A Dissertation Entitled Rho Gtpase Signaling Modulates

Home , CDC42, Netrin, RAC1, RhoG, RhoGEF domain

A Dissertation

entitled

Rho GTPase Signaling Modulates Neurotransmission in Caenorhabditis elegans

by Shuang Hu

Submitted to the Graduate Faculty as partial fulfillment of the requirements for the

Doctor of Philosophy Degree in Biology

Dr. Robert Steven, Committee Chair

Dr. Richard Komuniecki, Committee Member

Dr. Bruce Bamber, Committee Member

Dr. Song-Tao Liu, Committee Member

Dr. William Messer, Committee Member

Dr. Scott Molitor, Committee Member

Dr. Patricia R. Komuniecki, Dean College of Graduate Studies

The University of Toledo

May 2013

An Abstract of Rho GTPase Signaling Modulates Neurotransmission in Caenorhabditis elegans

by Shuang Hu Submitted to the Graduate Faculty as Partial fulfillment of the requirements for the Doctor of Philosophy Degree in Biology The University of Toledo May 2013

Rho family GTPases act as molecular switches in the regulation of diverse cellular functions, including cell division, gene transcription and neurotransmission. In the model organism Caenorhabditis elegans, the Kalirin and Trio ortholog UNC-73 contains two RhoGEF domains that specifically activate either Rac or Rho GTPases, respectively. The RhoGEF1 domain and the Rac pathway are required for axon guidance in neuronal development, while the RhoGEF2 domain is involved in the control of locomotion, possibly through the modulation of neurotransmission. The unc-73 gene is expressed as multiple isoforms that contain either one or both RhoGEF domains. This study focuses on defining the function of the UNC-73 RhoGEF2 containing isoforms and the Rho signaling pathway in the adult nervous system of C. elegans.

Animals with mutations in the unc-73 RhoGEF2 domain move more slowly than wild-type animals and this locomotory defect can be rescued by pan-neuronal expression of the UNC-73E isoform, which contains only the RhoGEF2 domain. In addition, unc-73

RhoGEF2 mutants are resistant to the cholinesterase inhibitor aldicarb and are hypersensitive to the cholinergic agonist levamisole, without any obvious changes in synaptic structure based on the localization of fluorescence-tagged synaptobrevin. These

iii results suggest that the UNC-73 RhoGEF2 isoforms may modulate synaptic strength and cholinergic signaling at the level of the neuromuscular junction. Neuropeptide signaling is also affected in UNC-73 RhoGEF2 mutants. RhoGEF2 mutants expressing the YFP- tagged neuropeptide NLP-21 in cholinergic motorneurons exhibit decreased axonal, but not somal, YFP fluorescence compared to wild-type animals, suggesting NLP-21::YFP packaging into vesicles or transport to the axons is inhibited. YFP fluorescence in the coelomocytes, an indicator of neuropeptide release, is also decreased in these animals.

These results suggest the UNC-73 RhoGEF2 isoforms may play a neuromodulatory role in the regulation of locomotion, perhaps by regulating dense core vesicle (DCV)- mediated neuropeptide signaling at the level of axonal transport or neuropeptide packaging.

The lethargic locomotory and aldicarb resistant phenotypes of unc-73 RhoGEF2 mutants were bypassed in animals with constitutively active (CA) Gαs signaling. Gαs(CA) expression is required in both muscles and neurons to rescue unc-73 locomotory phenotypes in RhoGEF2 mutants, as the expression in either muscle or neurons alone failed to yield substantial rescue. unc-73 RhoGEF2 mutants exhibited phenotypes similar to rab-2 and unc-31 mutants, both of which encode proteins that are involved in DCV- mediated signaling. Gαs(CA) signaling also rescues unc-31 and rab-2 locomotory defects, supporting our hypothesis that UNC-73 RhoGEF2 signaling may modulate DCV- mediated neurotransmission.

Together, these results indicate UNC-73 RhoGEF2 isoforms may regulate cholinergic signaling indirectly by modulating neuropeptide signaling and that the Gαs pathway acts downstream of, or in parallel to, the neuropeptide signaling pathway.

Acknowledgments

First I thank my advisor and mentor Dr. Robert Steven for his help and guidance throughout my graduate career. He gave me a precious opportunity for my Ph.D. study, and taught me many lab skills hands on. He is always willing to give a hand whenever I need help. Thanks him being gentle and patient to me. Second I would like to thank my husband Qing for his constantly support. I would not finish my Ph.D. without his encouragement and help. He also provided professional consultant on statistic problems. I also thank all the lab members in Dr. Steven’s lab, especially Todd Cramer, Thuy Tran, and John Farver for their companions and technical helps. We spent many long hours working together in the lab. I thank my committee members, Dr. Richard Komuniecki,

Dr. Bruce Bamber, Dr. Song-Tao Liu, Dr. William Messer, and Dr. Scott Molitor,

Committee Member. Dr. Komuniecki gave me many advices on my project, and his joint lab meeting broadened my view during my study. I thank other students in the biological department, Vera Hapiak, Gareth Harris, Yu Zhan, Haiying Li, Aaron Tipton, and Leah

Rider who were there to offer me suggestions. Dr. Bruce Bamber kindly let me use the software Velocity for fluorescent signal analysis. Dr. Maria Diakonova provided TR-

BSA for my coelomocyte experiment. Dr. Song-Tao Liu and Aaron Tipton shared their vectors and protocols to me for pull-down assay.

Table of Contents

Acknowledgments...... v

Table of Contents ...... vi

List of Tables ...... x

List of Figures ...... xi

List of Abbreviations ...... 1

1 Introduction ...... 2

1.1 Rho GTPase signaling in the nervous system ...... 2

1.2 RhoGEF proteins regulate Rho GTPase signaling as activators and scaffold

proteins ...... 3

1.3 UNC-73 is a C. elegans RhoGEF with two functional domains ...... 5

1.4 The model organism Caenorhabditis elegans ...... 10

1.5 C. elegans locomotory regulation ...... 12

1.6 Neurotransmission ...... 14

1.7 Heterotrimeric G protein pathways and neurotransmission ...... 16

1.8 RhoA/RHO-1 Signaling in neurotransmission ...... 17

1.9 Overall objectives: ...... 19

2 Materials And Methods...... 20

2.1 Worm strains ...... 20

2.2 Transgenic lines ...... 23

2.3 DNA constructs ...... 24

2.3.1 DNA constructs in C. elegans ...... 24

2.3.2 Construction of plasmids for protein expression in E. coli ...... 27

2.3.3 Construction of plasmids for protein expression in HEK293T cells ...... 29

2.4 Live animal assays ...... 29

2.4.1 Locomotion assays ...... 29

2.4.2 Reversal assay ...... 29

2.4.3 Drug sensitivity assays ...... 30

2.4.4 PMA treatment ...... 31

2.4.5 Coelomocyte endocytosis assay ...... 31

2.4.6 Body length measurement...... 32

2.5 Cell specific RNA interference ...... 32

2.6 Immunoblotting...... 34

2.7 Protein preparation and pull-down assays ...... 34

2.7.1 Protein preparation from E. coli...... 34

2.7.2 Protein preparation from HEK293T cells ...... 35

2.7.3 Pull-down assay ...... 35

vii

2.8 Microscopy and imaging analysis ...... 36

3 Results ...... 38

3.1 UNC-73 RhoGEF2 isoforms are neuronal Rho GEFs required for locomotion ... 38

3.1.1 Cell-specific rescue of unc-73 mutant locomotory defects...... 38

3.1.2 RHO-1 functions downstream of UNC-73 RhoGEF2 activity in vivo ...... 45

3.2 UNC-73 RhoGEF2 isoforms modulate cholinergic signaling...... 47

3.3 unc-73 mutants have neuropeptide processing defects ...... 52

3.4 Activated Gαs pathway mutants suppress unc-73 RhoGEF2 mutant defects ...... 55

3.4.1 The activated Gαs pathway is required in both muscles and nervous system

to fully rescue the locomotory phenotype of unc-73 mutants...... 55

3.4.2 Double mutants of unc-73 RhoGEF2 and the activated Gαs pathway are

hypersensitive to aldicarb and levamisole ...... 59

3.4.3 Gαs pathway activation alters neuropeptide levels ...... 61

3.5 The relationship between UNC-73 and other molecules in the DCV-mediated

signaling pathway ...... 64

3.5.1 rab-2,unc-31 and unc-73 RhoGEF2 mutants have similar slow movement

phenotypes that are rescued by Gαs pathway activation ...... 64

3.5.2 rab-2, unc-31 and unc-73 RhoGEF2 mutants have similar responses to

levamisole ...... 67

3.5.3 unc-73 RhoGEF2, rab-2 and unc-31 mutants respond similarly to PMA

treatment ...... 68

viii

3.5.4 Analysis of unc-73 RhoGEF2 double mutants with egl-3 and tom-1...... 70

3.5.5 gsa-1(gf) and unc-73(lf) mutants have a shorter body length than wildtype 73

3.6 An examination of the relationship between the UNC-73 RhoGEF2 pathway and

glutamate receptor signaling ...... 75

3.6.1 UNC-73 overexpression increases reversal frequency ...... 75

3.6.2 GLR-1 is mislocalized in unc-73 mutants ...... 77

3.7 The UNC-73 RhoGEF2 domain is most similar to mammalian Kalirin, Trio and

p63RhoGEF RhoGEF domains...... 79

4 Discussion ...... 83

4.1 UNC-73 RhoGEF2 isoforms and the Gαq pathway ...... 83

4.2 UNC-73 RhoGEF2 isoforms and the Gαs pathway ...... 85

4.3 UNC-73 is involved in DCV-mediated secretion pathways ...... 86

4.4 unc-73 RhoGEF2 mutant aldicarb and levamisole sensitivities ...... 88

4.5 UNC-73 RhoGEF2 isoform function and GLR-1 localization ...... 90

4.6 A model of UNC-73/Rho signaling in the regulation of neurotransmission ...... 92

References ...... 95

Appendix A UNC-73C1 interacting proteins ...... 111

Appendix B Cell specific RNAi ...... 117

List of Tables

Table 1. Single worm PCR primer pairs used to detect DNA mutations ...... 22

Table 2. PCR primers used to amplify cell-specific promoter regions...... 26

Table 3. PCR primers used to amplify cDNAs for protein expression...... 28

Table 4. PCR primers used to amplify the target gene fragments for cell-specific RNAi 33

Table 5. A list of cell specific promoters used in cell-specific rescue experiments ...... 41

List of Figures

Figure 1. Structure diagram of the UNC-73 RhoGEF2 isoforms...... 7

Figure 2. Neuronal expression of the UNC-73E isoform rescued the unc-73 slow

locomotory phenotype...... 42

Figure 3. promoter::UNC-73E::GFP cell-specific expression patterns...... 43

Figure 4. UNC-73E overexpression in different cell types in a wild-type background does

not increase locomotion rate...... 44

Figure 5. UNC-73 functions upstream of Rho to regulate locomotion...... 46

Figure 6. unc-73 mutants have altered cholinergic signaling, with no obvious changes in

basic synaptic structure...... 50

Figure 7. UNC-73E overexpression in a wild-type background did not change levamisole

sensitivity...... 51

Figure 8. Neuropeptide level is reduced in unc-73 mutants...... 53

Figure 9. Constitutive activation of the Gαs pathway rescues the unc-73 slow locomotion

defect...... 58

Figure 10. Activated Gαs pathway mutants and double mutants with unc-73 are

hypersensitive to aldicarb...... 60

Figure 11. Activated Gαs pathway mutants and the double mutants with unc-73 are

hypersensitive to levamisole...... 60

Figure 12. Neuropeptide levels are altered in a Gαs pathway mutant background...... 63

Figure 13. rab-2, unc-31, and unc-73 RhoGEF2 mutants have a similar slow locomotion

defect rescued by Gαs pathway activation...... 66

Figure 14. rab-2 and unc-31 mutants display altered sensitivity to levamisole and aldicarb.

...... 67

Figure 15. Animals treated with PMA display a similar coiled posture...... 69

Figure 16. Analysis of unc-73 RhoGEF2 double mutants with egl-3 and tom-1...... 72

Figure 17. The gsa-1(gf) and unc-73 RhoGEF2 (lf) mutants have a shorter body length

than wildtype...... 74

Figure 18. Overexpression of UNC-73E increases reversal frequency...... 76

Figure 19. unc-73 and rab-2 mutants have an altered GLR-1::GFP localization...... 78

Figure 20. The UNC-73 RhoGEF2 domain is homologous to Trio RhoGEF2, Kalirin

RhoGEF2, and p63RhoGEF RhoGEF domains...... 81

Figure 21. Multi-alignment of UNC-73RhoGEF2, Trio RhoGEF2, Kalirin RhoGEF2, and

p63RhoGEF RhoGEF domains...... 82

Figure 22. The working model of UNC-73 RhoGEF2 isoforms and Rho signaling in

neurotransmission...... 94

Figure 23. LIN-2 interacts with UNC-73C1 in vitro...... 113

Figure 24. F5512C.1 weakly interacts with UNC-73C1 in vitro...... 114

Figure 25. Cell-specific RNAi phenotypes for unc-73 and other related genes...... 119

xii

List of Abbreviations

ACh ...... Acetylcholine ACY ...... Adenylyl cyclase

CA ...... Constitutively active cAMP ...... Cyclic AMP

DAG ...... Diacylglycerol DCV ...... Dense core vesicle DGK ...... Diacylglycerol kinase DH ...... Dbl homology

EGL ...... Egg-laying defective phenotype ER ...... Endoplasmic reticulum

FLP ...... FMRF(Phe-Met-Arg-Phe)-amide-like neuropeptides

GAP...... GTPase-activating protein GDI ...... Guanine nucleotide dissociation inhibitor GEF ...... Guanine nucleotide exchange factor GFP ...... Green fluorescent protein GSA...... Gαs

IP3 ...... Inositol 1,4,5-triphosphate Is[D1]ev802 ...... The lethal unc-73 mutation rescued by the UNC-73D1 isoform

NGM ...... Nematode growth media NMDA ...... N-Methyl-D-aspartic acid or N-Methyl-D-aspartate

PAM ...... Peptidylglycine α-amidating monooxygenase PH ...... Pleckstrin homology PIP2 ...... Phosphatidylinositol 4,5-bisphosphate PKA...... Protein kinase A PLCβ ...... Phospholipase Cβ PMA ...... Phorbol ester (phorbol 12- myristate 13 acetate) PSD ...... Postsynaptic densities

RGS ...... Regulator of G-protein signaling

RNAi ...... RNA interference ROCK ...... Rho associated kinase

SV ...... Synaptic vesicle SEM ...... Standard error of the mean

UNC ...... Uncoordinated phenotype

Chapter 1

1 Introduction

The general goal of my study is to understand Rho GTPase signaling in the nervous system by examining the Rho GTPase guanine nucleotide exchange factor

(RhoGEF), UNC-73, of the model organism Caenorhabditis elegans. The study of this C. elegans RhoGEF will expand our knowledge of the molecular basis of signal transduction in the nervous system and help us understand the mechanisms of neurotransmission regulation in humans.

1.1 Rho GTPase signaling in the nervous system

Rho GTPases belong to the superfamily of Ras-related small GTPases that function as molecular switches in intracellular signaling pathways found in all eukaryotic cells (Jaffe & Hall, 2005). So far 22 mammalian genes have been identified encoding

Rho GTPases (Rossman et al, 2005), while only five Rho family GTPases are present in

C. elegans (Lundquist, 2006). The most extensively studied Rho GTPases are Rho, Rac, and Cdc42 in mammals, with Rho and Rac consisting of multiple family members with redundant or distinct functions (Govek et al, 2005). In C. elegans, the five Rho GTPases homologs are rho-1, rac-2, cdc-42, ced-10, and mig-2 (Lundquist, 2006).

Rho GTPases regulate diverse cellular functions, including cell division, gene transcription and regulation of the actin cytoskeleton. They act as molecular switches that have an “on” status when the GTPase is bound to GTP and “off” when bound to GDP

(Jaffe & Hall, 2005). Rho GTPases have intrinsic catalytic activity to hydrolyze GTP into GDP and the cycling between the GTP- and GDP-bound forms is regulated by GEFs and GAPs (Heasman & Ridley, 2008). GEFs activate GTPases by stabilizing a GTPase-

GEF intermediate complex that facilitates the replacement of GDP with GTP, since GTP is found in excess over GDP inside cells (Worthylake et al, 2000). Two groups of proteins inhibit GTPase signaling. GTPase-activating proteins (GAPs), which accelerate the intrinsic GTPase activity, promote GTP hydrolysis, and guanine nucleotide dissociation inhibitors (GDIs), which prevent GDP dissociation, also inhibit GTPase signaling (Schmidt et al, 2002).

In the nervous system, Rho GTPases regulate neuronal morphology, development and function (Hall & Lalli, 2010; Heasman & Ridley, 2008). In the case of axon outgrowth, CDC42 specifically regulates axon specification. CDC42 is activated at the tip of the presumptive axon to determine neuronal polarity (Schwamborn & Puschel, 2004).

Rac functions in axon guidance in response to guidance cues and is involved in dendritic spine development. For example, when the guidance receptor plexin binds the guidance cue semaphorin4D, plexin competitively binds Rac to leave the downstream kinase Pak free to regulate cytoskeletal structures allowing for axon retraction or turning (Vikis et al,

2002). In mammalian dendrites, constitutively activated Rac, RacV12, and dominant- negative Rac, RacN17, change the morphology of dendritic spines, which are small, actin-based protrusions involved in synaptic plasticity (Zhang et al, 2003a). At the axon

2 growth cone, a dynamic fan-shaped structure rich in actin and microtubules at the tip of the growing axon, Rho is required for growth cone retraction. Y27632, a chemical that inhibits downstream Rho signaling, abolished LPA induced growth cone retraction in cultured neurons (Zhang et al, 2003b). Rho is involved in neurotransmission as well. Rho is proposed to stimulate acetylcholine (ACh) neurotransmitter release presynaptically by binding and inhibiting diacylglycerol kinase to increase diacylglycerol (DAG) levels presynaptically (Hiley et al, 2006; McMullan et al, 2006).

1.2 RhoGEF proteins regulate Rho GTPase signaling as activators and scaffold proteins

Rho GTPases are essentially ubiquitous and broadly involved in multiple cellular events so their regulators such as GEFs and GAPs are required to determine the time and place for Rho activation. In mammals there are 22 Rho family GTPases regulated by

GEFs and GAPs (Heasman & Ridley, 2008) and over 60 RhoGEFs, the Rho GTPases activators (Schmidt & Hall, 2002). In C. elegans, there are five Rho family members: rho-1, cdc-42, rac-2, ced-10, and mig-2 (Lundquist, 2006) with 19 RhoGEFs (Schmidt &

Hall, 2002), which is about a 1:4 ratio of Rho GTPases and GEFs. This abundance of

RhoGEFs and the fact that each RhoGEF activates certain Rho GTPases reveals the highly differentiated and specifically regulated role of GEFs in Rho signaling.

The two most important domains of RhoGEF proteins are the RhoGEF domain and Pleckstrin homology (PH) domain. The RhoGEF domain directly interacts with the switch regions of the Rho GTPase for the GTP/GDP exchange activity. Three conserved regions, CR1-CR3, and the C-terminal α6 helix make up the domain core of the RhoGEF

3 domain. A highly conserved glutamate in CR1 is crucial for interaction with the Rho

GTPase switch 1 region and catalytic function (Rossman et al, 2002). A PH domain usually follows the RhoGEF domain at its C-terminus. PH domains bind weakly to phosphoinositides but their exact function is controversial (Rossman et al, 2005; Schmidt

& Hall, 2002). In some cases, the PH domain increases GEF activity, for example, the

RhoGEF-PH fragments of Trio have greater nucleotide exchanging ability than the

RhoGEF domain alone (Liu et al, 1998; Rossman & Campbell, 2000). In the case of Sos1, the PH domain inhibits RhoGEF domain activity by folding back onto the RhoGEF domain to prevent substrate binding (Soisson et al, 1998). The membrane docking functions of the PH domain contribute to intracellular localization that is essential for

Tiam1GEF function (Ma et al, 1997; Stam et al, 1997).

In addition to the RhoGEF and PH domains, other catalytic domains, such as protein or lipid interaction domains are commonly found in GEFs. For example, Vav1 contains a CH domain, a zinc-finger motif, one SH2 domain and two SH3 domains in addition to the RhoGEF and PH domains (Romero & Fischer, 1996). A RhoGAP domain is found in Abr, a RhoA-specific GEF (Tan et al, 1993), and a RasGEF domain is found in some Rac-specific GEFs such as Sos1 (Boriack-Sjodin et al, 1998). The presence of these interaction domains indicates that RhoGEFs may serve as scaffold proteins or signal localization hubs (Bos et al, 2007).

GEFs mediate signaling between heterotrimeric G proteins and the small monomeric GTPases such as Ras and Rho. This was first recognized in the study of

RhoGEFs containing a regulator of G protein signaling (RGS) domain, which functions as a GAP for heterotrimeric GTPases (Rossman et al, 2005). For example, p115RhoGEF

4 has dual functions to increase Gα13 subunit hydrolysis activity through the RGS domain and to activate RhoA through the RhoGEF domain (Hart et al, 1998). However, the interaction between a RhoGEF and Gα subunit is not always mediated by the RGS domain. For example, p63RhoGEF, which does not contain a RGS domain, also mediates a connection between Gαq and Rho signaling. Gαq directly binds to the C-terminus of the p63RhoGEF PH domain to enhance its GEF activity. The presence of activated Gαq was essential for p63RhoGEF to activate RhoA in vitro (Rojas et al, 2007). This Gα dependent activation was also shown with Gα q/EGL-30 and UNC-73E/Trio in C. elegans

(Williams et al, 2007).

RhoGEFs are important integrators of upstream signaling. For example, Ephexin, a Rho specific GEF, is activated upon the binding of EphinA to the EphA receptor when the growth cone reaches its target. In turn, Ephexin activates the Rho pathway and triggers the collapse of the growth cone (Sahin et al, 2005; Shamah et al, 2001). This Rho activated growth cone retraction is mediated by ROCK (Rho associated kinase), since

LPA induced growth cone retraction is abolished when a ROCK inhibitor is applied to cultured neurons (Zhang et al, 2003b).

1.3 UNC-73 is a C. elegans RhoGEF with two functional domains

The study is focused on the examination of the UNC-73 RhoGEF in C. elegans.

Full length UNC-73 contains two RhoGEF domains that specifically activate Rac and

Rho GTPases, respectively. Mutations affecting the Rac-specific RhoGEF1domain result in a developmental axon guidance defect and exhibit uncoordinated movement, which is the reason behind the unc name for the gene (Steven et al, 1998). Deletion of the Rho-

5 specific RhoGEF2 domain, as it occurs in the unc-73(ev802) allele, results in an early larval stage arrest phenotype due to weakened pharyngeal pumping. This defect is rescued by UNC-73D1 RhoGEF2 isoform expression specifically in the pharynx muscles

(Steven et al, 2005). The rescued viable Is[unc-73D1]unc-73(ev802) animals (notated as

Is[D1]; unc-73(ev802)), containing a construct that expresses UNC-73D1 to rescue the larval arrest, and another unc-73 RhoGEF2 allele ce362, which has a point mutation in the RhoGEF2 domain, have similar coordinated yet slow locomotory phenotypes which are distinct from the severe uncoordinated RhoGEF1mutant phenotype (Steven et al,

2005; Williams et al, 2007).

