Jarrell, Travis et al. (2009) International Worm Meeting "Network Graph Analysis of the C. elegans Male Posterior Connectome."

Emmons, Scott W., Xu, Meng, Wang, Yi, Jarrell, Travis, Hall, David H.

[International Worm Meeting, 2009]

Complex networks of all kinds, from social networks to the internet to power grids, have been investigated for a number of years to understand how their properties emerge from their structure. The C. elegans male posterior connectome presents a complex network of cellular interactions in which each cell is connected to an average of 21 other cells. Artificial neural networks in which nodes are "neurons" and edges are "synapses" have been attractive models of the brain because they can store "memories" as particular stable and recoverable patterns of activity. The results of these studies serve as guides for investigating how the male neural network functions to generate multi-step copulatory behavior. We have analyzed the male network using available software packages, including VisANT (http://visant.bu.edu). The posterior connectome can be represented as a weighted, directed graph with 229 nodes (all neurons and muscles) and 3222 edges (chemical synapses, gap junctions and neuromuscular junctions). Both chemical and electrical synapses are identified as morphological structures and can be assigned weights based on the number of 50 nm serial sections over which they extend. The graph structure has a number of features found in many natural networks, including the hermaphrodite nervous system. These features have important implications for its function. It is a small world graph, meaning the number of edges traversed when going between any pair of randomly chosen nodes is small, here about 3. Small world graphs are robust and capable of high speed information processing. The graph has a high cluster coefficient, meaning two neurons connected to a given neuron are likely to be connected to each other. The degree distribution, that is the distribution of the number of edges per node, is characterized by presence of a small number of nodes with very high degree. The most highly connected neurons in the male tail are PVV, PVX, PVZ and PDB, PVV being the highest with 65 different synaptic partners. These neurons form hubs and result in the network having a single giant component that includes essentially every cell. There is repeated recurrence of a feed-forward loop network motif involving the shared neurons LUAL and LUAR. By displaying the graph with grouping of cells by cell type or connectivity, pathways of information flow through the network can be visualized. We are interested to learn the extent to which the separate sub-behaviors of the copulatory sequence are subserved by distinct circuits or instead emerge as functional modes of the network as a whole.

Jarrell, Travis A. et al. (2013) International Worm Meeting "The connectome of the anterior nervous system of the C. elegans adult male."

Nguyen, Kenneth, Bloniarz, Adam E., Brittin, Christopher A., Cook, Steven J., Hall, David H., Emmons, Scott W., Xu, Meng, Jarrell, Travis A., Wang, Yi

[International Worm Meeting, 2013]

The innate behavioral repertoires of the two sexes of a species are guided by differing reproductive priorities. C. elegans male copulation is controlled by a neural network in the tail in which a majority of the neurons and muscles are specific to the male. But known differences in olfactory preferences and exploratory tendencies emanate from behaviors controlled by circuits in the head, where the complement of neurons is nearly identical in the two sexes. We determined connectivity in the anterior nervous system of the adult male from a 1,500 section-long thin section EM series extending from near the tip of the nose, through the nerve ring, and part way into the retrovesicular ganglion. This region contains the bulk of the synapses, excluding ventral cord nmj's. To make a comparison to the hermaphrodite, we re-reconstructed legacy Cambridge micrographs using our software, which allows us to score synaptic weights (see abstract by Cook et al). While our analysis is at an early stage, we can already see the essential result: in the adjacency matrices that display the connectivity, it is difficult to spot differences that appear greater than would be expected given the inherent variability of neuronal wiring. Known circuits in navigation and other responses are conserved. Thus behavioral differences likely emerge from differing circuit properties rather than differing connectivity. There are two possible exceptions: AIM synapses onto AIB and RIA synapses onto RIB in the male only. One set of male-specific synapses expected involves the male-specific head CEM sensory neurons, and the tail EF interneurons, which receive extensive input from the copulatory circuits in the tail and extend processes through the ventral nerve cord into the nerve ring. Both of these neuron classes have as their strongest targets the AVB command interneurons for forward locomotion. This suggests one of their functions may be to inhibit forward locomotion when a hermaphrodite is sensed or during copulation. They make additional connections to be further explored.

