Who invented optogenetics?

Wayne State University. Ever heard of it? Probably not. How about Zhuo-Hua Pan? No? No bell ringing? Let’s try a different approach: ever heard of Stanford University? Why, yes, it’s one of the most prestigious and famous universities in the world. And now the last question: do you know who Karl Deisseroth is? If you’re not a neuroscientist, probably not. But if you are, then you would know him as the father of optogenetics.

Optogenetics is the newest tool in the biology kit that allows you to control the way a cell behaves by shining a light on it (that’s the opto part). Prior to that, the cell in question must be made to express a protein that is sensitive to light (i.e. rhodopsin) either by injecting a virus or breeding genetically modified animals that express that protein (that’s the genetics part).

If you’re watching the Nobel Prizes for Medicine, then you would also be familiar with Deisseroth’s name as he may be awarded the Nobel soon for inventing optogenetics. Only that, strictly speaking, he did not. Or, to be fair and precise at the same time, he did, but he was not the first one. Dr. Pan from Wayne State University was. And he got scooped.98.png

The story is at length imparted to us by Anna Vlasits in STAT and republished in Scientific American. In short, Dr. Pan, an obscure name in an obscure university from an ill-famed city (Detroit), does research for years in an unglamorous field of retina and blindness. He figured, quite reasonably, that restoring the proteins which sense light in the human eye (i.e. photoreceptor proteins) could restore vision in the congenitally blind. The problem is that human photoreceptor proteins are very complicated and efforts to introduce them into retinas of blind people have proven unsuccessful. But, in 2003, a paper was published showing how an algae protein that senses light, called channelrhodopsin (ChR), can be expressed into mammalian cells without loss of function.

So, in 2004, Pan got a colleague from Salus University (if Wayne State University is a medium-sized research university, then Salus is a really tiny, tiny little place) to engineer a ChR into a virus which Pan then injected in rodent retinal neurons, in vivo. After 3-4 weeks he obtained the expression of the protein and the expression was stable for at least 1 year, showing that the virus works nicely. Then his group did a bunch of electrophysiological recordings (whole cell patch-clamp and voltage clamp) to see if shining light on those neurons makes them fire. It did. Then, they wanted to see if ChR is for sure responsible for this firing and not some other proteins so they increased the intensity of the blue light that the ChR is known to sense and observed that the cell responded with increased firing. Now that they saw the ChR works in normal rodents, next they expressed the ChR by virally infecting mice who were congenitally blind and repeated their experiments. The electrophysiological experiments showed that it worked. But you see with your brain, not with your retina, so the researchers looked to see if these cells that express ChR project from the retina to the brain and they found their axons in lateral geniculate and superior colliculus, two major brain areas important for vision. Then, they recorded from these areas and the brain responded when blue light, but not yellow or other colors, was shone on the retina. The brain of congenitally blind mice without ChR does not respond regardless of the type of light shone on their retinas. But does that mean the mouse was able to see? That remains to be seen (har har) in future experiments. But the Pan group did demonstrate – without question or doubt – that they can control neurons by light.

All in all, a groundbreaking paper. So the Pan group was not off the mark when they submitted it to Nature on November 25, 2004. As Anna Vlasits reports in the Exclusive, Nature told Pan to submit to a more specialized journal, like Nature Neuroscience, which then rejected it. Pan submitted then to the Journal of Neuroscience, which also rejected it. He submitted it then to Neuron on November 29, 2005, which finally accepted it. Got published on April 6, 2006. Deisseroth’s paper was submitted to Nature Neuroscience on May 12, 2005, accepted on July, and published on August 14, 2005… His group infected rat hippocampal neurons cultured in a Petri dish with a virus carrying the ChR and then they did some electrophysiological recordings on those neurons while shining lights of different wavelengths on them, showing that these cells can be controlled by light.