The unc-73 gene encodes at least eight differentially expressed isoforms. Among them, six isoforms contain just the Rho specific RhoGEF2 domain without the RhoGEF1 domain (Fig. 1). The smallest isoform, UNC-73E, includes only the RhoGEF2 RhoGEF and PH functional domains without any additional defined protein-protein interaction domains. These UNC-73 RhoGEF2 isoforms are divided into two groups that are differentially expressed in a cell specific manner in correlation with their functions. The

UNC-73 D1/D2 isoforms are expressed in the pharynx muscles and can rescue the ev802 young larvae arrest defect. C1, C2/F and E are expressed in the nervous system although

C1 is expressed in a limited subset of neurons while E is strongly expressed throughout the nervous system. Neuronal expression of either of the UNC-73 RhoGEF2 C1/C2 or E isoforms rescues the slow locomotory defect of [D1]unc-73(ev802) or unc-73(ce362) mutants (Hu et al, 2011; Steven et al, 2005). The focus of this study is defining the role of

UNC-73 RhoGEF2 function in the nervous system-mediated regulation of locomotion.

Figure 1. Structure diagram of the UNC-73 RhoGEF2 isoforms. Adopted from Steven et.al 2005. SH3, Src homology 3 domain, binds to proline rich motifs; DH, Dbl-homologous, which is another name for the RhoGEF domain; PH, pleckstrin homology domain, is associated with the DH domain in most Rho/Rac/Cdc42 GEFs; Ig, Immunoglobulin domain; FnIII, fibronectin type 3 domain.

The full length UNC-73 and its mammalian homologs Trio and Kalirin belong to the Trio family of GEFs, which are unusual in that they contain dual RhoGEF domains.

Trio and Kalirin have similar protein structures with three functional domains including two RhoGEFs and a protein serine/threonine kinase domain along with spectrin like repeats in the N-terminal half, and SH3, Ig and FnIII domains towards the C terminus

(Bateman & Van Vactor, 2001).

Trio was originally identified as a protein that interacts with the cytoplasmic region of LAR, a receptor-like tyrosine phosphatase (Debant et al, 1996). Trio also interacts with other targets, such as the actin cytoskeleton by associating with filamin and

Tara (Bellanger et al, 2000; Seipel et al, 2001), the Notch pathway by associating with the Netrin receptor DCC (Briancon-Marjollet et al, 2008; Forsthoefel et al, 2005), and with cadherin-11 (Backer et al, 2007). Differentially spliced Trio isoforms are expressed in diverse cell types including the nervous system and the skeletal muscles throughout development and in the adult (O'Brien et al, 2000; Portales-Casamar et al, 2006). Trio knock-out mice are embryonic lethal with abnormal hippocampus and olfactory bulb organization along with defects in skeletal muscles (O'Brien et al, 2000).

Kalirin was originally identified as a peptidylglycine α-amidating monooxygenase

(PAM) interacting protein. PAM converts peptide substrates such as neuropeptide Y into a biologically active α-amidated form (Alam et al, 1997). Kalirin isoforms are also widely expressed although the expression of Kalirin isoforms in adults is limited to the central nervous system (Hansel et al, 2001). Kalirin knock-out mice are viable, with morphological changes in cortical pyramidal neurons dendrites and behavioral defects

8 such as an age-dependent progressive impairment in working memory and reduced social approach (Cahill et al, 2009; Xie et al, 2010).

The RhoG/Rac1-specific GEF1 domain of the Trio family GEFs regulates cell migration, axon guidance, dendrite morphology, and NMDA receptor dependent synaptic plasticity by manipulating the cytoskeleton through the Rac GTPase pathway (Bellanger et al, 2000; Blangy et al, 2000; Kiraly et al, 2011; Lemtiri-Chlieh et al, 2011; Mandela &

Ma, 2012; Nicholson et al, 2012; van Rijssel et al, 2012). For example Kalirin-7, which contains the RacGEF domain alone, is enriched in the postsynaptic densities (PSD) in dendritic spines (Ma et al, 2003; Penzes et al, 2000), and was recently reported to interact with the NMDA glutamate receptor to regulate postsynaptic function and synaptic strength in events related to memory and learning (Kiraly et al, 2011).

Trio and Kalirin also function in the secretory pathway. PAM, the interaction partner of Kalirin is a neuropeptide processing enzyme that functions in DCVs (Alam et al, 1997; Mains et al, 1999). Kalirin and Trio modulate DCV maturation in pituitary derived AtT-20 neuroendocrine cells and their isoforms are differentially associated with

Golgi, immature DCVs, and endosomes (Ferraro et al, 2007). Increasing Kalirin or Trio activity depletes immature DCVs of their hormone cargo, while Kalirin or Trio inhibition increases the amount of mature hormone product in mature DCVs.

This role for Kalirin and Trio in the secretory pathway is mediated by their

RhoGEF1 domains, suggesting the UNC-73 RhoGEF1 domain may have a role in the secretory pathway. However, the function of the Trio family Rho specific RhoGEF2 domains has not been as well studied. The Trio RhoGEF2 domain directly interacts with

Gαq and RhoA (Rojas et al, 2007). Trio/RhoA regulates apical constriction in embryonic

9 development to establish cell polarization through ROCK (Plageman et al, 2011). Trio and RhoA are upregulated in Echo30 virus infected cells and activation of the RhoA pathway leads to neuronal cell death, likely contributing to aseptic meningitis and encephalitis, however, the specific mechanisms are not known (Lee et al, 2012). UNC-

73/RHO-1 regulates neurotransmission downstream of Gαq in parallel to PLCβ (Williams et al, 2007). UNC-73-RhoA signaling also contributes to innate immune responses in C. elegans through a MAP kinase pathway (McMullan et al, 2012).

p63RhoGEF is a Rho-specific GEF with only one GEF domain (Lutz et al, 2004). p63RhoGEF is not an obvious candidate as a member of the Trio GEF family as it lacks of protein-protein interaction domains other than the RhoGEF and PH domains. However, the p63RhoGEF C-terminal PH domain region is similar to the equivalent regions in the

Trio family RhoGEFs (Rojas et al, 2007), and the UNC-73E isoform, which has similarity to p63RhoGEF, has a similar ability to activate RHO-1 in vivo in C. elegans

(Williams et al, 2007). Therefore, p63RhoGEF could be a mammalian UNC-73

RhoGEF2 isoform homolog outside of the conventional Trio family.

1.4 The model organism Caenorhabditis elegans

C. elegans is a free living, transparent nematode introduced as a model organism in 1965 by Sydney Brenner. C. elegans has a rapid three day life cycle, is easy to grow under standard lab conditions, and is of small size with the adult worm about 1 mm in length. They exist as both hermaphrodites and males and reproduce by self-fertilization or mating. C. elegans is diploid so it is possible to study lethal mutations and introduce balancers and genetic markers (Riddle, 1997b). In 1998, the C. elegans genome was fully

10 sequenced and the fully annotated 100MB genome sequence is available online (Hodgkin,

2005). These characteristics make C. elegans a powerful tool in genetic and genomic studies. Many large scale genetic screens using classical genetics and RNAi techniques have been performed using C. elegans to identify new genes, phenotypes, and interacting proteins (Kamath & Ahringer, 2003; Scholer et al, 2012; Sieburth et al, 2005). The transparency of the nematode makes imaging the whole live organism possible and C. elegans research was instrumental to the development of green fluorescent protein (GFP) and confocal microscopy (Chalfie et al, 1994; White et al, 1987).

The adult hermaphrodite has a 302-cell nervous system and a completely mapped cell lineage from zygote to adult worm. The hard wired synaptic connections have been reconstructed based on the physical attachment and structural characters of neurons examined by electron microscopy (Riddle, 1997b). There are about 5000 chemical synapses, 2000 neuromuscular junctions, and 600 gap junctions (White et al, 1986). How this “wiring diagram” is modulated by neuropeptides is just beginning to be revealed

(Brezina, 2010). These fundamental studies provide the foundation for the use of C. elegans as a model organism in the study of neuroscience.

1.5 C. elegans locomotory regulation

The movement of the worm is regulated by the coordination of the nervous system and the body wall muscles. C. elegans neurons can be classified by function into three groups: receptor neurons (also known as sensory neurons), interneurons, and motoneurons. Between the motoneuron and body wall muscle, the dyadic neuromuscular junction connects the two systems with a chemical synapse (White et al, 1986).

Each functional group of neurons expresses specific molecules, the promoters of which can be used in cell-specific gene expression experiments. For instance, neuropeptides are produced in almost all neuronal types. The EGL-3/proprotein convertase is required for FMRFamide-like family neuropeptide production, and the egl-

3 promoter is expressed in most neurons (Husson et al, 2006). The sensory neurons have a sensilla structure to accept a signal either from the external chemical environment or from mechanical stress. For example, the osm-5 gene encodes a protein required for cilia formation, and osm-5 is expressed exclusively in sensory neurons (Haycraft et al, 2001).

Interneurons mediate information flow from sensory neurons to motoneurons. They innervate motoneurons to form the motor circuitry and direct locomotory behavior. For instance, forward movement is mediated by interneurons AVB and PVC, which innervate

DB and VB motoneurons, while backward movement is mediated by interneurons AVA,

AVD, and AVE, which innervate DA and VA motoneurons (Riddle, 1997a). In morphology and the utilization of neurotransmitters, the interneurons are more diverse than sensory or motoneurons. For example, the promoters for several genes were used in this thesis including tph-1, which is expressed in serotoninergic neurons (Sze et al, 2000), dat-1 in dopaminergic neurons (Nass et al, 2002), eat-4 in glutamatergic neurons (Lee et

12 al, 1999), and glr-1 which encodes the AMPA type glutamate receptor subunit that is expressed in the command interneurons (Zheng et al, 1999). The motoneurons of the ventral nerve cord innervate the ventral and dorsal body wall muscles in the main part of the worm body and form a feedback circuit to generate coordinated locomotion.

Motoneurons are classified into cholinergic and GABAergic by the type of neurotransmitter they release. The former expresses unc-17, which encodes an acetylcholine vesicular transporter (Alfonso et al, 1993), and the latter expresses unc-47, which encodes a GABA vesicular transporter (Eastman et al, 1999).

The mutants of these genes present various locomotory defects. Mutation of chemosensory neuron specific genes affecting the growth of the amphid and phasmid axons result in neuronal development defects, and cause the locomotory defects as well

(Hedgecock et al, 1985). The mechanosensory neurons also influence locomotory behavior and have been well studied (Goodman, 2006). The multi-dendritic sensory neurons PVD and FLP sense mechanical stimuli throughout the worm body, and defects in these neurons result in decreased movement and defective posture (Albeg et al, 2011;

Cohen et al, 2012). Animals with defects in the motoneurons have significantly impaired locomotory phenotypes. The unc-17 and unc-47 mutants have neurotransmitter filling defects in the motoneurons, and these two genes were named “uncoordinated” because of the movement phenotypes they display. As a result of reduced acetylcholine release at the neuromuscular junction, unc-17 mutants present a severe coiled posture on plates and are resistant to the acetylcholinesterase inhibitor aldicarb (Roghani et al, 1994; Varoqui et al,

1994). However, mutants with defective interneurons have more subtle locomotory defects. TPH-1 and DAT-1 affect locomotion by changing the response to the “food

13 signal” that is produced when animals are exposed to bacteria (Sawin et al, 2000).

Overall, tph-1 and dat-1 mutants with defects in dopamine or serotonin signaling, respectively, move in a relatively coordinated manner on plates (Nass et al, 2002; Sze et al, 2000). The glr-1 mutants have a defective escape response and a higher rate of spontaneous reversals than wildtype (Zheng et al, 1999). The fine-tuning of locomotory behavior therefore requires the input of multiple different types of neurons.

1.6 Neurotransmission

Neurotransmission at chemical synapses requires multiple steps including presynaptic exocytosis, neurotransmitter removal and recycling, and postsynaptic responses. Classical small molecule neurotransmitters such as acetylcholine (ACh) and

GABA, and neuromodulatory neurotransmitters such as neuropeptides undergo different mechanisms in neurotransmission (Sudhof & Malenka, 2008). Classical neurotransmitters are loaded and packed locally at synapses in small clear synaptic vesicles (SVs) and are released from presynaptic neurons into the synaptic cleft in several steps: recruitment, docking, priming and fusion (Sudhof & Rizo, 2011). Larger molecules such as neuropeptides and also monoamines are sorted and packed in large dense core vesicles

(DCVs) in the cell bodies and are transported down to synapses through the axons. The release of DCVs is not exclusively within the synaptic active zone as it is for SVs

(Husson et al, 2007).

Exocytosis is accomplished by conserved SNARE-mediated membrane fusion in general, but distinct molecules are required in the release mechanism for each vesicle type, although there is some controversy with respect to these studies. UNC-13/Munc13

14 is a DAG binding protein specifically required for SV fusion to the plasma membrane

(Aravamudan et al, 1999; Augustin et al, 1999; Richmond et al, 1999). UNC-31/CAPS is also a DAG binding protein with sequence homology to UNC-13, which appears to function exclusively in DCV release in C. elegans, but not in SV exocytosis (Speese et al,

2007). Both UNC-13 and UNC-31serve vesicle docking and priming functions (Lin et al,

2010; Richmond et al, 1999), and the two pathways converge on syntaxin, a SNARE component (Hammarlund et al, 2008). Although C. elegans UNC-31 appears to function exclusively in DCV release, the mammalian homolog CAPS appears to have an essential function in SV priming (Jockusch et al, 2007). Also, UNC-13 may not be restricted to SV functions since C. elegans unc-13 mutants have defects in DCV exocytosis as revealed by a neuropeptide assay (Sieburth et al, 2007). How the two pathways of SV- and DCV- mediated signaling intertwine in the mechanisms of neurotransmission must still be explored.

Classical small neurotransmitters are focused on the direct and immediate communication between neurons. For instance, acetylcholine is the major neurotransmitter at excitatory synapses while GABA mainly functions at inhibitory synapses at neuromuscular junctions in C. elegans (Jorgensen, 2005; Rand, 2007).

However, the components in DCVs, including monoamines and neuropeptides, mainly have a regulatory function in the fine-tuning of neurotransmission. For example, multiple neuropeptide genes were identified as modulators of C. elegans cholinergic signaling in an RNAi-based screen, including flp-1, nlp-12, ins-22 and ins-31 (Sieburth et al, 2005).

Neuropeptides encoded by nlp-3 are required to regulate ASH neuron-mediated aversive behavior in C. elegans (Harris et al, 2010). In mammals, the neuropeptides FF and VF

15 diminish the inhibitory effect of GABAergic neurons in the central nervous system

(Jhamandas et al, 2007). Neuropeptide Y and its receptors have multiple functions including regulation of ethanol tolerance (Thiele & Badia-Elder, 2003)

1.7 Heterotrimeric G protein pathways and neurotransmission

The network of heterotrimeric G protein signaling pathways regulating neuronal activity in C. elegans motoneurons has been extensively studied (Perez-Mansilla &

Nurrish, 2009).Presynaptically, Gαq/EGL-30 stimulates acetylcholine (ACh) release through an increase in the production of diacylglycerol (DAG) at synapses, which binds to and recruits UNC-13 to the synaptic vesicle release machinery (Lackner et al, 1999).

Gαq/EGL-30 activates phospholipase Cβ (PLCβ)/EGL-8, which hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) into inositol 1,4,5-triphosphate (IP3) and

DAG (Smrcka & Sternweis, 1993). UNC-73 RhoGEF2 activity functions downstream of

Gαo (Steven et al, 2005) and the UNC-73E isoform is a direct downstream factor of Gαq in parallel to EGL-8/ PLCβ (Williams et al, 2007). The relationship between UNC-73

RhoGEF activity and the heterotrimeric G protein signaling pathways is further explored in this study. The third branch of the heterotrimeric G protein signaling network regulating cholinergic neurotransmission in C. elegans is the Gαs/GSA-1 pathway.

Adenylyl cyclase/ACY-1 is the major effector of GSA-1 in growth and locomotory regulation in C. elegans and it catalyzes the conversion of ATP to cyclic AMP (cAMP).

Interestingly, DCV docking defects due to unc-31 mutations are rescued by PKA activation, PKA being a downstream kinase in the Gαs pathway (Zhou et al, 2007). The

16 precise nature of the interaction between the Gαs pathway and DCV release has not been defined.

1.8 RhoA/RHO-1 Signaling in neurotransmission

A recent review highlighted the role of RhoA in neuronal development and RhoA as a downstream factor in response to receptor activation (Tolias et al, 2011), but only a small number of studies suggest that RhoA/RHO-1 signaling affects neurotransmission.

In C. elegans, RHO-1 functions early in development as rho-1 null mutations are homozygous lethal (McMullan & Nurrish, 2011). However, the expression of RHO-1 is abundant in the adult nerve ring and anterior sensory organs, indicating RHO-1 likely also functions in the adult nervous system (Chen & Lim, 1994). The expression of constitutively activated RHO-1 in the cholinergic motoneurons causes a hyperactive locomotion rate and increased acetylcholine (ACh) release (McMullan et al, 2006). On the other hand, the expression of C3 transferase, a RHO-1 inhibitor, in the cholinergic motoneurons causes corresponding decreases in locomotion rate. Mammalian RhoA and

C. elegans RHO-1 bind and inhibit the enzyme diacylglycerol kinase (DGK) (Houssa et al, 1999; McMullan et al, 2006). RHO-1 inhibition of DGK-1 activity results in excess

DAG and upregulated recruitment of UNC-13 at synapses causing increased ACh release

(McMullan et al, 2006). The Gα12 pathway acting through the RhoGEF RHGF-1 functions upstream of RHO-1 in the regulation of ACh release (Hiley et al, 2006). These studies suggest activated RHO-1 plays a presynaptic role in classical neurotransmitter release and the experiments in this thesis further examine the role of the Rho signaling pathway in the regulation of neurotransmission.

In mammals, Rac1 is the only Rho family GTPase to associate with purified synaptic vesicles, although other family members including RhoA are present in rat brain synaptosomes (Doussau et al, 2000). If RhoA plays a regulatory role in synaptic vesicle release, it is unlikely to function by directly interacting with synaptic vesicles.

Normal neuronal function relies on endocytosis events such as presynaptic vesicle recycling and postsynaptic receptor dynamics. Indeed, RhoA is involved in both processes. Synaptic vesicle recycling is decreased in Oligophrenin1 (a RhoA-specific

GAP) null mice, indicating RhoA signaling down regulates synaptic vesicle recycling

(Khelfaoui et al, 2009). As shown in the same study, AMPA receptor endocytosis is also down regulated in neurons from Oligophrenin1 null mice and results in reduced long term depression (LTD). The link between RhoA signaling and LTD suggests that RhoA signaling may play a role in long term depression that relates to learning and memory functions. Importantly, defects in the RhoA signaling pathway can lead to human nervous system diseases such as mental retardation (Boda et al, 2010).

The RhoA signaling pathway also regulates neuroendocrine secretion

(Momboisse et al, 2011). It is proposed that activated RhoA associates with secretory granules to stabilize the surrounding actin cytoskeleton network to inhibit exocytosis of secretory granules in chromaffin cells (Gasman et al, 1998). However, RhoA is required for calcium-dependent secretory granule exocytosis from mast cells as observed using constitutively activated V14RhoA and the RhoA inhibitor C3 transferase in a separate study (Sullivan et al, 1999). Opposite to the action of RhoA in chromaffin cells, Cdc42 potentiates exocytosis and is required to assemble actin structure associated with the exocytotic machinery in chromaffin cells (Gasman et al, 1999). These studies link RhoA

18 to the regulation of secretory granule release, which may share a similar mechanism to that of DCV exocytosis in neurons.

1.9 Overall objectives:

The objectives of this research are to examine how UNC-73 RhoGEF2 isoforms regulate nervous system function. The specific aims of these studies are:

(1) Examine the relationship between UNC-73 RhoGEF2 activity and classical neurotransmitter release.

(2) Examine the relationship between UNC-73 RhoGEF2 activity and peptidergic neurotransmission.

(3) Characterize the interaction between UNC -73 RhoGEF2 and the heterotrimeric G protein signaling pathways.

Chapter 2

2 Materials And Methods

2.1 Worm strains

Strains were maintained at 21°C on plates containing standard nematode growth media (NGM) unless otherwise noted. The following strains were used in this study.

Double-mutant strains were constructed using standard genetic methods without additional marker mutations. Mutations in the double mutants were confirmed by PCR,

DNA sequencing, or outcrossing to him-5 males and examination of the F2 progeny phenotypes.

Strains used in this study are N2 Bristol (wild type), XA7314 unc-73(ev802) I, qaIs7312[unc-73D1; F25B3.3p::gfp] V, NL3231 acy-1(pk484) III, dpy-20(e1362) IV, pkIs296[hsp::gsa-1(QL) dpy-20(+)] X, KG518 acy-1(ce2), VC671 egl-3(ok979),

RM2221 egl-8 (md1971), egl-30(tg26), KG421 gsa-1(ce81), juIs1[unc-25p::snb-1::gfp],

KG532 kin-2(ce179), nuIs24[glr-1p::glr-1::gfp], nuIs183[unc-129p::nlp-21::YFP, myo-

2p::NLS::gfp] IV, QT47 nzIs1[HSp::rho-1(gf);ttx-3p::gfp] I , MT1093 rab-2(n501),

XA7314 unc-73(ev802) I, qaIs7312[unc-73D1; F25B3.3p::gfp] V, XA7330 qaEx7327

[unc-73D1; unc-73E; F25B3.3p::gfp], unc-73(ev802), CB928 unc-31(e928), XA7300 unc-73(ev802)/unc-11(e47) dpy-5 (e61), KG1278 unc-73(ce362), CB933 unc-17(e245),

GS1912 arIs37[pmyo-3::ssGFP, dpy-20(+)];dpy-20(e1282). The KG strains were a gift from Kenneth Miller (Oklahoma Medical Research Foundation), the nu alleles from

Joshua Kaplan (Harvard University), and QT strains from Stephen Nurrish (University

College London).

To detect DNA mutations, single worm lysate is obtained by lysing single adult worm in worm lysis buffer (50 mM KCl, 10 mM Tris pH 8.3, 2.5 mM MgCl2, 0.45% NP-

40, 0.45% Tween-20, 0.01% Gelatin, 100μg/ml proteinase K), by heating at 60°C for 1 hour, then increase to 95°C for 15 minutes to inactive proteinase K. 1μl of single worm lysate is used in each 10μl Single worm PCR (SWPCR) reaction, which reaction is scaled down from the commercial PCR kit instruction using HotMaster Taq (5 PRIME) or Taq

DNA Polymerase (NEB). The primers used to detect double mutants used in this study are listed as table 1.

Table 1. Single worm PCR primer pairs used to detect DNA mutations in animals by PCR genotyping or DNA sequencing. “WT primers” hybridize to the wild-type DNA sequence, while “Mutant primers” hybridize to the altered DNA sequence specific to that particular allele.

