Freie Universität Berlin

The marine snail Aplysia is perhaps best known for its Nobel-award-winning learning ability. In the classic work, a small water jet to the body of the animal was paired with electric shock to accomplish classical (or Pavlovian) conditioning. Since about the 1930s, it has been a longstanding debate among psychologists whether classical and operant conditioning can be explained by similar underlying learning mechanisms. To answer this question, new conditioning protocols have been established in Aplysia, to make them as comparable as possible.

The feeding behavior of this snail can be conditioned both operantly and classically. As you can see in this video:

Aplysia snails bite spontaneously and if every bite is followed by a reward (either a squirt of seaweed juice or an electric stimulus to the nerve mediating food reward), the animal will learn to bite more, even in the absence of food. This protocol thus establishes operant conditioning. Classical conditioning is established when you touch the lips of the animal with a brush and then feed the animal, irrespective of what it did before or after the brush stimulus. The effect of both protocols on feeding is the same: feeding is enhanced, either spontanous feeding (operant) or feeding elicited by a brush stimulus (classical).

The nice thing about Aplysia is that you can take its nervous system out, place it in a dish and still perform experiments. Using these techniques, researchers have found out that neuron B51 is critical for these kinds of learning: if B51 fires, the animal is feeding, if B51 is not firing, even if the mouthparts move, it is not a feeding movement. Thus, not surprisingly, operant learning makes B51 more likely to fire (Brembs et al. 2002). In classical conditioning, however, it’s the reverse: B51 is less likely to fire, even though there is more feeding after conditioning. Apparently, this can be explained by the sensory input from the brush to B51 being so potentiated, that it overcomes the reduction in B51 excitability.

How can two memories form in a single neuron? This question can now also be tackled, because then graduate student Fred Lorenzetti developed a single-neuron analog of both classical and operant conditioning. In operant conditioning the neuron is being depolarized until it fires, just as it would be during feeding, and then a puff of dopamine is applied, just as the esophageal nerve mediating the food-reward would do. In classical conditioning, the neurotransmitter Acetylcholine (ACh) is puffed onto the neuron (mimicking the brush-CS) and then dopamine (US).

Using this technique, Lorenzetti et al. have now found out that after such CS-US pairing in the dish, B51 responded more strongly to ACh puffs (just as intact animals responded with feeding to brush strokes), while general excitability of B51 was reduced, just like after classical conditioning of the intact animal. The D1-like dopamine receptor that mediates the changes after operant conditioning, doesn’t seem to be involved in the changes brought about by classical conditioning, as driving it with an agonist failed to bring about the changes in B51.

Thus, it seems as if dopamine acts through different receptors depending on what just happened prior to dopamine application: if the neuron fired during feeding, a general increase in excitability is observed, if the neuron is not firing, but ACh is applied, then a different dopamine receptor is activated and leads to a decrease in excitability.

Lorenzetti, F., Baxter, D., & Byrne, J. (2011). Classical Conditioning Analog Enhanced Acetylcholine Responses But Reduced Excitability of an Identified Neuron Journal of Neuroscience, 31 (41), 14789-14793 DOI: 10.1523/JNEUROSCI.1256-11.2011

A short communication was published in November 2011 (Neurobiologie) based on a study from 2005 (Journal of Neuroscience).

In the blowfly Phormia regina, the authors investigated the effect on PER when flies are (pre-) exposed to odors. They found that feeding threshold to sugar increased in the presence of the repellent D-limonene and decreased in the presence of the attractant dithiothreitol (DTT). The levels of tyramine (TA) and octopamine (OA) were changed with feeding threshold in flies experienced by flavored food. Injection experiments with OA, TA, or their agonist and antagonist indicated that TA more directly mediates feeding threshold determination.

In the very recent communication the authors cloned tyrosine decarboxylase (Tdc) and tyrosine beta-hydroxylase (Tbh) cDNAs from Phormia regina (PregTdc and PregTbh), the enzymes catalyzing OA and TA. PregTdc and PregTbh had different expression patterns throughout the animal. Exposure to the repellent indeed changed the expression of the enzymes in antenna. Localization of immunoreactive material and receptors gave a hint at the underlying mechanism how aversive information is transferred.

Ishida, Y., & Ozaki, M. (2011). Aversive odorant causing appetite decrease downregulates tyrosine decarboxylase gene expression in the olfactory receptor neuron of the blowfly, Phormia regina Naturwissenschaften DOI: 10.1007/s00114-011-0865-1

Nisimura, T. (2005). Experiential Effects of Appetitive and Nonappetitive Odors on Feeding Behavior in the Blowfly, Phormia regina: A Putative Role for Tyramine in Appetite Regulation Journal of Neuroscience, 25 (33), 7507-7516 DOI: 10.1523/JNEUROSCI.1862-05.2005

In a not so recent paper (January 2011), a nice report by Ann-Shyn Chiang group was published in PNAS.

Authors use both pharmacology and genetic trick to show that serotonin is involved in an aversive olfactory memory phase called ARM. Serotonin seems released by the DPM neuron (together with the amnesiac neuropeptide?), and the information seem to be received in the alpha/beta lobes of the mushroom bodies via the 5HT receptor A.

