University of Southern California Dissertations and Theses

Synaptic plasticity during neurodegeneration and axonal injury

(USC Thesis Other)

Synaptic plasticity during neurodegeneration and axonal injury

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Synaptic Plasticity During Neurodegeneration and Axonal Injury
by
Nancy Tran
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(NEUROSCIENCE)
May 2024
Copyright 2024 Nancy Tran

ii
Acknowledgements
I am forever grateful for my mentors, family, friends, and colleagues for their support and
guidance these last 6 years. This journey hasn’t been the easiest, but I am happy with how it
ended. None of this would have been possible without them.
I would first like to express gratitude for my advisor, Dr. Dion Dickman, and my
committee members: Drs. Karen Chang, Bruce Herring, Chien-Ping Ko, and Sean Curran.
Thank you to Dion for taking me in when I knew very little about neuroscience, and for letting
me work on the exploratory neurodegeneration and injury projects. Thanks also for giving me
space to become the best instructor and mentor to my undergraduates. This time allowed me to
realize exactly what I appreciate about science and what I should be fighting for. Finally, thank
you to my entire committee for their suggestions as I navigated my challenging projects.
As the first person in my family and one of the first in my friend group to pursue a
PhD, I wanted to give a heartfelt “Thank you!” to my family and friends. Through my postbaccalaureate, my Masters, and now my PhD, my family has always stood behind me. Even
though no one has a scientific background, they were happy to hear about my work and politely
nod along. I’m happy to say that after years of hearing “So, when do you get to graduate?” and
“You’ve been in school for so long; are you almost done?”, I can finally answer “I’m ready to
graduate, finish school, and get a real job!”. I am proud to be the first scientist in my family. But I
am even prouder to be a role model for my youngest niece who is now discovering the joys of
science. To my old Arizona friends, I appreciate you for keeping me sane and for being my
sounding board throughout the years. It was immensely helpful to talk about my experiences as
a graduate student, the trials and tribulations of research, and my general thoughts and feelings
with you. Our weekly Zoom hangouts during the pandemic were so important to me when
everything seemed to be falling apart in the world. Thank you also for joining my Zoom workouts
during lockdown – I’m glad that we could be a ‘fit family’ during those weird times. These ladies

iii
are the best! Lastly, a huge thank you to my little family here in LA. My biggest supporter was
my fiancé, James Jensen, who kept everything running smoothly at home when research and
teaching unfortunately kept me away. Thanks for the endless patience and validation during my
rant sessions. Also, thanks for being my food adventure buddy as we ate our way through LA.
He is the most amazing dog dad to our furry kids: Squirt, Dory (RIP), and Bruce – the best dogs
to ever grace this earth. I love you all.
I became more interested in science communication while working as a Teaching
Assistant (TA) and while mentoring my undergraduates in the Dickman Lab. So, the next people
I’d like to acknowledge are my fellow TA’s, my former students, and all my mentees. While
teaching came with frustrations and at times felt like it kept me from my research, I wouldn’t
trade in these experiences for more time at the bench. I would like to recognize the following
people for making the long grading nights bearable and for sharing fun stories about their
students: Anna Pushkin, Ilse Flores, Dr. Rory Spence, Aracely Simental-Ramos, Chenyu Yang,
Ariel Cohen, Daniel Hicks, Alicia Quihuis, Carinna Torgerson, and Po-Hsuan Huang. Next, I
want to extend my gratitude to my students. They made it clear that I was making a difference in
their education whenever they had an ‘Aha!’ or ‘Eureka!’ moment during discussion. Thanks for
the fantastic course evaluations at the end of each semester – they were all hilarious and too
kind. And a huge thank you to those that emailed me to let me know that they appreciated my
teaching and efforts. Those messages were greatly needed when I was at the lowest point of
my research.
Another group of students who made everything ‘worth it’ were my mentees. They
allowed me to flex my scientific storytelling skills and lead them through individual projects. They
are amazing women, and I can’t wait to see what they accomplish over the years. My first
mentee, Camille Rougier, worked with me briefly before the pandemic. We unfortunately,
couldn’t maintain this working partnership, but she helped me realize what kind of mentor I
should be. Next, thank you to Isha Sanghvi for her incredible help on my axonal transport defect

iv
project and to Lisa Tchitchkan for her amazing imaging on the STED compartments project.
Their tireless efforts were greatly appreciated, and I was so proud that they were able to present
their research at the Society for Neuroscience meeting. Finally, I am thankful for my “triple S”
team who helped me to push my final projects in the lab: Sara Wu for her work on the
degeneration and injury signaling project, Sarah Relle for her imaging analysis on the unc13A
project, and Sunny Lee for her degeneration and bouton analysis on the unc13A project. I hope
that I guided everyone through their research well and that I imparted a strong sense of
scientific integrity.
I cannot, in good conscience, forget about my colleagues and friends who made the dayto-day bearable. Firstly, a huge thank you to Dr. Sarah Perry for her valuable conversations
about research and teaching at smaller colleges. I’m sad that we only realized our love for
karaoke a few months before she had to move, but I’m glad that we had one last glorious
‘Hurrah!’ before she left. Secondly, I’m lucky to have a supportive workout crew who would join
me for F45 classes throughout the week. Thanks to Jerry Chien, Rebecca Stark, and Joshua
Martinez for being my accountability partners. I wouldn’t be this buff without them. Next, I
wanted to share my gratitude for Elizabeth Loxterkamp. It was so nice to have another person
passionate about teaching in the lab. Also, her stories about her rats never fail to make me
laugh. Last, but certainly not least, I am fortunate to call Jerry Chien my baymate, workout
partner, and good friend. It’s incredibly rare to meet someone who shares a similar research
background and sense of humor. I like to think that our daily banter lightened the mood in the
Dickman Lab. The world isn’t ready for our comedy podcast.

v
Table of Contents
Acknowledgements.....................................................................................................................ii
List of Figures ........................................................................................................................... vii
Abstract......................................................................................................................................ix
Chapter 1: Cells of the neuromuscular junction are acutely tuned to perturbations in neurons ... 1
1.1 Neurons undergoing neurodegeneration or injury attenuate intracellular processes ......... 2
1.2 Muscles sense distress in neurons and scale receptors accordingly................................. 3
1.3 Glia sense perturbations in neurons and become reactive to support or degrade them .... 4
Chapter 2: A Developmental arrest of Drosophila larvae elicits presynaptic depression and
enables prolonged studies of neurodegeneration....................................................................... 6
2.1 Abstract ............................................................................................................................ 7
2.2 Introduction....................................................................................................................... 8
2.3 Results ............................................................................................................................11
2.4 Discussion.......................................................................................................................19
2.5 Materials and Methods ....................................................................................................23
Chapter 3: Axonal transport defects reveal novel glutamate receptor plasticity .........................37
3.1 Abstract ...........................................................................................................................38
3.2 Introduction......................................................................................................................39
3.3 Results ............................................................................................................................42
3.4 Discussion.......................................................................................................................46
3.5 Materials and Methods ....................................................................................................49
Chapter 4: Stathmin is a key player in the Wallenda injury signaling pathway...........................60
4.1 Abstract ...........................................................................................................................61
4.2 Introduction......................................................................................................................62
4.3 Results ............................................................................................................................64
4.4 Discussion.......................................................................................................................67
4.5 Materials and Methods ....................................................................................................70
Chapter 5: Peripheral glia perform diverse roles during degeneration and injury.......................79
5.1 Abstract ...........................................................................................................................80
5.2 Introduction......................................................................................................................81
5.3 Results ............................................................................................................................83
5.4 Discussion.......................................................................................................................87
5.5 Materials and Methods ....................................................................................................89

vi
Chapter 6: Developing a Drosophila model to study loss of Unc13A and its involvement in ALSrelated neurodegeneration ........................................................................................................98
6.1 Abstract ...........................................................................................................................99
6.2 Introduction....................................................................................................................100
6.3 Results ..........................................................................................................................103
6.4 Discussion.....................................................................................................................106
6.5 Materials and Methods ..................................................................................................108
Chapter 7: Conclusion.............................................................................................................117
References .............................................................................................................................123
Appendices .............................................................................................................................140
Appendix A: Nancy’s fly storage notes: Where should I store my flies? ...............................141
Appendix B: Nancy’s Third Instar Larval Dissection Protocol and Notes (for IMAGING) ......144
Appendix C: Nancy’s Confocal Staining Protocol.................................................................152
Appendix D: Nancy’s Mounting Protocol for Confocal Imaging ............................................157

vii
List of Figures
Chapter 2
Figure 1: Synaptic strength remains stable throughout the life of an arrested third instar
larva ......................................................................................................................................28
Figure 2: Progressive synaptic growth and a concomitant accumulation of release sites
at ATI NMJs...........................................................................................................................29
Figure 3: Postsynaptic glutamate receptors accumulate and quantal size increases over
the ATI lifespan .....................................................................................................................30
Figure 4: Synaptic vesicle size remains unchanged across the ATI lifespan..........................32
Figure 5: A potent reduction in neurotransmitter release probability is expressed across
the ATI lifespan .....................................................................................................................33
Figure 6: Extending the larval stage reveals the progression of synaptic retractions in
stathmin mutants ...................................................................................................................34
Supplemental Figure 1: Presynaptic homeostatic potentiation can be induced and
expressed across the ATI lifespan.........................................................................................36
Chapter 3
Figure 1: Presynaptic homeostatic depression occurs normally in an injury signaling
background............................................................................................................................53
Figure 2: Presynaptic motor neurons of hiw + vGlut-OE animals exhibit altered
morphology............................................................................................................................54
Figure 3: Synaptic vesicle markers are diminished at hiw + vGlut-OE presynaptic
terminals................................................................................................................................55
Figure 4: vGlut overexpression exacerbates axonal transport defects in hiw mutants............56
Figure 5: Postsynaptic glutamate receptor abundance is enhanced when apposing
terminals are deprived of synaptic cargo ...............................................................................57
Figure 6: Schematic for proposed synaptic plasticity at synapses experiencing severe
reductions in axonal transport................................................................................................58
Supplemental Figure 1: Synaptic vesicle diameters increase in vGlut-OE animals ................59
Chapter 4
Figure 1: The stathmin mutation is epistatic to the highwire mutation ....................................73
Figure 2: BRP intensities are altered in hiw; stai double mutants...........................................74
Figure 3: vGlut intensities are consistently elevated in hiw; stai double mutants....................75
Figure 4: GluRIIA intensities are consistently decreased in hiw; stai double mutants.............76
Figure 5: GluRIIB intensities are consistently decreased in hiw; stai double mutants.............77
Figure 6: GluRIID intensities are decreased in segment A2 of hiw; stai double mutants ........78

viii
Chapter 5
Figure 1: Glia subtype-specific Gal4 drivers reveal their distinct expression patterns at
the Drosophila neuromuscular junction..................................................................................92
Figure 2: Ablating subperineurial glia ablates all peripheral glia.............................................93
Figure 3: Peripheral glia are less reactive in synapses undergoing enhanced
neurodegeneration ................................................................................................................94
Figure 4: Glia ablation in stathmin mutants reveals that glia may be required for neuronal
clearance...............................................................................................................................95
Figure 5: Glia may perform different functions for each motor neuron subtype ......................96
Figure 6: Peripheral glia are highly reactive at synapses experiencing exuberant growth......97
Chapter 6
Figure 1: Unc13-GFP colocalizes with other active zone proteins and apposes
glutamate receptor clusters .................................................................................................111
Figure 2: Unc13-GFP is a clean reporter for unc13 protein levels........................................112
Figure 3: unc13A mutations alone do not induce neurodegeneration...................................113
Figure 4: unc13A mutation in hTDP43 backgrounds does not enhance
neurodegeneration ..............................................................................................................114
Figure 5: unc13A mutation in TBPH backgrounds does not enhance neurodegeneration....115
Figure 6: stathmin mutation is sufficient to reduce unc13 levels...........................................116

ix
Abstract
Cells in the neuromuscular junction (NMJ) are incredibly resilient and capable of
adapting upon neuronal insult. When synaptic transmission is perturbed, the muscle is able to
sense this and scale its receptors appropriately. This is also true when the neuron is undergoing
injury-related signaling – synapses adapt to maintain a lower homeostatic threshold. Likewise,
glia can sense when neurons are damaged or dying and expedite the process of degeneration
and removal of debris. On the other hand, glia can also support neurons during their regrowth
after axonal injury. Finally, when undergoing neurodegeneration due to disease-like states or
cytoskeletal destabilization, proteins in the axon terminal and within the muscle membrane scale
correspondingly.
In chapter 2, we inhibit development to maintain our animals in the third instar stage.
While normal development allows us to study our animals for only one day, this manipulation
expands the experimental timeline to 28 days. The animal continues to grow until its peak size
on day 17, and then it declines until death. We find that regardless of age and size, synaptic
strength is maintained. Although muscle size, synapse size, and vesicular release sites grow
with the animal, homeostasis is preserved. Ultimately, the key factor is a reduction in release
probability to attenuate what would be excessive transmission. Finally, we show that this
developmental manipulation allows for studies of extended neurodegeneration.
Next, we ask whether Postsynaptic Homeostatic Depression (PHD) can be induced
during injury signaling in chapter 3. To induce PHD, we overexpress the vesicular glutamate
transporter (vGlut) to expand vesicles and increase quantal size. Then, we repeat this
manipulation in a highwire (hiw) mutant background to induce injury signaling and find that all of
the hallmarks of PHD are observed via electrophysiology. Knowing that injury actively alters the
neuron and can cause remodeling in the muscle, we interrogate the system. We first observe
aberrant morphology that is indicative of deficient axonal transport. The subsequent

x
experiments looking at protein intensity at axon terminals and within axons reveals an enhanced
issue with axonal trafficking. As a result, the receptor abundance is greatly increased, which is
unheard of in a synapse undergoing injury. We conclude that although large synaptic vesicles
are produced via vGlut-OE, trafficking is hindered, and only normally-side vesicles can be
released.
Chapter 4 focuses on the interactions between injury signaling and a microtubule
support protein, Stathmin (Stai). Once again, we employ the hiw mutation to induce chronic
injury signaling. Our first experiments with hiw and stai single mutants compared to the double
mutant reveal that Stai likely functions downstream of Hiw, due to the similarities of stai and the
double mutant. Further characterizations of the active zone protein, Bruchpilot, in the latter
conditions shows a reduction in synapses undergoing greater amounts of degeneration.
Analysis of vGlut levels shows an overall increase throughout the animal, which may explain the
general decrease of receptors at the postsynaptic compartment. Therefore, it is likely that injury
signaling within the neuron is enhancing plasticity that stai alone does not.
After looking at the well-studied cells of the NMJ, we turn our attention to the oftenforgotten glia in chapter 5. Since glia can both harm the system by promoting death and
neuronal clearance and help the system by maintaining and assisting with neuronal growth, we
think they are key players during neurodegeneration and injury states. After defining proper
tools for glial-specific manipulations, we develop a method of ablating all glia at the NMJ. We
find that glia are not very active in synapses undergoing neurodegeneration, at least, not at the
timepoint that we study them. Glial ablation studies, however, reveal that more boutons are
preserved without glia, implying that they are phagocytosing neurons earlier in development.
When we dissect the types of neurons at our synapses, we find that specifically Ib boutons are
eaten by glia, whereas glia maintain Is boutons instead. This provides interesting insight into
differential glial function depending on the synapse type at the same target postsynaptic cell.

xi
We complete this chapter with an observation that glia are highly reactive with overgrown
neurons, possibly to maintain the extensive branches.
In chapter 6, we attempt to develop a model to study how loss of TDP43 affects Unc13A
levels and if that promotes neurodegeneration. It is found that another function of TDP43 is to
prevent the inclusion of cryptic exons within specific genes. Cryptic exons cause deleterious
effects since they typically cause frameshifts that result in nonsense-mediated decay. The result
is a stark reduction in protein and either lead to diseases like ALS or increases disease
progression. The mammalian finds hinge on Unc13A specifically, but we do not have antibodies
or tagged constructs targeting that isoform, so we rely on Unc13-GFSTF that labels al Unc13
proteins. We find that our fly line is sufficient for our experiments. Despite the loss of Unc13A
leading to disease in the mammalian study, we do not see an increase in neurodegeneration.
This is also true when we couple our Unc13A mutation with either humanized TDP43 mutant
constructs or mutations in the fly ortholog, TBPH. We did find that stai leads to reductions in
overall Unc13 protein, which is an exciting finding for future studies of Stai and ALS.

1
Chapter 1: Cells of the neuromuscular junction are acutely tuned to
perturbations in neurons

2
1.1 Neurons undergoing neurodegeneration or injury attenuate intracellular processes
Neurons are highly plastic cells that are capable of assessing damage and adapting their
functions to maintain basal levels, even in the face of death. Massive perturbations, like
dysregulation of microtubules, does not outright kill animals despite the lack of structural support
and loss of axonal trafficking. Studies in flies show that loss of Stathmin (Stai), a microtubule
stabilizer, leads to degeneration and reduced axonal transport, but only in more posterior
segments (Duncan, Lytle, Zuniga, & Goldstein, 2013; E. R. Graf, Heerssen, Wright, Davis, &
DiAntonio, 2011). Surprisingly, more anterior segments show very little signs of
neurodegeneration, showing the resiliency of neurons. During ALS disease-like states neuronal
growth is limited, and synaptic transmission is maintained at a lower level, but they maintain
functionality to protect the overall animal (Perry, Han, Das, & Dickman, 2017).
Similar trends are apparent when the neuron is experiencing injury-like conditions due to
signaling as well. Activation of Wnd injury signaling by mutating its regulator, highwire (hiw;
(Asghari Adib, Smithson, & Collins, 2018; Grill, Murphey, & Borgen, 2016)) decreases synaptic
transmission (Collins, Wairkar, Johnson, & DiAntonio, 2006; Goel & Dickman, 2018; Wu,
Wairkar, Collins, & DiAntonio, 2005). hiw mutants also reduce protein expression and axonal
transport (J. Li et al., 2017), leading to a reduction of protein at the axon terminal.
Preservation of neuronal activity is key during injury as these cells are undergoing
regeneration as well as degeneration. Loss of the PHR family of proteins (Phr1/Hiw/RPM-1 in
mammals/flies/worms; (Grill et al., 2016)) leads to the activation of Dual Leucine Zipper
Kinase/Wallenda (DLK/Wnd; (Jin & Zheng, 2019; Xiong et al., 2010)) and the subsequent
activation of c-Jun N-terminal kinase (JNK; (Asghari Adib et al., 2018)). From this point JNK can
both promote regeneration and degeneration, but only in the appropriate axonal segments.
Given the complexity of this signaling pathway, it stands to reason that the neuron decreases
many of its functions to prioritize healing upon injury.

3
1.2 Muscles sense distress in neurons and scale receptors accordingly
In an effort to maintain synaptic strength, the muscle must sense alterations in
transmission and adjust its receptors accordingly (Pozo & Goda, 2010; G. Turrigiano, 2012).
Recently, our lab has published an article showing glutamate receptor downscaling due to
excess glutamate release (Y. Han et al., 2023). Interestingly, only the GluRIIA subtype are able
to homeostatically scale whereas the GluRIIB subtype is indifferent. In an unpublished set of
experiments, we see a bidirectionality in which reduced glutamate release leads to upscaling of
receptors. In the ALS model mentioned previously, receptors upscale to compensate for the
severe reduction in transmission (Perry et al., 2017). The scaling sufficiently restores mini EPSP
amplitude but fails to increase EPSP amplitude to the wild type baseline, due to the nearly
halving of the quantal content. It is clear that while synapses do their best to maintain
homeostasis, sufficient issues in the neuron prevent this.
A clear instance of blocked homeostasis is when our lab aimed to test whether
Presynaptic Homeostatic Potentiation (PHP) can be induced in a neuron experiencing injury
(Goel & Dickman, 2018). As mentioned in the previous sub-chapter, injured neurons reduce
transmission by half. Amazingly, the muscle can sense both the decreased release of glutamate
and the fact that the neuron is experiencing injury. It therefore reduces its receptive field by half
to sustain reduced synaptic strength. The animal appears otherwise normal and healthy,
considering its neurons exist in a limbo state of life and death. Despite the preference to
maintain lower receptor abundance, the muscle is capable of upscaling its receptors when also
faced with hypoinnervation. Thus, the muscle not only senses injury, but also added insults that
jeopardize the strength of the synapses, and perhaps the health of the overall animal. Finally,
while no one has published electrophysiological or GluR staining data on stai mutant animals,
we do show a decrease of the receptive field by half. Given our understanding of the injury
state, it is likely the synapses undergoing neurodegeneration have reduced quantal content and
that the muscles can sense all of this.

4
1.3 Glia sense perturbations in neurons and become reactive to support or degrade them
Peripheral glia are key players at NMJs for the maintenance, clearance, and repair of
neurons. When perisynaptic Schwann Cells (PSCs) in mammals are missing, neurons will grow
toward their muscle, make transient connections, then retract (Feng & Ko, 2008). Already
formed synapses also require these unsung heroes, as glial ablation leads to significant
neuronal retractions (Reddy, Koirala, Sugiura, Herrera, & Ko, 2003).
When neurons are damaged in the periphery, phagocytosis must occur to remove
debris. Microglia, the most well-known phagocytic glia, only exist in the central nervous system.
Therefore, peripheral neurons must rely on PSCs for clearance. Early in NMJ development,
neurons will contact multiple muscle cells, but only one partnership will remain and mature. Glia
can somehow sense “weak” connections between cells, then degrade and remove them
(Bishop, Misgeld, Walsh, Gan, & Lichtman, 2004; Ko & Robitaille, 2015). Beyond sensing weak
activity, PSCs surrounding lesioned neurons recruit actin. Then actomyosin structures constrict
to fragment the injured neuron and speed up clearance (Vaquie et al., 2019). Glia have been
seen invading fly NMJs to clear out neuronal debris and ghost boutons (Fuentes-Medel et al.,
2009).
Finally, glia promote a regenerative process that is unique to the peripheral nervous
system. When muscles are denervated, PSCs act as bridges to guide the growth of motor
neurons for reinnervation (Kang, Tian, Mikesh, Lichtman, & Thompson, 2014; Kang, Tian, &
Thompson, 2003; Kang, Tian, & Thompson, 2019; Y.-J. Son & W. J. Thompson, 1995; Y. J. Son
& W. J. Thompson, 1995). Early work by Son and colleagues show that PSCs surrounding the
damaged neuron send small processes to a nearly, healthy muscle fiber. The neuron from the
healthy NMJ then grows through the glial process to innervate the damaged NMJ. Kang’s oldest
work shows that if multiple motor neurons are damaged, the glia that surrounded them will
connect into one continuous tunnel. They propose that only one of the damaged neurons will
grow through all of the tunnels to reinnervate every muscle. Finally, more recent work by Kang

5
shows how these guiding glia remodel synapses post-injury. Given glia’s role in phagocytosis
and regrowth, it stands to reason that they are important for neurodegeneration and axonal
injury.

6
Chapter 2: A Developmental arrest of Drosophila larvae elicits
presynaptic depression and enables prolonged studies of
neurodegeneration

7
2.1 Abstract
Synapses exhibit an astonishing degree of adaptive plasticity in healthy and disease
states. We have investigated whether synapses also adjust to life stages imposed by novel
developmental programs for which they were never molded by evolution. Under conditions
where Drosophila larvae are terminally arrested, we have characterized synaptic growth,
structure and function at the neuromuscular junction (NMJ). While wild-type larvae transition to
pupae after 5 days, arrested third instar (ATI) larvae persist for 35 days, during which NMJs
exhibit extensive overgrowth in muscle size, presynaptic release sites, and postsynaptic
glutamate receptors. Remarkably, despite this exuberant growth, stable neurotransmission is
maintained throughout the ATI lifespan through a potent homeostatic reduction in presynaptic
neurotransmitter release. Arrest of the larval stage in stathmin mutants also reveals a degree of
progressive instability and neurodegeneration that was not apparent during the typical larval
period. Hence, an adaptive form of presynaptic depression stabilizes neurotransmission during
an extended developmental period of unconstrained synaptic growth. More generally, the ATI
manipulation provides a powerful system for studying neurodegeneration and plasticity across
prolonged developmental timescales.

8
2.2 Introduction
Synapses are confronted with extensive challenges during development, maturation,
and aging yet maintain stable information exchange. The dynamic and massive changes in
synapse growth, pruning and remodeling, coupled with intrinsic adjustments in neuronal
excitability can lead to unstable physiological activity. The resulting imbalances in excitation and
inhibition would propagate within neural circuits to undermine network stability. To adapt to such
challenges, synapses are endowed with the capacity to homeostatically adjust
neurotransmission while still permitting the flexibility necessary for Hebbian forms of plasticity
(Pozo & Goda, 2010; G. Turrigiano, 2012). The homeostatic control of neural activity operates
throughout organismal lifespan to balance the tension between stability and flexibility, and is
thought to break down in neurological and psychiatric diseases (Jie Li, Park, Zhong, & Chen,
2019; Nelson & Valakh, 2015; Wondolowski & Dickman, 2013). Although it is clear synapses
have the capacity to express both Hebbian and homeostatic forms of plasticity, how these
processes are integrated and balanced, particularly during development and aging, remain
enigmatic.
The Drosophila larval neuromuscular junction (NMJ) is an accessible and versatile
model for studying synaptic function, plasticity, and disease. This model glutamatergic synapse
has enabled fundamental insights into synaptic growth, transmission, homeostatic plasticity, and
injury (C Andrew Frank, James, & Müller, 2020; Menon, Carrillo, & Zinn, 2013). However,
studies in this system are limited by the relatively short larval period of 3-4 days before
pupariation, when NMJ accessibility is lost. This short temporal window limits the use of the third
instar larval NMJ as a model for interrogating dynamic processes over chronic time scales.
Intriguingly, recent studies on the signaling cascades in Drosophila that control the transition
from third instar to the pupal stage have revealed attractive targets for extending the duration of
the third instar (Gibbens, Warren, Gilbert, & O'Connor, 2011; Kim F. Rewitz, Yamanaka, Gilbert,
& O’Connor, 2009; Walkiewicz & Stern, 2009).

9
Developmental progression in Drosophila larvae is coordinated through two semiredundant signaling pathways via Torso and insulin-like receptors that ultimately lead to
ecdysone synthesis and release from the prothoracic gland (PG) to drive the transition from the
larval stage to (K. F. Rewitz, Yamanaka, Gilbert, & O'Connor, 2009; Walkiewicz & Stern, 2009;
Yamanaka, Rewitz, & O'Connor, 2013). A previous study reduced signaling through one arm of
this pathway to extend the third instar stage from 5 to 9 days, where the important observation
that NMJs continue to grow and function throughout this period was made (Miller, Ballard, &
Ganetzky, 2012). More recent work has demonstrated that loss of key transcription factors in
the PG, including Smox (dSMAD2), can disrupt both signaling pathways to fully arrest larval
development and prevent the transition to pupal stages (Gibbens et al., 2011; Ohhara,
Kobayashi, & Yamanaka, 2017) Remarkably, these arrested third instars (ATI) remain in the
larval stage until death. The development of ATI larvae now provides an opportunity to
characterize synaptic growth, function, and plasticity in a system of terminally persistent
expansion beyond normal physiological ranges and has the potential to reveal new insights into
processes such as neurodegeneration.
Here, we have developed an optimized approach to arrest Drosophila larvae at third
instar stages to characterize NMJ growth, function, and plasticity. We find that ATI larvae
continue to grow and survive for up to 35 days, where NMJs exhibit exuberant expansion in
both pre- and post-synaptic compartments. Interestingly, this growth should enhance synaptic
strength, yet no significant change is observed compared to baseline values. Instead, a potent
reduction in presynaptic neurotransmitter release maintains stable synaptic strength across the
life of an ATI larva. Finally, the ATI larvae enabled new insights into the progression of
neurodegeneration in stathmin mutants. Together, arresting larval development now provides a
powerful foundation to probe the mechanisms of synaptic growth, function, homeostatic
plasticity, and neurodegeneration at a model glutamatergic synapse in a genetically tractable
system.

10
In this project, I curated data for Fig. 3G and a locomotion assay for the reviewers’
rebuttal. I also analyzed the electron microscopy data in Fig. 4. Finally, I edited the manuscript
figures and text to comply with the journal’s requirements.

11
2.3 Results
Synaptic strength is maintained throughout the lifespan of an arrested third instar larva.
To arrest larval development at the third instar stage, we targeted genes that could
either disrupt both Torso and insulin signaling pathways or broadly inhibit the synthesis of
ecdysone synthesis in the PG, processes ultimately necessary for the transition to pupal stages
(Fig. 1A; (Gibbens et al., 2011; Ohhara et al., 2017)). We reasoned that if we could prevent the
release of ecdysone from the PG by knocking down a key transcript(s), pupation would be
delayed indefinitely (Yamanaka et al., 2013). Importantly, knock down of gene expression was
restricted to the PG by expression with phm-Gal4 (Ono et al., 2006), minimizing secondary
effects from possible functions in other tissues. We screened several lines described by other
investigators that prevent larval transitions to pupariation (Danielsen et al., 2016) and found that
a particular RNAi line targeting smox (dSMAD2), was the most effective, reliably preventing
pupation in nearly all animals (Fig. 1A,B). smox encodes a transcription factor required for
expression of both torso and insulin receptor genes in the PG (Gibbens et al., 2011), while also
having functions in a variety of other tissues (Brummel et al., 1999; Hevia, López-Varea,
Esteban, & de Celis, 2017). These developmentally arrested third instars (ATI) persist as larvae
and live up to 35 days after egg lay (AEL). Typical wild type larvae spend ~3 days in the third
instar stage before pupation and metamorphosis, living beyond 60 days AEL as adults (K. F.
Rewitz et al., 2009). For the first 5 days of development, ATI larvae appear largely unchanged
compared to wild type, but they fail to progress to become “wandering” third instars. Rather,
they continue to feed and gain body mass, peaking around 17 days AEL (ATI.17) and then
gradually losing body mass until dying soon after 33 days AEL (ATI.33) (Fig. 1B). For further
experiments, we compared wild type larvae at 5 days AEL (WT.5) to ATI larvae at varying time
points, including 5 days AEL (ATI.5), a time point similar to wild type; 17 days AEL (ATI.17), a
time corresponding to peak body mass; and 33 days AEL (ATI.33), a time near the terminal
stage of the ATI lifespan.

