Sleep and spine loss

Sleep and waking modulate spine turnover in the adolescent mouse cortex

Maret et al 2011 (Nature Neuroscience)

This is an intriguing little paper just appeared in Nature Neuroscience, looking at spine turnover in sensorimotor (a bit of a vague word this) cortex of adolescent (P23-44) YFP-expressing mice. They imaged the same section of dendrite through the skull of the same mouse on two separate occasions - quite a technical achievement - at the end of a period of wakefulness, and then again after the mouse's next sleeping period. In this situation they found a net loss of spines after the sleeping period, whereas if they reversed the order, imaging after sleep first, then after waking, they found a net gain of spines. These differences disappeared in adults.

So sleep seems to be an important period of synaptic pruning in adolescent mice. I wonder if this applies to humans as well. Teenagers seem to go through these periodic bouts of excessive sleeping. Perhaps they're not being lazy after all - they're engaging in important synaptic pruning, remodelling their brain circuits in preparation for adult life?

Seaside fun!

I'm writing this on the train from Devon back to London.

I've just spent a fortnight at the Marine Biological Association in Plymouth, attending the 28th annual Microelectrode Techniques Workshop. Two weeks of very intensive work - lectures, demonstrations, practical sessions in the lab, and an excellent lunch every day in the common room overlooking Plymouth Sound and the Americas Cup.

I have to say this course has been a real eye-opener for me - I first learned whole-cell patch clamping 3 years ago, but this is the first time I've really been taken through the physics and electronics of what the technique involves. It's one thing to know to press this button and twiddle that knob until the spiky bits go away, it's quite another to understand where that capacitative transient comes from, and how the compensation circuits work that can remove it from your signal. (This is no reflection on my teachers by the way - it reflects rather my own laziness and lack of curiosity.)

The very first day in the lab exemplified the approach - a solid lecture on electronics, followed by a happy day soldering up various op-amp circuits in various configurations: current to voltage converter, voltage follower, differentiator.

A huge amount of learning was packed into these two weeks. Lots of time in the lab, which was kitted out with many rigs of various kinds, gave me opportunities to try out several techniques, including iontophoretic cell injection with sharp electrodes, and single channel recording. The demonstrators were friendly, dedicated, and in many cases pretty distinguished. There was also a generous schedule of lectures, with highlights for me from Boris Barbour on amplifier electronics, and David Ogden on photolysis.

Another highlight was a fabulous morning spent on the MBA research vessel trawling for crabs and shrimps and miscellaneous denizens of the deep.

We finished about 7:30pm each day, and went in 6 days a week, which has left me tired but exhilarated - my head is buzzing with ideas for things I want to try out when I get back to the lab at UCL. I would recommend this course without hesitation to anyone who wants to be a better electrophysiologist. One clear lesson I have come away with is this - a good scientist does the homework to really understand the techniques they use. It's not good enough to just copy the protocol from a paper you read once, blindly following someone else's recipe.

Reading Week

This week is Reading Week, where the whole lab decamps down to my supervisor's lovely seaside home, and we spend the week reading neuroscience papers, and discussing them over huge meals and abundant bottles of wine. Alongside shorts, sunhat and various neuroscience texts, I also - with a vague sense of shame - packed GRR Martin's "Game of Thrones" to read in my spare time.

Which highlights a point of tension for me: should I even be reading for fun at all? Shouldn't all my reading hours be given over to papers, reviews and books about neuroscience? If I'm serious about this PhD shouldn't I be willing to sacrifice reading fiction for fun for a few years?

And it's not just a reading issue - what about my hobbies? When I was a wage-slave in the IT industry all those years, frustrated and bored at work much of the time, I lived for my free time, and hobbies were a big part of that. Paragliding, boardgaming, wargaming, mountain walking, cycling, drawing, even dancing for a couple of years. Now that I'm in the lab doing something interesting and exciting with my working day, it's not so urgently important for my happiness to be doing all these hobbies. Plus there's less time and energy left for them these days. But I am still pretty interested in some of them, especially walking and boardgaming and books. Should I still be making time for this stuff? My supervisor on my masters told me "science should be your hobby too". Even at this early stage, I can feel the totalizing pressure of science on my life, and I can see that for some of the really successful scientists it becomes an all-consuming passion. Should I let that happen to me? Or is there still room for a hobby or two in my life?

Terence Smith: imaging complex neuronal behavior in the enteric nervous system

I went to a fascinating seminar today at UCL given by Terence Smith of the University of Nevada. He was talking about his work on the enteric nervous system. Turnout was low - most of us are not that excited by anything without "hippocampus" in the title! But I was glad I went. Prof Smith described this fine mesh of ganglia and interconnecting fibres sitting between the smooth muscle layers of the gut. Did you know that the ENS has as many neurons as the spinal cord?! And what a fascinating array of neuron types - sensory, excitatory motor, inhibitory motor, pacemaker and many varieties of interneuron. All talking to each other by means of strange and unfamiliar (at least to me) neurotransmitters - 5HT, acetyl choline, ATP, NO and suchlike. Prof Smith and his lab have been using calcium sensitive dyes to track patterns of neuronal activation. The nice thing about the ENS is that it is a mammalian system that is also a tractable problem - at least compared with brain areas like the hippocampus. A relatively simple circuit producing a relatively small repertoire of "behaviours" - it should be possible to get to the bottom of this in the next few decades. Which is more than we could hope for with the hippocampus for example....

I'm getting a little worried about myself - I seem to be developing an interest in statistics.