There’s more on the saga with patent filings and a conference where Pan showed the ChR data in May 2005 and so on, you can read all about it in Scientific American. The magazine is just hinting to what I will say outright, loud and clear: Pan didn’t get published because of his and his institution’s lack of fame. Deisseroth did because of the opposite. That’s all. This is not about squabbles about whose work is more elegant, who presented his work as a scientific discovery or a technical report or whose title is more catchy, whose language is more boisterous or native English-speaker or luck or anything like that. It is about bias and, why not?, let’s call a spade a spade, discrimination. Nature and Journal of Neuroscience are not caught doing this for the first time. Not by a long shot. The problem is that they are still doing it, that is: discriminating against scientific work presented to them based on the name of the authors and their institutions. Personally, so I don’t get comments along the lines of the fox and the grapes, I have worked at both high profile and low profile institutions. And I have seen the difference not in the work, but in the reception.

Personally, so I don’t get comments along the lines of the fox and the grapes, I have worked at both high profile and low profile institutions. And I have seen the difference not in the work, but in the reception.

That’s my piece for today.

Source:  STAT, Scientific American.


1) Bi A, Cui J, Ma YP, Olshevskaya E, Pu M, Dizhoor AM, & Pan ZH (6 April 2006). Ectopic expression of a microbial-type rhodopsin restores visual responses in mice with photoreceptor degeneration. Neuron, 50(1): 23-33. PMID: 16600853. PMCID: PMC1459045. DOI: 10.1016/j.neuron.2006.02.026. ARTICLE | FREE FULLTEXT PDF

2) Boyden ES, Zhang F, Bamberg E, Nagel G, & Deisseroth K. (Sep 2005, Epub 2005 Aug 14). Millisecond-timescale, genetically targeted optical control of neural activity. Nature Neuroscience, 8(9):1263-1268. PMID: 16116447. DOI: 10.1038/nn1525. doi:10.1038/nn1525. ARTICLE 

By Neuronicus, 11 September 2016






THE FIRSTS: The Mirror Neurons (1988)

There are some neurons in the human brain that fire both when the person is doing some behavior and when watching that behavior performed by someone else. These cells are called mirror neurons and were first discovered in 1988 (see NOTE) by a group of researchers form the University of Parma, Italy, led by Giacomo Rizzolatti.

The discovery was done by accident. The researchers were investigating the activity of neurons in the rostral part of the inferior premotor cortex (riPM) of macaque monkeys with electrophysiological recordings. They placed a box in front of the monkey which had various objects in it. When the monkey pressed a switch, the content of the box was illuminated, then a door would open and the monkey reached for an object. Under each object was hidden a small piece of food. Several neurons were discharging when the animal was grasping the object. But the researchers noticed that some of these neurons ALSO fired when the monkey was motionless and watching the researcher grasping the objects!

The authors then did more motions to see when exactly the two neurons were firing, whether it’s related to the food or threatening gestures and so on. And then they recorded from some 182 more neurons while the monkey or the experimenter were performing hand actions with different objects. Importantly, they also did an electromyogram (EMG) and saw that when the neurons that were firing when the monkey was observing actions, the muscles did not move at all.

They found that some neurons responded to both when doing and seeing the actions, whereas some other neurons responded only when doing or only when seeing the actions. The neurons that are active when observing are called mirror neurons now. In 1996 they were identified also in humans with the help of positron emission tomography (PET).

88mirror - Copy
In yellow, the frontal region; in red, the parietal region. Credits: Brain diagram by Korbinian Brodmann under PD license; Tracing by Neuronicus under PD license; Area identification and color coding after Rizzolatti & Fabbri-Destro (2010) © Springer-Verlag 2009.

It is tragicomical that the authors first submitted their findings to the most prestigious scientific journal, Nature, believing that their discovery is worth it, and rightfully so. But, Nature rejected their paper because of its “lack of general interest” (Rizzolatti & Fabbri-Destro, 2010)! Luckily for us, the editor of Experimental Brain Research, Otto Creutzfeld, did not share the Nature‘s opinion.

Thousands of experiments followed the tremendous discovery of mirror neurons, even trying to manipulate their activity. Many researchers believe that the activity of the mirror neurons is fundamental for understanding the intentions of others, the development of theory of mind, empathy, the process of socialization, language development and even human self-awareness.

NOTE: Whenever possible, I try to report both the date of the discovery and the date of publication. Sometimes, the two dates can differ quite a bit. In this case, the discovery was done in 1988 and the publishing in 1992.