12
To investigate NMJs across the ATI lifespan, we first characterized muscle size and
passive electrical properties of the muscle. We observed a progressive gain in muscle size
across the ATI lifespan, where muscle surface area increased by over 50%, peaking at ATI.17
and then decreasing to ATI.33 (Fig. 1C,D). Consistent with this substantial increase in muscle
size, electrophysiological recordings of NMJs across the ATI lifespan revealed a massive
decrease in input resistance peaking around ATI.17 (Fig. 1D). Remarkably, despite these
changes in muscle size, synaptic strength (EPSP amplitude) remains constant across ATI NMJs
(Fig. 1E,F). Thus, as larvae grow and decline through an arrested third instar lifespan, synaptic
strength at the NMJ remains constant.
Presynaptic compartments at the NMJ progressively expand in ATI larvae.
Clearly, ATI NMJs maintain synaptic strength despite the substantial increase in muscle
size that progresses through arrested larval development. In principle, modulations to the
number of presynaptic release sites (N), the probability of release at each individual release site
(Pr), and/or the postsynaptic response to glutamate release from single synaptic vesicles
(quantal size, (Q)) could stabilize synaptic strength at these NMJs (Dittman & Ryan, 2019). We
first assessed synaptic growth to determine whether the number of presynaptic release sites
increases in proportion to the muscle surface area. During the conventional 3-4 day period of
larval development, there is a 100-fold expansion in the NMJ, with changes to the passive
electrical properties of the muscle and a concomitant growth of pre- and post-synaptic
compartments (Atwood, Govind, & Wu, 1993; Menon et al., 2013; Schuster, Davis, Fetter, &
Goodman, 1996). These changes are thought to scale NMJ function in parallel with growth and
maintain sufficient depolarization for muscle contraction (Davis & Goodman, 1998). However,
the progressive increase in muscle size at ATI NMJs poses a further challenge, where synapses
may need to expand to compensate for overgrowth. We therefore considered whether adaptive
changes in the growth of motor terminals and/or number of synapses served to stabilize

13
synaptic strength (EPSP amplitude). Using immunostaining, we instead found a progressive
enhancement in the neuronal membrane surface area and in the number of boutons per NMJ
throughout the ATI lifespan (Fig. 2A-D). In fact, the bouton to muscle area ratio even overshoots
the scaling that is normally observed at conventional development between first and third instar
larval stages (WT.5: 40 boutons/40,000 µm2
ratio (Schuster et al., 1996); ATI.17: 100
boutons/75,000 µm2
ratio; Table S1). Hence, motor neuron terminals grow in excess to muscle
growth.
Since NMJ boutons expand across ATI stages, we considered the possibility of a
compensatory reduction in the anatomical number of release sites. There is precedence for a
reduction in the density of active zones (AZs), independent of NMJ growth, to maintain synaptic
strength (P. Goel, M. Khan, et al., 2019; Ethan R Graf, Daniels, Burgess, Schwarz, & DiAntonio,
2009) . To identify individual presynaptic release sites, we immunostained NMJs with an
antibody against Bruchpilot (BRP), a central scaffolding protein that constitutes the “T-bar”
structure at AZs in Drosophila (Kittel et al., 2006). Since ~96% of release sites are labeled by
BRP at the fly NMJ (Akbergenova, Cunningham, Zhang, Weiss, & Littleton, 2018; Gratz et al.,
2019; Wagh et al., 2006), we defined an anatomical release site as a BRP punctum and
quantified these structures across the ATI lifespan. Interestingly, we found no significant
changes in BRP puncta density across ATI stages, with total BRP puncta number per NMJ
increasing in proportion to neuronal membrane area and bouton number (Fig. 2A-D; Table S1).
Finally, although the number of BRP puncta increased, the size and fluorescence intensity of
these puncta can be reduced at NMJs to compensate for synaptic overgrowth, reducing Pr at
individual release sites and maintaining overall synaptic strength (P. Goel et al., 2019; P. Goel,
M. Khan, et al., 2019). However, although BRP number at ATI NMJs increases to over threefold that of wild-type NMJs, no compensatory reduction in size and/ or intensity of BRP puncta
was observed (Fig. 2A lower panel; Table S1). Indeed, BRP puncta intensity was significantly
increased compared to WT.5 levels (Table S1), which may reflect the age-dependent increase

14
size and intensity and active zones documented at the fly NMJ (Akbergenova et al., 2018).
Thus, this anatomical analysis reveals an increase in the number of release sites (N) at ATI
NMJs, implying that other adaptations compensate for excessive growth in ATI larvae.
Postsynaptic receptor fields accumulate at the NMJ over the ATI lifespan.
Given the substantial increase in AZ number and intensity but stable synaptic strength,
we next considered the possibility that a reduction in the postsynaptic receptivity to
neurotransmitter (Q) may have offset the observed presynaptic overgrowth at ATI NMJs. One
possibility is that a reduction in the abundance, composition, and/or function of postsynaptic
glutamate receptors (GluRs) may have occurred at ATI NMJs. At the fly NMJ, two receptor
subtypes containing either GluRIIA- or GluRIIB- subunits form complexes with the essential
GluRIIC, GluRIID, and GluRIIE subunits to mediate the postsynaptic currents driving
neurotransmission (Qin et al., 2005). GluRIIA-containing receptors mediate larger current
amplitudes and slower decay kinetics compared to the GluRIIB-containing receptor counterparts
(T. H. Han, Dharkar, Mayer, & Serpe, 2015; Petersen, Fetter, Noordermeer, Goodman, &
DiAntonio, 1997) . We examined the postsynaptic GluRs using antibodies that specifically
recognize the GluRIIA- or GluRIIB- subunits, as well as the common GluRIID subunit (Fig. 3A).
Consistent with presynaptic overgrowth, total GluR puncta numbers per NMJ mirrored the
increase in presynaptic AZ number (Fig. 3B). Similarly, we observed a significant increase in the
abundance of all GluR subunits assessed at ATI NMJs revealed by enhanced fluorescence
intensity (Fig. 3C). Together, this demonstrates that postsynaptic receptor fields progressively
expand in number and abundance, mirroring the accumulation in presynaptic structures across
the ATI lifespan.
We next considered whether an apparent reduction in GluR functionality compensated
for the expansion of glutamate receptor fields at ATI NMJs. We determined GluR functionality
by electrophysiologically recording miniature events at ATI NMJs. Consistent with the increased

15
fluorescence intensity of all subunits, we observed an ~50% increase in mEPSP amplitude
compared to wild-type levels in ATI.17 larvae, an enhancement that persisted through ATI.33
(Fig. 3D,E). Consistent with increased presynaptic growth, we also observed an increase in
mEPSP frequency (Fig. 3F). Along with the imaging data suggesting there is no change in GluR
subunit stoichiometry, we observed no difference in mEPSP decay time constants (Fig. 3G).
Thus, both imaging and electrophysiological data indicate that postsynaptic GluR abundance
increases concurrently with presynaptic overgrowth at ATI NMJs.
In principle, it is possible that a change in the size of synaptic vesicles at ATI NMJs
could also contribute to the enhanced quantal size observed at ATI NMJs. We therefore
quantified synaptic vesicle diameter at NMJs of WT and ATI using electron microscopy (Fig.
4A). We observed no significant difference in the average vesicle diameter at NMJs of ATI
compared to wild type (Fig. 4B). Therefore, the enhanced mEPSP amplitudes of ATI NMJs are
due to an increase in postsynaptic GluR abundance, and yet synaptic strength remains stable
across the ATI lifespan. This suggests a homeostatic adjustment in presynaptic release
probability (Pr) is elicited at ATI NMJs that counteracts both the increased synapse number (N)
and enhanced postsynaptic sensitivity to neurotransmitter (Q). It is unlikely that a retrograde
signal from the muscle served to induce this reduction in Pr, as increased expression of
postsynaptic GluRs at the fly NMJ similarly enhances mEPSP amplitude but does not alter
presynaptic neurotransmitter release (X. Li, Goel, Wondolowski, Paluch, & Dickman, 2018;
Petersen et al., 1997). Thus, a reduction in Pr of sufficient magnitude must be induced at ATI
NMJs to fully counteract the increase in N and Q to maintain stable synaptic strength.
Synaptic strength at ATI NMJs is maintained through a potent homeostatic decrease in release
probability.
ATI larvae exhibit exuberant synaptic growth with accumulations of both pre- and postsynaptic components, resulting in an increased N and Q, factors that should enhance synaptic

16
strength. However, EPSP amplitudes remain stable across the ATI lifespan, implying
presynaptic release probability (Pr) must be substantially and precisely diminished to
compensate. To further test this idea, we calculated quantal content (the number of synaptic
vesicles released per stimulus) and found a substantial reduction at ATI NMJs (Fig. 5A). Next,
we assessed presynaptic function independently of mEPSP amplitude by performing failure
analysis, where repeated stimulations in low extracellular Ca2+ (0.15 mM) fail to elicit a response
in ~50% of trials in wild type. At ATI NMJs, failure rate was markedly increased (Fig. 5B),
consistent with reduced quantal content. Finally, we assayed paired-pulse ratios to gauge Pr. At
low extracellular Ca2+ (0.3 mM), paired-pulse facilitation (PPF) is observed at wild-type NMJs,
while paired-pulse depression (PPD) is found in elevated Ca2+ (1.5 mM) (Böhme et al., 2016; X.
Li, P. Goel, J. Wondolowski, et al., 2018). In ATI.17 and ATI.33 NMJs, PPF was significantly
increased while PPD was reduced, consistent with reduced Pr relative to wild type (Fig. 5C,D). It
is interesting to note that a similar phenomenon has been observed at the Drosophila NMJ in
the context of typical larval development, referred to as presynaptic homeostatic depression
(PHD). Here, mEPSP size is enhanced while quantal content is reduced to maintain normal
EPSP amplitudes (Daniels et al., 2004; Gavino, Ford, Archila, & Davis, 2015; X. Li, P. Goel, J.
Wondolowski, et al., 2018). While it is not clear that the mechanism of depression is shared
between later ATI time points and PHD, we can posit that a homeostatic reduction in
presynaptic release probability compensates for increased quantal size to maintain synaptic
strength across the ATI life span.
It has previously been shown that NMJs expressing PHD can also express other forms
of homeostatic plasticity, including a process referred to as presynaptic homeostatic potentiation
(PHP) (Gavino et al., 2015; P. Goel et al., 2019; X. Li, P. Goel, C. Chen, et al., 2018). To
induce PHP, we applied the postsynaptic GluR antagonist philanthotoxin-343 (PhTx) (C. A.
Frank, Kennedy, Goold, Marek, & Davis, 2006). 10 min incubation in PhTx reduces mEPSP
amplitude in both wild type and ATI NMJs, as expected (Fig. S1). In turn, EPSP amplitude is

17
maintained at baseline levels due to a retrograde, homeostatic increase in presynaptic
neurotransmitter release in wild type (Goel, Li, & Dickman, 2017). Similarly, PHP is robustly
expressed across ATI NMJs (Fig. S1). Thus, like PHD and other forms of homeostatic plasticity
studied at the Drosophila NMJ, the presynaptic inhibition observed at ATI NMJs can be
balanced with acute GluR challenge to express PHP and maintain stable synaptic strength.
Extending the larval stage reveals the progression of axonal degeneration in stathmin mutants.
In our final set of experiments, we considered whether ATI larvae could be utilized as
models for aging and/or neurodegeneration. We hypothesized that NMJs in ATI larvae were
unlikely to exhibit classical hallmarks of aging synapses. Although muscle integrity appears to
degrade in ATI.33 compared to earlier time points (Fig. 1), synaptic growth (Fig. 2), GluR
receptor fields (Fig. 3), and presynaptic function (Fig. 5) all appear similar in ATI.33 relative to
earlier time points. Indeed, while reductions in synaptic components and neurotransmission
have been observed at aging mammalian NMJs (L. Li, W.-C. Xiong, & L. Mei, 2018; Taetzsch &
Valdez, 2018), NMJ structure and function remains surprisingly robust in ATI larvae nearing
death, with no apparent defects in synaptic function or even PHP plasticity. One additional
canonical indicator of aging reported at mammalian NMJs includes presynaptic retractions and
fragmentation (Lei Li et al., 2018; Taetzsch & Valdez, 2018). We therefore assessed synaptic
retractions across the ATI lifespan using an established “footprint” assay, in which a
postsynaptic marker is observed to persist without a corresponding presynaptic marker (Eaton,
Fetter, & Davis, 2002; E. R. Graf et al., 2011; Perry et al., 2017). However, ATI NMJs, including
ATI.33, showed surprisingly stable synapses, with no apparent increases in footprints compared
with earlier time points (Fig. 6A,B). Together, these results indicate that NMJ structure, function,
and integrity remain surprisingly robust across all stages of ATI larvae, even at terminal periods,
and are therefore unlikely to serve as a compelling model for age-related synaptic decline.

18
Although NMJs remain structurally intact and stable across the lifespan of the ATI
larvae, this manipulation does enable a substantially longer time scale compared to the typical 5
days of larval development to investigate insults that contribute to neuronal degeneration. We
chose to characterize NMJ growth and stability in stathmin mutants extended through the ATI
manipulation. Stathmin is a tubulin-associated factor involved in maintaining the integrity of the
axonal cytoskeleton (Duncan et al., 2013; Lachkar et al., 2010). The mammalian homolog of
Drosophila stathmin (SCG10) is highly conserved and is thought to function as a surveillance
factor for axon damage and degenerative signaling (Shin et al., 2012). In Drosophila, loss of
stathmin leads to a marked increase in NMJ footprints, where more posterior segments show
increased severity relative to more anterior segments (E. R. Graf et al., 2011). Surprisingly,
stathmin mutants are still able to pupate and develop into adults. However, stathmin mutants
extended in larval stages by the ATI manipulation die shortly after 21 days AEL. We therefore
sought to use the ATI system to determine the impact of a prolonged phase of axonal instability
in stathmin mutants. Indeed, NMJs exhibit increased footprints in stai.13 (stathmin mutants
extended to ATI.13 time points) when compared to stai.5 controls in both frequency (Fig. 6C,D)
and severity (Fig. 6E), with the most severe retractions observed in posterior abdominal
segments (A3-A5). Finally, we tested whether NMJ growth increased in ATI-extended stathmin
NMJs, as it does in wild type. While control ATI synapses grow in bouton and BRP puncta
number between 5 and 13 days AEL, stathmin NMJs fail to consistently expand (Fig. 6F-H).
These experiments highlight the potential of the ATI system to be a useful tool for defining the
progression of neurodegeneration at the Drosophila NMJ, which is otherwise limited to short
larval stages.

19
2.4 Discussion
By arresting further maturation at third instar Drosophila larvae, we have been able to
accomplish a detailed study of NMJ structure, function, and plasticity over much longer
timescales than previously possible. This ATI larval system has revealed how the NMJ
maintains stable transmission over a vastly extended developmental timescale, where
persistent overgrowth in both pre- and post-synaptic compartments is offset through a potent
and homeostatic reduction in neurotransmitter release. Hence, this study not only provides
evidence for a potentially novel homeostatic signaling system that balances release probability
with synaptic overgrowth but now extends the temporal window to enable the characterization of
a variety of processes, including neurodegeneration, at a powerful model synapse.
As described by Miller et al. (2012), NMJs in third instar larvae that have been
developmentally arrested for at least a week beyond the normal time of pupariation continue to
grow and add new boutons. Here we extend this observation to larvae arrested at the third
instar for over 30 days, further demonstrating that mechanisms do not exist to suppress or
negatively regulate growth when developmental timing is artificially extended. During normal
larval growth from first to third instar, the body wall muscles undergo rapid and immense
expansion, growing nearly 100-fold in surface area within a few days (Menon et al., 2013).
Presynaptic terminals grow and add new boutons in parallel with muscle growth, presumably to
maintain stable NMJ strength. In effect, sufficient levels of muscle excitation is sustained
through a coordinated increase in all three parameters controlling synaptic physiology: N
(number of release sites), Pr (release probability at each site), and Q (quantal size) (Dittman &
Ryan, 2019). Hence, during typical stages of larval development, increasing muscle growth
requires a concomitant elaboration in NMJs, implying robust signaling systems exist to ensure
synaptic size, structure, and function expand in a coordinated manner. This tight structural
coupling between muscle fiber and NMJ growth is also observed in mammals and is thought to
be a primary mechanism for maintaining NMJ strength during post-developmental muscle

20
growth or wasting (Sanes & Lichtman, 1999, 2001). However, when the normal developmental
program is made to continue without terminating in pupariation, NMJ growth continues
apparently unchecked, posing a potential challenge of hyperexcitation. There is emerging
evidence that when NMJ growth is genetically perturbed, a redistribution of active zone material
or adaptations in synapse morphogenesis or postsynaptic neurotransmitter receptors can
maintain stable synaptic strength (P. Goel et al., 2019; P. Goel, M. Khan, et al., 2019; Ethan R
Graf et al., 2009). In the case of NMJ overgrowth in endophilin mutants, a homeostatic scaling
in active zone size compensates for increased number to lower release probability and maintain
stable synaptic strength (P. Goel et al., 2019). However, NMJs in ATI larvae do not appear to
utilize such strategies. Rather, a latent form of adaptive plasticity is revealed at ATI NMJs that is
sufficiently potent and precise to inhibit neurotransmitter release probability and compensate for
the overgrowth of both pre- and post-synaptic compartments.
The presynaptic inhibition of neurotransmitter release that maintains synaptic strength at
ATI NMJs is a potentially novel phenomenon of homeostatic plasticity. This form of presynaptic
depression appears to be an entirely functional change that reduces release probability, without
any apparent adaptations to active zone number, intensity or synaptic structure.
Electrophysiologically, the presynaptic inhibition demonstrated at ATI NMJs resembles
presynaptic homeostatic depression (PHD), a form of homeostatic plasticity characterized at the
Drosophila NMJ in which excess glutamate release induces a compensatory reduction in
release probability that maintains stable synaptic strength (Daniels et al., 2004; Gavino et al.,
2015; X. Li, P. Goel, J. Wondolowski, et al., 2018). Like PHD, the presynaptic inhibition at ATI
NMJs is not reflected in changes to the active zones or synaptic structure (P. Goel et al., 2019;
Gratz et al., 2019; X. Li, P. Goel, J. Wondolowski, et al., 2018). However, the only mechanisms
known to be capable of inducing PHD require enhanced synaptic vesicle size that results from
endocytosis mutants or overexpression of the vesicular glutamate transporter (Daniels et al.,
2004; Dickman, Horne, Meinertzhagen, & Schwarz, 2005b; P. Goel et al., 2019; Verstreken et

21
al., 2002). We found no evidence for changes in synaptic vesicle size at ATI NMJs, as the
enhanced postsynaptic glutamate receptor levels observed are sufficient to explain the
increased quantal size (Fig. 3) and direct measurement of vesicle size by EM revealed no
significant change from wild type (Fig. 4). Hence, if the homeostatic depression observed at ATI
NMJs is ultimately the same plasticity mechanism as PHD, then this would be the first condition
that does not require enlarged synaptic vesicle size. In this case, perhaps excess global
glutamate release from increased release sites at ATI NMJs induces the same homeostatic
plasticity that increased glutamate released from individual synaptic vesicles does. This would
be consistent with a “glutamate homeostat”, responding to excess presynaptic glutamate
release, necessary to induce and express PHD (X. Li, P. Goel, J. Wondolowski, et al., 2018).
Alternatively, the presynaptic inhibition triggered at ATI NMJs could be a novel form of
presynaptic homeostatic depression which is induced in response to synaptic overgrowth.
Interestingly, while increased postsynaptic glutamate receptors levels enhance mini size, no
adaptive change in presynaptic function results, which leads to a concomitant increase in
synaptic strength (DiAntonio, Petersen, Heckmann, & Goodman, 1999; X. Li, P. Goel, J.
Wondolowski, et al., 2018). One possibility is that a coordinated increase in both pre- and postsynaptic compartments may be necessary to induce the presynaptic inhibition observed at ATI
NMJs. The ATI model provides a unique opportunity to interrogate the interplay between
developmental growth, adaptive presynaptic inhibition, and other homeostatic signaling
systems.
Extending the larval stage through the ATI manipulation will circumvent limitations of the
brief time window provided by the standard developmental program. Although the ATI model
does not appear to exhibit the features described at aging mammalian NMJs (L. Li, W. C. Xiong,
& L. Mei, 2018; Taetzsch & Valdez, 2018), we have demonstrated its potential for modeling
neurodegenerative conditions by showing the extent of synaptic destabilization caused by loss
of stathmin that was not fully apparent when restricted to the normal short developmental period

22
in Drosophila larvae (E. R. Graf et al., 2011). By examining stathmin mutant phenotypes in ATIextended larvae, we were able to observe progressive, time-dependent retractions of
presynaptic terminals and gain further insight into stathmin’s role in normal NMJ growth and
stability. Consistent with the role of stathmin in flies, the mammalian homolog (SCG10) is
thought to be part of an axonal injury surveillance system, where it accumulates after injury and
is involved in regenerative signaling (Shin et al., 2012). More generally, previous studies of
degenerative disease models in the larval system have been limited by the brief timespan. For
example, one important ALS disease model in flies involves overexpression of repetitive RNAs
and peptides derived from the human C9ORF72 gene (Mizielinska et al., 2014; Xu et al., 2013).
However, while a variety of progressive and degenerative phenotypes are observed in
photoreceptors of adult flies, only the most toxic transgenes are capable of inducing substantial
neurodegeneration at the larval NMJ (Perry et al., 2017), likely due to the limited time frame of
typical larval development. The longer timescale enabled by the ATI model therefore provides
new opportunities to study progressive phenotypes during neuronal injury, stress, and
neurodegeneration in addition to the plasticity discussed above in a rapid and genetically
tractable system. Indeed, fly models of neurodegenerative conditions such as ALS,
Huntington’s, Parkinson’s and Alzheimer’s diseases (McGurk, Berson, & Bonini, 2015) can
benefit from the high resolution imaging and electrophysiological approaches established at the
larval NMJ. The powerful combination of established genetic tools, including binary expression
systems (Gal4/UAS, LexA, QF systems; (Venken, Simpson, & Bellen, 2011)) and emerging
CRISPR/Cas9 manipulations (Bier, Harrison, O’Connor-Giles, & Wildonger, 2018) with the ATI
model provides an exciting foundation to gain new insights into synaptic growth, structure,
function, plasticity, injury, and neurodegeneration over long times using the glutamatergic NMJ
as a model.

23
2.5 Materials and Methods
Fly Stocks: Drosophila stocks were raised at 25°C on standard molasses food. The w
1118
strain is used as the wild type control unless otherwise noted, as this is the genetic background
of the genetic mutants used in this study. ATI larvae were generated by crossing phm-GAL4
(Gibbens et al., 2011)to UAS-smox-RNAi (BDSC #41670). Stathmin mutations were introduced
into the ATI background (stai allele: BDSC #16165). All experiments were performed on thirdinstar larvae or arrested third instar larvae of both sexes and at various time points. A complete
list of all stocks and reagents used in this study, see Table S2.
Immunocytochemistry: Third-instar male or female larvae were dissected in ice cold 0
Ca2+ HL-3 and fixed in Bouin's fixative for 5 min as described (X. Chen et al., 2017b). Briefly,
larvae were washed with PBS containing 0.1% Triton X-100 (PBST) for 30 min, blocked for an
hour with 5% normal donkey serum in PBST, and incubated overnight in primary antibodies at
4°C followed by washes and incubation in secondary antibodies. Samples were mounted in
VectaShield (Vector Laboratories). The following antibodies were used: mouse anti-Bruchpilot
(nc82; 1:100; Developmental Studies Hybridoma Bank; DSHB); rabbit anti-DLG ((1:10,000;
(Pielage, Fetter, & Davis, 2005b)); guinea pig anti-vGlut ((1:2000; (Goel & Dickman, 2018));
mouse anti-GluRIIA (8B4D2; 1:100; DSHB); affinity purified rabbit anti-GluRIIB (1:1000; (Goel &
Dickman, 2018)), guinea pig anti-GluRIID ((1:1000; (Perry et al., 2017)). Donkey anti-mouse,
anti-guinea pig, and anti-rabbit Alexa Fluor 488-, Cyanine 3 (Cy3)-, and DyLight 405-
conjugated secondary antibodies (Jackson ImmunoResearch) were used at 1:400. Alexa Fluor
647 conjugated goat anti-HRP (Jackson ImmunoResearch) was used at 1:200.
Tetramethylrhodamine (TRITC)-conjugated phalloidin (Thermo Fisher) was used at 1:1000. All
antibody information is summarized in Table S2.

24
Confocal imaging and analysis: Samples were imaged using a Nikon A1R Resonant
Scanning Confocal microscope equipped with NIS Elements software and a 100x APO 1.4NA
oil immersion objective using separate channels with four laser lines (405, 488, 561, and 637
nm). For fluorescence quantifications of BRP intensity levels, z-stacks were obtained using
identical settings for all genotypes with z-axis spacing 0.5 µm within an experiment and
optimized for detection without saturation of the signal as described (Perry et al., 2017).
Boutons were counted using vGlut- and HRP-stained Ib NMJ terminals on muscle 4 of segment
A2-A4, considering each vGlut punctum to be a bouton. The general analysis toolkit in the NIS
Elements software was used for image analysis as described (Kikuma, Li, Kim, Sutter, &
Dickman, 2017b). Neuronal surface area was calculated by creating a mask around the HRP
channel that labels the neuronal membrane. BRP puncta number, area, and total BRP intensity
per NMJ were quantified by applying by using a bright-spot detection method and filters to
binary layers on the BRP labeled 488 channel in a manner similar to that previously described
(P. Goel et al., 2019). GluRIIA, GluRIIB, and GluRIID puncta intensities were quantified by
measuring the total sum intensity of each individual GluR punctum and these values were then
averaged per NMJ to get one sample measurement (n). For NMJ retraction analysis, footprints
were scored by eye as reported in (Eaton et al., 2002) on M6/7 segments A2-A5. Anti-DLG was
used as a postsynaptic marker and either anti-vGlut or anti-BRP for a presynaptic marker (wild
type controls yielded similar retraction scores for either presynaptic marker).
Electron Microscopy: EM analysis was performed as described previously (Russo et al.,
2019) with minor modifications. Wandering third-instar larvae were dissected in Ca2+
-free HL-3
and then fixed in 4% paraformaldehyde prepared by mixing 16% paraformaldehyde (aqueous,
EMS aspx#15700) in Phosphate Buffered Saline (PBS) at 4°C. The larval pelts were stored in
this fixative at 4°C. On the day of preparation, larvae were transferred to 1% glutaraldehyde in
distilled water for 1 hour at room temperature, then washed three times for 5 min in distilled

25
water. The larval pelts were then placed in 1% osmium tetroxide in distilled water 90 mins at
room temperature and in ferracyanide-reduced osmium tetraoxide for 90 mins. After washing
three times in water, larvae were dehydrated in a graded acetone series, infiltrated with 50%
uncatalyzed Spurr’s resin in acetone for 24 hours, followed by two 24 hour changes of 100%
uncatalyzed Spurr’s and polymerization in freshly catalyzed Spurr’s for 24 hours at 60°C. 60 nm
thin sections were collected on Formvar-coated 2x1 mm slot grids, stained with 2% uranyl
acetate and Reynolds lead citrate and imaged by transmission electron microscopy at 80 kV in
a Zeiss EM10 (Carl Zeiss Microscopy, White Plains, NY) equipped with an Erlangshen CCD
camera (Gatan, Pleasanton, CA). Images were analyzed blind to genotype using measurement
tools in the ImageJ software. The outer diameter of each synaptic vesicle within a 300 nm
radius from the center of an active zone within a bouton was quantified for all genotypes.
Electrophysiology: All dissections and recordings were performed in modified HL-3
saline (Dickman, Horne, Meinertzhagen, & Schwarz, 2005a; Kiragasi, Wondolowski, Li, &
Dickman, 2017; Stewart, Atwood, Renmger, Wang, & Wu, 1994) containing (in mM): 70 NaCl, 5
KCl, 10 MgCl2, 10 NaHCO3, 115 Sucrose, 5 Trehelose, 5 HEPES, and 0.4 CaCl2, pH 7.2.
Neuromuscular junction sharp electrode (electrode resistance between 10-30 MΩ) recordings
were performed on muscles 6 and 7 of abdominal segments A2 and A3 in wandering third-instar
larvae as described (Goel, Li, & Dickman, 2019). Recordings were performed on an Olympus
BX61 WI microscope using a 40x/0.80 water-dipping objective, and acquired using an Axoclamp
900A amplifier, Digidata 1440A acquisition system and pClamp 10.5 software (Molecular
Devices). Electrophysiological sweeps were digitized at 10 kHz and filtered at 1 kHz. Data were
analyzed using Clampfit (Molecular devices), MiniAnalysis (Synaptosoft), and Excel (Microsoft)
software.
Miniature excitatory postsynaptic potentials (mEPSPs) were recorded in the absence of
any stimulation and cut motor axons were stimulated to elicit excitatory postsynaptic potentials

26
(EPSPs). Average mEPSP, EPSP, and quantal content were calculated for each genotype by
dividing EPSP amplitude by mEPSP amplitude. Muscle input resistance (Rin) and resting
membrane potential (Vrest) were monitored during each experiment. Recordings were rejected if
the Vrest was above -60 mV, if the Rin was less than 5 MΩ, or if either measurement deviated by
more than 10% during the course of the experiment. Larvae were incubated with or without
philanthotoxin-433 (PhTx; Sigma; 20 μM) resuspended in HL-3 for 10 mins, as described
(Dickman & Davis, 2009; C. A. Frank et al., 2006).
Failure analysis was performed in HL-3 solution containing 0.15 mM CaCl2, which
resulted in failures in about half of the stimulated responses in wild-type larvae. A total of 40
trials (stimulations) were performed at each NMJ in all genotypes. Failure rate was obtained by
dividing the total number of failures by the total number of trials (40). Paired-pulse recordings
were performed at a Ca2+ concentration of 0.3 mM to assay facilitation (PPF) and 1.5 mM for
depression (PPD). Following the first stimulation, a second EPSC was evoked at an
interstimulus interval of 16.67 ms (60 Hz). Paired-pulse ratios were calculated as the EPSC
amplitude of the second response divided by the first response (EPSC2/EPSC1).
Experimental Design and Statistical Analysis: For electrophysiological and
immunostaining experiments, each NMJ terminal (muscle 6 for physiology, muscle 4 for
immunostaining analyses) is considered an n of 1 since each presynaptic motor neuron terminal
is confined to its own muscle hemi-segment. For these experiments, muscles 4 or 6 were
analyzed from hemi-segments A2-A4 from each larva, typically 2 NMJs/animal per experiment.
To control for variability between larvae within a genotype, NMJs were analyzed from at least 5
individual larvae. See Table S1 for additional details.
Statistical analysis was performed using GraphPad Prism (version 7.0) or Microsoft
Excel software (version 16.22). Data were assessed for normality using a D’Agostino-Pearson
omnibus normality test, which determined that the assumption of normality of the sample

27
distribution was not violated. Normally distributed data were analyzed for statistical significance
using a Student’s t-test (pairwise comparison) or an analysis of variance (ANOVA) and Tukey’s
test for multiple comparisons. Data were then compared using either a one-way ANOVA and
tested for significance using a Tukey’s multiple comparison test or using an unpaired 2-tailed
Student’s t-test with Welch’s correction. All data are presented as mean ±SEM. with varying
levels of significance assessed as p≤0.05 (*), p≤0.01 (**), p≤0.001 (***), N.S. = not significant.

28
Figure 1: Synaptic strength remains stable throughout the life of an arrested third instar larva.
(A) (Left) Schematic illustrating the signaling pathway that stimulates ecdysone synthesis in the
prothoracic gland prior to pupal formation. The transcription factor Smox is required for the
expression of both Torso and insulin receptors. (Right) Schematic comparing wild type
developmental timing and arrested maturation induced by smox-RNAi. (B) Representative
photographs of wild type third instar larvae (WT.5) and smox-RNAi (ATI) larvae at different time
points (days after egg lay (AEL)). Scale bar: 0.5 mm. (C) Representative images of larval body
walls stained with anti-phalloidin to highlight muscle structure. Muscle 4 surface area is outlined
in each image. Scale bar: 50 µm. (D) Graph summarizing muscle surface area measurements
(black) and muscle input resistance (grey) across the ATI lifespan. (E) Representative EPSP
traces for WT.5, ATI.5, ATI.17 and ATI.33 NMJs. (F) Average EPSP amplitudes for the
genotypes shown in (E). The sample size for each data set is indicated below the data point (D)
or in each bar (E). Error bars indicate ±SEM. One-way analysis of variance (ANOVA) test was
performed, followed by a Tukey’s multiple-comparison test. N.S.=not significant, p>0.05.

