Title: Tyramine functions independently of octopamine in the Caenorhabditis elegans nervous system
Summary: The year was 2005, and no one knew if tyramine was important in C. elegans. It was known that octopamine was found in C. elegans extracts, and that exogenous octopamine could manipulate C. elegans behavior, but no one had yet found a tyramine hydroxylase gene that could turn tyramine into octopamine in animals. And importantly, no one knew how octopamine and tyramine differed in their roles in worm behavior.
Until this paper was published.
Ok, so what is tyramine? Tyramine is decarboxylated tyrosine. ... what is octopamine? It's tyramine, with an extra hydroxyl group added at the beta carbon. Here's a general diagram you can stare at for a while. These are also closely related to dopamine, and I've left dopamine on the diagram too (this was taken from the dopamine wiki page).
(Note that the enzymes shown here are not necessarily all found in worms!)
Mark and coworkers first focused on finding the tyrosine decarboxylase and tyramine beta-hydroxylase genes in C. elegans. Focusing first on the second step in the pathway, they used a previously-identified tyramine beta-hydroxylase gene from drosophila to identify a single gene with significant homology (32% identity in the ORF) in the C. elegans genome. The gene encodes a 560 amino-acid protein.
By PCR screening a mutant library, they found two mutants with deletion alleles that remove pretty much the middle 3-4 exons of this gene. By HPLC, these mutants do not contain detectable octopamine, whereas wildtype animals contain 5 pmol per mg wet weight. (Ballparking a worm as mostly water, this corresponds to 5 umol per L or 5uM octopamine on average throughout the worm.) Perhaps not surprisingly, since this is the second step of the octopamine synthesis pathway, the precursor (tyramine) builds up to 20x WT levels!
Finding the gene for the first step of the pathway (tyrosine decarboxylase) was a bit more challenging. There were 5 putative aromatic amino acid decarboxylase (AADC) genes, based on similarity to DOPA decarboxylases (for making dopamine) in insects and mammals. They obtained mutants for all 5 alleles (this is apparently so easy for them that it literally gets one line in the manuscript), and found that only one changes the tyramine levels (as determined by 2D-TLC). They also noted that this mutant lacked octopamine (as would be expected!). No assays were done on tyrosine levels though - I wonder if these went up? Or alternatively on dopamine levels (which might have increased due to increased tyrosine -- or decreased because dopamine could potentially be synthesized by hydroxylating tyramine)?
As one final control, they note that in drosophila, DOPA decarboxylase can also act as a tyrosine decarboxyla (albeit at a much slower rate). They thus assay tyrosine decarboxylase activity in extracts from worms in which the DOPA decarboxylase has been knocked out, and not that it is effectively equal to the wild-type suggesting the gene they've found (they call it tdc-1) is the sole tyrosine decarboxylase in C. elegans.
Where are these genes expressed? In typical fashion, they 'generated rabbit polyclonal antibodies' like it was no big deal and validate them via western blotting. By immunostaining whole animals, they find that the enzyme for step 2 - tbh-1 (tyramine beta-hydroxylase) - is expressed pretty much in a single pair of head interneurons: RIC. I looked RIC up on wormbase, which provided no real context other than "RIC uses octopamine" which I guess I kind of already knew from writing the beginning of this paragraph. interestingly, in RIC, tbh-1 is predominantly NOT present in the soma, but instead in puncta in the processes in the nerve ring. I'm actually not sure how they know that those processes belong to RIC... but I'll take their word for it since they're the worm experts.
RIC shares gap junctions with ASH and AVK. ASH is the main nociceptive neuron, responding to nose touch, high osmolarity, detergents, and a handful of other 'noxious' stimuli. Wormatlas has significantly less to say about AVK. RIC also forms direct synapses onto AVA (the backwards command neuron) and SMD (4 dorsal motor head neurons) and SMBD (bilateral pair of dorsal head motor neurons).
Thinking back on this, Mark spoke this summer at Woods Hole, about some work in which he looked at cessation of head motion during backwards movement. That phenotype starts to make a little bit of sense in light of the connections mentioned above.
They also note that they find tbh-1 in gonadal sheath cells of adults.