Three hours after training, it is known that the aversive memory retention score (after one cycle of conditioning) depends on two memory systems: a labile, cold shock sensitive memory, and ARM (anesthesia resistant memory, this latter one does not depend on de nuovo protein synthesis). Authors used cold shock to show that their treatments are affecting ARM and not the labile phase.

They localized the cell responsible for this effect using DDC-RNAi: it is the famous DPM neurons, known to have gap-junction with the GABAergic APL neuron (important for short term memory) and previously discribed as the neuron expressing amnesiac, a neuropeptide also involved in memory formation. They use three different driver line to block activity of or prevent serotonin production in these neurons, which both lead to a phenocopy of the serotonin block.

Finally, they used RNAi to show that the 5HT receptor A is necessary in alpha/beta lobes for the formation of ARM.

Lee, P., Lin, H., Chang, Y., Fu, T., Dubnau, J., Hirsh, J., Lee, T., & Chiang, A. (2011). Serotonin-mushroom body circuit modulating the formation of anesthesia-resistant memory in Drosophila Proceedings of the National Academy of Sciences, 108 (33), 13794-13799 DOI: 10.1073/pnas.1019483108

In this focus paper in Behavioral and Brain Sciences, Crespi and Badcock argue that the psychiatric disorders autism and schizophrenia are opposite ends of a psychological spectrum. They claim that the natural variations in several factors such as sense of self, gaze, agency, social cognition, local versus global processing and others creates a spectrum at the end of which lie the two disorders. The idea is that autism involves a general pattern of constrained overgrowth, whereas schizophrenia involves undergrowth.

The authors provide a wealth of evidence supporting their claim. At the end of the paper, as usual for this journal, appear a list of peer comments, both supportive and critical of the authors arguments.

Clearly, the point that needs to be discussed is: at what point is the authors’ claim an oversimplification which hurts more than it helps? Oversimplifications are often exactly the kind of conceptual advance that is required to do experiments that otherwise nobody would have dared to do. The two disorders share commonalities, as the authors point out, but they also differ in many important aspects. At what point do the differences invalidate any claim of overarching commonalities?

Crespi, B., & Badcock, C. (2008). Psychosis and autism as diametrical disorders of the social brain Behavioral and Brain Sciences, 31 (03) DOI: 10.1017/S0140525X08004214

Recently it has been shown that hippocampal protein degradation via the ubiquitin proteasome system (UPS) is a prerequisite for reconsolidation of associative fear memory. Nevertheless the results for the role of hippocampal UPS mediated protein degradation in the consolidation of memory remain contradictory. Now Jarome et al. present data, suggesting that consolidation of fear memory depends on NMDA Receptor-triggered protein degradation in the amygdala. Further they show that the time courses (molecular marker & behavioral pharmacology) of protein degradation and protein synthesis induced by acquisition of fear memory are similar, suggesting that the processes are coupled. This is supported by their findings that MOV10 (a protein which silences synaptic protein synthesis) is tagged for degradation induced by acquisition. Concerning reconsolidation they confirm existing results mentioned above and add evidence that these processes are NMDAR dependent. Interestingly they show that the same targets tagged for degradation induced by acquisition (MOV10 & Shank) are also tagged induced by retrieval. The results of this comprehensive study show that consolidation and reconsolidation of fear memory depend on protein degradation in the amygdala.

Jarome, T., Werner, C., Kwapis, J., & Helmstetter, F. (2011). Activity Dependent Protein Degradation Is Critical for the Formation and Stability of Fear Memory in the Amygdala PLoS ONE, 6 (9) DOI: 10.1371/journal.pone.0024349

The good news first: this blog is now linked to both Twitter and Researchblogging.org (link to come), I will send access information via our mailing list.

The bad news: we don’t have a calendar function here on the blog such as that of our Yahoo Group, nor a file folder for the papers, nor a mailing list. This means we still have to use the Yahoo Group (i.e., we need to use two logins, one for the group and one for the blog).

In order to log in to the blog, you need to first register with your ZEDAT account by clicking on ‘anmelden’ here on the lower right hand side.
Once you have signed in, you need to send me your ZEDAT username and then I can manually enter you for the JC blog.
Once I have registered you, you will get an email asking to confirm the registration.
Convoluted, I know! I will have a meeting with the people in the CeDiS (hopefully soon) to work on improvements.

A point to be resolved is the place where the JC will take place. As the Heekeren lab is located in the ‘Silberlaube’ it would make sense to have the journal club there, right before lunch. Right now for us it’s on Mondays at 11am, I wonder if that would work for the interested Heekeren lab members? Do you have a room at this day and time, or should we move to a different day?
Any interested members of the Heekeren lab are recommended to sign up at the Yahoo Group (or suggest a better way of organizing a schedule with reminders and distribute files 🙂

Once we have a time and place, we sign up for dates until everyone has a spot. PDFs can be deposited in the folder on the Yahoo Group.

Let’s see if we can get this experiment off the ground!

This is, to my knowledge, the first blog of a journal club. I maybe wrong, but we might be making history. Check this space for new developments.