Thomas Biederer, PhD & Sreeganga Chandra, PhD

Name: Thomas Biederer, PhD & Sreeganga Chandra, PhD
Uploaded: 2025-10-03T12:44:05.71Z
Duration: 1 h 4 min 46 s

October 03, 2025

About the speakers

Sreeganga Chandra, PhD
Professor of Neurology & Neuroscience
Thomas Biederer, PhD
Professor of Neurology
Jaime Grutzendler, MD
Dr. Harry M. Zimmerman and Dr. Nicholas and Viola Spinelli Professor of Neurology and Neuroscience; Vice-Chair for Research, Neurology; Director, Center for Experimental Neuroimaging

Information

ID: 13487
To Cite: DCA Citation Guide

Download Transcript

00:14Alright.
00:20Hello?
00:22Hello? Hello?
00:24Can you hear me?
00:28Yes.
00:32Alright. We're gonna get going.
00:38Thank you all for, coming,
00:41to our,
00:42for the next month, we're
00:43gonna have what we are
00:45calling our,
00:47research,
00:50symposiums
00:51or bridging clinic what wasn't
00:53the
00:54I forgot about it. Bridging,
00:57you know, clinical,
00:59translational,
01:00and research,
01:02in our department
01:03and, sort of get to
01:05know,
01:06some of our researchers
01:08and,
01:09hopefully, ask questions and, you
01:11know, bring friend you know,
01:14find some common areas of
01:15interest. And, hopefully, this is
01:17gonna lead to
01:18some exciting research and
01:20translation,
01:21ultimately.
01:23So or friendship or friendship.
01:25Yeah. Exactly.
01:27Grants
01:28and etcetera.
01:30So,
01:32just to get going quickly,
01:33I mean, we have two
01:33distinguished, investigators from our
01:38there's gonna be a lot
01:39about,
01:40Parkinson's disease and neurodegeneration.
01:42And I'm gonna,
01:44have Vanessa introduce our distinguished
01:46speakers.
01:50Hello, everyone. Good afternoon. For
01:52those who don't know me,
01:53I'm Vanessa, one of the
01:54fourth year neurology residents,
01:56and welcome.
01:57Today, we're gonna have two
01:58for the price of one.
02:00So first, we're gonna have
02:02doctor, Bierder.
02:04He received a PhD in
02:05cell biology
02:06from,
02:07pardon my,
02:09German,
02:11the Humboldt Universitat,
02:12zu Berlin in Germany.
02:15After his postdoctoral,
02:16training with doctor Thomas,
02:18Sudoff,
02:20Doctor Bitter,
02:21became a faculty member here
02:23at Yale in two thousand
02:24and three. He took a
02:25position at Tufts in twenty
02:27thirteen before returning to Yale
02:28in twenty nineteen
02:30to join the faculty at
02:31the department of neurology.
02:33His current research
02:35addresses central questions to understand
02:37the biology of synopsis,
02:39the cellular structures that connect
02:41neurons into networks.
02:43First, how are new synapses
02:46formed?
02:47And second, which mechanisms control
02:50synapse,
02:51maintenance,
02:51and what makes them vulnerable
02:53and resilient in neurodegenerative
02:55disorders?
02:57To address these questions,
02:58his research group combines molecular
03:00studies
03:01with analysis of synaptic connectivity
03:03in the cortex and hippocampus,
03:07and mouse models.
03:09He leads a multidisciplinary
03:11team to address brain connectivity
03:13changes in Parkinson's disease, which
03:15he will talk about today.
03:18And doctor Chandra,
03:19our second speaker,
03:21received her PhD in chemistry
03:23from Purdue University.
03:25She pursued her interest in
03:27neuronal
03:28cell biology and neurodegeneration
03:30during her postdoctoral
03:31research.
03:33Doctor Chandra became a faculty
03:35member here at Yale,
03:37in two thousand and seven,
03:38where she's currently a professor
03:40in the department of neurology
03:41departments of neurology and neuroscience.
03:44Her research focuses on two
03:46interconnected
03:47themes,
03:48synapse man maintenance and neurodegeneration,
03:51with a specific emphasis on
03:53Parkinson's disease and related disorders.
03:55Doctor Chandra's lab
03:57specifically studies familial Parkinson's disease
03:59genes that encode synaptic proteins.
04:03She also explores the, the
04:04role of the endolysisomal
04:06pathway in the disease and
04:08examine presynaptic
04:09proteostasis
04:10and dysfunction,
04:11which they hypothesize
04:13are central to the early
04:15synaptic events,
04:16of this neurodegenerative
04:18disorders.
04:19Today, doctor Chandra will be
04:20discussing her research, in her
04:22talk named,
04:24elucidating the mechanisms of GBA
04:26linked Parkinson's and dementia with
04:28LVEE bodies.
04:31Now, well, help me to
04:32welcome them both.
04:42So
04:44thank everyone for,
04:46joining our research talks today.
04:48I think it's a splendid
04:49idea to,
04:51make, these efforts to better
04:53connect,
04:54clinical and basic research here
04:56within the department.
04:59As you can as you
05:00just heard,
05:02the, shared,
05:03area for both the talks
05:05today,
05:06is Parkinson's disease.
05:08And,
05:09I would like to start
05:10out by highlighting that,
05:13our research is actually in
05:15a part
05:16that,
05:17of the brain that normally
05:18not commonly studied in Parkinson's
05:20disease, namely in the cortex.
05:22And,
05:24the reason for this, is
05:25that,
05:27the majority of PD patients
05:29will experience cognitive impairments
05:31or even dementia. The trajectory,
05:33of course, depends is different
05:35for each of the patients,
05:36but,
05:37estimates are that up to
05:38eighty percent of patients will
05:40experience these cognitive
05:43challenges.
05:44And,
05:46we wanted to understand,
05:48what the mechanisms the cellular
