2000-2005A: A Third Major Branch of the Synaptic Signaling Network
As I started my new lab at OMRF in July of 2000, I had the ambitious goal of bridging the gap between synaptic function and the control of behaviors.
In my postdoc years I had contributed to discovering a network of "G Protein" signal transduction pathways that appeared to act in neurons to tightly control behaviors such as locomotion and egg laying in C. elegans (Adventures > 1996-1999 and 1996-2000). Since these pathways were conserved in all animals and were also present in the human brain, I and others hypothesized that they might be fundamentally important for turning synapses ON and OFF and that understanding the pathways might yield important insights into how behaviors are established (learned), maintained (remembered), and modified. |
Prior to studies by myself and others in C. elegans, G protein signaling had not been intensively studied using genetic approaches. Thus, despite an excellent body of biochemical work on G protein signaling, there remained significant gaps in our molecular understanding of the pathways as well as the big-picture understanding of how such signaling is used to control populations of synapses in behaving animals.
The genetic studies by myself and others in the late 1990's had raised more questions than they had answered, but they had also produced a framework for pursuing answers. For me, this framework was best understood through the following pathway model of the network.
The genetic studies by myself and others in the late 1990's had raised more questions than they had answered, but they had also produced a framework for pursuing answers. For me, this framework was best understood through the following pathway model of the network.
In this model, solid lines indicate that direct interactions are known or likely, while dashed lines and/ or large gaps indicate that interactions are only predicted or that there are possible missing components. Arrows and short perpendicular lines indicate activation and inhibition, respectively. Proteins that promote or inhibit locomotion and/ or neurotransmitter release are shown in red or green, respectively. Based on biochemical and genetic studies, EGL-30 (Gq) acts through EGL-8 (PLCbeta) to enzymatically produce the small signaling molecule DAG, which in turn positively regulates neurotransmitter release, in part via interactions with the DAG-binding protein UNC-13. The EGL-30 (Gq) pathway is negatively regulated by GOA-1 (Go), via an unknown mechanism, and the RGS protein EAT-16. Another RGS protein, EGL-10, negatively regulates GOA-1 (Go). DGK-1 (Diacylglycerol kinase) antagonizes the EGL-30 pathway by breaking down DAG.
I had just published our discovery of the novel, conserved protein RIC-8, which our genetic analysis suggested was important for maintaining the EGL-30 (Gq) pathway in a functional state (Adventures>1996-2000). As indicated by the dashed line extending from RIC-8 in the above model, we had no idea how RIC-8 exerted its effects or with which other pathways it might interact. Geneticists often use a technique call forward genetic screening when when they don't understand a process or pathway. "Forward genetics" means making random mutations in the germline of an organism (with a chemical like EMS), screening subsequent generations for mutations that disrupt the process, and then mapping the mutations to specific genes to identify which proteins the mutations disrupt. The beauty of forward genetics is that it lets the organism tell you what is important for the process you are studying. To find out more about RIC-8 and the Synaptic Signaling Network, we decided to do two large forward genetics screens. In the first, we started with ric-8 mutants and screened for mutations that rescued their paralysis. In the second, we mutated wild type animals and looked for mutations that caused hyperactive locomotion. Quite unexpectedly, all of the strongest mutations were rare mutations that caused the Gs pathway to turn ON constitutively. Three mutations were dominant, gain-of-function mutations in GSA-1 (Gs ) itself, two were dominant, gain-of-function mutations in ACY-1 (adenylyl cyclase), and three were loss of function mutations that inactivated the regulatory subunit of KIN-1 (Protein Kinase A), thus leading to constitutive PKA activity. This is the canonical Gs pathway that studies in fruit flies and Aplysia had shown was important for learning and memory! The best mutations improved the locomotion rate of the ric-8 mutant by about 40-fold, to levels slightly greater than the normal rate for wild type. The mutations also restored coordinated, sinusoidal movement to ric-8 mutants. However, mutants that had defects in synaptic vesicle fusion were not well rescued, indicating that RIC-8 and the Gs pathway functioned upstream of neurotransmitter release. We also acquired a transgenic strain from Ron Plasterk's lab in which we could induce activation of the Gs pathway in adults by heat shock. When we crossed the transgene into the ric-8 mutant, we found that we could still rescue ric-8's locomotion to greater than wild type levels by inducing activation of the Gs pathway in adults. This showed that ric-8 mutants don't have any permanent developmental defects, but are simply missing the signals they need to move. One of the most puzzling observations we made was that the mutants showed only mild to moderate hypersensitivity to aldicarb, which indicates that their overall levels of release of the neurotransmitter acetylcholine are only mildly to moderately increased (see Adventures>1993-1996). This contrasts with goa-1 (Go) mutants, which are about 40-fold more hypersensitive to aldicarb than wild type (see Adventures>1996-1999) but whose locomotion is not as hyperactive as the Gs pathway activation mutants. |
|
The Gs pathway is expressed in neurons and body wall muscle. Thus in the hyperactive Gs pathway activation mutants the pathway is hyper-activated in both muscle and neurons. Using transgenic experiments, we found that over-expressing the mutationally-activated protein only in muscle or only in the nervous system caused hyperactive locomotion in both cases (though to a lesser degree than the native mutation). However the paralysis of ric-8 mutants was only rescued through expression of the activated Gs pathway in the nervous system. Surprisingly, neither the muscle-activated Gs pathway nor the nervous system-activated Gs pathway affected overall levels of neurotransmitter release, since strains carrying these "single tissue" Gs activation mutations had wild type aldicarb sensitivity. A separate transgene that expressed the activated Gs pathway in both muscle and neurons caused aldicarb hypersensitivity and hyperactive locomotion similar to the native mutation.
