A New Generation of Tools For Experimental Modeling Of Brain Structure And Function

March 30, 2023

ID9770

To CiteDCA Citation Guide

00:00Let me introduce Doctor Andrelevchenko,
00:03who is the next speaker.
00:05He graduated from the Moscow
00:08Institute of Physics and Technology
00:10and obtained his doctor,
00:12obtained his doctoral degree from
00:14Columbia University and after
00:17postdoctoral experience at Caltech,
00:19he started his independent position at
00:21Johns Hopkins in 2001 and was recruited
00:24as the founding director of the Ill
00:27Systems Biology Institute in 2015.
00:29And he's also the John Malone
00:32professor of biomedical Engineering
00:34and professor of physics.
00:37Please.
00:44Hello and thanks for inviting me.
00:46And as as Angelica just said for
00:50engineers and physicists to be in the
00:53in the room with all of you is, is,
00:55is always the great pleasure and and very,
00:58very interesting to us.
01:01It's of course I'm an engineer,
01:02but I'm actually we do do quite a
01:04bit of biology and what you learn
01:07of course is being an engineer.
01:09If you develop tools and we know,
01:11all know with with tools you
01:15get to discover new tools,
01:17new discoveries and that's something
01:19that we've really enjoyed thoroughly
01:22on different different scales.
01:24And what I'd like to illustrate
01:26today is the use of these tools.
01:29And some insights that we can gain
01:31from them and the fact that you
01:33actually can do it on multiple
01:35different scales in terms of sales,
01:39small organoids or even larger structures,
01:45okay. So let's see if this. Yeah.
01:49So I always run out of time,
01:51hopefully not today.
01:52So I'd like to immediately thank all
01:55the people who helped us do this,
01:58as our collaborators as well as the
02:00members of the lab were truly in the
02:02trenches doing all of this work.
02:03Particularly for a lot of the
02:08work towards the ends of flora,
02:10because this has been a very,
02:11very exciting collaboration
02:12with the factory in the lab.
02:15And a lot of things are not shown are also
02:18very interesting to us and are always,
02:21almost always products of collaboration.
02:23And as on the hill,
02:24like I mentioned for example,
02:25we have been doing a lot of
02:27interesting things with her lab,
02:28which has been amazing.
02:30So what I'd like to really focus on
02:35is this combination of difference
02:37approaches and how they come together
02:40more specifically in the context of.
02:43The fact that as we've already
02:45heard the tissues and you know,
02:48even if you go smaller on the
02:50level of digital cells,
02:52what you see is that the
02:54environment is not uniform.
02:55The environment can present cells
02:57and tissues with gradients,
02:59for example of cues of morphogens.
03:03And this can occur both for stages
03:06in development or in cancer,
03:09for example, progression where
03:10cells may migrate gradients.
03:12Could be in collective cell migration or
03:16reorganization of tissues in development.
03:19It could be in homeostasis and wound repair.
03:23So many, many instances.
03:25And of course if you take
03:28developmental biology,
03:29I guess more or less,
03:32you know, almost anywhere,
03:33at least where I took it,
03:35you find that there is this model,
03:37beautiful model due to Lewis Wolpert.
03:40Of the French flag,
03:42essentially suggesting that in
03:43the gradient of morphogens,
03:45you can have multiple different
03:48fates emerging due to the overall
03:51level of morphogens,
03:52sort of sort of like the cells
03:54responding to the doses,
03:55different doses of the input.
03:57And it does frequently happen in
04:00tissues and hopefully towards the
04:02end I'm going to show you an example
04:04that Flora has really introduced us to.
04:09Now,
04:09so how do you do experiments to
04:12try to understand the influence
04:14of graded
04:15inputs, whether you look at cell migration
04:18or this developmental processes where
04:21you have multiple different responses.
04:25So there are there's a tradition
04:27in by engineering that you may or
04:29may not be aware of for kind of a
04:31progression of different steps in
04:32development of different tools.
04:34And these are just some of the examples
04:37essentially historically how the
04:39gradient studies have been done and
04:41almost always what you see and the ones
04:43at the bottom are the ones that we developed.
04:47The ones at the top are the
04:49ones that were used before.
04:50You see that it's a kind of
04:52two different ideas.
04:53One is the flow.
04:54In the flow you have mixing of
04:57liquids and within liquids you
04:59can have different doses of.
