Modeling Neuroepigenetic Disorders In 2D And 3D Neurons

March 30, 2023

ID9778

To CiteDCA Citation Guide

00:00It's my pleasure to
00:02introduce Doctor Yang,
00:05who is received this MD from Shanghai
00:09Medical College of Punan University and
00:12his PhD at Baylor College of Medicine.
00:15And he completed his pediatric residency
00:18and Clinical Genetics Fellowship at
00:20Texas Children's Hospital at Baylor,
00:22at Baylor and then joined the Yale
00:25Department of Genetics in 2019.
00:27His research focuses are the discovery of
00:30rare diseases and modeling neurogenetics
00:33and neuro epigenetic disorders using
00:35human derived ipscs and mutant mice.
00:48Thank you. All right.
00:57Well, thank you for and hyphen
00:59for organize this event.
01:01Also for the invitation,
01:02I shoot the clothes.
01:04I'm not stem cell biologist,
01:06I'm not biomedical engineer either.
01:09So while I'm here,
01:13sorry, how do I convince?
01:17So how do we get interesting for the
01:20bring organize IPSC and 2DS3 neuron?
01:24So I'm kind of called genetics. I see.
01:26The patient in the clinical. Sorry
01:30for real genetic disorder
01:31and then back to the lab.
01:33Wow, that didn't work. Sorry.
01:40Go one more time.
01:45Oh, OK. Thank you. That means we
01:47need every step together, right? So.
01:52Next lab, we can try modeling the
01:55genetic new epigenetic primary aspect
01:57of brain development in the lab,
01:59primarily using the Moss model
02:01or other animal model,
02:03understanding the function of gene,
02:05understanding the disease mechanism,
02:07passive Physiology as a physician,
02:09definitely interesting develop
02:10a treatment back to the family.
02:13However, as probably I would say
02:17almost all the success we learn
02:19primary from the brain from the mice.
02:22Did not translate well come to the
02:25human we fail miserably for many,
02:27many of the successful exciting
02:30story for mice.
02:31So that's what we're asking could
02:34we kind of study for the IPSC,
02:37derive patient, derive 2DS3 neuron,
02:39feel this gap, but just this challenge.
02:42So I'm going to give you two
02:44example in in my lab,
02:45probably focus on the
02:47first one the time I say.
02:49So I hope you feel these two disorder
02:52are very interesting in general.
02:53So this is Endrum syndrome which is
02:56oftentimes many syndrome named by physician,
02:58the first recognized name endrum,
03:02it's a very classical severe end of
03:04the new developmental disorder IQ,
03:06it's very low IQ like 20,
03:08they don't speech at all,
03:09don't have any speech.
03:11More challenges and they have
03:13very severe epilepsy and almost
03:161/3 is medically intractable.
03:18It's very devastating to the family.
03:21It's very interesting molecular basis,
03:24a primary genetic defect IS15Q11Q13
03:28and matured patient have a 15Q11Q13
03:31deletion cross this region,
03:34but interestingly because
03:35it's imprinting related.
03:37So the paternal delition delition
03:39come from paternal chromosome
03:41caused a complete separate syndrome
03:43called the pero alloy and with
03:45the delition come from mother
03:47maternal alloy cause the end German.
03:50So over the time we know the gene
03:53response for this larger delition
03:55it's ubiquitin protein like this 3A.
03:58As more interesting this region is,
04:00we also know maternal duplication
04:03only maternal duplication from mother.
04:06Costs about 1 to 2% in the Ed Pass
04:09autism but not paternal so you can
04:11see it's very very interesting.
04:13If a duplication for father
04:15relatively normal
04:18so. So with with over the time we
04:22learned this is more complex sort of
04:24a genetic epigenetic defect majority
04:26of the logic deletion we have a poor
04:28mutation in the maternal chromosome you
04:30put in like H3A we have another two
04:33class of a where kind of uniprantal
04:35dysomy 2 comes come from same parents
04:37and of imprinting defect that's also
04:40small number for case color you B3
04:43gainer function contributor for autism.
04:47So now you put in like a G is an open.
04:51H3A some of you very familiarly percolation
04:53pathway and the most interesting to
04:55us is this is epigenetic phenomenon
04:57kind of imprinting the expression
04:59for the gene in the next generation.
05:01It's depend where this come from.
05:03So for the ENDROOM gene actually it's
05:06very interesting which is first kind
05:09of new specific imprinting gene in non
05:11neuro both Gene Express both earlier express.