Allele WT primer Mutant primer Notes unc-73(ce362) RS62/RS424 RS37/RS421 unc-73(ev802) RS62/RS35 RS160/RS31 rho-1 RS852/RS853 RS856/RS857 egl-3(ok979) RS539/RS540 RS539/RS531 53°C annealing temp gsa-1(ce81) RS283/RS291 RS283/RS289 acy-1(ce2) RS298/RS962 RS298/RS963 no more than 30 cycles acy-1(pk484) RS335/RS336 RS337/RS336 Mutant reaction use 59°C annealing temp, 1.5min egl-8(n488) RS640/RS641 RS645/RS706 extension time, yield 1.3kb product dgk-1(ok1462) RS918/RS919 RS920/RS921 dgk-1(nu62) RS961/RS245 RS243/RS958 tom-1(ok285) RS922/RS923 RS924/RS925 rab-2(n501) RS938/RS940 RS939/RS940 unc-31(e928) UT38/39 UT38/40 Allele Sequencing primers Notes 50°C annealing temp, yield kin-2(ce179) RS287/RS288 500bp product acy-1(ce2) RS286/RS298 500bp product dgk-1(sy428) RS820/RS821 dgk-1(nu62) RS820/RS821 gsa-1(Q208L) RS283/RS315 for cDNA GFP arIS37 UT33/UT35 for GFP construct

2.2 Transgenic lines

Standard microinjection techniques were used to generate stable transgenic C. elegans lines carrying extrachromosomal DNA arrays. For the unc-73(ev802) rescue experiments, between 50 and 100 ng/ml of the gene-Xp::unc-73E::gfp constructs were mixed with 50–100 ng/ml of pXS2 encoding unc-73D1 and injected into unc-

73(ev802)/unc-11 dpy-5 without a cotransformation marker. pXS2 was added to rescue unc-73(ev802) lethality. Progeny were screened for stable expression of the extrachromosomal array and homozygous unc-73(ev802) lines containing the array were established. For unc-73(ce362) rescue experiments, transgenic lines were first established in a wildtype background, by injection of 100 ng/ml gene-Xp::unc-73E::gfp without a cotransformation marker, and then the lines were crossed into unc-73(ce362). Exceptions were the dat-1p and eat-4p constructs, which were injected into wild type at 50 ng/ml along with 50 ng/ml unc-122p::gfp (coelomocyte-specific promoter) as a cotransformation marker. PCR-based genotyping was used to confirm unc-73(ev802) and unc-73 (ce362) rescued lines were homozygous. Gain-of-function cDNA (P260S) constructs [KG#81 myo-3p::acy-1(gf) and KG#83 rab-3p::acy-1(gf)] (Reynolds et al,

2005; Schade et al, 2005) at 10 ng/ml with 50 ng/ml unc-122p::gfp and 90 ng/ml herring sperm DNA were injected into N2 to establish stable transgenic lines that were crossed into unc-73(ce362). The synapse marker KP#282 acr-2p::snb-1::cfp (Nurrish et al, 1999) at 50 ng/ml mixed with 50 ng/ml pRF4 rol-6 (su1006dm) and 50 ng/ml herring sperm

DNA were injected into N2 and unc-73(ce362) animals to establish stable lines. The plasmid KP#315 acr-2p::egl-30(Q209L) (Lackner et al, 1999) at 10 ng/ml mixed with 50 ng/ml F25B3.3p::gfp (a cotransformation marker expressed in neurons) and 70 ng/ml

23 herring sperm DNA were injected into N2 and unc-73(ce362) to establish stable transgenic lines. At least three transgenic lines were isolated for each injected plasmid mix and it was confirmed that each set of lines had the same phenotype and/or GFP expression pattern.

2.3 DNA constructs

2.3.1 DNA constructs in C. elegans

The gene-Xp::unc-73E::gfp cell-specific promoter constructs were made by directionally cloning promoter DNA into the unique NotI (5’) and EcoRV or EcoRI (3’) sites of the unc-73E::gfp plasmid pXS6. In this construct the unc-73E transcript is encoded by cDNA for the first two exons with the remainder encoded by a genomic DNA fragment extending 150 bp beyond the 39 end of the coding region of the transcript.

Promoters generated by PCR using wildtype genomic DNA or the plasmids pPD49.83

(heat-shock promoter; from Andy Fire, Stanford University, Palo Alto, CA), pPD136.64

(myo-3 promoter; from Andy Fire), or pRM621 (modified unc-17 promoter as described in Charlie et al. 2006; from Jim Rand, Oklahoma Medical Research Foundation) as the template were cloned into pJET (Fermentas) or pGEM (Promega) PCR cloning vectors and confirmed by sequencing. We used the same promoter regions previously defined in mutant rescue experiments. Two promoters were not generated by PCR, but directly cloned from plasmids. pBY103 (from Dave Pilgrim, University of Alberta, AB, Canada) contains the 1.2-kb HindIII/EcoRI unc-119 promoter fragment and pJL35 (from Dr.

Bruce Bamber, University of Toledo) contains the 1.2-kb unc-47 promoter fragment.

The pan-neuronal constitutively activated GSA-1(CA) construct rab-3p::gsa-1(gf) was made by replacing the heat-shock promoter in plasmid KG341 (Dr. Kenneth Miller,

Oklahoma Medical Research Foundation) with the rab-3 promoter by adding

StuI/BamHI sites. The KG341 plasmid was cut with HindIII and filled in with Klenow fragment, then cut with BamHI to ligate with the StuI/BamHI digested rab-3 promoter from rab-3p::pJET. The pan-neuronal gain of function rho-1 construct rab-3p::rho-1(gf) was generated by cutting the rho-1(gf) cDNA from unc-73Ep::rho-1(gf) plasmid (Dr.

Robert Steven, University of Toledo) and ligating into an empty pPD49.26 vector with

NheI/SacI sites. The rab-3 promoter was produced by PCR from genomic DNA with primers RS892/RS893 containing PstI/BamHI sites and cloned into pJET. The promoter was cut from this clone and ligated into the newly constructed rho-1(gf)pPD49.26 plasmid. The lin-2p::UNC-73E::GFP was generated by fusion PCR to fuse the 5.1kb lin-

2 promoter PCR product with UNC-73E from EGmNP plasmid to generate a 9.2kb product. Three independent fusion PCR products are mixed for injection.

The PCR primers and the templates of the promoter regions are listed in table 2.

Table 2. PCR primers used to amplify cell-specific promoter regions.

Promoter Primers Template Size Ref RS565 atat GCGGCCGC tatcaatcatttcagaaatttgtg egl-3p Genomic DNA 4kb (Kass et al, 2001) RS566 ATCG GATATC accttcaattgaaaaatatttattg RS546 tcaaacataagcatcggatcac lin-2p Genomic DNA 5.1kb (Hoskins et al, 1996) RS861 ATCAGATTAGCACACGGCATtctgttttagctgaaaaatgct RS438 gCA ggC ggC CgC TTC AAT Cgg TCg ACC gAA CAg flp-22p Genomic DNA 2.7kb (Kim & Li, 2004) RS439 gCTg gATATC TT TTT gTg TAT ATC CTg AAA T RS436 gCA ggC ggC CgC TgT AgT ACg TgA CTG TAg CCC flp-4p Genomic DNA 3.1kb (Kim & Li, 2004) RS437 CTg CgA TATC TgT CgC TgT AgT TgT CTg RS434 gAC ggC ggC CgC TTg CAA TAC TAA TAg ggA gTg C tph-1p Genomic DNA 3.1kb (Sze et al, 2000) RS460 CGT CCA GCT GAT GAT TGA AGA GAG CAA TGC TAC C RS683 atat GCGGCCGC cccggaacagtcgaaagtc unc-47p pJL35 1.2kb (Eastman et al, 1999) RS684 atcg GAATTC taatgaaataaatgtgacgctgt RS749 GAAT gcggccgc aaatgaatttgtttattgaataaatac glr-1p Genomic DNA 700bp (Zheng et al, 1999) RS750 ACTG GATATC tgtgaatgtgtcagattgggt UT10 GCGGCCGCgatctcatttgttcaatcaatca osm-5p Genomic DNA 240bp (Haycraft et al, 2001) UT11 GAATTCtaagaaaagtgttctcagaagaa RS458 GCA GGC GGC CGC TCC ATG AAA TGG AAC TTG AAT C dat-1p Genomic DNA 700bp (Nass et al, 2002) RS433 gCAggAATTCggCTAAAAATTgTTgAgATTCgAgTAAACC RS594 GATT GcggCCgc GGATAATAAGGGATAGAGACAG C1p Genomic DNA 1kb (Steven et al, 2005) RS646 AGCT gaattc TTTAATGATGGAATATATATAGGAC

2.3.2 Construction of plasmids for protein expression in E. coli

Proteins used in the pull-down assay were generated in E. coli. The C. elegans genes were cloned from a cDNA pool then inserted into protein expression vectors either by restriction enzyme cut/ligation (Fisher Fermentas) or by Gateway system (Invitrogen). pDEST17 for N-terminus His tag, pDEST15 for N-terminus GST tag.

His::LIN-2 in pET-30a and GST::UNC-73 in pGEX-4T1 plasmids were made by

John Farver (Farver, 2009). DYS-1B cDNA in pDEST-17 plasmid was made by Kyoung

Jo. NUF (F55C12.1) was cloned with UT114/UT115 from C. elegans cDNA library, put in pDONR221 entry clone vector, then recombined into pDEST15 and pDEST17 destination vectors. Clone cDNA of full-length UNC-73C1, D1, and E isoforms for

Gateway system. The primers used in cloning are listed as below.

Table 3. PCR primers used to amplify cDNAs for protein expression.

Gene name Primers Size UT114 GGGGACAAGTTTGTACAAAAAAGCAGGCTGG ATGACCGACGTGTATGAGTCG NUF 1.4kb UT115 GGGGACCACTTTGTACAAGAAAGCTGGGTG TTACCCGTTAACGGGATCTATCT UT162 GGGGACAAGTTTGTACAAAAAAGCAGGCTGG atggtgataaaatgtttcacttgc UNC-73C1 2.8kb UT163 GGGGACCACTTTGTACAAGAAAGCTGGGTG tcacatttccgttccatcaaaac UT164 GGGGACAAGTTTGTACAAAAAAGCAGGCTGG atgcgacggtctcaaactgtg UNC-73D1 2.8kb UT163 GGGGACCACTTTGTACAAGAAAGCTGGGTG tcacatttccgttccatcaaaac UT165 GGGGACAAGTTTGTACAAAAAAGCAGGCTGG atgccgtgtgctaatctgatg UNC-73E 1.7kb UT166 GGGGACCACTTTGTACAAGAAAGCTGGGTG tcattttccaaccggtggcg

2.3.3 Construction of plasmids for protein expression in HEK293T cells

The DYS, NUF, UNC-73C1 and D1 cDNA were also recombined into mammalian cell expression vectors by using Gateway system. The mammalian expression vectors pDEST26 were used to generate His tagged protein, pECE for GFP tagged protein, and pDCG-XhoI were used to generate GST tagged protein. The mammalian expression vectors were gifts from Dr. Song-Tao Liu’s lab (University of

Toledo).

2.4 Live animal assays

2.4.1 Locomotion assays

Rates of locomotion were determined by body bends assay at room temperature as described previously (Steven et al, 2005). Healthy and well fed young adult animals were transferred to NGM assay plates without bacteria and left for at least 20 sec before counting. Locomotion rate is defined as the number of body bends exhibited in 20 sec of uninterrupted forward movement. If an animal stopped moving or reversed direction, the count was abandoned. One body bend was defined as a complete cycle of terminal bulb motion starting from the top position of the sinusoidal wave track through to the bottom and back to the top. Between 25 and 30 animals were examined for each strain.

2.4.2 Reversal assay

Healthy and well fed young adult animals were transferred to NGM assay plates without bacteria and left for at least 1 minute before counting. Spontaneous reversal is

29 defined as the number of direction changes from forward to backward on the plates without additional stimulus in 3 minutes (Hart (ed.), 2006). Between 25 and 30 animals were examined for each strain.

2.4.3 Drug sensitivity assays

The aldicarb sensitivity protocol was modified from previously described protocol

(Mahoney et al, 2006). Larval stage four (L4) worms were picked the day before the assay was performed and left to grow overnight on plates with nematode growth media

(NGM) and bacteria. The next day the young adults were transferred to NGM plates containing 1.0 mM aldicarb (Chem Services). Plates were stored at 4°C for 1 to 5 days and seeded with bacteria the day of use. Approximately 30 animals per strain were examined blind for paralysis every 10 min for 2 hr, using a Leica MZ6 stereomicroscope.

Animals were defined as paralyzed if after a maximum of three worm pick taps on the head and three taps on the tail there was no body movement; however, animals exhibiting only small foraging movements of the head or small movements of the tail were still considered paralyzed. The assay was repeated three times for each strain and an average for each time point was calculated.

Levamisole sensitivity assays were performed as described above for aldicarb sensitivity. Staged young adult worms were examined on seeded NGM plates containing

0.5 mM levamisole (Sigma, St. Louis) over a period of 1 hr. Plates were prepared as described above for aldicarb and at least three trials were performed for each strain.

2.4.4 PMA treatment

NGM plates containing the phorbol ester (phorbol 12- myristate 13 acetate, PMA)

(Enzo Life Sciences) were made as previously described (Reynolds et al, 2005). PMA was prepared in 100% EtOH to make a 10mM stock solution. Then 20μl of PMA stock or

20μl of EtOH was added to 40ml of pre-autoclaved NGM medium to make either NGM plates with 5μM PMA or control plates with EtOH carrier alone, respectively.

Approximately 15–20 adult worms for each strain were placed on separate PMA or control (ethanol carrier only) plates at room temperature and videotaped after 2.5 hours.

PMA treatment of the lethargic mutants increased the locomotion rates, but they tended to coil and reverse frequently, which made it impossible to measure the locomotion rates in our body bend assays. Video images were captured with a Dino-Lite AM413T digital microscope in a “Worm Tracker” rig and processed using Worm Tracker 2.0 software

(Schafer Lab).

2.4.5 Coelomocyte endocytosis assay

Coelomocyte endocytosis in C. elegans strains was examined by monitoring

Texas Red-conjugated bovine serum albumin (TR-BSA) (Invitrogen) uptake by coelomocytes (Zhang et al, 2001). Synchronized young adults were injected with a short pulse of 1 mg/ml TR-BSA into the pseudocoelom in the region of the pharynx between the metacorpus and the terminal bulb. Animals were left to recover on seeded NGM plates for 1 hr and then collected in a drop of ice-cold 1% paraformaldehyde (Electron

Microscopy Sciences) in M9 on a slide with a 2% agarose pad. They were maintained in

31 the drop for at least 20 min and up to 2 hr, on ice, as all the injected worms were collected, and then covered with a cover slip and observed under the microscope.

2.4.6 Body length measurement

Larval stage four (L4) worms were picked and imaged live on Olympus dissecting scope. The imaged worms were collected and cultured on the standard NGM plates. After

12, 24, and 36 hours, the same batch of worms were imaged again under the same conditions. Worm lengths were measured by drawing a curved line along the midline of the worm shadow from the head to the tail, tip to tip. At least 25 worms from each strain were measured.

2.5 Cell specific RNA interference

Cell-specific RNAi constructs were made by fusion PCR as previously described

(Esposito et al, 2007). Promoter regions were amplified from either C. elegans genomic

DNA or a promoter-containing plasmid using a high fidelity PCR kit, such as iProof

(Bio-Rad). Approximately 500bp of the targeted gene was amplified from a cDNA library pool. The PCR for the promoter and gene fragments were repeated in triplicate and tested for size and purity by gel electrophoresis before mixing to use as the template for the fusion PCR. The fusion PCR was also performed in triplicate using the PCR

Extender System (5 Prime). A mixture of sense and anti-sense fusion PCR product at equal concentrations was microinjected along with the pan-neuronal GFP cotransformation marker F25B3.3p::gfp into N2 animals, using standard microinjection

32 techniques. The primers used to generate the specific target gene fragments for fusion

PCR are listed in table 4.

Table 4. PCR primers used to amplify the target gene fragments for cell-specific RNAi.

Gene Name Primers RS863 TTCAGAGAATAACACGCTATCG unc-73 PH1 RS864 TACTCCAAGACCCAATGGC RS624 TGACATGATGCAAGTCGGTCGT unc-73 PH2 RS625 CTCTCCTGTCACCTCCAGGG RS661 TGGCTGCGATTAGAAAGAAG rho-1 RS662 CCTCACGAATTCCGTCCTTA RS655 GCGAGCAGAAGCGAATAAAT egl-30 RS656 ACACGAACACGCAGAATGTC RS673 ATTAACTCTGGATTTTCGGATCAA unc-31 RS674 TCCTTGTGTAATTAACCAATGCAG RS649 TGGAAACGATCCAAGTCTCC myo-3 RS650 CGATGAAATGCTTACGCTCA

2.6 Immunoblotting

SDS-PAGE was typically run at 100v for 100 minutes in a Bio-Rad cartridge with running buffer (15.1g Tris, 72g Glycine, 50ml 10% SDS, add H2O up to 5L). Gels were transferred to a PVDF membrane (Millipore) in the Western buffer (19g Tris, 90g

Glycine, 62.5ml 10% SDS, 1L Methanol, add H2O up to 5L) with the transfer unit at

400mA for 100 minutes. The transferred membrane was blocked in blocking buffer (5% non-fat milk, 1% Tween-20 in 1x TBS) for 1 hour at room temperature. The primary antibody (Santa Cruz) was applied (SC803 for anti-His at 1:600, or SC-459 for anti-GST at 1:2000) by rocking overnight at 4°C. The membrane was washed with 1x TBST for 15 minutes, four times. The HRP conjugated secondary antibody goat-anti-rabbit (Pierce) was applied (at 1:1000 for anti-His, or at 1:10,000 for anti-GST) rocking 1 hour at room temperature, then the membrane was washed with 1x TBST for 15 minutes, four times.

The film was developed with the Western Blotting Substrate (Pierce).

2.7 Protein preparation and pull-down assays

2.7.1 Protein preparation from E. coli

DNA constructs were transformed into the E. coli CodonPlus strain (Agilent

Technologies, gift from Dr. Song-Tao Liu’s lab) and liquid cultures incubated at 30°C to reach O.D. 0.6, then induced with 1mM IPTG either at 30°C for additional 4 hours or at

16°C overnight. Bacterial pellets were resuspended in either GST lysis buffer (50mM

Tris pH 7.5, 150mM NaCl, 5% glycerol, 0.1% Triton X-100, 1 mM TCEP), or His lysis buffer (50mM Tris pH 7.5, 300mM NaCl, 5% glycerol, 0.1% Triton X-100, 1 mM TCEP,

10mM imidizole), at a ratio of 5ml buffer per 1gm pellet with protease inhibitor cocktail set II (Calbiochem) and lysozyme to a final concentration of 0.1mg/ml. The lysates were sonicated on ice with 8 times of 15 seconds at 13% amplitude with 15 seconds pauses.

The supernatant of the whole cell lysate were obtained after two centrifuges at 14,000g at

4°C and supernatant was removed and transferred to a new tube.

2.7.2 Protein preparation from HEK293T cells

HEK293T cells (gift from Dr. Fan Dong’s lab) were cultured under standard conditions in total medium (mixture of 500ml DMEM, 5ml 200mM Glutamine, 2.5ml

5000U/ml Pen/Strep, 50ml BCS or FBS). Tagged UNC-73C1 and DYS-1B constructs were engineered as described above. These DNA constructs were transfected with

TransIT-LT1 reagent (Mirus Bio) following the manufacturer’s instructions. Cells were collected 24 hours after transfection, washed with PBS twice, then resuspended the cell pellet by gently pipeting in cell lysis buffer (1x PBS, 10% glycerol, 0.1% NP-40, 1mM

TCEP) containing protease inhibitor cocktail set III (Calbiochem), at a ratio of 150μl cell lysis buffer per 50μl cell pellet. The cell suspension was incubated on ice for 20 minutes then centrifuged at 14,000g at 4°C and the supernatant was removed and transferred to a new tube.

2.7.3 Pull-down assay

Combine the bacterial whole cell lysates containing each of exogenous proteins or obtain the mammalian whole cell lysate containing both interested proteins at 1:1 ratio.

The cell lysates mixture was incubated and rocked for 2 hours at 4°C. Then 5 μl of

35 prepared Glutathione Sepharose (GE Healthcare) 50% slurry was applied into the mixture, incubated and rocked for an additional 2 hours at 4°C. The beads were collected by centrifugation at 500g for 5 minutes at 4°C, and washed with GST-wash buffer (50mM

Tris pH 7.5, 300mM NaCl, 5% glycerol, 1 mM TCEP) for 3 times. After the final spin the beads were drained carefully, 10μl of 2x SDS sample buffer was applied to the beads.

The samples were boiled and loaded on an SDS-PAGE gel for analysis.

2.8 Microscopy and imaging analysis

Worms were immobilized with 30 mg/ml 2,3-butanedione monoxime (BDM)

(Sigma) or 10 mM levamisole (Sigma) in M9. Images were captured with a Hamamatsu

ORCA-ER camera mounted on a Leica DMRA2 microscope and processed using

Openlab software (Improvision, Lexington, MA) or captured with a QICAM camera

(QImaging) mounted on a Zeiss Axiophot microscope and processed with Q Capture Pro

(QImaging). Confocal microscopy images were obtained with an Olympus Fluoview

300/IX70 confocal microscope. For NLP-21::YFP and TR-BSA analysis stacks of 0.4-

μm-thick optical images were captured with Olympus Fluoview 5.0 software and fluorescence intensity was quantified using Volocity software (Improvision). Images were tightly cropped to contain the cells or axons of interest and fluorescent objects within the images were identified automatically by intensity. The threshold value for this process was kept constant for each experiment and was appropriately chosen to eliminate

“background” fluorescence as observed in wildtype animals. Object intensities were quantified and summed for each image. Small objects containing less than 30 voxels

(coelomocytes), 6 voxels (cell bodies), or 4 voxels (dorsal axons) were eliminated since

36 these likely corresponded to nonspecific background specks. The arbitrary fluorescent unit of each measurement was standardized to the average fluorescent unit of wild type obtained for that day. Synchronized young adults with dorsal cord axon bundles oriented toward the objective were imaged using a UPlanApo 40x objective. The fluorescence intensity was quantified as the arbitrary fluorescent unit per unit length. The DA6 and

DB6 cell bodies in the region posterior to the vulva were imaged with a UPlanApo 100x objective in synchronized young adults with the ventral cord oriented toward the objective. Fluorescence intensity was quantified as the arbitrary fluorescent unit per two cell bodies. Single posterior coelomocytes oriented toward the objective were imaged in

L4 larvae using the UPlanApo 100x objective. Fluorescence intensity was quantified as the arbitrary fluorescent unit per coelomocyte.

Chapter 3

3 Results

3.1 UNC-73 RhoGEF2 isoforms are neuronal Rho GEFs required for locomotion

3.1.1 Cell-specific rescue of unc-73 mutant locomotory defects

Mutations affecting the UNC-73 RhoGEF2 domain-containing isoforms lower the rate of C. elegans locomotion without affecting the coordination of animal movements

(Steven et al, 2005; Williams et al, 2007). These isoforms have differential expression patterns, but they are all expressed to some extent in the nervous system (Steven et al,

2005). The small UNC-73E isoform was used as a representative RhoGEF-2 isoform in tissue-specific transgenic rescue experiments to better define where UNC-73 RhoGEF-2 activity is required for wild-type rates of locomotion. The lethargic mutants used in this study were unc-73(ce362), which has a point mutation severely reducing RhoGEF2 activity, and Is[D1]; unc-73(ev802), which carries the pharyngeal expressed unc-73D1 transgene to rescue ev802 lethality (Steven et al, 2005; Williams et al, 2007). UNC-73E expression in the nervous system (unc-119 or unc-73E promoter), but not in the body wall muscles (myo-3 promoter) rescued the Is[D1]; unc-73(ev802) and unc-73(ce362) lethargic movement phenotypes as quantified by measuring the number of body bends in a twenty second interval off food (Hu et al, 2011) (Fig. 2) . unc-73 mutant animals

38 expressing the unc-73E::gfp transgene in the nervous system moved at a rate comparable to that of wild-type N2, while the movement of unc-73 mutants expressing the same transgene in the muscles was not significantly different from that of the lethargic Is[D1]; unc-73(ev802) or unc-73(ce362) animals without the unc-73E::gfp transgene (Fig. 2).

Expression of unc-73E::gfp in a large subset of neurons using the uncharacterized lin-2 promoter also rescued the unc-73 RhoGEF2 mutant lethargy (Fig. 2 and 3). This observation is consistent with a proposed interaction between the UNC-73 RhoGEF2 isoforms and the LIN-2/CASK protein (see Appendix A ) (Farver, 2009).

