering and Synaptic Physiology at the Drosophila Neuromuscular Junction. J. Neurosci. 34, 2910–2920.
Keshishian, H., Chiba, A., Chang, T.N., Halfon, M.S., Harkins, E.W., Jarecki, J., Wang, L., Anderson, M., Cash, S., and Halpern, M.E. (1993). Cellular mechanisms governing synaptic development in Drosophila melanogaster. J. Neurobiol. 24, 757–787.
Kidd, T., Bland, K.S., and Goodman, C.S. (1999). Slit is the midline repellent for the robo receptor in Drosophila. Cell 96, 785–794.
Kim, J., Jones, B.W., Zock, C., Chen, Z., Wang, H., Goodman, C.S., and Anderson, D.J. (1998). Isolation and characterization of mammalian homologs of the Drosophila gene glial cells missing. Proc. Natl. Acad. Sci. U. S. A. 95, 12364–12369.
Kinchen, J.M., Cabello, J., Klingele, D., Wong, K., Feichtinger, R., Schnabel, H., Schnabel, R., and Hengartner, M.O. (2005). Two pathways converge at CED-10 to mediate actin rearrangement and corpse removal in C. elegans. Nature 434, 93–99.
Kinrade, E.F., Brates, T., Tear, G., and Hidalgo, a (2001). Roundabout signalling, cell contact and trophic support confine longitudinal glia and axons in the Drosophila CNS. Development 128, 207–216. 171
Klämbt, C., Jacobs, J.R., and Goodman, C.S. (1991). The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance. Cell 64, 801–815.
Kocks, C., Cho, J.H., Nehme, N., Ulvila, J., Pearson, A.M., Meister, M., Strom, C., Conto, S.L., Hetru, C., Stuart, L.M., et al. (2005). Eater, a transmembrane protein mediating phagocytosis of bacterial pathogens in Drosophila. Cell 123, 335–346.
Koelle, M.R., Talbot, W.S., Segraves, W.A., Bender, M.T., Cherbas, P., and Hogness, D.S. (1991). The Drosophila EcR gene encodes an ecdysone receptor, a new member of the steroid receptor superfamily. Cell 67, 59–77.
Konishi, M., and Akutagawa, E. (1987). Hormonal control of cell death in a sexually dimorphic song nucleus in the zebra finch. Ciba Found. Symp. 126, 173–185.
Krashes, M.J., and Waddell, S. (2011a). Drosophila appetitive olfactory conditioning. Cold Spring Harb. Protoc. 2011, pdb.prot5609.
Krashes, M.J., and Waddell, S. (2011b). Drosophila aversive olfactory conditioning. Cold Spring Harb. Protoc. 2011, pdb.prot5608.
Kumar, S. (2007). Caspase function in programmed cell death. Cell Death Differ. 14, 32– 43.
Kunisaki, Y., Masuko, S., Noda, M., Inayoshi, A., Sanui, T., Harada, M., Sasazuki, T., and Fukui, Y. (2004). Defective fetal liver erythropoiesis and T lymphopoiesis in mice lacking the phosphatidylserine receptor. Blood 103, 3362–3364.
Kuo, C.T., Jan, L.Y., and Jan, Y.N. (2005). Dendrite-specific remodeling of Drosophila sensory neurons requires matrix metalloproteases, ubiquitin-proteasome, and ecdysone signaling. Proc. Natl. Acad. Sci. U. S. A. 102, 15230–15235.
Kuo, C.T., Zhu, S., Younger, S., Jan, L.Y., and Jan, Y.N. (2006). Identification of E2/E3 ubiquitinating enzymes and caspase activity regulating Drosophila sensory neuron dendrite pruning. Neuron 51, 283–290.
Kuraishi, T., Nakagawa, Y., Nagaosa, K., Hashimoto, Y., Ishimoto, T., Moki, T., Fujita, Y., Nakayama, H., Dohmae, N., Shiratsuchi, A., et al. (2009). Pretaporter, a Drosophila protein serving as a ligand for Draper in the phagocytosis of apoptotic cells. Eur. Mol. Biol. Organ. J. 28, 3868–3878.
Kurant, E. (2011). Keeping the CNS clear: glial phagocytic functions in Drosophila. Glia 59, 1304–1311. 172
Kurant, E., Axelrod, S., Leaman, D., and Gaul, U. (2008). Six-microns-under acts upstream of Draper in the glial phagocytosis of apoptotic neurons. Cell 133, 498–509.
Kurucz, E., Márkus, R., Zsámboki, J., Folkl-Medzihradszky, K., Darula, Z., Vilmos, P., Udvardy, A., Krausz, I., Lukacsovich, T., Gateff, E., et al. (2007). Nimrod, a putative phagocytosis receptor with EGF repeats in Drosophila plasmatocytes. Curr. Biol. 17, 649–654.