These results and others led us to hypothesize that the Gs pathway activation may cause synapses that are optimal for locomotion to be turned ON and release more neurotransmitter, but that many non-optimal acetylcholine-utilizing motor neuron synapses release normal levels of acetylcholine in the Gs pathway activation mutants. This would result in overall (whole animal) levels of release that are only mildly to moderately elevated. Additionally, there appears to be a retrograde communication link/ pathway between the muscles and the nervous system that is driven by the Gs pathway.
Schade MA, Reynolds NK, Dollins CM, and Miller KG (2005). Mutations that rescue the paralysis of Caenorhabditis elegans ric-8 (Synembryn) mutants activate the Gas pathway and define a third major branch of the synaptic signaling network. Genetics. 169(2), 631-649. PMCID: PMC1449092.
Published back-to-back with Reynolds et al., 2005.
Cited by 75 papers as of April, 2018.
These results and others led us to hypothesize that the Gs pathway activation may cause synapses that are optimal for locomotion to be turned ON and release more neurotransmitter, but that many non-optimal acetylcholine-utilizing motor neuron synapses release normal levels of acetylcholine in the Gs pathway activation mutants. This would result in overall (whole animal) levels of release that are only mildly to moderately elevated. Additionally, there appears to be a retrograde communication link/ pathway between the muscles and the nervous system that is driven by the Gs pathway.
Schade MA, Reynolds NK, Dollins CM, and Miller KG (2005). Mutations that rescue the paralysis of Caenorhabditis elegans ric-8 (Synembryn) mutants activate the Gas pathway and define a third major branch of the synaptic signaling network. Genetics. 169(2), 631-649. PMCID: PMC1449092.
Published back-to-back with Reynolds et al., 2005.
Cited by 75 papers as of April, 2018.
Remaining Important Questions
-In our large genetic screens, we also isolated gain-of-function mutations in EGL-30 (Gq) that constitutively turn on the Gs pathway and these were also strong suppressors of ric-8 mutants. Also, our previous studies suggested that RIC-8 functioned upstream of EGL-30 (Gq) (Adventures>1996-2000). Our findings in the current study are consistent with RIC-8 also functioning upstream of the Gs pathway. We further tested this idea genetically in Adventures>2000-2005B, but it wasn't until years later that this was demonstrated biochemically (see Related Studies below).
-The discovery of a third major branch of the Synaptic Signaling Network posed a major challenge for us. We now had two major non-redundant pathways that were required for locomotion, but they seemed to be quite different in their effects on neurotransmitter release. What is the relationship of the Gs pathway to the Go-Gq system that we and others had discovered earlier? We pursued this question intensively in Adventures>2000-2005B and Adventures>2006.
-The discovery of a third major branch of the Synaptic Signaling Network posed a major challenge for us. We now had two major non-redundant pathways that were required for locomotion, but they seemed to be quite different in their effects on neurotransmitter release. What is the relationship of the Gs pathway to the Go-Gq system that we and others had discovered earlier? We pursued this question intensively in Adventures>2000-2005B and Adventures>2006.
Fun Factoids about this Study
-We did the two forward genetic screens in this study on a huge scale, even by C. elegans standards. Together, the 2 screens represent 33-fold genomic coverage, meaning that we screened enough animals to obtain 33 loss-of-function mutations in each average size gene in the C. elegans genome. We did this because we wanted to identify rare dominant mutations, if they existed, and rare reduction-of-function mutations in genes with lethal null phenotypes. As it turns out the statistics were favorable for us and we identified both kinds of mutations in this screen (the kin-2 PKA regulatory subunit and pde-4 cAMP phosphodiesterase reduction-of-function mutants are lethal as nulls).
