Microcircuits Ed Boyden Synthetic Neurobiology Laboratory - - PowerPoint PPT Presentation

Microcircuits

Ed Boyden

Synthetic Neurobiology Laboratory Massachusetts Institute of Technology Invited lecture at Stanford in CS379C on April 8, 2013

Today’s outline

STATISTICAL MICROCIRCUITS ‘PRECISE’ MICROCIRCUITS Goal: not to saturate with examples of microcircuits, but to pick a few canonical examples that highlight the kind of information that one gets.

Microcircuits

• Role of a ‘precise’ anatomical connection in a

physiological or behavioral function

– Usually, but not always, considered as a circuit that is spatially compact

• This is an assumption: precise wiring spans multiple

regions, but this is less studied

• Cell type A connected to Cell type B

– As you remember from lecture 1, cell types are defined by

• Shape
• Molecular composition
• Firing rate
• These arguments are largely statistical in nature

– For a computer: might conclude “transistors are connected to capacitors”

• Is this helpful?
• Specific connections, when altered, cause specific behavioral

defects

– Schizophrenia, epilepsy, etc. – It tells us something about the brain that when one kind of connection is altered, you can get a specific behavioral deficit – Evolution has let these (connection ßà disorder) links slide

Statistical Microcircuits

Drive a region, the region rings like a bell

(Alilain et al., 2008)

Function of a specific connection

• A model of obsessive compulsive disorder:

molecular deletion of Sapap3, scaffolding protein important for glutamate receptor presence, and synaptic strength (Welch et al., 2007)

• Leads to excessive grooming
• Striatal intervention reduces phenotype

Statistical Microcircuits: Looking Within

• Last time: cortical regions classified by
• cytoarchitectonics. This time, peer within:

(Silberberg et al., 2005)

Inhibitory neurons

• Basket cells are connected by gap junctions and

also GABA synapses

– Inhibitory neurons often reciprocally connected within class

(Galarreta and Hestrin, 1999) (Hestrin and Galarreta, 2005)

Statistical microcircuits can yield insights into statistical dynamics

• Low-order dynamics can emerge from statistical

connectivity

• Tonic drive to the interneuron circuit would be

predicted to lead to oscillations

Connexin 36 KO (Hormudzi et al., 2001)

Example: optical stimulation of inhibitory neurons

• Cardin et al 2009

Molecules make sense in the context of the specific cell type they’re in: schizophrenia

• Blocking NMDA reduces

gamma oscillations in a slice model of such

• scillations (Cunningham et al.,

2006)

• Less gamma oscillations

in schizophrenic patients

(Spencer et al., 2004)

Gamma oscillations involve inhibitory interneurons

(Cunningham Et al., 2003)

Molecules in the context of cells: Example: GABA receptor and NMDA receptor

• In patients: histological changes in subclasses of

interneurons (Lewis and Volk 2005)

– See reductions in GAT-1 staining, the GABA transporter – In, amongst other cells, PV+ neurons

• “chandelier cell”

A possible pipeline from molecule to macroscopic phenotype

• NMDA dysfunction à problems with

interneurons à less gamma oscillations

• Brain stimulation is done pretty ad hoc right now
• By moving beyond the molecule, we can start to

think about whether brain stimulation could correct deficits directly

Excitatory cells: how many do you need, to get a response? (Huber et al., 2008)

Schematic of the behavioural apparatus and reward contingencies. The mouse initiates a trial by sticking its snout into the central port. Photostimuli are applied during a stimulation period (300 ms) accompanied by a series of bright blue light flashes delivered to the behavioural arena (30 Hz, 300 ms) to mask possible scattered light from the portable light source. The mouse then decides to enter either the left or the right port for a water reward. If a photostimulus was present, the choice

• f the left port was rewarded with a drop of

water (hit, green star) whereas the choice of the right port lead to a short timeout (4 s, miss, red star). If the stimulus was absent, only the choice of the right port was rewarded with reward (correct reject, green circle) whereas the left port lead to a timeout (4 s, false alarm, red circle).

DAT-Cre + AAV-FLEX-ChR2-tdTomato

Kim et al. (2012) PLoS One 7(4):e33612

Finding circuits in the brain that can mediate reward

• Dopamine

neurons: implicated in reward and addiction, but largely through pharmacologic al and electrical means

• Is a brief

activation of them sufficient to drive reward?

Kim et al. (2012) PLoS One 7(4):e33612

light stimulation no light stimulation

Kim et al. (2012) PLoS One 7(4):e33612

Transgenic mice expressing original-N. pharaonis halorhodopsin in hypocretin neurons

Tsunematsu et al. (2011) Journal of Neuroscience 31(29): 10529-10539.

Light silences the neurons, resulting in slow-wave sleep

Tsunematsu et al. (2011) Journal of Neuroscience 31(29): 10529-10539.