Additional promoters were also used to examine the cellular requirements of

UNC-73E expression for wild-type locomotion (Table 5). UNC-73E expression in different classes of neurons including ciliated sensory, dopaminergic, GABAergic, glutamatergic, serotonergic, and cholinergic neurons did not rescue the Is[D1]; unc-

73(ev802) or unc-73(ce362) lethargic phenotypes when the constructs were injected individually or together in various combinations (Fig. 2 and 3). In contrast, UNC-73E expression in peptidergic neurons using the egl-3 promoter (Kass et al, 2001) did rescue unc-73(ce362) and Is[D1]; unc- 73(ev802) slow movement phenotypes (Fig. 2 and 3). egl-3 encodes a proprotein convertase that is copackaged in dense core vesicles with neuropeptides and is expressed in peptidergic neurons, a large subset of neurons, which includes the cholinergic motor neurons (Kass et al, 2001). Although UNC-73E expression driven by the egl-3 promoter was widespread in the nervous system (Fig. 3), these results suggested that UNC-73 RhoGEF-2 isoforms might play a role in peptidergic neurotransmission to influence locomotion rates. However, UNC-73E::GFP expression in

39 more restricted subsets of peptidergic neurons using the flp-4 and flp-22 promoters failed to rescue (Fig. 2 and 3).

The UNC-73C1 RhoGEF2 isoform rescues unc-73 RhoGEF2 mutant lethargy

(Steven et al, 2005). UNC-73C1::GFP is expressed in a limited number of neurons suggesting that UNC-73 RhoGEF2 activity is not required in all neurons for wild-type movement and select neurons may be important for Rho pathway regulation of locomotion rate (Steven et al, 2005). However, UNC-73C1::GFP does not rescue the unc-

73 lethargy, as the GFP tag may be interfering with UNC-73C1 function (Steven et al,

2005). The same unc-73C1 5’ regulatory sequence used in the unc-73C1 rescuing construct (Steven et al, 2005) was used to drive UNC-73E expression and examine if

UNC-73 RhoGEF2 activity in this limited number of neurons could rescue unc-73

RhoGEF2 mutant lethargy, but rescue was not observed (Fig. 2 and 3).

Table 5. A list of cell specific promoters used in cell-specific rescue experiments

Gene Encoded Protein Cellular Expression References Name Plasma membrane dopamine dat-1 Dopaminergic neurons (Nass et al, 2002) transporter eat-4 Vesicular glutamate transporter Glutamatergic neurons (Lee et al, 1999) egl-3 Proprotein convertase Peptidergic neurons (Kass et al, 2001) tph-1 Tryptophan hydroxylase Serotoninergic neurons (Sze et al, 2000) flp-4 FMRFamide-like Peptide A subset of peptidergic neurons (Kim & Li, 2004) flp-22 FMRFamide-like Peptide A subset of peptidergic neurons (Kim & Li, 2004) AMPA-type ionotropic glutamate A subset of neurons, including the (Hart et al, 1995; glr-1 receptor subunit command interneurons Maricq et al, 1995) A membrane associated guanylate Throughout the nervous system, (Hoskins et al, 1996) lin-2 kinase (MAGUK) family protein body-wall muscles (Fig. 3) (Miller et al, 1983) promoter modified myo-3 Myosin heavy chain (MHC) Body-wall muscles as described (Charlie et al, 2006b) Ortholog of the murine polycystic osm-5 kidney disease gene Tg737 and Ciliated sensory neurons (Haycraft et al, 2001) human IFT88 Synaptic acetylcholine transporter unc-17 Cholinergic motoneurons (Alfonso et al, 1993) (VAChT) Transmembrane vesicular GABA unc-47 GABAergic motoneurons (McIntire et al, 1997) transporter A subset of head neurons and unc-73C1 UNC-73 C1 isoform ventral cord, and some of the (Steven et al, 2005) neuron-associated sheath cells. Throughout the nervous system, and some of the neuron- unc-73E UNC-73 E isoform associated sheath cells. Also in (Steven et al, 2005) the excretory canals and the somatic gonad. (Maduro & Pilgrim, unc-119 Myristoyl-binding protein Throughout the nervous system 1995; Wright et al, 2011)

Figure 2. Neuronal expression of the UNC-73E isoform rescued the unc-73 slow locomotory phenotype. unc-73 mutants move more slowly than wildtype off food. Movement is represented by the number of body bends observed in a 20 second interval. The slow locomotory phenotype was rescued by expressing UNC-73E pan-neuronally (unc-119 promoter) or in large subsets of neurons (2lin- and egl-3 promoters). B) The slow locomotory phenotype was not rescued by UNC-73E expression in smaller subsets of neurons including specific peptidergic (flp-22 promoter) or serotoninergic (tph-1 promoter) neurons, alone or in combination with expression in motoneurons (unc-17 promoter). In these strains movement is represented by the number of body turns observed in a 40 second interval as they contain the pRF4 cotransformation marker and thus exhibit a roller phenotype. The nonUnc control strain is the heterozygous unc-73 mutant strain Ex[pRF4] unc- 73(ev802)/unc-11 dpy-5. The Ex[pRF4; D1] unc-73(ev802) lethargic mutant is the background unc-73 mutant used in these experiments. C) UNC-73E expression in additional neuronal subsets did not rescue the unc-73 RhoGEF2 mutant lethargic phenotype. Table 5 lists the expression patterns expected from the promoters used in these experiments. Select expression patterns observed in the strains used in these experiments are given in Fig 3. Data are presented as a mean ± SEM and analyzed by Student’s t-test. “**” p < 0.0001, indicating significantly different movement rates compared to unc-73 mutant animals under identical test conditions.

Figure 3. promoter::UNC-73E::GFP cell-specific expression patterns. All images show the same head region of the animal with anterior to the right and ventral down. Neurons identified by morphology and position are labeled. The observed expression patterns are consistent with previously reported expression patterns obtained with these promoters (Table 5). The transgene present in each animal is indicated on the image (“EG” is an abbreviation for UNC-73E::GFP). Images were produced from flattened confocal image stacks.

Figure 4. UNC-73E overexpression in different cell types in a wild-type background does not increase locomotion rate. “**” p < 0.0001, in comparison to N2 under identical test conditions.

3.1.2 RHO-1 functions downstream of UNC-73 RhoGEF2 activity in vivo

The UNC-73 RhoGEF2 domain functions as a Rho specific GEF in vitro (Spencer et al, 2001; Williams et al, 2007) and UNC-73 RhoGEF2 isoforms with a point mutation in the RhoGEF2 domain fail to rescue the unc-73 RhoGEF2 mutant locomotory phenotype (Steven et al, 2005; Williams et al, 2007). To determine if the Rho signaling pathway functions downstream of UNC-73 RhoGEF2 activity in vivo, constitutively activated RHO-1expression was tested for its ability to rescue the unc-73 RhoGEF2 lethargic phenotype.

A cDNA encoding constitutively active RHO-1(G14V), referred to as rho-1(gf), was used in these experiments (McMullan et al, 2006). The rho-1 (gf) cDNA was driven by a heat-shock promoter to temporally control RHO-1(G14V) expression. rho-1(gf) expression in heat-shocked adult transgenic animals partially rescued the unc-73(ce362) locomotory defect (6.8 ± 0.3 body bends per minute, n = 61 compared to unc-73(ce362)

3.5 ± 0.2, n = 32; Fig. 5). These results suggest that UNC-73 RhoGEF2 isoforms function through RHO-1 to physiologically regulate locomotion in developed animals.

Surprisingly however, rho-1(gf) driven by the unc-73E endogenous promoter did not significantly rescue the unc-73 RhoGEF2 slow locomotion phenotype (Fig. 5). rho-1(gf) expression in ventral cord motor neurons using the modified unc-17 promoter also failed to rescue, but this might be expected since UNC-73 RhoGEF2 isoform expression restricted to the ventral cord motor neurons did not rescue the unc-73 RhoGEF2 lethargic phenotype. Expression of these rho-1(gf) constructs in a wild-type background did not change the locomotion rate. This excluded the possibility the transgenes slowed the movement of animals expressing the rho-1(gf) transgene.

Figure 5. UNC-73 functions upstream of Rho to regulate locomotion. Constitutively activated RHO-1(gf) expression induced by heat-shock in adult worms partially rescued the unc-73(ce362) slow locomotion phenotype. Data are presented as a mean ± SEM and analyzed by two-tailed Student’s t test. “**” p < 0.0001.

3.2 UNC-73 RhoGEF2 isoforms modulate cholinergic signaling.

Since UNC RhoGEF2 activity functions in the adult nervous system to control the rate of locomotion (Figs. 2 and 5) (Hu et al, 2011), pharmacological agents were used to test if unc-73 RhoGEF2 mutants have any neurotransmission defects. Defects in C. elegans cholinergic neurotransmission can be tested with pharmacological agents such as aldicarb and levamisole.

Aldicarb is an antagonist of acetylcholinesterase, which is present in the synaptic cleft to decrease the level of acetylcholine after its release from the presynaptic neuron.

Exposure of C. elegans to aldicarb causes an accumulation of acetylcholine at neuromuscular junctions and other cholinergic synapses. Excess acetylcholine causes the body wall muscles to over-contract, which paralyzes the animal. Animals exposed to aldicarb are assayed by calculating the percentage of paralyzed worms in a time-course manner (Rand, 2007). Heterotrimeric G protein pathway mutants, such as egl-8 or egl-30, paralyze more slowly on aldicarb plates compared to wild-type worms because the animals release less acetylcholine into the neuromuscular junction (Lackner et al, 1999).

Consequently, aldicarb resistance is suggestive of defects in cholinergic neurotransmission and has been utilized to screen for neurotransmission defective mutants. Two unc-73 RhoGEF2 mutant alleles, Is[D1]ev802 and ce362, were tested and both consistently showed a weak resistance to aldicarb (Fig. 6A and B). A mildly aldicarb resistant egl-8(md1971) mutant was used as a control (Miller et al, 1999). The aldicarb resistance of both unc-73 alleles was fully rescued by UNC-73E RhoGEF2 isoform expression (Fig. 6A and B).

Instead of being caused by a decrease in presynaptic neurotransmitter release, aldicarb resistance can also result from decreased postsynaptic body wall muscle responses to acetylcholine. Levamisole is commonly used to examine C. elegans postsynaptic responses. Levamisole is an acetylcholine receptor agonist. Animals exposed to levamisole became paralyzed over time, again as a result of muscle over- contraction. Mutants, such as unc-29, that have defective acetylcholine receptors are resistant to the effects of levamisole (Rand, 2007). Both Is[D1]ev802 and ce362 mutants were hypersensitive to levamisole (Fig. 6C), and the sensitivity was substantially rescued by UNC-73E RhoGEF2 isoform expression. The unc-29 positive control strain was insensitive to levamisole during the whole time course of the experiment, while unc-73 mutants began to paralyze within 20 minutes under the same conditions. The impact of unc-73 mutants with cell-specific UNC-73E isoform expression on levamisole was also tested. The neuronal expression (Np) using rab-3 promoter, the muscular expression (Mp) using myo-3 promoter, and the motoneuronal expression (MNp) using unc-17 promoter of UNC-73E expression yielded hypersensitivity in the levamisole assay (Fig. 6D). It was examined if UNC-73E overexpression in a wild-type background altered levamisole sensitivity, but no difference in levamisole sensitivity from wildtype was observed (Fig.

7).

To examine if the cholinergic neurotransmission alterations in the unc-73

RhoGEF2 mutants was due to a defect in synaptic structure at neuromuscular junctions, synapse structure was examined at a gross level with the fluorescence-tagged SNB-1 synaptic vesicle marker expressed in either cholinergic (acr-2p::SNB-1::CFP) (Nurrish et al, 1999) or in GABAergic ( unc-25p::SNB-1::GFP) motoneurons (Hallam & Jin, 1998).

SNB-1/synaptobrevin-associated fluorescence localizes to synaptic active zones and forms puncta along C. elegans dorsal cord axons (Nonet et al, 1998). No obvious differences in synaptic structure between the unc-73 mutants and wildtype were observed

(Fig. 6E).These results indicate that UNC-73 RhoGEF2 mutants have altered levels of cholinergic signaling, but these alterations are not likely due to changes in basic synapse structure at the neuromuscular junction.

Figure 6. unc-73 mutants have altered cholinergic signaling, with no obvious changes in basic synaptic structure. A and B) Two unc-73 alleles were slightly resistant to 1mM aldicarb on food. UNC-73E expression under the control of its endogenous promoter restored unc-73 mutant sensitivity to aldicarb. The mildly aldicarb resistant egl-8(md1971) strain was used as a control. C and D) unc-73 mutants were hypersensitive to 500nM levamisole. UNC-73E expression partially rescued unc-73 mutant levamisole hypersensitivity. Partial rescue was also obtained with UNC-73E expression in the nervous system (Np; rab-3p) but not in just the cholinergic motoneurons (MNp; unc-17p) or in muscles (Mp; myo-3p). A levamisole resistant unc-29 strain was used as a control. Error bars indicate SEM, three independent experiments were combined for each strain. E) The distribution of synaptic puncta did not obviously change in the unc-73 mutant background compared to wild-type in the fluorescent images of the fluorescence-tagged synaptic marker SNB-1. Images were focused on the dorsal nerve cord at 1000 magnification.

Figure 7. UNC-73E overexpression in a wild-type background did not change levamisole sensitivity. Error bars indicate SEM, three independent experiments were combined for each strain.

3.3 unc-73 mutants have neuropeptide processing defects

Neurotransmission includes two main mechanisms involving either “classical” neurotransmitters released through SVs or neuromodulators released from DCVs. SV neurotransmission defects in unc-73 RhoGEF2 mutants were assayed as described above.

DCV-mediated neurotransmission was tested in the same mutants using a neuropeptide release assay developed independently by the Kaplan and Jorgensen labs (Sieburth et al,

2007; Speese et al, 2007).

Fluorescence intensity derived from NLP-21 neuropeptide, expressed in motoneurons and tagged with YFP, was measured from different locations in unc-73 mutant animals. Fluorescent signal produced from the cell body represents neuropeptide level before transport to synapses, fluorescence along the dorsal cord axon bundle represents neuropeptide in transport before release, and fluorescence from the coelomocytes represents the level of neuropeptide released from neurons. Coelomocytes are C. elegans scavenger cells that continuously take up molecules from the body cavity fluid, including neuropeptides secreted by the neurons. The fluorescence intensity from axon bundles and coelomocytes decreased in unc-73(ce362) mutants compared to a wildtype background, while cell body fluorescence did not change significantly (Fig. 8).

These results suggest that UNC-73 RhoGEF2 isoforms are involved in neuropeptide- mediated signaling processes.

Figure 8. Neuropeptide level is reduced in unc-73 mutants. A) Representative confocal image stacks of animals expressing NLP-21::YFP from cholinergic motor neurons in wild-type and unc-73mutant backgrounds. Upper left: DA6 and DB6 cell bodies along the ventral cord. Lower left: posterior coelomocytes. Right: axon bundles in the dorsal cord. Insets show the punctate fluorescence expected from DCVs carrying fluorescently tagged neuropeptide along the dorsal cord. B) The confocal images were used to quantify unc-129p::NLP-21::YFP derived fluorescent signals standardized to the wild-type background. Fluorescent signals inunc-73 mutants were decreased in axons and coelomocytes, but not in the cell bodies. C) Confocal image stacks of representative coelomocytes containing TR-BSA in wild-type and mutant backgrounds. D) Standardized TR-BSA derived fluorescent signals 1 hour after injection. No significant difference between wild type and unc-73(ce362) suggested that unc-73 mutant coelomocyte function is intact. Data are presented as a mean ± SEM and analyzed by Welch Two Sample t-test. “*” p < 0.01, “**” p < 0.001.

Mutants defective in coelomocyte uptake could affect the results of the neuropeptide release assay. To exclude the possibility that unc-73 mutants affect fluid uptake by coelomocytes, a control experiment was performed by injecting Texas Red conjugated BSA (TR-BSA) into the pseudocoelom of mutant and wild-type animals, and measuring the accumulation of TR-BSA derived fluorescence in the coelomocytes after a one hour period (Zhang et al, 2001). This assay reveals the ability of coelomocytes to take up fluid from the pseudocoelom. In the control assay, there was no significant difference between the unc-73 mutant and wild-type backgrounds (Fig. 8C, D). The control experiment suggested the decreased coelomocyte fluorescence intensity observed in the neuropeptide release assay was not due to a coelomocyte uptake defect.

3.4 Activated Gαs pathway mutants suppress unc-73 RhoGEF2 mutant defects

Mutants with elevated Gαs pathway activity have a phenotype opposite that of unc-73 RhoGEF2 mutants. For example, gsa-1(ce81) and acy-1(ce2) have point mutations resulting in constitutively active GSA-1/Gαs and ACY-1 (Adenylyl cyclase) respectively. Both ce81 and ce2 animals are hyperactive, moving faster and bending more often than wildtype in locomotion assays, and they are hypersensitive to aldicarb compared to the aldicarb resistant unc-73 mutants (Schade et al, 2005). To examine the relationship between the UNC-73 and Gαs pathways, double mutants were made with unc-73 loss of function and gsa-1 pathway gain of function mutants.

3.4.1 The activated Gαs pathway is required in both muscles and nervous system to

fully rescue the locomotory phenotype of unc-73 mutants

Activated Gαs/GSA-1 pathway mutants suppress the unc-73 RhoGEF2 mutant locomotory defect indicating the Gαs/GSA-1 pathway functions downstream of or in parallel to UNC-73 RhoGEF2 activity (Fig. 9A) (Hu et al, 2011). To investigate which cells require activated Gαs pathway activity to rescue unc-73 lethargic movement, transgenic constitutively activated (CA) ACY-1(P260S) constructs driven by different cell-specific promoters were used (the ACY-1(CA) DNA constructs were kindly provided by Kenneth Miller, Oklahoma Medical Research Foundation). ACY-1(CA) was required in both muscles and the nervous system to rescue the slow locomotion phenotype of unc-73 RhoGEF2 mutants (Fig. 9B). Unlike transgenic UNC-73 nervous system expression, which rescued the unc-73 lethargic phenotype, pan-neuronal transgenic ACY-1(CA) expression did not rescue unc-73 RhoGEF2 mutant lethargy.

ACY-1(CA) expression in body-wall muscles increased the rate of movement of unc-73 mutants (by 33%), but the rate did not reach a wild-type level. These results suggest that the Gαs pathway in both neurons and body-wall muscles functions downstream of or in parallel to UNC-73 RhoGEF2 activity in the regulation of locomotion.

ACY-1 was used instead of GSA-1 in these rescue experiments because constitutively activated GSA-1 expression in the nervous system under the control of the rab-3 promoter caused a slow movement phenotype in a wildtype background (data not shown), which is a phenotype opposite that of the activated Gαs pathway mutant gsa-

1(ce81). Transgenic animals expressing constitutively active GSA-1 under the control of the gsa-1 promoter results in the death of some neurons (Korswagen et al, 1997), which may be the reason for the slow movement phenotype observed in animals containing rab-3p::gsa-1(gf). Expression of wild-type GSA-1 under the control of its own promoter results in hyperactive movement (Korswagen et al, 1997), so it appears that neurons have different responses to the overexpression of either the wild-type or constitutively active forms of GSA-1.

UNC-73E was identified as a downstream factor of Gαq/EGL-30 in parallel to

PLCβ/EGL-8, but the role of UNC-73 RhoGEF2 activity in the regulation of locomotion was not determined (Williams et al, 2007). As described above, unc-73 mutants have an obvious locomotory defect that is rescued by an activated Gαs pathway (Fig. 9) (Hu et al,

2011). The reported ability of Gαq to activate UNC-73E suggests that Gαs activation should also rescue the phenotype of loss of function Gαq mutations. However, it was previously shown that the activated Gαs/GSA-1 pathway does not rescue the locomotory phenotype of Gαq/EGL-30 mutants (Schade et al, 2005). To examine the interaction

56 between Gαq/EGL-30 and Gαs/GSA-1 in more detail an additional egl-30 allele was used.

EGL-30/Gαq null mutants have an embryonic lethal phenotype (Brundage et al, 1996), therefore the viable allele ad805 was tested in locomotion assays. ad805 animals had an extremely slow locomotion rate (zero bends per 20 seconds), likely because of abolished neurotransmitter release at the neuromuscular junction (Lackner et al, 1999). This locomotory defect was not rescued by the hyperactive Gαs pathway mutant allele acy-

1(ce2) even though the same ce2 mutation fully rescued the unc-73 RhoGEF2 slow movement phenotype (Fig. 9C).

Figure 9. Constitutive activation of the Gαs pathway rescues the unc-73 slow locomotion defect. A) The gain-of-function gsa-1(ce81) allele suppressed the lethargic locomotion phenotype of unc-73 RhoGEF2 mutants. B) Transgenic expression of constitutively activated ACY-1(P260S) (acy-1(gf)) rescued the locomotory defect of unc-73(ce362) mutants when acy-1(gf) was expressed in both the nervous system, using the rab-3 promoter (NSp), and the body wall muscles, using the myo-3 promoter (Mp). Nervous system specific expression alone was not sufficient to rescue. Body wall muscle expression of constitutively activated GSA-1 or ACY-1 slightly, but significantly, increased the locomotory rate in an unc-73 mutant background. C) Loss-of-function Gαq pathway mutants egl-30(ad805) and egl-8(n488) have slow movement phenotypes. The hyperactive Gαs gain of function mutant acy-1(ce2) did not rescue the severe ad805 slow movement defect. Data are presented as a mean ± SEM and analyzed by two-tailed Student’s t test. “**” p < 0.0001, significantly different under identical test conditions. The acy-1(gf) DNA constructs were provided by Kenneth Miller (Oklahoma Medical Research Foundation).

3.4.2 Double mutants of unc-73 RhoGEF2 and the activated Gαs pathway are

hypersensitive to aldicarb and levamisole

The unc-73 RhoGEF2 mutants are resistant to aldicarb and hypersensitive to levamisole while activated Gαs pathway mutants are hypersensitive to both aldicarb and levamisole (Fig. 10 and 11) (Charlie et al, 2006b; Schade et al, 2005). The drug sensitivities of Gαs pathway double mutants with unc-73 were also tested. The aldicarb hypersensitivity of the constitutively active Gαs pathway mutants indicates these mutants release more acetylcholine into the synapse or they are more sensitive to acetylcholine in comparison to wildtype (Fig. 10) (Schade et al, 2005). Increased acetylcholine release and sensitivity are likely factors involved in the activated Gαs pathway rescue of unc- 73

RhoGEF-2 lethargy as gsa-1(ce81) unc-73(ce362) and Is [D1]; gsa-1(ce81) unc-

73(ev802) double mutants were hypersensitive to both aldicarb and 0.5 mM levamisole

(Fig. 10 and 11). Increased acetylcholine release may be the more significant factor in aldicarb hypersensitivity since gsa-1(ce81) mutants are actually resistant to a lower concentration of 0.1 mM levamisole (Schade et al, 2005). It therefore appears that there is a correlation between the response to aldicarb and locomotion rate. Lethargic unc-73

RhoGEF2 mutants are resistant to aldicarb while the faster moving activated Gαs pathway mutants and the double mutants with unc-73 are hypersensitive to aldicarb. Interestingly, however, levamisole sensitivity did not correlate with the locomotory phenotypes of unc-

73 RhoGEF2 and Gαs pathway mutants. For example, rescued gsa-1(ce81) unc-73(ce362) double mutants with wild-type locomotion rates and lethargic unc-73(ce362) single mutants were both strongly hypersensitive to levamisole (Fig. 11). Similarly, transgenic

59 animals that are lethargic due to reduced RHO-1 activity or hyperactive as a result of rho-

1(gf) expression are both hypersensitive to levamisole (McMullan et al, 2006).

Figure 10. Activated Gαs pathway mutants and double mutants with unc-73 are hypersensitive to aldicarb. A and B) Gain-of-function gsa-1(ce81) unc-73 double mutants are hypersensitive to aldicarb similar to gsa-1(ce81) mutants alone. Error bars indicate SEM. Three independent experiments were combined for each strain.

Figure 11. Activated Gαs pathway mutants and the double mutants with unc-73 are hypersensitive to levamisole. Gain-of-function gsa-1(ce81) mutants and gsa-(ce81) double mutants with unc-73(ce362) are hypersensitive to levamisole. Error bars indicate SEM. Three independent experiments were combined for each strain.