-We screened for 28 weekly cycles. We did the screens "clonally", meaning that my technician, Nicky Reynolds, would hand pick 2000 adult F1 progeny of mutagenized animals onto individual wells of 24-well plates and then hand-picked them off the next day after allowing them to lay progeny overnight. Then, 3 1/2 - 4 days later, I individually screened each well and isolated any mutants I saw. I remember that the strongest Gs pathway gain-of-function mutants from the ric-8 suppressor screen barely survived. I had to baby them along for weeks!
-Although the C. elegans genome had been sequenced by this point, this was before the days of routine whole-genome sequencing to identify mutations, so we had to map the mutations at high resolution relative to Single Nucleotide Polymorphisms (SNPs). There were no routine protocols for doing this, so we adapted high throughput PCR-based screening methods from the C. elegans Gene Knockout Project based in Bob Barstead's lab at OMRF. We mapped representative mutations in each gene to 150 - 250 Kb intervals and then looked for candidate genes in the interval. Soon, a pattern emerged (the Gs pathway).
-Our genetic screens produced the first native, germline Gs and adenylyl cyclase gain-of-function mutants isolated in an animal system and the first whole-animal non-targeted PKA regulatory subunit mutations.
-As an example of the kinds of mutations you can get when you screen at high resolution, one of our gsa-1 gain-of-function mutations hit the catalytic arginine of Gs that is the precise target of ADP ribosylation by cholera toxin (Cholera infection causes diarrhea by modifying that amino acid to constitutively activate Gs).
-This is the "learning and memory" pathway made famous by studies of learning mutants in Drosophila and Erik Kandel's memory studies in Aplysia.
-Our study was published back-to-back with our first follow up study (Reynolds et al., 2005; Adventures>2000-2005B).
-We screened for 28 weekly cycles. We did the screens "clonally", meaning that my technician, Nicky Reynolds, would hand pick 2000 adult F1 progeny of mutagenized animals onto individual wells of 24-well plates and then hand-picked them off the next day after allowing them to lay progeny overnight. Then, 3 1/2 - 4 days later, I individually screened each well and isolated any mutants I saw. I remember that the strongest Gs pathway gain-of-function mutants from the ric-8 suppressor screen barely survived. I had to baby them along for weeks!
-Although the C. elegans genome had been sequenced by this point, this was before the days of routine whole-genome sequencing to identify mutations, so we had to map the mutations at high resolution relative to Single Nucleotide Polymorphisms (SNPs). There were no routine protocols for doing this, so we adapted high throughput PCR-based screening methods from the C. elegans Gene Knockout Project based in Bob Barstead's lab at OMRF. We mapped representative mutations in each gene to 150 - 250 Kb intervals and then looked for candidate genes in the interval. Soon, a pattern emerged (the Gs pathway).
-Our genetic screens produced the first native, germline Gs and adenylyl cyclase gain-of-function mutants isolated in an animal system and the first whole-animal non-targeted PKA regulatory subunit mutations.
-As an example of the kinds of mutations you can get when you screen at high resolution, one of our gsa-1 gain-of-function mutations hit the catalytic arginine of Gs that is the precise target of ADP ribosylation by cholera toxin (Cholera infection causes diarrhea by modifying that amino acid to constitutively activate Gs).
-This is the "learning and memory" pathway made famous by studies of learning mutants in Drosophila and Erik Kandel's memory studies in Aplysia.
-Our study was published back-to-back with our first follow up study (Reynolds et al., 2005; Adventures>2000-2005B).
Related Studies
We followed up on this study with three major publications. Reynolds et al., 2005 was co-published with this study and investigated the relationship of RIC-8 to the Gs pathway and the relationship of the Gs pathway to the Gq pathway (Adventures>2000-2005B). Charlie et al., 2006A "completes" the Gs pathway by showing that the two large genetic screens in this study also netted a single rare mutation in pde-4, which encodes the cAMP Phosphodiesterase that is part of the Gs pathway and an ortholog of the Drosophila "Dunce" learning and memory mutant (Adventures>2004-2006). Charlie et al., 2006B further investigates the relationship of the Gs and Gq pathways to each other and to UNC-31 (CAPS), which, among other possible functions, is involved in the release of neuropeptides by dense core vesicles (Adventures>2006).