Example: how anesthesia modulates functional connectivity of networks recruited by SI pyramidal cells

Opto-fMRI: a translational bridge between animal (causal) and human (behavior/disease) Idea generalizes: Opto-EEG, Opto-ECOG, etc.

Desai et al. (2011) Journal of Neurophysiology 105(3):1393-405. Kahn et al. (2011) Journal of Neuroscience 31(42):15086-15091.

Using cell-type specific control to understand the meaning of neurostimulation

• Electrical stimulation: has heterogeneous effects
• n neurons

– Example: some neurons can be completely silenced by electrical stimulation – Engaging interneurons? Direct silencing?

(Butovas and Schwarz, 2003)

Example: pyramidal cells can drive broad inhibition

• Usually you think of a synapse from a cell onto

another cell being weak

– But, can be strong enough to drive downstream neurons to spike – or to exclusively drive another entire cell class to function (Kapfer et al., 2007)

What mediates this inhibition?

• Somatostatin-positive interneurons in layers 2/3

and 5

Prefrontal Cortex

• Example: prefrontal cortex for working memory –

hypothesize that maintenance of firing over time is important – look for ‘loops’ (Wang et al., 2006)

– 1,233 double patch clamp recordings

PFC: 89% have two apical dendrites (‘complex’, cPC) 47% of pairs reciprocally connected VC: 87% have one apical dendrite (‘simple’, sPC) 18% of pairs reciprocally connected

Cerebellum

• Cerebellar microcircuit – used to try and justify

models of motor learning (Marr, Albus, 60’s)

(Boyden et al., 2004)

A simpler, take-to-the-bank cerebellar microcircuit conclusion

(Vos et al., 1999)

Problem: Canonical Microcircuit really means ‘Statistical’ Microcircuit

• These are descriptions of pairwise

correlations in connectivity

• Often used to generate unfalsifiable models; to

disprove them is too much work

• Current efforts are mostly spent on moving from

pairwise to n-wise descriptions of connectivity

– Currently, n is between 3 and 4 – A lot of this is taking new tools to perform

Caveat experimentor

• Most of these experiments are done in slice

– Across multiple ages, species, brain regions, etc. – More synapses in the hippocampus ~2 hours of slicing, than a perfusion-fixed animal (Kirov et al., 1999)

Extend correlations in connectivity

• Patch-clamp four rat visual cortex layer 5 pyramidal cells

at a time; analyze the connectivity (Song et al., 2005)

Sampling bias: look at nearby cells. Hard to scale beyond 3; very laborious. Only statistical, still.

Extend correlations in space

• Find follower cells by stimulating with patch pipette,

calcium imaging other cells (Kosloski et al., 2001)

Yellow = big interneurons, green = pyramidal, red = fusiform interneurons Small n. Only see super-strong followers. Lose some geometry in the slice.

Photostimulation-assisted mapping

• Connected excitatory neurons, receive common

excitatory inputs (Yoshimura, et al., 2005)

Excitatory neurons get common inhibition regardless of connectivity to one another

Photostimulation-assisted mapping

(Yoshimura et al., 2006)

Channelrhodopsin-2-assisted mapping

(Petreanu et al., 2007)

Left, laminar positions of recorded cells ipsilateral or contralateral to the electroporated hemisphere. Solid symbols indicate cells showing EPSCs in response to photostimulation of ChR2-positive axons and open symbols indicate cells that did not show EPSCs. Triangles, pyramidal cells; diamonds, stellate cells; blue circles, fast-spiking interneurons. Right, the fraction of stellate and pyramidal cells receiving input from L2/3 cells.

Going beyond ‘statistical connectivity’

• Still an “A projects to B” kind of problem
• Moving to higher-order statistics is hard
• True understanding requires reconstruction of a

circuit all the way from input to output

– The worm, C. elegans – Connectomics

Driving a single cell causes flip of the whole- cortex EEG (Li et al, 2009)

350 to 600 ms per step, 30 to 40 steps with 4- to 4.5-s intervals,

• r 8 ms per pulse, 5 to 15 pulses per burst, 40 to 80 bursts with

2- to 2.5-s intervals

• C. elegans

“There are 302 neurons in the nervous system of

• C. elegans; this number is invariant between
• animals. Each neuron has a unique

combination of properties, such as morphology, connectivity and position, so that every neuron may be given a unique label. Groups of neurons that differ from each other only in position have been assigned to classes. There are 118 classes that have been made using these criteria, the class sizes ranging from 1 to 13. Thus C. elegans has a rich variety of neuron types in spite of having only a small total complement of neurons. This is in marked contrast to structures such as the mammalian cerebellum, which contains more than 1010 neurons (Braitenberg & Atwood 1958) and yet has only five classes of component neuron (Eccles et al. 1967).” White..Brenner, Phil. Trans. Royal Soc. London. Series B, Biol Scien. Vol.314, Issue 1165 (Nov 12, 1986), 1-340