I am currently reading two stats books with relish:

Edward Tufte "The Visual Display of Quantitative Information"

Peter Dalgaard "Introductory Statistics with R'

The Tufte especially I'm finding well nigh unputdownable. A beautiful argument embodied in a beautiful book.

It all started with an inspirational little stats course at UCL earlier this year, which began with the story of the Challenger shuttle disaster in 1986. At a meeting the night before the launch all the data needed to predict the failure of the O-rings was presented by the engineers to the managers, but in such an opaque way that the message failed to get across and the flight went ahead anyway.

So good stats can save lives, folks!

Later on the lecturer introduced some ideas from Tufte, and showed us some nice graphs done with R.

So now I'm a bit hooked. Playing with R in my spare time. And thinking about super-clear ways to display the data I hope to get (one day!) from my experiment. And today I was even asked about stats by some of my lab-mates. Not sure if I answered correctly, but need to be careful here - getting a reputation.

Is synaptic scaling important in Alzheimer’s Disease?

Ruppin & Reggia (1994) A Neural Model of Memory Impairment in Diffuse Cerebral Atrophy. British Journal of Psychiatry (1995) 166, 19-28.

I've been looking for papers about homeostatic synaptic plasticity in Alzheimer's or other neurodegenerative diseases. Not much at all. Which is odd. Intuitively you would think that homeostatic effects like synaptic scaling must be an important factor in the progression, if not the aetiology, of the disease.

But I did turn up this suggestive neural network paper from 1994.

The authors built an attractor neural network. I'm not 100% sure what that is, but my understanding is that it is a network of interconnected model neurons which tends to move to one of a finite number of stable states. With the appropriate learning rules the network can be trained to "remember" a number of input patterns.

They subjected their attractor network (which had several 100 model neurons) to a gradual loss of neurons or synapses. They found that if they built in a compensatory strengthening of surviving synapses (analogous to synaptic scaling) then memories degraded gracefully rather than suddenly. They also found that remote memories are spared in comparison with recent memories (with or without scaling), and that scaling increases the number of false positives. All these results are reminiscent of the clinical features of Alzheimer's Disease.

Which doesn't prove that synaptic scaling is important in Alzheimer's, of course. But it strongly suggests that it's at least worth taking a look.

Sacred vessels

The rule of St. Benedict says, apparently, that you should treat your wheelbarrow as you would the sacred vessels of the altar.

In other words: respect your tools!

So today I plan to give the venerable electrophysiology rig that I use every day, with its upright Olympus microscope and online confocal microscope, some tender loving care.

The bath is flooding again, which was giving me lots of problems when I was imaging yesterday, but once I have sorted that out I will also clean up the stage and the objectives and the condensor and just generally wipe away the accumulated salt and dust. Also tidy away the old pipettes and clamps and bits and pieces that people leave lying around on the gas table. Make the old girl feel loved.

It's the right thing to do. St. Benedict said.

How to write a neuroscience textbook

I've been leaving the trusty old iPod at home and instead carrying "Dendritic Spines" (Rafael Yuste, MIT Press, 2010) with me for on-tube entertainment. Also for propping up in front of my sandwiches during my lunch "break". It's a long time since I have come across such a readable science text. My previous big purchase, "Barrel Cortex" by Woolsey and Fox, had large tracts of impenetrable dullness. But Yuste's book, as well as being less than half the price, just whizzes along. Considering the fairly technical nature of its subject matter, it's superbly readable.

I think part of the secret is that Dr Yuste has a big idea, and his writing buzzes with the excitement of getting this across to you. After all, if spines are the physical substrate of memory and cognition, then it's pretty important to understand how they work.

Plus it has a great chapter on Ramon y Cajal, my neuroscience hero....

Viva post mortem

In December I submitted my first year dissertation, and last week I faced a viva on this dissertation. Opinions from colleagues and supervisors varied as to whether or not this was really an exam, and how much I should relax about it. Naturally (me being me) I did not relax about it. One feature I knew that this viva (exam or not) shared with real exams was that failure was a possibility, and that failure had consequences.

So this mattered.

I was lucky enough to be able to invite two scientists I like and get on with to examine me. But still - people can show a whole new side of their personality in this kind of role. And I hate the idea of making a fool of myself in front of people I respect. So I was tense for at least a week beforehand, and preparing diligently.

In the event, the whole thing was very amiable - enjoyable even. Defending my dissertation was a pretty minor part of it. Most of our time was spent chewing over the implications of different technical choices as I go forward with my project. Chemical LTP or glutamate uncaging? Organotypic or acute slices?

And I passed. And got some pretty nice feedback. So I can go on....

Journal club fun

McGuiness et al, "Presynaptic NMDARs in the Hippocampus Facilitate Transmitter Release at Theta Frequency", Neuron 68, 1109–1127, December 22, 2010

I go to two journal clubs, one nice the other nasty. Today was the nasty one, where we looked at the above article. It was the usual pattern - a paper that I thought was quite impressive, turned out to have feet of clay when subjected to the withering scrutiny of the assembled critical intelligences.

In this study, they used Ca-sensitive dyes to image Ca-transients in presynaptic boutons of the CA3-CA1 pathway, in response to action potentials. They claim that the distribution of Ca-transient amplitudes is bimodal. Various drug manipulations, including the use of norketamine to block (internally) only presynaptic NMDARs, demonstrate that the less frequent, large transients are dependent on presynaptic NMDARs. Trouble is, after fig 1 they don’t show us the data – just their opaque Bayesian analysis. And what is worse, they use the same control data in most of their figures.