05:14In the neuron in the brain only
05:17maternal allele expressed.
05:19So that's the how interesting this
05:21phenomenon is and over the time we
05:23and many other understanding for the
05:26mechanism how could this they cell
05:28type specific as a less specific
05:31infinite phenomena happen.
05:32It's actually due to a very very long,
05:34almost mega based long non coding a RNA.
05:38Are part of what we called also antisense
05:40for the UB3 gene only expressed from
05:43paternal chromosome then silence for
05:46the sense on the paternal console.
05:48So that's the mechanism and we also
05:51generate a many Moss model over
05:53the time to study this mechanism.
05:55Overall Moss model provide many many
05:58valuable insight however that's.
06:02Also the capitulate a lot of the human
06:05phenotype reasonable well especially
06:07we're interesting for the epilepsy
06:10or abnormal EEG.
06:11However then we take this one on try
06:14understanding it because as you know
06:15one set of patient have no control for
06:18the seizure or lifetime which is very,
06:20very challenging.
06:21So we try understanding,
06:22use the muscle model industry
06:24understanding why this epilepsy so
06:27common just highlight one phenomenon
06:29we use this is very specific.
06:31To Physiology phenomena measure the
06:34action potential with particular
06:36folks on fast component after
06:39hyperpolar polarization.
06:40We realized in this particular engine
06:42mouse model in the brain and neuron
06:45this FHP is increased and we have done
06:47a lot of work using the biochemical
06:50molecular and linked it to the Ek channel.
06:53It's enhanced function for BK
06:56channel contribute this phenomena.
06:58Then we also show this link to indeed
07:00in link to the epilepsy in the mice,
07:02which is you can use the antagonist
07:05Hassel and can reduce the amplitude
07:08and the frequency eventually suppress
07:11the seizure in the mice.
07:13So that's all good.
07:14The question come to whether this
07:16is translated between the human
07:18so that come to what we got into
07:21the IPSC 2D and three you are.
07:24That's one time I moved to the
07:26Yale and then I realized the world
07:28class of the stem cell center.
07:30So I talked to Hifi and India and the child
07:33home and say hey why we just do this
07:36create a repository for the Andrew IPIC.
07:40So luckily we got a very generous support
07:43for the fast foundation for Andrew
07:45and therapeutic they give very general
07:48support and right before the COVID.
07:50So over the last three years
07:51even during the COVID,
07:52we were able to generate.
07:55One day cell line IPS IPSC cell line from
07:59different genotype including the control.
08:01This is free to everyone here.
08:02If you're interested you can just e-mail me
08:05are free to distribute it to each of you.
08:07So that allow us to really ask the
08:10question whether phenomenon we study from
08:12mice is the translator from the mice.
08:15So I'll just give you a few slide and
08:17the summers are published already.
08:18And then to shows the phenomena,
08:21same phenomena we observed from
08:23the mice which you can see the FHHP
08:25which indeed in IPSC the wife the
08:282D new one in the cortical new one
08:30in this is enhanced the two.
08:32So that's increased the frequency
08:35suggestion the hyperexcitable new
08:36one and then we all can can use
08:39in the sort of rescue indeed it
08:41shows that deficiency UPC is a
08:43responsibly electrophysiology.
08:44Phenomena.
08:44Now we got to choose also biochemical assay.
08:49Indeed it's correlated for the BK
08:52channel function in the 2D neuron.
08:55Now you can show also the
08:57paxilance same as in vivo,
08:58in the mice can suppress
09:00this hyperexatibility,
09:01but in the 2D neuron,
09:05now we move on together.
09:13That's we cannot advance. So that's
09:21right. All right. Maybe I did
09:23something which I should not do.
09:28So then when you organize 3D neurons, it
09:30shows very similar discovery of funding to.
09:33Yeah, same thing, the increase the.
09:39Frequency then the pass and
09:41suppress the hyperexcitability.
09:43Of course the question will come
09:45to whether you can recapitulate
09:46epileptic form in the brain organelle.
09:49That's question we still have not
09:51get into that very very very detail.
09:53So that's all looks good because we
09:55can study the mice to translate the
09:57human allow us to the confidence
09:59maybe indeed we can allow these
10:01two system to testing additional.
10:04Particular for treatment strategy.
10:06So one of the thing we are working on
10:09right now is to try to using this strategy.
10:12As I told you,
10:13I could not get into the very detail.
10:15So the mechanism regulate this
10:17imprinting is due to the antisense
10:20Bay long megabase continuous
10:22antisense long line coding RNA.