05:01Compound that you're interested in and
05:03that gradually may generate the gradient.
05:06The issue is that of course
05:07cells don't like flow of liquid.
05:09For the most part,
05:11they die almost immediately.
05:12They're very,
05:13very sensitive to what happens
05:15with the environment.
05:16If you shake, you know your flask.
05:19Sometimes you'll see a cell death,
05:21and that's what you face when you try
05:23to introduce cells into anything that
05:25flows when you hear microfluidics,
05:27microfluidics.
05:29Fluidics parts should really generate
05:32immediately some trepidation for you
05:35because anything again that flows
05:38almost always there are instances
05:40where flow is important for sure,
05:43for example in the material cells,
05:44but generally cells are really sensitive
05:46to that and they really don't like it.
05:48So that is something that we recognized
05:51almost immediately started when
05:53we started working with cells and
05:55then we had to develop a different.
05:57Series of devices where it's
06:00really all diffusion, right?
06:02It's not nothing flows.
06:03The cells are actually in this beautiful,
06:05very steady environment,
06:07but there are gradients and so with that.
06:11So you have this example on
06:13the left here where you have.
06:17I'm not, I'm not going to mess it. Okay.
06:19So you have this idea of a source,
06:22for example of growth factory GF.
06:25And the cells in between
06:26looking like fish here,
06:28but this is a cell and there are this
06:31nice channels connecting source of
06:33the GF and medium without the GF and
06:36the certain gradient developing and
06:38the the cells actually do respond.
06:40And so you start seeing how they run
06:43very happily, they run into each other,
06:45they collide.
06:46You can study what happens when that
06:49when that occurs and you see this
06:51beautiful chains of cells going up and down.
06:56And let me try again,
06:58maybe it will work, maybe not.
07:02So we wanted to really extend
07:04that a little bit to you know,
07:07we've heard that cells actually live in
07:09softer media and they're surrounded by
07:11the cells that communicate the form tissues,
07:14tissue ensembles.
07:15And so can we really extend
07:19this analysis now to?
07:21What really is at the center of today's
07:23workshop and that is organized.
07:25And so this paper was the first
07:27attempt to do that where it wasn't.
07:30It was an organoid of breast tissue
07:33that you'll see next and rather than.
07:37Brain tissue that we'll see a bit
07:40later and this was surf the question,
07:42okay.
07:42So what happens,
07:44we know that in the breast there's branching,
07:48there's in the periods of lactation
07:51especially or around that time
07:53there may be a very significant
07:55reorganization of the breast tissue.
07:57And in that process in the adults
08:00what happens is there is a growth
08:02and branching of various surf.
08:04Parts of the of the tissue and that
08:08is is triggered by e.g F the same
08:11same compound that I showed you in
08:13the in the last movie can trigger
08:15migration of cells in a directive way.
08:18So what happens with tissues?
08:21And So what you can see is that
08:22you again you can sort of extend
08:24this technology to the same idea,
08:26have a gradient now if e.g F over
08:29much larger distance and organized
08:31are embedded now in the space.
08:34And you start seeing that they
08:36actually start branching.
08:37This organoid that initially was
08:38the kind of a spherical thing,
08:41begins to branch in a very directed
08:43way towards the source of e.g F
08:46what's interesting is that you can
08:48either induce it or naturally have
08:50some single cells around and they
08:52actually don't sense these gradients.
08:54So in spite of the movie that I
08:56just showed you in the previous,
08:58the the ligand concentration
08:59gradients in this case are so shallow.
09:02That individual cells just don't respond,
09:04they don't sense the gradients,
09:05only the tissues can sense the gradients,
09:08which implies that there is some
09:09sort of cell cell communication.
09:11And so our analysis suggested that's
09:15really the cell cell communication
09:18mechanisms in this process,
09:20this one branch forming here that
09:23will branch more is,
09:25is really all of that is mediated
09:28by calcium signaling.
09:29And calcium communication between the
09:32channels connecting the cells.
09:33So one thing that I want to
09:36emphasize already in the two
09:38examples that I showed you where
09:40you in both cases you saw movies,
09:42is that you really benefit not
09:44only from the ability to generate
09:46gradients or grows such organoids.
09:49But also from the fact that
09:51you can really peer into life,
09:53either life cells or life in
09:55this case life organoias,
09:56and see the dynamics of the processes
09:58that are of interest to you.
