Tbr1 Function in Cell Fate and Synaptogenesis

May 05, 2023

Information

Aghajanian Lecture: Tbr1 Function in Cell Fate and Synaptogenesis

John Rubenstein, MD, PhD, Professor of Psychiatry, University of California San Francisco

ID9900

To CiteDCA Citation Guide

00:03No, no, we'll talk. Got it, got it.
00:11And the the very generous introductions,
00:17sorry for the 18 issues.
00:20Look at this, Alexa.
00:22Nicole might be able to do that.
00:23You try, you know,
00:25that don't gloat this.
00:27You hear a lot of disasters out there.
00:30Golden Gate Park is still fine
00:35and this is a view of Golden
00:37Gate Park for the hill were at
00:40UCSF Mount Parnassus campuses.
00:44So I'll be telling you about a
00:46transcription factor called TV R1 which
00:49is really important and early portable
00:52development in self based specification.
00:55But as you'll see it has several roles
00:57as time moves on in development.
01:00Roles that I think are perhaps,
01:02maybe even more remain to cognitive
01:06disorders and then psychiatric disorders,
01:11in particular to autism.
01:12But I think it could be and could be
01:16thinking about other neuropsychiatric
01:19illnesses as well.
01:21Well, give a little historical
01:23perspective to this.
01:24To those of you who've known me this long,
01:27it's about 36 years ago.
01:30Began a screen for genes that are.
01:32We were hoping to find genes that
01:35are expressed much more during
01:37development of the forebrain
01:38than in the adult forebrain,
01:40with a simple idea that if they're
01:42expressed really highly during development,
01:44maybe they're important for development.
01:47And we use a technique oldfashioned
01:49technique called subtractive vibration
01:51with directional CD and A libraries we
01:54compared in the mouse and embryonic day 15.
01:57Forebrain 2 an adult forebrain,
02:04and using that technique we found a
02:06gene which we called healing Cell
02:08Phone Embryonic Subtraction #1,
02:11or Test one, which unfortunately had
02:14to have its name change in DL X2.
02:16I'm not it's not so bad about
02:18losing the test one for that reason,
02:20but you'll see in a second
02:21why it was sad to lose that.
02:28Test 1 encodes A homeo
02:31domain transcription factor.
02:33They're probably the most common
02:34kinds of transcription factors
02:36that are important in self fate
02:37regulation during development.
02:39And this is a picture of the
02:42embryonic mouse brain and the
02:44white stuff in the embryonic mouse
02:46brain shows where the RNA for test
02:48one is very strongly expressed
02:52here in the. Ganglia, primordia,
02:54this big lump of cells called
02:57the ganglionic eminences,
02:58which is below the cortex,
02:59which is not labeled,
03:01and it's also very strongly
03:03expressed in the Pantera,
03:04diencephalan and hypothalamus. Of
03:09course, when we did this first experiment,
03:10we had no idea what we were looking
03:12at because we had in medical school,
03:14you don't have any training in what
03:16the embryonic brain looks like.
03:17At least we didn't have poshco
03:20at standards we didn't Poshco.
03:21I might have known what this is, but.
03:23No, nobody who I working with knew
03:26what any of these things were.
03:30And then I visited jail in 1990
03:33when I met for the first time.
03:36At least let's say 10 of you who are
03:38in this room and I've been friends
03:40with you for that long and I gave my
03:43first talk on Test 1 / D L X2 and maybe
03:46one of you or two remember that talk.
03:49But. It was a huge experience for
03:51me to to come and meet you all.
03:54I almost took a job here and it
03:58was sad not to be able to do that.
04:00It was gratifying that I have
04:03life not lifelong,
04:04but half my lifelong friends that are
04:07who I met that day and have continued to
04:10see over the years and in many venues.
04:17We we went through our everyonic
04:19subtraction looking for other
04:20interesting genes and found many others.
04:22The the second most interesting
04:24in my opinion was test 56.
04:26So that's the 56 gene that we
04:29looked at and we named it a TB R1.
04:34Where T encephalon or A2 on cephalon
04:38express brain gene one or T box
04:40brain gene one has many reasons
04:43for its name and you'll see what
04:45its real reason is in a second.
04:49So unlike the DLX gene which was
04:52expressed in the basal ganglia,
04:54tea bear one not expressed in the basal
04:57ganglia it's it surrounds the basal
05:00ganglia expressed in the cerebral
05:02cortex and then a little domain
05:05in the that was called eminential
05:07palamide and then part of the
05:10hypothallus and and as it turned out.
05:16The DLX gene is expressed in
05:19progenitors and in Gabourgic
05:21neurons and in Gabourgic neurons.
05:24So DLX expression is pretty much
05:26the same as it as a gene like
05:28flutamic acid D carboxylase or
05:31the vesicular GABA transporter.