  1. di Pellegrino G, Fadiga L, Fogassi L, Gallese V, & Rizzolatti G (October 1992). Understanding motor events: a neurophysiological study. Experimental Brain Research, 91(1):176-180. DOI: 10.1007/BF00230027. ARTICLE  | Research Gate FULLTEXT PDF
  2. Rizzolatti G & Fabbri-Destro M (Epub 18 Sept 2009; January 2010). Mirror neurons: From discovery to autism. Experimental Brain Research, 200(3): 223-237. DOI: 10.1007/s00221-009-2002-3. ARTICLE  | Research Gate FULLTEXT PDF 

By Neuronicus, 15 July 2016

Intracranial recordings in human orbitofrontal cortex

81 ofc - CopyHow is reward processed in the brain has been of great interest to neuroscience because of the relevance of pleasure (or lack of it) to a plethora of disorders, from addiction to depression. Among the cortical areas (that is the surface of the brain), the most involved structure in reward processing is the orbitofrontal cortex (OFC). Most of the knowledge about the human OFC comes from patients with lesions or from imaging studies. Now, for the first time, we have insights about how and when the OFC processes reward from a group of scientists that studied it up close and personal, by recording directly from those neurons in the living, awake, behaving human.

Li et al. (2016) gained access to six patients who had implanted electrodes to monitor their brain activity before they went into surgery for epilepsy. All patients’ epilepsy foci were elsewhere in the brain, so the authors figured the overall function of OFC is relatively intact.

While recording directly form the OFC the patients performed a probabilistic monetary reward task: on a screen, 5 visually different slot machine appeared and each machine had a different probability of winning 20 Euros (0% chances, 25%, 50%, 75% and 100%), fact that has not been told to the patients. The patients were asked to press a button if a particular slot machine is more likely to give money. Then they would use the slot machine and the outcome (win 20 or 0 Euros) would appear on the screen. The patients figured out quickly which slot machine is which, meaning they ‘guessed’ correctly the probability of being rewarded or not after only 1 to 4 trails (generally, learning is defined in behavioral studies as > 80% correct responses). The researchers also timed the patients during every part of the task.

Not surprisingly, the subjects spent more time deciding whether or not the 50% chance of winning slot machine was a winner or not than in all other 4 possibilities. In other words, the more riskier the choice, the slower the time reaction to make that choice.

The design of the task allowed the researchers to observe three 3 phases which were linked with 3 different signals in the OFC:

1) the expected value phase where the subjects saw the slot machine and made their judgement. The corresponding signal showed an increase in the neurons’ firing about 400 ms after the slot machine appeared on the screen in moth medial and lateral OFC.

2) the risk or uncertainty phase, when subjects where waiting for the slot machine to stop its spinners and show whether they won or not (1000-1500 ms). They called the risk phase because both medial and lateral OFC had the higher responses when there was presented the riskiest probability, i.e. 50% chance. Unexpectedly, the OFC did not distinguish between the winning and the non-wining outcomes at this phase.

3) the experienced value or outcome phase when the subjects found out whether they won or not. Only the lateral OFC responded during this phase, that is immediately upon finding if the action was rewarded or not.

For the professional interested in precise anatomy, the article provides a nicely detailed diagram with the locations of the electrodes in Fig. 6.

The paper is also covered for the neuroscientists’ interest (that is, is full of scientific jargon) by Kringelbach in the same Journal, a prominent neuroscientist mostly known for his work in affective neuroscience and OFC. One of the reasons I also covered this paper is that both its full text and Kringelbach’s commentary are behind a paywall, so I am giving you a preview of the paper in case you don’t have access to it.

Reference: Li Y, Vanni-Mercier G, Isnard J, Mauguière F & Dreher J-C (1 Apr 2016, Epub 25 Jan 2016). The neural dynamics of reward value and risk coding in the human orbitofrontal cortex. Brain, 139(4):1295-1309. DOI: http://dx.doi.org/10.1093/brain/awv409. Article

By Neuronicus, 25 March 2016

Autism cure by gene therapy

shank3 - Copy

Nothing short of an autism cure is promised by this hot new research paper.

Among many thousands of proteins that a neuron needs to make in order to function properly there is one called SHANK3 made from the gene shank3. (Note the customary writing: by consensus, a gene’s name is written using small caps and italicized, whereas the protein’s name that results from that gene expression is written with caps).