Landmesser, L., and Pilar, G. (1972). The onset and development of transmission in the chick ciliary ganglion. J. Physiol. 222, 691–713.
Landmesser, L., and Pilar, G. (1974). Synaptic transmission and cell death during normal ganglionic development. J. Physiol. 241, 737–749.
Lauber, K., Bohn, E., Kröber, S.M., Xiao, Y.J., Blumenthal, S.G., Lindemann, R.K., Marini, P., Wiedig, C., Zobywalski, A., Baksh, S., et al. (2003). Apoptotic cells induce migration of phagocytes via caspase-3-mediated release of a lipid attraction signal. Cell 113, 717–730.
Lee, T., and Luo, L. (1999). Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron 22, 451–461.
Lee, T., and Luo, L. (2001). Mosaic analysis with a repressible cell marker (MARCM) for Drosophila neural development. Trends Neurosci. 24, 251–254.
Lee, G., Wang, Z., Sehgal, R., Chen, C.-H., Kikuno, K., Hay, B., and Park, J.H. (2011). Drosophila caspases involved in developmentally regulated programmed cell death of peptidergic neurons during early metamorphosis. J. Comp. Neurol. 519, 34–48.
Lee, T., Lee, A., and Luo, L. (1999). Development of the Drosophila mushroom bodies: sequential generation of three distinct types of neurons from a neuroblast. Dev. Cambridge Engl. 126, 4065–4076.
Lee, T., Marticke, S., Sung, C., Robinow, S., and Luo, L. (2000). Cell-autonomous requirement of the USP/EcR-B ecdysone receptor for mushroom body neuronal remodeling in Drosophila. Neuron 28, 807–818.
Leiserson, W.M., Harkins, E.W., and Keshishian, H. (2000). Fray, a Drosophila serine/threonine kinase homologous to mammalian PASK, is required for axonal ensheathment. Neuron 28, 793–806.
LeVay, S., Stryker, M.P., and Shatz, C.J. (1978). Ocular dominance columns and their development in layer IV of the cat’s visual cortex: a quantitative study. J. Comp. Neurol. 179, 223–244. 173
LEVI-MONTALCINI, R., and HAMBURGER, V. (1951). Selective growth stimulating effects of mouse sarcoma on the sensory and sympathetic nervous system of the chick embryo. J. Exp. Zool. 116, 321–361.
Li, M.O., Sarkisian, M.R., Mehal, W.Z., Rakic, P., and Flavell, R.A. (2003). Phosphatidylserine receptor is required for clearance of apoptotic cells. Science 302, 1560–1563.
Licciardo, S.D. (2006). Neuron Glia Communications During Axon Scaffold Assembly in Drosophilia (Clark University).
Lilly, M., and Carlson, J. (1990). smellblind: a gene required for Drosophila olfaction. Genetics 124, 293–302.
Liu, Q. a, and Hengartner, M.O. (1998). Candidate adaptor protein CED-6 promotes the engulfment of apoptotic cells in C. elegans. Cell 93, 961–972.
Logan, M.A., Hackett, R., Doherty, J., Sheehan, A., Speese, S.D., and Freeman, M.R. (2012). Negative regulation of glial engulfment activity by Draper terminates glial responses to axon injury. Nat. Neurosci. 15, 1–10.
Luo, L., and O’Leary, D.D.M. (2005). Axon retraction and degeneration in development and disease. Annu. Rev. Neurosci. 28, 127–156.
MacDonald, J.M., Beach, M.G., Porpiglia, E., Sheehan, A.E., Watts, R.J., and Freeman, M.R. (2006). The Drosophila cell corpse engulfment receptor Draper mediates glial clearance of severed axons. Neuron 50, 869–881.
Magnus, T., Chan, A., Savill, J., Toyka, K. V., and Gold, R. (2002). Phagocytotic removal of apoptotic, inflammatory lymphocytes in the central nervous system by microglia and its functional implications. J. Neuroimmunol. 130, 1–9.
Maisonpierre, P.C., Belluscio, L., Squinto, S., Ip, N.Y., Furth, M.E., Lindsay, R.M., and Yancopoulos, G.D. (1990). Neurotrophin-3: a neurotrophic factor related to NGF and BDNF. Science 247, 1446–1451.
Mallat, M., Marín-Teva, J.L., and Chéret, C. (2005). Phagocytosis in the developing CNS: more than clearing the corpses. Curr. Opin. Neurobiol. 15, 101–107.
Manaka, J., Kuraishi, T., Shiratsuchi, A., Nakai, Y., Higashida, H., Henson, P., and Nakanishi, Y. (2004). Draper-mediated and phosphatidylserine-independent phagocytosis of apoptotic cells by Drosophila hemocytes/macrophages. J. Biol. Chem. 279, 48466– 48476. 174
Mariani, J., and Changeux, J.P. (1981). Ontogenesis of olivocerebellar relationships. I. Studies by intracellular recordings of the multiple innervation of Purkinje cells by climbing fibers in the developing rat cerebellum. J. Neurosci. 1, 696–702.