3.4.3 Gαs pathway activation alters neuropeptide levels

Constitutive activation of the Gαs pathway rescues the unc-73 RhoGEF2 mutant lethargic phenotype (Fig. 9). Since C. elegans Gαs pathway activation increases DCV- mediated exocytosis (Zhou et al, 2007), one hypothesis is that the unc-73 RhoGEF2 mutant lethargy and decreased neuropeptide level are compensated for by the increased rate of neuropeptide release from DCVs due to constitutive Gαs pathway activation. To test this hypothesis NLP-21::YFP derived fluorescence was examined in gain of function gsa-1(ce81) mutants and ce81 ce362 double mutants. The amount of neuropeptide released in the ce81; ce362 double mutants, however, was not increased relative to the unc-73(ce362) mutant alone. Neuropeptide release in ce81 ce362 mutants, as measured by neuropeptide uptake by coelomocytes, was 52.4% that of wildtype while the gsa-

1(ce81) mutant alone was 18.6% of the wild-type level (Fig. 12A).

Since the coelomocyte fluorescence was significantly decreased in the ce81 mutant, TR-BSA uptake in ce81 coelomocytes was examined. The TR-BSA level in ce81 mutants was increased (Fig. 12B), indicating ce81 coelomocytes are more active in the endocytosis of molecules from the environment or they exhibit a decreased rate of TR-

BSA degradation. The altered coelomocyte physiology in ce81 mutants, as revealed by the TR-BSA measurements, made it difficult to interpret the NLP-21::YFP derived fluorescence data from ce81 mutant coelomocytes.

Observations of NLP::YFP derived fluorescence were extended to axons and cell bodies (Fig. 12A). ce81 fluorescence intensity in cell bodies was significantly higher than that observed in the wild-type background, but was lower in axons. ce81 ce362 fluorescence intensity in the cell bodies was also strongly increased relative to wildtype

(Fig. 12A), to a level similar to that of the ce81 single mutant. The fluorescence intensity of the double mutant in the axons was not significantly different from the wild-type background. These results reveal that GSA-1 activity in cholinergic motor neurons significantly increases neuropeptide level in the cell bodies, but whether this increase in neuropeptide level is involved in the gsa-1 mediated rescue of unc-73 RhoGEF2 lethargy is still to be determined.

In addition, neuropeptide levels were examined in the Gαs pathway weak loss of function mutant acy-1(pk484) as it might be expected that Gαs pathway loss of function mutants present a phenotype opposite to that of the gsa-1 gain of function mutants. Since the gsa-1 null mutants arrest development in the L1 stage (Korswagen et al, 1997), the milder acy-1(pk484) allele was used in the neuropeptide release assay (Fig. 12A). pk484 did not show any differences from wildtype, in the cell bodies, axons, or coelomocytes.

As pk484 is a mild mutation (Moorman & Plasterk, 2002) it is possible a more severe loss of function mutation would reveal defects in this assay.

Figure 12. Neuropeptide levels are altered in a Gαs pathway mutant background. A) Standardized fluorescence intensity from unc-129p::nlp-21::yfp in the DA6 and DB6 motor neuron cell bodies, dorsal cord axon bundles and posterior coelomocytes of the indicated strains. The activated Gαs pathway mutant gsa-1(ce81), unc-73(ce362) and the ce81 ce362 double mutant had less YFP accumulation in the coelomocytes. However, ce81 and ce81;ce362 fluorescence was greatly increased in the cell bodies. The loss-of- function Gαs pathway mutant acy-1(pk484) displayed neuropeptide levels comparable to wildtype. B) Standardized fluorescence intensity of Texas Red from anterior coelomocytes in the TR-BSA control assay. The activated Gαs pathway mutant gsa- 1(ce81) had more Texas Red accumulated in the coelomocytes, indicating more active coelomocyte uptake. Data are presented as a mean ± SEM and analyzed by Welch Two Sample t-test. “**” p < 0.0001.

3.5 The relationship between UNC-73 and other molecules in the DCV-mediated signaling pathway

Since UNC-73 RhoGEF2 activity is unlikely to play a central role in classical synaptic vesicle release mechanisms in neurotransmission, and unc-73 RhoGEF2 mutants had altered neuropeptide levels, an important question is how unc-73 compares to other genes functioning in DCV-mediated signaling pathways. The unc-73 phenotype was compared and genetic interactions were examined with several genes that play important roles in neuropeptide production, processing or release including unc-31, rab-2, egl-3 and tom-1.

3.5.1 rab-2,unc-31 and unc-73 RhoGEF2 mutants have similar slow movement

phenotypes that are rescued by Gαs pathway activation

The lethargic yet coordinated locomotion phenotype of unc-73 RhoGEF-2 mutants is separate from the severe uncoordinated movement phenotypes of genes such as snb-1 (synaptobrevin), which are fundamental to the mechanisms of synapse function involving synaptic vesicles, and is perhaps more suggestive of a role for RhoGEF-2 activity in the modulatory mechanisms controlling locomotion rate (Nonet et al, 1998;

Steven et al, 2005). For example, C. elegans RAB2 (RAB-2) function is critical for the modulatory regulation of locomotion through its role in DCV maturation (Edwards et al,

2009; Sumakovic et al, 2009). Importantly, the rab-2 locomotion phenotype (Chun et al,

2008; Park & Horvitz, 1986) is very similar to the unc-73 RhoGEF2 slow movement phenotype; animals are extremely slow on food, but increase their speed off food to a rate about one-third that of wild type (Fig.13). UNC-31/CAPS is a DAG binding protein

64 required for DCV docking and the exocytosis of hormones and neuropeptides from neuroendocrine cells (Avery et al, 1993; Speese et al, 2007). unc-31 loss of function mutants also have a lethargic movement phenotype (Speese et al, 2007). The unc-73, unc-

31 and rab-2 mutant worms are similar in that the animals move in a coordinated manner, but can be described as having an “unmotivated” phenotype as their rate of locomotion is much slower than wildtype.

As observed with unc-73, the rab-2 slow movement phenotype was suppressed by activation of the GSA-1 pathway. rab-2(n501); acy-1(ce2) double mutants had locomotion rates more similar to wildtype (Fig. 13A). Similarly, unc-31(e928) mutants were also rescued by GSA-1 pathway activation (Charlie et al, 2006a). In cell-specific rescuing experiments, ACY-1(CA; P260S) expression in both muscle and the nervous system rescued the rab-2 locomotory defect, but neuronal expression alone failed to rescue (Fig. 13B). This is similar to what was observed with unc-73 RhoGEF2 mutants.

However, neuronal expression of ACY-1(CA) significantly increased unc-31 mutant locomotion rate while expression in muscles did not (Fig. 13B).

Figure 13. rab-2, unc-31, and unc-73 RhoGEF2 mutants have a similar slow locomotion defect rescued by Gαs pathway activation. A) rab-2(n501) and unc-31(e928) have a locomotory defect similar to unc-73 mutants. The rab-2 defect is rescued by the gain-of-function acy-1(ce2) mutant. B) Transgenic ACY-1(CA) expression was required the in both the nervous system and body wall muscles to rescue rab-2, similar to what was observed with unc-73 RhoGEF2 mutants. Neuronal ACY-1(CA) expression and the combination of neuronal and muscular expression rescued unc-31 locomotory defects. Data are presented as a mean ± SEM and analyzed by two-tailed Student’s t test. “**” p < 0.0001.

3.5.2 rab-2, unc-31 and unc-73 RhoGEF2 mutants have similar responses to levamisole

To further compare rab-2, unc-31, and unc-73 RhoGEF2 mutant phenotypes, rab-

2 and unc-31 mutants were exposed to aldicarb and levamisole as described above for the unc-73 mutants. In the drug sensitivity assays, rab-2(n501) had similar aldicarb resistance to unc-73(ce362) mutants (Fig. 14A), while unc-31(e928) did not have altered sensitivity to aldicarb compared to wildtype in my acute drug response assay (Fig. 14A).

Similar results with rab-2 and unc-31 were also observed with the unc-31 analysis performed using a chronic aldicarb assay (Charlie et al, 2006a; Edwards et al, 2009;

Sumakovic et al, 2009). Both unc-31(e928) and rab-2(n501) were hypersensitive to levamisole (Fig. 14B). However, unc-31(e928) mutants have a wild-type levamisole response when assayed by measuring body length contraction (Charlie et al, 2006a).

Figure 14. rab-2 and unc-31 mutants display altered sensitivity to levamisole and aldicarb. A) rab-2(n501) mutants are resistant to aldicarb similar to unc-73(ce362) animals. However, unc-31(e928) mutants have a wild-type response to aldicarb. The mildly aldicarb resistant egl-8(md1971) strain was used as control. B) Both rab-2(n501) and unc-31(e928) mutants were hypersensitive to levamisole. Levamisole resistant unc-29 animals were used as a control. Error bars indicate SEM. Three independent experiments were combined for each strain.

3.5.3 unc-73 RhoGEF2, rab-2 and unc-31 mutants respond similarly to PMA treatment

Phorbol ester (PMA) was used to further characterize the lethargic mutants. PMA mimics the biological effects of DAG and stimulates neurotransmission by binding to proteins such as UNC-13 and PKC to stimulate vesicle release and bypass the requirements of upstream pathways (Lou et al, 2008). Mutants with PKC pathway defects, such as tpa-1, which encodes C. elegans PKC, are resistant to PMA induced growth arrest (van der Linden et al, 2003).

Wildtype, unc-73(ce362), rab-2(n501) and unc-31(e928) were stimulated by

PMA treatment and displayed a coiled posture likely the result of increased muscle contraction (Fig. 15). The mutant animals were also rescued for their slow movement phenotypes (data not shown). The locomotion rate of the PMA treated animals could not be measured using the body bend assay due to the severe coiling displayed by these animals. This result suggests that the downstream mechanisms of synaptic vesicle mediated neurotransmission are intact in unc-73, rab-2 and unc-31 mutants.

Figure 15. Animals treated with PMA display a similar coiled posture. Wildtype and the unc-73(ce362), rab-2(n501), and unc-31(e928) mutants display a similar coiled posture in response to PMA exposure. Animals were incubated on NGM plates containing either 20μl of the EtOH carrier or 20μl of 10 mM PMA. Pictures were taken after 2.5 hour under the same condition.

3.5.4 Analysis of unc-73 RhoGEF2 double mutants with egl-3 and tom-1.

EGL-3/proprotein convertase catalyzes the key peptide precursor cleavage step in

FMRFamide-like family neuropeptide production and as expected the level of

FMRFamide-like neuropeptides decreases in the egl-3(ok979) null mutants (Husson et al,

2006). The egl-3(ok979) allele was used in this study to investigate the relationship between UNC-73 RhoGEF2 activity and neuropeptide related signaling. The locomotion rate of the egl-3(ok979) mutant was decreased off food in the body bend assay (Fig. 16A)

(Kass et al, 2001).This slow locomotion phenotype in egl-3 mutants was also rescued by the activated Gαs pathway mutant gsa-1(ce81) (Fig. 16A), similar to ce81 rescue of unc-

73 RhoGEF2 mutants (Fig. 9A). egl-3(ok979) mutants also showed altered drug sensitivity defects similar to unc-73 RhoGEF2 mutants, with resistance to aldicarb (Fig.

16C) and hypersensitivity to levamisole (Fig. 16D), as reported previously (Jacob &

Kaplan, 2003). Again, similar to what was observed with unc-73 RhoGEF2 mutants egl-

3(ok979) aldicarb resistance was suppressed by the activated gsa-1(ce81) mutant (Fig.

16C). The similarities between the phenotypes of unc-73 and egl-3 are consistent with both genes functioning in DCV-mediated signaling pathways.

Since my results indicated UNC-73 RhoGEF-2 isoforms affect the level of neuromodulatory protein in the DCVs of peptidergic neurons, I looked for evidence of a genetic interaction between unc-73(ce362) and the egl-3(ok979) null mutant. The unc-

73(ce362);egl-3(ok979) double mutant moved slower than either single mutant (1.1 ±

0.2 bends/20sec), suggesting EGL-3 and the UNC-73 RhoGEF-2 isoforms act in parallel pathways (Fig. 16A). The hyperactive ce81 mutant also rescued the locomotion rate of the severely slow ce362;ok979 double mutant to a rate faster than that of wildtype (Fig.

16A). The slow moving double mutant was also more severely hypersensitive to levamisole than either single mutant, and had similar hypersensitivity to aldicarb as the ce81 mutant alone (Fig. 16).

These results correlate with the ce81 suppression of the unc-73 mutants and suggest that UNC-73 may function in parallel to EGL-3 mediated neuropeptide production. The results also indicate the Gαs pathway functions downstream of neuropeptide and DCV-mediated signaling pathways and can fully bypass defects in neuropeptide production resulting from the lack of EGL-3.

TOM-1/Tomosyn is a syntaxin interacting protein that negatively regulates UNC-

31-dependent DCV exocytosis (Gracheva et al, 2007). The tom-1(ok285) deletion allele is hypersensitive to aldicarb suggesting TOM-1 also negatively regulate synaptic vesicle release (Dybbs et al, 2005). The tom-1(ok285);[D1]unc-73(ev802) double mutant was made to investigate if these two genes genetically interact. tom-1(ok285) animals did not show a higher locomotory rate as might have been expected (Fig. 16B) since increased cholinergic signaling can correlate with a higher locomotion rate, as in worms containing gain-of-function RHO-1 (McMullan et al, 2006). The tom-1(ok285);[D1]unc-73(ev802) double mutant was as slow as the unc-73 single mutant (Fig. 16B), indicating tom-1 did not suppress the unc-73 lethargic movement.

Figure 16. Analysis of unc-73 RhoGEF2 double mutants with egl-3 and tom-1. A) egl-3(ok979) null mutants have a slow locomotion phenotype that can be rescued by the activated Gαs pathway mutant gsa-1(ce81). The unc-73(ce362);egl-3(ok979)double mutant moved slower than either single mutant and the double mutants’ slow movement is suppressed by gsa-1(ce81). B) unc-73 RhoGEF2 lethargy is not altered in a tom- 1(ok285) mutant background. Data are presented as a mean ± SEM and analyzed by two- tailed Student’s t test. “**” p < 0.0001. C) egl-3(ok979) mutants are resistant to aldicarb while ok936 mutants in a gain-of-function gsa-1(ce81) mutant background are hypersensitive to aldicarb. D) egl-3(ok979) and unc-73(ce362);egl-3(ok979)double mutants are hypersensitive to levamisole. Error bars indicate SEM. Three independent experiments were combined for each strain.

3.5.5 gsa-1(gf) and unc-73(lf) mutants have a shorter body length than wildtype

It was also observed that the gsa-1(ce81) gain of function double mutant with unc-73 was shorter than either single mutant, while the length of each single mutant was shorter than wildtype (Fig. 17). This observation indicates UNC-73 RhoGEF2 isoforms and the Gαs pathway are involved in controlling animal length during development, possibly through DCV signaling affecting the function of proteins or growth factors in addition to neuropeptides.

For example, TGFβ signaling may be affected in unc-73 RhoGEF2 mutants. The

C. elegans TGFβ family member UNC-129 has a locomotory phenotype, but it is thought to result from axon guidance defects during development (Colavita et al, 1998).

Mutations in other TGFβ signaling molecules can suppress unc-2 mutant calcium channel lethargy, but the mechanisms are not defined (Estevez et al, 2004). C. elegans TGFβ pathways are more well known for the control of dauer formation and importantly, animal size, a trait affected in unc-73 RhoGEF-2 mutants (Figure. 17). Also relevant are observations that vertebrate TGFβ affects the activity of mature synapses at the neuromuscular junction (Chin et al, 2002; Fong et al, 2010).

Figure 17. The gsa-1(gf) and unc-73 RhoGEF2 (lf) mutants have a shorter body length than wildtype. The gsa-1(ce81) and [D1]ev802 mutants were about 80% of the length of wildtype after they reached adulthood. The [D1] ev802 ce81 double mutant was even shorter, about 60% of wild-type length. Staged worms were kept on standard seeded NGM plates made from the same batch. The body length was measured every 12 hours from the L4 “crescent vulva” stage using captured images and measuring the length of a line drawn along the midline of the worm. Error bars indicate SEM, three independent experiments were combined for each strain.

3.6 An examination of the relationship between the UNC-73 RhoGEF2 pathway and glutamate receptor signaling

3.6.1 UNC-73 overexpression increases reversal frequency

Wild-type animals spontaneously change their directions from forward to backward with a specific frequency. Spontaneous reversal frequency is defined by the nervous system and is affected by environmental cues (Hart (ed.), 2006). Mutations in glutamatergic neurotransmission can change the reversal frequency. glr-1 encodes an

AMPA-type ionotropic glutamate receptor subunit (Hart et al, 1995). In C. elegans, glr-1 mutants exhibit defects in the backing response to strong mechanical stimuli (Hart et al,

1995; Maricq et al, 1995), and a higher spontaneous reversal frequency on food (Zheng et al, 1999).

Interestingly, UNC-73E isoform overexpression caused a higher spontaneous reversal rate (16.6 ±1.5 reversals in 3 minutes off food) than wildtype (6.3 ±0.6) (Fig.

18). This hyper reversal phenotype was also exhibited by animals expressing egl-3p::unc-

73E, which drives UNC-73E expression in peptidergic neurons, and in animals expressing UNC-73E in GLR-1 expressing neurons through glr-1p::unc-73E (Fig. 18).

Control animals containing unc-47p::unc-73E did not exhibit a hyper reversal frequency

(Fig. 18). These animals expressed unc-73E in the GABAergic motoneurons, which do not modulate the reversal responses. These results indicate that UNC-73 RhoGEF2 activity may function in glutamatergic neurons to regulate reversal behavior by controlling GLR-1 in a cell autonomous manner. The expression of constitutively activated RHO-1 in the nervous system using rab-3p::rho(gf) or unc-73Ep::rho-1(gf) did not change the reversal rate (Fig. 18). These results indicate that a higher level of UNC-

73 RhoGEF2 activity may cause a hyper-reversal phenotype independent of Rho. A wildtype reversal frequency was also observed in animals expressing acy-1(gf) in the nervous system using the rab-3 promoter indicating Gs pathway activation in the nervous system does not influence reversal frequency.

Figure 18. Overexpression of UNC-73E increases reversal frequency. Expression of the UNC-73E isoform in a wild-type background significantly elevates the spontaneous reversal rate off food. Hyper-reversal is observed when UNC-73E is expressed using the unc-73E (pan-neuronal), egl-3 (peptidergic neurons) or glr-1 (glutamate receptor neurons) promoters. RHO-1(gf) or ACY-1(gf) expression did not increase the reversal rate. Data are presented as a mean ± SEM and analyzed by two- tailed Student’s t test. “**” p < 0.0001.

3.6.2 GLR-1 is mislocalized in unc-73 mutants

Since UNC-73 RhoGEF2 activity increases glutamate modulated reversal frequency and rab-2 affects GLR-1 glutamate receptor localization (Chun et al, 2008) while exhibiting locomotion and drug sensitivity phenotypes similar to unc-73, it is possible that unc-73 mutants may also affect GLR-1 localization. To test this an integrated nuIs24[glr-1p::glr-1::gfp] line was crossed into an unc-73(ce362) mutant background and the fluorescent signal from the GLR-1::GFP fusion protein was observed.

GLR-1 is expressed in many neurons including the single pair of PVQ neurons.

The PVQ cell bodies are located in the lumbar ganglia in the tail of the worm, and they extend neurites into the ventral cord and up into the nerve ring in the anterior. GLR-

1::GFP distribution in the PVQ axons was altered in unc-73 mutants, presenting as a visible bright fluorescent spot located between the cell body and the ventral cord (Fig.

19). This phenotype is apparent in the unc-73(ce362) RhoGEF2 mutants (28.4%) and unc-73(e936) and rh40 RhoGEF1 mutants (37.1% and 21.9% respectively), while it is only rarely seen in wild-type animals (2.7%), which typically display an even distribution of GLR-1::GFP fluorescence in this region. These fluorescent spots may indicate a receptor trafficking defect with the receptor accumulating at the beginning of the axon instead of traveling properly down the axon after leaving the cell body.

GLR-1::GFP localization was similarly examined in the background of two genes, lin-10(e1439) and rab-2(n501), which are involved in receptor trafficking. LIN-10 is required for proper localization of GLR-1 receptors to the synaptic puncta along the ventral nerve cord and to the Golgi within cell bodies (Glodowski et al, 2005). Only 4.7% of lin-10 animals had an altered GLR-1::GFP distribution similar to that seen in the unc-

73 mutants, which was not significantly different from wildtype, however, 33.2% of rab-

2(n501) mutants displayed the GLR-1 accumulation defect (Fig. 19).

Total Normal With Defect Background p-value Number Number Percentage Number Percentage N2 149 145 97.3% 4 2.7% lin-10(e1439) 296 282 95.3% 14 4.7% 0.4363 unc-73(ce362) 222 159 71.6% 63 28.4% 6.865e-10 unc-73(e936) 280 176 62.9% 104 37.1% 1.233e-14 unc-73(rh40) 192 150 78.1% 42 21.9% 6.178e-07 rab-2(n501) 370 247 66.8% 123 33.2% 5.467e-13

Figure 19. unc-73 and rab-2 mutants have an altered GLR-1::GFP localization. A) Representative images of PVQ neurons expressing GLR-1::GFP in the indicated genetic backgrounds. Wild-type animals rarely display the GLR-1::GFP accumulations (arrowheads) seen in the mutant animals. Arrowheads indicate the bright spot just out of the cell body. B) Quantification of the GLR-1::GFP accumulation defects observed in the mutant strains. The p value was calculated using a X-squared test.

3.7 The UNC-73 RhoGEF2 domain is most similar to mammalian Kalirin, Trio and p63RhoGEF RhoGEF domains

RhoGEFs share relatively low percentage of conserved DNA sequences even among the RhoGEFs with same GTPase specificity (Rossman et al, 2005). Mammalian

Trio and Kalirin, Drosophila Trio and C. elegans UNC-73 belong to the Trio family of

RhoGEFs because they share similar protein structures containing two RhoGEF domains

(Bateman & Van Vactor, 2001). A shorter mammalian Rho-specific GEF p63RhoGEF shares structural similarity with the Trio RhoGEF2 and PH2 domains as well as a C- terminal α helix extension of the PH domain (Rojas et al, 2007). Both Trio and p63RhoGEF are activated by Gq so these observations suggest p63RhoGEF is a possible paralogue to a Trio RhoGEF2 isoform. Since the GEF function is accomplished by the RhoGEF domain instead of PH domain, the RhoGEF domains of 26 Rho-specific

GEFs in human and C. elegans were compared.

The conserved RhoGEF domain is about 180 amino acids long and interacts directly with GDP-bound Rho GTPases for GEF function. The protein sequence of six

RhoGEF domains of C. elegans Rho-specific GEFs were compared with twenty RhoGEF domains from human RhoGEF proteins. A phylogenetic tree was generated by MEGA5

(Fig. 20). UNC-73 RhoGEF2 domain was closely grouped with its known homologs Trio

RhoGEF2 and Kalirin RhoGEF2 as well as p63RhoGEF. Other C. elegans RhoGEFs showed similarity to their human homologs as reported, such as ECT-2 to ECT2, or

CGEF-1 to Dbl (Rossman et al, 2005).

To establish a consensus RhoGEF sequence, UNC-73 RhoGEF2, Trio RhoGEF2,

Kalirin RhoGEF2, and p63RhoGEF RhoGEF domains were compared (Fig. 21). The

79 multi-alignment result showed these four sequences are highly conserved with the amino acid identities of UNC-73 RhoGEF2 with p63RhoGEF RhoGEF, Trio RhoGEF2, and

Kalirin RhoGEF2 domains at 49%, 55%, and 48%, respectively (blastp (NCBI)). The multi-alignment result generated a consensus sequence among these four domains

(ClustalW2 (EBI); Fig. 21).