10:00So these devices really allow you to analyze,
10:04you know,
10:05communication between the cells and
10:07what happens with the cells and
10:10shoot movies like that and analyze
10:12in this case the calcium signaling.
10:15So we wanted to really continue doing
10:17this and of course the challenge the the
10:20more interesting structures the more
10:21complex structures occur in the brain.
10:23And in the brain there may be different
10:26types of analysis and this was termed
10:29for the first time brain on the chip.
10:31Now in the Community of Engineers,
10:35anything on a chip, you'll,
10:36you'll hear longer on the chip,
10:37brain on chip it's just a big name
10:40it's it's not more than that.
10:41But here it's,
10:43it's,
10:43it's it's something that allows
10:46you to start modeling the presence
10:49of multiple cell types in the self
10:53organizing networks and again do it
10:55in such a way that can visualize this
10:57in great detail and see what happens.
11:01So here again you can start
11:03with the progenitor cells.
11:04They will not necessarily form and organize,
11:07they can form clusters of cells that
11:09are connected by bundles of axons.
11:11At some point you can couple that to a
11:14layer of endothelial cells and mimic the
11:17blood brain barrier that actually forms here.
11:20And you can introduce some drugs here,
11:22for example,
11:23on this side,
11:24and study how they can potentially be
11:27penetrating this blood brain barrier.
11:29There's a basement membrane
11:31that will form here.
11:32And so even though of course
11:33it's not the real tissue,
11:35it starts having some interesting
11:36features that you can use.
11:38To start exploring what happens,
11:40there is a communication as was just
11:43mentioned by antalicate between the
11:45neuronal cells as they different
11:48shades and endothelial cells.
11:50And it can introduce,
11:51which we did in this case.
11:52Also project cells into a more
11:55developed network and see how the
11:57presence of both neuronal cells and
12:00in the field cells make may control
12:03the behavior of this progenitor
12:05cells neural progeneral cells.
12:07And you can again introduce the gradients
12:09now of variety of different cues.
12:12So I could be B&amp;P we've heard about some
12:16examples have already been enunciated
12:19and so you can have introduction
12:21of multiple gradients can attended
12:23that can potentially either Dr.
12:25migration of the cells and how they
12:27position them themselves now in the
12:29more realistic model of this brain
12:32tissue or what happens to them in different.
12:35Concentration within this gradient,
12:37right.
12:38And so you can study that and of
12:40course again you can visualize
12:42what happens within
12:43the clusters. So again,
12:44this is calcium imaging before you
12:46saw that in the memory tissue,
12:48organoid in this case,
12:50it's neural tissue, not yet organoid,
12:52it's cluster of cells,
12:54but you can already see how they
12:58show this neuronal phenotypes,
12:59how they communicate with each
13:00other and they can visualize
13:01calcium signaling in them.
13:05So of course I already mentioned
13:08flora multiple times, and I think
13:11she's going to talk about that more.
13:13So I'm not going to go into biology of
13:17of what happens with embryos and what
13:19happens with the development of the brain
13:21very much other than to say that again,
13:24there are gradients, of course,
13:26in the developing embryo.
13:28Of multiple morphogens and
13:30multiple signals that really
13:32define axis for developing embryo.
13:35And it again is of interest to see
13:38what happens either with cells,
13:39single cells or cell ensembles
13:41that they just showed you,
13:42or even with organoids that may be
13:46exposed the gradients of various
13:48cues and it could be the actual.
13:50Signaling molecules for something
13:52that mimics them.
13:53For example,
13:53this G SK3 inhibitor is widely used
13:55and used by flora quite a bit.
13:57And so we tried to develop this radiance
14:00and analyze the outcomes of that.
14:03And again as I said,
14:05Flora will likely,
14:06I don't know exactly what
14:07she's going to talk about,
14:08but she will talk a little
14:10bit more about this.
14:12And so this devices again are
14:13very similar in some sense,
14:15but but have now been optimized
14:17and developed and that does
14:19take quite a bit of time.
14:20Forebrain organize that are much
14:22more complex than the ensembles
14:24that they showed you before.
14:26And this is what you do,
14:28is you try to optimize all of this.
14:29You do both analysis and experiments and
14:33modeling of what happens in such devices.