Lazaro-Pena, Maria I. et al. (2011) International Worm Meeting "An approach towards identification of cell recognition molecules involved in synapse specificity in the nematode Caenorhabditis elegans."

Bloniarz, Adam, Lazaro-Pena, Maria I., Jarrell, Travis, Emmons, Scott W.

[International Worm Meeting, 2011]

How neurons recognize their correct synaptic partners remains largely an unsolved problem. Additional synapse-level connectivity from the new field of connectomics reveal the complexity of this process. Neurons in the circuits of the C. elegans male tail that govern mating are joined in a complex network by over 4000 chemical and 4000 electrical synapses. Each neuron makes synapses with many other neurons (average 15-20). We measure the strength of these interactions from the number of serial EM sections over which they occur. Every neuron interacts strongly with some of its partners and weakly with others, forming a smooth distribution from strong to weak. Fifty percent of the total synaptic load is carried by the weaker set of interactions (<20 sections). To determine whether the weaker set was specific or random synaptic noise, we ranked the neurons by the similarity of their connectivity. Ranking neurons according to their weaker connections gave the same result as ranking them by their strong interactions: left/right homologs and other sets of neurons thought to be equivalent ranked as most similar in connectivity. Therefore, at least some of the weaker connections are specific. To determine whether the strength of synaptic interaction is a consequence of the amount of cell contact, we have begun a volumetric reconstruction of the neurons in the adult male pre-anal ganglion. Preliminary results indicate an absence of correlation between the amount of cell contact and the strength of synaptic interaction, implying the presence of cell-specific as well as neighborhood-specific recognition functions. A model to explain these observations hypothesizes that there are a large number of recognition molecules in every neuron mediating synapse formation, the number that match determining the strength of connection. Involvement of a large number of recognition molecules can explain why it has been difficult to find single gene loss-of-function mutations that alter connectivity. To test this hypothesis, we will select a pair of cells that are respectively synaptic and non-synaptic neighbors of the same cell. For example B-type ray neurons strongly synapse onto EF(1-3) interneurons while the A-type ray neurons contact EF(1-3) but do not synapse with them. Using cell-type gene expression profiling, we hope to identify genes expressed in the B-type neurons but not the A-type neurons that are candidates for mediating EF targeting. Their function can be tested by expressing them in the A-type neurons and examining whether this induces synapses with EF(1-3).

Emmons, Scott W. et al. (2009) International Worm Meeting "The posterior connectome of the C. elegans male."

Albertson, Donna G., Wang, Yi, Hall, David H., Xu, Meng, Thomson, Nicole, Emmons, Scott W., Jarrell, Travis

[International Worm Meeting, 2009]

For many biological systems knowledge of structure is key to understanding function. It was Sydney Brenner''s insight that the structure of the C. elegans nervous system could be determined and analyzed by means of genetics that provided the inspiration for C. elegans research (Brenner, 1974). For over 20 years, the completed wiring diagram of the C. elegans hermaphrodite has afforded a unique basis for genetic studies of worm behavior. Among C. elegans behaviors, the most complex motor program is the multi-step mating behavior of the male. Reconstruction of the male nervous system was initiated along with that of the hermaphrodite in the 1970''s (Sulston, Albertson and Thomson, 1980), but its completion has awaited development of the modern PC. Using electron micrographs from the MRC we digitized and analyzed using a software platform for annotation of electron micrographs from the computer screen, we have determined the connectivity among the neurons and muscles in the male tail, the posterior connectome. Reconstruction of the anterior nervous system is underway. The male posterior connectome consists of the interconnections among the processes of 175 neurons (85 male-specific and 90 shared with the hermaphrodite) and 65 muscles (41 male-specific and 24 shared). These cells are joined together in a complex network by some 8000 synapses, 4000 chemical and 4000 electrical, more than are contained in the entire hermaphrodite nervous system. The networks of chemical and electrical synapses are largely overlapping, suggesting parallel routes of information transfer and processing. Male-specific and shared neurons and muscles are fully integrated together in the network. Many of the shared neurons are sexually dimorphic, not only in having a more branching architecture and having synapses with male-specific neurons and muscles, but also in lacking some and gaining other synaptic interactions amongst themselves. In spite of its complex network architecture, potential circuits for the various steps of the mating program can be discerned in the connectivity diagram, in some cases revealing previously unsuspected roles for individual neurons or classes of neurons. The results provide an unprecedented opportunity not only to understand how a decision-making, multifunctional neural network processes sensory information in a coherent manner, selecting a choice among alternate behavioral outputs in a goal-oriented behavior, but also an opportunity and a challenge to understand how this incredibly complex structure, the connectome, is specified by the genome.