Figure 20. The UNC-73 RhoGEF2 domain is homologous to Trio RhoGEF2, Kalirin RhoGEF2, and p63RhoGEF RhoGEF domains. A phylogenetic tree was generated by the Maximum Likelihood method, tested by Bootstrapping 500 times and based on the amino acid sequences of listed RhoGEF domains from six C. elegans and twenty human Rho-specific GEFs using MEGA5. The C. elegans sequences are marked by red dots. The numbers at the branches indicate confidence values.

Figure 21. Multi-alignment of UNC-73RhoGEF2, Trio RhoGEF2, Kalirin RhoGEF2, and p63RhoGEF RhoGEF domains. The bottom line indicates the consensus Trio-family Rho-specific RhoGEF amino acid sequence. Aligned with ClustalW2 (EBI).

Chapter 4

4 Discussion

In this study, it was demonstrated that UNC-73 RhoGEF2 isoforms function in the nervous system to regulate locomotion rate most likely by modulating both synaptic vesicle-mediated and dense core vesicle-mediated neurotransmission in C. elegans.

Genetic analysis revealed UNC-73/Rho signaling interacts with heterotrimeric G protein pathways indicating a complex signaling network operates in the regulation of locomotory rate. Analysis of glutamate neurotransmitter receptor localization in unc-73 mutants also revealed that UNC-73/Rho signaling affects postsynaptic function.

4.1 UNC-73 RhoGEF2 isoforms and the Gαq pathway

The heterotrimeric Gαq pathway is an excitatory pathway that stimulates acetylcholine neurotransmitter release at the neuromuscular junction in C. elegans (Miller et al, 1999). The Rho-specific p63RhoGEF, a mammalian homolog of the UNC-73

RhoGEF2 isoform, directly binds to Gαq and RhoA (Rojas et al, 2007). UNC-73E was also identified as a direct Gαq downstream factor in a screen for suppression of the hyperactive locomotion and slow growth resulting from constitutively activated EGL-

30/Gαq (Williams et al, 2007).

In this study it was observed that unc-73 RhoGEF2 mutants have phenotypes that are distinct from Gαq pathway mutants suggesting UNC-73 RhoGEF2 activity many not always function downstream of Gαq. unc-73 RhoGEF2 mutants move slower than wildtype, and respond to prodding by increasing their rate of movement, while egl-

30(ad805) animals are essentially paralyzed and do not respond to prodding (Fig. 2A and

9C) (Lackner et al, 1999; Steven et al, 2005). In drug sensitivity assays, unc-73 RhoGEF2 mutants much weaker aldicarb resistance compared to egl-8 and egl-30 (Fig. 6) (Lackner et al, 1999). Levamisole hypersensitivity was observed in unc-73 mutants (Fig. 6), but not in egl-30 or egl-8 mutants (Lackner et al, 1999), although egl-8(md1971) was hypersensitive to levamisole in another study (Williams et al, 2007). EGL-30 modulates locomotion through its activity in cholinergic motor neurons (Lackner et al, 1999), but

UNC-73 RhoGEF2 isoform effects on locomotion involve neurons in addition to cholinergic motor neurons (Fig. 2). A final difference between the unc-73 RhoGEF2 mutants and the egl-30 pathway mutants is that an activated Gαs pathway suppresses unc-

73 mutant defects in locomotion rate, as well as aldicarb sensitivity, and axonal neuropeptide levels (Fig. 9, 10 and 12), yet Gαs pathway activation does not rescue egl-

30 mutant lethargy (Fig. 9C). UNC-73E is likely a direct downstream Gαq/EGL-30 effector with regard to egg laying and growth (Williams et al, 2007); however, the relationship between EGL-30 and UNC-73E in the modulation of locomotion is not as clear and may involve one or more additional factors upstream of UNC- 73 (“?” in Figure

22). UNC-73 RhoGEF2 activity may therefore have EGL-30 independent functions.

4.2 UNC-73 RhoGEF2 isoforms and the Gαs pathway

Gαs and its downstream factor cAMP regulate DCV and synaptic vesicle release in multiple cell types and are required for the modulation of synaptic plasticity in mammals through the modification of ion channels, the synaptic release machinery, or transcription factors (Brandon et al, 1997; Seino & Shibasaki, 2005). In this study it was observed that Gαs pathway activation suppresses unc-73 mutant defects in locomotion, drug sensitivity and neuropeptide level (Fig. 9, 10 and 12).

In the locomotion assay, expression of constitutively activated ACY-1 in both the nervous system and the body wall muscles was essential for the rescue of unc-73

RhoGEF2 mutant lethargic locomotion. Activated GSA-1(CA) or ACY-1(CA) expression in body wall muscles partially rescued the unc-73 RhoGEF2 mutant locomotion defect indicating Gαs pathway activation in muscles can at least partially bypass neuronal unc-73 RhoGEF2 defects (Fig. 9). Since complete rescue requires Gαs pathway activation in both the nervous system and body wall muscles suggests the Gαs pathway may play distinct roles in specific cells.

Mutants with loss of function defects in the Gαq pathway show a correlation between strong aldicarb resistance and a slow locomotory rate. The slow movement and aldicarb resistance are believed to be due to a decrease in neurotransmitter release at the neuromuscular junction (Lackner et al, 1999). Similarly, the faster locomotion rate of activated Gαs pathway mutants correlates with hypersensitivity to aldicarb (Fig. 10 and

12) (Schade et al, 2005). However, a direct correlation between the rate of locomotion and aldicarb resistance is not observed for all mutants including unc-73, rab-2, and unc-

31. These mutants all have a very slow rate of movement, but they are only weakly

85 resistant to aldicarb. If the weak aldicarb resistance suggests only a mild classical neurotransmitter release defect, it seems unlikely to account for the dramatically decreased locomotory rate of unc-73, rab-2 and unc-31 mutants. Furthermore, these mutants have a high level of hypersensitivity to levamisole that is not observed in Gαq pathway loss of function mutants. These observations suggest the UNC-73, RAB-2 and

UNC-31 proteins may regulate locomotion in a similar mechanism that is different from the Gαq pathway.

Constitutive activation of the Gαs pathway rescues the unc-73 slow locomotion defect, however, activated Gαs pathway mutants do not suppress egl-30/Gαq mutants. The

Gαq pathway is considered a central excitatory pathway in the regulation of classical SV- mediated neurotransmission, while the Gαs pathway also stimulates SV-mediated neurotransmission and requires the Gαq pathway for this stimulation. The Gαs pathway most likely functions to regulate large dense core vesicle mediated neurotransmission mechanisms in C. elegans since activation of the pathway can bypass defects in

CAPS/UNC-31 based DCV release (Charlie et al, 2006a; Zhou et al, 2007).

4.3 UNC-73 is involved in DCV-mediated secretion pathways

Dense core vesicles (DCVs) compose a distinct set of vesicles in the neuron.

Synaptic vesicles (SVs) are packed with classical neurotransmitters such as ACh, while

DCVs contains neuromodulators such as monoamines and neuropeptides. Both types of vesicles take part in the mechanisms of regulated secretion during neurotransmission.

UNC-73 RhoGEF2 isoforms play a role in the DCV-mediated mechanisms of neurotransmission as revealed by a neuropeptide assay (Fig. 8). Neuropeptide levels

86 indirectly measured by NLP-21::YFP derived fluorescence intensity decreased in the axons and coelomocytes of unc-73 RhoGEF2 mutants in comparison to wildtype. The unc-73 mutants also share similar phenotypes with rab-2 and unc-31, two genes known to be involved in the DCV-mediated mechanisms of neurotransmission (Fig. 13 and 14). In addition to the similar phenotypes, activated Gαs pathway mutants similarly rescued the slow moving defects of unc-73, unc-31 and rab-2 mutants.

Although the locomotory and drug sensitivity defects of unc-73 mutants were rescued by activated Gαs pathway mutants and transgenes, it could not be determined if

Gαs pathway activation rescues unc-73 neuropeptide release defects. Neuropeptide levels were altered in activated Gαs pathway mutants (Fig. 12). The activated Gαs pathway mutant gsa-1(ce81) has an increased neuropeptide-derived YFP level in cell bodies, but a lower level in coelomocytes compared to wildtype (Fig. 12). However, it is possible that the lower level of coelomocyte fluorescence in GSA-1 pathway mutants is due to coelomocyte uptake defects since gsa- 1(ce81) mutant coelomocytes took up more Texas

Red conjugated BSA than wildtype in an assay for coelomocyte function (Fig. 12). The decreased NLP-21::YFP derived fluorescence in ce81 coelomocytes may therefore represent either a lower level of neuropeptide release from neurons, or perhaps a faster rate of YFP turnover in the coelomocytes of ce81 mutants. The increase in ce81 coelomocyte TR-BSA levels indicates that the ce81 mutation does, in fact, affect coelomocyte activity. The increased NLP-21::YFP derived fluorescence in ce81 cell bodies may indicate more neuropeptide is produced in ce81 mutants, but this possibility does not correlate with the dramatic decrease in YFP fluorescence observed in ce81 coelomocytes. If there is indeed more neuropeptide produced in ce81 mutants, then more

NLP-21::YFP derived fluorescence would be expected in coelomocytes, although there may be changes in ce81 coelomocyte function as mentioned above. Another possibility is that ce81 defects in neuropeptide transport out of the cell body results in an accumulation of NLP::YFP derived fluorescence in the cell body and a corresponding decrease of neuropeptide release from axons. However, vesicle transport defects have not been reported in ce81 mutants so far.

It is even a question as to whether the YFP fluorescence measured in this neuropeptide release assay correlates to the level of actual NLP-21 neuropeptide present in the coelomocytes. As there may be unknown changes to the secretory pathways in the mutants that were examined using the neuropeptide release assay, caution is required in interpreting the cell body, axon and coelomocyte fluorescence levels observed in this recently developed assay. Electron microscopy could be used in future experiments to directly examine if unc-73 RhoGEF2 mutant neurons have alterations in the number or distribution of DCVs.

4.4 unc-73 RhoGEF2 mutant aldicarb and levamisole sensitivities

unc-73 RhoGEF2 mutants have a distinct phenotype being resistant to aldicarb yet hypersensitive to levamisole. There are only fifteen genes reported as Ric (resistant to aldicarb) with a normal or hypersensitive levamisole response. Among these mutants, unc-75 (CELF), unc-18 (Munc18), snt-1 (Synaptotagmin) and unc-41 (Stonin) have significant levamisole hypersensitivity similar to unc-73 RhoGEF2 mutants (Miller et al,

1996). UNC-75, a putative pre-mRNA splicing factor, modulates GABAergic and cholinergic neurotransmission (Loria et al, 2003). UNC-18 interacts with the SNARE

88 complex and functions in synaptic vesicle exocytosis (Dulubova et al, 2007; McEwen &

Kaplan, 2008). SNT-1 is required for synaptic vesicle exocytosis and clathrin-mediated endocytosis (Jorgensen et al, 1995; Nonet et al, 1993). UNC-41 binds to SNT-1 and mediates SNT-1 sorting during endocytosis (Jung et al, 2007).

The aldicarb sensitivity assay measurements presumably reflect the accumulation of ACh at all cholinergic synapses in the organism and differences from wildtype may involve either pre-synaptic ACh release defects or post-synaptic ACh signaling defects.

The levamisole sensitivity assay tests the response of levamisole-sensitive ACh receptors, which contain UNC-29, UNC-38, UNC-63, LEV-1, and LEV-8 subunits (Rand, 2007;

Richmond & Jorgensen, 1999). These levamisole-sensitive receptors in C. elegans predominantly localize to the body wall muscles, but UNC-29 and UNC-63 are found in motoneurons as well (Culetto et al, 2004; Gally et al, 2004). Therefore resistance or hypersensitivity to levamisole in a particular mutant may represent changes to acetylcholine receptors at the neuromuscular junction, at neuron-neuron synapses or at both types of synapses. The unc-73 drug sensitivity phenotypes suggest UNC-73

RhoGEF2 activity may have a role in the regulation of receptor strength, possibly through mechanisms involving the secretory pathway. Future examination of motoneuron and body wall muscle electrophysiology should provide insight into the unc-73 RhoGEF2 mutant levamisole and aldicarb sensitivity phenotypes.

RhoA regulates receptor dynamics in endocrine cells, which is consistent with my observations of RhoA pathway function in neurons. Constitutively active RhoA inhibits clathrin-mediated endocytosis of epidermal growth factor (EGF) receptor (Lamaze et al,

1996). Rho inhibition of EGF receptor endocytosis is mediated by phosphorylation of

89 endophilinA1 by the Rho effector ROCK (Kaneko et al, 2005). The RhoA GEF involved in EGF receptor regulation was identified as Vav2, which delays EGF receptor internalization and degradation (Thalappilly et al, 2010). Taken together, in the regulation of EGF receptor endocytosis, RhoA activated by Vav2 turns on ROCK to phosphorylate endophilinA1, which disassociates from the EGF receptor complex upon phosphorylation thus decreasing the internalization of EGF receptor.

This series of studies on the RhoA signaling pathway regulating EGF receptor dynamics indicates the possibility of similar regulation of neuronal receptors. Indeed,

Oligophrenin RhoGAP mutations result in elevated RhoA/ROCK activity, which in turn reduces both synaptic vesicle endocytosis and AMPA receptor internalization and may be the cause of X-linked mental retardation in humans (Khelfaoui et al, 2009; Nadif Kasri et al, 2009). How UNC-73/Rho signaling regulates levamisole sensitivity will be interesting to investigate in future studies. It is possible UNC-73/Rho signaling may regulate levamisole sensitive ACh receptors in multiple neuronal types to alter cholinergic neurotransmission or UNC-73/Rho signaling may regulate muscle levamisole sensitive

ACh receptor strength through unidentified neuromodulators released from neurons.

4.5 UNC-73 RhoGEF2 isoform function and GLR-1 localization

GLR-1, an AMPA glutamate receptor is mislocalized in unc-73 RhoGEF2 and lin-2 mutants (Fig. 19). GLR-1::GFP distribution in the mutant PVQ neurons is often visible as a bright fluorescent spot located at the start of the axon close to the cell body.

The altered localization of GLR-1::GFP suggests a receptor trafficking defect. The receptor may not be transported properly down the axon after leaving the cell body, and

90 may instead accumulate at the beginning of the axon. This accumulation could be the result of over-expressed GLR-1, however, similar accumulations were not observed in wild-type and lin-10 animals, which leads us to believe altered GLR-1 localization is an interesting unc-73 mutant phenotype requiring further investigation.

LIN-2 and LIN-10 are components of a conserved protein complex involved in receptor localization (Kaech et al, 1998). LIN-2 and LIN-10 interact with each other in C. elegans, and the GLR-1 glutamate receptor is mislocalized along motoneuron axons in lin-10 mutants (Glodowski et al, 2005). However, the altered GLR-1 localization observed in my experiment only appeared in lin-2 and unc-73 mutants, but not lin-10 animals. These results suggest that UNC-73 may interact with LIN-2 in a receptor regulation pathway functioning independent of the LIN-2/LIN-10 complex.

The observed receptor accumulation suggests a defect in the mechanisms of receptor sorting and/or transport. Potential future experiments include an examination for changes in the secretory and endocytic pathways of unc-73 mutants using fluorescent markers specific to distinct sub-cellular locations within these pathways. To address the hypothesis that UNC-73 RhoGEF-2 isoforms function in the secretory pathway, the intensity levels, size and distribution patterns of fluorescently labeled markers of selected compartments of the secretory pathway in neurons of unc-73 RhoGEF2 mutant animals would be examined using confocal microscopy. Changes in the gross structure of one or more secretory compartments may be observed in unc-73 RhoGEF2 mutant animals.

Possible fluorescently labeled intracellular markers of the secretory pathway to be used include ssVenus::KDEL for the ER (Chun et al, 2008), MANS II for the Golgi

(Sumakovic et al, 2009), Syntaxin-5 for the trans-Golgi-network (Chun et al, 2008), and

2xFYVE for endosomes (Sumakovic et al, 2009).

The UNC-73 isoform cellular distribution patterns were examined using functional GFP tagged UNC-73 isoform specific constructs (Steven et al, 2005), however, the subcellular localization of the UNC-73 isoforms was not carefully examined. The

UNC-73E isoform is evenly distributed in the cytoplasmic region of cells including the axons and cell bodies of neurons, which might be expected for this isoform, which lacks the protein domains found in other isoforms offering greater potential to mediate localization to specific regions in the cell. Other UNC-73 isoforms, which have one or more sec14, spectrin, PH, SH3, immunoglobin or fibronectin type III domains, were often noticeably punctate in the their subcellular distributions, including the cell bodies of neurons (Robert Steven, personal communication). The subcellular locations of GFP tagged UNC-73 isoforms could be identified by co-localization with the same subcellular secretory pathway marker proteins described above. Positive identification of UNC-73 isoforms in particular locations of the secretory pathway would substantiate the hypothesis that UNC-73 is required for the sorting or transport of neurotransmitter receptors in neurons.

4.6 A model of UNC-73/Rho signaling in the regulation of neurotransmission

A working model of UNC-73 RhoGEF2 function is diagrammed in Figure 22. C. elegans locomotion is regulated by classical neurotransmitters, such as ACh, released from synaptic vesicles and neuromodulators released from dense core vesicles. Synaptic vesicle signaling is regulated by two Gα pathways: the Gαq pathway (EGL-30EGL-

8DAG) and the Gα12 pathway (GPA-12RHGF-1RHO-1DAG) (Lackner et al,

1999; McMullan et al, 2006). Dense core vesicle mediated signaling can regulate locomotion “directly” by interacting with receptors on muscles to modulate muscle activity (Nelson et al, 1998), or “indirectly” by modulating SV signaling originating from motor neurons or interneurons further upstream. The Gαs pathway may regulate locomotion “directly” by functioning to modulate muscle electrophysiology possibly downstream of a neuropeptide receptor (Seino & Shibasaki, 2005) or “indirectly” by modulating either SV or DCV signaling from neurons (Charlie et al, 2006a; Schade et al,

2005; Zhou et al, 2007).

UNC-73 RhoGEF2 isoforms and the Rho signaling pathway regulate locomotion by at least two possible mechanisms: 1) As a direct downstream effector of EGL-30/Gαq acting in parallel to EGL-8/PLCβ to regulate SV signaling as revealed by altered unc-73

RhoGEF2 mutant responses to aldicarb and levamisole. 2) Through the regulation of

DCV-mediated neuromodulator signaling as revealed by unc-73 RhoGEF2 mutant defects in neuropeptide levels monitored using NLP-21::YFP. An activated Gαs pathway can compensate for unc-73 RhoGEF2 mutant defects, most likely at multiple levels since

Gαs pathway activation is required in both neurons and muscles for the highest level of rescue.

In conclusion, my study has extended the knowledge of Rho signaling mechanisms in the process of neuromodulatory regulation of locomotory behavior and highlighted the importance of C. elegans as a model system in the study of neuroscience.

Figure 22. The working model of UNC-73 RhoGEF2 isoforms and Rho signaling in neurotransmission. The UNC-73/Rho signaling pathway interacts with heterotrimeric Gα pathways regulating neurotransmission through multiple mechanisms. One mechanism of Rho activation of SV signaling is through the interaction and inhibition of DGK-1 to increase synaptic vesicle signaling under control of the Gα12/RHGF-1 pathway. Rho also regulates dense core vesicle signaling by an unknown mechanism to affect locomotion under the activation of the Gαq/UNC-73 pathway. Aside from the direct interaction of UNC-73 with EGL-30, other unknown upstream factor(s) regulating UNC-73/Rho likely exist as well. An activated Gαs pathway compensates for multiple unc-73 mutant defects but the mechanism is not known. Additional explanations are provided in the text. Blue boxes: heterotrimeric Gα proteins, Red boxes: Rho GTPase; Solid arrows: direct interaction; Dashed arrow: indirect or unknown mechanism of interaction.

References

Alam MR, Johnson RC, Darlington DN, Hand TA, Mains RE, Eipper BA (1997) Kalirin, a cytosolic protein with spectrin-like and GDP/GTP exchange factor-like domains that interacts with peptidylglycine alpha-amidating monooxygenase, an integral membrane peptide-processing enzyme. J Biol Chem 272(19): 12667-12675

Albeg A, Smith CJ, Chatzigeorgiou M, Feitelson DG, Hall DH, Schafer WR, Miller DM, 3rd, Treinin M (2011) C. elegans multi-dendritic sensory neurons: morphology and function. Mol Cell Neurosci 46(1): 308-317

Alfonso A, Grundahl K, Duerr JS, Han HP, Rand JB (1993) The Caenorhabditis elegans unc-17 gene: a putative vesicular acetylcholine transporter. Science 261(5121): 617-619

Aravamudan B, Fergestad T, Davis WS, Rodesch CK, Broadie K (1999) Drosophila UNC-13 is essential for synaptic transmission. Nat Neurosci 2(11): 965-971

Augustin I, Rosenmund C, Sudhof TC, Brose N (1999) Munc13-1 is essential for fusion competence of glutamatergic synaptic vesicles. Nature 400(6743): 457-461

Avery L, Bargmann CI, Horvitz HR (1993) The Caenorhabditis elegans unc-31 gene affects multiple nervous system-controlled functions. Genetics 134(2): 455-464

Backer S, Hidalgo-Sanchez M, Offner N, Portales-Casamar E, Debant A, Fort P, Gauthier-Rouviere C, Bloch-Gallego E (2007) Trio controls the mature organization of neuronal clusters in the hindbrain. J Neurosci 27(39): 10323-10332

Bateman J, Van Vactor D (2001) The Trio family of guanine-nucleotide-exchange factors: regulators of axon guidance. J Cell Sci 114(Pt 11): 1973-1980

Bellanger JM, Astier C, Sardet C, Ohta Y, Stossel TP, Debant A (2000) The Rac1- and RhoG-specific GEF domain of Trio targets filamin to remodel cytoskeletal actin. Nat Cell Biol 2(12): 888-892

Bessou C, Giugia JB, Franks CJ, Holden-Dye L, Segalat L (1998) Mutations in the Caenorhabditis elegans dystrophin-like gene dys-1 lead to hyperactivity and suggest a link with cholinergic transmission. Neurogenetics 2(1): 61-72

Blangy A, Vignal E, Schmidt S, Debant A, Gauthier-Rouviere C, Fort P (2000) TrioGEF1 controls Rac- and Cdc42-dependent cell structures through the direct activation of rhoG. J Cell Sci 113 ( Pt 4): 729-739

Boda B, Dubos A, Muller D (2010) Signaling mechanisms regulating synapse formation and function in mental retardation. Curr Opin Neurobiol 20(4): 519-527

Boriack-Sjodin PA, Margarit SM, Bar-Sagi D, Kuriyan J (1998) The structural basis of the activation of Ras by Sos. Nature 394(6691): 337-343

Bos JL, Rehmann H, Wittinghofer A (2007) GEFs and GAPs: critical elements in the control of small G proteins. Cell 129(5): 865-877

Brandon MA, Voglmaier S, Siddiqi AA (1997) Molecular characterization of a Dictyostelium G-protein alpha-subunit required for development. Gene 200(1-2): 99-105

Brezina V (2010) Beyond the wiring diagram: signalling through complex neuromodulator networks. Philos Trans R Soc Lond B Biol Sci 365(1551): 2363-2374

Briancon-Marjollet A, Ghogha A, Nawabi H, Triki I, Auziol C, Fromont S, Piche C, Enslen H, Chebli K, Cloutier JF, Castellani V, Debant A, Lamarche-Vane N (2008) Trio mediates netrin-1-induced Rac1 activation in axon outgrowth and guidance. Mol Cell Biol 28(7): 2314-2323