10:24So the one strategy is if we can
10:27disrupt this long line coding RNA
10:29and then you can reactivate the gene.
10:31On matpat and chromosome is supposed
10:33as like a gene therapy can approach.
10:36So the ASO has proved it's effective and
10:40it's also in the phase one trial right now.
10:42So we think about it with whether
10:44we should do a more permanent fix.
10:46The ASO we need every month sort of
10:48the eye spinal injection and now we
10:51work on this with CRISPR additive.
10:55So working with collaborative with the.
10:58People from the biomedical engineer as
11:01deliver crisp to brain is a challenge.
11:03We actually got this piece of very
11:05exciting data by our AIS in the lab
11:08shall now shows you we use this
11:11chemical modified I MP conjugated
11:13CAS 9 protein and and gala and
11:16they together deliver IT intracego
11:18injection which you can see.
11:20Amazingly this deliver the editing in
11:22the cortical neuro and cerebellum.
11:25Which you can see this a new M cell,
11:27then this green cell,
11:29it's a reactivation after editing
11:32the anti sense.
11:33It's almost like 70% efficiency for
11:36this coach similar to the cortico
11:39that is pretty amazing.
11:41So we feel like this will be the next step.
11:43We are watching actually active working
11:45on the 2D and the 3D neuron right now,
11:47see if for the same delivery it's
11:50effective if entry before we
11:52go to FDI&amp;D and to the human.
11:55So I'm gonna switch the GAIL
11:58for second disorder,
12:00which I hope that you will
12:01find also very interesting.
12:03In the same scene I could
12:05genetically related disorder.
12:06So this is a patient I saw about four
12:08or five years ago in the clinic.
12:10It's a very similar to engerman but
12:12that's definitely severity it's 11
12:14mile and moderated compare engerman
12:15and to me it's a severe and and this
12:18is a moderate and with autism as
12:20a predominant feature intellectual
12:22disability interesting they have
12:24macrosuppony which the big brain is
12:26the bigger they have a low percentage
12:28of low frequent preference of epilepsy
12:30too and so so this is the the boy
12:34and what he was nine years old and then.
12:37Later on then was to find interesting
12:40another end of a phenotype.
12:42This is the
12:44sort of a longitudinal sort of picture
12:48from the infant to when he was thirty.
12:52I hope you probably say okay,
12:53that's probably not 30 years old face,
12:57it's probably more old than that.
12:58So that's a premature Asian phenotype
13:00that's we think it's also the other end.
13:02It's very interesting.
13:03That's a delay early new
13:06development somehow later on.
13:07Is actually accelerated aging process
13:11so that's clear a puzzle and and then
13:13we're looking for the genetic we we
13:15did a whole accident in the clinic we
13:17identified the first mutation is patient
13:19other colleague from UK also similar
13:23time 2017 2018 report a few other
13:26case and eventually last five years
13:28will accumulate almost 100 case now.
13:31So what you notice quickly matured
13:33for mutation in the C terminal domain.
13:37And majority for them it's
13:39A-frame shift mutation.
13:40What's more interesting when you
13:42do the computational prediction,
13:44the open reading frame,
13:45actually it's quite interesting,
13:47you can see here if you do the open
13:51reading from open reading from prediction,
13:54regardless where the mutation
13:57location near all and very very same
14:00tail about for the amino acid at N.
14:04So that's to me is a little unusual.
14:07I'm a geneticist,
14:08I see a lot of patient database.
14:10I have not seen this kind of
14:12phenomenon very often.
14:13If it happen one of you,
14:14you have some case like this talking
14:16to me and we kind of working
14:18together to figure out this puzzle.
14:20Interestingly we also generate and
14:23antibody specifically against this tail.
14:26The antibody actually very easy
14:28to generate because this tail if
14:30you against the genome or podium
14:32actually pretty unique.
14:33So allow you to very quickly
14:35generate this antibody.
14:36Now you're testing the patient IP.
14:38At the same time we generate about a
14:41IPSC cell line from this patient too.
14:45So you can see this IPSC cell
14:47line and this abnormal tail,
14:50it's indeed it's stable.
14:52So that raised the question
14:54whether this abnormal tail.
14:56It's actually gain of function or
14:58dominant net function because when
14:59you now call this gene in the mice,
15:01they have no significant phenotype.
15:04But in the habazygs in the human,
15:06it's very definitely a very severe phenotype.
15:09So that we interesting also this
15:11is same mutation or same mutation
15:13in the mouse gene or mouse gene,
15:16you won't be able to create the same tail.