Reynolds NK, Schade MA, and Miller KG (2005). Convergent, RIC-8 – dependent G alpha signaling pathways in the Caenorhabditis elegans synaptic signaling network. Genetics. 169(2), 651-670. PMCID: PMC1449085.
Charlie NK, Thomure AM, Schade MA, and Miller KG (2006). The Dunce cAMP phosphodiesterase PDE-4 negatively regulates Gs – dependent and Gs – independent cAMP pools in the Caenorhabditis Elegans synaptic signaling network. Genetics 173(1), 111-130. PMCID: PMC1461419.
Charlie NK, Schade MA, Thomure AM, and Miller KG. (2006). Presynaptic UNC-31 (CAPS) is required to activate the Gs pathway of the Caenorhabditis elegans synaptic signaling network. Genetics. 172(2): 943-961. PMCID: PMC1456257.
In 2003, Greg Tall published his discovery that mammalian RIC-8 is guanine nucleotide exchange factor for Gq and Go that interacts directly with those proteins. He also found that RIC-8 interacted directly with Gs, but he could not demonstrate GEF activity the RIC-8 Gs interaction in that first study. Later, he and another lab were able to demonstrate GEF activity of RIC-8 for Gs.
Tall, GG, Krumins AM and AG Gilman (2003). Mammalian Ric-8A (Synembryn) Is a Heterotrimeric G alpha Protein Guanine Nucleotide Exchange Factor. J Biol Chem 278: 8356-8362.
Chan, P, Gabay M, Wright, FA, Tall, G.G. (2011). RIC-8B is a GTP-dependent G Protein alpha S Guanine Nucleotide Exchange Factor. J Biol Chem 286: 19932-42.
Ximena, R, Pasten P, MartInez, S, et al. (2008) xRIC-8 is a GEF for Gs and participates in maintaining meiotic arrest in Xenopus laevis oocytes. J Cell Physiol 214: 673-80.
Before we isolated our native mutants with Gs pathway hyperactivation, two other studies from the labs of Ron Plasterk and Josh Kaplan were published in which they overexpressed transgenic, constitutively active Gs in C. elegans neurons. These studies found that this killed neurons and caused permanent paralysis. When we looked for neuronal death in our native dominant mutants, we found only a low level (about 1 in 300 neurons; Schade et al., 2005). So, unlike the transgenic strains, the native dominant mutations do not cause widespread neuronal death, as also seems self-evident from the hyperactive locomotion phenotype.
Korswagen, H. C., A. M. van der Linden and R. H. Plasterk, 1998 G protein hyperactivation of the Caenorhabditis elegans adenylyl cyclase SGS-1 induces neuronal degeneration. EMBO J. 17: 5059-5065.
Berger, A. J., A. C. Hart and J. M. Kaplan, 1998 Gs-induced neurodegeneration in Caenorhabditis elegans. Journal of Neuroscience 18: 2871-2880.
The Plasterk lab produced two very useful Gs pathway reagents that we used in this study and several future studies: a null mutant in acy-1 (larval arrested) that was nicely balanced with a Dpy mutation, and a strain containing a genomically integrated transgene that allows inducible expression of the Gs pathway using the heat-shock promoter.
Moorman, C., and R. H. Plasterk, 2002 Functional characterization of the adenylyl cyclase gene sgs-1 by analysis of a mutational spectrum in Caenorhabditis elegans. Genetics 161: 133-142.
Korswagen, H. C., J. Park, Y. Ohshima and R. H. Plasterk, 1997 An activating mutation in a Caenorhabditis elegans Gs protein induces neural degeneration. Genes and Development 11: 1493-1503.
Our Gs pathway activation mutants have been useful tools for other labs. Yishi Jin and Andrew Chisholm's lab at UC San Diego used them to demonstrate that cAMP (the small molecule produced by the Gs pathway) promotes axonal regeneration in the laser-induced axon severing model they use. Dave Raizen's lab at Penn used them to analyze the role of the Gs pathway in sleep.
Ghosh-Roy, A., Wu, Z., Goncharov, A., Jin, Y., and Chisholm, A.D. (2010). Calcium and
cyclic AMP promote axonal regeneration in Caenorhabditis elegans and require DLK-1
kinase. J. Neurosci. 30, 3175-3183.
Belfer, S.J,, Chuang, H.S., Freedman, B.L., Yuan, J., Norton, M., Bau, H.H., and
Raizen, D,M. (2013). Caenorhabditis-in-drop array for monitoring C. elegans quiescent
behavior. Sleep 36, 689-698G.
