IB404 - Caenorhabditis elegans 1 – Feb 81. C. elegans was being studied by a few UK naturalists in the 1960s when South African Sydney Brenner at Cambridge, England decided to leave bacterial molecular genetics for a simple animal that would allow detailed developmental and neurobiological/behavioral studies. He is now retired at the Salk Institute in San Diego.2. He recruited John Sulston to join him, and Sulston undertook the remarkable serial EM sectioning that allowed identification of every one of the 996 cells in the hermaphrodite (males have a few more cells). Sulston went on to run half of the genome project and later the British component of the human genome at Sanger Center.3. With a Harvard PhD, Robert Horvitz joined Sulston’s effort and discovered apoptosis (programmed cell death) in worms, then moved back to MIT. They received the Physiology/Medicine Nobel Prize in 2002, understood as a prize to the C. elegans field in general as a model animal.

4. The worm community is largely derived from Brenner’s postdocs and students, including Bob Waterston, who ran the WashU genome center.

5. They sequenced the genome using a clone-by-clone approach, with a physical map of ±17,500 cosmid clones with inserts of 20-40 kb in about 700 clusters. Gaps were spanned by an overlapping map of ±3,500 YACs with 150-500 kb inserts. This map was organized using fingerprinting of restriction enzyme digest patterns of these clones, looking for overlaps.

Lines at the top are YAC clones (names start with a Y). Below are cosmid clones. The yellow boxed cosmids are those sequenced.

6. A draft genome was published in 1998, and finished to the last base pair in 2001. 7. The genome is about 100 Mbp in six roughly equal size chromosomes, I-V and X.8. Genes are evenly spread across the chromosomes (light blue), but highly conserved genes (those with yeast matches – dark blue) tend to be in the middle of chromosomes, while various repeats and transposons are near the ends. This is the opposite of other animals, perhaps related to their distributed centromeres (worms are holocentric, with no single centromere).

9. Genes are generally reasonably small, with generally short introns, with a clear peak around 60 bp, but some introns contain transposons, and a few genes are in large introns of other genes.

10. Initial annotation using various automated methods came up with ~19,000 proteins-coding genes, and refinement has brought that near to ~20,000.

11. Initial analyses could only compare proteins with E. coli, yeast, and a subset of available human genes. Remarkably, about 36% of the predicted C. elegans genes had confident human matches, supporting model status for this worm.

12. Worm/yeast comparisons allow identification of basic eukaryotic cell proteins, as well as those specific to multicellular worms. Most fundamental proteins are simple orthologs, but cytoskeleton proteins have many paralogs in worms (red), having duplicated within animals.

13. Worms have many classes of proteins absent from yeasts, as well as many highly expanded protein families containing conserved domains. I worked up their ~1800 putative chemoreceptors (7TM proteins)!

14. About 15% of nematode genes are organized into operons of 2-8 genes, analogous to the operons of bacteria. The genes in these operons are of course transcribed together as a polycistronic transcript and hence are coordinately expressed, but the downstream genes are translated as a result of a trans-splicing event in which a short RNA leader is spliced onto the front of the internal gene mRNAs (SL in figure below).

Some classes of genes are commonly in operons, e.g. those involved in RNA degradation, the basic machinery of transcription, RNA splicing and translation, as well as those that encode mitochondrial proteins; generally ancient conserved genes. Other gene classes are never in operons, e.g. the ~1800 chemoreceptor genes.

The 5'-most gene is mitochondrial ATPase inhibitor-1 (mai-1). This gene is not TRANS-SPLICED. The two downstream genes, glyceraldehyde 3-phosphate dehydrogenase-2 (gpd-2) and glyceraldehyde 3-phosphate dehydrogenase-3 (gpd-3), encode isoforms of a glycolytic enzyme.

15. WormBase is a remarkable database of all features of C. elegans biology, largely centered around the genome sequence and annotation. Many “tracks” can be viewed, including here the gene name, model of exons/introns, alternative splicing, cDNA and EST sequence evidence, C. briggsae alignments, RNAi experiments, and ORFeome project, e.g. synaptotagmin is an 8 exon gene, with alternative splicing of exon 6.

In lieu of generating mutants of every gene, screens of knockdowns of gene expression can be conducted using RNAi in worms. The neat trick is that one can feed worms bacteria expressing double-stranded RNA and RNAi against that gene is induced. So a region of each gene is amplified by PCR from worms, cloned into a special plasmid vector (right), and these recombinant plasmids are transformed into a mutant strain of E. coli that does not have much RNase activity. A library of ~18,000 such bacterial strains was generated and then worms fed each of these and phenotypes examined as in a mutant screen. E.g. a GFP transgene (a) was knocked down by feeding a RNAi construct for 24 (b) or 48 (c) hours.

About 10% of genes knocked down had obvious phenotypes such as lethality, sterility, or growth defects. Most of these genes encode ancient conserved proteins involved in fundamental cellular processes. Genes yielding viable phenotypes were more likely involved in signaling and other functions. ±700 genes were lethals or just 5% of those tested, which seems somewhat low to me, probably because the knockdowns are not complete; they are leaky.

Lethal

Other phenotype

Another RNAi screen looked for worms that stored more or less fat, indicated by a red fat-binding dye. Major players in this pathway were identified, such as the insulin-like protein (daf-2) and transcription factors that regulate its expression (daf-16), and various enzymes involved in fat metabolism. For example, in daf-2 mutants that accumulate fat, RNAi against a fatty acid elongase counteracts that - last two panels. Remarkably about 20 of these ±400 genes are 7TM chemoreceptors, perhaps indicating importance of detection of fats either during feeding or during accumulation.

Another RNAi screen searched for genes that when knocked down led to increased lifespan. It was already known that a remarkable variety of mutant genes can lead to increased lifespan in worms, up to about 50% per gene and much more for combinations of mutants. The inevitable conclusion is that much of lifespan regulation involves evolved shortness of lifespan. This RNAi screen detected numerous mitochondrial function genes, e.g. cytochrome oxidase subunits.

The interpretation is that some, but not complete, knockdown of mitochondrial function reduces oxidative stress on the organism, much as restricted calorie diets in mice, monkeys, and other animals increases lifespan.