Brundage L, Avery L, Katz A, Kim UJ, Mendel JE, Sternberg PW, Simon MI (1996) Mutations in a C. elegans Gqalpha gene disrupt movement, egg laying, and viability. Neuron 16(5): 999-1009

Cahill ME, Xie Z, Day M, Photowala H, Barbolina MV, Miller CA, Weiss C, Radulovic J, Sweatt JD, Disterhoft JF, Surmeier DJ, Penzes P (2009) Kalirin regulates cortical spine morphogenesis and disease-related behavioral phenotypes. Proc Natl Acad Sci U S A 106(31): 13058-13063

Caruana G (2002) Genetic studies define MAGUK proteins as regulators of epithelial cell polarity. Int J Dev Biol 46(4): 511-518

Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Green fluorescent protein as a marker for gene expression. Science 263(5148): 802-805

Charlie NK, Schade MA, Thomure AM, Miller KG (2006a) Presynaptic UNC-31 (CAPS) is required to activate the G alpha(s) pathway of the Caenorhabditis elegans synaptic signaling network. Genetics 172(2): 943-961

Charlie NK, Thomure AM, Schade MA, Miller KG (2006b) The Dunce cAMP phosphodiesterase PDE-4 negatively regulates G alpha(s)-dependent and G alpha(s)-

96 independent cAMP pools in the Caenorhabditis elegans synaptic signaling network. Genetics 173(1): 111-130

Chen K, Featherstone DE (2011) Pre and postsynaptic roles for Drosophila CASK. Mol Cell Neurosci 48(2): 171-182

Chen W, Lim L (1994) The Caenorhabditis elegans small GTP-binding protein RhoA is enriched in the nerve ring and sensory neurons during larval development. J Biol Chem 269(51): 32394-32404

Chin J, Angers A, Cleary LJ, Eskin A, Byrne JH (2002) Transforming growth factor beta1 alters synapsin distribution and modulates synaptic depression in Aplysia. J Neurosci 22(9): RC220

Chun DK, McEwen JM, Burbea M, Kaplan JM (2008) UNC-108/Rab2 regulates postendocytic trafficking in Caenorhabditis elegans. Mol Biol Cell 19(7): 2682-2695

Cohen E, Yemini E, Schafer W, Feitelson DG, Treinin M (2012) Locomotion analysis identifies roles of mechanosensory neurons in governing locomotion dynamics of C. elegans. J Exp Biol

Colavita A, Krishna S, Zheng H, Padgett RW, Culotti JG (1998) Pioneer axon guidance by UNC-129, a C. elegans TGF-beta. Science 281(5377): 706-709

Culetto E, Baylis HA, Richmond JE, Jones AK, Fleming JT, Squire MD, Lewis JA, Sattelle DB (2004) The Caenorhabditis elegans unc-63 gene encodes a levamisole- sensitive nicotinic acetylcholine receptor alpha subunit. J Biol Chem 279(41): 42476- 42483

Debant A, Serra-Pages C, Seipel K, O'Brien S, Tang M, Park SH, Streuli M (1996) The multidomain protein Trio binds the LAR transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate rac-specific and rho-specific guanine nucleotide exchange factor domains. Proc Natl Acad Sci U S A 93(11): 5466-5471

Doussau F, Gasman S, Humeau Y, Vitiello F, Popoff M, Boquet P, Bader MF, Poulain B (2000) A Rho-related GTPase is involved in Ca(2+)-dependent neurotransmitter exocytosis. J Biol Chem 275(11): 7764-7770

Dulubova I, Khvotchev M, Liu S, Huryeva I, Sudhof TC, Rizo J (2007) Munc18-1 binds directly to the neuronal SNARE complex. Proc Natl Acad Sci U S A 104(8): 2697-2702

Dybbs M, Ngai J, Kaplan JM (2005) Using microarrays to facilitate positional cloning: identification of tomosyn as an inhibitor of neurosecretion. PLoS Genet 1(1): 6-16

Eastman C, Horvitz HR, Jin Y (1999) Coordinated transcriptional regulation of the unc- 25 glutamic acid decarboxylase and the unc-47 GABA vesicular transporter by the Caenorhabditis elegans UNC-30 homeodomain protein. J Neurosci 19(15): 6225-6234

Edwards SL, Charlie NK, Richmond JE, Hegermann J, Eimer S, Miller KG (2009) Impaired dense core vesicle maturation in Caenorhabditis elegans mutants lacking Rab2. J Cell Biol 186(6): 881-895

Esposito G, Di Schiavi E, Bergamasco C, Bazzicalupo P (2007) Efficient and cell specific knock-down of gene function in targeted C. elegans neurons. Gene 395(1-2): 170-176

Estevez M, Estevez AO, Cowie RH, Gardner KL (2004) The voltage-gated calcium channel UNC-2 is involved in stress-mediated regulation of tryptophan hydroxylase. J Neurochem 88(1): 102-113

Farver J (2009) Defining Roles for the Caenorhabditis elegans UNC-73 Isoforms in Neurotransmission and Muscle Function. Unpublished master thesis, University of Toledo, Toledo, USA

Ferraro F, Ma XM, Sobota JA, Eipper BA, Mains RE (2007) Kalirin/Trio Rho guanine nucleotide exchange factors regulate a novel step in secretory granule maturation. Mol Biol Cell 18(12): 4813-4825

Fong SW, McLennan IS, McIntyre A, Reid J, Shennan KI, Bewick GS (2010) TGF-beta2 alters the characteristics of the neuromuscular junction by regulating presynaptic quantal size. Proc Natl Acad Sci U S A 107(30): 13515-13519

Forsthoefel DJ, Liebl EC, Kolodziej PA, Seeger MA (2005) The Abelson tyrosine kinase, the Trio GEF and Enabled interact with the Netrin receptor Frazzled in Drosophila. Development 132(8): 1983-1994

Francis R, Waterston RH (1991) Muscle cell attachment in Caenorhabditis elegans. J Cell Biol 114(3): 465-479

Gally C, Eimer S, Richmond JE, Bessereau JL (2004) A transmembrane protein required for acetylcholine receptor clustering in Caenorhabditis elegans. Nature 431(7008): 578- 582

Gasman S, Chasserot-Golaz S, Hubert P, Aunis D, Bader MF (1998) Identification of a potential effector pathway for the trimeric Go protein associated with secretory granules. Go stimulates a granule-bound phosphatidylinositol 4-kinase by activating RhoA in chromaffin cells. J Biol Chem 273(27): 16913-16920

Gasman S, Chasserot-Golaz S, Popoff MR, Aunis D, Bader MF (1999) Involvement of Rho GTPases in calcium-regulated exocytosis from adrenal chromaffin cells. J Cell Sci 112 ( Pt 24): 4763-4771

Giugia J, Gieseler K, Arpagaus M, Segalat L (1999) Mutations in the dystrophin-like dys- 1 gene of Caenorhabditis elegans result in reduced acetylcholinesterase activity. FEBS Lett 463(3): 270-272

Glodowski DR, Wright T, Martinowich K, Chang HC, Beach D, Rongo C (2005) Distinct LIN-10 domains are required for its neuronal function, its epithelial function, and its synaptic localization. Mol Biol Cell 16(3): 1417-1426

Goodman MB (2006) Mechanosensation. WormBook

Govek EE, Newey SE, Van Aelst L (2005) The role of the Rho GTPases in neuronal development. Genes Dev 19(1): 1-49

Gracheva EO, Burdina AO, Touroutine D, Berthelot-Grosjean M, Parekh H, Richmond JE (2007) Tomosyn negatively regulates CAPS-dependent peptide release at Caenorhabditis elegans synapses. J Neurosci 27(38): 10176-10184

Haenggi T, Fritschy JM (2006) Role of dystrophin and utrophin for assembly and function of the dystrophin glycoprotein complex in non-muscle tissue. Cell Mol Life Sci 63(14): 1614-1631

Hall A, Lalli G (2010) Rho and Ras GTPases in axon growth, guidance, and branching. Cold Spring Harb Perspect Biol 2(2): a001818

Hallam SJ, Jin Y (1998) lin-14 regulates the timing of synaptic remodelling in Caenorhabditis elegans. Nature 395(6697): 78-82

Hammarlund M, Watanabe S, Schuske K, Jorgensen EM (2008) CAPS and syntaxin dock dense core vesicles to the plasma membrane in neurons. J Cell Biol 180(3): 483-491

Hansel DE, Quinones ME, Ronnett GV, Eipper BA (2001) Kalirin, a GDP/GTP exchange factor of the Dbl family, is localized to nerve, muscle, and endocrine tissue during embryonic rat development. J Histochem Cytochem 49(7): 833-844

Harris G, Mills H, Wragg R, Hapiak V, Castelletto M, Korchnak A, Komuniecki RW (2010) The monoaminergic modulation of sensory-mediated aversive responses in Caenorhabditis elegans requires glutamatergic/peptidergic cotransmission. J Neurosci 30(23): 7889-7899

Hart (ed.) AC (2006) Behavior. WormBook

Hart AC, Sims S, Kaplan JM (1995) Synaptic code for sensory modalities revealed by C. elegans GLR-1 glutamate receptor. Nature 378(6552): 82-85

Hart MJ, Jiang X, Kozasa T, Roscoe W, Singer WD, Gilman AG, Sternweis PC, Bollag G (1998) Direct stimulation of the guanine nucleotide exchange activity of p115 RhoGEF by Galpha13. Science 280(5372): 2112-2114

Haycraft CJ, Swoboda P, Taulman PD, Thomas JH, Yoder BK (2001) The C. elegans homolog of the murine cystic kidney disease gene Tg737 functions in a ciliogenic pathway and is disrupted in osm-5 mutant worms. Development 128(9): 1493-1505

Heasman SJ, Ridley AJ (2008) Mammalian Rho GTPases: new insights into their functions from in vivo studies. Nat Rev Mol Cell Biol 9(9): 690-701

Hedgecock EM, Culotti JG, Thomson JN, Perkins LA (1985) Axonal guidance mutants of Caenorhabditis elegans identified by filling sensory neurons with fluorescein dyes. Dev Biol 111(1): 158-170

Hiley E, McMullan R, Nurrish SJ (2006) The Galpha12-RGS RhoGEF-RhoA signalling pathway regulates neurotransmitter release in C. elegans. EMBO J 25(24): 5884-5895

Hodgkin J (2005) Introduction to genetics and genomics. WormBook

Horgan CP, Hanscom SR, Jolly RS, Futter CE, McCaffrey MW (2010) Rab11-FIP3 binds dynein light intermediate chain 2 and its overexpression fragments the Golgi complex. Biochem Biophys Res Commun 394(2): 387-392

Horgan CP, McCaffrey MW (2009) The dynamic Rab11-FIPs. Biochem Soc Trans 37(Pt 5): 1032-1036

Horgan CP, Oleksy A, Zhdanov AV, Lall PY, White IJ, Khan AR, Futter CE, McCaffrey JG, McCaffrey MW (2007) Rab11-FIP3 is critical for the structural integrity of the endosomal recycling compartment. Traffic 8(4): 414-430

Hoskins R, Hajnal AF, Harp SA, Kim SK (1996) The C. elegans vulval induction gene lin-2 encodes a member of the MAGUK family of cell junction proteins. Development 122(1): 97-111

Houssa B, de Widt J, Kranenburg O, Moolenaar WH, van Blitterswijk WJ (1999) Diacylglycerol kinase theta binds to and is negatively regulated by active RhoA. J Biol Chem 274(11): 6820-6822

Hu S, Pawson T, Steven RM (2011) UNC-73/trio RhoGEF-2 activity modulates Caenorhabditis elegans motility through changes in neurotransmitter signaling upstream of the GSA-1/Galphas pathway. Genetics 189(1): 137-151

100

Husson SJ, Clynen E, Baggerman G, Janssen T, Schoofs L (2006) Defective processing of neuropeptide precursors in Caenorhabditis elegans lacking proprotein convertase 2 (KPC-2/EGL-3): mutant analysis by mass spectrometry. J Neurochem 98(6): 1999-2012

Husson SJ, Mertens I, Janssen T, Lindemans M, Schoofs L (2007) Neuropeptidergic signaling in the nematode Caenorhabditis elegans. Prog Neurobiol 82(1): 33-55

Jacob TC, Kaplan JM (2003) The EGL-21 carboxypeptidase E facilitates acetylcholine release at Caenorhabditis elegans neuromuscular junctions. J Neurosci 23(6): 2122-2130

Jaffe AB, Hall A (2005) Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol 21: 247-269

Jantsch-Plunger V, Gonczy P, Romano A, Schnabel H, Hamill D, Schnabel R, Hyman AA, Glotzer M (2000) CYK-4: A Rho family gtpase activating protein (GAP) required for central spindle formation and cytokinesis. J Cell Biol 149(7): 1391-1404

Jeyifous O, Waites CL, Specht CG, Fujisawa S, Schubert M, Lin EI, Marshall J, Aoki C, de Silva T, Montgomery JM, Garner CC, Green WN (2009) SAP97 and CASK mediate sorting of NMDA receptors through a previously unknown secretory pathway. Nat Neurosci 12(8): 1011-1019

Jhamandas JH, Simonin F, Bourguignon JJ, Harris KH (2007) Neuropeptide FF and neuropeptide VF inhibit GABAergic neurotransmission in parvocellular neurons of the rat hypothalamic paraventricular nucleus. Am J Physiol Regul Integr Comp Physiol 292(5): R1872-1880

Jing J, Tarbutton E, Wilson G, Prekeris R (2009) Rab11-FIP3 is a Rab11-binding protein that regulates breast cancer cell motility by modulating the actin cytoskeleton. Eur J Cell Biol 88(6): 325-341

Jockusch WJ, Speidel D, Sigler A, Sorensen JB, Varoqueaux F, Rhee JS, Brose N (2007) CAPS-1 and CAPS-2 are essential synaptic vesicle priming proteins. Cell 131(4): 796- 808

Jorgensen EM (2005) GABA. WormBook

Jorgensen EM, Hartwieg E, Schuske K, Nonet ML, Jin Y, Horvitz HR (1995) Defective recycling of synaptic vesicles in synaptotagmin mutants of Caenorhabditis elegans. Nature 378(6553): 196-199

Jung N, Wienisch M, Gu M, Rand JB, Muller SL, Krause G, Jorgensen EM, Klingauf J, Haucke V (2007) Molecular basis of synaptic vesicle cargo recognition by the endocytic sorting adaptor stonin 2. J Cell Biol 179(7): 1497-1510

101

Kaech SM, Whitfield CW, Kim SK (1998) The LIN-2/LIN-7/LIN-10 complex mediates basolateral membrane localization of the C. elegans EGF receptor LET-23 in vulval epithelial cells. Cell 94(6): 761-771

Kamath RS, Ahringer J (2003) Genome-wide RNAi screening in Caenorhabditis elegans. Methods 30(4): 313-321

Kaneko T, Maeda A, Takefuji M, Aoyama H, Nakayama M, Kawabata S, Kawano Y, Iwamatsu A, Amano M, Kaibuchi K (2005) Rho mediates endocytosis of epidermal growth factor receptor through phosphorylation of endophilin A1 by Rho-kinase. Genes Cells 10(10): 973-987

Kass J, Jacob TC, Kim P, Kaplan JM (2001) The EGL-3 proprotein convertase regulates mechanosensory responses of Caenorhabditis elegans. J Neurosci 21(23): 9265-9272

Khelfaoui M, Pavlowsky A, Powell AD, Valnegri P, Cheong KW, Blandin Y, Passafaro M, Jefferys JG, Chelly J, Billuart P (2009) Inhibition of RhoA pathway rescues the endocytosis defects in Oligophrenin1 mouse model of mental retardation. Hum Mol Genet 18(14): 2575-2583

Kim K, Li C (2004) Expression and regulation of an FMRFamide-related neuropeptide gene family in Caenorhabditis elegans. J Comp Neurol 475(4): 540-550

Kiraly DD, Lemtiri-Chlieh F, Levine ES, Mains RE, Eipper BA (2011) Kalirin binds the NR2B subunit of the NMDA receptor, altering its synaptic localization and function. J Neurosci 31(35): 12554-12565

Korswagen HC, Park JH, Ohshima Y, Plasterk RH (1997) An activating mutation in a Caenorhabditis elegans Gs protein induces neural degeneration. Genes Dev 11(12): 1493- 1503

Lackner MR, Nurrish SJ, Kaplan JM (1999) Facilitation of synaptic transmission by EGL-30 Gqalpha and EGL-8 PLCbeta: DAG binding to UNC-13 is required to stimulate acetylcholine release. Neuron 24(2): 335-346

Lamaze C, Chuang TH, Terlecky LJ, Bokoch GM, Schmid SL (1996) Regulation of receptor-mediated endocytosis by Rho and Rac. Nature 382(6587): 177-179

Lee JW, Yeo SG, Kang BH, Lee HK, Kim JW, Lee SH, Kim KS, Cheon DS (2012) Echovirus 30 Induced Neuronal Cell Death through TRIO-RhoA Signaling Activation. PLoS One 7(5): e36656

Lee RY, Sawin ER, Chalfie M, Horvitz HR, Avery L (1999) EAT-4, a homolog of a mammalian sodium-dependent inorganic phosphate cotransporter, is necessary for glutamatergic neurotransmission in caenorhabditis elegans. J Neurosci 19(1): 159-167

102

Lemtiri-Chlieh F, Zhao L, Kiraly DD, Eipper BA, Mains RE, Levine ES (2011) Kalirin-7 is necessary for normal NMDA receptor-dependent synaptic plasticity. BMC Neurosci 12: 126

Lin XG, Ming M, Chen MR, Niu WP, Zhang YD, Liu B, Jiu YM, Yu JW, Xu T, Wu ZX (2010) UNC-31/CAPS docks and primes dense core vesicles in C. elegans neurons. Biochem Biophys Res Commun 397(3): 526-531

Liu X, Wang H, Eberstadt M, Schnuchel A, Olejniczak ET, Meadows RP, Schkeryantz JM, Janowick DA, Harlan JE, Harris EA, Staunton DE, Fesik SW (1998) NMR structure and mutagenesis of the N-terminal Dbl homology domain of the nucleotide exchange factor Trio. Cell 95(2): 269-277

Loria PM, Duke A, Rand JB, Hobert O (2003) Two neuronal, nuclear-localized RNA binding proteins involved in synaptic transmission. Curr Biol 13(15): 1317-1323

Lou X, Korogod N, Brose N, Schneggenburger R (2008) Phorbol esters modulate spontaneous and Ca2+-evoked transmitter release via acting on both Munc13 and protein kinase C. J Neurosci 28(33): 8257-8267

Lundquist EA (2006) Small GTPases. WormBook

Lutz S, Freichel-Blomquist A, Rumenapp U, Schmidt M, Jakobs KH, Wieland T (2004) p63RhoGEF and GEFT are Rho-specific guanine nucleotide exchange factors encoded by the same gene. Naunyn Schmiedebergs Arch Pharmacol 369(5): 540-546

Ma AD, Brass LF, Abrams CS (1997) Pleckstrin associates with plasma membranes and induces the formation of membrane projections: requirements for phosphorylation and the NH2-terminal PH domain. J Cell Biol 136(5): 1071-1079

Ma XM, Huang J, Wang Y, Eipper BA, Mains RE (2003) Kalirin, a multifunctional Rho guanine nucleotide exchange factor, is necessary for maintenance of hippocampal pyramidal neuron dendrites and dendritic spines. J Neurosci 23(33): 10593-10603

Maduro M, Pilgrim D (1995) Identification and cloning of unc-119, a gene expressed in the Caenorhabditis elegans nervous system. Genetics 141(3): 977-988

Mahoney TR, Luo S, Nonet ML (2006) Analysis of synaptic transmission in Caenorhabditis elegans using an aldicarb-sensitivity assay. Nat Protoc 1(4): 1772-1777

Mains RE, Alam MR, Johnson RC, Darlington DN, Back N, Hand TA, Eipper BA (1999) Kalirin, a multifunctional PAM COOH-terminal domain interactor protein, affects cytoskeletal organization and ACTH secretion from AtT-20 cells. J Biol Chem 274(5): 2929-2937

103

Mandela P, Ma XM (2012) Kalirin, a key player in synapse formation, is implicated in human diseases. Neural Plast 2012: 728161

Maricq AV, Peckol E, Driscoll M, Bargmann CI (1995) Mechanosensory signalling in C. elegans mediated by the GLR-1 glutamate receptor. Nature 378(6552): 78-81

McEwen JM, Kaplan JM (2008) UNC-18 promotes both the anterograde trafficking and synaptic function of syntaxin. Mol Biol Cell 19(9): 3836-3846

McIntire SL, Reimer RJ, Schuske K, Edwards RH, Jorgensen EM (1997) Identification and characterization of the vesicular GABA transporter. Nature 389(6653): 870-876

McMullan R, Anderson A, Nurrish S (2012) Behavioral and immune responses to infection require Galphaq- RhoA signaling in C. elegans. PLoS Pathog 8(2): e1002530

McMullan R, Hiley E, Morrison P, Nurrish SJ (2006) Rho is a presynaptic activator of neurotransmitter release at pre-existing synapses in C. elegans. Genes Dev 20(1): 65-76

McMullan R, Nurrish SJ (2011) The RHO-1 RhoGTPase modulates fertility and multiple behaviors in adult C. elegans. PLoS One 6(2): e17265

Miller DM, 3rd, Ortiz I, Berliner GC, Epstein HF (1983) Differential localization of two myosins within nematode thick filaments. Cell 34(2): 477-490

Miller KG, Alfonso A, Nguyen M, Crowell JA, Johnson CD, Rand JB (1996) A genetic selection for Caenorhabditis elegans synaptic transmission mutants. Proc Natl Acad Sci U S A 93(22): 12593-12598

Miller KG, Emerson MD, Rand JB (1999) Goalpha and diacylglycerol kinase negatively regulate the Gqalpha pathway in C. elegans. Neuron 24(2): 323-333

Momboisse F, Houy S, Ory S, Calco V, Bader MF, Gasman S (2011) How important are Rho GTPases in neurosecretion? J Neurochem 117(4): 623-631

Moorman C, Plasterk RH (2002) Functional characterization of the adenylyl cyclase gene sgs-1 by analysis of a mutational spectrum in Caenorhabditis elegans. Genetics 161(1): 133-142

Nadif Kasri N, Nakano-Kobayashi A, Malinow R, Li B, Van Aelst L (2009) The Rho- linked mental retardation protein oligophrenin-1 controls synapse maturation and plasticity by stabilizing AMPA receptors. Genes Dev 23(11): 1289-1302

Nass R, Hall DH, Miller DM, 3rd, Blakely RD (2002) Neurotoxin-induced degeneration of dopamine neurons in Caenorhabditis elegans. Proc Natl Acad Sci U S A 99(5): 3264- 3269

104

Nelson LS, Rosoff ML, Li C (1998) Disruption of a neuropeptide gene, flp-1, causes multiple behavioral defects in Caenorhabditis elegans. Science 281(5383): 1686-1690

Nicholson DA, Cahill ME, Tulisiak CT, Geinisman Y, Penzes P (2012) Spatially restricted actin-regulatory signaling contributes to synapse morphology. J Neurochem 121(6): 852-860

Nonet ML, Grundahl K, Meyer BJ, Rand JB (1993) Synaptic function is impaired but not eliminated in C. elegans mutants lacking synaptotagmin. Cell 73(7): 1291-1305

Nonet ML, Saifee O, Zhao H, Rand JB, Wei L (1998) Synaptic transmission deficits in Caenorhabditis elegans synaptobrevin mutants. J Neurosci 18(1): 70-80

Nurrish S, Segalat L, Kaplan JM (1999) Serotonin inhibition of synaptic transmission: Galpha(0) decreases the abundance of UNC-13 at release sites. Neuron 24(1): 231-242

O'Brien SP, Seipel K, Medley QG, Bronson R, Segal R, Streuli M (2000) Skeletal muscle deformity and neuronal disorder in Trio exchange factor-deficient mouse embryos. Proc Natl Acad Sci U S A 97(22): 12074-12078

Park EC, Horvitz HR (1986) Mutations with dominant effects on the behavior and morphology of the nematode Caenorhabditis elegans. Genetics 113(4): 821-852

Penzes P, Johnson RC, Alam MR, Kambampati V, Mains RE, Eipper BA (2000) An isoform of kalirin, a brain-specific GDP/GTP exchange factor, is enriched in the postsynaptic density fraction. J Biol Chem 275(9): 6395-6403

Perez-Mansilla B, Nurrish S (2009) A network of G-protein signaling pathways control neuronal activity in C. elegans. Adv Genet 65: 145-192

Plageman TF, Jr., Chauhan BK, Yang C, Jaudon F, Shang X, Zheng Y, Lou M, Debant A, Hildebrand JD, Lang RA (2011) A Trio-RhoA-Shroom3 pathway is required for apical constriction and epithelial invagination. Development 138(23): 5177-5188

Portales-Casamar E, Briancon-Marjollet A, Fromont S, Triboulet R, Debant A (2006) Identification of novel neuronal isoforms of the Rho-GEF Trio. Biol Cell 98(3): 183-193

Rand JB (2007) Acetylcholine. WormBook

Reynolds NK, Schade MA, Miller KG (2005) Convergent, RIC-8-dependent Galpha signaling pathways in the Caenorhabditis elegans synaptic signaling network. Genetics 169(2): 651-670

Richmond JE, Davis WS, Jorgensen EM (1999) UNC-13 is required for synaptic vesicle fusion in C. elegans. Nat Neurosci 2(11): 959-964

105

Richmond JE, Jorgensen EM (1999) One GABA and two acetylcholine receptors function at the C. elegans neuromuscular junction. Nat Neurosci 2(9): 791-797

Riddle DL (1997a) C. elegans II, Plainview, N.Y.: Cold Spring Harbor Laboratory Press.