05:33They're in the same cells.
05:34In fact, DLX regulates these
05:36fundamental Gabourgic properties
05:38wherever you are in the forebrain,
05:41but the rest of the brain,
05:43it's not expressed,
05:44so it's a forebrain specific.
05:46Transcription factor that's
05:48involved with gabiergic cell fate,
05:52differentiation and function,
05:55and TV R1, by contrast,
05:57is only in glutamateurgic neurons.
06:00And so we began to get the idea
06:02that you can have transcriptional
06:04pathways that separate out these
06:06two fundamental cell types,
06:08excitatory cells and inhibitory cells.
06:10And by differentially controlling
06:12the activity of the DLX or TV R1,
06:15you can change the balance of
06:17of the function of excitatory
06:20or inhibitory neurons.
06:28And these two genes were became the
06:30foundation of much of the work that
06:32I've done the last 30 years at UCSL.
06:36So the real reason we named
06:38the genes what they were.
06:40Is because my daughter's name is Tess.
06:44That's Test 1. My my son's name is Thomas,
06:47that's Thomas Braden Rubenstein.
06:50That's why they have their names.
06:51That's why I was upset that
06:53the test turned into DL X2.
06:56She answers.
06:56She answers the both now.
07:02So I'm going to start by telling
07:04you I'm only going to be telling
07:05you about TB R1 in this talk.
07:07I'm going to tell you about TBR one's.
07:10Function from the very beginning
07:13of making the cerebral cortex
07:16into into an adultery,
07:18even through something called
07:19the grandpa mouse experiment.
07:22So kind of a life lifelong function of
07:25T ver one in controlling many parts
07:27of how you make cortical neurons.
07:30They're giving their identity
07:32and connections. I'm
07:38going to start by showing you something.
07:40Pictures of where T bear ones expressed
07:42because although it's kind of boring
07:44their anatomy, it sets the stage
07:46for thinking about what it does. So
07:52here's the In teaching hybridization done
07:54by the Allenbrae Institute of T bear one,
07:56the brown black cells are T bear one
08:00expressing cells, and you can see
08:02it's only in the cerebral cortex.
08:04And in this part of the hypothalamus,
08:07it's essentially nowhere else.
08:10In the central nervous system,
08:11except for some deep cerebellar nuclei.
08:15So it's highly highly specific for the
08:18for the forebrain, as I mentioned,
08:21it's only in citatory neurons in the cortex,
08:28and it's asking you to sign into Dropbox.
08:34And within the developing cortex
08:36it's not in the progenitors,
08:38so it's not in dividing cells.
08:40It turns on once the cells become post
08:43mitotic and are migrating and we call
08:46the intermediate zone and then it those
08:49cells coalesce and form the developing
08:51cortical plate and it's expressing
08:54the earliest born cortical neurons.
08:56I'll ask posture what those are.
08:57Now earliest born critical neurons are the
09:01Cahabretzia cells, the subplate and layer 6.
09:05So what you're looking at at
09:07this stage is probably 15 or so.
09:09Primarily layer 6 and subplate and cahabats.
09:12Your cells all coalesced in
09:14this early cortical plate.
09:17You look at the specialty bear one.
09:20In the neonatal mouse cortex,
09:23expression is concentrated in layer 6.
09:27If you look carefully in the rosel parts
09:30of layer five you can see scattered cells.
09:33So at this stage it's primarily
09:35layer 6 and layer 5.
09:39I don't think it's ever in layer 4.
09:40I'm not sure about.
09:42Then I can correct me.
09:43But then it turns on in layer 2-3,
09:45later on
09:49again, but only in excitatory neurons.
09:53And then in an adult Peter,
09:54one expression is maintained
09:56again most strongly in layer 6,
09:59in about half the cells in layer 5:00,
10:01and then in many cells in layers 2 and three.
10:04And I'll tell you mainly about
10:05what TB I one is doing in layer
10:076 and a little bit about what TB
10:09I one's doing in layer 5,
10:10and I don't know what TB I
10:11one's doing in layer 2-3 yet.
10:17So our one of our first papers
10:19on TB I-1 function shown here.
10:22The title says everything about it.
10:24It regulates differentiation
10:25of the preplate in layer 6.
10:27The preplate is basically
10:29the subplate and the car.
10:31That's just cells and there.
10:35It's the several things.
10:39Within Chen, we showed that TB R1
10:45controls the identity of layer 6.
10:48So ordinarily this transcription factor
10:51C tip 2 is in layer 5 and not layer 6,
10:54But in the TB R1 constituent
10:56of loss of function unit,
10:58this transcription factor C tip
11:012 is now expressed in layer 6.
11:03So TB R1 has a major role in depressing
11:07layer 5 identity in layer 6 cells,
11:11and if you start messing around
11:13with layer 6 in this way,
11:14a lot of bad things happen.