15:19So it's only to the human coating.
15:22So that create a little challenge
15:24to to manipulate in the mice but
15:26of course we can't do it.
15:28We made in the humanized mouse model
15:30by engineering the entire human gene
15:32in the marketing of the car knocking
15:35and replace the anti mouse genome.
15:37So that's ongoing.
15:39We're just talking to people outside
15:42like it's a mouse have azyg have
15:44mild phenol type homozygic actually
15:46end the post Natal early so.
15:48So,
15:48but in any way so we will say OK
15:50that would be good to kind of look a
15:52little more this is a very definitely
15:54because my title is a new epigenetic.
15:56So it's a, it's a H1 link protein
15:59as you know probably have about
16:0011 H one link some of the somatic
16:03form is one of five somatic form.
16:06So this H1 link and the function
16:08for H1 link in largely we still
16:10don't know because over the last
16:1325 years lot of people study cold.
16:15For histone, very, very detailed,
16:17but this is almost like a forgotten
16:19histone for over the last decade.
16:21But now you can come to interesting for
16:24many people because the human disease link.
16:26So again it's a link histone with
16:29the link DNA and it's core histone.
16:31Presumably function is making the
16:34chromity more compact but it's a
16:36very basic component of the histone,
16:39but the link to very selectively
16:41human neuro behaviour and neuro
16:44developmental phenotype.
16:45So, so that's what we would generate
16:47a panel of an IPSL again,
16:49again a child home and in yon.
16:51And for this effort,
16:53sometimes people ask and say
16:55why you didn't do it at Duke.
16:57And I would say, well,
16:58yeah, we want to do at Duke,
17:00but we don't have a facility like here.
17:03I find I left two years ago
17:04before I joined the Duke.
17:06But that's what we lost high
17:07final deal and now we don't have
17:09the stem cell facility there.
17:10So that's what's asking you should have
17:13taken advantage of for we have indeed
17:15excellent facility environment here.
17:17So again remind you this phenotype
17:20patient have a macrocell,
17:21a big cell,
17:22a big brain.
17:23So you look at the cell proliferation
17:26indeed somehow suggesting that
17:28maybe correlate the human phenotype
17:30they are proliferated faster.
17:32Then both in IPSA and MPSC new one early
17:37precursor and we also did it for the I,
17:39I and they seek and looking for.
17:42Whether it's a chromatin structure
17:44affect the downstream transcription
17:46indeed is when you have this mutation
17:49chromatin sounds like more relaxing.
17:51So more gene up regulated and
17:53surprising to us it's you actually
17:56see the set of a gene regulated,
17:58it's actually still in the
18:00chromatin related gene.
18:01Many ask actually chromatin related gene.
18:03So we are very,
18:05very intriguing by this founding.
18:07We also look at because this is a
18:09chromatin structural later protein,
18:10we asking whether that's actually
18:13indeed affect the chromatin nuclear
18:15morphology or chromatin structure.
18:17So we did a EM for this IPIC and
18:20and the neuron we look in the
18:22morphology of the nucleolo and
18:24the nucleus indeed that's alter
18:25the morphology from the nucleus.
18:28So indeed because the unique
18:30this protein alteration which
18:32are not mimicking in the mice.
18:35So that I PRC derived 2D and 3D new
18:38one actually indeed it's the one way
18:41to to do more sort of investigation.
18:43I do have time to show you the
18:45organized other data on different time
18:47and almost close perfect time there.
18:49So I hope that by using the two
18:52example illustrated 21 is that
18:54you don't have to be the stem
18:56cell biologist to study stem cell work.
18:59Second is I hope you also.
19:02Data from if you study mice
19:03actually indeed it would be good to
19:06translate what your mouse discovery
19:07of funding from mutant mice to one
19:09more step between the human device,
19:12this neuron before moving to the
19:15we call it FDI&amp;D study because
19:18we are fail many many occasion,
19:22many example everything is
19:23beautiful in the mice,
19:25it's a fail in the human stage.
19:28So large of the people in the lab
19:31primarily I think the people Kung
19:33Yun is the primary thing is now here
19:36working on the andrewman IPSC and
19:39IPSC organelle and then so nice data
19:42from the mouse for the IP and then
19:45a few other kind of also work in the
19:47similar sort of standing project.
19:49And then of course without your stem
19:53cell center support this kind of large
19:55scale production will not be the feasible.
19:58And all the support I've received and
20:00also the collaboration with Duke and
20:03the EOS for the engine work together.
20:05Thank you.