29
Figure 2: Progressive synaptic growth and a concomitant accumulation of release sites at ATI
NMJs. (A) (Top) Representative images of muscle 4 NMJs at WT.5 and several ATI time points
stained with anti-vGlut (synaptic vesicle marker) and anti-HRP (neuronal membrane). (Bottom)
Representative portion of the synapses above marked with anti-BRP (active zone marker).
Scale bars: 10 µm (top) or 5 µm (bottom). (B-D) Graphs showing the average neuronal
membrane surface area (B), bouton number (C), and BRP puncta number (D) per muscle 4
NMJ for WT.5 and the indicated ATI time points. The sample size for each data set is indicated
in each bar. Error bars indicate ±SEM. One-way analysis of variance (ANOVA) test was
performed, followed by a Tukey’s multiple-comparison test. **p≤0.01; ***p≤0.001; N.S.=not
significant, p>0.05.

30

31
Figure 3: Postsynaptic glutamate receptors accumulate and quantal size increases over the ATI
lifespan. (A) Representative images of the indicated GluR subunit staining at NMJ terminals of
muscle 4 in wild type (WT.5) and the indicated ATI time points. Scale bar: 5 µm. Quantification
of GluRIID puncta number per NMJ (B) and GluR puncta sum fluorescence intensity (C) in the
indicated genotypes. The value for wild type at day 5 is shown by a dashed line (WT.5, n=40).
(D) Representative mEPSP traces of WT.5 and the indicated ATI time points. (E-G)
Quantification of mEPSP amplitude (E), mEPSP frequency (F), and decay time constant (G) in
the indicated genotypes. The sample size for each data set is indicated in each bar. Error bars
indicate ±SEM. Two-tailed Student’s T-Test was performed. *p≤0.05; **p≤0.01; ***p≤0.001;
N.S.=not significant, p>0.05.

32
Figure 4: Synaptic vesicle size remains unchanged across the ATI lifespan. (A) Representative
electron micrographs of synaptic vesicles at NMJ terminals of wild type (WT.5) and ATI larvae at
the indicated time points. Scale bar: 50 nm. (B) Quantification of synaptic vesicle diameter in the
indicated genotypes reveal no significant differences. The sample size for each data set is
indicated in each bar. Error bars indicate ±SEM. One-way analysis of variance (ANOVA) test
was performed, followed by a Tukey’s multiple-comparison test. N.S.=not significant, p>0.05.

33
Figure 5: A potent reduction in neurotransmitter release probability is expressed across the ATI
lifespan. (A) (Left) Schematic illustrating reduced synaptic strength at later ATI time points.
(Right) Quantal content calculated from EPSP and mEPSP data in Figs 1, 3. (B) (Left)
Representative traces following stimulation used to perform failure analysis. Grey traces
indicate failures and colored traces indicate evoked responses. Eight traces are shown for each
genotype. (Right) Quantification of failure rates for each genotype. (C) Representative two
electrode voltage clamp (TEVC) traces in lowered extracellular Ca2+ showing paired pulse
facilitation for each genotype (left) and quantification of the paired pulse ratio (right). (D)
Representative TEVC traces showing paired pulse depression in elevated extracellular Ca2+ for
each genotype (left) and the paired pulse ratio (right). The sample size for each data set is
indicated in each bar. Error bars indicate ±SEM. Two-tailed Student’s T-Test was performed.
**p≤0.01; ***p≤0.001; N.S.=not significant, p>0.05.

34

35
Figure 6: Extending the larval stage reveals the progression of synaptic retractions in stathmin
mutants. (A) Representative images of ATI.5 and ATI.33 synapses stained with presynaptic
(BRP; green) and postsynaptic (DLG; magenta) markers demonstrating a lack of synaptic
retractions at these stages. Scale bar: 5 µm. (B) Quantification of % NMJs at muscle 6/7 with
one or more footprints observed across the ATI lifespan. The sample size for each data set is
shown above each bar. The value for wild type at day 5 is shown by a dashed line (WT.5, n=9).
(C) Representative BRP and DLG images of stathmin mutant NMJs in an ATI background
(stai.5 and stai.13, see Table S1 for full genotypes) showing footprints (DLG staining without
corresponding BRP marked with an asterisk). Scale bar: 5 µm. (D) Quantification of NMJs with
one or more footprints in stai.5 and stai.13 animals. (E) Quantification of footprints per NMJ
separated by segment in stathmin mutants demonstrating more severe retractions on posterior
segments (stai.5: A2, n=21; A3, n=21; A4, n=22; A5, n=21; stai.13: A2, n=20; A3, n=20; A4,
n=19; A5, n=20). (F) Representative images of ATI and stathmin ATI NMJs at muscle 4 (days 5
and 13) stained with HRP and BRP showing a failure of NMJs to growth in stai mutants. Scale
bar: 10 µm. Quantification of bouton number (G; the sample size for each data set is indicated in
each bar.) and BRP puncta number (H; ATI.5, n=15; ATI.13, n=7; stai.5, n=8; stai.13, n=8) per
NMJ on segment A4 for the indicated genotypes. Error bars indicate ±SEM. Two-tailed
Student’s T-Test was performed. *p≤0.05; **p≤0.01; ***p≤0.001; N.S.=not significant, p>0.05.

36
Figure S1: Presynaptic homeostatic potentiation can be induced and expressed across the ATI
lifespan. (A) (Top) Schematic illustrating baseline and +PhTx conditions at NMJs for each
genotype. (Bottom) Representative EPSP and mEPSP traces for each genotype at baseline and
+PhTx. (B-D) Quantification of mEPSP amplitude (B), EPSP amplitude (C), and mEPSP and
quantal content values following PhTx application normalized as a percentage of baseline
values (-PhTx) (D) in the indicated genotypes. (Baseline: WT.5, n = 5; ATI.5, n = 7; ATI.17, n =
8; ATI.25, n = 7. +PhTx: WT.5, n = 7; ATI.5, n = 7; ATI.17, n = 11; ATI.25, n = 10). Error bars
indicate ±SEM. Two-tailed Student’s T-Test was performed. **p≤0.01; ***p≤0.001; N.S. = not
significant, p>0.05.

37
Chapter 3: Axonal transport defects reveal novel glutamate receptor
plasticity

38
3.1 Abstract
Neurons in the peripheral nervous system are incredibly resilient to a variety of stress.
Homeostatic mechanisms that are induced to overcome synaptic transmission perturbations are
well known and are apparent in both the peripheral and central nervous systems. A unique
adaptation that neurons in the periphery has developed is the ability to both regenerate and
degenerate specific regions of their axons following an injury. This unique control is governed by
injury-related Dual leucine zipper-bearing kinase signaling in mammals, and Wallenda signaling
in Drosophila. Here, we interrogate whether Presynaptic Homeostatic Depression (PHD) via
vGlut overexpression (vGlut-OE) can be induced when a neuron is experiencing injury. First, we
find that these neurons appear to undergo PHD. However, we are unclear if the same wellknown mechanism occurs during active injury signaling. We then assess how the presynaptic
neuron experiencing injury is affected by vGlut-OE and find that axonal trafficking is deficient
since more protein is observed within nerves and fewer proteins are found at the axon terminal.
Finally, our analysis of the postsynaptic compartment reveals an astonishing upscaling of
receptors. Therefore, we believe that the classic PHD mechanism is not involved in this injury
with vGlut-OE condition. Instead, this is an entirely different form of synaptic plasticity where
transmission is severely reduced, leading to a concomitant increase of receptors, revealing a
novel form of postsynaptic plasticity.

39
3.2 Introduction
Neuronal injury is detrimental to organisms as it can impede circuit function in the central
nervous system (CNS) or hinder sensory and motor transmission in the peripheral nervous
system (PNS). Incredibly, neurons in the PNS are protected due to Dual leucine zipper-bearing
kinase (DLK) signaling in mammals, and Wallenda (Wnd) signaling in Drosophila (Asghari Adib
et al., 2018). DLK/Wnd activation triggers a two-part program that selectively regenerates the
proximal axon and degenerates distal segments (Hao & Collins, 2017; Hao et al., 2019; Shin,
Ha, Kim, Cho, & DiAntonio, 2019; Watkins et al., 2013; Xiong et al., 2010). Several cellular
stressors have been implicated in Wnd activation - the most apparent is axotomy or nerve crush
(Jin & Zheng, 2019). Cytoskeletal perturbations due to pharmacological agents (Valakh et al.,
2015) and genetic disruptions (Massaro et al., 2009) also trigger Wnd signaling. Even relatively
innocuous stress, like the accumulation of synaptic proteins in cell bodies, induce the Wnd
cascade (Li et al., 2017). While there are several triggers for Wnd activation, the most
consistent is the inhibition or loss of the upstream Phr1/Highwire/Rpm-1 (PHR family) protein
that normally degrades Wnd (Grill et al., 2016). For this thesis work, we utilize a wellcharacterized hiw mutant (hiwΔN) to chronically turn on injury-related signaling.
Both the loss of Hiw and the enhanced activity of Wnd are known to induce intracellular
modulations. A striking consequence of either condition is an unrestrained growth of synapses
at the fly neuromuscular junction (NMJ) (Wan et al., 2000; Wang et al., 2013). The resulting
structure is spindly with nearly double the number of boutons. However, each bouton is much
smaller in size and expresses lower levels of proteins. This is likely due to the transcriptional
downregulation by active Wnd in response to protein accumulation at the soma due to axonal
defects or overexpression (J. Li et al., 2017). Hiw is even shown to promote synaptic
transmission (Russo et al., 2019; Wu et al., 2005) . While these types of regulation are expected
within the cell experiencing active injury, interesting adaptations are also observed in partnered
cells. Past work in the Dickman Lab reveals that synapses undergoing injury-related signaling

40
alter their homeostatic threshold to stabilize lower synaptic strength (Goel & Dickman, 2018).
Pragya Goel’s study asks if Presynaptic Homeostatic Potentiation (PHP) can be induced during
active Wnd signaling but finds that it is blocked – this is the basis for my own dissertation work
with a different form of homeostasis.
Instead of potentiation, I focus instead on Presynaptic Homeostatic Depression (PHD)
here. The first mention of PHD occurs upon overexpressing the vesicular glutamate transporter
(vGlut) gene (Daniels et al., 2004). This manipulation increases the number of vGlut proteins
embedded in synaptic vesicles, leading to enlarged vesicles that are loaded with greater
amounts of glutamate. As a result, releasing a single vesicle yields a larger mini EPSP
amplitude. Interestingly, quantal content reduced by half, implying that total vesicular release is
hindered. Earlier work from the Dickman Lab show that a presynaptic component inhibits
neurotransmitter release (X. Li, P. Goel, J. Wondolowski, et al., 2018). Later, Xiling defines the
presynaptic glutamate-gated chloride channel (GluCl) as the player for autocrine inhibition of
transmission (X. Li et al., 2021). The release of excess glutamate is the key trigger for GluCl
activation and PHD induction. My work will assess whether PHD can occur during injury and if
so, whether it occurs by the same mechanism.
With electrophysiological experiments performed by Jerry Chien, we observe what we
initially consider to be PHD. However, my NMJ imaging reveals morphological defects with
enlarged terminal boutons and a robust decrease in synaptic vesicle proteins. Imaging within
the motor nerves by my undergraduate mentee, Isha Sanghvi, makes it clear that proteins are
experiencing “traffic jams” and are stuck within axons. Interestingly, my postsynaptic imaging
shows that the receptors triple in abundance, which may be a result of reduced synaptic cargo
reaching the axon terminals. This then leads us to hypothesize that enlarged vesicles fail to
traffic down the axon during injury conditions, meaning that only normal-sized vesicles can be
transported. To test this, we employ the skills of Cristian Pinales at Oak Crest Institute of
Science to image the synaptic vesicles near vesicular release sites via electron microscopy.

41
Unfortunately, this cannot be done due to the poor quality of the sick animal tissue. However, in
the end, we believe that we have found a novel form of glutamate receptor plasticity.

42
3.3 Results
Presynaptic homeostatic depression occurs normally in an injury signaling background
To answer the question of whether PHD can be induced during injury, we employed the
classic method of vGlut overexpression (vGlut-OE; (Daniels et al., 2004)) in motor neurons. This
increased the number of vGlut proteins embedded in synaptic vesicle membranes, allowing for
individual vesicles to be filled with more glutamate. An interesting side effect of this manipulation
is that synaptic vesicles increase in size and their density around active zone scaffolds is
decreased (Fig. S1A-D). As a result, vGlut-OE enhances miniature EPSP (mEPSP) amplitudes
in the wild type background (Fig. 1B). The evoked EPSP amplitudes are generally unchanged
(Fig. 1C) compared to the wild type baseline, leading to a resulting decrease in quantal content
(Fig. 1D). These trends match previous research and point to a decrease in vesicular release
per action potential (Daniels et al., 2004; X. Li, P. Goel, J. Wondolowski, et al., 2018).
For PHD to occur during injury signaling, we expected to see similar physiological
trends. Since our lab previously reported that hiw reduces quantal content and downscales
receptor fields compared to wild type (Goel & Dickman, 2018), we made comparisons between
hiw + vGlut-OE and hiw alone. At first glance, we noticed a robust doubling of mEPSP
amplitude (Fig. 1F) and no difference in EPSP amplitude (Fig. 1G). As a result, we observed a
severe decrease, by nearly half, in the quantal content (Fig. 1H). Based on these results, we
hypothesized that PHD was in fact occurring the same way in both wild type (Fig. 1A) and
injury-related signaling (Fig. 1E) conditions. We thought that the main differences in the vGlutOE backgrounds compared to baseline was the release of fewer enlarged vesicles and that the
postsynaptic receptive fields remained unchanged. However, previous studies have shown
synaptic plasticity due to the loss of hiw (Goel & Dickman, 2018), so we decided to characterize
key proteins in the pre- and postsynapse to verify if the synaptic composition was truly the same
in both wild type and injured genetic backgrounds.

43
Presynaptic motor neurons of hiw + vGlut-OE animals exhibit altered morphology
We first began our investigation with the presynaptic motor neuron since this is the cell
actively experiencing injury signaling. Immediately, we noticed a difference in morphologies in
the hiw conditions compared to wild type. Wild type alone and vGlut-OE synapses appear
generally similar in their morphologies (Fig. 2A). Quick bouton counting at muscles 6/7 (Fig. 2B)
and analyzing the terminal boutons (Fig. 2C) revealed nearly identical bouton numbers and
terminal bouton areas. Therefore, we concluded that vGlut overexpression in a wild type
background only affected synaptic vesicle conformation and transmission, but not neuronal
growth. This was not the case for animals undergoing injury signaling, however. At baseline, hiw
mutants’ synapses exhibit exuberant growth (Wan et al., 2000; Wu et al., 2005) with spindly
synapses consisting entirely of numerous, tiny boutons (Fig. 2D, left). With added vGlut-OE, we
observed synapses with branches composed of similarly small boutons that terminated into
comically large boutons (Fig. 2D, right). Surprisingly, the bouton numbers were the same
between both conditions (Fig. 2E) despite differences in growth leading to massive terminal
boutons (Fig. 2F). Due to the extreme swelling, we thought there might be issues with axonal
transport. A previous report showed that mutations in a dynactin subunit caused major swelling
in terminal boutons (Lloyd et al., 2012) as motor proteins and endosomal vesicles accumulated.
Other studies showed that activated Wnd and downstream JNK signaling also regulated
kinesin- and dynein-mediated axonal transport (Liu, 2017; Verhey, 2007), supplying support for
our hypothesis.
Synaptic vesicle markers are diminished at hiw + vGlut-OE presynaptic terminals
To assess the possible axonal transport defects, we first quantified the level of
expression for synaptic vesicle markers at the presynaptic terminal. To do this, we
immunolabeled vesicular glutamate transporter (vGlut), Synapsin (Syn), and Synaptotagmin
(Syt) (Fig. 3A, C), imaged the NMJ, and compared the mean fluorescence intensities between

44
animal types. Interestingly, we did not see a difference in Syn or Syt protein levels for vGlut-OE
(Fig. 3B), but we did see an expected increase of vGlut levels since this is the expression that
we were manipulating. Despite large synaptic vesicles, it appeared that axonal trafficking
occurred normally and that synaptic vesicle proteins could make it to the presynaptic terminal
without issues. We then turned to our injury-related signaling animals and assessed the same
properties. At this point, we were surprised to see that all protein levels were reduced in hiw +
vGlut-OE (Fig. 3D) compared to hiw. Syn and Syt levels were reduced by more than half. The
most interesting result was that vGlut levels were reduced by 25% despite the high level of
genetic overexpression. We now hypothesized that Wnd activation, which already reduces
trafficking (J. Li et al., 2017), along with large vesicular cargo compounded transport defects. If
synaptic products could not make it to the presynaptic terminal, then they were likely trapped
and aggregating somewhere upstream in the motor neuron.
vGlut overexpression exacerbates axonal transport defects in hiw mutants
The most logical region to look for a “traffic jam” of proteins were within the axons of
motor nerves. These contain microtubule tracks that motor proteins could walk along to carry
cargo to and from the synaptic terminal. If trafficking was indeed hindered, then we would
expect to see an increase of proteins within nerve bundles. For this assay, we maintained the
central nervous system (CNS) during our dissections to preserve the motor nerves running from
the ventral nerve cord (VNC; the fly equivalent of the spinal cord) to the NMJ.
Quick observations of the nerves revealed obvious trafficking defects in animals with
added vGlut-OE (Fig. 4A). When looking at Syn and vGlut levels in vGlut-OE animals, we did
not see an increase of Syn within axons, but we did see a greater than five-fold increase of
vGlut (Fig. 4B). The Syn result made sense since we saw wild type levels of this protein at the
axon terminal. The vGlut intensity was unexpected since we saw a large amount of this protein
at the terminal end. We could not claim that the high axonal intensity was due to trafficking

45
defects, however. It is possible that the overexpression of vGlut produces so much protein that
large amounts are transported down the axons normally.
We then turned to studying the effect of injury signaling on transport. In a hiw mutant, we
did not see a difference in either vGlut or Syn levels (Fig. 4C). Although we expected to see
more “traffic jams”, it is possible that active Wnd signaling reduced the gene expression of both
(J. Li et al., 2017). Therefore, protein levels would also decrease, leading to an overall decrease
in the measurable protein visible within axons. So even if transport was hindered, an overall
reduction of protein gave the perception of normal protein intensities in axons. Finally, we
assayed protein trafficking in hiw + vGlut-OE animals (Fig. 4D). As with hiw, we did not see a
difference in Syn protein expression within axons, which may have been due to downregulation
of gene expression. However, the vGlut levels robustly increased by 14-fold in the motor nerves.
Given the knowledge that Wnd signaling should decrease gene expression, it was surprising to
see that vGlut was unaffected in these animals. Due to the extreme blockage of vGlut in the
nerves, it appeared that our intensified trafficking defects hypothesis was supported.
Postsynaptic glutamate receptor abundance is enhanced when apposing terminals are deprived
of synaptic cargo
Now with a clearer understanding of plasticity in the neuron, we then turned to the
postsynaptic muscle to study potential compensation due to the reduction of synaptic vesicle
proteins. Upon comparing wild type and vGlut-OE synapses (Fig. 5A), we saw an overall
reduction of receptors with a greater decrease in GluRIIA-type receptors (Fig. 5B). This
matched our lab’s recent findings with the release of excess glutamate (Y. Han et al., 2023).
Opposite trends were apparent with our injury conditions (Fig. 5C). Incredibly, the receptive field
increased three-fold in hiw + vGlut-OE animals, with a greater increase in GluRIIA levels (Fig.
5D). While this matches general trends that we have seen in our lab with reduced vGlut levels
(unpublished data not shown here), the magnitude of scaling is astonishing and unexpected.

46
3.4 Discussion
While this work is proposed as a PHD partner to our previous studies of injury-related
signaling and PHP (Goel & Dickman, 2018), it soon becomes clear that this is an entirely
different type of synaptic compensation. In our past work, we show that loss of hiw blocks PHP
signaling and induction. Naturally, we hypothesize that PHD would also be hindered. However,
electrophysiology shows characteristic traits of homeostatic depression (Fig. 1). We now
understand the mechanism for PHD due to vGlut-OE (X. Li et al., 2021; X. Li, P. Goel, J.
Wondolowski, et al., 2018). The trigger is the release of excess glutamate to activate a
presynaptic glutamate-gated chloride channel (GluCl). Once GluCl opens, further release of
vesicles is inhibited, resulting in an autocrine inhibition of synaptic transmission (Fig. 6A). For
our injury system to undergo PHD, we expect to see a similar activation of GluCl upon excess
glutamate release.
Our assessment of presynaptic neurons reveals axonal trafficking defects that are
enhanced by having enlarged vesicles in an injury background (hiw + vGlut-OE; Figs. 2-4). This
leads us to believe there are deficiencies in the presynaptic motor neuron. An analysis of the
receptor field shows a tripling in abundance (Fig. 5). Upon seeing this, we then question
whether large vesicles are being trafficked to and fusing at the presynaptic terminal. If large
vesicles are being released to a three-fold increase of receptors, then the mEPSP amplitude
should be much larger than what we observe (Fig. 1F). Therefore, we hypothesize that only
smaller vesicles are being trafficked and released in hiw + vGlut-OE synapses (Fig. 6B). We
had hoped to verify this via electron microscopy, but the tissue samples from these sickened
animals are too poor quality to process for imaging (Fig. S1). Without the enlarged vesicles
releasing excess glutamate, it is clear that the mechanism for this system’s synaptic
compensation relies on a massive upscaling of receptor numbers and decrease in vesicular
release.

47
While hiw mutants are potent activators of Wnd injury signaling, we cannot ignore
Highwire’s many functions. It is involved with Bone Morphogenetic Protein (BMP) and Wnd→cJun N-terminal kinase (JNK) signaling (Grill et al., 2016; McCabe et al., 2004) functioning as an
E3 ubiquitin ligase. It stands to reason that Hiw may be involved in many other pathways. More
specifically, this powerful protein regulates axonal and dendritic growth, synaptic transmission,
and direct axonal degeneration (Russo et al., 2019; Wang et al., 2013; Wu et al., 2005; Xiong et
al., 2012). To fully ensure that what we are observing is entirely due to Wnd injury signaling and
not one of Hiw many other functions, it would be beneficial to repeat all of the experiments in a
Wnd overexpression (Wnd-OE) genetic background. The limitation here is with the
Gal4/Upstream Activation Sequence (UAS) expression system (Brand & Perrimon, 1993) since
a single Gal4 can activate all UAS lines in an animal. Unfortunately, per tests in our lab, WndOE cannot be driven by OK6-Gal4 (the diver for vGlut-OE). Therefore, we would have to rely on
a separate binary expression system, LexA/LexAOp (Szüts & Bienz, 2000). While OK6-LexA
and LexAOp-vGlut both exist, the genetics proves to be too difficult since we would have to
manipulate three out of the four fly chromosomes within a single animal.
It is unknown at this point why the glutamate receptor field expands to the extent that it
does. However, evidence from our lab shows that a reduction in vGlut expression fills fewer
synaptic vesicles (Daniels et al., 2006) and that transmission leads to an increase in receptors
(unpublished data). Therefore, it is possible that the muscle is sensing its injured partner and
the reduced glutamate release and is compensating by embedding more receptors in its
membrane. To test this, we need to release greater amounts of glutamate into the synapse
while bypassing the issues with axonal transport. An attractive manipulation to consider is a
mutation in the minibrain (mnb) kinase (C. K. Chen et al., 2014). Mutations in this gene results
in defective endocytosis at the axon terminal to produce larger vesicles and a larger mEPSP
amplitude upon release, phenocopying vGlut-OE (Y. Han et al., 2023). If mnb and hiw mutants

48
display a downscaling of receptors compared to hiw + vGlut-OE, then our hypothesis would be
supported.

49
3.5 Materials and Methods
Fly stocks: Drosophila stocks were raised at 25°C on standard molasses food. The w
1118
strain is used as the wild type control unless otherwise noted as this is the genetic background
in which all genotypes are bred. The following fly stocks were used: UAS-vGlut (Daniels et al.,
2004), OK6-Gal4 (Aberle et al., 2002), hiwΔN (Wu et al., 2005). All other stocks were obtained
from Bloomington Drosophila Stock Center (BDSC): w
1118 (#5905).
Immunocytochemistry: Third-instar larvae were dissected in ice cold 0 Ca2+ HL-3 and
immunostained using a standard protocol as described (Goel et al., 2017; Y. Han et al., 2023).
In brief, larvae were either fixed in Bouin’s fixative for 5 min (Sigma, HT10132-1L), 100% icecold methanol for 5 min, or 4% paraformaldehyde (PFA) for 10 min. Larvae were then washed 3
times with phosphate buffered saline (PBS) containing 0.1% Triton X-100 (PBST) for 30 minutes
total, blocked with 5% Normal Donkey Serum followed by overnight incubation in primary
antibodies at 4°C. Preparations were then washed 3 times in PBST for 30 minutes total,
incubated in secondary antibodies for 2 hours at room temperature, washed 3 times in PBST for
30 minutes total, and equilibrated in 70% glycerol/PBS. Prior to imaging, samples were mounted
in VectaShield (Vector Laboratories). The following primary antibodies were used: mouse antiGluRIIΑ (8B4D2; 1:50; Developmental Studies Hybridoma Bank (DSHB)); rabbit anti-GluRIIIB
(1:1000; (Perry et al., 2017)); guinea pig anti-GluRIID (1:1000; (Perry et al., 2017)); mouse antisynapsin/SYNORF1 (3C11; 1:10; DSHB); guinea pig anti-vGlut (1:2000, (X. Chen et al.,
2017a)); rabbit anti-synaptotagmin1 (1:2500; (Mackler, Drummond, Loewen, Robinson, & Reist,
2002)); mouse anti-bruchpilot (nc82; 1:100; DSHB). Alexa Fluor-647 conjugated goat anti-HRP
(1:200; Jackson ImmunoResearch) and Donkey anti-mouse, -guinea pig, and -rabbit conjugated
Alexa Fluor 488, Cy3, and DyLight 405 secondary antibodies (Jackson ImmunoResearch) were
used at 1:400.

50
Imaging and analysis: Samples were imaged as (Perry et al., 2017) using a Nikon A1R
Resonant Scanning Confocal microscope equipped with NIS Elements software and a 100x
APO 1.4NA oil immersion objective using separate channels with four laser lines (405 nm, 488
nm, 561 nm, and 647 nm). For fluorescence intensity quantifications of Syn, Syt, vGlut, BRP,
GluRIIA, GluRIIB and GluRIID, z-stacks were obtained on the same day using identical gain and
laser power settings with z-axis spacing between 0.15-0.20 µm (for NMJ imaging) or 0.5 µm (for
nerve imaging) for all genotypes within an individual experiment. 30 µm of nerve bundles were
analyzed for axonal transport defects. Maximum intensity projections were utilized for
quantitative image analysis using the general analysis toolkit of NIS Elements software. The
fluorescence intensity levels of Syn, Syt, vGlut, BRP, GluRIIA, GluRIIB and GluRIID
immunostaining were quantified by applying intensity thresholds and filters to binary layers in
the 405 nm, 488 nm, and 561 nm channels. The mean intensity for each channel was quantified
by obtaining the average total fluorescence signal for each individual punctum and dividing this
value by the puncta area. The sum intensity for GluRIIA, GluRIIB, and GluRIID was quantified
as the total fluorescence signal of each individual GluR punctum. A mask was created around
the HRP channel, used to define the neuronal membrane, and only puncta within this mask
were analyzed to eliminate background signals. Bouton counting was done by labeling
synapses with vGlut and counting all positively stained puncta. All measurements based on
confocal images were taken from synapses acquired from at least six different animals.
Electron Microscopy: EM analysis was performed as described previously (Russo et al.,
2019) with minor modifications. Wandering third-instar larvae were dissected in Ca2+-free HL-3
and then fixed in 4% paraformaldehyde prepared by mixing 16% paraformaldehyde (aqueous,
EMS aspx#15700) in Phosphate Buffered Saline (PBS) at 4°C. The larval pelts were stored in
this fixative at 4°C. On the day of preparation, larvae were transferred to 1% glutaraldehyde in
distilled water for 1 hour at room temperature, then washed three times for 5 min in distilled

51
water. The larval pelts were then placed in 1% osmium tetroxide in distilled water 90 mins at
room temperature and in ferracyanide-reduced osmium tetraoxide for 90 mins. After washing
three times in water, larvae were dehydrated in a graded acetone series, infiltrated with 50%
uncatalyzed Spurr’s resin in acetone for 24 hours, followed by two 24 hour changes of 100%
uncatalyzed Spurr’s and polymerization in freshly catalyzed Spurr’s for 24 hours at 60°C. 60 nm
thin sections were collected on Formvar-coated 2x1 mm slot grids, stained with 2% uranyl
acetate and Reynolds lead citrate and imaged by transmission electron microscopy at 80 kV in
a Zeiss EM10 (Carl Zeiss Microscopy, White Plains, NY) equipped with an Erlangshen CCD
camera (Gatan, Pleasanton, CA). Images were analyzed blind to genotype using measurement
tools in the ImageJ software. The outer diameter of each synaptic vesicle within a 300 nm
radius from the center of an active zone within a bouton was quantified for all genotypes.
Electrophysiology: All dissections and recordings were performed in modified HL-3
saline (X. Li et al., 2021; Perry et al., 2022) containing (in mM): 70 NaCl, 5 KCl, 10 MgCl2, 10
NaHCO3, 115 Sucrose, 5 Trehelose, 5 HEPES, and 0.4 CaCl2, pH 7.2. Neuromuscular junction
sharp electrode (electrode resistance between 10-30 MΩ) recordings were performed on
muscles 6 and 7 of abdominal segments A2 and A3 in wandering third-instar larvae as
described (Perry et al., 2020). Recordings were performed on an Olympus BX61 WI microscope
using a 40x/0.80 water-dipping objective, and acquired using an Axoclamp 900A amplifier,
Digidata 1440A acquisition system and pClamp 10.5 software (Molecular Devices).
Electrophysiological sweeps were digitized at 10 kHz and filtered at 1 kHz. Data were analyzed
using Clampfit (Molecular devices), MiniAnalysis (Synaptosoft), and Excel (Microsoft) software.
Miniature excitatory postsynaptic potentials (mEPSPs) were recorded in the absence of any
stimulation and cut motor axons were stimulated to elicit excitatory postsynaptic potentials
(EPSPs). Average mEPSP, EPSP, and quantal content were calculated for each genotype by
dividing EPSP amplitude by mEPSP amplitude. Muscle input resistance (Rin) and resting

52
membrane potential (Vrest) were monitored during each experiment. Recordings were rejected if
the Vrest was above -60 mV, if the Rin was less than 5 MΩ, or if either measurement deviated by
more than 10% during the course of the experiment.
Statistical analysis: Data were analyzed using GraphPad Prism or Microsoft Excel
software. Sample values were tested for normality using the D’Agostino & Pearson omnibus
normality test which determined that the assumption of normality of the sample distribution was
not violated. Data were then compared using a one-way ANOVA and tested for significance
using a Tukey’s multiple comparison test. All data are presented as mean +/-SEM; p denotes
the level of significance assessed as p≤0.05 (*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****);
ns=not significant.

53
Figure 1: Presynaptic homeostatic depression occurs normally in an injury signaling
background. (A) Schematics and representative traces of mEPSP and EPSP showing that
overexpressing vGlut results in larger vesicles, increased mEPSP amplitude, and a reduction in
vesicular release. Genotypes: wild type (w
1118); vGlut-OE (w; OK6-GAL4/UAS-vGlut). (B-D)
Quantification of mEPSP amplitudes (B), EPSP amplitudes (C), and quantal content (D) in the
genotypes represented in (A). (E) Schematics and representative traces of mEPSP and EPSP
showing that overexpressing vGlut in hiw mutants results in larger vesicles, increased mEPSP
amplitude, and a reduction in vesicular release. Genotypes: highwire (hiw
ΔN); hiw + vGlut-OE
(hiwΔN; OK6-GAL4/UAS-vGlut). (F-H) Quantification of mEPSP amplitudes (F), EPSP
amplitudes (G), and quantal content (H) in the genotypes represented in (E). Data points in the
graphs represent one synapse. Error bars indicate ±SEM. One-way ANOVA test was
performed, followed by a Tukey’s multiple comparison test. * P≤0.05; **** P≤0.0001; ns, not
significant (P>0.05). Electrophysiology experiments were performed by Jerry Chien.

