BOYDEN, Ed, Tools for Brain Research, MIT, April 14, 2016
<p class="MsoNormal">BOYDEN, Ed, Tools for Brain Research, MIT, April 14, 2016
This is in the Stanford Neurosci Institute series on Thursdays in Clark Aud today.</p>
<p>full title: Tools for analyzing and repairing complex biological systems.</p>
<p>he is a prof at MIT; this is an <strong>SRO crowd</strong> in Clark Aud at Stfd.
(Ed is a superstar who just shared a $3M Breakthrough Prize with his former
Stanford associate, Karl Deisseroth.)</p>
<p>introduced by our Prof. Jenn Raymond (she tells many great stories;
she first interviewed Ed when he was 19 (and already had a masters!)
he did PhD with Prof. Raymond here .
Abstract of this talk:
To enable the understanding and repair of the brain, we are creating tools that enable molecular-resolution maps of large scale biological systems, as well as technologies for observing and controlling high-speed physiological dynamics in such systems.
First, we have developed a method for imaging large 3-D specimens with nanoscale precision, by <strong>embedding them in a dense swellable polymer</strong>, mechanically homogenizing the specimen, and exposing it to water – which causes it to <strong>expand isotropically severalfold.</strong>
This method, which we call <strong>expansion microscopy (ExM), </strong>enables scalable, inexpensive diffraction-limited microscopes to do <strong>large-volume nanoscopy.</strong>
Second, we have collaboratively developed strategies to<strong> image fast physiological processes in 3-D with millisecond precision, and used them to acquire neural activity maps </strong>throughout small organisms.
Third, we have collaboratively developed nanotechnological and robotic methods to <strong>record high-speed electrical events with single cell resolution in living mammalian brain</strong>.
Finally, we have developed a set of genetically-encoded reagents, known as<strong> optogenetic tools, that when expressed in specific neurons, enable their electrical activities to be precisely driven</strong> or silenced in response to millisecond timescale pulses of light. In this way we aim to enable the systematic mapping, dynamical observation, and control of complex biological systems like the brain.
<strong>Ed begins</strong> *</p>
<p>focus today is on tech dev for brain research.</p>
<p>SIZE scales (in studying the brain)... vary hugely...</p>
<p>from ion channels (nm) to one meter long axons (in corticospinal neurons)</p>
<p>and TIME scales...</p>
<p>from milliseconds to seconds and even to years!</p>
<p>so, how to cross all these huge scales of space and time?</p>
<p>eg.... h ow to map brain circuits with molec precision?</p>
<p><strong>can we magnify the objects of study themselves!!!</strong>
(this is stunning research and new to most of our SRO audience)
we use sodium polyacrylate..</p>
<p>it's a polymer that swells in water (found in baby diapers!)</p>
<p>see Chen Tillber Boyden in SCIENCE in 2015...
and this work has been commercialized by ESBat Expansion Tech Inc
(Ed is founder = ESB)</p>
<p>we embed brain in polyacrylate and then add water...</p>
<p><strong>want to expand the components</strong> (but brain fights back!)</p>
<p><strong>Expansion MIcroscopy = ExM</strong></p>
<p>want to copy the INFO represented by the structures to the fluorophors
(that bind to them using antibodies = Abs) (but then don't need the orig tissue)</p>
<p>add Ab + fluor + polymer (and that binds to the tissue) and
then just add water...</p>
<p>it expands... so, the structures spread apart (and then they are much easier to see...
can then be seen with light microscopy!)</p>
<p>he shows before and after with confocal microscopy...</p>
<p>so it induces some alteration in the structures (but only abt 1% distortion)</p>
<p>do before microscopy with</p>
<p>SR-SIM = super resol struc illumination microscopy and after (it is 4.5X size)
and look with confocal micro.</p>
<p>does it work on intact tissue? Yes.</p>
<p>he shows that <strong>ExM preserves global morphology of brain</strong> using YFP expressing neurons...</p>
<p>a bonus: we can use existing fixatives and fluorophors... big advantage..</p>
<p>stains for Homer and Basson...</p>
<p>shows dendritic spines... can see this with regular microscopy...</p>
<p>we can use inexpensive and scalable optical methods.</p>
<p>shows perisomatic synapses... and bizarre shape of chromatin in nucleus.</p>
<p>also can look at tumors eg breast CA in humans..</p>
<p>see Yongxin and Beck study...</p>
<p>and can even expand bacteria. (images)</p>
<p>can viz. even with modified webcams!!</p>
<p>our current focus... make all this easier; and can we expand even bigger?</p>
<p>anchoring proteins and then destroying them (by digestion using proteases)</p>
<p>but secodnary Abs bearing fluorophores survive the protease K treatment...</p>
<p>he shows application to Brainbow mice... with proExM...</p>
<p>beta barrels survive this process...</p>
<p>work of Paul Tillberg and Dawen Cai...</p>
<p>this process and its result is intrinsically multiscale in space
(Bob Desimone Lab looks at non-human primate brain)</p>
<p>and he has a collab with George. Church Lab</p>
<p>v cool process for doing <strong>in situ RNA imaging</strong> after in situ amplification a la Church.</p>
<p><strong>we can readout K channels, kinases, glutamate receptos, forkhead box</strong>, etc.</p>
<p>want to see the RNAs registered with the background micro histology...</p>
<p>collab with Xiaowei Zhuang at Harvard...using STORM imaging...
illustrations of that work... he shows</p>
<p>pre-expansion: images of microtubules in red vs
post-expansion in green (with confocal microscopy)
of microtubules...</p>
<p>we can preserve isotropy down to tens of nm.</p>
<p>also note: the expanded samples are almost completely transparent...