Riddle DL (1997b) C. elegans II, Plainview, N.Y.: Cold Spring Harbor Laboratory Press.

Roghani A, Feldman J, Kohan SA, Shirzadi A, Gundersen CB, Brecha N, Edwards RH (1994) Molecular cloning of a putative vesicular transporter for acetylcholine. Proc Natl Acad Sci U S A 91(22): 10620-10624

Rojas RJ, Yohe ME, Gershburg S, Kawano T, Kozasa T, Sondek J (2007) Galphaq directly activates p63RhoGEF and Trio via a conserved extension of the Dbl homology- associated pleckstrin homology domain. J Biol Chem 282(40): 29201-29210

Romero F, Fischer S (1996) Structure and function of vav. Cell Signal 8(8): 545-553

Rossman KL, Campbell SL (2000) Bacterial expressed DH and DH/PH domains. Methods Enzymol 325: 25-38

Rossman KL, Der CJ, Sondek J (2005) GEF means go: turning on RHO GTPases with guanine nucleotide-exchange factors. Nat Rev Mol Cell Biol 6(2): 167-180

Rossman KL, Worthylake DK, Snyder JT, Siderovski DP, Campbell SL, Sondek J (2002) A crystallographic view of interactions between Dbs and Cdc42: PH domain-assisted guanine nucleotide exchange. EMBO J 21(6): 1315-1326

Sahin M, Greer PL, Lin MZ, Poucher H, Eberhart J, Schmidt S, Wright TM, Shamah SM, O'Connell S, Cowan CW, Hu L, Goldberg JL, Debant A, Corfas G, Krull CE, Greenberg ME (2005) Eph-dependent tyrosine phosphorylation of ephexin1 modulates growth cone collapse. Neuron 46(2): 191-204

Sawin ER, Ranganathan R, Horvitz HR (2000) C. elegans locomotory rate is modulated by the environment through a dopaminergic pathway and by experience through a serotonergic pathway. Neuron 26(3): 619-631

Schade MA, Reynolds NK, Dollins CM, Miller KG (2005) Mutations that rescue the paralysis of Caenorhabditis elegans ric-8 (synembryn) mutants activate the G alpha(s) pathway and define a third major branch of the synaptic signaling network. Genetics 169(2): 631-649

Schmidt A, Hall A (2002) Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev 16(13): 1587-1609

106

Schmidt S, Diriong S, Mery J, Fabbrizio E, Debant A (2002) Identification of the first Rho-GEF inhibitor, TRIPalpha, which targets the RhoA-specific GEF domain of Trio. FEBS Lett 523(1-3): 35-42

Scholer LV, Moller TH, Norgaard S, Vestergaard L, Olsen A (2012) Isolating Genes Involved with Genotoxic Drug Response in the Nematode Caenorhabditis elegans Using Genome-Wide RNAi Screening. Methods Mol Biol 920: 27-38

Schwamborn JC, Puschel AW (2004) The sequential activity of the GTPases Rap1B and Cdc42 determines neuronal polarity. Nat Neurosci 7(9): 923-929

Seino S, Shibasaki T (2005) PKA-dependent and PKA-independent pathways for cAMP- regulated exocytosis. Physiol Rev 85(4): 1303-1342

Seipel K, O'Brien SP, Iannotti E, Medley QG, Streuli M (2001) Tara, a novel F-actin binding protein, associates with the Trio guanine nucleotide exchange factor and regulates actin cytoskeletal organization. J Cell Sci 114(Pt 2): 389-399

Shamah SM, Lin MZ, Goldberg JL, Estrach S, Sahin M, Hu L, Bazalakova M, Neve RL, Corfas G, Debant A, Greenberg ME (2001) EphA receptors regulate growth cone dynamics through the novel guanine nucleotide exchange factor ephexin. Cell 105(2): 233-244

Sieburth D, Ch'ng Q, Dybbs M, Tavazoie M, Kennedy S, Wang D, Dupuy D, Rual JF, Hill DE, Vidal M, Ruvkun G, Kaplan JM (2005) Systematic analysis of genes required for synapse structure and function. Nature 436(7050): 510-517

Sieburth D, Madison JM, Kaplan JM (2007) PKC-1 regulates secretion of neuropeptides. Nat Neurosci 10(1): 49-57

Smrcka AV, Sternweis PC (1993) Regulation of purified subtypes of phosphatidylinositol-specific phospholipase C beta by G protein alpha and beta gamma subunits. J Biol Chem 268(13): 9667-9674

Soisson SM, Nimnual AS, Uy M, Bar-Sagi D, Kuriyan J (1998) Crystal structure of the Dbl and pleckstrin homology domains from the human Son of sevenless protein. Cell 95(2): 259-268

Speese S, Petrie M, Schuske K, Ailion M, Ann K, Iwasaki K, Jorgensen EM, Martin TF (2007) UNC-31 (CAPS) is required for dense-core vesicle but not synaptic vesicle exocytosis in Caenorhabditis elegans. J Neurosci 27(23): 6150-6162

Spencer AG, Orita S, Malone CJ, Han M (2001) A RHO GTPase-mediated pathway is required during P cell migration in Caenorhabditis elegans. Proc Natl Acad Sci U S A 98(23): 13132-13137

107

Stam JC, Sander EE, Michiels F, van Leeuwen FN, Kain HE, van der Kammen RA, Collard JG (1997) Targeting of Tiam1 to the plasma membrane requires the cooperative function of the N-terminal pleckstrin homology domain and an adjacent protein interaction domain. J Biol Chem 272(45): 28447-28454

Steven R, Kubiseski TJ, Zheng H, Kulkarni S, Mancillas J, Ruiz Morales A, Hogue CW, Pawson T, Culotti J (1998) UNC-73 activates the Rac GTPase and is required for cell and growth cone migrations in C. elegans. Cell 92(6): 785-795

Steven R, Zhang L, Culotti J, Pawson T (2005) The UNC-73/Trio RhoGEF-2 domain is required in separate isoforms for the regulation of pharynx pumping and normal neurotransmission in C. elegans. Genes Dev 19(17): 2016-2029

Sudhof TC, Malenka RC (2008) Understanding synapses: past, present, and future. Neuron 60(3): 469-476

Sudhof TC, Rizo J (2011) Synaptic vesicle exocytosis. Cold Spring Harb Perspect Biol 3(12)

Sullivan R, Price LS, Koffer A (1999) Rho controls cortical F-actin disassembly in addition to, but independently of, secretion in mast cells. J Biol Chem 274(53): 38140- 38146

Sumakovic M, Hegermann J, Luo L, Husson SJ, Schwarze K, Olendrowitz C, Schoofs L, Richmond J, Eimer S (2009) UNC-108/RAB-2 and its effector RIC-19 are involved in dense core vesicle maturation in Caenorhabditis elegans. J Cell Biol 186(6): 897-914

Sze JY, Victor M, Loer C, Shi Y, Ruvkun G (2000) Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutant. Nature 403(6769): 560- 564

Tan EC, Leung T, Manser E, Lim L (1993) The human active breakpoint cluster region- related gene encodes a brain protein with homology to guanine nucleotide exchange proteins and GTPase-activating proteins. J Biol Chem 268(36): 27291-27298

Thalappilly S, Soubeyran P, Iovanna JL, Dusetti NJ (2010) VAV2 regulates epidermal growth factor receptor endocytosis and degradation. Oncogene 29(17): 2528-2539

Thiele TE, Badia-Elder NE (2003) A role for neuropeptide Y in alcohol intake control: evidence from human and animal research. Physiol Behav 79(1): 95-101

Tolias KF, Duman JG, Um K (2011) Control of synapse development and plasticity by Rho GTPase regulatory proteins. Prog Neurobiol 94(2): 133-148

108

Towers PR, Lescure P, Baban D, Malek JA, Duarte J, Jones E, Davies KE, Segalat L, Sattelle DB (2006) Gene expression profiling studies on Caenorhabditis elegans dystrophin mutants dys-1(cx-35) and dys-1(cx18). Genomics 88(5): 642-649 van der Linden AM, Moorman C, Cuppen E, Korswagen HC, Plasterk RH (2003) Hyperactivation of the G12-mediated signaling pathway in Caenorhabditis elegans induces a developmental growth arrest via protein kinase C. Curr Biol 13(6): 516-521 van Rijssel J, Kroon J, Hoogenboezem M, van Alphen FP, de Jong RJ, Kostadinova E, Geerts D, Hordijk PL, van Buul JD (2012) The Rho-GEF Trio controls leukocyte transendothelial migration by promoting docking structure formation. Mol Biol Cell

Varoqui H, Diebler MF, Meunier FM, Rand JB, Usdin TB, Bonner TI, Eiden LE, Erickson JD (1994) Cloning and expression of the vesamicol binding protein from the marine ray Torpedo. Homology with the putative vesicular acetylcholine transporter UNC-17 from Caenorhabditis elegans. FEBS Lett 342(1): 97-102

Vikis HG, Li W, Guan KL (2002) The plexin-B1/Rac interaction inhibits PAK activation and enhances Sema4D ligand binding. Genes Dev 16(7): 836-845

White JG, Amos WB, Fordham M (1987) An evaluation of confocal versus conventional imaging of biological structures by fluorescence light microscopy. J Cell Biol 105(1): 41- 48

White JG, Southgate E, Thomson JN, Brenner S (1986) The structure of the nervous system of the nematode Caenorhabditis elegans. Philos Trans R Soc Lond B Biol Sci 314(1165): 1-340

Williams SL, Lutz S, Charlie NK, Vettel C, Ailion M, Coco C, Tesmer JJ, Jorgensen EM, Wieland T, Miller KG (2007) Trio's Rho-specific GEF domain is the missing Galpha q effector in C. elegans. Genes Dev 21(21): 2731-2746

Wilson GM, Fielding AB, Simon GC, Yu X, Andrews PD, Hames RS, Frey AM, Peden AA, Gould GW, Prekeris R (2005) The FIP3-Rab11 protein complex regulates recycling endosome targeting to the cleavage furrow during late cytokinesis. Mol Biol Cell 16(2): 849-860

Worthylake DK, Rossman KL, Sondek J (2000) Crystal structure of Rac1 in complex with the guanine nucleotide exchange region of Tiam1. Nature 408(6813): 682-688

Wright KJ, Baye LM, Olivier-Mason A, Mukhopadhyay S, Sang L, Kwong M, Wang W, Pretorius PR, Sheffield VC, Sengupta P, Slusarski DC, Jackson PK (2011) An ARL3- UNC119-RP2 GTPase cycle targets myristoylated NPHP3 to the primary cilium. Genes Dev 25(22): 2347-2360

109

Xie Z, Cahill ME, Penzes P (2010) Kalirin loss results in cortical morphological alterations. Mol Cell Neurosci 43(1): 81-89

Zhang H, Webb DJ, Asmussen H, Horwitz AF (2003a) Synapse formation is regulated by the signaling adaptor GIT1. J Cell Biol 161(1): 131-142

Zhang XF, Schaefer AW, Burnette DT, Schoonderwoert VT, Forscher P (2003b) Rho- dependent contractile responses in the neuronal growth cone are independent of classical peripheral retrograde actin flow. Neuron 40(5): 931-944

Zhang Y, Grant B, Hirsh D (2001) RME-8, a conserved J-domain protein, is required for endocytosis in Caenorhabditis elegans. Mol Biol Cell 12(7): 2011-2021

Zheng Y, Brockie PJ, Mellem JE, Madsen DM, Maricq AV (1999) Neuronal control of locomotion in C. elegans is modified by a dominant mutation in the GLR-1 ionotropic glutamate receptor. Neuron 24(2): 347-361

Zhou KM, Dong YM, Ge Q, Zhu D, Zhou W, Lin XG, Liang T, Wu ZX, Xu T (2007) PKA activation bypasses the requirement for UNC-31 in the docking of dense core vesicles from C. elegans neurons. Neuron 56(4): 657-669

Zhou S, Chen L (2011) Neural integrity is maintained by dystrophin in C. elegans. J Cell Biol 192(2): 349-363

110

Appendix A

UNC-73C1 interacting proteins

Potential UNC-73 interacting proteins were identified in a yeast two hybrid assay performed by John Farver using the UNC-73C1 isoform (Farver, 2009). I focused on

LIN-2, F5512C.1 and DYS-1 to confirm the interaction in vitro and initiate the characterization of the expression patterns of these proteins with the help of undergraduate students.

1. LIN-2/CASK

LIN-2/CASK belongs to the membrane associated guanylate kinase (MAGUK) protein family (Caruana, 2002). LIN-2, LIN-7, and LIN-10 form a protein complex that is required for receptor localization and is conserved in mammals (Kaech et al, 1998). The role of LIN-2 in C. elegans vulva formation during development was extensively studied, but LIN-2 function in the nervous system was not reported. However, Drosophila CASK plays pre- and post-synaptic roles in neuromuscular junction formation (Chen &

Featherstone, 2011). In C. elegans, the LIN-2 interacting protein LIN-10 is essential for

GLR-1 glutamate receptor localization (Glodowski et al, 2005). In mammals, CASK is involved in NMDA receptor sorting and trafficking in hippocampal neurons (Jeyifous et

111 al, 2009). It is intriguing and suggestive of an UNC-73/LIN-2 interaction that UNC-73

RhoGEF2 isoform overexpression displays a glutamate receptor-like increased reversal frequency phenotype and GLR-1 receptors are mislocalized in unc-73 RhoGEF2 mutants while LIN-2 interacting proteins are also required for glutamate receptor localization.

In an in vitro GST pull down assay, His::LIN-2 bound to GST::UNC-73 (Fig

23C). The interaction of LIN-2 and UNC-73 was therefore confirmed in vitro. Also consistent with the physical interaction between UNC-73C1 and LIN-2 is the fact that both proteins are expressed in the nervous system, particularly in the nerve ring (Steven et al, 2005) (Robert Steven, personal communication). The implications of this interaction remain unknown.

112

Figure 23. LIN-2 interacts with UNC-73C1 in vitro. Pull down of GST::UNC-73C1 brought down His::LIN-2(88-360aa). LIN-2(88-360) interacts with UNC-73C1 in vitro. GST::UNC-73C1 from E. coli was bound to glutathione sepharose beads and incubated with E. coli lysate expressing His::LIN-2(88- 360). Bound proteins were separated and imunoblotted with anti-His antibody. The blot was then stripped and reprobed with anti-GST antibody. Two percent of the His::LIN- 2(88-360) and 0.4 percent of the GST::UNC-73C1 input cell lysates are also shown. GST::UNC73C1 (130kDa) and GST dimer (56kDa) are indicated with arrowheads

113

Figure 24. F5512C.1 weakly interacts with UNC-73C1 in vitro. UNC-73 and F5512C.1 appear to weakly interact in an in vitro GST pull down assay. GST::UNC-73C1 from E. coli was bound to glutathione sepharose beads and incubated with E. coli lysate expressing His::F5512C.1. Bound proteins were separated and imunoblotted with anti-His antibody. The blot was then stripped and reprobed with anti- GST antibody of the GST::UNC-73C1 input cell lysates are also shown.

114

2. F55C12.1/Rab11-FIP3/NUF

F55C12.1 is a C. elegans homolog of the Rab11 effector Rab11-FIP3, which belongs to the Rab11 family of interacting proteins (Horgan & McCaffrey, 2009). Rab11-

FIP3 regulates endocytosis by associating to the recycling endosome membrane (Horgan et al, 2007; Wilson et al, 2005). Rab11-FIP3 is also involved in actin cytoskeleton regulation and binds to dynein (Horgan et al, 2010; Jing et al, 2009). These studies provide a possibility that F5512C.1/Rab11-FIP3 and UNC-73 along with Rho signaling could functionally interact.

A weak in vitro interaction between F55C12.1 and UNC-73C1 was observed in a

GST-pull down experiment (Fig. 24). Also consistent with the physical interaction between UNC-73C1 and F55C12.1 is the fact that both are expressed in the nervous system (Steven et al, 2005) (Robert Steven, personal communication). The yeast two hybrid, GST pull-down and expression data are consistent with UNC-73C1 interacting with F55C12.1/Rab11-FIP3 in vivo.

3. DYS-1/dystrophin

DYS-1/dystrophin is member of the dystrophin-glycoprotein complex (DGC) in muscles, which anchors the extracellular matrix to the cytoskeleton (Francis & Waterston,

1991). Dystrophin also localizes to synapses in the nervous system and neuromuscular junctions (Haenggi & Fritschy, 2006). Dystrophin mutations in humans result in

Duchenne muscular dystrophy (DMD) and other hereditary diseases. DMD features muscle degeneration, mental retardation, has no cure and affects 1 in 3,600 boys. In C. elegans, the dys-1 null mutant does not show muscle degeneration, but is hypersensitive

115 to aldicarb indicating potential neurotransmission defects (Bessou et al, 1998). However, this hypersensitivity may be due to a decrease of acetylcholinesterase activity instead of impaired neurotransmission (Giugia et al, 1999). Despite many studies of muscular DYS-

1function of, only recently has the role of DYS-1 in the nervous system been addressed.

DYS-1 is required to maintain neuronal integrity in C. elegans (Zhou & Chen, 2011). dys-1 mutants have a locomotory defect characterized by excessive head movements, which is rescued by muscle DYS-1 expression, and a defect in neuronal integrity characterized by mislocalization of cholinergic motoneuron cell bodies, which is rescued by neuronal DYS-1 expression. In a microarray screening, RHO-1 was down regulated in dys-1 mutants (Towers et al, 2006). The DYS-1 neuronal function and the regulation of

RHO-1 expression by DYS-1 indicate the possibility of a functional interaction between

DYS-1 and UNC-73/Rho signaling in the nervous system.

To examine DYS-1 interactions with UNC-73C1 in vitro, DYS-1B and UNC-

73C1 expression was attempted in E. coli and HEK293 cells, but DYS-1B could not be produced in either cell type. The in vitro expression of DYS-1 failed in E. coli Codon

Plus bacteria cells grown at 16°C in a range of IPTG concentrations. Neither His- nor

GST-tagged DYS-1B fusion protein expression was detected in cell lysates by Western blotting and probing with corresponding antibodies. Western blotting and probing with the same antibodies failed to detect DYS-1B fusion proteins in the lysates of transfected

HEK293 cells as well. As it may have been an issue with DYS-1B expression in yeast cells (Farver, 2009), DYS-1B overexpression may have toxic effects on bacterial and mammalian cell growth. Finding the optimal condition for DYS-1B expression is the main obstacle to characterizing a potential interaction with UNC-73 in vitro.

116

Appendix B

Cell specific RNAi

To explore which neurons are crucial to mediate the locomotory regulation of

UNC-73 RhoGEF2 isoforms, cell-specific RNAi (Esposito et al, 2007) was performed along with the previous described cell-specific rescue experiments.

The efficiency of knock-down was monitored by expressing the UNC-73

RhoGEF2 specific RNAi construct in an unc-73E::GFP background. If all UNC-73

RhoGEF2 isoforms were specifically knocked-down, the GFP signal associated with

UNC-73E RhoGEF2 isoform expression should decrease. The normally strong expression of UNC-73E in the ventral nerve cord was much weaker and the cell bodies along the cord were no longer visible in the RNAi animals (Fig. 25B). The mid-body axonal expression of the UNC-73E isoform was also decreased. These results suggest that the cell specific RNAi construct significantly decreased the expression of UNC-73

RhoGEF2 isoforms. Another control was knocking out the UNC-73 RhoGEF2 isoforms in the pharyngeal muscles using the myo-2 promoter. Since UNC-73 RhoGEF2 isoforms are required for pharyngeal pumping to feed (Steven et al, 2005), the myo-2p::UNC-

73PH2 RNAi worms were expected to be larval arrested similar to unc-73(ev802). That was indeed the case (the worm lines could not be maintained, data not shown).

117

Pan-neuronal unc-73 RhoGEF2 RNAi (unc-119p::UNC-73PH2 RNAi) decreased the locomotion rate of the transgenic worms, but not to a rate as low as that of unc-

73(ce362) mutants (Fig. 25A). Similarly, EGL-30, EGL-8, UNC-13, GSA-1, or UNC-31 knock-down in cholinergic motor neurons (unc-17p) did not decrease the locomotion rates as expected based on previous genetic analysis (Lackner et al, 1999; Miller et al,

1999; Richmond et al, 1999; Speese et al, 2007). This may be a result of variations in the dosage of the RNAi construct or the strength of the promoter affecting the RNAi results

(Esposito et al, 2007).

The RNAi experiments also yielded some unexpected results. First, all RNAi strains driven by the unc-17 promoter showed an unexpected coiled phenotype, similar to that seen in unc-17 mutant. This suggested these particular RNAi constructs affect unc-17 gene expression complicating the interpretation of the phenotype generated by RNAi.

Second, unc-119p::rho-1RNAi transgenic worms were unable to be maintained. The neuronal RHO-1 knock-down resulted in elongated unhealthy F1s that did not develop any progeny (data not shown; worm lines could not be maintained). This phenotype may result from the connection between Rho signaling and fecundity, whereby RHO-1 affects egg-laying and brood size in adult C. elegans (McMullan & Nurrish, 2011), and RHO-1

RNAi that is not cell specific causes early embryonic lethality (Jantsch-Plunger et al,

2000).

118

B C

Figure 25. Cell-specific RNAi phenotypes for unc-73 and other related genes. A) unc-73 RhoGEF2 pan-neuronal RNAi using the unc-119 promoter decreased locomotion rate, while muscle knock-down using the myo-3 promoter did not change the rate significantly. RNAi on other related genes resulted in impaired locomotion rates that were not as severe as the corresponding genetic mutants. Data are presented as a mean ± SEM. B) UNC-73E::GFP fluorescence decreased in RNAi strains indicating the efficiency of UNC-73 RhoGEF2 isoform knock-down. Left panels: ventral cord with arrow heads indicating cell bodies along the ventral cord. Right panel: mid-body of the worm, showing axonal fluorescence and arrows indicating motor neuron commissures. All pictures were captured using the same exposure time. C) All unc-17p RNAi animals had a coiled phenotype similar to unc-17 mutants.

119