54
Figure 2: Presynaptic motor neurons of hiw + vGlut-OE animals exhibit altered morphology. (A)
Upper images: Representative muscle 4 synapses. HRP (magenta) labels the neuronal
membrane and vGlut (green) marks the boutons. Lower images: Representative string of
boutons labeled with HRP (gray) to portray typical bouton sizes. Genotypes: wild type (w
1118);
vGlut-OE (w; OK6-GAL4/UAS-vGlut). (B-C) Graphs showing average bouton numbers counted
at muscles 6/7 (B) and average terminal bouton area (C) for the respective genotypes in (A). (D)
Upper images: Representative muscle 4 synapses. HRP (magenta) labels the neuronal
membrane and vGlut (green) marks the boutons. Lower images: Representative string of
boutons labeled with HRP (gray) to portray typical bouton sizes. Genotypes: highwire (hiw
ΔN);
hiw + vGlut-OE (hiwΔN; OK6-GAL4/UAS-vGlut). (E-F) Graphs showing average bouton numbers
counted at muscles 6/7 (E) and average terminal bouton area (F) for the respective genotypes
in (D). Data points in the graphs represent one synapse. Error bars indicate ±SEM. One-way
ANOVA test was performed, followed by a Tukey’s multiple comparison test. **** P≤0.0001; ns,
not significant (P>0.05).

55
Figure 3: Synaptic vesicle markers are diminished at hiw + vGlut-OE presynaptic terminals. (A)
Representative images of synaptic vesicles protein (vesicular glutamate transporter, gray;
Synaptotagmin, green; and Synapsin, magenta) expression in muscle 4 NMJ terminals.
Genotypes: wild type (w
1118); vGlut-OE (w; OK6-GAL4/UAS-vGlut). (B) Quantification of mean
fluorescence intensity for each synaptic vesicle marker in the genotypes indicated in (A). All
values are normalized to wild type data. (C) Representative images of synaptic vesicles protein
(vGlut, gray; Syt, green; and Syn, magenta) expression in muscle 4 NMJ terminals. Genotypes:
highwire (hiw
ΔN); hiw + vGlut-OE (hiwΔN; OK6-GAL4/UAS-vGlut). (D) Quantification of mean
fluorescence intensity for each synaptic vesicle marker in the genotypes indicated in (C). All
values are normalized to highwire mutant data. The dotted lines in the graphs denote 100%
intensity for the respective baselines. Data points in the graphs represent one synapse. Error
bars indicate ±SEM. One-way ANOVA test was performed, followed by a Tukey’s multiple
comparison test. **** P≤0.0001; ns, not significant (P>0.05).

56
Figure 4: vGlut overexpression exacerbates axonal transport defects in hiw mutants. (A)
Representative images of nerve bundles to assess axonal transport. Top row: HRP (gray) labels
the neuronal membranes in the bundle. These images define the boundaries of expression for
vGlut (green) and Syn (magenta) shown in rows 2 and 3. Row 4: HRP labeling for the nerves
used to analyze Bruchpilot (BRP, cyan) axonal expression in row 5. Genotypes: wild type
(w
1118); vGlut-OE (w; OK6-GAL4/UAS-vGlut).; highwire (hiw
ΔN); hiw + vGlut-OE (hiwΔN; OK6-
GAL4/UAS-vGlut). (B-D) Quantification of vGlut, Syn, and BRP mean fluorescence intensity
within axons of vGlut-OE (B), hiw (C), or hiw + vGlut-OE (D) animals. All values were
normalized to wild type data. The dotted lines in the graphs denote 100% intensity for the
baseline wild type. Data points in the graphs represent one nerve. Error bars indicate ±SEM.
One-way ANOVA test was performed, followed by a Tukey’s multiple comparison test. ****
P≤0.0001; ns, not significant (P>0.05). Some imaging and all data analysis were performed by
Isha Sanghvi.

57
Figure 5: Postsynaptic glutamate receptor abundance is enhanced when apposing terminals are
deprived of synaptic cargo. (A) Representative images of glutamate receptor subtypes GluRIIA
(IIA, green) and GluRIIB (IIB, magenta) along with total glutamate receptor abundance
represented by GluRIID (IID, gray) at muscle 4. Genotypes: wild type (w
1118); vGlut-OE (w; OK6-
GAL4/UAS-vGlut). (B) Quantification of sum fluorescence intensity of IIA, IIB, and IID puncta for
the genotypes in (A). Data in this graph was normalized to wild type. (C) Representative images
of glutamate receptor subunits at muscle 4. Genotypes: highwire (hiw
ΔN); hiw + vGlut-OE (hiwΔN;
OK6-GAL4/UAS-vGlut). (D) Quantification of sum fluorescence intensity of IIA, IIB, and IID
puncta for the genotypes in (C). Data in this graph was normalized to highwire. The dotted lines
in the graphs denote 100% intensity for the respective baselines. Data points in the graphs
represent one synapse. Error bars indicate ±SEM. One-way ANOVA test was performed,
followed by a Tukey’s multiple comparison test. * P≤0.05; *** P≤0.001; **** P≤0.0001; ns, not
significant (P>0.05).

58
Figure 6: Schematic for proposed synaptic plasticity at synapses experiencing severe
reductions in axonal transport. (A) Schematic depicting presynaptic homeostatic depression
(PHD) in a wild type background. Upon overexpression of vGlut, vesicles become enlarged,
releasing more glutamate, and triggering an auto-inhibition of vesicle release. (B) Schematic
depicting a proposed model of presynaptic homeostatic depression (PHD) in a highwire mutant
(injury signaling) background. Although vGlut is overexpressed, large vesicles fail to release at
the presynaptic terminal due to issues with axonal transport. However, the postsynaptic receptor
abundance increases three-fold to compensate for the reduced vesicular release.

59
Supplemental Figure 1: Synaptic vesicle diameters increase in vGlut-OE animals. (A)
Representative electron micrographs of synaptic vesicles at terminals. Genotypes: wild type
(w
1118); vGlut-OE (w; OK6-GAL4/UAS-vGlut).; highwire (hiw
ΔN); hiw + vGlut-OE (hiwΔN; OK6-
GAL4/UAS-vGlut). (B) Quantification of average synaptic vesicle (SV) diameters in wild type,
vGlut-OE, and hiw animals. (C) Relative frequencies quantification showing the general
distribution of SV diameters for each genotype. (D) Quantification of SV density in a 300nm
radius from the center of an active zone (not shown in the micrographs). Data points in the
graphs represent one micrograph. Error bars indicate ±SEM. One-way ANOVA test was
performed, followed by a Tukey’s multiple comparison test. * P≤0.05; **** P≤0.0001; ns, not
significant (P>0.05). After dissections and fixation, samples were prepared and imaged by
Cristian Pinales at Oak Crest Institute of Science.

60
Chapter 4: Stathmin is a key player in the Wallenda injury signaling
pathway

61
4.1 Abstract
The peripheral nervous system is endowed with a unique ability to regenerate after
axonal injury or stress. Before regrowth can occur, the neuron is suspended in limbo where it
must determine which parts of it must die, and which parts may live. This decision is mediated
by both Wallenda/Dual Leucine zipper Kinase signaling and Stathmin-mediated axonal
surveillance. As a key microtubule-associated protein (MAP) and cytoskeletal stabilizer,
Stathmin plays an unknown role in the injury signaling pathway. Mutations in this gene may
trigger Wnd signaling, or this same signaling may require the activity of Stathmin protein
downstream for synaptic growth. We aim to dissect the interplay between this MAP and injury
signaling. First, we find that Stathmin is necessary for hiw mutant morphology. Next, we
characterize remodeling of the pre- and postsynaptic compartments in our injury with
microtubule destabilization double mutant model. We observe a drastic reduction in Bruchpilot
protein in more posterior segments of the animal, but an overall increase of vesicular glutamate
transporter protein throughout the animal in our double mutant compared to stathmin mutants
alone. On the postsynaptic side, we see an overall downscaling of receptors when comparing
the same sets of animals. While our model may phenocopy the morphology and
neurodegeneration of a stathmin mutant, it is clear that the added injury signaling is inducing
unique synaptic plasticity.

62
4.2 Introduction
The nervous system must be highly adaptable in the face of insults to ensure that the
entire network is protected and remains functional. This plasticity is apparent even at the single
neuron level. When axons become damaged or are stressed, they induce injury-related
signaling mediated by Wallenda (Wnd) in Drosophila or the Dual Leucine Zipper Kinase (DLK)
in mammals (Asghari Adib et al., 2018; Jin & Zheng, 2019), and the cell must decide to undergo
apoptosis or repair itself (Watkins et al., 2013). This pathway is highly conserved across
organisms (Goodwani et al., 2020; Yan, Wu, Chisholm, & Jin, 2009) and has been implicated in
neurodegenerative diseases like Alzheimer’s Disease and Amyotrophic Lateral Sclerosis (Le
Pichon et al., 2017; Patel et al., 2017), and in developmental disorders like Fragile X Syndrome
(Russo & DiAntonio, 2019). Wnd signaling activates both degenerative (Hao et al., 2019) and
regenerative (Shin et al., 2019; Wang et al., 2013) programs in distinct axonal compartments,
however, only a few key downstream players have been identified.
The Wnd signaling pathway is regulated by an E3 ubiquitin ligase, Highwire (Hiw), that
targets Wnd protein for degradation. During injury or after stress, hiw is degraded, allowing for
the activation of Wnd and the onset of injury signaling. Due to this interaction, Wnd can be
activated genetically by removing its regulator, Hiw, or by overexpressing Wnd itself (Collins et
al., 2006; Goel & Dickman, 2018; Grill et al., 2016; J. Li et al., 2017). Alternatively, Wnd
signaling can be induced physically via nerve crush, laser axotomy, or pharmacology that
disrupts the cytoskeletal structure (Jin & Zheng, 2019; Valakh, Frey, Babetto, Walker, &
DiAntonio, 2015). For my dissertation work here, I utilize a hiw mutant, hiwΔN. These animals’
synapses exhibit exuberant morphologies, implying that Hiw regulates synaptic growth and
associates with microtubules (Lewcock, Genoud, Lettieri, & Pfaff, 2007; Wang et al., 2013; Wu
et al., 2005). Injury-related signaling can be activated by cytoskeletal disruptions to induce cell
death or stabilize the cytoskeleton for repair. Given this unique “chicken and egg” scenario

63
between Wnd signaling and cytoskeletal dynamics, we aim to dissect their relationship by
employing a microtubule associated protein and stabilizer, stathmin.
Stathmin (stai) is a microtubule-associated protein (MAP; (Lachkar et al., 2010)) whose
function is necessary for fly NMJ stability (E. R. Graf et al., 2011). Loss of stai is sufficient to
degrade the cytoskeleton (Duncan et al., 2013), resulting in neurons retracting from their muscle
partners and impairing axonal transport (Duncan et al., 2013; Eaton et al., 2002; E. R. Graf et
al., 2011). stai mutants also show a drastic reduction of bouton numbers toward the more
posterior segments in larvae. Due to these phenotypes, staiB200 animals are robust models for
neurodegeneration. Similar roles for stai are seen in its mammalian orthologs. Loss of Superior
Cervical Ganglion 10 (SCG10) proteins within distal regions of axons after injury induces rapid
degeneration (Shin et al., 2012), and accumulations of SCG10 in the damaged ends of proximal
axons promotes regrowth (Shin, Geisler, & DiAntonio, 2014). Interestingly, SCG10 is rapidly
degraded due to phosphorylation by JNK proteins, further tying stathmin to Wnd. Due to these
perplexing mechanisms, I aim to uncover the interactions between stai function and Wnd
signaling.
To so this, we work with the hiwΔN
, staiB200, and the double mutant to quickly understand
their relationship. Surprisingly, we observe that the double mutant precisely phenocopies the
staiB200 mutant alone – the percentage of retractions and boutons counts matched at every
segment that we studied. This was first seen by myself, then verified by my undergraduate
mentor, Sara Wu. Thus, we hypothesize that Stai is epistatic to Hiw. However, it appears that
this is only the case with general synaptic morphology. Imaging of synapses by myself followed
by imaging analysis by Sara Wu reveals that key pre- and postsynaptic protein expressions are
altered in the double mutant relative to the single staiB200 mutant. Therefore, it is likely that injury
signaling is causing synaptic plasticity independently of stai-mediated neurodegeneration.

64
4.3 Results
The stathmin mutation is epistatic to the highwire mutation
To determine the association between injury signaling and stai function, we first
characterized the morphology and neurodegeneration at fly neuromuscular junctions (NMJs;
Fig. 1A). Since it is known that staiB200 induces greater levels of degeneration in more posterior
segments (E. R. Graf et al., 2011), we imaged synapses from segments A2 to A5 in larvae. We
noticed immediately that the stai + hiw double mutant looked identical to the stai mutant alone.
Further investigation of the bouton numbers showed a non-significant difference in both of these
conditions (Fig. 1B). Within each genotype, we saw the expected decrease in numbers as we
moved posteriorly. hiwΔN synapses consistently displayed double the boutons whereas staiB200
and stai + hiw both had half the boutons compared to wild type. This provided clear evidence
that morphology was phenocopied in both stai conditions.
We then thought to quantify levels of neurodegeneration at each segment by using the
well-established “footprint” assay (Eaton et al., 2002; E. R. Graf et al., 2011). For this type of
analysis, both the pre- and postsynaptic cells are immunolabeled with known markers. The
presynaptic motor neuron is typically stained with anti-vesicular glutamate transporter (vGlut) or
anti-bruchpilot (BRP), and the postsynaptic muscle is stained with anti-discs large (DLG; Fig.
1C). Any DLG expression seen without vGlut or BRP is marked as a “footprint” and defines the
NMJ as one that is experiencing retractions (Fig. 1D). Retractions were rarely seen in wild type
and hiwΔN throughout the larvae but were abundant in segment A5 for both staiB200 and stai +
hiw synapses. We interpreted this as increased degeneration in the posterior animal regions.
BRP intensities are altered in hiw; stai double mutants
Although the overall morphology and extent of degeneration were the same in the stai
conditions, we questioned whether there were differences at the molecular level. For the
remaining characterizations, we focused on the synapses with the most extreme phenotypes,

65
segments A2 (Fig. 2A) and A5 (Fig. 2C). We first analyzed the mean intensities of the active
zone scaffold, BRP (Fig. 2B) as a proxy for the amount of protein present at the synapse. hiw
mutants showed no difference at A2, but then were significantly reduced at A5 compared to wild
type. Similar trends were seen with the double mutant. However, stai mutants alone consistently
had reduced fluorescence intensity at both segments. It is possible that injury signaling played a
bigger role in BRP expression and trafficking than loss of stathmin.
vGlut intensities are consistently elevated in hiw; stai double mutants
We then turned to a different type of presynaptic player, vGlut, a synaptic vesicle protein
and bouton-filling reporter (Fig. 3A,C). There was a consistent reduction in expression in all
animals and in both segments (Fig. 3B). vGlut was reduced by about the same amount in both
segments in hiwΔN, which matches previous reports of how injury signaling reduces protein
expression (J. Li et al., 2017; Xiong et al., 2010). Likewise, staiB200 animals had a reduced level
of vGlut throughout the animal, with a greater decrease in segment A5. Given our
understanding of cytoskeletal dysregulation and deficient axonal transport due to loss of stai,
this result makes sense since both aspects are exacerbated in posterior regions (Duncan et al.,
2013). The most striking finding from this analysis is that vGlut expression is higher in stai + hiw
double mutants compared to stai.
GluRIIA intensities are consistently decreased in hiw; stai double mutants
After analyzing the effects of injury signaling and microtubule destabilization in the
presynaptic cell, we then turned our attention to the postsynapse. We initially looked at the
GluRIIA subunit that is an essential part of the “A Type” glutamate receptor in flies (Fig. 4A-C).
This type of receptor is considered to be stronger because it desensitizes slowly, allowing for
more current to pass through (DiAntonio et al., 1999). In a hiw mutant background, IIA receptor
levels decreased throughout the animal, matching remodeling trends seen previously (Goel &

66
Dickman, 2018). Receptors were downscaled in the A2, but not the A5 segment, in stai mutants.
Interestingly, IIA expression was always lower in the double mutant compared to wild type, but
consistently higher than in stathmin mutants.
GluRIIB intensities are consistently decreased in hiw; stai double mutants
The next logical step after studying the “A type” receptors was to study the other major
type at the NMJ. GluRIIB, or “B type” receptors play a redundant role at our favorite synapse.
Although they are considered to be weaker due to their faster desensitization kinetics, they are
sufficient to maintain synaptic function upon the loss of GluRIIA (DiAntonio et al., 1999). Quick
visual analysis of IIB puncta reveals similar expression trends as IIA in all conditions (Fig. 5A,C).
The quantifications also show similar trends as stated before (Fig. 5B). Once again, animals
undergoing injury (hiwΔN) exhibit reduced IIB levels compared to wild type. Unlike as seen with
the A type receptors, stai mutants had a consistent downscaling of B types throughout the
animal. Finally, stai + hiw double mutants had an ever further reduction of expression levels.
GluRIID intensities are decreased in segment A2 of hiw; stai double mutants
We finished the postsynaptic analysis with quantification of the total receptive field. For
this, we labeled one of the common subunits, GluRIID (Fig. 6A,C; (Qin et al., 2005)). In general,
the trends should match what we observed with each respective receptor type previously
analyzed. In fact, we saw a substantial reduction of total receptor expression in all mutant
conditions (Fig. 6B). At segment A2, the double mutant showed a reduction in IID levels, but the
A5 segment only had a modest, non-significant reduction.

67
4.4 Discussion
Here we have uncovered the previously unknown relationship between injury signaling
induced by loss of Hiw and microtubule destabilization by loss of Stathmin. Morphology and
degeneration analysis of the double mutants shows a clear resemblance to stathmin mutants
(Fig 1). Therefore, an attractive interpretation is that Stathmin is epistatic to Highwire – meaning
that hiw mutant phenotypes can only occur with proper stai expression. It is likely that Stai
functions downstream of Hiw since it is known that the hiw mutant growth phenotype requires
Wnd then JNK activity, and that mammalian SCG10 is a known target of JNK (Collins et al.,
2006; Shin et al., 2012). While JNK regulation of Stai has not been shown in flies, it stands to
reason that it can be targeted the same way as its mammalian ortholog. Additionally, there is
evidence that chronic disruptions to the cytoskeleton activates JNK signaling along with loss of
hiw to promote synaptic growth (Massaro, Pielage, & Davis, 2009). The fact that we do no
observe enhanced growth in the double mutant implies that Stai is not involved upstream in this
pathway.
Despite our evidence, we cannot deny the possibility that loss of stai leads to activation
of Wnd injury signaling and that the observed phenotypes are due to other molecular players. It
is well established that perturbations of microtubule structure and axonal transport are sufficient
to activate injury signaling (Asghari Adib et al., 2018; Li et al., 2017; Valakh et al., 2015). To
assess whether staiB200 activates Wnd and JNK signaling, we can utilize the reporter puc-lacZ
(Hao et al., 2019; J. Li et al., 2017; Xiong et al., 2010). This tool is localized to motor neuron
nuclei in the fly ventral nerve cord and its expression intensity can be used as a readout for Wnd
and JNK activation. We could generate a fly expressing both the stai mutation and the puc-lacZ
reporter to assess whether lack of stai leads to injury signaling.
Interpreting the synaptic plasticity in the presynaptic neuron is difficult due to the varying
trends of expression. With BRP (Fig. 2), we do not observe a difference in protein levels in hiw
alone or with the double mutants. However, we do see a reduction in stai mutants. Interestingly,

68
we then notice a reduction of BRP levels in hiw and the double mutant, with a greater decrease
in the latter. It is possible that while Wnd activation downregulates transcription (J. Li et al.,
2017), added issues with axonal transport in stai mutants (Duncan et al., 2013) greatly reduced
protein levels at axon terminals. This makes sense especially in segment A5 due to the greater
distance that proteins have to travel – any issues with trafficking would be more apparent given
the length of the axon. Conversely, the expression trends for vGlut (Fig. 3) are more
stereotyped in both segments. Already at A2, we observe a decrease in all genotypes. This may
differ from the BRP results due to the differences in transporting active zone and synaptic
vesicle protein precursors (Maas et al., 2012; Okada, Yamazaki, Sekine-Aizawa, & Hirokawa,
1995; Shapira et al., 2003). Segment A5 yields similar protein levels with an even greater
increase in staiB200, which again may be due to stronger axonal transport defects. Regardless of
segment however, stai + hiw synapses have greater vGlut protein expression compared to stai
alone. It is currently unknown how or why these animals overcompensate vGlut expression
when faced with both types of insults.
Glutamate receptor scaling has a more reliable trend of downscaling in our mutants (Fig.
4-6). It is well-known that muscles can sense injury in the presynaptic motor neurons and
reduce receptor abundance (Goel & Dickman, 2018). When comparing individual receptor types
and total receptors between hiwΔN and the double mutant, we see a greater number of GluRs in
the latter. Therefore, more than injury-related signaling must be affecting the muscle. However,
the double mutant has fewer GluRs compared to stai. It is attractive to think that differences in
transmission may be the cause. The enhancement in vGlut expression in the double mutant
may indicate a greater release of glutamate, which would lead to greater downscaling of
receptors (Y. Han et al., 2023). This hypothesis cannot be supported without electrophysiology
to verify differences in synaptic transmission.
Highwire is a powerful protein that is involved with many signaling pathways as an E3
ubiquitin ligase. Here, we extensively discuss its involvement with JNK signaling (Grill et al.,

69
2016; Massaro et al., 2009), but it is also a key player in Bone Morphogenetic Protein (BMP)
signaling (McCabe et al., 2004). Due to its involvement in various pathways as a key regulator,
it is possible that the hiw mutation is affecting more than Wnd signaling. To cover our bases, we
should repeat these experiments in a Wnd overexpression (Wnd-OE) background to eliminate
confounding factors.

70
4.5 Materials and Methods
Fly stocks: Drosophila stocks were raised at 25°C on standard molasses food. The w
1118
strain is used as the wild type control unless otherwise noted as this is the genetic background
in which all genotypes are bred. The following fly stock was used: hiwΔN (Wu et al., 2005). All
other stocks were obtained from Bloomington Drosophila Stock Center (BDSC): w
1118 (#5905)
and staiB200 (#16165).
Immunocytochemistry: Third-instar larvae were dissected in ice cold 0 Ca2+ HL-3 and
immunostained using a standard protocol as described (Goel et al., 2017; Y. Han et al., 2023).
In brief, larvae were either fixed in Bouin’s fixative for 5 min (Sigma, HT10132-1L) or 100% icecold methanol for 5 min. Larvae were then washed 3 times with phosphate buffered saline
(PBS) containing 0.1% Triton X-100 (PBST) for 30 minutes total, blocked with 5% Normal
Donkey Serum followed by overnight incubation in primary antibodies at 4°C. Preparations were
then washed 3 times in PBST for 30 minutes total, incubated in secondary antibodies for 2
hours at room temperature, washed 3 times in PBST for 30 minutes total, and equilibrated in
70% glycerol/PBS. Prior to imaging, samples were mounted in VectaShield (Vector
Laboratories). The following primary antibodies were used: mouse anti-GluRIIΑ (8B4D2; 1:50;
Developmental Studies Hybridoma Bank (DSHB)); rabbit anti-GluRIIIB (1:1000; (Perry et al.,
2017)); guinea pig anti-GluRIID (1:1000; (Perry et al., 2017)); guinea pig anti-vGlut (1:2000, (X.
Chen et al., 2017a)); mouse anti-bruchpilot (nc82; 1:100; DSHB); mouse anti-discs large (4F3;
1:100; DSHB); rabbit anti-discs large (1:10,000; (Pielage, Fetter, & Davis, 2005a)). Alexa Fluor647 conjugated goat anti-HRP (1:200; Jackson ImmunoResearch) and Donkey anti-mouse, -
guinea pig, and -rabbit conjugated Alexa Fluor 488, Cy3, and DyLight 405 secondary antibodies
(Jackson ImmunoResearch) were used at 1:400.

71
Imaging and analysis: Samples were imaged as described (Perry et al., 2017) using a
Nikon A1R Resonant Scanning Confocal microscope equipped with NIS Elements software and
a 100x APO 1.4NA oil immersion objective using separate channels with four laser lines (405
nm, 488 nm, 561 nm, and 647 nm). For fluorescence intensity quantifications of vGlut, BRP,
GluRIIA, GluRIIB and GluRIID, z-stacks were obtained on the same day using identical gain and
laser power settings with z-axis spacing of 0.125 µm for all genotypes within an individual
experiment. X- and Y-axis pixel sizes were set to 40 nm for proper sampling to then deconvolve
the image (Scientific Volume Imaging, SVI, Huygen’s software). SVI Huygen’s Object Analysis
software was then used to perform 3D analysis of BRP puncta, vGlut bouton expression, and
receptor clusters to determine “object” voxel numbers and sum intensity values. Mean intensity
was calculated by dividing sum intensity by the voxel numbers for each object. All
measurements based on confocal images were taken from NMJs acquired from at least six
different animals.
Retraction scoring: Retractions were scored as described originally (Eaton et al., 2002).
In brief, a retraction is defined when a postsynaptic reporter is seen without a partnered
presynaptic marker. Anti-DLG was our postsynaptic reporter of choice, and either anti-BRP or
anti-vGlut were used as our presynaptic reporter. Any synapse with even a single bouton
lacking either anti-BRP or anti-vGlut was scored as a retraction. Analysis was performed at
muscles 6/7 from segments A2 to A5.
Statistical analysis: Data were analyzed using GraphPad Prism or Microsoft Excel
software. Sample values were tested for normality using the D’Agostino & Pearson omnibus
normality test which determined that the assumption of normality of the sample distribution was
not violated. Data were then compared using a one-way ANOVA and tested for significance
using a Tukey’s multiple comparison test. All data are presented as mean +/-SEM; p denotes

72
the level of significance assessed as p≤0.05 (*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****);
ns=not significant. Statistical interpretations are provided for each figure in their respective
legends.

73
Figure 1: The stathmin mutation is epistatic to the highwire mutation. (A) Representative images
of muscle 6/7 synapses at segments A2 (top row) and A5 (bottom row) for the following
genotypes: wild type (w
1118); hiw (hiwΔN); stai (w; staiB200); stai + hiw (hiwΔN; staiB200). Horseradish
perosidase (HRP; gray) labels motor neurons. (B) Quantification of bouton numbers (at muscle
6/7) in the genotypes indicated in (A) from segments A2 to A5. (C) Representative image of
retractions in segment A5 of a stai mutant. Bruchpilot (BRP; green) marks the presynapse, and
discs large (DLG; magenta) marks the postsynapse. The asterisks denote areas of retractions
where the motor neuron has degenerated and pulled away from its partnered muscle. (D)
Quantification of NMJs with retractions (at muscle 6/7) in the genotypes indicated in (A) from
segments A2 to A5. Data points in (A) represent one synapse and data points in (D) represent
one animal. Error bars indicate ±SEM. One-way ANOVA test was performed, followed by a
Tukey’s multiple-comparison test. *** P≤0.001; **** P≤0.0001; ns, not significant (P>0.05).
Statistical markings directly above each bar represents the comparison to wild type. Statistics
comparing stai to stai + hiw are denoted above the lines hovering over the respective bars.
These experiments were first performed by Nancy Tran, then later confirmed by Sara Wu.

74
Figure 2: BRP intensities are altered in hiw; stai double mutants. (A) Representative images of
BRP puncta in segment A2 for the following genotypes: wild type (w
1118); hiw (hiwΔN); stai (w;
staiB200); stai + hiw (hiwΔN; staiB200). (B) Quantification of BRP mean fluorescence intensity for
hiw, stai, and stai + hiw in segment A2 (left) or segment A5 (right). All data is normalized to wild
type. (C) Representative images of BRP puncta in segment A5 for the genotypes indicated in
(A). Data points represent single “objects” defined by the SVI analysis program. Error bars
indicate ±SEM. One-way ANOVA test was performed, followed by a Tukey’s multiplecomparison test. **P<0.01; *** P≤0.001; **** P≤0.0001; ns, not significant (P>0.05). Statistical
markings directly above each bar represents the comparison to wild type. Statistics comparing
stai to stai + hiw are denoted above the lines hovering over the respective bars. Nancy Tran
stained and imaged the samples, then Sara Wu analyzed the images.

75
Figure 3: vGlut intensities are consistently elevated in hiw; stai double mutants. (A)
Representative images of vGlut in segment A2 for the following genotypes: wild type (w
1118); hiw
(hiwΔN); stai (w; staiB200); stai + hiw (hiwΔN; staiB200). (B) Quantification of vGlut mean
fluorescence intensity for hiw, stai, and stai + hiw in segment A2 (left) or segment A5 (right). All
data is normalized to wild type. (C) Representative images of vGlut in segment A5 for the
genotypes indicated in (A). Data points represent single “objects” defined by the SVI analysis
program. Error bars indicate ±SEM. One-way ANOVA test was performed, followed by a
Tukey’s multiple-comparison test. *P≤0.05; **P<0.01; **** P≤0.0001. Statistical markings directly
above each bar represents the comparison to wild type. Statistics comparing stai to stai + hiw
are denoted above the lines hovering over the respective bars. Nancy Tran stained and imaged
the samples, then Sara Wu analyzed the images.

76
Figure 4: GluRIIA intensities are consistently decreased in hiw; stai double mutants. (A)
Representative images of GluRIIA puncta in segment A2 for the following genotypes: wild type
(w
1118); hiw (hiwΔN); stai (w; staiB200); stai + hiw (hiwΔN; staiB200). (B) Quantification of GluRIIA
mean fluorescence intensity for hiw, stai, and stai + hiw in segment A2 (left) or segment A5
(right). All data is normalized to wild type. (C) Representative images of GluRIIA puncta in
segment A5 for the genotypes indicated in (A). Data points represent single “objects” defined by
the SVI analysis program. Error bars indicate ±SEM. One-way ANOVA test was performed,
followed by a Tukey’s multiple-comparison test. **** P≤0.0001; ns, not significant (P>0.05).
Statistical markings directly above each bar represents the comparison to wild type. Statistics
comparing stai to stai + hiw are denoted above the lines hovering over the respective bars.
Nancy Tran stained and imaged the samples, then Sara Wu analyzed the images.

77
Figure 5: GluRIIB intensities are consistently decreased in hiw; stai double mutants. (A)
Representative images of GluRIIB puncta in segment A2 for the following genotypes: wild type
(w
1118); hiw (hiwΔN); stai (w; staiB200); stai + hiw (hiwΔN; staiB200). (B) Quantification of GluRIIB
mean fluorescence intensity for hiw, stai, and stai + hiw in segment A2 (left) or segment A5
(right). All data is normalized to wild type. (C) Representative images of GluRIIB puncta in
segment A5 for the genotypes indicated in (A). Data points represent single “objects” defined by
the SVI analysis program. Error bars indicate ±SEM. One-way ANOVA test was performed,
followed by a Tukey’s multiple-comparison test. **** P≤0.0001; ns, not significant (P>0.05).
Statistical markings directly above each bar represents the comparison to wild type. Statistics
comparing stai to stai + hiw are denoted above the lines hovering over the respective bars.
Nancy Tran stained and imaged the samples, then Sara Wu analyzed the images.