Wang, Yi et al. (2009) International Worm Meeting "The C. elegans male posterior connectome matures during early adulthood."

Wang, Yi, Emmons, Scott W., Bernstein, Max, Albertson, Donna G., Thomson, Nicole, Jarrell, Travis, Xu, Meng, Hall, David H.

[International Worm Meeting, 2009]

We are carrying out reconstructions of the C. elegans male nervous system from serial section electron micrographs created at the MRC in Cambridge, UK, during the 1970''s. At that time, several adult animals of different ages were sectioned and photographed through critical regions of the male tail containing the neural circuits that underlie copulatory behavior. Our most complete reconstruction is that of an animal known as N2Y, annotated as an "old adult." We have compared selected neurons in the pre-anal ganglion of N2Y to corresponding neurons in a second worm, JSI, annotated as a "young adult." The neural network of N2Y is more complex than that of JSI. The number of synapses and the number of synaptic partners of individual neurons increases more than fourfold. Comparison of individual neuron maps revealed that the reason appears to be that neuron processes in JSI are not fully grown out. The absence in JSI of many apparently major synaptic interactions present in N2Y made it appear doubtful whether JSI could have mated properly. Accordingly, we tested the mating ability of young males immediately after their molt to adulthood. When presented with five paralyzed hermaphrodites, only one of 6 just-matured males mated during the first 3 hr, whereas males matured overnight mate within the first 20 min. Mating began thereafter, suggesting 3-5 hr are necessary for maturation of the male nervous system. Maturation of the connectivity during adulthood raised the possibility that experience could influence the wiring process and might improve performance. In order to generate mature males that had never experienced sensory inputs associated with mating or mating-type behaviors, we allowed L4 males to mature overnight in liquid. When placed with hermaphrodites on plates, such males mated immediately and performed as well as males matured on plates with other animals. Conversely, the performance of several day old, experienced males was not improved over that of inexperienced males. Therefore we found no evidence that mating competence either requires or is improved by experience. It appears that the pattern of synaptic interactions necessary for efficient mating is fully established a few hours after the L4/adult molt and is sufficiently well-specified genetically to support mating behavior.

Oren-Suissa, Meital et al. (2015) International Worm Meeting "Her Brain, His Behavior: Dimorphic neuronal connectivity and behavior are established by sex-specific synapse pruning during development."

Oren-Suissa, Meital, Hobert, Oliver

[International Worm Meeting, 2015]

Sexually reproducing animals display sex-specific behaviors wired onto dimorphic connectivity patterns in the nervous system (Jarrell et al., 2012). The mechanisms underlying the development of sexually dimorphic nervous systems that consists mainly of shared neuronal types remain largely unknown.Here we dissect the regulatory programs that specify sexually dimorphic neuron identity. Specifically, we focus on the PHB sensory neurons, which function as chemosensory cells that negatively modulate reversals to odorant repellents (Hilliard et al., 2002). In hermaphrodites, PHBs synapse heavily onto command interneurons (AVAs) that control locomotion. Using in vivo trans-synaptic labeling (Feinberg et al., 2008), we confirm the EM data from Jarrell et al. that demonstrates these connections are absent in adult males, and instead, connections are made with AVG and LUA interneurons and male specific motor neurons. We find that PHBs in males do not process the same sensory cues as in hermaphrodites, but instead are repurposed to process cues involved in mating behavior. Inducible neuronal silencing of PHBs results in defective contact-induced backward locomotion of males during mating, as well as defective vulva location. Strikingly, we find that PHB-AVA synaptic connections are shared by both sexes at the L1 stage, therefore resulting in the sensory modulation of locomotion by repellents in both sexes. Later in development, the PHBs synapses to AVA interneurons are pruned in males, generating a sexually diverged synaptic state that generates dimorphic behaviors. Using cell-specific neuronal masculinization, we show that the pre-synaptic cell autonomously controls sex-specific pruning. Conversely, PHB-AVA synapses are pruned in hermaphrodites with masculinized PHBs, and the antagonistic behavior is lost. The regulatory programs that trigger this specific dimorphic connectivity are controlled by several members of the evolutionary conserved family of DM domain Zn-finger transcription factors, which function in neuron-type specific modules. dmd-5 and dmd-11 are dimorphically expressed in the male AVG neuron and are required for synaptic connectivity and male mating behavior.Our results suggest that sexual identity of individual neurons defines sex-specific synaptic targets and allows for diversification of behavioral outputs with a limited set of shared neurons.