11:17I'll just tell you one of them.
11:19Ordinarily layer 6 is the main
11:21output for axons that grow
11:23and innervate the thalamus,
11:24and that's what's shown in this picture here.
11:26And then the TB R1 mutant.
11:28The axons are made,
11:30but they stop in the basal ganglia.
11:32Around the boundary with the hypothalamus
11:35and they never innovate the thalamus.
11:37So Tegre was really important in
11:41the program for the connection
11:43of the cortex to the thalamus.
11:48A very brief summary of what TR does
11:51does early at E 10.5 is important
11:54for generations of rexial cells.
11:56These cells are fundamental for making the
11:59laminar organization of the frebal cortex.
12:02That E 12.5, it's really important
12:04we think for the fate of layer 6,
12:07and we begin to think that the fate switch
12:10for layer 5 May start as early as E 12.5.
12:13And it's important for the connections
12:16of the cortex to the thalamus which
12:19is takes place in this interval
12:21between 12.5 and let's say 16.5.
12:26So this is that's what happens
12:27with you have a T01.
12:29Constitutive loss of function unit i.e.
12:31There's no T row one protein at all.
12:33All these bad things happens
12:35to to the cortex.
12:38So the bottom line is T row one.
12:41In T row one, layer 6 neurons
12:44transformed into layer 5 type similar.
12:48Not exactly, but pretty close to being
12:50a layer 5 type cortical neuron. And
12:56this is where I left it in the mid.
12:59Around 2005, 2010 or so and then I
13:05worked from Yale from Stefan Sanders,
13:08Nenod and and Matt State showed that a
13:12mutation of TB R1 was highly implicated
13:15in causing some forms of autism.
13:18And so when Matt came to UCSF he and
13:21then talked about this and he told me
13:23about the result ahead of time and let me
13:26get involved back into the TB R1 story.
13:28And that helped fund me getting
13:30back into TB R1 at work.
13:35So with that impetus I decided to look
13:40to see if I could understand how TB R1
13:43mutations might increase the risk for autism.
13:47And part of that was to begin to
13:49look at mice that are heterozygote
13:50for the loss of function tier one,
13:52because the humans who have autism or
13:56heterozygotes have loss of function alleles.
13:59And then the you know the Hail Mary was
14:03maybe find some possible treatment so
14:05that but that was that was a pipe dream
14:12and this work was done primarily by
14:14a post doc named Sivash Darbondy.
14:16And Sivash's first paper on this
14:18subject is this this one down here.
14:28OK, so for this experiment we
14:29decided to not use a constituent
14:32loss of function allele.
14:33We wanted to make a conditional mutation
14:36because we wanted to begin to dissect
14:38what T bear was was doing at different
14:40times and different cell types.
14:43So for that purpose we made what's
14:46no call the flopsed allele,
14:48which just means that we're we now
14:51genetically can manipulate when
14:53we delete T bear one.
14:55And this was done with the
14:57collaboration Matt State and Ben
14:59Chen and Ben's at Santa Cruz.
15:04So having built a floss the wheel for
15:07TBO and then we wanted to find the
15:10appropriate 3 recombinations so that we
15:13could delete it where and when we wanted.
15:16And we chose two layer specific
15:193 recombinations that turned
15:21on around every like day 18.5.
15:23Which is about 8 days after TB R1 turns on.
15:27So these mice will be developing with
15:30normal TB R1 until just before birth.
15:33So all the things that I've been
15:35telling you about that go wrong with
15:37TB R1 in the null mutant should not
15:40or may not happen in this mutant,
15:42but we will. We'll find out.
15:44But he goes.
15:47So one tree that we use is called.
15:49Doesn't really matter,
15:50it's called NTS R1 Creek.
15:52And it's really good at deleting tibro
15:55one in cortical layer 6 and again
15:57it turns around on around E 18.5.
16:02The other three we use is called
16:05R BP4 Creed and it deletes tibro
16:08one in layer 5 and nicely.
16:11It also includes the prefrontal cortex.
16:15The layer 6 creed is not active
16:17in the in the prefrontal cortex
16:19and we couldn't interrogate.
16:20Cortex function in the layer 6,
16:23but we could do it in the layer 5.
16:25That'll come into play later in the talk.
16:31So let me first tell you about the
16:34conditional deletion of TB R1 in layer 6,
16:37and I'll probably just be calling
16:39these TBR one layer 6 mutants.
16:42So we started off doing an RNARNA
16:46sequencing and comparing the RNA levels.
16:49And neonal cortex of wild type versus
16:53the conditional layer 6 mutant.
16:56And this pretty picture so-called
16:59volcano plot shows genes that are
17:01up regulated in red and genes
17:03that are down regulated in blue.
17:05So let's first look at the blue ones.