78
Figure 6: GluRIID intensities are decreased in segment A2 of hiw; stai double mutants. (A)
Representative images of GluRIID puncta in segment A2 for the following genotypes: wild type
(w
1118); hiw (hiwΔN); stai (w; staiB200); stai + hiw (hiwΔN; staiB200). (B) Quantification of GluRIID
mean fluorescence intensity for hiw, stai, and stai + hiw in segment A2 (left) or segment A5
(right). All data is normalized to wild type. (C) Representative images of GluRIID puncta in
segment A5 for the genotypes indicated in (A). Data points represent single “objects” defined by
the SVI analysis program. Error bars indicate ±SEM. One-way ANOVA test was performed,
followed by a Tukey’s multiple-comparison test. **P<0.01; **** P≤0.0001; ns, not significant
(P>0.05). Statistical markings directly above each bar represents the comparison to wild type.
Statistics comparing stai to stai + hiw are denoted above the lines hovering over the respective
bars. Nancy Tran stained and imaged the samples, then Sara Wu analyzed the images.

79
Chapter 5: Peripheral glia perform diverse roles during degeneration
and injury

80
5.1 Abstract
Glial cells are incredibly versatile synaptic members. In both the central and peripheral
nervous systems, they can perform both positive and negative roles. They are known to
promote neuronal growth, mediate synaptogenesis, induce death, and clear debris. While there
is some insight into how perisynaptic Schwann Cells act after physical injury, it is unknown how
peripheral glia act with neurons experiencing injury signaling or during neurodegeneration.
Here, we attempt to study glial function at the Drosophila neuromuscular junction in these
aforementioned conditions. We first verify our genetic tools to isolate the differently glial cell
types in the fly periphery. Next, we utilize those tools to cleanly ablate all glia without
compromising the animal. We then find that glia may play several roles at the same synapse
depending on the developmental timeframe and neuronal partner. Finally, we discover that glia
may play a major role in maintaining synapses with excessive growth.

81
5.2 Introduction
The neuromuscular junction (NMJ) is a tripartite synapse consisting of a motor neuron,
muscle, and glial cells (Hall & Sanes, 1993). Sadly, glia are typically forgotten and, as such, are
understudied key players in the system. Unfortunately, this is even more so at the Drosophila
NMJ, which is disappointing considering the versatility of this model in studying axonal injury,
neurodegeneration, and other stressors to the synapse. Thankfully, we can gain insights from
prior studies in other organisms and in the central nervous system (CNS).
Glia in the CNS not only perform supportive and maintenance roles, but also synaptic
formation and plasticity roles under normal conditions (Nicola J. Allen & Barres, 2009). Microglia
are known for their phagocytic ability to clear debris, but they are also important for myelin
maintenance, guidance for dopaminergic neurons, and survival of specific interneurons (Bohlen,
Friedman, Dejanovic, & Sheng, 2019). Astrocytes are important for synaptogenesis and
maintenance of newly formed synapses (N. J. Allen & Barres, 2005). However, astrocytes and
microglia become “reactive” when neurons are damaged to promote cell death and clearance
(Aldskogius, Liu, & Svensson, 1999; Liddelow & Barres, 2017). Amazingly, blocking this
reactivity is sufficient to prevent death (Liddelow et al., 2017). While studies in the CNS tell us
what happens during the early stages of damage and injury, neurons here are incapable of
repairing. Therefore, we must now turn to the peripheral nervous system (PNS) to understand
glial involvement in neuronal regeneration.
The NMJ serves as a powerful model synapse for studying the interactions between the
cells in its tripartite structure: non-myelinating perisynaptic Schwann Cells (PSCs), motor
neurons, and muscles (Hall & Sanes, 1993; Sanes & Lichtman, 1999, 2001). Early in
development when neurons grow towards developing myotubes, PSCs are seen extending
beyond the axons as guides in mice (Barik, Li, Sathyamurthy, Xiong, & Mei, 2016) and frogs
(Herrera, Qiang, & Ko, 2000). Glia in flies also promote growth and synaptogenesis (FuentesMedel et al., 2012). In vertebrate systems, when neurons extend towards muscles in the

82
absence of glia, their axons will form a transient connection before retracting (Feng & Ko, 2008),
highlighting the importance of glia in this process. The same cellular mechanisms are seen after
injury. After crushing or cutting a nerve, PSCs extend projections out to nearby healthy
synapses (Kang et al., 2003; Reynolds & Woolf, 1992; Y.-J. Son & W. J. Thompson, 1995; Y. J.
Son & W. J. Thompson, 1995). During this phase, PSCs peel away from the old synapse and
form new connections on other parts of the muscle (Kang et al., 2014). The postsynaptic
compartment is remodeled as the uncovered, old synapses lose AChRs and the nascent
synapses form AChR clusters. Thus, developmental programs are turned on again after
neuronal damage to promote regrowth and reinnervation.
We have mentioned how glia perform positive roles in the PNS, however, they are
capable of performing negative roles as well. PSCs can fragment, eliminate, and engulf neurons
in mammals (Bishop et al., 2004; Ko & Robitaille, 2015; Vaquie et al., 2019) and flies (FuentesMedel et al., 2009). These responses imply that glia can sense nearby damaged neurons.
Drosophila peripheral glia have not been studied in the context of neurodegeneration until
recently (Ozdowski, Wentzell, Engert, Abbott, & Sherwood, 2020). This study finds that glia
became “reactive” and send out processes, providing a foundation for my dissertation work
here.
To understand the roles of Drosophila glia, we must first isolate the different types of
peripheral glia. There are three main glial subtypes at the fly NMJ: the wrapping glia (WG), the
subperineurial glia (SPG), and the perineurial glia (PG) that all concentrically ensheath the
axons (Freeman & Doherty, 2006). The key layer for glial stability is the SPG layer (Leiserson,
Harkins, & Keshishian, 2000) and will be the target of my manipulations. Jerry Chien in our lab
has identified glial subtype expression and has developed a clean method of ablating all glia by
driving pro-apoptotic genes, reaper (rpr) and head involution defective (hid), in the
subperineurial glia (SPG) layer. I will use these tools to understand how glia are involved during
neurodegeneration and axonal injury.

83
5.3 Results
Glia subtype-specific Gal4 drivers reveal their distinct expression patterns at the Drosophila
neuromuscular junction
To characterize the three glial subtypes at the Drosophila NMJ, Jerry first characterized
their localization. This was accomplished via the Gal4/Upstream Activating Sequence binary
expression system (Brand & Perrimon, 1993) and generous gifts from Dr. Chun Han at Cornell.
He developed specific Gal4 lines that only expressed in the NMJ and not the CNS. These lines
were used to drive expression of GFP within membranes (Fig. 1A) and compared to neuronal
membranes. We found that wrapping glia (WG) expressed entirely in the nerve bundle (Fig. 1A
top). Perineurial glia (PG) and Subperineurial (SPG) both expressed in the nerve and extended
toward the terminals without invading bouton regions (Fig. 1A middle and bottom). With this
expression data, we knew that these lines were clean and could be used for future experiments.
Ablating subperineurial glia ablates all peripheral glia
The benefit of these tools is to specially manipulate each glial type and gain insights into
their functions. Jerry systematically ablated each layer by using the three Gal4 lines to drive
apoptosis via UAS-rpr, UAS-hid (Fig. 2A). To verify whether only the single layer was ablated
compared to the entire glial population, we utilized the UAS-Infrared Fluorescent Protein (UASIFP) within each Gal4 line as the subtype reporter. Overall glial expression was gauged with a
reporter driven by a pan-glial promoter, Repo-tdTomato. We found that WG were only partially
ablated since we could see IFP aggregates in the nerve and that all glia were preserved (Fig.
2A top). PG ablation yielded no noticeable ablation of either perineurial glia or total glia (Fig. 2A
middle). However, SPG ablation killed all glia at the NMJ, as evidenced by the lack of tdTomato
expression (Fig. 2A bottom). Given this data, it was clear that the SPG layer was necessary for
total glial health. This also verified that our manipulation was sound and that it would be a
powerful method of removing all glia from NMJs for future experiments.

84
Peripheral glia are less reactive in synapses undergoing enhanced neurodegeneration
We wanted to uncover glial roles during neurodegeneration, so we employed the
stathmin (stai) mutation, staiB200. Stai in flies is a microtubule-associated protein that promotes
the stability of microtubules. Upon loss of stai, axon bundles are disorganized, axonal transport
is hindered, and motor neurons retract from muscles (Duncan et al., 2013; E. R. Graf et al.,
2011). Neurodegeneration became progressively worse at more posterior abdominal segments.
We first coupled Repo-tdTomato with staiB200 and observed glia projecting into bouton spaces
(Fig. 3A), which is highly uncharacteristic of glia. We then scored the number of boutons that
were invaded by glia, modeling our analysis on Ozdowski’s published work (Ozdowski et al.,
2020). Stai-induced neurodegeneration is more pronounced at posterior segments (E. R. Graf et
al., 2011), so we analyzed glial invasions from segments A2 to A5 (Fig. 3B). We found that glia
were more reactive in stai segment A2 compared to wild type, but not at the other segments.
We interpreted this to mean that glia were actively sensing the synapses experiencing low
levels of degeneration. This is possibly in preparation to degrade and phagocytose the boutons.
However, the glia at the more posterior segments were less active since they already cleared
neuronal material and had switched to a maintenance role.
Glia ablation in stathmin mutants reveals that glia may be required for neuronal clearance
We wanted to better understand the role of glia during neurodegeneration, so we ablated
glia and observed the effects on synaptic morphology. Visual analysis of the most extreme
synapses at segments A2 and A5 showed that ablating glia preserved boutons (Fig. 4A). Formal
quantification of bouton numbers in all segments revealed interesting trends (Fig. 4B). We found
that bouton numbers between stai and stai + glia ablation were similar in segment A2. However,
that difference widened progressively as we moved posteriorly. Unfortunately, there was no
statistical difference between the two groups, but that might be due to the low sample sizes in

85
this preliminary data. We posited that glia may be playing phagocytic roles at the synapses
undergoing enhanced neurodegeneration.
Glia may perform different functions for each motor neuron subtype
The previous experiment looked at total bouton numbers. However, the NMJ is actually
populated with two types of motor neurons, MN-Ib and MN-Is (Aponte-Santiago & Littleton,
2020; Newman et al., 2017), each with their own unique properties (He et al., 2023). Several
people in lab have observed that MN-Is seemed to be more susceptible to stress and are the
first type to die (unpublished data). So, we thought to look at glial function with each type of
neuron between stai and stai + glia ablation conditions. The two types were differentiated by
discs large (DLG) expression, with Ib neurons apposing postsynaptic densities with greater
levels of DLG (Fig. 5A). We saw that greater numbers of Ib boutons were preserved upon glial
ablation, especially in segment A5 (Fig. 5B). Conversely, fewer Is boutons were seen with glial
ablation throughout the animal (Fig. 5C). We took this to mean that glia performed phagocytic
functions with Ib neurons and maintenance functions with Is boutons. Again, while the data was
not significant, this might be due to the low sample size. However, it is interesting to think that
glia may have distinct roles for different neuron types innervating the same target.
Peripheral glia are highly reactive at synapses experiencing exuberant growth
Up to this point, we had looked at glia during wild type conditions and during
neurodegeneration. However, we were still curious how glia worked at synapses experiencing
injury where both regeneration and degeneration were occurring. To accomplish this, we utilized
the highwire mutation, hiwΔN to activate Wnd injury signaling (Asghari Adib et al., 2018; Jin &
Zheng, 2019; Watkins et al., 2013). These mutants display exuberant neuronal growth (Fig. 6A),
but even more so for glia (Fig. 6B). Since hiw affects the whole animal uniformly, we focused
our attention to our standard segments A2 and A3. We saw an incredible increase of glial

86
invasions, with a greater ratio of invaded boutons compared to the total bouton number in our
conditions (data not shown). Based on this relationship, we hypothesized that glia were
maintaining the massively overgrown neurons.

87
5.4 Discussion
Here we have gained insight to how peripheral glia respond during neurodegeneration
and axonal injury. We first defined the expression regions for the three types of glia (wrapping,
subperineurial, and perineurial) at the Drosophila NMJ using clean Gal4 driver lines (Fig. 1). We
find that only the SPG and PG layers interact with NMJs after branching off from the nerve
bundle. After confirming the efficacy of these lines, we then use them to ablate each layer (Fig.
2). We are pleased to find that ablating the vital SPG layer sufficiently kills all glia. This provides
a tool to study how the loss of glia at the NMJ affects the system without outright killing the
animals, which is what happens when apoptosis is induced in all glia. Now, we have an
established toolkit for ablating peripheral glia.
With these tools in hand, we then look at glial activity during neurodegeneration caused
by microtubule instability with the loss of stai (Duncan et al., 2013; E. R. Graf et al., 2011). We
initially observe that glia are surprisingly dormant in synapses experiencing heightened
degeneration (Fig. 3), leading us to believe that glia are maintaining synapses. However, loss of
glia shows that more of the overall synaptic structure is preserved in destabilized synapses (Fig.
4), making us think that glia are actually phagocytosing damaged neurons. Even more
confusingly, this seems to be the case for MN-Ib, but MN-Is rely on glia for maintenance (Fig. 5).
To test the hypothesis that glia are selectively phagocytosing Ib neurons, we can employ
mutations in the engulfment receptor, Draper (MacDonald et al., 2006; McLaughlin, PerryRichardson, Coutinho-Budd, & Broihier, 2019). If we observe increased bouton numbers
compared to stai alone, then that would support our hypothesis. Alternatively, it is entirely
possible that peripheral glia are phagocytosing degenerating neurons, then switching to a
maintenance role. When we ablate glia, we are removing them from the NMJ early in
development and are observing a preservation of boutons. Therefore, it is plausible that
phagocytosis occurs in earlier developmental stages. When we look at glial dynamics as a
whole, we are observing a current “snapshot” of glial dynamics. So, it is possible that at glia

88
exist to maintain synapses in later larval stages. A clean method to observe this shift in
functions is live imaging of neurons and glia throughout the larval lifespan.
A limitation of our system is that neurodegeneration can only be studied for the few days
that our animals are at the third instar stage. Thankfully, we can genetically arrest development
and prevent pupation (Perry et al., 2020), at least, if our system allows for these manipulations.
Unfortunately, both glial ablation and developmental arrest rely on the Gal4/UAS system (Brand
& Perrimon, 1993). A single Gal4 line is capable of activating all UAS constructs in the system,
which would kill the animal in this case. We would have to employ a different binary expression
system, LexA/LexAOp (Szüts & Bienz, 2000; Venken et al., 2011), for one of our manipulations.
Sadly, our attempt to generate this system failed.
Finally, we look at glial dynamics during genetically-induced injury, a state with both
degenerative and regenerative signaling (Fig. 6). Here we find that glia are much more active in
a system that causes excess neuron growth, and we conclude that the glia are maintaining the
synapse. The next logical step is to ablate glia, however, we once again face limitation due to
genetics. hiwΔN and UAS-rpr, UAS-hid are on the X chromosome and attempts to recombine all
components into a single animal failed. An alternative is to CRISPR a hiw mutant in the rpr, hid
background, but our lab currently does not have an efficient hiw-guide RNA line for
mutagenesis.
With these quick experiments and preliminary data, we have uncovered potentially
interesting glial roles during active degeneration and axonal injury. The most interesting of
which is a potential for glia to perform varying roles for two motor neurons innervating the same
muscle. This work provides a foundation for exciting research about glia during disease states in
a genetically power Drosophila model.

89
5.5 Materials and Methods
Fly stocks: Drosophila stocks were raised at 25°C on standard molasses food. The w
1118
strain is used as the wild type control unless otherwise noted as this is the genetic background
in which all genotypes are bred. The following fly stocks were used: R57G03-Gal4, UAS-CD4-
IFP-2A-H01, R47A02-Gal80 (“wrapping glia Gal4”; gift from Chun Han); R26B09-Gal4, UASCD4-IFP-2A-H01 (“subperineurial glia Gal4”; gift from Chun Han); R20B11-Gal4, UAS-CD4-IFP2A-H01, R40B03-Gal80 (“perineurial glia Gal4”; gift from Chun Han); UAS-CD4-tdGFP (gift from
Chun Han); Repo-CD4-tdTomato (gift from Chun Han); UAS-rpr, UAS-hid (Zhou et al., 1997);
hiwΔN (Wu et al., 2005). All other stocks were obtained from Bloomington Drosophila Stock
Center (BDSC): w
1118 (#5905) and staiB200 (#16165).
Immunocytochemistry: Third-instar larvae were dissected in ice cold 0 Ca2+ HL-3 and
immunostained using a standard protocol as described (Goel et al., 2017; Y. Han et al., 2023).
In brief, larvae were either fixed in Bouin’s fixative for 5 min (Sigma, HT10132-1L), 100% icecold methanol for 5 min, or 4% paraformaldehyde (PFA) for 10 min. Larvae were then washed 3
times with phosphate buffered saline (PBS) containing 0.1% Triton X-100 (PBST) for 30 minutes
total, blocked with 5% Normal Donkey Serum followed by overnight incubation in primary
antibodies at 4°C. Preparations were then washed 3 times in PBST for 30 minutes total,
incubated in secondary antibodies for 2 hours at room temperature, washed 3 times in PBST for
30 minutes total, and equilibrated in 70% glycerol/PBS. Prior to imaging, samples were mounted
in VectaShield (Vector Laboratories). The following primary antibodies were used: guinea pig
anti-vGlut (1:2000, (X. Chen et al., 2017a)); mouse anti-discs large (4F3; 1:100; DSHB). Alexa
Fluor-647 conjugated goat anti-HRP (1:200; Jackson ImmunoResearch) and Donkey antimouse and -guinea pig conjugated Alexa Fluor 488, Cy3, and DyLight 405 secondary antibodies
(Jackson ImmunoResearch) were used at 1:400.

90
Imaging: Samples were imaged as described (Perry et al., 2017) using a Nikon A1R
Resonant Scanning Confocal microscope equipped with NIS Elements software, a 60X APO
1.4NA oil immersion objective (for muscle 6/7 imaging), and a 100x APO 1.4NA oil immersion
objective (for muscle 4 imaging) using separate channels with four laser lines (405 nm, 488 nm,
561 nm, and 647 nm).
Bouton counting and retraction scoring: Synapses were labeled with synaptic vesicle
protein anti-vGlut and each punctum of vGlut was counted as a bouton. Anti-DLG staining in the
postsynaptic density defined Ib (with stronger signals) vs. Is (with weaker signals) bouton types.
Retractions were scored as described originally (Eaton et al., 2002). In brief, a retraction is
defined when a postsynaptic reporter is seen without a partnered presynaptic marker. Anti-DLG
was our postsynaptic reporter of choice, and either anti-BRP or anti-vGlut were used as our
presynaptic reporter. Any synapse with even a single bouton lacking either anti-BRP or antivGlut was scored as a retraction. Analysis was performed at muscles 6/7 from segments A2 to
A5.
Reactive glia analysis: Bouton spaces (labeled with HRP) co-expressing the glia reporter
(Repo-tdTomato) were counted to gauge the reactivity of glia in various conditions. Analysis
was performed at muscles 6/7 from segments A2 to A5.
Statistical analysis: Data were analyzed using GraphPad Prism or Microsoft Excel
software. Sample values were tested for normality using the D’Agostino & Pearson omnibus
normality test which determined that the assumption of normality of the sample distribution was
not violated. Data were then compared using a one-way ANOVA and tested for significance
using a Tukey’s multiple comparison test. All data are presented as mean +/-SEM; p denotes
the level of significance assessed as p≤0.05 (*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****);

91
ns=not significant. Statistical interpretations are provided for each figure in their respective
legends.

92
Figure 1: Glia subtype-specific Gal4 drivers reveal their distinct expression patterns at the
Drosophila neuromuscular junction. (A) Representative images of muscle 4 synapses showing
motor neurons (labeled with horseradish peroxidase, HRP, in gray) and their associated glia
(labeled with green fluorescent protein, GFP, in green). Drivers for each glial subtype are used
to drive GFP expression. Genotypes: wrapping glia (w; ; wpG-Gal4/UAS-CD4-tdGFP);
perineurial glia (w; ; pG-Gal4/UAS-CD4-tdGFP); subperineurial glia (w; ; spG-Gal4/UAS-CD4-
tdGFP). This experiment was performed by Jerry Chien.

93
Figure 2: Ablating subperineurial glia ablates all peripheral glia. (A) Representative images of
muscle 4 synapses showing motor neurons (HRP, gray), glial-subtypes (labeled with infrared
fluorescent protein, IFP, in green), and total glia (labled with tdTomato, in red). Drivers for each
glial subtype are used to drive IFP expression. Genotypes: wrapping glia (w, UAS-Rpr, UASHid/+; Repo-tdTomato/+; wpG-Gal4, UAS-CD4-IFP/+); perineurial glia (w, UAS-Rpr, UAS-Hid/+;
Repo-tdTomato/+; pG-Gal4, UAS-CD4-IFP/+); subperineurial glia (w, UAS-Rpr, UAS-Hid/+;
Repo-tdTomato/+; spG-Gal4, UAS-CD4-IFP/+). This experiment was performed by Jerry Chien.

94
Figure 3: Peripheral glia are less reactive in synapses undergoing enhanced
neurodegeneration. (A) Representative images of muscle 6/7 synapses at segment A2. Motor
neurons are labeled with HRP (magenta) and glia are labeled with tdTomato (green). Zoomed
insets show representative regions where glia sit at bouton margins (arrows) or where glia
invade bouton spaces (arrowheads). Genotypes: wild type (w
1118); stai (w; staiB200). (B)
Quantification of the number of boutons invaded by reactive glia in the genotypes indicated in
(A) and from segments A2 to A5. Data points represent synapses. Error bars indicate ±SEM.
One-way ANOVA test was performed. **** P≤0.0001; ns, not significant (P>0.05).

95
Figure 4: Glia ablation in stathmin mutants reveals that glia may be required for neuronal
clearance. (A) Representative images of muscle 6/7 synapses at segments A2 (top row) or A5
(bottom row). Vesicular glutamate transporter (vGlut; green) immunostaining marks boutons.
HRP (magenta) labels neuronal membranes. Genotypes: wild type (w
1118); stai (w; staiB200); stai
+ glia ablation (w, UAS-Rpr, UAS-Hid/+; staiB200; spG-Gal4/+). (B) Quantification of bouton
numbers per NMJ in the genotypes indicated in (A) and from segments A2 to A5. Data points
represent synpases. Error bars indicate ±SEM. One-way ANOVA test was performed, followed
by a Tukey’s multiple-comparison test. * P≤0.05; *** P≤0.001; **** P≤0.0001; ns, not significant
(P>0.05). Statistical markings directly above each bar represents the comparison to wild type.
Statistics comparing stai to stai + hiw are denoted above the lines hovering over the respective
bars.

96
Figure 5: Glia may perform different functions for each motor neuron subtype. (A)
Representative images of muscle 6/7 synapses at segments A2 (top row) or A5 (bottom row).
Discs large (DLG; gray) labels the postsynaptic density apposing motor neurons. Differences in
DLG expression is used to differentiate between Ib and Is neurons. Genotypes: wild type (w
1118);
stai (w; staiB200); stai + glia ablation (w, UAS-Rpr, UAS-Hid/+; staiB200; spG-Gal4/+). (B-C)
Quantification of Ib (B) or Is (C) bouton numbers per NMJ in the genotypes indicated in (A) and
from segments A2 to A5. Data points represent synpases. Error bars indicate ±SEM. One-way
ANOVA test was performed, followed by a Tukey’s multiple-comparison test. * P≤0.05; **
P≤0.01; *** P≤0.001; **** P≤0.0001; ns, not significant (P>0.05). Statistical markings directly
above each bar represents the comparison to wild type. Statistics comparing stai to stai + hiw
are denoted above the lines hovering over the respective bars.

97
Figure 6: Peripheral glia are highly reactive at synapses experiencing exuberant growth. (A)
Representative images of muscle 6/7 synapses at segment A2. Motor neurons are labeled with
HRP (magenta) and glia are labeled with tdTomato (green). Zoomed insets show representative
regions where glia sit at bouton margins (arrows) or where glia invade bouton spaces
(arrowheads). Genotypes: wild type (w
1118); hiw (hiwΔN). (B) Quantification of the number of
boutons invaded by reactive glia in the genotypes indicated in (A) and from segments A2 to A5.
Data points represent synapses. Error bars indicate ±SEM. One-way ANOVA test was
performed. **** P≤0.0001.

98
Chapter 6: Developing a Drosophila model to study loss of Unc13A
and its involvement in ALS-related neurodegeneration

99
6.1 Abstract
TDP43 is a potent regulator of mRNA splicing, processing, and trafficking. It is well
known that mislocalization and aggregation in the cytoplasm is a hallmark of Amyotrophic lateral
sclerosis (ALS) and related diseases. However, TDP43 roles in preventing cryptic exon
inclusion in key genes has recently gained traction. When TDP43 is missing, and a key trigger
in the form of single base substitutions occurs, these cryptic exons fail to be spliced. The result
is the addition of an extra, small exon that shifts the reading frame, causing nonsense-mediated
decay. Thus far, Unc13A and STMN2 have been identified as key targets for TDP43 protection.
Cryptic exon inclusions in either gene leads to ALS. Here, we attempt to generate a fly model to
study how the loss of TDP43 leads to reductions in Unc13A, and if this leads to enhanced
neurodegeneration. Without an antibody against Unc13A, we rely on a tagged reporter construct
and find that it works well. We find that a mutation in Unc13A does not lead to increased
degeneration, nor do double mutants with humanized TDP43 or fly TBPH. Finally, we find that
mutations in a microtubule stabilizing gene, stathmin, is sufficient to reduce Unc13 protein
levels, leaving a promising foundation for future work.

100
6.2 Introduction
Amyotrophic lateral sclerosis (ALS) is a disease marked by the loss of motor neuron
function ultimately leading to patient paralysis and death (Taylor, Brown, & Cleveland, 2016).
Unfortunately, there is currently no cure for the disease. Current treatments only prolong
lifespans of patients for a few months. The identified causes of familial ALS are mutations in
superoxide dismutase 1 (SOD1; (Taylor et al., 2016)) and a repeat of the (GGGGCC or G4C2)
sequence within the non-coding region of the chromosome 9 open reading frame 72 (C9orf72)
gene (DeJesus-Hernandez et al., 2011; Renton et al., 2011). The focus for my dissertation work
is the transactivate response DNA binding protein 43 kDa (TDP43).
Both sporadic and familial cases of ALS present with an accumulation of TDP43 in motor
neuron cytoplasm. Specifically, the aggregation of ubiquitinated TDP43, known as inclusion
bodies, acts as a strong predictor for ALS pathology (Neumann et al., 2006). As its name
suggests, TDP43 is a DNA- and RNA-binding protein (DBP; RBP) that regulates expression of
genes (Freibaum, Chitta, High, & Taylor, 2010). Protein localization analyses show that TDP43
closely associates with other messenger RBPs and can be seen at the axon terminal of mouse
motor neurons (Fallini, Bassell, & Rossoll, 2012; Narayanan et al., 2013). Thus, TDP43 also
likely plays a role in anterograde axonal transport to the terminal end. This was confirmed when
three different mutations resulted in hindered transport of TDP43-positive granules (Alami et al.,
2014) and transcripts for those same mutations were mislocalized without proper TDP43
function (Rotem et al., 2017). It is now known that TDP43 serves as an adapter that binds
mRNA to kinesin motors in a messenger ribonucleoprotein (mRNP) complex (Sahoo, Smith,
Perrone-Bizzozero, & Twiss, 2018). In addition to the aforementioned key functions, TDP43 has
been shown to prevent cryptic exon inclusions (Ling, Pletnikova, Troncoso, & Wong, 2015).
Cryptic exons (CEs) are smaller exons that are embedded within introns and are
normally spliced out. When specific triggers occur, these CEs are maintained in the mature
transcript (Mehta, Brown, Ward, & Fratta, 2023), leading to frameshifts and nonsense-mediated

101
decay. Typically, the result is a severe reduction in protein levels leading to disease pathology.
A recent study finds that loss of TDP43 leads to CE inclusions in an active zone gene, Unc13A
(Ma et al., 2022). Another article states that TDP43 works along with other heterogeneous
nuclear ribonucleoproteins (hnRNPs) to control CE inclusions within the same gene. Unc13A
being implicated in ALS is a surprise since it is involved with SNARE complex priming and
defining synaptic vesicle release sites (Jahn & Fasshauer, 2012; Reddy-Alla et al., 2017),
neither of which is obviously involved with ALS. Ma and colleagues find that while Unc13A CE
inclusions do not directly lead to disease, they are good predictors for enhanced disease
progression. They identify key single nucleotide polymorphisms (SNPs) located within the same
intron as the cryptic exon, providing a potential trigger for CE inclusion. While the presence of
cryptic exons in the genome is not fatal, the perfect storm leading to their inclusion can be.
Besides Unc13A, another gene has been implicated with CE inclusion leading to ALS.
Genetic screens using siTDP43 reveal that STMN2 (the gene encoding stathmin-2 or SCG10) is
especially sensitive to the TDP43 knock down (Klim et al., 2019). STMN2/SCG10 is part of the
stathmin family of microtubule-associated proteins that stabilizes the cytoskeleton. Axotomized
mouse neurons turn on JNK signaling that depletes SCG10, leading to axonal degeneration
(Shin et al., 2012) which likely contributes to the loss of motor neurons in ALS. However,
SCG10 is spared in TDP43 knock down when JNK is inhibited (Klim et al., 2019). In a more
direct approach, rescue of STMN2 expression after loss of TDP43 also restores axonal
regeneration (Melamed et al., 2019). Both studies by Klim et al. and Melamed et al. show
reduced SCG10 in ALS patient spinal cords and that loss of SCG10 leads to Frontotemporal
Lobar Degeneration due to TDP43 Proteinopathy (FTLD-TDP). Therefore, SCG10 is a key
maintenance protein for axon health and that loss of STMN2 contributes to ALS pathology.
In this dissertation work, we show that we have a consistent reporter for Unc13 protein
levels. We also show that mutations in Unc13A alone do not result in enhanced
neurodegeneration. Follow-up work with human TDP43 mutations or its fly ortholog, TBPH,

102
along with the Unc13A- also do no increase signs of degeneration. Finally, analysis of Unc13
protein levels in stathmin mutants reveals a reduction in the animal, with lower levels at
synapses experiencing greater degeneration. I generated all necessary fly lines and prepared
animals for our experiments weekly. I dissected, stained, and imaged Unc13-GFSTF animals,
and one of my undergraduate mentees, Sarah Relle, analyzed the images for mean intensity.
Another undergraduate mentee, Sunny Lee, performed the retraction experiments and analysis
with guidance from me.