(Sulston and Horvitz, 1977)

http://www.wormatlas.org/MoW_built0.92/nervous_system.html

• C. elegans
• FIGURE 16. Transverse sections through the anterior regions of the ventral ganglion (a) and process identification

(b). The ventral cord enlarges at its anterior end and contains some of the neuropile of the cells of the ventral

• ganglion. Several of the processes in this region have characteristic shapes and synapses; these differences

facilitate their identification. Processes from SMB flatten out and sandwich each other; AIY makes characteristic dyadic and triadic synapses onto the RIB, AIZ and RIA; FLP makes prominent synapses onto AVA; RIGL/R run near the centre of the neuropile making extensive gap junctions to each other; the processes of AIA enlarge and make several synapses to AIB and RIF. The neuropile in this region is bounded on the ventral and lateral sides by thin sheet-like processes from CEPsh. Processes from the excretory glands run along the dorsal surface of the neuropile and eventually end in the ventral regions of the nerve ring. The four cell bodies above the neuropile are always in this configuration and can be readily identified. The rest of the cell bodies in the ventral ganglion are more variable in their relative positions. *WA editors'note: PVPL and PVPR are inadvertently mislabeled in this figure. It should be PVPL on the right side of VC and PVPR on the left.

http://www.wormatlas.org/MoW_built0.92/nervous_system.html

• C. elegans
• “There are, perhaps, two fundamental questions in the field of neurobiology: how

neurons organize themselves during development into specifically interconnected networks, and how such a network functions. A knowledge of the detailed structure of a nematode's nervous system does not in itself provide any answers to these questions, but it does at least provide a framework within which it is possible to pose rather more specific questions.”

Molecular correlations across cells of a network

• Different neurons, in the same network,

coordinate their channel expression across cells (Schulz et al., 2006)

– The pyloric rhythm of the STG of the crab C. borealis.

Automating Electron Microscopy with Serial Section Backscatter Electron Microscopy

$750,000 for a full setup. (Jurrus et al., 2006) (Denk and Horstmann, 2004)

Linking neural activity to underlying connectomes

• (Bock et al., 2011; Briggman et al., 2011)

Super-resolution Microscopy

(Hell, 2007)

Super-resolution Microscopy

Willig et al., 2006 – STED of synaptotagmin Rust et al., 2006 – STORM of photoswitchable Cy5 Betzig et al., 2006 – PALM of photoactivatable GFP

Other papers relevant for this week

READINGS Lewis DA, Hashimoto T, Volk DW. Cortical inhibitory neurons and schizophrenia. Nat Rev Neurosci. 2005 Apr;6(4):312-24. http://www.nature.com/nrn/journal/v6/n4/abs/nrn1648.html Esclapez M, Hirsch JC, Ben-Ari Y, Bernard C. Newly formed excitatory pathways provide a substrate for hyperexcitability in experimental temporal lobe epilepsy. J Comp Neurol. 1999 Jun 14;408(4):449-60. http://www3.interscience.wiley.com/cgi-bin/abstract/61005935/ ABSTRACT?CRETRY=1&SRETRY=0 Tseng KY, Kasanetz F, Kargieman L, Riquelme LA, Murer MG. Cortical slow oscillatory activity is reflected in the membrane potential and spike trains of striatal neurons in rats with chronic nigrostriatal lesions. J Neurosci. 2001 Aug 15;21(16):6430-9. http://www.jneurosci.org/cgi/content/full/21/16/6430 SUPPLEMENTARY READINGS Jay R. Gibson, Michael Beierlein and Barry W. Connors Two networks of electrically coupled inhibitory neurons in neocortex Nature 402, 75-79 (4 November 1999) | doi:10.1038/47035 http://www.nature.com/nature/journal/v402/n6757/abs/402075a0.html Pascual O, Casper KB, Kubera C, Zhang J, Revilla-Sanchez R, Sul JY, Takano H, Moss SJ, McCarthy K, Haydon PG. Astrocytic purinergic signaling coordinates synaptic networks.

• Science. 2005 Oct 7;310(5745):113-6.

http://www.sciencemag.org/cgi/content/full/310/5745/113 Petreanu L, Huber D, Sobczyk A, Svoboda K. Channelrhodopsin-2-assisted circuit mapping of long-range callosal projections. Nat Neurosci. 2007 May;10(5):663-8. Epub 2007 Apr 15. http://www.nature.com/neuro/journal/v10/n5/abs/nn1891.html Denk W, Horstmann H. Serial block-face scanning electron microscopy to reconstruct three- dimensional tissue nanostructure. PLoS Biol. 2004 Nov;2(11):e329. http://biology.plosjournals.org/perlserv/?request=get- document&doi=10.1371/journal.pbio.0020329&ct=1 Sul JY, Orosz G, Givens RS, Haydon PG. Astrocytic Connectivity in the Hippocampus. Neuron Glia Biol. 2004 Feb;1(1):3-11. http://www.pubmedcentral.nih.gov/articlerender.fcgi? tool=pubmed&pubmedid=16575432