(Note: These notes from talks I sat in today at SICB are largely unedited. Apologies for typos!)
I sat in the neuroecology symposium. It started with Richard Zimmer, talking about the concept of keystone molecules: those that have more impact that. You would predict by abundance, started off with tetrodotoxin, a defense molecule in salamanders.
Then, he moved on to talking about a glycoprotein used as a contact cue in rocky beaches. bArnacles seem to make it as part of their shell, which gets used as both a settlement cue by other barnacles and as a predatory cue by predatory whelks.
Moved to starfish as keystone predators, and starfish will come to a single compound, another glycoprotein. KEYSTONin protein drives predation on mussels. Take it out, and you completely lose starfish predation. There are hints mussels use the same molecule to build their shells. Closing question is how many effects described as biotic interactions are actually chemical ones.
Ashlee Rowe was next. I had worked with her before in a nonciception sympsium. She starts out pointing that many species have developed neurotoxins to interact with ion channels as defenses. One subset of these neurotoxins cause extreme pain. In particular, she is working with the Arizona bark scorpion Centroides scorpions. Their neurotoxins bins to sodium and potassium channels in neurons. The neurotoxins cause spontaneous, prolonged action potentials.
These scorpions are preyed upon by grasshopper mice, however. Grasshopper mice also eat non-painful scorpions. The mice are quite resistant to the venom, not killed, and causing little pain. The venom activates action potentials, but the signal is blocked quickly so the signal doesn't reach the grasshopper mouse brain.
But drop the mice know the difference in pain? Grasshopper mice prefer a non-painful scorpion over a painful one. But when the stingers were blocked, mice much more likely to eat the painful scorpion (when sized matched). And naive mice will go for scorpions at random - if stings are blocked. Is the stings are unblocked, the captive bred mice won't bother to attack the scorpions at all. In the real world, the painful scorpions are abundant, but skinny and painful, so there are trade offs.
Jessica Fox was next on fly mechanoreceptors. What determines a fly compared to other flying insects is that they have a sensory organ: the halteres. Haltere have 100-300 sensible at the base, which provides balance. They cannot fly without halteres. The halteres are incredibly precise in tracking the phase of their movements, up to as fast as their machine would got (150 Hz). The sensory population shows range fractionation to the phase of the halteres movement.
Some flies activate their halteres when walking. One of the most recent fly clades quite consistently move their halteres while walking. These movements are not driven by flight muscles, but by intrinsic haltere muscles. When walking, the halteres can take any phase relationship.
They created a fly Olympiad for flies with ablated halteres. The only task that haltere ablated flies had a problem with was walking up walls and falling off walls when given a surprising vibration (cup drop). Talking about mutations in "early" and "late" stages of evolution (whatever those are) in guppies. Lots of convergence in guppy phenotype. Looking at gene expression pathways in these guys.
Kim Hoke was taking about responses across different timescales. What determines the response of an animal to another member of its species at any given time. Many possibilities, but one one happens.
When lab-reared animals are reared with predator cues, several protein expressions change. She is also able compare fishes from different drainages. There is some concordance from different lineages, but there is a lot of smear. It's not a tight relationship. She is suggesting that network homeostasis could drive coordinated divergence, but this was not the case. Instead, there seems to be a few central genes in the module that are evolving quickly. Possibility are connected genes evolve in long term, but not short term. So convergence in phenotype has only moderate similarities in gene expression.
Observation around the halfway point in neuroecology symposium: both neurobiology and ecology seems to be optional in calling your research “neuroecology.”
Gabby Nevitt was studying chemical comminucation in pelagic seabirds like albatrosses, seabirds, petrels, and so on. How do they find each other to mate? Because they mate for life. They have huge olfactory turbinates and olfactory bulbs.
The major histocompatibility complex (MHc) often implication in mate choice, but very little work has been done in the wild, it tends to be underpowered, and the results are not consistent. So she went looking at MHC a storm petrel, which is a very common bird.