103
6.3 Results
Unc13-GFSTF colocalizes with other active zone proteins and apposes glutamate receptor
clusters
This project was designed to compare the levels of Unc13A and define whether that
correlated with neurodegeneration. Unfortunately, we did not possess antibodies targeting
Unc13A protein, so we relied on a reporter instead. Unc13-GFSTF labeled both isoforms,
Unc13A and Unc13B, and provided a readout for total Unc13 protein at synapses (Dannhäuser
et al., 2022). To verify its localization, we co-expressed the reporter with a known active zone
scaffold, Bruchpilot (BRP), and a common glutamate receptor subunit, GluRIIC (Fig. 1). We
found that our tag colocalized precisely with the presynaptic BRP and apposed GluR clusters.
We felt this reporter was at least a good marker for our experiments.
Unc13-GFSTF is a clean reporter for unc13 protein levels
To gauge the ability of Unc13-GFSTF to be a reporter for protein levels, we analyzed
fluorescence intensity of the GFP in the tag in various genetic conditions. Endogenous GFP was
imaged, the images were deconvolved to remove excess background, and then they were
analyzed for mean intensity. The conditions we assessed were the homozygous stock, a
heterozygous animal, an animal with one copy of the mutant (Unc13-P84200), and an animal
with one copy of a deficiency (Fig. 2A). Deficiencies (Dfs) in the fly system are chromosomes
where a large portion is deleted; we typically select a Df line that covers only slightly more than
our gene of interest’s locus. We found a reduction in GFP intensity between 40 and 50% (Fig.
2B), which gave us confidence that Unc13-GFSTF was a reliable reporter for our future
experiments.
unc13A mutations alone do not induce neurodegeneration

104
Per the recent publication by Ma, we thought that loss of Unc13A would lead to
neurodegeneration (Ma et al., 2022). We noticed an obvious reduction in GFP expression,
which makes sense since an Unc13A mutant leaves only Unc13B protein (Fig. 3A). Formal
quantification revealed a 40% decrease in total Unc13 protein levels (Fig. 3B). Unfortunately,
retraction scoring (as described in (Eaton et al., 2002; E. R. Graf et al., 2011) in our mutant
revealed a non-significant difference compared to wild type (Fig. 3C). Thus Unc13A- alone is
insufficient to cause degeneration in a fly model.
unc13A mutations in hTDP43 backgrounds does not enhance neurodegeneration
Next, we thought to couple TDP43 mutations with Unc13Ato potentially compound the
manipulations and induce degeneration at our NMJs. We worked with “humanized” fly lines in
which the gene locus for the fly ortholog of TDP43 is replaced with normal or mutant human
TDP43 with substitutions seen in familial ALS (Fig. 4A; (Chang & Morton, 2017)). At baseline,
hTDP43 did not reduce Unc13 levels, however the hTDP43.Q331K and hTDP43.M337V did (by
19% and 5%, respectively; Fig. 4B). These modest reductions in Unc13 did not produce above
wild type levels of retractions (Fig. 4C). Loss of Unc13A in all conditions reduced Unc13-GFP
intensity, as expected (Fig. 4A). Interestingly, neither hTDP43 nor hTDP43.M337V reduced
GFP intensity below that of Unc13A- alone, but hTDP43.Q331K did (Fig. 4B). Sadly, there was
no increase of neurodegeneration in any condition, despite the further reduction of Unc13 levels
in the Q331K condition (Fig. 4C).
unc13A mutations in TBPH backgrounds does not enhance neurodegeneration
It is possible that the humanized TDP43 had no effect on the fly system, so we turned to
mutations in the fly ortholog, TAR DNA-binding protein-43 homolog (TBPH; (Feiguin et al.,
2009). We found that the TBPHΔ142 mutant somehow increased Unc13 levels by 63% (Fig.
5A,B). Once again, we observed an approximately 50% decrease in Unc13 proteins with

105
Unc13A-
, however, the added TBPH mutant only decreased it by 33% (Fig. 5A,B). We observed
a significant increase of retractions in TBPHΔ142 + Unc13Acompared to wild type, but the
difference was non-significant compared to Unc13A- alone (Fig. 5C).
stathmin mutation is sufficient to reduce Unc13 levels
Finally, we questioned if other conditions causing neuronal insults could reduce Unc13
levels. We focused on the stathmin (stai) mutant that destabilizes microtubules and causes
motor neurons to retract away from the muscles they are innervating (Duncan et al., 2013; E. R.
Graf et al., 2011). It is used as a potent neurodegeneration model where retractions become
more apparent as you move toward the posterior end. While imaging, we observed a striking
reduction of Unc13-GFP intensity in stai animals at segments A2 and A5 (Fig. 6A). There was
only 47% of the baseline GFP expression at A2, and only 42% at A5 (Fig. 6). Amazingly, stai
caused a greater reduction of GFP than Unc13A- did. It seemed that there was a positive
correlation between the level of neurodegeneration experienced by the synapse and the amount
of Unc13 expressed there. This provides exciting possible connections between Stathmin
function and ALS via Unc13A.

106
6.4 Discussion
Our work here has provided the foundation for a future project looking at Unc13 levels
during neurodegeneration. We verify that Unc13-GFSTF expresses well in the active zone and
across from glutamate receptors (Fig. 1). We also confirm that this reporter scales appropriately
with our heterozygous and mutant controls (Fig. 2). Also, approximately 50% of total Unc13
remains after loss of Unc13A (Fig. 3). Unfortunately, although there is evidence that CE
inclusion and reductions of Unc13A lead to neurodegeneration (Ma et al., 2022), we do not see
that in our Drosophila model. Combining humanized TDP43 (hTDP43) with Unc13A-
(Fig. 4) and
the fly TBPH ortholog with Unc13A-
(Fig 5) yields no additional neurodegeneration. The most
interesting finding from our work is that stai mutants reduce Unc13 proteins more than Unc13Adoes (Fig. 6). We find that the decrease in protein is even more pronounced with greater
amounts of neurodegeneration and microtubule instability. If this project is to be continued, it
would be best to use Unc13A- and Unc13B specific antibodies or markers to properly gauge
each isoform’s protein expression. While Unc13-GFSTF works well, we cannot dissect the two
types of Unc13 proteins with this reporter.
While cryptic exons do exist in the Drosophila genome, they are involved with recursive
splicing (Brian Joseph, Kondo, & Lai, 2018; B. Joseph, Scala, Kondo, & Lai, 2022). In this form
of splicing, introns are removed in a piecemeal manner by first splicing out the intronic region
between an upstream exon and a cryptic exon. These are specifically “recursive splicing” CEs
(RS-CEs) with tandem splice acceptor and slice donor sites known as the ratchet point (RP).
Once the upstream intron is removed, the RP splice donor at the 3’ end of the CE and an RSCE splice donor at the 5’ end of the CE remains. Typically, the RP splice donor is used to fully
remove the RS-CE, however, mutations in the RP splice donor allows for CE inclusion. Thus far,
there have been no reports of this recursive splicing mechanism leading to disease.
Finally, the finding that stai leads to a generous loss of Unc13 is exciting as this may be
the molecular connection between the loss of stathmin and progression to ALS or FTLD-TDP

107
(Melamed et al., 2019). Although, it is entirely possible that axonal trafficking defects due to the
loss of stai may reduce Unc13 protein at the axon terminal (Duncan et al., 2013). Further
intensity and retraction analysis must be performed in a stai + Unc13A- mutant. This is an
exciting interaction and process to pursue in future projects.

108
6.5 Materials and Methods
Fly stocks: Drosophila stocks were raised at 25°C on standard molasses food. The w
1118
strain is used as the wild type control unless otherwise noted as this is the genetic background
in which all genotypes are bred. The following fly stocks were used: Unc13A-
(this study);
Unc13A-
, Unc13-GFSTF (this study). All other stocks were obtained from Bloomington
Drosophila Stock Center (BDSC): w
1118 (#5905); Unc13-GFSTF (#97202); Df(4)02 (Unc13
Deficiency, #7084); hTDP43 (#93125); hTDP43.Q331K (#93127); hTDP43.M337V (#93126);
TBPHΔ142 (#93600); staiB200 (#16165) or the Kyoto stock center: Unc13 P84200 (#101911).
Molecular Biology: Unc13A- and Unc13A-
, Unc13-GFSTF mutants were generated using
a CRISPR/Cas9 genome editing strategy as described (Kikuma, Li, Kim, Sutter, & Dickman,
2017a). To generate Unc13A- mutants, a double guide RNA (2xgRNA) line that targeted the first
exon (5’ TCTCTTCATATTCTAACAAG 3’; 5’ TCAAGCCATTACGCAAATAC 3’) were each
cloned into the pU6 vector. Constructs were sent to BestGene Inc. (Chino Hill, CA) for targeted
insertion into the attP40 site on the second chromosome. 2xgRNA flies were crossed to a BamCas9 line on the third chromosome to induce active germline CRISPR mutagenesis, and 20
independent lines were screened by PCR to identify Unc13A- mutants. To generate the
Unc13A-
, Unc13-GFP line, the 2xgRNA flies were crossed to Bam-Cas9; Unc13-GFSTF flies to
induce active germline CRISPR mutagenesis, and 20 independent lines were screened by PCR
to identify mutants.
Immunocytochemistry: Third-instar larvae were dissected in ice cold 0 Ca2+ HL-3 and
immunostained using a standard protocol as described (Goel et al., 2017; Y. Han et al., 2023).
In brief, larvae were either fixed in 100% ice-cold methanol for 5 min to preserve endogenous
GFP from the Unc13-GFSTF fly stock. Larvae were then washed 3 times with phosphate
buffered saline (PBS) containing 0.1% Triton X-100 (PBST) for 30 minutes total, blocked with

109
5% Normal Donkey Serum followed by overnight incubation in primary antibodies at 4°C.
Preparations were then washed 3 times in PBST for 30 minutes total, incubated in secondary
antibodies for 2 hours at room temperature, washed 3 times in PBST for 30 minutes total, and
equilibrated in 70% glycerol/PBS. Prior to imaging, samples were mounted in VectaShield
(Vector Laboratories). The following primary antibodies were used: rabbit anti-GluRIIC (1:2000;
(Goel & Dickman, 2018)); mouse anti-bruchpilot (nc82; 1:100; DSHB). Alexa Fluor-647
conjugated goat anti-HRP (1:200; Jackson ImmunoResearch) and Donkey anti-mouse and -
rabbit conjugated Alexa Fluor Cy3 and DyLight 405 secondary antibodies (Jackson
ImmunoResearch) were used at 1:400.
Imaging and analysis: Samples were imaged as described (Perry et al., 2017) using a
Nikon A1R Resonant Scanning Confocal microscope equipped with NIS Elements software and
a 100x APO 1.4NA oil immersion objective using separate channels with four laser lines (405
nm, 488 nm, 561 nm, and 647 nm). For fluorescence intensity quantifications of Unc13-GFP
and BRP, z-stacks were obtained on the same day using identical gain and laser power settings
with z-axis spacing of 0.125 µm for all genotypes within an individual experiment. X- and Y-axis
pixel sizes were set to 40 nm for proper sampling to then deconvolve the image (Scientific
Volume Imaging, SVI, Huygen’s software). SVI Huygen’s Object Analysis software was then
used to perform 3D analysis of Unc13-GFP and BRP puncta to determine “object” voxel
numbers and sum intensity values. Mean intensity was calculated by dividing sum intensity by
the voxel numbers for each object. All measurements based on confocal images were taken
from NMJs acquired from at least six different animals.
Retraction scoring: Retractions were scored as described originally (Eaton et al., 2002).
In brief, a retraction is defined when a postsynaptic reporter is seen without a partnered
presynaptic marker. Anti-DLG was our postsynaptic reporter of choice, and anti-vGlut was used

110
as our presynaptic reporter. Any synapse with even a single bouton lacking anti-vGlut was
scored as a retraction. Analysis was performed at muscles 6/7 from segments A2 to A5.
Statistical analysis: Data were analyzed using GraphPad Prism or Microsoft Excel
software. Sample values were tested for normality using the D’Agostino & Pearson omnibus
normality test which determined that the assumption of normality of the sample distribution was
not violated. Data were then compared using a one-way ANOVA and tested for significance
using a Tukey’s multiple comparison test. All data are presented as mean +/-SEM; p denotes
the level of significance assessed as p≤0.05 (*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****);
ns=not significant. Statistical interpretations are provided for each figure in their respective
legends.

111
Figure 1: Unc13-GFP colocalizes with other active zone proteins and apposes glutamate
receptor clusters. Representative images of muscle 4 boutons expressing the active zone
scaffold bruchpilot (BRP; gray), Unc13 tagged with green fluorescent protein (Unc13-GFP;
green), a common glutamate receptor subunit (GluRIIC; magenta), and the merged image
(multicolor panel on the left). Genotype: Unc13-GFP (w; ; ; Unc13-GFSTF). Images have been
stabilized and deconvolved with the SVI Huygens Software.

112
Figure 2: Unc13-GFP is a clean reporter for unc13 protein levels. (A) Representative images of
muscle 4 boutons from various Unc13 conditions expressing Unc13-GFP (green). Genotypes:
Unc13-GFP (w; ; ; Unc13-GFSTF); Unc13-GFP/+ (w; ; ; Unc13-GFSTF/+); Unc13-GFP/Unc13-
P84200 (w; ; ; Unc13-GFSTF/ Unc13-P84200); Unc-13-GFP/Unc13 Df (w; ; ; Unc13-GFSTF/
Df(4)02). Images have been stabilized and deconvolved with the SVI Huygens Software. (B)
Quantification of mean fluorescence intensity of GFP in the genotypes mentioned in (A). All data
is normalized to homozygous Unc13-GFP (baseline) intensity. Data points represent single
“objects” defined by the SVI analysis program. Error bars indicate ±SEM. One-way ANOVA test
was performed, followed by a Tukey’s multiple-comparison test. **** P≤0.0001. Nancy Tran
stained and imaged the samples, then Sarah Relle analyzed the images.

113
Figure 3: unc13A mutations alone do not induce neurodegeneration. (A) Representative images
of muscle 4 boutons with total Unc13 expression (top) or Unc13B expression (bottom). Unc13-
GFP (green) is the reporter for expression. Genotypes: Unc13-GFP (w; ; ; Unc13-GFSTF);
Unc13A-
, Unc13-GFP (w; ; ; Unc13A-
, Unc13-GFSTF). Images have been stabilized and
deconvolved with the SVI Huygens Software. (B) Quantification of mean fluorescence intensity
of GFP in Unc13A-
, Unc13-GFP normalized to the baseline Unc13-GFP levels. Data points
represent single “objects” defined by the SVI analysis program. (C) Quantification of NMJs
(segments A2 to A5) with retractions in Unc13A- animals. Comparisons are made to retraction
data in wild type (w1118) animals. Data points represent animals. Error bars indicate ±SEM.
One-way ANOVA test was performed, followed by a Tukey’s multiple-comparison test. ****
P≤0.0001; ns, not significant (P>0.05). Nancy Tran stained and imaged the Unc13-GFP
samples, then Sarah Relle analyzed the images. Sunny Lee performed the complete retraction
scoring experiments.

114
Figure 4: unc13A mutation in hTDP43 backgrounds does not enhance neurodegeneration. (A)
Representative images of muscle 4 boutons expressing Unc13-GFP reporter in the baseline
conditions (top) or Unc13A- conditions (bottom). Various hTDP43 insertions are assessed here.
Genotypes: Unc13-GFP (w; ; ; Unc13-GFSTF); Unc13-GFP + Unc13A-
(w; ; ; Unc13A-
, Unc13-
GFSTF); hTDP43 (w; hTDP43; ; Unc13-GFSTF); hTDP43 + Unc13A-
(w; hTDP43; ; Unc13A-
,
Unc13-GFSTF); hTDP43.Q331K (w; hTDP43.Q331K; ; Unc13-GFSTF); hTDP43.Q331K +
Unc13A-
(w; hTDP43.Q331K; ; Unc13A-
, Unc13-GFSTF); hTDP43.M337V (w; hTDP43.M337V; ;
Unc13-GFSTF); hTDP43.M337V + Unc13A-
(w; hTDP43.M337V; ; Unc13A-
, Unc13-GFSTF).
Images have been stabilized and deconvolved with the SVI Huygens Software. (B)
Quantification of mean fluorescence intensity of GFP in the genotypes indicated in (A)
normalized to the baseline Unc13-GFP levels. Data points represent single “objects” defined by
the SVI analysis program. (C) Quantification of NMJs (segments A2 to A5) with retractions for
the following genotypes: Unc13A-
(w; ; ; Unc13A-
); hTDP43 (w; hTDP43); hTDP43 + Unc13A-
(w; hTDP43; ; Unc13A-
); hTDP43.Q331K (w; hTDP43.Q331K); hTDP43.Q331K + Unc13A-
(w;
hTDP43.Q331K; ; Unc13A-
); hTDP43.M337V (w; hTDP43.M337V); hTDP43.M337V + Unc13A-
(w; hTDP43.M337V; ; Unc13A-
). Comparisons are made to retraction data in wild type (w1118)
animals. Data points represent animals. Error bars indicate ±SEM. One-way ANOVA test was
performed, followed by a Tukey’s multiple-comparison test. * P≤0.05; *** P≤0.001; ****
P≤0.0001; ns, not significant (P>0.05). Statistical markings directly above each bar in (B)
represents the comparison to Unc13-GFP. Statistics comparing Unc13A- data to each other are
denoted above the brackets in (B). Comparisons amongst all genotypes in (C) revealed no
statistical significance. Nancy Tran stained and imaged the Unc13-GFP samples, then Sarah
Relle analyzed the images. Sunny Lee performed the complete retraction scoring experiments.

115
Figure 5: unc13A mutation in TBPH backgrounds does not enhance neurodegeneration. (A)
Representative images of muscle 4 boutons expressing Unc13-GFP reporter in the baseline
conditions (top) or unc13A- conditions (bottom). Genotypes: Unc13-GFP (w; ; ; Unc13-GFSTF);
Unc13-GFP + Unc13A-
(w; ; ; Unc13A-
, Unc13-GFSTF); TBPHΔ142 (w; TBPHΔ142; ; Unc13-
GFSTF); TBPHΔ142 + Unc13A-
(w; TBPHΔ142; ; Unc13A-
, Unc13-GFSTF). Images have been
stabilized and deconvolved with the SVI Huygens Software. (B) Quantification of mean
fluorescence intensity of GFP in the genotypes indicated in (A) normalized to the baseline
Unc13-GFP levels. Data points represent single “objects” defined by the SVI analysis program.
(C) Quantification of NMJs (segments A2 to A5) with retractions for the following genotypes:
Unc13A-
(w; ; ; Unc13A-
); TBPHΔ142 (w; TBPHΔ142); TBPHΔ142 + Unc13A-
(w; TBPHΔ142; ;
Unc13A-
). Comparisons are made to retraction data in wild type (w1118) animals. Data points
represent animals. Error bars indicate ±SEM. One-way ANOVA test was performed, followed by
a Tukey’s multiple-comparison test. * P≤0.05; **** P≤0.0001; ns, not significant (P>0.05).
Comparisons amongst all genotypes in (B) yielded **** significance. Statistical markings directly
above each bar in (C) represents the comparison to wild type. Statistics comparing Unc13Adata to each other are denoted above the line in (C). Nancy Tran stained and imaged the
Unc13-GFP samples, then Sarah Relle analyzed the images. Sunny Lee performed the
complete retraction scoring experiments.

116
Figure 6: stathmin mutation is sufficient to reduce unc13 levels. (A) Representative images of
muscle 4 boutons expressing the Unc13-GFP reporter in the base stock and in stai mutants.
Segments A2 (top) and A5 were analyzed. Genotypes: Unc13-GFP (w; ; ; Unc13-GFSTF); stai +
Unc13-GFP (w; staiB200; ; Unc13-GFSTF). Images have been stabilized and deconvolved with
the SVI Huygens Software. (B) Quantification of mean fluorescence intensity of GFP in stai +
Unc13-GFP normalized to the baseline Unc13-GFP levels at each respective segment. Data
points represent single “objects” defined by the SVI analysis program. Error bars indicate ±SEM.
One-way ANOVA test was performed, followed by a Tukey’s multiple-comparison test. ***
P≤0.001; **** P≤0.0001. Nancy Tran stained and imaged the Unc13-GFP samples, then Sarah
Relle analyzed the images.

117
Chapter 7: Conclusion

118
Our work with various types of neurodegeneration, axonal injury, and general neuronal
insults has revealed interesting synaptic compensation. Neurons are highly resilient and
adaptable, so while plasticity is to be expected, the nature of plasticity surprises us. We have
interrogated how preventing an animal from transitioning into its development alters its synaptic
functionality. But, even more importantly, we develop a power tool to extend future experiments
on neurodegeneration and/or with disease states. Next, we define how injury signaling along
with large synaptic cargo hinders axonal transport, leading to a concomitant upscaling in the
postsynaptic receptor abundance. Following that, we dissect the paradoxical relationship
between injury signaling and a known cytoskeletal stability disruptor, which is also known to be
a surveillance factor for injury. We then shift to the lesser known third player in the synapse,
glia. We find that they perform diverse roles when the neuron at the NMJ is experiencing either
injury signaling or neurodegeneration. Finally, we attempt to develop a Drosophila model to
study how TDP43 mutants affect Unc13A protein levels, and how that all leads to
neurodegeneration.
We first arrest our animals in the third instar stage and extend our studies of these
animals from only one day to ~28 days. We immediately notice an obvious increase in animal
size and the growth of muscle over this time frame. Interestingly, synaptic transmission remains
stable over the duration of the arrested third instar (ATI) lifespan. Looking at the neuronal
morphology, we observe a rapid growth of the terminal relative to the muscle (Schuster et al.,
1996). We also see an increase in anatomical release sites (Akbergenova et al., 2018; Wagh et
al., 2006). At this point, we think that surely the postsynaptic field will downscale to compensate
for the assumed increased transmission, but we instead see an increase in glutamate receptor
abundance. With an upregulation of both release sites and receptors, something else must be
downregulated to maintain stable transmission. In the end, we find that release probability is
hindered. Finally, and most applicably to the rest of the work here, we utilize the developmental
arrest toolkit to extend our studies of neurodegeneration in third instars.

119
Our next study attempts to answer whether PHD can be induced during injury signaling
as a follow-up to our PHP and injury story (Goel & Dickman, 2018). To induce PHD, we
overexpress vGlut to generate large vesicles that release more glutamate into the synapse
(Daniels et al., 2004). Secondly, we genetically induce injury with the hiwΔN mutant to activate
Wnd (Asghari Adib et al., 2018; Collins et al., 2006). Initial electrophysiological experiments
reveal that synapses can undergo what appears to be PHD: the mEPSP amplitudes are
increased and EPSP amplitude remains consistent, leading to halved quantal content. However,
knowing that the presynaptic neuron is actively experiencing injury-related signaling, we
continue to interrogate the system. The most striking, obvious phenotype is that hiw + vGlut-OE
synapses have enlarged terminal boutons. Previous reports have shown a terminal swelling due
to defects in dynein transport, and with Wnd involvement in axonal transport, we then
hypothesize there are issues with trafficking (Liu, 2017; Lloyd et al., 2012; Verhey, 2007).
Images at the axon terminal show a stark reduction in synaptic vesicle proteins in our
manipulation, and imaging in the nerve bundles shows greater “traffic jams” of vGlut protein in
hiw + vGlut-OE than in vGlut-OE alone. As a result of this, we then turn to the postsynaptic side
to observe scaling that offsets the reduced transmission (Pozo & Goda, 2010; G. G. Turrigiano,
2008). Amazingly, we observe a tripling of the receptor abundance. At this point, it is clear that
large vesicles are not being released, because we would see much larger mEPSP amplitudes
with the upscaled receptors. We hypothesize that the injury signaling with the enlarged vesicles
only allows for the release of normally-sized vesicles. Unfortunately, we do not have electron
microscopy data to verify this. Since the classic mechanism for PHD induction involves the
release of excess glutamate, we are not witnessing PHD here (X. Li et al., 2021; X. Li, P. Goel,
J. Wondolowski, et al., 2018). Instead, we have found a compensatory mechanism that
drastically increases receptor abundance to maintain homeostasis in an injury signaling
background. While transmission does not return to wild type levels, it does return to the
threshold defined by hiwΔN, as shown previously (Goel & Dickman, 2018).

120
Our following project probes the enigmatic interactions between a microtubule stabilizer,
Stathmin (Stai), and injury signaling. Disruptions in the cytoskeleton are known to induce Wnd
injury signaling (J. Li et al., 2017; Valakh et al., 2015), so it is possible that loss of stai could turn
on injury signaling. However, research in the mammalian ortholog of Stathmin, SCG10, shows
that it is a key surveillance factor during injury. It is degraded by JNK at the distal ends, but its
accumulation in the proximal axon promotes growth (Shin et al., 2014; Shin et al., 2012).
Therefore, it is unclear when and how Stai functions during injury signaling in Drosophila. We
first generally characterized hiwΔN and staiB200 single mutants, then the double mutants. In both
bouton counts and retraction scoring for neurodegeneration, we observe a similarity between
stai + hiw compared to staiB200 alone in all segments that we viewed. This provides clear
evidence that stai is needed for hiwΔN morphology and we conclude that Stai is epistatic to Hiw.
Although the overall morphology is the same, we want to see if there is remodeling at the preand postsynaptic compartments. When comparing stai + hiw to staiB200, we observe an increase
of BRP at proximal segment A2, but then a decrease at distal segment A5, and vGlut has a
more stereotyped increase at both segments. We attribute the differences in protein expression
to the axonal transport modalities and distances traveled (Maas et al., 2012; Okada et al., 1995;
Shapira et al., 2003). In the same conditions, we see an overall reduction of receptors, with a
greater reduction of GluRIIB in our double mutants. It may be possible the transmission is
enhanced in stai + hiw and that the postsynaptic side is compensating by downscaling (Y. Han
et al., 2023). Despite the similarities in morphology and neurodegeneration, it is clear that injury
signaling is causing remodeling beyond that of staiB200 alone. Also, it is likely that Stai functions
downstream of Hiw.
Beyond the motor neuron and muscle cells, we are curious how glia respond to
stressors in the neuron. There is evidence that after injury, glia will send projections out to
healthy synapses (Kang et al., 2003; Y.-J. Son & W. J. Thompson, 1995; Y. J. Son & W. J.
Thompson, 1995). This shows that glia are highly dynamic and capable of sensing when their

121
partners are experiencing insults. It is also known that glia facilitate neuronal degradation and
clearance (Bishop et al., 2004; Fuentes-Medel et al., 2009; Ko & Robitaille, 2015; Vaquie et al.,
2019), making them important members for neurodegeneration. On the opposite side of the
spectrum, glia can also promote growth and guidance (Barik et al., 2016; Fuentes-Medel et al.,
2012; Herrera et al., 2000), which are key for assisting injured neurons. In this study, we assess
if similar processes occur at the fly NMJ. We first verify tools to manipulate subsets of glia and
then use these tools to kill all glia. We then observe how they respond to neurodegeneration
and find that they are less reactive in synapses that have undergone high levels of
degeneration. Interestingly, ablating glia in the same condition shows a preservation of overall
boutons (a majority of which being Ib). This implies that glia may have a phagocytic role early in
development and preferentially for Ib boutons, but then switches to a maintenance role later in
the larval life and for Is boutons specifically. Finally, we observe that glia are highly dynamic
when the neuron is growing exuberantly. We take this to mean that the glia are maintaining the
extensive growths. These findings provide exciting insight into peripheral glia “reactivity” due to
neuronal perturbations.
Finally, we attempt to generate a model for TDP43 mutant-mediated loss of Unc13A and
how this leads to neurodegeneration. It has recently been discovered that TDP43 and other
hnRNPs prevent the inclusion of cryptic exons into Unc13A (Koike et al., 2023; Ma et al., 2022).
When cryptic exons are preserved, they lead to protein reductions due to nonsense-mediated
decay. In the case of Unc13A, these inclusions promote the progression of ALS. We hope to
study these interactions in the fly. First, we utilize a reporter construct that labels all Unc13
protein (both Unc13A and Unc13B isoforms). We find that this fly line localizes in the
presynaptic terminal as we expect and that it scales well based on our heterozygous and mutant
controls. Next, mutations in Unc13A yield an approximately 50% reduction in our reporter.
However, despite the genetic loss of this gene we do not observe neurodegeneration.
Subsequent experiments with hTDP42 or TBPH coupled with Unc13Ayields variations in

122
overall Unc13 levels, but no increases in neurodegeneration. Lastly, we find that
neurodegeneration due to stai mutations leads to a drastic decrease of Unc13 protein. This is
exciting as this could be a reason why cryptic exon inclusions in the mammalian version of
Stathmin, STMN2, leads to ALS and FTLD-TDP (Klim et al., 2019; Melamed et al., 2019). While
we do not have a model matching the mammalian data, we may be able to study Stai and ALS
in Drosophila.
My dissertation work here has increased our understanding of synaptic plasticity,
including in glia, when neurons experience neurodegeneration via cytoskeletal dysregulation or
disease, and during injury-related signaling. We develop a tool that extends our animals’
lifespan for long-term degeneration studies. Next, we find that receptors are significantly
upscaled when faced with reduced neurotransmission and injury signaling from the neuronal
partner. Third, we find that Stai possibly functions downstream of Hiw and that active injury
signaling causes further remodeling than that induced by stai. Then, we learn that glia can
become reactive during neurodegeneration and neuronal injury. We also find that they likely
play differential roles depending on development and neuron type. Finally, we confirm that loss
of TDP43 does not alter Unc13 in a way that promotes neurodegeneration. A more important
finding is that stai leads to loss of Unc13, which may be important for ALS disease progression.
Altogether, this work provides a foundation for future projects looking at synaptic remodeling
during neurodegeneration, injury, and disease.

123
References
Aberle, H., Haghighi, A. P., Fetter, R. D., McCabe, B. D., Magalhães, T. R., & Goodman, C. S.
(2002). wishful thinking encodes a BMP type II receptor that regulates synaptic growth in
Drosophila. Neuron, 33(4), 545-558. doi:10.1016/s0896-6273(02)00589-5
Akbergenova, Y., Cunningham, K. L., Zhang, Y. V., Weiss, S., & Littleton, J. T. (2018).
Characterization of developmental and molecular factors underlying release
heterogeneity at Drosophila synapses. Elife, 7. doi:10.7554/eLife.38268
Alami, N. H., Smith, R. B., Carrasco, M. A., Williams, L. A., Winborn, C. S., Han, S. S. W., . . .
Taylor, J. P. (2014). Axonal transport of TDP-43 mRNA granules is impaired by ALScausing mutations. Neuron, 81(3), 536-543. doi:10.1016/j.neuron.2013.12.018
Aldskogius, H., Liu, L., & Svensson, M. (1999). Glial responses to synaptic damage and
plasticity. J Neurosci Res, 58(1), 33-41.
Allen, N. J., & Barres, B. A. (2005). Signaling between glia and neurons: focus on synaptic
plasticity. Curr Opin Neurobiol, 15(5), 542-548. doi:10.1016/j.conb.2005.08.006
Allen, N. J., & Barres, B. A. (2009). Glia — more than just brain glue. Nature, 457(7230), 675-
677. doi:10.1038/457675a
Aponte-Santiago, N. A., & Littleton, J. T. (2020). Synaptic Properties and Plasticity Mechanisms
of Invertebrate Tonic and Phasic Neurons. Front Physiol, 11, 611982.
doi:10.3389/fphys.2020.611982
Asghari Adib, E., Smithson, L. J., & Collins, C. A. (2018). An axonal stress response pathway:
degenerative and regenerative signaling by DLK. Curr Opin Neurobiol, 53, 110-119.
doi:10.1016/j.conb.2018.07.002
Atwood, H., Govind, C., & Wu, C. F. (1993). Differential ultrastructure of synaptic terminals on
ventral longitudinal abdominal muscles in Drosophila larvae. Journal of neurobiology,
24(8), 1008-1024.
Barik, A., Li, L., Sathyamurthy, A., Xiong, W. C., & Mei, L. (2016). Schwann Cells in
Neuromuscular Junction Formation and Maintenance. J Neurosci, 36(38), 9770-9781.
doi:10.1523/JNEUROSCI.0174-16.2016

124
Bier, E., Harrison, M. M., O’Connor-Giles, K. M., & Wildonger, J. (2018). Advances in
engineering the fly genome with the CRISPR-Cas system. Genetics, 208(1), 1-18.
Bishop, D. L., Misgeld, T., Walsh, M. K., Gan, W. B., & Lichtman, J. W. (2004). Axon branch
removal at developing synapses by axosome shedding. Neuron, 44(4), 651-661.
doi:10.1016/j.neuron.2004.10.026
Bohlen, C. J., Friedman, B. A., Dejanovic, B., & Sheng, M. (2019). Microglia in Brain
Development, Homeostasis, and Neurodegeneration. Annu Rev Genet, 53, 263-288.
doi:10.1146/annurev-genet-112618-043515
Böhme, M. A., Beis, C., Reddy-Alla, S., Reynolds, E., Mampell, M. M., Grasskamp, A. T., . . .
Babikir, H. (2016). Active zone scaffolds differentially accumulate Unc13 isoforms to
tune Ca2+ channel–vesicle coupling. Nature neuroscience, 19(10), 1311-1320.
Brand, A. H., & Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates
and generating dominant phenotypes. Development, 118(2), 401-415.
doi:10.1242/dev.118.2.401
Brummel, T., Abdollah, S., Haerry, T. E., Shimell, M. J., Merriam, J., Raftery, L., . . . O’Connor,
M. B. (1999). The Drosophila activin receptor baboon signals through dSmad2 and
controls cell proliferation but not patterning during larval development. Genes &
development, 13(1), 98-111.
Chang, J.-C., & Morton, D. B. (2017). Drosophila lines with mutant and wild type human TDP-43
replacing the endogenous gene reveals phosphorylation and ubiquitination in mutant
lines in the absence of viability or lifespan defects. PLoS One, 12(7), e0180828.
doi:10.1371/journal.pone.0180828
Chen, C. K., Bregere, C., Paluch, J., Lu, J. F., Dickman, D. K., & Chang, K. T. (2014). Activitydependent facilitation of Synaptojanin and synaptic vesicle recycling by the Minibrain
kinase. Nat Commun, 5, 4246. doi:10.1038/ncomms5246
Chen, X., Ma, W., Zhang, S., Paluch, J., Guo, W., & Dickman, D. K. (2017a). The BLOC-1
Subunit Pallidin Facilitates Activity-Dependent Synaptic Vesicle Recycling. eNeuro, 4(1).
doi:10.1523/ENEURO.0335-16.2017
Chen, X., Ma, W., Zhang, S., Paluch, J., Guo, W., & Dickman, D. K. (2017b). The BLOC-1
Subunit Pallidin Facilitates Activity-Dependent Synaptic Vesicle Recycling. eNeuro, 4(1),
1-18.