Marín-Teva, J.L., Cuadros, M.A., Calvente, R., Almendros, A., and Navascués, J. (1999). Naturally occurring cell death and migration of microglial precursors in the quail retina during normal development. J. Comp. Neurol. 412, 255–275.
Marín-Teva, J.L., Dusart, I., Colin, C., Gervais, A., van Rooijen, N., and Mallat, M. (2004). Microglia promote the death of developing Purkinje cells. Neuron 41, 535–547.
Martin, S.J., Finucane, D.M., Amarante-Mendes, G.P., O’Brien, G.A., and Green, D.R. (1996). Phosphatidylserine externalization during CD95-induced apoptosis of cells and cytoplasts requires ICE/CED-3 protease activity. J. Biol. Chem. 271, 28753–28756.
Maynard, T.M., Wakamatsu, Y., and Weston, J.A. (2000). Cell interactions within nascent neural crest cell populations transiently promote death of neurogenic precursors. Development 127, 4561–4572.
Motoyama, N., Wang, F., Roth, K.A., Sawa, H., Nakayama, K., Negishi, I., Senju, S., Zhang, Q., and Fujii, S. (1995). Massive cell death of immature hematopoietic cells and neurons in Bcl-x-deficient mice. Science 267, 1506–1510.
Nagaosa, K., Okada, R., Nonaka, S., Takeuchi, K., Fujita, Y., Miyasaka, T., Manaka, J., Ando, I., and Nakanishi, Y. (2011). Integrin βν-mediated phagocytosis of apoptotic cells in Drosophila embryos. J. Biol. Chem. 286, 25770–25777.
Nagata, S., Hanayama, R., and Kawane, K. (2010). Autoimmunity and the clearance of dead cells. Cell 140, 619–630.
Nakamura, H., and O’Leary, D.D. (1989). Inaccuracies in initial growth and arborization of chick retinotectal axons followed by course corrections and axon remodeling to develop topographic order. J. Neurosci. 9, 3776–3795.
Nave, K.-A. (2010). Myelination and support of axonal integrity by glia. Nature 468, 244–252.
Nguyen, J. V, Soto, I., Kim, K.-Y., Bushong, E.A., Oglesby, E., Valiente-Soriano, F.J., Yang, Z., Davis, C.O., Bedont, J.L., Son, J.L., et al. (2011). Myelination transition zone astrocytes are constitutively phagocytic and have synuclein dependent reactivity in glaucoma. Proc. Natl. Acad. Sci. U. S. A. 108, 1176–1181.
175
Nickols, J.C., Valentine, W., Kanwal, S., and Carter, B.D. (2003). Activation of the transcription factor NF-kappaB in Schwann cells is required for peripheral myelin formation. Nat. Neurosci. 6, 161–167.
Nikolaev, A., McLaughlin, T., O’Leary, D.D.M., and Tessier-Lavigne, M. (2009). APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature 457, 981–989.
Nonaka, S., Nagaosa, K., Mori, T., Shiratsuchi, A., and Nakanishi, Y. (2013). Integrin αPS3/βν-mediated phagocytosis of apoptotic cells and bacteria in Drosophila. J. Biol. Chem. 288, 10374–10380.
O’Leary, D.D., and Koester, S.E. (1993). Development of projection neuron types, axon pathways, and patterned connections of the mammalian cortex. Neuron 10, 991–1006.
O’Leary, D.D.M., and McLaughlin, T. (2005). Mechanisms of retinotopic map development: Ephs, ephrins, and spontaneous correlated retinal activity. Prog. Brain Res. 147, 43–65.
Oppenheim, R.W. (1991). Cell Death During Development Of The Nervous System. Annu. Rev. Neurosci. 14, 453–501.
Palgi, M., Lindström, R., Peränen, J., Piepponen, T.P., Saarma, M., and Heino, T.I. (2009). Evidence that DmMANF is an invertebrate neurotrophic factor supporting dopaminergic neurons. Proc. Natl. Acad. Sci. U. S. A. 106, 2429–2434.
Paolicelli, R.C., Bolasco, G., Pagani, F., Maggi, L., Scianni, M., Panzanelli, P., Giustetto, M., Ferreira, T.A., Guiducci, E., Dumas, L., et al. (2011). Synaptic pruning by microglia is necessary for normal brain development. Science 333, 1456–1458.
Park, D., Tosello-Trampont, A.-C., Elliott, M.R., Lu, M., Haney, L.B., Ma, Z., Klibanov, A.L., Mandell, J.W., and Ravichandran, K.S. (2007). BAI1 is an engulfment receptor for apoptotic cells upstream of the ELMO/Dock180/Rac module. Nature 450, 430–434.