Shi, Cheng et al. (2019) International Worm Meeting "Insulin-like peptides and the mTOR-TFEB pathway protect C. elegans hermaphrodites from Mating-induced Death."

Shi, Cheng, Murphy, Coleen

[International Worm Meeting, 2019]

C. elegans lifespan is shortened by mating, but must be delayed long enough for successful reproduction. Susceptibility to brief mating-induced death increases with age; surprisingly, this is not due to declining health, but to loss of protection upon self-sperm depletion. Self-sperm maintains expression of a DAF-2 insulin-like antagonist, INS-37, which promotes the nuclear localization of HLH-30/TFEB, a pro-longevity regulator. Mating induces the agonist INS-8, promoting HLH-30 nuclear exit and subsequent death. In opposition to the protective role of HLH-30 and DAF-16/FOXO, TOR/LET-363 and the IIS-regulated Zn-finger transcription factor PQM-1 promote seminal-fluid-induced killing. Self-sperm maintenance of nuclear HLH-30/TFEB allows hermaphrodites to resist mating-induced death, increasing the chances that mothers will survive through reproduction. The hijacking of the IIS pathway by males is combated by the mother's expression of an insulin antagonist that keeps her healthy through the activity of pro-longevity factors, as long as she has her own sperm to utilize. ** If possible, I would prefer that this abstract please be considered with the abstract submitted by Lauren N. Booth, Travis J. Maures, Robin W. Yeo, and Anne Brunet ("Self-sperm induce resistance to the detrimental effects of sexual encounters with males in hermaphroditic nematodes").

Cook, Steven J. et al. (2017) International Worm Meeting "Investigating the structure and sexual dimorphisms of the C. elegans connectome."

Bulow, Hannes E., Brittin, Christopher A., Nguyen, Ken C.Q., Hall, David H., Tang, Leo T.-H., Jarrell, Travis A., Cook, Steven J., Yakovlev, Maksim, Emmons, Scott W., Wang, Yi, Hobert, Oliver, Bayer, Emily A.

[International Worm Meeting, 2017]

We present the first whole-animal maps of synaptic connectivity, including anatomical connection strength, of both adult sexes of a species. Our results are based on analyses of legacy and new serial section electron micrographs (EMs) of the C. elegans nervous system and the tissues it innervates. In a graph representation of connectivity, the pathways of information flow can be arranged hierarchically, revealing a largely feed forward structure of shallow (1-5 synapses) depth. Our reconstruction has revealed that muscles and other end-organs are more extensively cross-connected than previously reported. Sensory information converges and diverges widely throughout the fully-connected neural network. The sexes differ not only by the addition of sex-specific neurons and muscles, but also at numerous points in the connectivity of shared neurons. Differences between the hermaphrodite and male reconstructions could be either inter-individual differences or differences due to genetic sex. To distinguish between these possibilities, we examined a subset of 7 synaptic connections that were respectively stronger in the male reconstruction, 4 that were stronger in the hermaphrodite reconstruction, and 4 that were similar, using in vivo trans-synaptic labeling. In each instance, the difference seen in the reconstructions was confirmed in multiple animals. Extrapolating these results to the number of statistically significant differences in the reconstructions, we conclude that there is an unexpectedly large number of sexually dimorphic connections. These connections were mainly located in the nerve ring, and embedded within the connectome at least one synapse away from any sex-specific neuron. Our results showed that AVA receives sex-specific input from ADL, ASH, and AVF in the hermaphrodite, while RIB receives sex-specific input from IL1, IL2, and RIA in the male. AIM, which has been reported to change its neurotransmitter from glutamate in the hermaphrodite to acetylcholine in the male, makes a strong male-specific connection to AIB. These hubs of sex-specific connectivity also maintained the majority of their sex-shared output. Our results suggest that the genetic sex of the nervous system allows for diverse synaptic patterns in a relatively small nervous system.