This protein is important for the correct assembly of synapses and previous work has shown that if you delete its gene in mice they show autistic-like behavior. Similarly, some people with autism, but by far not all, have a deletion on Chromosome 22, where the protein’s gene is located.

The straightforward approach would be to restore the protein production into the adult autistic mouse and see what happens. Well, one problem with that is keeping the concentration of the protein at the optimum level, because if the mouse makes too much of it, then the mouse develops ADHD and bipolar.

So the researchers developed a really neat genetic model in which they managed to turn on and off the shank3 gene at will by giving the mouse a drug called tamoxifen (don’t take this drug for autism! Beside the fact that is not going to work because you’re not a genetically engineered mouse with a Cre-dependent genetic switch on your shank3, it is also very toxic and used only in some form of cancers when is believed that the benefits outweigh the horrible side effects).

In young adult mice, the turning on of the gene resulted in normalization of synapses in the striatum, a brain region heavily involved in autistic behaviors. The synapses were comparable to normal synapses in some aspects (from the looks, i.e. postsynaptic density scaffolding, to the works, i.e. electrophysiological properties) and even more so in others (more dendritic spines than normal, meaning more synapses, presumably). This molecular repair has been mirrored by some behavioral rescue: although these mice still had more anxiety and more coordination problems than the control mice, their social aversion and repetitive behaviors disappeared. And the really really cool part of all this is that this reversal of autistic behaviors was done in ADULT mice.

Now, when the researchers turned the gene on in 20 days old mice (which is, roughly, the equivalent of the entering the toddling stage in humans), all four behaviors were rescued: social aversion, repetitive, coordination, and anxiety. Which tells us two things: first, the younger you intervene, the more improvements you get and, second and equally important, in adult, while some circuits seem to be irreversibly developed in a certain way, some other neural pathways are still plastic enough as to be amenable to change.

Awesome, awesome, awesome. Even if only a very small portion of people with autism have this genetic problem (about 1%), even if autism spectrum disorders encompass such a variety of behavioral abnormalities, this research may spark hope for a whole range of targeted gene therapies.

Reference: Mei Y, Monteiro P, Zhou Y, Kim JA, Gao X, Fu Z, Feng G. (Epub 17 Feb 2016). Adult restoration of Shank3 expression rescues selective autistic-like phenotypes. Nature. doi: 10.1038/nature16971. Article | MIT press release

by Neuronicus, 19 February 2016


A third way of neuronal communication

electrical fields - CopyEvery neuroscience or biology textbook will tell you that neurons communicate with one another in two ways: via chemical synapses (one neuron releases some substances that change the membrane potential of the receiving neuron) or via electrical synapses (neurons physically share some membrane proteins that allows the electrical impulse to go from one neuron to another). (Nota bene: for the taxonomic nitpickers, I decided that the other non-conventional forms of communication, like diffusion, fall in one or the other of the two categories above).

Now Qiu et al. (2015) have some evidence that there may be another way of neuronal chatting. Long ago (in 1924), Hans Berger observed that neurons have rhythmic and spontaneous electrical activity. The rhythmic activity of an entire population of neurons is called a wave and can be detected with rudimentary tools such as EEG. Qiu et al. reasoned that the speed of such a traveling wave is too slow to be explained by conventional neuronal communication (of which we know quite a lot).

So the researchers ran some computer simulations to test if diffuse electrical fields can explain the speed of a wave, instead of the conventional fast communications. In other words, can local, weak, slow endogenous fields sweep the brain by recruiting nearby neurons? The computer simulations said it is possible, with a very slow speed of 10 cm/s.

Then they recorded the electrical activity of mouse hippocampus isolated in a Petri dish. When the researchers blocked the electrical field there was a decrease in the speed of the wave, meaning the field is required for the wave speed observed.

To be fair, this is not a new idea; even in the early eighties these fields were suggested, because the wave persisted even when the other two ways of communicating have been blocked. And in the early noughts it was shown that a neural network is much more sensitive to electrical field manipulation than individual neurons. What makes this paper interesting is that the authors show that the endogenous electrical fields are not too weak to underlie the wave, as previously thought. The work has implications for the study of epilepsy, sleep, and memory.