Park, S.-Y., Kang, K.-B., Thapa, N., Kim, S.-Y., Lee, S.-J., and Kim, I.-S. (2008). Requirement of adaptor protein GULP during stabilin-2-mediated cell corpse engulfment. J. Biol. Chem. 283, 10593–10600.
Parker, L., Padilla, M., Du, Y., Dong, K., and Tanouye, M. a (2011). Drosophila as a model for epilepsy: bss is a gain-of-function mutation in the para sodium channel gene that leads to seizures. Genetics 187, 523–534.
Parnaik, R., Raff, M., and Scholes, J. (2000). Differences between the clearance of apoptotic cells by professional and non-professional phagocytes. Curr. Biol. 10, 857–860. 176
Pavlidis, P., Ramaswami, M., and Tanouye, M. a (1994). The Drosophila easily shocked gene: a mutation in a phospholipid synthetic pathway causes seizure, neuronal failure, and paralysis. Cell 79, 23–33.
Penn, a. a., Riquelme, P.A., Feller, M.B., and Shatz, C.J. (1998). Competition in retinogeniculate patterning driven by spontaneous activity. Science 279, 2108–2112.
Pereanu, W., Shy, D., and Hartenstein, V. (2005). Morphogenesis and proliferation of the larval brain glia in Drosophila. Dev. Biol. 283, 191–203.
Pereanu, W., Spindler, S., Cruz, L., and Hartenstein, V. (2007). Tracheal development in the Drosophila brain is constrained by glial cells. Dev. Biol. 302, 169–180.
Philips, J.A., Rubin, E.J., and Perrimon, N. (2005). Drosophila RNAi screen reveals CD36 family member required for mycobacterial infection. Science 309, 1251–1253.
Platel, J.-C., Dave, K. a, Gordon, V., Lacar, B., Rubio, M.E., and Bordey, A. (2010). NMDA receptors activated by subventricular zone astrocytic glutamate are critical for neuroblast survival prior to entering a synaptic network. Neuron 65, 859–872.
Rämet, M., Pearson, A., Manfruelli, P., Li, X., Koziel, H., Göbel, V., Chung, E., Krieger, M., and Ezekowitz, R.A. (2001). Drosophila scavenger receptor CI is a pattern recognition receptor for bacteria. Immunity 15, 1027–1038.
Reddien, P.W., and Horvitz, H.R. (2004). The engulfment process of programmed cell death in caenorhabditis elegans. Annu. Rev. Cell Dev. Biol. 20, 193–221.
Reddien, P.W., Cameron, S., and Horvitz, H.R. (2001). Phagocytosis promotes programmed cell death in C. elegans. Nature 412, 198–202.
Ren, Y., and Savill, J. (1998). Apoptosis: the importance of being eaten. Cell Death Differ. 5, 563–568.
Riedl, S.J., and Shi, Y. (2004). Molecular mechanisms of caspase regulation during apoptosis. Nat. Rev. Mol. Cell Biol. 5, 897–907.
Riethmacher, D., Sonnenberg-Riethmacher, E., Brinkmann, V., Yamaai, T., Lewin, G.R., and Birchmeier, C. (1997). Severe neuropathies in mice with targeted mutations in the ErbB3 receptor. Nature 389, 725–730.
Rival, T., Soustelle, L., Strambi, C., Besson, M.T., Iché, M., and Birman, S. (2004). Decreasing glutamate buffering capacity triggers oxidative stress and neuropil degeneration in the Drosophila brain. Curr. Biol. 14, 599–605. 177
Robinow, S., Talbot, W.S., Hogness, D.S., and Truman, J.W. (1993). Programmed cell death in the Drosophila CNS is ecdysone-regulated and coupled with a specific ecdysone receptor isoform. Dev. Cambridge Engl. 119, 1251–1259.
Rogulja-Ortmann, A., Lüer, K., Seibert, J., Rickert, C., and Technau, G.M. (2007). Programmed cell death in the embryonic central nervous system of Drosophila melanogaster. Development 134, 105–116.
Roldán, A., Gogg, S., Ferrini, M., Schillaci, R., and De Nicola, A.F. (1997). Glucocorticoid regulation of in vitro astrocyte phagocytosis. Biocell 21, 83–89.
Rosenthal, A., Goeddel, D. V, Nguyen, T., Lewis, M., Shih, A., Laramee, G.R., Nikolics, K., and Winslow, J.W. (1990). Primary structure and biological activity of a novel human neurotrophic factor. Neuron 4, 767–773.
Roy, B., Singh, A.P., Shetty, C., Chaudhary, V., North, A., Landgraf, M., Vijayraghavan, K., and Rodrigues, V. (2007). Metamorphosis of an identified serotonergic neuron in the Drosophila olfactory system. Neural Dev. 2, 20.