17:07Hebrew one is down regulated.
17:09Thank God because we deleted it.
17:11It's RNA was normal.
17:12That would look bad for our experiment.
17:15OK, so we're off to a good start.
17:18And one of the genes that's
17:20most downregulated is T LE4,
17:21which then I can tell you is
17:24super important for layer 6.
17:25So that's that was really good.
17:28And then a gene that's upregulated.
17:30And I can also tell you this Fez F2.
17:33Fez F2 is a key regulator of layer 5
17:36identity and it's expressed in layer 5.
17:39But this is telling us already
17:41that something's wrong with layer
17:436 and it's missing normal levels
17:45of T LE4 and has too much.
17:47Layer 5 expression.
17:48In it.
17:54We did a bunch of in C twos using
17:56probes based upon the genes that
17:58were found in the RN A/C analysis,
17:59and here's just two of them.
18:01Both of them are markers of layer 6,
18:05so Fox B2 and TT LE4.
18:10They're strongly expressed in layer 6,
18:12and very little in the case of
18:14Fox B2 and superficial layers.
18:16Killy 4 has this back when I don't
18:18know if that's kill expression or not,
18:20but the strong expression is in
18:23layer six in the in the heterozygote
18:27there's not much of an effect.
18:29You might think there's a
18:31could be a slight decrease,
18:32particularly in this one,
18:34but not a strong heterozygote effect.
18:36But in the homozygote 5B2
18:40expression is greatly reduced.
18:42Until E4 expression is greatly reduced,
18:44although not eliminated in the well,
18:46the subplate and also a
18:49superficial layer 6 okay.
18:51So this is interesting because TV R1
18:53has been there for eight days now,
18:56we took it away for about 3 or 4 days and
19:00you the layer 6 identity whittles away.
19:03So TV R1 is not only required for initiation
19:06of state specification of layer six,
19:09you need to keep TV R1 there.
19:12New Natalie in order to
19:14maintain layer 6 identity.
19:16We were surprised by this,
19:17but that's what that's
19:18that's the way it goes.
19:24Now let's look at layer 5 markers.
19:28There's this B CL11B which is same
19:31as C tip 2 which I showed earlier.
19:33Beautiful layer 5 expression lower
19:36in layer 6 and Fez F2 almost almost
19:40specific for layer five at this age.
19:43Almost nothing in layer 6.
19:45Now look at the heterozygote.
19:46This is really surprising.
19:47Let's just go straight to the Feds up two.
19:51Here's the layer 5 expression.
19:52Now in the heterozygote there is Feds
19:55up two at reasonable levels in layer 6.
19:59So even as a heterozygote,
20:01your layer 5 is in trouble.
20:02I'm sorry, your layer 6 is in trouble.
20:04If you're a TBR 01,
20:05no new and one can think about
20:07the meaning of that for a human
20:10who's heterozygote for TB 01.
20:11Having a mixed identity of layer 6
20:17and then this is the hobo
20:18Zygo and you get this.
20:20It looks like almost equivalent in
20:21case of peasant 2 layer 5 and layer
20:256 expression and similar for P CL11B.
20:31So
20:36I told you what early keyword
20:38one function does over here on
20:40the far on your far left when you
20:42delete keyword one around 18.5.
20:47Layer 6 takes on properties of layer five.
20:51We don't know how long,
20:53how far out that goes,
20:54but it'll be just don't.
20:55But we know that during this period
20:59that they changes that molecular
21:03possible fake changes prominent.
21:10Okay. Now we wanted to look what's
21:12what's the ramification of these changes
21:14in transcription factor expression
21:16on the cellular properties of the
21:19of the layers mutant layer 6 cells.
21:22So we've next looked at their dendrites.
21:28Normally layer 6 cells have
21:30this fascinating property.
21:31I don't know if you can see it very well.
21:33They're aprical dendrites wrote
21:35to here in in the layer 5 and
21:37they don't go into layer 6 much.
21:39So there's. A very strong
21:41regulation of dendrite properties.
21:45In layer six they would have that
21:48their little forest of dendrites
21:50only go up halfway up the cortex.
21:55On the other hand, in the TB R1
21:58mutant this looks straight down.
22:00Here many of their axons grow
22:02all the way up to layer one,
22:05this low pass and even in the heterozygote.
22:09It's a it's a partially penetrate phenotype.
22:12Some of these dendrites grow up there.
22:15Ramification of that that now means
22:16that layer 6 cells are going to
22:18get inputs from places that they
22:20never get inputs from before.
22:22So that's going to change the wiring
22:23diagram of the cortex and we'll
22:26confuse the circuit functions.
22:30Obviously we can use the circuit
22:31functions of the cortex.
22:36So with these conditional mutants,
22:38we can see that heroin has a persistent
22:42effect on repressing layer 5 identity,
22:45which includes the layer 6 cells
22:50send their dendrites to layer one,
22:52which is a property of layer 5 cells.