Reynolds NK, Schade MA, and Miller KG (2005). Convergent, RIC-8 – dependent G alpha signaling pathways in the Caenorhabditis elegans synaptic signaling network. Genetics. 169(2), 651-670. PMCID: PMC1449085.
Charlie NK, Thomure AM, Schade MA, and Miller KG (2006). The Dunce cAMP phosphodiesterase PDE-4 negatively regulates Gs – dependent and Gs – independent cAMP pools in the Caenorhabditis Elegans synaptic signaling network. Genetics 173(1), 111-130. PMCID: PMC1461419.
Charlie NK, Schade MA, Thomure AM, and Miller KG. (2006). Presynaptic UNC-31 (CAPS) is required to activate the Gs pathway of the Caenorhabditis elegans synaptic signaling network. Genetics. 172(2): 943-961. PMCID: PMC1456257.
In 2003, Greg Tall published his discovery that mammalian RIC-8 is guanine nucleotide exchange factor for Gq and Go that interacts directly with those proteins. He also found that RIC-8 interacted directly with Gs, but he could not demonstrate GEF activity the RIC-8 Gs interaction in that first study. Later, he and another lab were able to demonstrate GEF activity of RIC-8 for Gs.
Tall, GG, Krumins AM and AG Gilman (2003). Mammalian Ric-8A (Synembryn) Is a Heterotrimeric G alpha Protein Guanine Nucleotide Exchange Factor. J Biol Chem 278: 8356-8362.
Chan, P, Gabay M, Wright, FA, Tall, G.G. (2011). RIC-8B is a GTP-dependent G Protein alpha S Guanine Nucleotide Exchange Factor. J Biol Chem 286: 19932-42.
Ximena, R, Pasten P, MartInez, S, et al. (2008) xRIC-8 is a GEF for Gs and participates in maintaining meiotic arrest in Xenopus laevis oocytes. J Cell Physiol 214: 673-80.
Before we isolated our native mutants with Gs pathway hyperactivation, two other studies from the labs of Ron Plasterk and Josh Kaplan were published in which they overexpressed transgenic, constitutively active Gs in C. elegans neurons. These studies found that this killed neurons and caused permanent paralysis. When we looked for neuronal death in our native dominant mutants, we found only a low level (about 1 in 300 neurons; Schade et al., 2005). So, unlike the transgenic strains, the native dominant mutations do not cause widespread neuronal death, as also seems self-evident from the hyperactive locomotion phenotype.
Korswagen, H. C., A. M. van der Linden and R. H. Plasterk, 1998 G protein hyperactivation of the Caenorhabditis elegans adenylyl cyclase SGS-1 induces neuronal degeneration. EMBO J. 17: 5059-5065.
Berger, A. J., A. C. Hart and J. M. Kaplan, 1998 Gs-induced neurodegeneration in Caenorhabditis elegans. Journal of Neuroscience 18: 2871-2880.
The Plasterk lab produced two very useful Gs pathway reagents that we used in this study and several future studies: a null mutant in acy-1 (larval arrested) that was nicely balanced with a Dpy mutation, and a strain containing a genomically integrated transgene that allows inducible expression of the Gs pathway using the heat-shock promoter.
Moorman, C., and R. H. Plasterk, 2002 Functional characterization of the adenylyl cyclase gene sgs-1 by analysis of a mutational spectrum in Caenorhabditis elegans. Genetics 161: 133-142.
Korswagen, H. C., J. Park, Y. Ohshima and R. H. Plasterk, 1997 An activating mutation in a Caenorhabditis elegans Gs protein induces neural degeneration. Genes and Development 11: 1493-1503.
Our Gs pathway activation mutants have been useful tools for other labs. Yishi Jin and Andrew Chisholm's lab at UC San Diego used them to demonstrate that cAMP (the small molecule produced by the Gs pathway) promotes axonal regeneration in the laser-induced axon severing model they use. Dave Raizen's lab at Penn used them to analyze the role of the Gs pathway in sleep.
Ghosh-Roy, A., Wu, Z., Goncharov, A., Jin, Y., and Chisholm, A.D. (2010). Calcium and
cyclic AMP promote axonal regeneration in Caenorhabditis elegans and require DLK-1
kinase. J. Neurosci. 30, 3175-3183.
Belfer, S.J,, Chuang, H.S., Freedman, B.L., Yuan, J., Norton, M., Bau, H.H., and
Raizen, D,M. (2013). Caenorhabditis-in-drop array for monitoring C. elegans quiescent
behavior. Sleep 36, 689-698G.