125
Collins, C. A., Wairkar, Y. P., Johnson, S. L., & DiAntonio, A. (2006). Highwire restrains synaptic
growth by attenuating a MAP kinase signal. Neuron, 51(1), 57-69.
doi:10.1016/j.neuron.2006.05.026
Daniels, R. W., Collins, C. A., Chen, K., Gelfand, M. V., Featherstone, D. E., & DiAntonio, A.
(2006). A single vesicular glutamate transporter is sufficient to fill a synaptic vesicle.
Neuron, 49(1), 11-16. doi:10.1016/j.neuron.2005.11.032
Daniels, R. W., Collins, C. A., Gelfand, M. V., Dant, J., Brooks, E. S., Krantz, D. E., & DiAntonio,
A. (2004). Increased expression of the Drosophila vesicular glutamate transporter leads
to excess glutamate release and a compensatory decrease in quantal content. J
Neurosci, 24(46), 10466-10474. doi:10.1523/JNEUROSCI.3001-04.2004
Danielsen, E. T., Moeller, M. E., Yamanaka, N., Ou, Q., Laursen, J. M., Soenderholm, C., . . .
Rewitz, K. F. (2016). A Drosophila Genome-Wide Screen Identifies Regulators of Steroid
Hormone Production and Developmental Timing. Dev Cell, 37(6), 558-570.
doi:10.1016/j.devcel.2016.05.015
Dannhäuser, S., Mrestani, A., Gundelach, F., Pauli, M., Komma, F., Kollmannsberger, P., . . .
Paul, M. M. (2022). Endogenous tagging of Unc-13 reveals nanoscale reorganization at
active zones during presynaptic homeostatic potentiation. Frontiers in cellular
neuroscience, 16. doi:10.3389/fncel.2022.1074304
Davis, G. W., & Goodman, C. S. (1998). Genetic analysis of synaptic development and
plasticity: homeostatic regulation of synaptic efficacy. Current opinion in neurobiology,
8(1), 149-156.
DeJesus-Hernandez, M., Mackenzie, I. R., Boeve, B. F., Boxer, A. L., Baker, M., Rutherford, N.
J., . . . Rademakers, R. (2011). Expanded GGGGCC hexanucleotide repeat in
noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron,
72(2), 245-256. doi:10.1016/j.neuron.2011.09.011
DiAntonio, A., Petersen, S. A., Heckmann, M., & Goodman, C. S. (1999). Glutamate receptor
expression regulates quantal size and quantal content at the Drosophila neuromuscular
junction. Journal of Neuroscience, 19(8), 3023-3032.
Dickman, D. K., & Davis, G. W. (2009). The schizophrenia susceptibility gene dysbindin controls
synaptic homeostasis. Science, 326(5956), 1127-1130. doi:10.1126/science.1179685

126
Dickman, D. K., Horne, J. A., Meinertzhagen, I. A., & Schwarz, T. L. (2005a). A Slowed
Classical Pathway Rather Than Kiss-and-Run Mediates Endocytosis at Synapses
Lacking Synaptojanin and Endophilin. Cell, 123(3), 521-533. Retrieved from
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation
&list_uids=16269341
Dickman, D. K., Horne, J. A., Meinertzhagen, I. A., & Schwarz, T. L. (2005b). A slowed classical
pathway rather than kiss-and-run mediates endocytosis at synapses lacking
synaptojanin and endophilin. Cell, 123(3), 521-533. doi:10.1016/j.cell.2005.09.026
Dittman, J. S., & Ryan, T. A. (2019). The control of release probability at nerve terminals. Nature
Reviews Neuroscience, 20(3), 177-186.
Duncan, J. E., Lytle, N. K., Zuniga, A., & Goldstein, L. S. (2013). The Microtubule Regulatory
Protein Stathmin Is Required to Maintain the Integrity of Axonal Microtubules in
Drosophila. PLoS One, 8(6), e68324. doi:10.1371/journal.pone.0068324
Eaton, B. A., Fetter, R. D., & Davis, G. W. (2002). Dynactin is necessary for synapse
stabilization. Neuron, 34(5), 729-741.
Fallini, C., Bassell, G. J., & Rossoll, W. (2012). The ALS disease protein TDP-43 is actively
transported in motor neuron axons and regulates axon outgrowth. Hum Mol Genet,
21(16), 3703-3718. doi:10.1093/hmg/dds205
Feiguin, F., Godena, V. K., Romano, G., D'Ambrogio, A., Klima, R., & Baralle, F. E. (2009).
Depletion of TDP-43 affects Drosophila motoneurons terminal synapsis and locomotive
behavior. FEBS Lett, 583(10), 1586-1592. doi:10.1016/j.febslet.2009.04.019
Feng, Z., & Ko, C. P. (2008). The role of glial cells in the formation and maintenance of the
neuromuscular junction. Ann N Y Acad Sci, 1132, 19-28. doi:10.1196/annals.1405.016
Frank, C. A., James, T. D., & Müller, M. (2020). Homeostatic control of Drosophila
neuromuscular junction function. Synapse, 74(1), e22133.
Frank, C. A., Kennedy, M. J., Goold, C. P., Marek, K. W., & Davis, G. W. (2006). Mechanisms
underlying the rapid induction and sustained expression of synaptic homeostasis.
Neuron, 52(4), 663-677. doi:10.1016/j.neuron.2006.09.029
Freeman, M. R., & Doherty, J. (2006). Glial cell biology in Drosophila and vertebrates. Trends
Neurosci, 29(2), 82-90. doi:10.1016/j.tins.2005.12.002

127
Freibaum, B. D., Chitta, R. K., High, A. A., & Taylor, J. P. (2010). Global Analysis of TDP-43
Interacting Proteins Reveals Strong Association with RNA Splicing and Translation
Machinery. Journal of Proteome Research, 9(2), 1104-1120. doi:10.1021/pr901076y
Fuentes-Medel, Y., Ashley, J., Barria, R., Maloney, R., Freeman, M., & Budnik, V. (2012).
Integration of a retrograde signal during synapse formation by glia-secreted TGF-beta
ligand. Curr Biol, 22(19), 1831-1838. doi:10.1016/j.cub.2012.07.063
Fuentes-Medel, Y., Logan, M. A., Ashley, J., Ataman, B., Budnik, V., & Freeman, M. R. (2009).
Glia and muscle sculpt neuromuscular arbors by engulfing destabilized synaptic boutons
and shed presynaptic debris. PLoS Biol, 7(8), e1000184.
doi:10.1371/journal.pbio.1000184
Gavino, M. A., Ford, K. J., Archila, S., & Davis, G. W. (2015). Homeostatic synaptic depression
is achieved through a regulated decrease in presynaptic calcium channel abundance.
Elife, 4. doi:10.7554/eLife.05473
Gibbens, Y. Y., Warren, J. T., Gilbert, L. I., & O'Connor, M. B. (2011). Neuroendocrine
regulation of Drosophila metamorphosis requires TGFbeta/Activin signaling.
Development, 138(13), 2693-2703. doi:10.1242/dev.063412
Goel, P., Bergeron, D. D., Bohme, M., Nunnelly, L., Lehmann, M., Buser, C., . . . Dickman, D. K.
(2019). Homeostatic scaling of active zone scaffolds maintains global synaptic strength.
J Cell Biol, In Press.
Goel, P., & Dickman, D. (2018). Distinct homeostatic modulations stabilize reduced postsynaptic
receptivity in response to presynaptic DLK signaling. Nat Commun, 9(1), 1856.
doi:10.1038/s41467-018-04270-0
Goel, P., Khan, M., Howard, S., Kim, G., Kiragasi, B., Kikuma, K., & Dickman, D. (2019). A
Screen for Synaptic Growth Mutants Reveals Mechanisms That Stabilize Synaptic
Strength. J Neurosci, 39(21), 4051-4065. doi:10.1523/JNEUROSCI.2601-18.2019
Goel, P., Li, X., & Dickman, D. (2017). Disparate Postsynaptic Induction Mechanisms Ultimately
Converge to Drive the Retrograde Enhancement of Presynaptic Efficacy. Cell Rep,
21(9), 2339-2347. doi:10.1016/j.celrep.2017.10.116
Goel, P., Li, X., & Dickman, D. (2019). Estimation of the Readily Releasable Synaptic Vesicle
Pool at the Drosophila Larval Neuromuscular Junction. Bio Protoc, 9(1).
doi:10.21769/BioProtoc.3127

128
Goodwani, S., Fernandez, C., Acton, P. J., Buggia-Prevot, V., McReynolds, M. L., Ma, J., . . .
Ray, W. J. (2020). Dual Leucine Zipper Kinase Is Constitutively Active in the Adult
Mouse Brain and Has Both Stress-Induced and Homeostatic Functions. Int J Mol Sci,
21(14). doi:10.3390/ijms21144849
Graf, E. R., Daniels, R. W., Burgess, R. W., Schwarz, T. L., & DiAntonio, A. (2009). Rab3
dynamically controls protein composition at active zones. Neuron, 64(5), 663-677.
Graf, E. R., Heerssen, H. M., Wright, C. M., Davis, G. W., & DiAntonio, A. (2011). Stathmin is
required for stability of the Drosophila neuromuscular junction. J Neurosci, 31(42),
15026-15034. doi:10.1523/JNEUROSCI.2024-11.2011
Gratz, S. J., Goel, P., Bruckner, J. J., Hernandez, R. X., Khateeb, K., Macleod, G. T., . . .
O'Connor-Giles, K. M. (2019). Endogenous tagging reveals differential regulation of
Ca2+ channels at single active zones during presynaptic homeostatic potentiation and
depression. Journal of Neuroscience, 39(13), 2416-2429.
Grill, B., Murphey, R. K., & Borgen, M. A. (2016). The PHR proteins: intracellular signaling hubs
in neuronal development and axon degeneration. Neural Development, 11(1), 8.
doi:10.1186/s13064-016-0063-0
Hall, Z. W., & Sanes, J. R. (1993). Synaptic structure and development: the neuromuscular
junction. Cell, 72 Suppl, 99-121. doi:10.1016/s0092-8674(05)80031-5
Han, T. H., Dharkar, P., Mayer, M. L., & Serpe, M. (2015). Functional reconstitution of
Drosophila melanogaster NMJ glutamate receptors. Proc Natl Acad Sci U S A, 112(19),
6182-6187. doi:10.1073/pnas.1500458112
Han, Y., Goel, P., Chen, J., Perry, S., Tran, N., Nishimura, S., . . . Dickman, D. (2023). Excess
glutamate release triggers subunit-specific homeostatic receptor scaling. Cell Rep,
42(7), 112775. doi:10.1016/j.celrep.2023.112775
Hao, Y., & Collins, C. (2017). Intrinsic mechanisms for axon regeneration: insights from injured
axons in Drosophila. Curr Opin Genet Dev, 44, 84-91. doi:10.1016/j.gde.2017.01.009
Hao, Y., Waller, T. J., Nye, D. M., Li, J., Zhang, Y., Hume, R. I., . . . Collins, C. A. (2019).
Degeneration of Injured Axons and Dendrites Requires Restraint of a Protective JNK
Signaling Pathway by the Transmembrane Protein Raw. J Neurosci, 39(43), 8457-8470.
doi:10.1523/JNEUROSCI.0016-19.2019

129
He, K., Han, Y., Li, X., Hernandez, R. X., Riboul, D. V., Feghhi, T., . . . Dickman, D. (2023).
Physiologic and Nanoscale Distinctions Define Glutamatergic Synapses in Tonic vs
Phasic Neurons. The Journal of Neuroscience, 43(25), 4598-4611.
doi:10.1523/jneurosci.0046-23.2023
Herrera, A. A., Qiang, H., & Ko, C. P. (2000). The role of perisynaptic Schwann cells in
development of neuromuscular junctions in the frog (Xenopus laevis). J Neurobiol, 45(4),
237-254. doi:10.1002/1097-4695(200012)45:4<237::aid-neu5>3.0.co;2-j
Hevia, C. F., López-Varea, A., Esteban, N., & de Celis, J. F. (2017). A search for genes
mediating the growth-promoting function of TGFβ in the Drosophila melanogaster wing
disc. Genetics, 206(1), 231-249.
Jahn, R., & Fasshauer, D. (2012). Molecular machines governing exocytosis of synaptic
vesicles. Nature, 490(7419), 201-207. doi:10.1038/nature11320
Jin, Y., & Zheng, B. (2019). Multitasking: Dual Leucine Zipper-Bearing Kinases in Neuronal
Development and Stress Management. Annu Rev Cell Dev Biol, 35, 501-521.
doi:10.1146/annurev-cellbio-100617-062644
Joseph, B., Kondo, S., & Lai, E. C. (2018). Short cryptic exons mediate recursive splicing in
Drosophila. Nature Structural & Molecular Biology, 25(5), 365-371. doi:10.1038/s41594-
018-0052-6
Joseph, B., Scala, C., Kondo, S., & Lai, E. C. (2022). Molecular and genetic dissection of
recursive splicing. Life Sci Alliance, 5(1). doi:10.26508/lsa.202101063
Kang, H., Tian, L., Mikesh, M., Lichtman, J. W., & Thompson, W. J. (2014). Terminal Schwann
cells participate in neuromuscular synapse remodeling during reinnervation following
nerve injury. J Neurosci, 34(18), 6323-6333. doi:10.1523/JNEUROSCI.4673-13.2014
Kang, H., Tian, L., & Thompson, W. (2003). Terminal Schwann cells guide the reinnervation of
muscle after nerve injury. J Neurocytol, 32(5-8), 975-985.
doi:10.1023/B:NEUR.0000020636.27222.2d
Kang, H., Tian, L., & Thompson, W. J. (2019). Schwann cell guidance of nerve growth between
synaptic sites explains changes in the pattern of muscle innervation and remodeling of
synaptic sites following peripheral nerve injuries. J Comp Neurol, 527(8), 1388-1400.
doi:10.1002/cne.24625

130
Kikuma, K., Li, X., Kim, D., Sutter, D., & Dickman, D. K. (2017a). Extended Synaptotagmin
Localizes to Presynaptic ER and Promotes Neurotransmission and Synaptic Growth in
Drosophila. Genetics, 207(3), 993-1006. doi:10.1534/genetics.117.300261
Kikuma, K., Li, X., Kim, D., Sutter, D., & Dickman, D. K. (2017b). Extended Synaptotagmin
Localizes to Presynaptic ER and Promotes Neurotransmission and Synaptic Growth in
Drosophila. Genetics.
Kiragasi, B., Wondolowski, J., Li, Y., & Dickman, D. K. (2017). A Presynaptic Glutamate
Receptor Subunit Confers Robustness to Neurotransmission and Homeostatic
Potentiation. Cell Rep., 19(13), 2694-2706.
Kittel, R. J., Wichmann, C., Rasse, T. M., Fouquet, W., Schmidt, M., Schmid, A., . . . Willig, K. I.
(2006). Bruchpilot promotes active zone assembly, Ca2+ channel clustering, and vesicle
release. Science, 312(5776), 1051-1054.
Klim, J. R., Williams, L. A., Limone, F., Guerra San Juan, I., Davis-Dusenbery, B. N., Mordes, D.
A., . . . Eggan, K. (2019). ALS-implicated protein TDP-43 sustains levels of STMN2, a
mediator of motor neuron growth and repair. Nat Neurosci, 22(2), 167-179.
doi:10.1038/s41593-018-0300-4
Ko, C. P., & Robitaille, R. (2015). Perisynaptic Schwann Cells at the Neuromuscular Synapse:
Adaptable, Multitasking Glial Cells. Cold Spring Harb Perspect Biol, 7(10), a020503.
doi:10.1101/cshperspect.a020503
Koike, Y., Pickles, S., Estades Ayuso, V., Jansen-West, K., Qi, Y. A., Li, Z., . . . Prudencio, M.
(2023). TDP-43 and other hnRNPs regulate cryptic exon inclusion of a key ALS/FTD risk
gene, UNC13A. PLoS Biol, 21(3), e3002028. doi:10.1371/journal.pbio.3002028
Lachkar, S., Lebois, M., Steinmetz, M. O., Guichet, A., Lal, N., Curmi, P. A., . . . Ozon, S.
(2010). Drosophila stathmins bind tubulin heterodimers with high and variable
stoichiometries. J Biol Chem, 285(15), 11667-11680. doi:10.1074/jbc.M109.096727
Le Pichon, C. E., Meilandt, W. J., Dominguez, S., Solanoy, H., Lin, H., Ngu, H., . . . Lewcock, J.
W. (2017). Loss of dual leucine zipper kinase signaling is protective in animal models of
neurodegenerative disease. Sci Transl Med, 9(403). doi:10.1126/scitranslmed.aag0394
Leiserson, W. M., Harkins, E. W., & Keshishian, H. (2000). Fray, a Drosophila Serine/Threonine
Kinase Homologous to Mammalian PASK, Is Required for Axonal Ensheathment.
Neuron, 28(3), 793-806. doi:https://doi.org/10.1016/S0896-6273(00)00154-9

131
Lewcock, J. W., Genoud, N., Lettieri, K., & Pfaff, S. L. (2007). The ubiquitin ligase Phr1
regulates axon outgrowth through modulation of microtubule dynamics. Neuron, 56(4),
604-620. doi:10.1016/j.neuron.2007.09.009
Li, J., Park, E., Zhong, L. R., & Chen, L. (2019). Homeostatic synaptic plasticity as a
metaplasticity mechanism—a molecular and cellular perspective. Current opinion in
neurobiology, 54, 44-53.
Li, J., Zhang, Y. V., Asghari Adib, E., Stanchev, D. T., Xiong, X., Klinedinst, S., . . . Collins, C. A.
(2017). Restraint of presynaptic protein levels by Wnd/DLK signaling mediates synaptic
defects associated with the kinesin-3 motor Unc-104. Elife, 6. doi:10.7554/eLife.24271
Li, L., Xiong, W.-C., & Mei, L. (2018). Neuromuscular junction formation, aging, and disorders.
Annual review of physiology, 80, 159-188.
Li, L., Xiong, W. C., & Mei, L. (2018). Neuromuscular Junction Formation, Aging, and Disorders.
Annu Rev Physiol, 80, 159-188. doi:10.1146/annurev-physiol-022516-034255
Li, X., Chien, C., Han, Y., Sun, Z., Chen, X., & Dickman, D. (2021). Autocrine inhibition by a
glutamate-gated chloride channel mediates presynaptic homeostatic depression. Sci
Adv, 7(49), eabj1215. doi:10.1126/sciadv.abj1215
Li, X., Goel, P., Chen, C., Angajala, V., Chen, X., & Dickman, D. K. (2018). Synapse-specific
and compartmentalized expression of presynaptic homeostatic potentiation. Elife, 7.
doi:10.7554/eLife.34338
Li, X., Goel, P., Wondolowski, J., Paluch, J., & Dickman, D. (2018). A Glutamate Homeostat
Controls the Presynaptic Inhibition of Neurotransmitter Release. Cell Rep, 23(6), 1716-
1727. doi:10.1016/j.celrep.2018.03.130
Liddelow, S. A., & Barres, B. A. (2017). Reactive Astrocytes: Production, Function, and
Therapeutic Potential. Immunity, 46(6), 957-967. doi:10.1016/j.immuni.2017.06.006
Liddelow, S. A., Guttenplan, K. A., Clarke, L. E., Bennett, F. C., Bohlen, C. J., Schirmer, L., . . .
Barres, B. A. (2017). Neurotoxic reactive astrocytes are induced by activated microglia.
Nature, 541(7638), 481-487. doi:10.1038/nature21029
Ling, J. P., Pletnikova, O., Troncoso, J. C., & Wong, P. C. (2015). TDP-43 repression of
nonconserved cryptic exons is compromised in ALS-FTD. Science, 349(6248), 650-655.
doi:10.1126/science.aab0983

132
Liu, J.-J. (2017). Regulation of dynein-dynactin-driven vesicular transport. Traffic, 18(6), 336-
347. doi:https://doi.org/10.1111/tra.12475
Lloyd, Thomas E., Machamer, J., O'Hara, K., Kim, Ji H., Collins, Sarah E., Wong, Man Y., . . .
Kolodkin, Alex L. (2012). The p150Glued CAP-Gly Domain Regulates
Initiation of Retrograde Transport at Synaptic Termini. Neuron, 74(2), 344-360.
doi:10.1016/j.neuron.2012.02.026
Ma, X. R., Prudencio, M., Koike, Y., Vatsavayai, S. C., Kim, G., Harbinski, F., . . . Gitler, A. D.
(2022). TDP-43 represses cryptic exon inclusion in the FTD-ALS gene UNC13A. Nature,
603(7899), 124-130. doi:10.1038/s41586-022-04424-7
Maas, C., Torres, V. I., Altrock, W. D., Leal-Ortiz, S., Wagh, D., Terry-Lorenzo, R. T., . . .
Garner, C. C. (2012). Formation of Golgi-derived active zone precursor vesicles. J
Neurosci, 32(32), 11095-11108. doi:10.1523/jneurosci.0195-12.2012
MacDonald, J. M., Beach, M. G., Porpiglia, E., Sheehan, A. E., Watts, R. J., & Freeman, M. R.
(2006). The Drosophila cell corpse engulfment receptor Draper mediates glial clearance
of severed axons. Neuron, 50(6), 869-881. doi:10.1016/j.neuron.2006.04.028
Mackler, J. M., Drummond, J. A., Loewen, C. A., Robinson, I. M., & Reist, N. E. (2002). The
C(2)B Ca(2+)-binding motif of synaptotagmin is required for synaptic transmission in
vivo. Nature, 418(6895), 340-344. doi:10.1038/nature00846
Massaro, C. M., Pielage, J., & Davis, G. W. (2009). Molecular mechanisms that enhance
synapse stability despite persistent disruption of the spectrin/ankyrin/microtubule
cytoskeleton. J Cell Biol, 187(1), 101-117. doi:10.1083/jcb.200903166
McCabe, B. D., Hom, S., Aberle, H., Fetter, R. D., Marques, G., Haerry, T. E., . . . Haghighi, A.
P. (2004). Highwire regulates presynaptic BMP signaling essential for synaptic growth.
Neuron, 41(6), 891-905. doi:10.1016/s0896-6273(04)00073-x
McGurk, L., Berson, A., & Bonini, N. M. (2015). Drosophila as an In Vivo Model for Human
Neurodegenerative Disease. Genetics, 201(2), 377-402.
doi:10.1534/genetics.115.179457
McLaughlin, C. N., Perry-Richardson, J. J., Coutinho-Budd, J. C., & Broihier, H. T. (2019). Dying
Neurons Utilize Innate Immune Signaling to Prime Glia for Phagocytosis during
Development. Dev Cell, 48(4), 506-522 e506. doi:10.1016/j.devcel.2018.12.019

133
Mehta, P. R., Brown, A. L., Ward, M. E., & Fratta, P. (2023). The era of cryptic exons:
implications for ALS-FTD. Mol Neurodegener, 18(1), 16. doi:10.1186/s13024-023-
00608-5
Melamed, Z., Lopez-Erauskin, J., Baughn, M. W., Zhang, O., Drenner, K., Sun, Y., . . .
Cleveland, D. W. (2019). Premature polyadenylation-mediated loss of stathmin-2 is a
hallmark of TDP-43-dependent neurodegeneration. Nat Neurosci, 22(2), 180-190.
doi:10.1038/s41593-018-0293-z
Menon, K. P., Carrillo, R. A., & Zinn, K. (2013). Development and plasticity of the Drosophila
larval neuromuscular junction. Wiley Interdiscip Rev Dev Biol, 2(5), 647-670.
doi:10.1002/wdev.108
Miller, D. L., Ballard, S. L., & Ganetzky, B. (2012). Analysis of synaptic growth and function in
Drosophila with an extended larval stage. J Neurosci, 32(40), 13776-13786.
doi:10.1523/JNEUROSCI.0508-12.2012
Mizielinska, S., Grönke, S., Niccoli, T., Ridler, C. E., Clayton, E. L., Devoy, A., . . . Isaacs, A. M.
(2014). C9orf72 repeat expansions cause neurodegeneration in
Drosophila through arginine-rich proteins. Science, 345(6201), 1192-1194.
doi:doi:10.1126/science.1256800
Narayanan, R. K., Mangelsdorf, M., Panwar, A., Butler, T. J., Noakes, P. G., & Wallace, R. H.
(2013). Identification of RNA bound to the TDP-43 ribonucleoprotein complex in the adult
mouse brain. Amyotroph Lateral Scler Frontotemporal Degener, 14(4), 252-260.
doi:10.3109/21678421.2012.734520
Nelson, Sacha B., & Valakh, V. (2015). Excitatory/Inhibitory Balance and Circuit Homeostasis in
Autism Spectrum Disorders. Neuron, 87(4), 684-698. doi:10.1016/j.neuron.2015.07.033
Neumann, M., Sampathu, D. M., Kwong, L. K., Truax, A. C., Micsenyi, M. C., Chou, T. T., . . .
Lee, V. M. (2006). Ubiquitinated TDP-43 in frontotemporal lobar degeneration and
amyotrophic lateral sclerosis. Science, 314(5796), 130-133.
doi:10.1126/science.1134108
Newman, Z. L., Hoagland, A., Aghi, K., Worden, K., Levy, S. L., Son, J. H., . . . Isacoff, E. Y.
(2017). Input-Specific Plasticity and Homeostasis at the Drosophila Larval
Neuromuscular Junction. Neuron, 93(6), 1388-1404 e1310.
doi:10.1016/j.neuron.2017.02.028

134
Ohhara, Y., Kobayashi, S., & Yamanaka, N. (2017). Nutrient-Dependent Endocycling in
Steroidogenic Tissue Dictates Timing of Metamorphosis in Drosophila melanogaster.
PLoS Genet, 13(1), e1006583. doi:10.1371/journal.pgen.1006583
Okada, Y., Yamazaki, H., Sekine-Aizawa, Y., & Hirokawa, N. (1995). The neuron-specific
kinesin superfamily protein KIF1A is a unique monomeric motor for anterograde axonal
transport of synaptic vesicle precursors. Cell, 81(5), 769-780. doi:10.1016/0092-
8674(95)90538-3
Ono, H., Rewitz, K. F., Shinoda, T., Itoyama, K., Petryk, A., Rybczynski, R., . . . O'Connor, M. B.
(2006). Spook and Spookier code for stage-specific components of the ecdysone
biosynthetic pathway in Diptera. Dev Biol, 298(2), 555-570.
doi:10.1016/j.ydbio.2006.07.023
Ozdowski, E. F., Wentzell, J. S., Engert, S. M., Abbott, H., & Sherwood, N. T. (2020).
Suppression of spastin Mutant Phenotypes by Pak3 Loss Implicates a Role for Reactive
Glia in AD-HSP. Front Neurosci, 14, 912. doi:10.3389/fnins.2020.00912
Patel, S., Meilandt, W. J., Erickson, R. I., Chen, J., Deshmukh, G., Estrada, A. A., . . . Siu, M.
(2017). Selective Inhibitors of Dual Leucine Zipper Kinase (DLK, MAP3K12) with Activity
in a Model of Alzheimer’s Disease. Journal of Medicinal Chemistry, 60(19), 8083-8102.
doi:10.1021/acs.jmedchem.7b00843
Perry, S., Goel, P., Tran, N. L., Pinales, C., Buser, C., Miller, D. L., . . . Dickman, D. (2020).
Developmental arrest of Drosophila larvae elicits presynaptic depression and enables
prolonged studies of neurodegeneration. Development, 147(10).
doi:10.1242/dev.186312
Perry, S., Han, Y., Das, A., & Dickman, D. (2017). Homeostatic plasticity can be induced and
expressed to restore synaptic strength at neuromuscular junctions undergoing ALSrelated degeneration. Hum Mol Genet, 26(21), 4153-4167. doi:10.1093/hmg/ddx304
Perry, S., Han, Y., Qiu, C., Chien, C., Goel, P., Nishimura, S., . . . Dickman, D. (2022). A
glutamate receptor C-tail recruits CaMKII to suppress retrograde homeostatic signaling.
Nat Commun, 13(1), 7656. doi:10.1038/s41467-022-35417-9
Petersen, S. A., Fetter, R. D., Noordermeer, J. N., Goodman, C. S., & DiAntonio, A. (1997).
Genetic Analysis of Glutamate Receptors in Drosophila Reveals a Retrograde Signal
Regulating Presynaptic Transmitter Release. Neuron, 19(6), 1237-1248.
doi:10.1016/S0896-6273(00)80415-8