22:56So we have a molecular and cellular
22:59transformation of layer 6 towards layer 5
23:02identity in these conditional mutants and.
23:05Partial phenotypes in the heterozygous
23:13OK with Alex Nord we've looked at the
23:18locations in the on the chromosomes where
23:21the TB R1 transcription factor binds.
23:23We've identified which promoters and which
23:26candidate enhancers have TB R1 binding
23:31and we able to.
23:34Identify the canonical in vivo
23:36T row and binding motif motifs.
23:39There's there's a lot of variety
23:42of this and there's a lot more
23:44to do with to understand what
23:45those different movies mean
23:50based upon where keeper ones binding
23:53on the chromosomes and the RNA changes
23:57that happened at Tiro on mutants.
24:00And through these genomic binding sites,
24:02we can identify the candidate
24:04enhancers for the genes that
24:06whose expression has changed.
24:09We can develop enhancer transcription
24:11assays to test whether these enhancers
24:14are in fact regulated by TBRO One.
24:17Finding all of this data,
24:18we've begun to define transcriptional
24:21network regulated by TBRO One.
24:24And here's a here's a cartoon showing
24:26some of the things we've found.
24:28The T row one functions both as an activator.
24:31That's what these green arrows means,
24:33as well as a repressor.
24:35That's what the Red Arrows means.
24:37We think that most of these
24:42this pathway shown here are
24:44direct T rowing functions,
24:46not it could be also indirect,
24:48but you think a lot of this is
24:51direct and the genes that have
24:54the double asterisks on them.
24:57PC11A, OX B2, GRIN 2B,
24:59WIN 7B, Box P1 are also autism
25:06risk alleles risk genes,
25:09so perhaps TB R1 is critical
25:12in regulating many of the
25:15D chains that cause autism.
25:23So the summary of this first part of the
25:27talk is that TB R1 is really important.
25:30In cell identity, for excitatory
25:33neurons of the cerebral cortex,
25:35it's required to initiate and
25:37to maintain layer 6 identity,
25:39which then leads to the taking on of
25:42layer 5 morphological properties,
25:48the and we think that the layer and
25:50and obviously molecular properties
25:52and many of these tea bear one
25:54targets are autism risk genes.
26:01Okay. So now we went into
26:03new territory for us.
26:04We're very good at doing in C twos and
26:06learning about the molecular changes.
26:08Now we went to look at the synapses
26:10in these mutant layer 6 cells
26:14and synaptogenesis is most,
26:16you know in the mouse takes
26:19place postnatally being around
26:21postnatal layer 5 and accelerating.
26:23During the first week and second week,
26:30what Sivas did was look at synapses
26:34primarily in layer 5 because
26:37in the wild type or layer six
26:38don't go up to any other layers.
26:40So we chose layer 5:00 during the
26:42wild type to the conditional unit and
26:45then we did immunofluorescence assays
26:47for pre and post synaptic markers,
26:50excitatory and inhibitory synapses.
26:52And then measure the densities of the
26:55synaptic process of these synaptic
26:58elements as a function of length of the
27:02Axon and or dendrite that we were measuring.
27:08And we saw a very easy relate very
27:10simple relationship that the density
27:13of synaptic structures called
27:15butans very wild type heterozygote
27:19homozygote and we call it linear
27:21but it decreases as a function of.
27:24How much cheaper?
27:24Or when you have and this was true for
27:27both excitatory and inhibitory synapses,
27:29both were greatly reduced and heterozygote
27:31had the intermediate phenotype.
27:38Then with Victa Sohal and his
27:41graduate student Sarah Robinson.
27:43The electrophysiology of these mutants,
27:48and I won't show you a lot of squiggles,
27:50I'll just tell you wild type.
27:53And this was a normal squiggle and the
27:56homozygote is very quiet and less activity.
27:59The heterozygote was somewhere in between.
28:04So being a TV on mutant is very
28:07bad for your synaptic formation,
28:10and since you have less synapses,
28:11there's less activity.
28:12That's not too surprised.
28:16So now we add to all these other problems.
28:19They have reduced excitatory and inhibitory
28:22synaptic densities. And activities.
28:27Now we're beginning to get a
28:28little bit more psychiatric.
28:29You can see these not only is the
28:31identity of layer 6 messed up,
28:33so the connectivity's bad,
28:35that's not good.
28:36But also the amount of input
28:38being processed through the
28:39mutant cells is greatly reduced.
28:45So post Natal ETR ones required for
28:47normal number of excitator inhibitory
28:49synapsis onto layer 6 pyramidal neurons.
28:55And then we wanted to know why is,
28:57why are the number of synapses reduced?
28:59And we think we have a grasp of that answer.