Samsam, M., Mi, W., Wessig, C., Zielasek, J., Toyka, K. V, Coleman, M.P., and Martini, R. (2003). The Wlds mutation delays robust loss of motor and sensory axons in a genetic model for myelin-related axonopathy. J. Neurosci. 23, 2833–2839.
Sauter, B., Albert, M.L., Francisco, L., Larsson, M., Somersan, S., and Bhardwaj, N. (2000). Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J. Exp. Med. 191, 423–434.
Savill, J., Dransfield, I., Gregory, C., and Haslett, C. (2002). A blast from the past: clearance of apoptotic cells regulates immune responses. Nat. Rev. Immunol. 2, 965–975.
Schafer, D.P., Lehrman, E.K., Kautzman, A.G., Koyama, R., Mardinly, A.R., Yamasaki, R., Ransohoff, R.M., Greenberg, M.E., Barres, B. a, and Stevens, B. (2012). Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron 74, 691–705.
Scheib, J.L., Sullivan, C.S., and Carter, B.D. (2012). Jedi-1 and MEGF10 signal engulfment of apoptotic neurons through the tyrosine kinase Syk. J. Neurosci. 32, 13022– 13031.
Schlegel, R. a, and Williamson, P. (2001). Phosphatidylserine, a death knell. Cell Death Differ. 8, 551–563. 178
Schmalbruch, H., Jensen, H.J., Bjaerg, M., Kamieniecka, Z., and Kurland, L. (1991). A new mouse mutant with progressive motor neuronopathy. J. Neuropathol. Exp. Neurol. 50, 192–204.
Schmidt, H., Rickert, C., Bossing, T., Vef, O., Urban, J., and Technau, G.M. (1997). The embryonic central nervous system lineages of Drosophila melanogaster. II. Neuroblast lineages derived from the dorsal part of the neuroectoderm. Dev. Biol. 189, 186–204.
Schoenmann, Z., Assa-Kunik, E., Tiomny, S., Minis, A., Haklai-Topper, L., Arama, E., and Yaron, A. (2010). Axonal degeneration is regulated by the apoptotic machinery or a NAD+-sensitive pathway in insects and mammals. J. Neurosci. 30, 6375–6386.
Schreiber, J., Sock, E., and Wegner, M. (1997). The regulator of early gliogenesis glial cells missing is a transcription factor with a novel type of DNA-binding domain. Proc. Natl. Acad. Sci. U. S. A. 94, 4739–4744.
Schubiger, M., Wade, A.A., Carney, G.E., Truman, J.W., and Bender, M. (1998). Drosophila EcR-B ecdysone receptor isoforms are required for larval molting and for neuron remodeling during metamorphosis. Dev. Cambridge Engl. 125, 2053–2062.
Schubiger, M., Tomita, S., Sung, C., Robinow, S., and Truman, J.W. (2003). Isoform specific control of gene activity in vivo by the Drosophila ecdysone receptor. Mech. Dev. 120, 909–918.
Schwabe, T., Bainton, R.J., Fetter, R.D., Heberlein, U., and Gaul, U. (2005). GPCR signaling is required for blood-brain barrier formation in Drosophila. Cell 123, 133–144.
Seeger, M., Tear, G., Ferres-Marco, D., and Goodman, C.S. (1993). Mutations affecting growth cone guidance in Drosophila: genes necessary for guidance toward or away from the midline. Neuron 10, 409–426.
Sepp, K.J., Schulte, J., and Auld, V.J. (2001). Peripheral glia direct axon guidance across the CNS/PNS transition zone. Dev. Biol. 238, 47–63.
Shatz, C.J., and Stryker, M.P. (1978). Ocular dominance in layer IV of the cat’s visual cortex and the effects of monocular deprivation. J. Physiol. 281, 267–283.
Shen, S., Li, J., and Casaccia-Bonnefil, P. (2005). Histone modifications affect timing of oligodendrocyte progenitor differentiation in the developing rat brain. J. Cell Biol. 169, 577–589.
Shindler, K.S., Latham, C.B., and Roth, K.A. (1997). Bax deficiency prevents the increased cell death of immature neurons in bcl-x-deficient mice. J. Neurosci. 17, 3112– 3119. 179
Shu, T., and Richards, L.J. (2001). Cortical axon guidance by the glial wedge during the development of the corpus callosum. J. Neurosci. 21, 2749–2758.
Simon, D.J., Weimer, R.M., McLaughlin, T., Kallop, D., Stanger, K., Yang, J., O’Leary, D.D.M., Hannoush, R.N., and Tessier-Lavigne, M. (2012). A caspase cascade regulating developmental axon degeneration. J. Neurosci. 32, 17540–17553.
Smits, E., Criekinge, W. Van, Plaetinck, G., and Bogaert, T. (1999). The human homologue of Caenorhabditis elegans CED-6 specifically promotes phagocytosis of apoptotic cells Elke Smits *, Wim Van Criekinge *, Geert Plaetinck and Thierry Bogaert. 1351–1354.
Sofroniew, M. V (2005). Reactive astrocytes in neural repair and protection. Neuroscientist 11, 400–407.
Sofroniew, M. V, and Vinters, H. V (2010). Astrocytes: biology and pathology. Acta Neuropathol. 119, 7–35.
Sokolowski, J.D., Nobles, S.L., Heffron, D.S., Park, D., Ravichandran, K.S., and Mandell, J.W. (2011). Brain-specific angiogenesis inhibitor-1 expression in astrocytes and neurons: Implications for its dual function as an apoptotic engulfment receptor. Brain Behav. Immun. 25, 915–921.
Sonnenfeld, M.J., and Jacobs, J.R. (1995). Macrophages and glia participate in the removal of apoptotic neurons from the Drosophila embryonic nervous system. J. Comp. Neurol. 359, 644–652.
Soustelle, L., Besson, M.-T., Rival, T., and Birman, S. (2002). Terminal glial differentiation involves regulated expression of the excitatory amino acid transporters in the Drosophila embryonic CNS. Dev. Biol. 248, 294–306.
Southwell, D.G., Paredes, M.F., Galvao, R.P., Jones, D.L., Froemke, R.C., Sebe, J.Y., Alfaro-Cervello, C., Tang, Y., Garcia-Verdugo, J.M., Rubenstein, J.L., et al. (2012). Intrinsically determined cell death of developing cortical interneurons. Nature 491, 109– 113.
Sretavan, D.W., and Shatz, C.J. (1986). Prenatal development of retinal ganglion cell axons: segregation into eye-specific layers within the cat’s lateral geniculate nucleus. J. Neurosci. 6, 234–251.
Stacey, S.M., Muraro, N.I., Peco, E., Labbé, A., Thomas, G.B., Baines, R. a, and van Meyel, D.J. (2010). Drosophila glial glutamate transporter Eaat1 is regulated by fringe- mediated notch signaling and is essential for larval locomotion. J. Neurosci. 30, 14446– 14457. 180
Stellwagen, D., and Shatz, C.J. (2002). An instructive role for retinal waves in the development of retinogeniculate connectivity. Neuron 33, 357–367.
Stevens, B. (2008). Neuron-astrocyte signaling in the development and plasticity of neural circuits. Neurosignals. 16, 278–288.
Stevens, B., Allen, N.J., Vazquez, L.E., Howell, G.R., Christopherson, K.S., Nouri, N., Micheva, K.D., Mehalow, A.K., Huberman, A.D., Stafford, B., et al. (2007). The classical complement cascade mediates CNS synapse elimination. Cell 131, 1164–1178.
Stork, T., Engelen, D., Krudewig, A., Silies, M., Bainton, R.J., and Klämbt, C. (2008). Organization and function of the blood-brain barrier in Drosophila. J. Neurosci. 28, 587– 597.
Su, H.P., Nakada-Tsukui, K., Tosello-Trampont, A.-C., Li, Y., Bu, G., Henson, P.M., and Ravichandran, K.S. (2002). Interaction of CED-6/GULP, an adapter protein involved in engulfment of apoptotic cells with CED-1 and CD91/low density lipoprotein receptor- related protein (LRP). J. Biol. Chem. 277, 11772–11779.
Sur, M., Humphrey, A.L., and Sherman, S.M. (1982). Monocular deprivation affects X- and Y-cell retinogeniculate terminations in cats. Nature 300, 183–185.
Sutherland, J.D., Kozlova, T., Tzertzinis, G., and Kafatos, F.C. (1995). Drosophila hormone receptor 38: a second partner for Drosophila USP suggests an unexpected role for nuclear receptors of the nerve growth factor-induced protein B type. Proc. Natl. Acad. Sci. U. S. A. 92, 7966–7970.
Talbot, W.S., Swyryd, E. a, and Hogness, D.S. (1993). Drosophila tissues with different metamorphic responses to ecdysone express different ecdysone receptor isoforms. Cell 73, 1323–1337.
Tansey, F. a, and Cammer, W. (1998). Differential uptake of dextran beads by astrocytes, macrophages and oligodendrocytes in mixed glial-cell cultures from brains of neonatal rats. Neurosci. Lett. 248, 159–162.
Tao, J., Wu, H., Lin, Q., Wei, W., Lu, X.-H., Cantle, J.P., Ao, Y., Olsen, R.W., Yang, X.W., Mody, I., et al. (2011). Deletion of astroglial Dicer causes non-cell-autonomous neuronal dysfunction and degeneration. J. Neurosci. 31, 8306–8319.
Thomaidou, D., Mione, M.C., Cavanagh, J.F., and Parnavelas, J.G. (1997). Apoptosis and its relation to the cell cycle in the developing cerebral cortex. J. Neurosci. 17, 1075–1085.
181
Thomas, A., Lee, P.-J., Dalton, J.E., Nomie, K.J., Stoica, L., Costa-Mattioli, M., Chang, P., Nuzhdin, S., Arbeitman, M.N., and Dierick, H.A. (2012). A versatile method for cell- specific profiling of translated mRNAs in Drosophila. PLoS One 7, e40276.
Thummel, C.S. (1996). Flies on steroids--Drosophila metamorphosis and the mechanisms of steroid hormone action. Trends Genet. 12, 306–310.
Torigoe, K., Tanaka, H.F., Takahashi, A., Awaya, A., and Hashimoto, K. (1996). Basic behavior of migratory Schwann cells in peripheral nerve regeneration. Exp. Neurol. 137, 301–308.
Truman, J. (1990). Metamorphosis of the central nervous system of Drosophila. J. Neurobiol. 21, 1072–1084.
Truman, J.W., Talbot, W.S., Fahrbach, S.E., and Hogness, D.S. (1994). Ecdysone receptor expression in the CNS correlates with stage-specific responses to ecdysteroids during Drosophila and Manduca development. Dev. Cambridge Engl. 120, 219–234.
Tung, T.T., Nagaosa, K., Fujita, Y., Kita, A., Mori, H., Okada, R., Nonaka, S., and Nakanishi, Y. (2013). Phosphatidylserine recognition and induction of apoptotic cell clearance by Drosophila engulfment receptor Draper. J. Biochem. 153, 483–491.
Ullian, E.M., Sapperstein, S.K., Christopherson, K.S., and Barres, B. a (2001). Control of synapse number by glia. Science 291, 657–661.
Venegas, V., and Zhou, Z. (2007). Two alternative mechanisms that regulate the presentation of apoptotic cell engulfment signal in Caenorhabditis elegans. Mol. Biol. Cell 18, 3180–3192.
Vincent, S., Vonesch, J.L., and Giangrande, A. (1996). Glide directs glial fate commitment and cell fate switch between neurones and glia. Development 122, 131–139.
Visanji, N.P., Wong, J.C., Wang, S.X., Cappel, B., Kleinschmidt-Demasters, B.K., Handler, M.H., Ochi, A., Otsubo, H., Rutka, J.T., Go, C., et al. (2012). A proteomic analysis of pediatric seizure cases associated with astrocytic inclusions. Epilepsia 53, e50–4.
Waller, a. (1850). Experiments on the Section of the Glossopharyngeal and Hypoglossal Nerves of the Frog, and Observations of the Alterations Produced Thereby in the Structure of Their Primitive Fibres. Philos. Trans. R. Soc. London 140, 423–429.
Walsh, M.K., and Lichtman, J.W. (2003). In vivo time-lapse imaging of synaptic takeover associated with naturally occurring synapse elimination. Neuron 37, 67–73. 182
Wang, C., and Youle, R.J. (2009). The role of mitochondria in apoptosis*. Annu. Rev. Genet. 43, 95–118.
Wang, X., Wu, Y.-C., Fadok, V.A., Lee, M.-C., Gengyo-Ando, K., Cheng, L.-C., Ledwich, D., Hsu, P.-K., Chen, J.-Y., Chou, B.-K., et al. (2003). Cell corpse engulfment mediated by C. elegans phosphatidylserine receptor through CED-5 and CED-12. Science (80-. ). 302, 1563–1566.
Watts, R.J., Hoopfer, E.D., and Luo, L. (2003). Axon pruning during Drosophila metamorphosis: evidence for local degeneration and requirement of the ubiquitin- proteasome system. Neuron 38, 871–885.
Watts, R.J., Schuldiner, O., Perrino, J., Larsen, C., and Luo, L. (2004). Glia engulf degenerating axons during developmental axon pruning. Curr. Biol. 14, 678–684.
Weimann, J.M., Zhang, Y.A., Levin, M.E., Devine, W.P., Brûlet, P., and McConnell, S.K. (1999). Cortical neurons require Otx1 for the refinement of exuberant axonal projections to subcortical targets. Neuron 24, 819–831.
Weissman, A. (2001). Themes and variations on ubiquitylation. Nat. Rev. Mol. Cell Biol. 2, 169–178.
White, F. a, Keller-Peck, C.R., Knudson, C.M., Korsmeyer, S.J., and Snider, W.D. (1998). Widespread elimination of naturally occurring neuronal death in Bax-deficient mice. J. Neurosci. 18, 1428–1439.
White, K.P., Hurban, P., Watanabe, T., and Hogness, D.S. (1997). Coordination of Drosophila metamorphosis by two ecdysone-induced nuclear receptors. Science 276, 114–117.
Wiesel, T., and Hubel, D. (1963). Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 26, 1003–1017.