135
Pielage, J., Fetter, R. D., & Davis, G. W. (2005a). Presynaptic spectrin is essential for synapse
stabilization. Curr Biol, 15(10), 918-928. doi:10.1016/j.cub.2005.04.030
Pielage, J., Fetter, R. D., & Davis, G. W. (2005b). Presynaptic Spectrin Is Essential for Synapse
Stabilization. Curr. Biol., 15, 918-928.
Pozo, K., & Goda, Y. (2010). Unraveling Mechanisms of Homeostatic Synaptic Plasticity.
Neuron, 66(3), 337-351. doi:10.1016/j.neuron.2010.04.028
Qin, G., Schwarz, T., Kittel, R. J., Schmid, A., Rasse, T. M., Kappei, D., . . . Sigrist, S. J. (2005).
Four different subunits are essential for expressing the synaptic glutamate receptor at
neuromuscular junctions of Drosophila. J Neurosci, 25(12), 3209-3218.
doi:10.1523/JNEUROSCI.4194-04.2005
Reddy-Alla, S., Böhme, M. A., Reynolds, E., Beis, C., Grasskamp, A. T., Mampell, M. M., . . .
Walter, A. M. (2017). Stable Positioning of Unc13 Restricts Synaptic Vesicle Fusion to
Defined Release Sites to Promote Synchronous Neurotransmission. Neuron, 95(6),
1350-1364.e1312. doi:10.1016/j.neuron.2017.08.016
Reddy, L. V., Koirala, S., Sugiura, Y., Herrera, A. A., & Ko, C. P. (2003). Glial cells maintain
synaptic structure and function and promote development of the neuromuscular junction
in vivo. Neuron, 40(3), 563-580. doi:10.1016/s0896-6273(03)00682-2
Renton, A. E., Majounie, E., Waite, A., Simon-Sanchez, J., Rollinson, S., Gibbs, J. R., . . .
Traynor, B. J. (2011). A hexanucleotide repeat expansion in C9ORF72 is the cause of
chromosome 9p21-linked ALS-FTD. Neuron, 72(2), 257-268.
doi:10.1016/j.neuron.2011.09.010
Rewitz, K. F., Yamanaka, N., Gilbert, L. I., & O'Connor, M. B. (2009). The insect neuropeptide
PTTH activates receptor tyrosine kinase torso to initiate metamorphosis. Science,
326(5958), 1403-1405. doi:10.1126/science.1176450
Rewitz, K. F., Yamanaka, N., Gilbert, L. I., & O’Connor, M. B. (2009). The Insect Neuropeptide
PTTH Activates Receptor Tyrosine Kinase Torso to Initiate Metamorphosis. Science,
326(5958), 1403-1405. doi:doi:10.1126/science.1176450
Reynolds, M. L., & Woolf, C. J. (1992). Terminal Schwann cells elaborate extensive processes
following denervation of the motor endplate. J Neurocytol, 21(1), 50-66.
doi:10.1007/bf01206897

136
Rotem, N., Magen, I., Ionescu, A., Gershoni-Emek, N., Altman, T., Costa, C. J., . . . Perlson, E.
(2017). ALS Along the Axons - Expression of Coding and Noncoding RNA Differs in
Axons of ALS models. Sci Rep, 7, 44500. doi:10.1038/srep44500
Russo, A., & DiAntonio, A. (2019). Wnd/DLK Is a Critical Target of FMRP Responsible for
Neurodevelopmental and Behavior Defects in the Drosophila Model of Fragile X
Syndrome. Cell Rep, 28(10), 2581-2593 e2585. doi:10.1016/j.celrep.2019.08.001
Russo, A., Goel, P., Brace, E. J., Buser, C., Dickman, D., & DiAntonio, A. (2019). The E3 ligase
Highwire promotes synaptic transmission by targeting the NAD-synthesizing enzyme
dNmnat. EMBO Rep, 20(3). doi:10.15252/embr.201846975
Sahoo, P. K., Smith, D. S., Perrone-Bizzozero, N., & Twiss, J. L. (2018). Axonal mRNA
transport and translation at a glance. J Cell Sci, 131(8). doi:10.1242/jcs.196808
Sanes, J. R., & Lichtman, J. W. (1999). DEVELOPMENT OF THE VERTEBRATE
NEUROMUSCULAR JUNCTION. Annual Review of Neuroscience, 22(Volume 22,
1999), 389-442. doi:https://doi.org/10.1146/annurev.neuro.22.1.389
Sanes, J. R., & Lichtman, J. W. (2001). Induction, assembly, maturation and maintenance of a
postsynaptic apparatus. Nature Reviews Neuroscience, 2(11), 791-805.
doi:10.1038/35097557
Schuster, C. M., Davis, G. W., Fetter, R. D., & Goodman, C. S. (1996). Genetic Dissection of
Structural and Functional Components of Synaptic Plasticity. I. Fasciclin II Controls
Synaptic Stabilization and Growth. Neuron, 17(4), 641-654. doi:10.1016/S0896-
6273(00)80197-X
Shapira, M., Zhai, R. G., Dresbach, T., Bresler, T., Torres, V. I., Gundelfinger, E. D., . . . Garner,
C. C. (2003). Unitary assembly of presynaptic active zones from Piccolo-Bassoon
transport vesicles. Neuron, 38(2), 237-252. doi:10.1016/s0896-6273(03)00207-1
Shin, J. E., Geisler, S., & DiAntonio, A. (2014). Dynamic regulation of SCG10 in regenerating
axons after injury. Exp Neurol, 252, 1-11. doi:10.1016/j.expneurol.2013.11.007
Shin, J. E., Ha, H., Kim, Y. K., Cho, Y., & DiAntonio, A. (2019). DLK regulates a distinctive
transcriptional regeneration program after peripheral nerve injury. Neurobiol Dis, 127,
178-192. doi:10.1016/j.nbd.2019.02.001

137
Shin, J. E., Miller, B. R., Babetto, E., Cho, Y., Sasaki, Y., Qayum, S., . . . DiAntonio, A. (2012).
SCG10 is a JNK target in the axonal degeneration pathway. Proc Natl Acad Sci U S A,
109(52), E3696-3705. doi:10.1073/pnas.1216204109
Son, Y.-J., & Thompson, W. J. (1995). Nerve sprouting in muscle is induced and guided by
processes extended by schwann cells. Neuron, 14(1), 133-141.
doi:https://doi.org/10.1016/0896-6273(95)90247-3
Son, Y. J., & Thompson, W. J. (1995). Schwann cell processes guide regeneration of peripheral
axons. Neuron, 14(1), 125-132. doi:10.1016/0896-6273(95)90246-5
Stewart, B. A., Atwood, H. L., Renmger, J. J., Wang, J., & Wu, C. F. (1994). Improved stability
of Drosophila larval neuromuscular prepa- rations in haemolymph-like physiological
solutions. J. Comp. Physiol., 175, 179-191.
Szüts, D., & Bienz, M. (2000). LexA chimeras reveal the function of Drosophila Fos as a
context-dependent transcriptional activator. Proceedings of the National Academy of
Sciences, 97(10), 5351-5356. doi:doi:10.1073/pnas.97.10.5351
Taetzsch, T., & Valdez, G. (2018). NMJ maintenance and repair in aging. Curr Opin Physiol, 4,
57-64. doi:10.1016/j.cophys.2018.05.007
Taylor, J. P., Brown, R. H., Jr., & Cleveland, D. W. (2016). Decoding ALS: from genes to
mechanism. Nature, 539(7628), 197-206. doi:10.1038/nature20413
Turrigiano, G. (2012). Homeostatic synaptic plasticity: local and global mechanisms for
stabilizing neuronal function. Cold Spring Harbor perspectives in biology, 4(1), a005736.
Turrigiano, G. G. (2008). The self-tuning neuron: synaptic scaling of excitatory synapses. Cell,
135(3), 422-435. doi:10.1016/j.cell.2008.10.008
Valakh, V., Frey, E., Babetto, E., Walker, L. J., & DiAntonio, A. (2015). Cytoskeletal disruption
activates the DLK/JNK pathway, which promotes axonal regeneration and mimics a
preconditioning injury. Neurobiol Dis, 77, 13-25. doi:10.1016/j.nbd.2015.02.014
Vaquie, A., Sauvain, A., Duman, M., Nocera, G., Egger, B., Meyenhofer, F., . . . Jacob, C.
(2019). Injured Axons Instruct Schwann Cells to Build Constricting Actin Spheres to
Accelerate Axonal Disintegration. Cell Rep, 27(11), 3152-3166 e3157.
doi:10.1016/j.celrep.2019.05.060

138
Venken, K. J., Simpson, J. H., & Bellen, H. J. (2011). Genetic manipulation of genes and cells in
the nervous system of the fruit fly. Neuron, 72(2), 202-230.
doi:10.1016/j.neuron.2011.09.021
Verhey, K. J. (2007). Motor Proteins: Trafficking and Signaling Collide. Current Biology, 17(18),
R804-R806. doi:https://doi.org/10.1016/j.cub.2007.07.005
Verstreken, P., Kjaerulff, O., Lloyd, T. E., Atkinson, R., Zhou, Y., Meinertzhagen, I. A., & Bellen,
H. J. (2002). Endophilin mutations block clathrin-mediated endocytosis but not
neurotransmitter release. Cell, 109(1), 101-112.
Wagh, D. A., Rasse, T. M., Asan, E., Hofbauer, A., Schwenkert, I., Dürrbeck, H., . . . Qin, G.
(2006). Bruchpilot, a protein with homology to ELKS/CAST, is required for structural
integrity and function of synaptic active zones in Drosophila. Neuron, 49(6), 833-844.
Walkiewicz, M. A., & Stern, M. (2009). Increased insulin/insulin growth factor signaling
advances the onset of metamorphosis in Drosophila. PLoS One, 4(4), e5072.
doi:10.1371/journal.pone.0005072
Wan, H. I., DiAntonio, A., Fetter, R. D., Bergstrom, K., Strauss, R., & Goodman, C. S. (2000).
Highwire regulates synaptic growth in Drosophila. Neuron, 26(2), 313-329.
doi:10.1016/s0896-6273(00)81166-6
Wang, X., Kim, J. H., Bazzi, M., Robinson, S., Collins, C. A., & Ye, B. (2013). Bimodal control of
dendritic and axonal growth by the dual leucine zipper kinase pathway. PLoS Biol, 11(6),
e1001572. doi:10.1371/journal.pbio.1001572
Watkins, T. A., Wang, B., Huntwork-Rodriguez, S., Yang, J., Jiang, Z., Eastham-Anderson,
J., . . . Lewcock, J. W. (2013). DLK initiates a transcriptional program that couples
apoptotic and regenerative responses to axonal injury. Proceedings of the National
Academy of Sciences, 110(10), 4039-4044. doi:doi:10.1073/pnas.1211074110
Wondolowski, J., & Dickman, D. (2013). Emerging links between homeostatic synaptic plasticity
and neurological disease. Frontiers in cellular neuroscience, 7, 223.
Wu, C., Wairkar, Y. P., Collins, C. A., & DiAntonio, A. (2005). Highwire function at the
Drosophila neuromuscular junction: spatial, structural, and temporal requirements. J
Neurosci, 25(42), 9557-9566. doi:10.1523/JNEUROSCI.2532-05.2005

139
Xiong, X., Hao, Y., Sun, K., Li, J., Li, X., Mishra, B., . . . Collins, C. A. (2012). The Highwire
ubiquitin ligase promotes axonal degeneration by tuning levels of Nmnat protein. PLoS
Biol, 10(12), e1001440. doi:10.1371/journal.pbio.1001440
Xiong, X., Wang, X., Ewanek, R., Bhat, P., Diantonio, A., & Collins, C. A. (2010). Protein
turnover of the Wallenda/DLK kinase regulates a retrograde response to axonal injury. J
Cell Biol, 191(1), 211-223. doi:10.1083/jcb.201006039
Xu, Z., Poidevin, M., Li, X., Li, Y., Shu, L., Nelson, D. L., . . . Wingo, T. S. (2013). Expanded
GGGGCC repeat RNA associated with amyotrophic lateral sclerosis and frontotemporal
dementia causes neurodegeneration. Proceedings of the National Academy of Sciences,
110(19), 7778-7783.
Yamanaka, N., Rewitz, K. F., & O'Connor, M. B. (2013). Ecdysone control of developmental
transitions: lessons from Drosophila research. Annu Rev Entomol, 58, 497-516.
doi:10.1146/annurev-ento-120811-153608
Yan, D., Wu, Z., Chisholm, A. D., & Jin, Y. (2009). The DLK-1 kinase promotes mRNA stability
and local translation in C. elegans synapses and axon regeneration. Cell, 138(5), 1005-
1018. doi:10.1016/j.cell.2009.06.023
Zhou, L., Schnitzler, A., Agapite, J., Schwartz, L. M., Steller, H., & Nambu, J. R. (1997).
Cooperative functions of the reaper and head involution defective genes in
the programmed cell death of Drosophila central nervous system
midline cells. Proceedings of the National Academy of Sciences, 94(10), 5131-
5136. doi:doi:10.1073/pnas.94.10.5131

140
Appendices

141
Appendix A: Nancy’s fly storage notes: Where should I store my flies?
*A note about our incubators: We currently have two 18°C and two 25°C incubators. Put
all your trays into the same 18°C incubator and 25°C incubator! If the temperature ever
fluctuates, then at least ALL your flies will be subjected to the same conditions.
18°C Incubator
This is great for long-term storage! The longest you should leave flies here is 6 weeks.
The average life cycle is 18 days here (some mutant lines may take even longer).
What type of flies do I leave here?
o Backup stocks of flies that I have at room temp. and at 25 degrees
o Stocks that you need to maintain but might not need right now
o All collected virgins
Flip vials in the 18°C every 4 or 5 weeks. I prefer every 4 weeks because my flies are
sick, so they benefit from more frequent flipping. Yours may be healthier, so you can push it to 5
weeks.
Room Temperature
Here’s a nice middle ground between the two extreme temperatures for raising flies and
is considered to be between 20°C and 25°C. The temperature fluctuates throughout the day and
depending on the season, so this isn’t the most reliable temperature to work with.
What type of flies do I leave here?
o My “working stocks” that I am currently using for genetics
o Flies I need quick access to
Flip vials raised at room temperature every 2-3 weeks.

142
25°C Incubator
Flies raised in this temperature will develop the fastest – 10 day average life cycle.
What type of flies do I leave here?
o ALL crosses. I want the next generation of adults or the larval progeny ASAP for
my genetics and experiments
o Gal4 drivers and balancers
o Any flies you are raising up for virgin collections
Flip vials raised in the 25°C every 1-2 weeks. These flies develop quickly, so provide
them with fresh food often.
For some experiments and genetics, you may flip vials more often (every 2-3 days).
My virgin collection strategy:
Use the temperatures and developmental timelines to your advantage!
1) Leave trays with vials for virgin collecting in the 18°C overnight. This slows the flies
down and reduces the number that eclose (pop out of the pupal casings) overnight.
2) Collect virgin female flies between 9 and 10 am – the most flies will pop out at this time.
Look for the green waste (meconium) in their abdomen.
*Completely clear the vials of adults during these early collections!*
3) Place this tray in the 25°C incubator for up to 7 hours. If you properly cleared all adults
out earlier, every female fly you see will be a virgin (regardless of morphology.)
*When in doubt, look for the meconium.

143
4) After the second collection of the day, place the tray back in the 18°C for overnight
storage.
5) Repeat this cycle every day for up to 18 days (this only applies to crosses)! You’ll avoid
collecting flies from the F2 generation by tossing out vials on day 18.

144
Appendix B: Nancy’s Third Instar Larval Dissection Protocol and Notes (for IMAGING)
Equipment to gather before the dissection:
• Ice bucket filled with ice
• 2 forceps
o I prefer one with a sharp tip and one with a blunt tip
• 1 pair of scissors
• Pre-cut insect pins
o Shorter pins are easier to manipulate. Only use the sharp ends and toss the
leftover portions
o I recommend prepping extras because you WILL lose pins
• Plate(s) with silicone at the bottom
• 4 glass pipettes with bulbs
o 1 each for HL3, PBST, fixative, waste solutions
• 1 labeled Eppendorf tube filled with PBST to store dissected filets
Reagents:
• HL3 (hemolymph-like saline); this should be on ice
• PBST (Phosphate-buffered saline with TritonX-100 detergent); this should be on ice
• Fixative
o Bouin’s fixative (yellow solution made of picric acid, acetic acid, and
formaldehyde); this can be kept at room temperature
o Paraformaldehyde (abbreviated as PFA); this should be on ice
o Ethanol or Methanol; these should be on ice

145
***There are special cases where you’ll stain a non-permeabilized set of tissues.
• DO NOT use reagents with PBST for these experiments – TritonX-100 pokes holes in
cell membranes and allows antibodies to enter the cytoplasm.
• Instead, use PBS for your washes and dilute all reagents in PBS.

146
Dissection steps:
1) Select your wandering third instar larvae and wash them in distilled water to remove any
food stuck on their cuticle.
2) Move them to your dissection plate with pre-cut pins.
a) We don’t typically image the posterior half of larvae, so we want to carefully pin
and preserve the anterior half.
3) Pin the posterior (tail) end, then the anterior (head) end.
a) We don’t typically image the posterior half of larvae, so we want to carefully pin
and preserve the anterior half.
b) I found that stabilizing the posterior first, then using my second pair of forceps to
gently hold the larvae in place helped me pin the head with minimal damage.
c) DO NOT stretch the animal in this anterior-posterior axis yet – it’ll make lateral
stretching more difficult later.
4) Add 1-2 drops of cold HL3 to cover the larva.
5) Use scissors to cut the animal from posterior to anterior, forking into a “Y” shape near
the anterior pin.
Anterior pin, you’ll see the
black mouth hooks here
Posterior pin, you’ll see
two brown dots here

147
6) At this point, the guts and fat body (fluffy white tissue) will spill out.
7) Rotate your plate 180° so that the posterior is facing the top of your field of view.
Carefully complete the incision by cutting toward your posterior pin.
a) If you don’t do this step, you’ll leave a pouch at the posterior end. Cleaning and
stretching will be extremely annoying unless you fully open the animal.
8) From here, the goal is to clear the body wall of internal tissues (but not the CNS).
Carefully pull out the tissues with forceps or cut them out with your scissors.
DO NOT damage the body wall! The muscles we’ll image are on this wall.
9) Once most/all the tissues have been removed, place four pins at the sides of the body
wall to open it up and flatten it out into a hexagonal filet.
a) I aim for the A1 segment and then midway down the body wall to get the best
stretch of the A2 and A3 segments for imaging.
b) Once all 6 pins have been placed, I further stretch the tissue by manipulating one
pin at a time and gently pulling outward. A good stretch here will save you hours
at the confocal since you’ll have fewer optical sections to scan.
A1-A5
segments

148
***If you’re bouton counting at muscles 6 and 7, your goal is to stretch them to get a
beautiful “X” shaped synapse with ALL Ib and Is boutons visible.
• I recommend placing your pins closer to the midline to achieve the best stretch. Only do
this if you don’t plan on imaging muscle 4 later.
10) Finish cleaning the filets and carefully remove the CNS using forceps.
11) Remove all HL3 and floating debris using your ‘waste’ pipette.
a) Add more HL3 as needed to remove all debris.
12) Add 1-2 drops of fixative and incubate for the necessary amount of time:
a) Bouin’s for 5 minutes
b) PFA for 10 minutes
c) Ethanol or methanol for 5 minutes
d) DO NOT incubate longer than this - you might over-fix your tissues and prevent
antibody binding.
13) Remove the fixative using your ‘waste’ pipette and rinse quickly with PBST on the plate.
a) Use two pipettes here: one for PBST only and one for your waste.
b) I usually add three drops of PBST to each larva for rinses, and I rinse three
times.
c) The filets will still look yellowed (if using Bouin’s) and that’s fine! Later washes
will clean this up.
14) Make your genotype cuts at this point (when the tissues are clean, but still pinned down
to the silicone).

149
15) Carefully remove the 6 pins using both of your forceps: one to hold the body wall down
and another to pull the pins out. If you aren’t careful, you may rip your filet!
16) Transfer your filets to an Eppendorf tube with 1mL of PBST. You’ll add filets to this tube
until you’re ready to start the staining process
a) Before my first wash on the rocker, I’ll remove the liquid and add 1mL of fresh
PBST.
b) When working with Bouin’s you’ll see some of the fixative leech out into the
PBST in the tube. Replacing the PBST before your first wash will clear the yellow
stain from your tissues.
***For those planning to image endogenous GFP or RFP samples, protect your samples
from light from the fixation period to the end of the staining protocol.
• While fixing the tissues, I’ll cover my petri dish and samples with a box lid. I’ll also wrap
my Eppendorf tube in foil since samples sit there until you’ve finished dissecting.
• All washes on the rocker should be done under the box lid.

150
Clean-up procedure:
1) Wash your silicone-lined plate with distilled water. Dab dry with a KimWipe.
2) Fold a new KimWipe and spray a little bit of distilled water on it, then use this to wipe
down your forceps and scissors. Wait for everything to dry before you cap/cover and
store away.
3) Remove bulbs from the glass pipettes. Toss the pipettes in the glass waste bin (green
with a white top).
4) Dump your ice into the sink.
5) Put all reagents away:
o HL3, PBST, PFA, and the alcohol fixatives will go back into the fridge
o Bouin’s will go back in a tray on the benchtop.

151
Where do I find….?
• Ice buckets: stacked up next to the main sink in the lab
• HL3: in the glass door fridge next to Dion’s office
• PBST: this is also in the glass door fridge; look for ‘PBS with 0.03% TritonX-100’
• Bouin’s fixative: there is a bottle on the chemical shelves or in the fume hood; aliquot a
small amount for your use
• Ethanol and methanol: in the “Flammable Liquid” cabinet beneath the fume hood; aliquot
a small amount
• PFA: frozen 10X stock solutions are stored in 10mL tubes at the bottom of the -30°C
freezer
o ask around the lab for an aliquot before mixing some up
o this must be diluted and used within one week

152
Appendix C: Nancy’s Confocal Staining Protocol
Recipes:
o 5% Normal Donkey Serum (NDS) = 500µL + 9.5mL of PBST
o PFA = 500µL + 4.5mL of PBS
Pre-absorption Protocol:
1) Dissect 4-8 animals – these will be tossed later, so don’t worry about carefully dissecting
or stretching these.
2) Fix the filets normally and quickly wash the tissues on the plate.
3) Wash 3 times in PBST for 10-20 mins each on the rocker.
4) DO NOT BLOCK these tissues!
a) We pre-absorb antibodies to remove non-specific binding and reduce background
noise. Blocking solutions cover non-specific sites and prevent antibody binding,
which is the exact opposite of what we want for pre-absorptions.
5) Add your primary cocktail and place your tube on the rocker.
6) Incubate overnight at 4° or for 2 hours at room temperature.
7) Remove the cocktail and store it in a clean, labeled tube.
8) Toss the tube with the filets.

153
Staining steps:
1) Dissect and fix your animals - see my dissection protocol for detailed notes.
• Fix for 5 minutes in Bouin’s fixative
• Fix for 10 minutes in Paraformaldehyde (PFA)
• Fix for 5 minutes in Ethanol (EtOH) or Methanol (MeOH)
**These times are firm! If you over-fix your tissues, you may prevent antibodies from
binding their targets.
2) Wash 3 times in PBST for 10-20 minutes each on the rocker.
a) If you don’t have time to block your samples, at least perform step 2 to wash off
fixatives. After the washes, remove all liquid and add fresh PBST to the tube. Then
store the tube in the fridge FOR ONLY ONE NIGHT.
3) Remove PBST and add 5% NDS to block the tissues on the rocker: 500µL for 1-16
animals or 1mL for 17-32 animals. Block for 30 mins - 1 hr.
a) You can get away with a 30-minute block for BRP, vGlut, and DLG.
b) GluR antibodies and RBP require the full 1 hr.
c) anti-BRP is such a clean antibody that you don’t have to block your tissues at all.
**Blocking prevents unspecific binding from the antibodies. Keep this in the fridge when
not in use. Throw it away if something starts to grow in it.
4) Add primary antibodies directly into the blocking solution in the tube. Incubate overnight
on the shaker in the 4°C fridge.
a) Primaries can be used up to 3 times. If the signal becomes weak before then, throw
them away and use fresh primaries.

154
b) Instead of an overnight incubation, you can incubate primaries for 1-2 hours at room
temperature instead.
**If you’re reusing an old primary cocktail, then remove all liquid from the tube before
you add the pre-made cocktail.
Antibody Dilution
Volume
(for 500µL NDS)
Bouton
Counting
Guinea pig α-vGlut 1:100 5µL
Mouse α-DLG 1:100 5uL
Glutamate
Receptor Analysis
Guinea pig α-GluRIID 1:100 5µL
Mouse α-GluRIIA 1:50 10µL
Rabbit α-GluRIIB 1:100 5µL
5) Wash 3 times in PBST for 10-20 minutes each on the rocker.
6) Remove PBST and add secondary antibodies. Incubate for 2 hours at room temperature
IN THE DARK (under the box lid) on the rocker.
a) Secondaries are light-sensitive, so keeping them in the dark prevents
photobleaching/loss of signal. These can only be used ONCE.
**You can instead incubate samples with secondaries overnight in the 4°C fridge. But
you must protect the samples from light!!

155
Antibody Dilution
Volume
(for 500µL NDS)
Bouton
Counting
α-guinea pig 488 1:400 1.25µL
α-mouse cy3 1:400 1.25µL
α-HRP 647 1:200 2.5µL
Glutamate
Receptor Analysis
α-guinea pig 405 1:400 1.25µL
α-mouse 488 1:400 1.25µL
α-rabbit cy3 1:400 1.25µL
α-HRP 647 1:200 2.5µL
7) Wash 3 times in PBST for 10-20 minutes each on the rocker IN THE DARK.
8) Remove PBST and add 70% glycerol to the tube – this helps tissues equilibrate with
glycerol (the base for our mounting agent).
**Tissues will float to the top, but they’ll sink over time as they infuse with glycerol.
9) Store your tissues in the fridge (in the dark) until you’re ready to mount them onto a
microscope slide.
a) Mount within one week. A few of us have noticed that tissues stored for longer
appear “hazy”.

156
Why do we use Normal Donkey Serum?
• NDS contains antibodies that bind to all non-specific sites. This prevents our primaries
from binding those same sites and giving us “false positive” staining.
o Why do we use Donkey serum specifically? Since this serum contains antibodies
that will coat the entire tissue, we DO NOT want to use serum from animals that our
primary antibodies were raised in (so no Rabbit, Mouse, Guinea pig, or Chicken sera
allowed!) If we used one of these three serum types, we’d see the corresponding
secondary antibody binding everywhere. You’ll notice that all our secondaries were
raised in Donkey – it’s good to use Normal Serum from the animal that your
secondaries were raised in.
• Pre-absorbing antibodies also increases the specificity of our primaries.
• Primary vs. Secondary antibodies and why we use them: See “Indirect” method below.

157
Appendix D: Nancy’s Mounting Protocol for Confocal Imaging
1) Clean your microscope slide and coverslip with ethanol and a Kim Wipe.
2) Label your slide with:
o Genotypes or crosses (write these in the order that you’ll mount your tissues)
o Males or females (usually for X chromosome genetics)
o Primary and secondary antibodies
o I’ll shorthand this by writing “vGlut GP 405”, which means the primary was antivGlut raised in a guinea pig, and the secondary was anti-GP with a 405
fluorophore
o The date
3) With a pair of forceps, carefully transfer your tissues from the tube to your microscope
slide.
a) Don’t worry about the tissue arrangement now. We’ll organize everything once all
filets are on the slide.
b) I typically do this with my scope light off to avoid photobleaching.
4) Arrange your tissues by genotype/cross into a single row.
a) Match your genotypes with the labeling from Step 2!
b) I like to place my tissues in order from best to worst so that I’ll always image the best
synapses first. If I run out of time or the confocal breaks down, I don’t mind “wasting”
my worst tissues.
c) I recommend arranging your tissues into a grid to improve your imaging experience.

158
5) Flip the filets so that the muscles are facing upward.
a) Sometimes flipping your filets over a few times will help you see a difference
between the muscles and cuticle.
6) This is based on personal preference. The microscope inverts images, so if you want
things to look right-side up, mount your tissues upside-down (with the mouth hooks
toward the bottom of the slide.)
7) Dab excess glycerol from the slide with a KimWipe.
8) Using a razor blade, cut off the anterior and posterior portions of each filet and discard
them. These are the thickest parts of the animal, so removing them will make our sample
much thinner.
a) Avoid cutting into the abdominal segments that you’ll image later!!! So, maintain
abdominal segments A2 and A3 (or A2 - A5 if you’re working with a stathmin (stai)
mutant).
9) Soak up excess 70% glycerol surrounding your samples with a KimWipe.
10) Apply one drop of the VectaShield mounting agent onto your samples.
11) Using a pair of forceps, carefully spread out the VectaShield to cover all filets and fill out
the area that your coverslip will cover.
a) This will reduce drifting of samples when you apply your coverslip.

159
12) Add the coverslip at an angle and ensure VectaShield fills the entire area.
a) If you have excess, then soak it up with a KimWipe.
b) If you need more, apply a small drop at the coverslip’s edge and allow it to seep
between the slide and coverslip.
13) Seal the coverslip with clear nail polish. Allow the polish to dry before imaging.

Abstract (if available)

Abstract Cells in the neuromuscular junction (NMJ) are incredibly resilient and capable of adapting upon neuronal insult. When synaptic transmission is perturbed, the muscle is able to sense this and scale its receptors appropriately. This is also true when the neuron is undergoing injury-related signaling – synapses adapt to maintain a lower homeostatic threshold. Likewise, glia can sense when neurons are damaged or dying and expedite the process of degeneration and removal of debris. On the other hand, glia can also support neurons during their regrowth after axonal injury. Finally, when undergoing neurodegeneration due to disease-like states or cytoskeletal destabilization, proteins in the axon terminal and within the muscle membrane scale correspondingly.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Molecular mechanisms underlying the bi-directional control of presynaptic homeostatic plasticity

PDF

Engineering genetic tools to illustrate new insights into the homeostatic control of synaptic strength

PDF

Translational regulation and endosomal trafficking in synaptic adaptation to stress

PDF

Synaptic transmission, nutrient sensors, and aging in Drosophila melanogaster

PDF

Dual genetic screens for mutants in synaptic homeostatic plasticity and a characterization of insomniac as a regulator for retrograde homeostatic signaling

PDF

Presynaptic glutamate receptors and auxiliary subunits in neurotransmission and homeostatic potentiation

PDF

Uncovering cell type-specific roles of proteins involved in glutamatergic synapse regulation

PDF

Engineering 2D & 3D microphysiological systems for interrogating skeletal muscle tissues

PDF

The role of RhoGEFs in glutamatergic synapse development and human cognitive disorders

PDF

Uncovering synapse-specific roles of proteins implicated in complex brain disorders via novel and targeted approaches

PDF

The role of Rho-GEF signaling in synapse function and autism-related disorders

PDF

The role of the cofilin/Limk1 signaling pathway in axon growth during development and regeneration

PDF

Nebula/DSCR1 upregulation preserves axonal transport and memory function in a Drosophila model for Alzheimer's disease

PDF

On the Chow-Witt rings of toric surfaces

PDF

Characterizing the hippocampal synaptic and sleep abnormalities of a mouse model of human chromosome 16p11.2 microdeletion

PDF

Endogenous H2O2 signaling positively regulates the release of neuropeptides during a neuron-gut axis mediated oxidative stress response in caenorhabditis elegans

PDF

Differential regulation of anatomical and functional visual plasticity by NgR1

PDF

Making the most of at-home music: some student music careers are taking off outside the classroom

PDF

The role of a novel integrin activator in the maintenance of stem cell niche and activity-induced synaptic structural modifications in Drosophila

PDF

Elucidating neurodevelopmental consequences of syngap1 mutations and inactivated functional regions in human iPSC-derived neurons

Asset Metadata

Creator Tran, Nancy Lan (author)

Core Title Synaptic plasticity during neurodegeneration and axonal injury

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Degree Conferral Date 2024-05

Publication Date 05/17/2024

Defense Date 04/23/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag axon,Drosophila,injury,larva,neurodegeneration,neuromuscular junction,NMJ,OAI-PMH Harvest,plasticity,scaling,stress,synapse

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Herring, Bruce (committee chair), Chang, Karen (committee member), Dickman, Dion (committee member)

Creator Email nancy.l.tran@gmail.com,nancyltr@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113940240

Unique identifier UC113940240

Identifier etd-TranNancyL-12962.pdf (filename)

Legacy Identifier etd-TranNancyL-12962

Document Type Dissertation

Format theses (aat)

Rights Tran, Nancy Lan

Internet Media Type application/pdf

Type texts

Source 20240517-usctheses-batch-1155 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu