New York Area C. elegans Discussion Group

Organizers: Jane Hubbard, Skirball Institute of Biomolecular Medicine; Shai Shaham, Rockefeller University; Cathy Savage-Dunn, Queens College CUNY

Each meeting will include four presentations by graduate students, postdocs, or by laboratory heads. Talks are selected from area laboratories by the program committee with an emphasis on new and emerging data.

The Discussion Group will celebrate the 2008 Nobel Prize in Chemistry with Dr. Martin Chalfie.

Speaker Abstracts

AMPH-1/Amphiphysin/BIN1: A Novel Regulator of Endocytic Recycling
Saumya Pant (Barth Grant), Rutgers University

Receptor-mediated endocytosis is crucial for the internalization of receptor-bound ligands of many types. Cargo-laden internalized vesicles fuse with early endosomes. Endosome maturation allows sorting with degradation in the late endosomes and lysosomes or recycling to the plasma membrane to participate in further rounds of endocytosis. Endocytic recycling maintains cellular homeostasis of membrane and lipid components and in polarized cells it allows accurate sorting to maintain distinct apical and basolateral domains. The Grant Lab utilizes Caenorhabditis elegans as a model system to uncover the molecular players involved in endocytic recycling. We have previously established that the RME-1 protein is found on the endocytic recycling compartment and mediates endocytic recycling in several C. elegans tissues. RME-1 is a conserved protein homologous to four mammalian proteins EHD1-EHD4. The protein is an ATPase with an ATP binding N-terminal P-loop that is critical for membrane association and homo-oligomerization. The recent EHD2 crystal structure reveals that, while being ATPases, EHD/RME-1 family proteins bear structural similarity to the clathrin-coated pit GTPase Dynamin. The C-termini of RME-1 family proteins contain a single Eps15-homology (EH) domain. Classically, EH domains are found in endocytosis proteins functioning as protein interaction motifs, binding target proteins through Asparagine-Proline-Phenylalanine (NPF) motifs. Mammalian EHD proteins can bind to interacting proteins such as Syndapin, Rabenosyn5, EHBP1, and Numb in this manner.

Utilizing an RNAi based approach in a GFP-RME-1 expressing worm strain, we screened for proteins causing alteration in recycling endosome morphology. We focused on predicted worm proteins bearing multiple NPF motifs. Through this screen we identified C. elegans AMPH-1/Amphiphysin as a novel regulator of endocytic recycling. We utilized genetic analysis to establish a novel role for AMPH-1 in the regulation of recycling endosome morphology and recycling of transmembrane cargo. We demonstrate that endogenous RME-1 colocalizes with and associates with endogenous AMPH-1, and this association has a role in recycling. We show that in vitro recombinant purified RME-1 and AMPH-1 bind and tubulate phosphatidylserine liposomes independently, but their properties are distinct when in complex. We propose that the AMPH-1/RME-1 complex functions at the recycling endosome, promoting tubulation and/or scission of cargo carriers destined for the plasma membrane. We further show phylogenetic conservation of the role of Amphiphysin family proteins in endosome function using siRNA analysis in HeLa cells.

Distinct Neuronal Soluble Guanylate Cyclases Detect Increases and Decreases in Environmental Oxygen Levels
Manuel Zimmer (Cori Bargmann), Rockefeller University

Animals monitor internal and external oxygen (O₂) levels using specialized sensory systems to drive adaptive behaviors. Atypical soluble guanylate cyclases (sGCs) have been proposed to act as rapid neuronal O₂ sensors, but the mechanism for encoding different O₂ levels is unknown. Using quantitative behavioral assays and calcium imaging of neural activity in Caenorhabditis elegans, we show that two sets of sGC-expressing sensory neurons detect opposite changes in environmental oxygen levels to drive O₂-dependent behaviors. BAG sensory neurons are activated by downshifts from high to low O₂, and URX sensory neurons are activated by upshifts from low to high O₂. Atypical sGCs encoded by gcy-31and gcy-33are expressed in BAG neurons and are required to detect O₂ downshifts, and sGCs encoded by gcy-35 and gcy-36 are expressed in URX neurons and are required to detect O₂ upshifts. The reciprocal sensory properties of BAG and URX are directly specified by sGCs, as misexpression of URX sGC genes in BAG neurons causes these cells to detect O₂ upshifts. These results provide genetic evidence that sGCs act as primary O₂ sensors, and show that diversification of sGCs allows sensory neurons to generate separate representations of increases and decreases in environmental oxygen.

Automated Screening for Cell Fate Mutants
Maria Doitsidou (Oliver Hobert), Columbia University

We describe an automated method to isolate mutant Caenorhabditis elegans that do not appropriately execute cellular differentiation programs. We used a fluorescence-activated sorting mechanism implemented in the COPAS Biosort machine to isolate mutants with subtle alterations in the cellular specificity of GFP expression. This methodology is considerably more efficient than comparable manual screens and enabled us to isolate mutants in which dopamine neurons do not differentiate appropriately.

Caenorhabditis elegans Mutant Cloning by Whole-genome Sequencing
Sumeet Sarin (Oliver Hobert), Columbia University

Identification of the molecular lesion in Caenorhabditis elegans mutants isolated through forward genetic screens usually involves time-consuming genetic mapping. We used Illumina deep sequencing technology to sequence a complete, mutant C. elegans genome and thus pinpointed a single-nucleotide mutation in the genome that affects a neuronal cell fate decision. This constitutes a proof-of-principle for using whole-genome sequencing to analyze C. elegans mutants.

Evolution of Introns in Caenorhabditis and other Rhabditid Species
Karin Kiontke (David Fitch), New York University

Since the discovery of spliceosomal introns in 1977, their origin and evolution was the a matter of much attention. It is likely that spliceosomal introns evolved in Eukaryota. Recent studies suggest that the genome of the eukaryote ancestor was intron rich, but in extant species, the number of introns varies by 5 orders of magnitude. With increasing numbers of genome projects, it is now clear that introns are gained and lost and that the rate of gains and losses varies across clades and genes. The comparison of whole genomes of this small number of distantly related species, however, cannot asses at which scale these rate variations occur and how frequent intron evolution events really are.

We studied intron evolution in one highly conserved gene (RNA polymerase II) of 48 closely related rhabditid nematode species and several outgroup representatives. This approach allows us to resolve intron evolution events at a fine scale. Within nematodes, we find between one (e.g. in C. briggsae) and 20 introns (in Rhabditoides regina) per species in 50 different positions along 2800 nt of the gene. We used weighted maximum parsimony (MP) and two likelihood methods to reconstruct intron gains and losses within rhabditid nematodes based on a molecular phylogeny which was derived from the sequences of three genes. Our analysis showed that even within this clade, the rate of intron loss and gain is highly variable. Some lineages show barely any change, whereas in other lineages, losses and/or gains were frequent. Caenorhabditis experienced the most dramatic intron loss. There is no evidence that gains and losses are correlated. Even though all applied models assumed losses to be considerably more frequent than gains, we found at least twice as many instances of parallel gains of introns in the same position than was estimated previously. Among these parallel gains are at least two regains in positions where an intron was lost earlier in the lineage. Such regains were thought to be highly improbable. Using weighted MP and maximum likelihood, we reconstructed the sequence around all intron insertion sites within rhabditids. These sequences concur with the "protosplice sequence" proposed by earlier authors. We conclude that within nematodes, intron evolution events are more frequent than previously thought and occur at greatly different rates. Parallel gains of introns in the same position are frequent, consistent with the finding that introns are inserted into non-random sites.