Williams, D.W., Kondo, S., Krzyzanowska, A., Hiromi, Y., and Truman, J.W. (2006). Local caspase activity directs engulfment of dendrites during pruning. Nat. Neurosci. 9, 1234–1236.
De Wit, J., Hong, W., Luo, L., and Ghosh, A. (2011). Role of leucine-rich repeat proteins in the development and function of neural circuits. Annu. Rev. Cell Dev. Biol. 27, 697– 729.
Woldeyesus, M.T., Britsch, S., Riethmacher, D., Xu, L., Sonnenberg-Riethmacher, E., Abou-Rebyeh, F., Harvey, R., Caroni, P., and Birchmeier, C. (1999). Peripheral nervous 183
system defects in erbB2 mutants following genetic rescue of heart development. Genes Dev. 13, 2538–2548.
Wu, H.-H., Bellmunt, E., Scheib, J.L., Venegas, V., Burkert, C., Reichardt, L.F., Zhou, Z., Fariñas, I., and Carter, B.D. (2009). Glial precursors clear sensory neuron corpses during development via Jedi-1, an engulfment receptor. Nat. Neurosci. 12, 1534–1541.
Wu, Y., Singh, S., Georgescu, M.-M., and Birge, R.B. (2005). A role for Mer tyrosine kinase in alphavbeta5 integrin-mediated phagocytosis of apoptotic cells. J. Cell Sci. 118, 539–553.
Wyss-Coray, T., Loike, J.D., Brionne, T.C., Lu, E., Anankov, R., Yan, F., Silverstein, S.C., and Husemann, J. (2003). Adult mouse astrocytes degrade amyloid-beta in vitro and in situ. Nat. Med. 9, 453–457.
Xiong, W., and Montell, C. (1995). Defective Glia Induce Neuronal Apoptosis in the. 14, 581–590.
Xiong, W.C., Okano, H., Patel, N.H., Blendy, J. a, and Montell, C. (1994). repo encodes a glial-specific homeo domain protein required in the Drosophila nervous system. Genes Dev. 8, 981–994.
Yao, T.P., Forman, B.M., Jiang, Z., Cherbas, L., Chen, J.D., McKeown, M., Cherbas, P., and Evans, R.M. (1993). Functional ecdysone receptor is the product of EcR and Ultraspiracle genes. Nature 366, 476–479.
Yu, X., Lu, N., and Zhou, Z. (2008). Phagocytic Receptor CED-1 Initiates a Signaling Pathway for Degrading Engulfed Apoptotic Cells. PLoS Biol. 6, 20.
Yuan, J.Y., and Horvitz, H.R. (1990). The Caenorhabditis elegans genes ced-3 and ced-4 act cell autonomously to cause programmed cell death. Dev. Biol. 138, 33–41.
Yuan, L.L., and Ganetzky, B. (1999). A glial-neuronal signaling pathway revealed by mutations in a neurexin-related protein. Science 283, 1343–1345.
Yuasa, Y. (2003). Drosophila homeodomain protein REPO controls glial differentiation by cooperating with ETS and BTB transcription factors. Development 130, 2419–2428.
Zhai, Q., Wang, J., Kim, A., Liu, Q., Watts, R., Hoopfer, E., Mitchison, T., Luo, L., and He, Z. (2003). Involvement of the ubiquitin-proteasome system in the early stages of Wallerian degeneration. Neuron 39, 217–225.
Zhang, L.I., and Poo, M.M. (2001). Electrical activity and development of neural circuits. Nat. Neurosci. 4 Suppl, 1207–1214. 184
Zheng, X., Wang, J., Haerry, T.E., Wu, A.Y.-H., Martin, J., O’Connor, M.B., Lee, C.- H.J., and Lee, T. (2003). TGF-beta signaling activates steroid hormone receptor expression during neuronal remodeling in the Drosophila brain. Cell 112, 303–315.
Zhou, Z., Caron, E., Hartwieg, E., Hall, a, and Horvitz, H.R. (2001a). The C. elegans PH domain protein CED-12 regulates cytoskeletal reorganization via a Rho/Rac GTPase signaling pathway. Dev. Cell 1, 477–489.
Zhou, Z., Hartwieg, E., and Horvitz, H.R. (2001b). CED-1 is a transmembrane receptor that mediates cell corpse engulfment in C. elegans. Cell 104, 43–56.
Ziegenfuss, J.S., Biswas, R., Avery, M. a, Hong, K., Sheehan, A.E., Yeung, Y.-G., Stanley, E.R., and Freeman, M.R. (2008). Draper-dependent glial phagocytic activity is mediated by Src and Syk family kinase signalling. Nature 453, 935–939.
Ziegenfuss, J.S., Doherty, J., and Freeman, M.R. (2012). Distinct molecular pathways mediate glial activation and engulfment of axonal debris after axotomy. Nat. Neurosci. 15, 979–987.