Having examined over 1,000 birds, they are seeing a lot of long tails in the alleles I in the population. But there is no assortative mating based on MHC in the petrels. There are a lot of underpowered studies out there, contributing to controversy.
They have also developed a chick choice assay, based on fake nests. Chicks can tell their parents' scent, and pick it over an MHC matched scent.
Jeremy Niven is looking at energy efficiency of potassium channels in eyes. Wants to know the trade off between energy consumption and information fidelity. The cost of action potential is related to movement of ions across the membranes: in particular, the long term potassium sodium pumps.
The size and shape of action potentials vary significantly in their energy cost. Squid action potential is wasteful! Spending way more energy than it needs to. Cortical relay cells are very efficient.
The overlap between sodium and potassium currents is a major driver of the energy costs of neurons. And neurons vary a lot in this regard.
To address this, Jeremy looks at fly photoreceptors. These neurons are non-spiking. They can respond to single photon, very little signal at low light, and reasonable signal in high light.
If the fly photoreceptors process more information, they pay a high energy cost. Modeled ATP costs per bit. Big flies can get up to pretty high bit rates, but high energy costs, too. Slow and fast photoreceptors also have different properties.
Why have a voltage gated potassium channel in the icky photoreceptor? By changing the resistance of the cell membrane, because the amount of membrane in a photoreceptor is constrained. Potassium channels are improving bandwidth and lowering energy costs. Eyes pull back on maximum information, and save huge amounts of energy by doing so.
Michael Markham was looking at metabolic costs of communication signals. For bats and electric fish, you can’t stay quiet without going “blind,” because communication is linked to sensory signaling. The cost of an action potential is a cost for sodium: 1 ATP for every 3 sodium ions.
Weakly electric fish modulate their electric organ discharges to save energy between day and night, or depending on social system. But some fish generate hundred of discharges every second it is alive. Eigenmammia terminates it action potential with a sodium gated potassium channel. Seems to be a mechanism to reduce overlap you see in sodium and potassium currents, which is wasteful. 30% more efficient was predicted.
The cost of the electric organ discharge is one of the most expensive known in the animal kingdom. You can see increases in amplitude depending on when the anime is fed. Food pumps up the amplitude of the signal, but this is not due to an absolute energetic limitation. It seems to be modulated by leptin.
Next question is how do the ion pumps keep up with the demands for ions of the channels running the action potentials? There are a lot of them, for one, but they are probably still not enough: there are probably specialized high volume sodium/potassium pumps.
Markham proposes that electric fish could be sentinel species for climate change, they are living on the energetic edge to live in these muddy waters. They are probably going to be very sensitive to changes in primary productivity.
Jeff Riffell is looking at insect olfactory neuroecology. The interaction between plants and pollinators have huge, community wide effects that structure communities. Sometimes, the communities are wide, with many pollinators per species, but sometimes it is more specific.
Platanthera are interesting orchids, for which some are mostly pollinated by mosquitoes (which is unusual), while some are flies, moths, etc.. How do their scents cause pollinator attraction? You can make good predictions about what pollinates a species based on what odor chemicals the plant is generating.
Of the mosquito pollinated orchids, only 3 species of mosquitoes act at pollinators. The odors seem to be very specific attractants.
Giving the odor to a flying mosquito in the lab changes the flight behaviour of mosquitoes: they “surge” in wing eat frequency. They respond to the scent of a moth-pollinated orchid scent by slowing a bit, although it smells much more intense to a human.
There are some antennas lobe neurons that respond very strongly to the orchid scents, while others are suppressed. The population of neurons are able to generalize between the mosquito attractive scents versus other scents, like moth pollinated orchid scents.
Marie Suver was looking at visually guided flying in fruit flies. VS and HS cells are visual neurons that seem to detect self-motion. She was able to characterize their responses to pitch and roll, and the ehnext step was to track what happened downstream from those. She found several neurons whose response was predicted by the tuning of the HS and VS neurons.