14:35Ultimately,
14:36of course what you look at is
14:39the outcome of the genetic level
14:42of expression of different
14:43markers of different tissues you
14:45can play with concentrations,
14:46different concentrations you can.
14:49Do different types of analysis in
14:53terms of for example different
14:55hydrogel composition of collagen.
14:57We just again heard from Angelica
15:00that the extra cell metrics is
15:02very important and so how does
15:04it affected how the gradients of
15:06different morphogens affected.
15:07And of course even visually you
15:10can see that depending on where
15:11you are in the gradient,
15:13in this case still 1 dimensional gradient.
15:17It really defines the outcome in
15:19terms of the differentiation and
15:21expression of the differentiation markers.
15:23Now we want to really take it
15:25beyond this and since there are
15:28multiple axis for differentiation,
15:29if you have dorsal ventral AT axis and so on,
15:34it would be wonderful to develop
15:36now fields of this morphogens
15:38that may be 2 dimensional or
15:40high dimensional so that embryos
15:42positioned in different parts of this.
15:44Field can have different combinations
15:47of morphogens affecting their
15:51differentiation and So what you really
15:55get if you succeed in experiment
15:57like this is a really snapshot of
16:00not just one combination of different
16:02inputs or different factors,
16:04but multiple combinations present in
16:06a sort of A2 dimensional those distribution.
16:10And of course by modulating what you.
16:13Do with these devices you can get
16:15all sorts of different gradients
16:17and gradient distributions,
16:19and again you can do various
16:21types of experimental tests
16:23and simulations of all this.
16:24But ultimately, again as we make
16:26hopefully we'll hear from Flora and
16:28this is done very much with her lab,
16:31you can really now begin to examine
16:35within this two-dimensional fields
16:37what happens with the expression
16:39now of markers that will tell you.
16:43About the AP&amp;DV or Dorsodential
16:47here access imposed by Sony
16:50Hedgehog and this compound that
16:51they mentioned before which is the
16:54G SK3 inhibitor and really look at
16:56the rated changes in distribution
17:00of multiple expression markers and
17:02ultimately how it corresponds to the
17:06differentiation in the actual embryo now.
17:10I meant to insert conclusion slide
17:12but it didn't come out and so well
17:14I'll tell you a little bit just
17:16to summarize a couple of thoughts,
17:17I'll leave you with that and also
17:20with what we want to do next.
17:23So again I think one of the things I
17:26would like to, I wanted to illustrate
17:28it's no not the only challenge,
17:30but one of the challenges again is how
17:32do you screen multiple conditions,
17:34how do you explore the influence of graded?
17:38Inputs, it could be attractants,
17:40growth factors, it could be morphogens,
17:43could be other things.
17:45So how do you study that and do you
17:47really have technologies to do that?
17:49And the answer is that the technology,
17:51at least for that part,
17:52is becoming more and more mature
17:55and applicable to multiple scales.
17:57You know, you can play with cells,
17:58you can play with cell small cell
18:01ensembles or even larger organized
18:04and you really can explore that.
18:06One huge limitation of course,
18:08is the actual size of the organoids
18:10that we are playing with, which is,
18:13as already has been mentioned,
18:14is really limited.
18:16For example by the vascularization.
18:17We don't really have vascularization
18:20of organoids,
18:21and therefore whatever nutrients
18:23or oxygen they get,
18:25they get by diffusion from
18:27the surrounding medium,
18:28and that means that you cannot
18:30grow them beyond a certain size
18:32before they become necrotic.
18:33In the at the center and that means
18:36that that's actually one of the
18:38bigger challenges that flora and I
18:40have been discussing quite a bit.
18:41And then and Helica mentioned as well,
18:43how do you really vascularize organoids
18:45and how do you introduce that component,
18:49especially because vascular
18:51component can affect formation
18:53of this structures as well.
18:55So there's crosstalk between
18:58vasculature and neural tissue.
19:01So this it's a huge challenge for
19:03the whole community and we're
19:04trying to tackle it now.
19:05And beyond that of course,
19:07how can we really build it up in
19:10terms of the complexity of both the
19:12tissues themselves and the fields
19:14of different inputs that this such
19:16tissues can experience and do it in
19:19the relatively high throughput fashion.
19:20So this is what we are super excited
19:22right now about and this technology
19:24really has been progressing.
19:25So hopefully it will be widely used.
19:27As as it becomes available to you.
19:29Thank you very much.