29:04So we concentrated on one of the key
29:08components at least of excitatory synapses,
29:10which are the spines.
29:12Is the structure on the dendrites of the
29:15excitatory dendrites center of the neurons.
29:18We use this fantastic conflict of microscopy.
29:21System that has a computer hooked
29:23up to it and what it does is it
29:25gives you a picture like this where
29:27the the dendrites are these white
29:29beautiful guys and the spines are
29:32colored blue by the computer.
29:33Like I can understand this,
29:36we can count blue blobs.
29:39That's what a control looks like.
29:41And the tibra on mutant
29:47many fewer blue blobs.
29:49Now the blue blobs are. Mature spines.
29:52The immature spines are these thin,
29:56thin, filamentous things.
29:57So the mutant has fewer mature spines,
30:00and I don't know if it has more,
30:02but I think it does
30:03filamentous immature spines.
30:04So there's something wrong in the T barrel.
30:071 dendrite to promote the maturation of its
30:12immature spines to become mature spines,
30:16and you need to have mature spines
30:18for excitatory synapses of form.
30:20I cannot say why the inhibitory synapses
30:22are are having problems, but they
30:29they. This is a qualification at 2 pages,
30:33postnatal day five and postnatal 2021 of the
30:37of the measuring these blue mature spines.
30:41There's a wild type and here's a homozygote.
30:44There might be fewer mature spines.
30:47And then the later stages,
30:48we looked at wild type heterozygote
30:51homozygote and we again see this
30:54quantitative effect not only in the
30:56homozygote but in the heterozygote as well.
31:03Then we wanted to see if we can
31:05understand what might be a molecular
31:07mechanism that underlies failure
31:08and maturation of the spines.
31:11We went back to our RNAC
31:15and looked at some of the.
31:17Genes that are up or down regulated
31:21and I just point out one of them
31:22because this is the one that turned
31:23out to be the most useful which is
31:25a protein called Win 7B to secreted
31:28which early in development is super
31:31important in regional specification
31:33and selfate specification.
31:34But Patricia Salinas several many years
31:36ago showed how the wind proteins are
31:39also important in synapse formation
31:41and in Axon growth and and targeting.
31:46Then there are a bunch of other
31:48blue dots that are interesting as
31:49well which I won't tell you about.
31:53So when 7B is normally most strongly
31:57expressive layer 6 and then
32:00either when heterozygote may be
32:04reduced and the homozygote greatly
32:07reduced and then for some reason
32:08top of the express in layer 2-3.
32:11I don't understand that.
32:12Let's just pay attention to the reduction
32:15of T of Win 70 in in the in layer 6:00.
32:21So we wanted to ask whether or
32:23not we could restore synapses
32:26by destroying Win 70 expression.
32:28So we took a new Natal cortex,
32:31wild type or mutant and dissected it off
32:36and and dissociating and gruiting culture.
32:40And then let those cultures mature for
32:42about two weeks and then measure the
32:45number of synapses in those culture
32:47on to the layer 6 cells which would
32:48be red because of the tea tomato,
32:50which I didn't tell you about that
32:52you can see which cells are mutant
32:54that way we harvest days, day zero
32:58at day one we transpect with the DNA
33:03expression vector for win 7B and then.
33:0914 days later we count synapse numbers
33:13by these immunosum rest and sassay and
33:15we did this with about 5 different genes
33:18that were down regulating the T year only.
33:21And in this case only one
33:23of them showed big effect.
33:25I'll show you one that did not
33:27have an effect and one that did.
33:28So this is the control experiment with no
33:32transfection a wild type and homozygote.
33:35This is just showing what we already
33:37knew that there's decreased.
33:38Synapses we transpected with CAD
33:42here and eight which Josh Sainz
33:44had shown was down regulated in
33:46the retina of T VO1 mutants.
33:48We thought maybe that might be important
33:50it it didn't help and then when we use
33:54Win 7B it showed nearly complete rescue.
33:58That was our in vitro synapse assay.
34:01We followed up with an in vivo assay
34:04where we used a retrovirus vector.
34:10Let's factor that only rest of
34:12win 70 if the cells express free,
34:14so we use the NTS. R1 mouse has the
34:17pre and layer 6 and infect those and
34:27so we infect the babies are born P0,
34:31infected the next day at P1 and
34:33then at P28 we sacrifice the
34:36animal and stain them. And this is
34:43this is a similar but
34:48not a move, but sequential.
34:49It's not allowing me to
34:51touch to make it sequential.
34:56Anyway, what you would have seen here,
34:57but I could have made that work,
34:59is that if we could restore the
35:01synapses win 7B of the TB R1
35:05heterozygote and homozygote.
35:07And in the wild,
35:08type adding when 7B didn't do
35:09anything is the same synapse density.
35:16OK how much time?
35:17What time is it? So I have 1103.