Reference: Qiu C, Shivacharan RS, Zhang M, & Durand DM (2 December 2015). Can Neural Activity Propagate by Endogenous Electrical Field?, Journal of Neuroscience, 35(48): 15800-15811; doi: 10.1523/JNEUROSCI.1045-15.2015 Article | Full Text via Research Gate | ScienceAlert cover

By Neuronicus, 11 February 2016

CCL11 found in aged but not young blood inhibits adult neurogenesis

vil - Copy
Portion of Fig. 1 from Villeda et al. (2011, doi: 10.1038/nature10357) describing the parabiosis procedure. Basically, under complete anesthesia, the peritoneal membranes and the skins of the two mice were sutured together. The young mice were 3–4 months (yellow) and old mice were 18–20 months old (grey).

My last post was about parabiosis and its sparse revival as a technique in physiology experiments. Parabiosis is the surgical procedure that joins two living animals allowing them to share their circulatory systems. Here is an interesting paper that used the method to tackle blood’s contribution to neurogenesis.

Adult neurogenesis, that is the birth of new neurons in the adult brain, declines with age. This neurogenesis has been observed in some, but not all brain regions, called neurogenic niches.

Because these niches occur in blood-rich areas of the brain, Villeda et al. (2011) wondered if, in addition with the traditional factors required for neurogenesis like enrichment or running, blood factors may also have something to do with neurogenesis. The authors made a young and an old mouse to share their blood via parabiosis (see pic.).

Five weeks after the parabiosis procedure, the young mouse had decreased neurogenesis and the old mouse had increased neurogenesis compared to age-matched controls. To make sure their results are due to something in the blood, they injected plasma from an old mouse into a young mouse and that also resulted in reduced neurogenesis. Moreover, the reduced neurogenesis was correlated with impaired learning as shown by electrophysiological recordings from the hippocampus and from behavioral fear conditioning.

So what in the blood does it? The authors looked at 66 proteins found in the blood (I don’t know the blood make-up, so I can’t tell if 66 is a lot or not ) and noticed that 6 of these had increased levels in the blood of ageing mice whether linked by parabiosis or not. Out of these six, the authors focus on CCL11 (unclear to me why that one, my bet is that they tried the others too but didn’t have enough data). CCLL11 is a small signaling protein involved in allergies. So the authors injected it into young mice and Lo and Behold! there was decreased neurogenesis in their hippocampus. Maybe the vampires were onto something, whadda ya know? Just kidding… don’t go around sucking young people’s blood!

This paper covers a lot of work and, correspondingly, has no less than 23 authors and almost 20 Mb of supplemental documents! The story it tells is very interesting and as complete as it gets, covering many aspects of the problems investigated and many techniques to address those problems. Good read.

Reference: Villeda SA, Luo J, Mosher KI, Zou B, Britschgi M, Bieri G, Stan TM, Fainberg N, Ding Z, Eggel A, Lucin KM, Czirr E, Park JS, Couillard-Després S, Aigner L, Li G, Peskind ER, Kaye JA, Quinn JF, Galasko DR, Xie XS, Rando TA, Wyss-Coray T. (31 Aug 2011). The ageing systemic milieu negatively regulates neurogenesis and cognitive function. Nature. 477(7362):90-94. doi: 10.1038/nature10357. Article | FREE Fulltext PDF

By Neuronicus, 6 January 2016

You were not my first choice either!

Sexually receptive mice females prefer a Lego brick over a male if their oxytocin neurons are silenced.

Over the past five years or so, dopamine stepped down from the role of the “love molecule” in favor of oxytocin, a hormone previously known mostly for its crucial role in pregnancy, labor, delivery, lactation, and breastfeeding. Since some interesting discoveries in monogamous vs. polygamous voles (a type of rodent) pointing to oxytocin as essential for bonding, many studies implicated the chemical in all sorts of behaviors, from autistic to trusting, from generosity to wound healing.

Nakajima, Görlich, & Heintz (2015) add to that body of knowledge by finding that only a small group of cells in the medial prefrontal cortex express oxytocin receptors: a subpopulation of somatostatin cortical interneurons. Moreover, these neurons are gender dimorphic, meaning they differ from male to female: the female ones have twice as many action potentials upon application of oxytocin as compared to male’s.