The descending inter neurons, in turn, cause various body responses, each of which is a various body axes. You see similar patterns in several vertebrates, including pigeons and rabbits, leading to some speculation that this is a common control system.
Emma Coddington looks at cortisol in salamander clasping. Vasotocin seems to be criticized cal, acting as a gate to all sensorimotor activist in clasping. Cortisol released by stress stops the clasping related behaviours.
The timing of these two hormones, and clasping, matters. The hormone or behaviour changes the salamander's response to stress (cotrtisol).
In thinking about this, she found there were limitations to thinking about time frames. Usually people just called them acute and chronic treatments. Chronic in particular was ill defined and confusing.
Cortisol interferes with the endocytosis of vasotocin in a seasonally-specific way. In no breeding animals, cortisol does very little, but has bigger effects in breeding season. Cortisol also effects brain regions differentially, as you'd expect.
Corticosterone does not affect intrinsic neural properties, but seems to tweak synaptic communication.
Hannah Ter Hofstede asks how a new signal can evolve? After all, you normally think of a sender and a received, and "signal" implies adaptation. Not incidental. There can be precursors to new signals in either a sender or receiver. But the sensitive should exist before the evolution of a signal. A sender is generally not expected to make a signal similar to a predator, say, but rather to exploit something a receiver already finds attractive.
One group of crickets, the eneopterans, produces very high frequency calls though, that normally would be categorized by crickets as an aversive bat-like sound. These female crickets don't walk towards male calling songs; the male approaches the female, who produces a brief vibrating after each male call.
Crickets will fly away from high frequency sounds, and when standing on the ground, all crickets tested show a startle response to a high frequency sound. This could be a precursor to the female "push-up."
The scenario is that males exploited female startle, so they didn't have to wait for females to come to them. Females no longer produce startle response to just any high frequency sound; it has to be the specific timing that males make. So it's now a true signal.
The females responses are tuned to about 12-15kHz, roughly that of AN2 bat detector in most species. The neural responses... Did not see any response to 5 kHz, seeming to imply AN1 is inactive or lost. AN2 type responses seemed fine. But it's not clear if the ascending neurons, especially, the AN1 homologous, are lost or just modified. All the ascending neurons respond up in righ frequency.
Exploiting predator cues may be more common than wet bought, starting off as a way for sender to exploit the receiver.
Later that night, away from the neuroscology symposium, Vincent Careau give the George Bartholomew lecture. I was grumpy about his characterization of crayfish as non-charismatic, but oh well. He was interested in the evolution of physiology, and how physiology affected evolutionary patterns. Are physiology I, performance and behaviour related in evolution? Are these complex traits evolving in a correlated way?
BMR varies by six fold even after you take body mass into account. There are a lot of explanations for this, but one might have been overlooked is behaviour. Used an open field test used in psychology, and found BMR was negatively correlated with open field behaviour.
His first species though, showed no clear relationship between the two, but later, once pedigree was taken into account, there did seem to be a correlation between the two. So how were these apparently contradictory results? It looks like it may be a species specific effect.
Next project was looking at effects of botflies on a chipmunk population. Botfly larvae on chipmunks was correlated with high BMR. Botfly infection also caused reduction in overwinter survival. If a chipmunk ever had more than 4 botflies, they didn't survive. Suggests that this is due to a reduction in oxygen consumption ability me which chimpmunks need to come out of torpor.
Careau next talked about selection for wheel running in mice. In 20 generations, selected mice run three times as much as control lines. But after generation 20, there seemed to be no increase in wheel running, suggesting there is a selection limit.
He hypothesized that wheel running on days 1-4 might constrain wheel running on days 5-6. Selection was apparent over 10 generations.
As an aside, he showed the equipment that has been used to gather the wheel running data, the experiment was started in 1993, and is still running on the same equipment: a DOS PC and floppy disks.