35:20We got fine time. OK yeah,
35:22I should be coming pretty quick.
35:25So then we use the TVI one layer 5:00,
35:28so we could begin to ask what
35:31happens to the pretrial cortex,
35:32because as psychiatrists,
35:34we know that the pretrial cortex.
35:36Has to be important.
35:38Pretty much.
35:38That's what we
35:41and wonderful work from
35:43Patricia Goldman and Rakesh,
35:44and we aren't That and others In Pasco.
35:48I've shown that prefrontal cortex is
35:50important and not now it's figuring out
35:52how you get a lot of prefrontal cortex
35:54and Cartec About Tabby Raman, in fact.
35:56Beautiful work showing how the thalamus has
35:59an important role written with signaling,
36:01important prefrontal cortex
36:03maturation and identity in any case.
36:05So we wanted to do something in
36:07pretrial CORTEX for all those reasons
36:09and we use the layer 5 deletion to
36:13do that and that paper was published
36:16to that's the reference down below.
36:22OK. And we wanted to see whether we
36:26could make a difference in these mice
36:29by restoring wind signaling to their
36:32behavior and pretrial CORTEX function.
36:35And so I learned from my next door neighbor,
36:39Ben Cheyat, that lithium is
36:43a agonist or wind signaling.
36:48Because years ago Ben's teacher
36:51and Randy Moon had shown during
36:53development of Zenith this it lithium
36:55functions of winds with agonist.
36:59And lithium of course is
37:01fairly safe drug for humans.
37:04And then we also use a G SP3 beta inhibitor,
37:07which will also activate when signaling
37:10to see whether we could change the
37:13synapse biology of these illumines,
37:17Okay. So I'm just going to tell
37:19you about the lithium experiment.
37:21And so Civash gave one intraperitoneal
37:25injection of lithium at post Natal 32
37:31and then just one day later.
37:34Harvest the brain as we've learned
37:36from Ben that you could increase
37:39synapses within 24 hours with lithium
37:42in his synapse assays in vitro.
37:44And then we did our handy dandy blue assay.
37:47So this is just a control showing
37:50you that normal Iran mutant,
37:52Iran mutant has less mature
37:54blue spines. So
38:01this is the controls well as it's the control
38:05without lithium control with lithium.
38:07No big difference in the lithium didn't
38:10do much to the wild type animals.
38:12Here's the mutant that has
38:15less blue density and then we
38:17we increase the blue density.
38:18If you compare this to this picture,
38:23see if I modified the result
38:26and showed that there was
38:31rescue. The difference between
38:32the wild type and the mutant
38:33density was not significant.
38:35Even the mutants didn't quite
38:36get all the way up there.
38:38So amazingly, this is magic to me
38:42that lifting within one day would
38:44make mature synapses out of those
38:47mutant immature synapses one day,
38:58and then the next thing we
38:59wanted to find out was, was it.
39:01Would this be true for an adult
39:03or even grandpa mouse mutant,
39:06for his whole life had decreased synapses?
39:09Could we restore the synapses
39:12in this situation? And so this,
39:15that's what this paper is about. I'm
39:21just showing a couple of there.
39:23So we gave one dose of postnatal 830 and
39:27then analyzed six months later one we did.
39:36And so this is the control
39:38of this grandpa mouse.
39:41It still has lower synapse
39:43density and windows. I don't know.
39:47It's hard to believe,
39:48so don't believe it,
39:49but this is what we found.
39:52How could this be possible?
39:55I'll just say quickly that one way
39:58to be possible is that those immature
40:00synapses had a lot going for them.
40:02They had almost everything they
40:04needed to become mature synapses,
40:06but they needed the kick.
40:09And when signaling evidently was
40:12enough to give me that kick,
40:15another gene that we found,
40:17rescue synapses in our in vitro assay,
40:19is a kinesin,
40:20which is a motor protein that moves
40:23presynaptic vesicles in the position.
40:25And so a simple hypothesis would
40:28be that the lithium somehow allows
40:31the synaptic vesicles to move and
40:35dock in posing the post precinct.
40:38Postnaptic membrane and fuse and
40:41bring in the necessary components
40:43for the for let's say NMDA receptors,
40:47maybe after receptors to then
40:49allow that synapse to form.
40:51I don't know how this would
40:52work for the average synapsis.
40:58Then with excellent neurophysiologists
41:01at UCSF, Andrew Nelson and Kevin Bender,
41:05they looked at these old mice.
41:07Indivo using slices and this is the control
41:13showing that at this age they still have
41:16less frequency of miniature and EPSC's.
41:20Here's the here's the wild
41:22type and here's the null.
41:24I don't know that we showed that for the
41:27heterozygote in this experiment and then
41:31again with our single dose of lithium
41:34post natalate 30 in this case testing.