And here is the more interesting part:
– Females in the sexually receptive phase of their estrus whose oxytocin neurons were silenced preferred to interact with a Lego brick over a male mouse (which, as you might have guessed, in not what they typically choose).
– Females that were not in their sexually receptive phase when their oxytocin neurons were silenced still preferred to interact with a mouse (male or female) over the Lego brick.
– Silencing of other neurons had no effect on their choice.
– Silencing had no effect on the males.

Hm… there are such things out there as oxytocin inter-nasal sprays… How soon do you think until the homeopaths, naturopaths, and other charlatans market oxytocin as a potent aphrodisiac? And it will take some deaths until the slow machine of beaurocracy turns its wheels and tightens the regulations on the accessibility to the hormone. Until then… as the cartoons say, don’t try this at home! Go buy some flowers or something for your intended one… it would work better, trust me on this.

Reference: Nakajima M, Görlich A, & Heintz N (9 October 2014). Oxytocin modulates female sociosexual behavior through a specific class of prefrontal cortical interneurons. Cell. 159(2): 295–305. doi:10.1016/j.cell.2014.09.020. Article | FREE FULLTEXT PDF

By Neuronicus, 23 October 2015

Giving up? Your parvalbumin neurons may have something to do with it

Cartoon from http://i393.photobucket.com/albums/pp20/saisi24/dontgivedup.jpg, licensing unknown
Cartoon from Photobucket, licensing unknown.

One of the most ecologically-valid rodent models of depression is the learned helplessness paradigm. You get a rat or a mouse and you confine it in a cage with an electrified grid. Then you apply mild foot shocks at random intervals and of random duration for an hour (which is one session). The mouse initially tries to escape, but there is no escape; the whole floor is electrified. After a couple of sessions, the mouse doesn’t try to escape anymore; it gives up. Even when you put the mouse in a cage with an open door, so it can flee to no-pain freedom, it doesn’t attempt to do so. The interpretation is that the mouse has learned that it cannot control the environment, no matter what he does, he’s helpless, so why bother? Hence the name of the behavioral paradigm: learned helplessness.

All antidepressants on the market have been tested at one point or another against this paradigm; if the drug got the mouse to try to escape more, then the drug passed the test.

Just like in the higher vertebrate realm, there are a few animals who keep trying to escape longer than the others, before they too finally give up; we call these resilient.

Perova, Delevich, & Li (2015) looked at a type of neuron that may have something to do with the capacity of some of the mice to be resilient; the parvalbumin interneurons (PAI) from the medial prefrontal cortex (mPFC). These neurons produce GABA, the major inhibitory neurotransmitter in the brain, and modulates the activity of the nearby neurons. Thanks to the ability to genetically engineer mice to have a certain kind of cell fluoresce, the researchers were able to identify and subsequently record from and manipulate the function of the PAIs. These PAIs’ response to stimulation was weaker in helpless animals compared to resilient or controls. Also, inactivation of the PAI via a designer virus promotes helplessness.

Reference: Perova Z, Delevich K, & Li B (18 Feb 2015). Depression of Excitatory Synapses onto Parvalbumin Interneurons in the Medial Prefrontal Cortex in Susceptibility to Stress. The Journal of Neuroscience, 35(7):3201–3206. doi: 10.1523/JNEUROSCI.2670-14.2015. Article | FREE FULLTEXT PDF

By Neuronicus, 21 October 2015

Making new neurons from glia. Fully functional, too!

NeuroD1 transforms glial cells into neurons. Summary of the first portion of the Guo et al. (2014) paper.
Fig. 1. NeuroD1 transforms glial cells into neurons. Summary of the first portion of the Guo et al. (2014) paper.

Far more numerous than the neurons, the glial cells have many roles in the brain, one of which is protecting an injury site from being infected. In doing so, they fill up the injury space, but they also prohibit other neurons to grow there.

Guo et al. (2015) managed to turn these glial cells into neurons. Functioning neurons, that is, fully integrated within the rest of the brain network! They did it in a mouse model of stab injury and a mouse model of Alzeihmer’s in vivo. Because a mouse is not a man, they also metamorphosized human astrocytes into functioning glutamatergic neurons in a Petri dish, that is in vitro.