(assists microscopy)</p>
<p>our goal... <strong>nanoscale imaging across entire brain...
</strong>(so combine nm scale with multi-cm scale of entire brain)...</p>
<p> <strong>2nd half of the talk</strong></p>
<p>he shifts focus to living systems (above was in preserved specimens)</p>
<p>we'd like to be able to readout the dynamics of cells (neurons eg)</p>
<p>how can you record biologic dynamics with precise temporal and broad spatial scales?</p>
<p>this is a a new direciton for our group.</p>
<p>see eg Adamala Martin-alarcon and Boyden in 2016 in press.</p>
<p>he shows Pumilio homolog 1
it's made of a series of 36 amino acid repeats..
each 36 AA binding unit ...</p>
<p>we want to observe RNAs as they are expressed or translated (by ribosomes)
in living cells!!!
method: stick one half of GFP on to each half of the molecule of interest
and then when the two halves of the molecule bind , only then do they glow...
(cuz the halves of GFP are also brought together) so you can monitor when they bind!
(when the light turns on!)</p>
<p>this is done with PUMs...</p>
<p>he shows Pumby n-mers... these can be of arbitrary length...</p>
<p>so, you get more GFP created, if there is more tranlation of the protein by ribosomes...</p>
<p>can also use other fluorophors eg( we can split luciferase...)</p>
<p>and the glow is proportional to the amount of protein being made.</p>
<p>so, can monitor RNA translation in living cells...</p>
<p>this is early work in HeLa cells (not in neurons yet)</p>
<p>the amount of fluor is proportional to strength of binding of PUMS
(cuz the POM binding leads to more fluor eg GFP or luciferase)</p>
<p>open problem.... can we put in multiple fluorophors
(to make signal brighter (cuz, just one molec is hard to see)?</p>
<p><strong>3rd part of talk: control of neural activity</strong> **</p>
<p>12 years ago he started work with Karl D on ChR ......(when Ed was a PhD student here)</p>
<p>the initial work on optogenetics was done by Oesterhelt and Stockenious in 1971 !!! ...
that was on bacteriorhodopsins...</p>
<p>then work on halorhodopsins was in 1977</p>
<p>2010 work on archaerhodopsins...</p>
<p>see Boyden in 2011 Faculty of 1000 Bio Reports 3:11...</p>
<p>now, he gives an update on this...</p>
<p>Now we're mining genomic diversity in protists...</p>
<p>he will tell us two stories... (and one new one)</p>
<p>Amy Chuong did this work in our group.</p>
<p>he illuminates a cubic cm of his finger tip with his laser pointer...
note: <strong>RED light is much better for penetration of tissues</strong> than green or blue.</p>
<p>consider JAWS a novel halorhodopsin ...</p>
<p>we can use red light to go up to 3mm deep in the brain to silence neurons... (with jaws)</p>
<p>also see Klapoetke in 2014 Nat Methods... in algae...</p>
<p>one had a great red light drivable photo current...</p>
<p>"Chrimson" (name of an opsin) was the best that we found
(it has peak in the red part of spectrum)</p>
<p>we can drive neurons to fire spikes even with near IR light...</p>
<p>Chronos (is another opsin)... a v fast channelrhodopsin...</p>
<p>but how does it work? (it has much better light sensitivity)</p>
<p>and perhaps we can <strong>control 2 different sets of neurons with different colors</strong></p>
<p>put chrimson in one set of neurons and drive those using red light</p>
<p>and say Chronos in another set of neurons and drive those with blue...</p>
<p>together they have zero cross talk in control of spikes...</p>
<p>and can use them to trigger synapatic release events...</p>
<p>he shows a window of intensities of blue light
where you can drive chronos neurons but Chrimson neurons are still below threshold for firing.</p>
<p> <strong>so, how many different populations of neruons can you control?</strong> </p>
<p>let's say we project a hologram into brain and
sculpt the light in the brain... but need high temporal accuracy....</p>
<p>he shows we can get spikes even with only a 3 msec stim....
we can get 100 hz response... work by Valentina Emiliani...</p>
<p>everything is posted on our website...his lab does lots of outreach...(transfer of tech to other labs)</p>
<p>Bill N comment: he's seen analogous work in retina to selectively stim photoreceptors
Bill suggests poss of using similar strategy for doing selective stim of neurons...</p>
<p>audience Q: how do you get specific intensity of light (to get selective response (of eg chronos vs chrimson)</p>
<p>Q how abt screening for other polymers....
(I chat with Louise Giam (Sidhof Lab) and Paul Hoerbelt (Malenka Lab))
</p>