41:37Not as long,
41:38but one to two months later you
41:40got restoration of the EMSEPSCS.
41:43They're both by neuroanatomy or his,
41:47his Histology assays as well
41:50as electrophysiology assays.
41:51We have convergent evidence that T PL-1
41:55mutants as adults have reduced synapses,
41:58but those can be rescued
42:01morphologically and physiologically
42:02with a single dose of lithium.
42:07So the summary of Part 2,
42:09which is our synaptic phenotypes,
42:11either one promotes excitatory
42:12inhibitory synapses both on layer
42:145 and layer six animal cells.
42:16It does that in part by promoting
42:20Keep Win 7B expression.
42:23And these synapse, spine and synaptic
42:26defects are rapidly and stably
42:28corrected by a single dose of lithium,
42:31which perdures for several months.
42:35I didn't show you this,
42:36but this is really fascinating.
42:37They also have a very mild
42:38cortic thalamic defect.
42:39The axons broke to the thalamus but
42:42don't enter the medial thalamus.
42:44But giving lithium one day you
42:46get a teeny amount of growth,
42:48but you don't need much
42:49get that extra growth.
42:51We don't know whether that corrects
42:52any Physiology in the thalamus.
42:56Okay. Now just three minutes. Finish up.
43:02Because so Hall is a graduate,
43:04Mark Turner had looked at social
43:07behavioral defects and associate
43:11physiological measures in the
43:13preparal cortex in the TV R1 mutants
43:16and the effect of lithium on those.
43:20So this is using a interaction acid
43:24through index of socialization of mice.
43:28Wild type and the tibro and mutants.
43:29So the tibro and layer 5,
43:31the guys had the preampron cortex
43:33problem, have less socialization
43:41and you give them lithium.
43:43So we gave them lithium, waited four weeks,
43:45did the social assay and we got some
43:48improvement on their socialization.
43:51I trust this because my lab
43:52did not do the social assay.
43:54We did. We gave the lithium,
43:56we gave them the mice and then.
43:58I hope they were blinded.
43:59I can't remember they're better than blinded.
44:01But let's just assume they are blinded.
44:04Okay. Then they got fancier.
44:06They put a one of these endomicroscopes
44:09into the prefrontal cortex and measure
44:12calcium imaging as a as an indication
44:14of whether a given behavior was
44:17activating cohorts of prefrontal cortex
44:19neurons during a social behavior.
44:22And they they this is their behavioral
44:25testing either the social assay before.
44:28The lithium and then after the lithium
44:30they also did elevate plus maze which
44:31I'm not going to tell you about.
44:33And their social acid basically is
44:36they introduce a mouse our mouse to
44:40mouse one they rested for a while
44:42introduced to an object just into another
44:44mouse and then they reintroduced to
44:46one of these familiar mice and they
44:48measure the interaction time as their
44:52index of socialization and while
44:55while they're doing that they're.
44:57They're measuring calcium imaging in
45:00the prefrontal cortex. This is right.
45:03I get fuzzy on this stuff.
45:05They get these kinds of descriptions of,
45:10well, this is the wild type
45:12calcium imaging of cohorts of
45:14neurons and prefrontal cortex,
45:15and that's what the distribution of
45:18amplitude and number of cells looks like.
45:21And then that's what looks like
45:23natiro mutant and the difference?
45:25Between this shape and this shape
45:26is statistically significant
45:27with the P value of point O1.
45:30So Tibo and mutants have problems interacting
45:34with a new mouse or a familiar mouse.
45:38This is just a in vivo calcium
45:41imaging correlation of the social
45:43thing that I showed you.
45:45Then they did it after lithium and they
45:48got indistinguishable P values for these.
45:54Indices showing that we could rescue in
45:58vivo Physiology with lithium as well.
46:02So
46:07that's the end of that data.
46:09Summary stories tells you is social
46:11deficits that are present in the
46:13layer 5 units are rescued by lithium
46:16and the in vivo activity of layer
46:195 neurons in the medial prefrontal
46:22cortex are also rescued by lithium.
46:27So where, where does this take us?
46:30Or maybe suggest that one can have
46:34a therapy for at least humans who
46:37are heterozygote or the tiger one,
46:40no mutation with lithium or
46:43other kinds of wind agonists.
46:47And I think, you know,
46:49that's the pipe dream.
46:50And here's a little bit more of a pipe dream,
46:51maybe other forms of cognitive
46:54disorders of child and childhood.
46:57Have at least as a component
46:59of their phenotype this type of
47:02problem with synapse formation
47:04that might be amenable to treatment
47:07with increasing wind signaling and
47:11that's it. I I tried to mention I
47:16think I mentioned everybody here
47:18they are again and those are the
47:19Blue Angels flying over the Golden
47:21Gate Bridge recommend coming to.
47:31So we'll manage questions both.