It is an elegant paper that crossed all the Ts and dotted all the Is. They went to a lot of double checking in different ways (see Fig. 1) to make sure their fantastic claim is for real (this kind of double, triple, quadruple checking is what gets a paper into the Big Name journals, like Cell). Needles to say, the findings show a tremendous therapeutic potential for people with central nervous system injuries, like paralyses, strokes, Alzheimer’s, Parkinson’s, Huntington, tumor resections, and many many more. Certainly worth a read!

Reference: Guo Z, Zhang L, Wu Z, Chen Y, Wang F, & Chen G (6 Feb 2014, Epub 19 Dec 2013). In vivo direct reprogramming of reactive glial cells into functional neurons after brain injury and in an Alzheimer’s disease model. Cell Stem Cell, 14(2):188-202. doi: 10.1016/j.stem.2013.12.001. Article | FREE FULLTEXT PDF | Cell cover

By Neuronicus, 18 October 2015

I can watch you learning

Human Stereotaxic System. Photo credit: The Mind Project
Human Stereotaxic System. Photo credit: The Mind Project

Recording directly form the healthy living human brain has always been a coveted goal of many neuroscientists, thus bypassing the limitations of non-invasive techniques or animal work. But, understandably, nobody would seek or grant approval for inserting an electrode in the healthy living human brain, on moral and ethical grounds. The next best thing is to insert an electrode into the not so healthy living human brain.

Ison, Quiroga, & Fried (2015) got lucky and gained access to 14 patients with intractable epilepsy that had electrodes implanted in their brain to find where the seizure focus is (for possible surgical resection later on). Using these electrodes, they recorded the activity of single neurons within the medial temporal lobe (MTL, a brain area paramount for learning) while the patients performed some simple association tasks. First, they presented images of places, people, and animals to the patients to see “which (if any) of the recorded neurons responded to a picture” (p. 220). When they got a neuron responding to something, they rushed out, did some data and image processing, and after an hour they started the experiment. Which was showing the patient the picture to which the neuron responded to (e.g. Stimulus 1 = patient’s daughter) overimposed on a background that the neuron did not respond to (e.g. Stimulus 2 = the Eiffel tower). After one single trial (although there was some variability), the patients learned the associations (i.e. Stimulus 3 = daughter in front of Eiffel tower) and this learning was mirrored by how the neuron responded. Namely, the neuron increased its activity by 200% to 400% (counted in spikes per second) when shown the previously un-responded to image alone (i.e. Stimulus 2).

Excerpt from Fig. 5 from Ison, Quiroga, & Fried (2015).
Excerpt from Fig. 5 from Ison, Quiroga, & Fried (2015). “Average normalized neural activity (black squares) and behavioral responses (green circles) to the non-preferred stimulus as a function of trial number. Data were aligned to the learning time (relative trial number 0)”, i.e. when they showed the composite image between Stimulus 1 and Stimulus 2. “Note that the neural activity follows the sudden increase in behavioral learning”.

The authors recorded from over 600 neurons from various MTL regions, out of which 51 responded to a Stimulus 1. From these, only half learned, that is, they increased their activity when Stimulus 2 was shown. For the picky specialist, the cells were both Type 1 and Type 2 neurons, located 6 in the hippocampus, 4 in the entorhinal cortex, 11 in the parahippocampal cortex, and 1 in the amygdala. And the authors controlled for familiarity, attentional demands, and other extraneous variables (with some very fancy and hard to follow stats, I might add).

The paper settles an old psychology dispute. Do we learn an association gradually or at once? In other words, do we learn gradually that A and B occur together, or do we learn that the first time we are shown A and B together and the next trials serve just to refine and consolidate the new knowledge? Ison, Quiroga, & Fried (2015) data show that learning happens at once, in an all-or-none fashion.

Reference: Ison, M. J., Quian Quiroga, R., & Fried, I. (1 July 2015). Rapid Encoding of New Memories by Individual Neurons in the Human Brain. Neuron, 87(1): 220-30. doi: 10.1016/j.neuron.2015.06.016. Article | FREE PDF

By Neuronicus, 4 October 2015