Bayer, Emily A et al. (2017) International Worm Meeting "Monoamines control the feeding-dependent emergence of"

Hobert, Oliver, Bayer, Emily A

[International Worm Meeting, 2017]

Sexual dimorphism is a widely acknowledged biological phenomenon, yet the mechanisms underlying specific developmental dimorphisms are largely unknown. As a result of the defined connectome available for both C. elegans males and hermaphrodites, it is clear that there are dimorphic synaptic wiring differences of neurons that are shared between male and hermaphrodites (1). Several of these dimorphisms are found in the phasmid sensory neurons, whose chemical synapses onto the command interneurons are sexually dimorphic, suggesting that they have sex-specific functions. Using GRASP transsynaptic labeling technology we have previously shown that these sexually dimorphic synaptic connections arise from a mixed juvenile state, with many eventually dimorphic adult connections present in both sexes initially and then restricted by sex-specific synaptic pruning (2). I have found that this sex-specific synaptic pruning is compromised in males that have been subjected to periods of starvation during the larval stages, whereas starved hermaphrodites are apparently unaffected. Correspondingly, males that have undergone periods of larval starvation retain their ability to respond to aversive cue, an ability normally lost due to synaptic pruning. I found that this starvation effect on synaptic pruning is mimicked by exogenously added octopamine in the presence of food, and completely rescued by exogeously added serotonin during starvation. The SSRI fluoxetine (Prozac) can also function to modulate the effects of starvation on male synaptic pruning. I will describe an octopamine receptor and a serotonin receptor that are required for this process. Taken together, our results show that experience can impact neuronal patterning in a sex-specific manner. 1. Jarrell TA, Wang Y, Bloniarz AE, Brittin CA, Xu M, Thomson JN, Albertson DG, Hall DH, Emmons SW. 2012. Science 337, 437-444. 2. Oren-Suissa M, Bayer EA, Hobert O. 2016. Nature, 533, 206-211.

Bayer, Emily et al. (2015) International Worm Meeting "Investigating the genetic basis of neuronal dimorphisms in C. elegans."

Oren-Suissa, Meital, Hobert, Oliver, Bayer, Emily

[International Worm Meeting, 2015]

Sexual dimorphism is a widely acknowledged biological phenomenon, yet the mechanisms underlying specific developmental dimorphisms are largely unknown. As a result of the defined connectome available for both C. elegans males and hermaphrodites, it is clear that there are dimorphic wiring differences in shared neurons between the adult animals of the two sexes (1). Several of these dimorphisms are found in the phasmid sensory neurons, whose synapses onto the command interneurons are sexually dimorphic, suggesting that they have sex-specific functions. I am confirming the predicted sex-specific functions of the phasmid neurons using behavioral assays. To study how these dimorphic patterns of connectivity (and function) are established, I am using transsynaptic labeling (GRASP technology; 2) to visualize both male-specific and hermaphrodite-specific synaptic connections of phasmid neurons. Sexual determination is regulated across many invertebrate and vertebrate species by the highly-conserved doublesex/mab (DM) domain genes. In C. elegans, the DM domain class contains 11 paralogs, the majority of which have no known function. I analyzed expression patterns for 8 of the dmd genes in both larval and adult stages in the two sexes to identify dimorphisms, and identified dmd-4 as a candidate for dimorphic regulation in the PHA and PHB phasmid neurons. dmd-4 is an embryonic lethal gene, so I am generating a conditional knockout allele and perform mosaic rescue analysis to determine if dmd-4 is part of the genetic regulatory system for dimorphism in the phasmid neurons, and if this function is cell-autonomous. Ultimately, we will seek to elucidate the genetic regulatory logic of dimorphisms in shared neurons.1. Jarrell TA, Wang Y, Bloniarz AE, Brittin CA, Xu M, Thomson JN, Albertson DG, Hall DH, Emmons SW. 2012. Science 337, 437-444.2. Feinberg EH, VanHoven MK, Bendesky A, Wang G, Fetter RD, Shen K, Bargmann CI. 2008. Neuron 57(3), 353-363.