As has seemingly become the characteristic organization in this paper, they shift back from the 2nd step of octopamine synthesis to the first step (tyrosine decarboxylase), and note that tdc-1 is diffuse throughout cell bodies and processes, in RIC and the gonadal sheath cells, suggesting that these cells do indeed produce both enzymes required to make octopamine, and further that the cells probably produce tyramine in the cytosol, and then in the vesicle or during vesicle packing, convert it to octopamine.
tdc-1 was also expressed in some cells that do not express tbh-1, suggesting that these cells might use tyramine directly! These cells included RIM (head motor neurons) and the UV1 uterine cells.
tdc-1 and tbh-1 mutants were viable, 'healthy' and had normal brood sizes. Therefore octopamine signaling is clearly not essential. On the other hand, tdc-1 mutants (which have neither octopamine or tyramine) lay their eggs earlier than normal, whereas tbh-1 mutants do not, suggesting that egg-laying timing is modulated by tyramine. Similarly, exogeneous tyramine delays egg laying. This is opposite of the effects of serotonin (secreted by the HSN neurons), which induces egg laying and the tph-1 mutant is egg-laying defective. The tph-1; tdc-1; double mutant is also egg-laying defective. Put differently, in the absence of both serotonin (egg-laying-promoting) and tyramine (egg-laying-delaying), the basal behavior of the worm is delayed egg-laying.
Aside from egg-laying defects, tdc-1 mutants don't suppress head motion during soft-head-touch-induced reversals, and tbh-1 mutants do, suggesting a role for tyramine. Soft head touch is thought to be sensed by ALM/AVM mechanosensory neurons. Interestingly, backwards locomotion isn't always associated with cessation of head motion. During spontaneous reversal, or after harsh touch (e.g. a platinum wire, which is sensed by PVD apparently), reversals occur where head motion doesn't cease - it's only during soft touch that head motion ceases. Note that in tdc-1 mutants, worms still back up (though for slightly less time than WT, and in a jerky manner) - they just fail to suppress their head motions.
As background, apparently there are 8 radially symmetric sections to the head muscles, which are innervated by SMD and RMD (cholinergic and coupled by gap junctions), RME (GABAergic) and RIM which is tyraminergic.
They argue that RIM tyraminergic signaling relaxes the head and suppresses head oscillation during backing up. Ablating RIM in wild type animals eliminated the ability to suppress head oscillations in reversal after soft touch. Ablating RIC had no effect on head oscillations.
AVA ablations had a similar effect as RIM ablations (RIM and AVA share a gap junction!). AVA-ablated animals still back in response to anterior touch, but they're uncoordinated, and they do not suppress suppress head motion.
Questions that came up while I was reading this: (oh man, this paper is like two theses, it took me forever to get through it and I had a lot of questions)
- It's interesting to think about how ablating a cell might affect other gap-junction-neighboring cells. Presumably it at least briefly makes a huge shunt? Does this shunt get repaired?
- How would a worm navigating a 3d environment distinguish soft touch from a self-induced stimulus (e.g, running into a low ceiling?)
- RIM's tyraminergic signaling seems like it might have a overriding effect on other inputs to the head muscles (or the other head motor neurons directly). (An alternative hypothesis is that ALM/AVM shut down these other inputs so that the muscles only receive RIM's input). How does this work? How does RIM shut down head motion?
- Is there a head motion CPG? How well worked out is this? They speculate about it in this paper
- Could you image octopamine formation in vesicles? Does it occur during packaging? Or is the enzyme colocalized in vesicles and simply rapidly converts all tyramine into octopamine?
- What is the concentration and number of neurotransmitter molecules in a single synaptic vesicle? wikipedia claims that the V1 visual cortex has vesicles with a mean diameter of 40 nm (0.03 aL) [no word on how different this might be for C elegans]. One molecule per vesicle is still at a concentration of 50 uM. For enzymes with nanomolar Km, all the enzymes are operating at their maximal velocity! Strikingly, a bit of googling suggests that in neuromuscular junctions, there could be 10000 molecules/vesicle, yielding 10s of mM concentration of neurotransmitter in a vesicle!
- There are all these ideas of a 'minimal cell' in microbiology and cell biology. What constitutes a minimal organism? Clearly, tyramine and octopamine are not 'necessary' for survival and reproduction. Similarly with serotonin, as we saw a few days back.
- Could you construct a minimal worm?
- On a related note - could you construct a humanized worm? (Not relevant to this paper specifically, but something I wanted to get written down regardless).
- Are tyramine and octopamine secreted extra-synaptically? or are they only transmitted locally?