05:50mechanisms are that underlie these
05:52cognitive impairments because there's really,
05:54at this point, no therapeutic
05:56intervention for these, problems.
06:00What you see here on
06:01the right is actually one
06:02of the leads that we
06:03already had.
06:05What you see here is
06:06a tissue section from cortex
06:08of a PD patients.
06:10And, what you see in
06:12green
06:13is the staining for a
06:15protein,
06:16named synuclein. It's actually, in
06:17this case, it's be a
06:19form of the protein that
06:20is phosphorylated,
06:21which marks the aggregated forms
06:22of synuclein.
06:24And the abundance of these
06:26synuclein aggregates is not only
06:28a hallmark of PDs or
06:29as Lewy bodies or as
06:31Lewy neurites.
06:32Also, the abundance
06:34correlates with the extent of
06:35cognitive decline. So the project
06:37that I will present today
06:38is really based on modeling
06:41these synuclein pathologies.
06:45And, just to recap a
06:47little bit of the literature,
06:48as I mentioned,
06:50dementia or mild cognitive impairments
06:52at least is very common
06:53in PD.
06:55And there already have been
06:57studies, in human patients that
06:59showed that activity patterns in
07:01the cortex are also disrupted.
07:03And, as I just said,
07:05the best,
07:06indicator
07:07of, dementia or progression to
07:09dementia
07:09is then the abundance of
07:11the synuclein aggregates in the
07:12cortex.
07:14And so together,
07:15our hypothesis is
07:17that these,
07:18synuclein pathologies,
07:21cause damage to vulnerable neuron
07:23types and vulnerable synapses and
07:25that underlies or at least
07:26contributes
07:27to the cognitive impairments.
07:31Now with this,
07:34big question of how has
07:35connectivity
07:36changed, we try to simplify
07:38it and we go back
07:39to a synaptic level.
07:42And there's,
07:43ample data that support that,
07:45this is a relevant factor
07:47in in human PD patients.
07:49What you see on the
07:50left is an older study
07:51that used immunostaining
07:52for a synaptic protein,
07:54and that showed that the
07:55extent of synapse loss that
07:57occurs in the cortex in
07:58PD patients is almost as
08:00high as the extent of
08:01synapse loss in patients with
08:02Alzheimer's disease.
08:05Here on the right,
08:07much more advanced studies
08:09pioneered by colleagues here at
08:10Yale.
08:11In this case, this is
08:12a paper by Dave Matuski,
08:14using the PET ligand that
08:17was developed here at Yale
08:18to visualize,
08:20the presynaptic protein s v
08:21two,
08:23which is a wonderful tool
08:25to,
08:26represent,
08:27synapses
08:28in line in humans and
08:30actually do longitudinal studies.
08:32For me, this is still
08:33breathtaking that you can visualize
08:34synapses in a living brain
08:36and analyze synapse density. And
08:38what Dave and his colleagues
08:39found is that one of
08:40the areas that is impacted
08:42in PD patients
08:44as they undergo cognitive impairments
08:47is, the cortex.
08:48And, actually, also, he could
08:50find that, the extent of
08:51synapse loss correlates with cognitive
08:53impairments.
08:55So there's a strong precedent
08:57that from human patients that
08:58is relevant to analyze,
09:00synaptic vulnerability
09:02in Parkinson's disease.
09:05Together with,
09:07several colleagues, we try to
09:09get at the molecular underpinnings.
09:11And what I show you
09:12here are data generated by
09:14Mike Henderson at the Fan
09:15Adel Institute.
09:17He tried to obtain mechanistic
09:20leads by looking at gene
09:21expression patterns,
09:23and he analyzed,
09:24the cortex of PD patients,
09:28typically patients with dementia,
09:30and compared this to a
09:31mouse model. I will introduce
09:33the mouse model in the
09:34next slide. But,
09:36the theme is that there's
09:37very strongly
09:39shared,
09:40gene expression changes in both
09:42the human patients and in
09:44our mouse model.
09:46And one of the,
09:48clusters that was, down regulated
09:51in both,
09:52humans and the mouse model
09:54are synaptic genes. And I
09:56can show you this here.
09:56So this is the area
09:58that is here highlighted.
10:00What you see here,
10:02is named,
10:03genes. These are all genes
10:04that are downregulated when they
10:06are shown in blue.
10:08And as you can see,
10:09it's involves both presynaptic
10:12and postsynaptic proteins.
10:15One protein that I can
10:16already highlight because it will
10:17show up
10:18in, at the end of
10:20my presentation. I hear these
10:21transsynaptic
10:22interactions.
10:24So I will come back
10:24to that later.
10:26But we have a number
10:27of of of leads. And
10:29when we use gene ontology
10:31analysis,
10:32we can pinpoint that these,
10:34changes
10:34affect both the synaptic vesicle
10:37cycle or proteins involved in
10:39the synaptic vesicle cycle and
10:41also proteins involved in postsynaptic
10:43organization. So the,
10:45damage hits on both sides
10:46of the synapse.
10:48Now we try to,
10:51analyze this. And so the
10:52first question then larger, so
10:55the the main part of
10:56my talk,
10:58is how we can,
10:59analyze or what what we
11:00can learn about
11:02synapse loss in this p
11:03d in PD.
11:05And here, I would like
11:07to briefly introduce,
11:09the model that we are
11:10using,
11:11which, as I mentioned, is
11:12based on, synuclein pathology.
11:15And the data that I
11:15show here are generated by
11:17Saroj Sar, who is here
11:18in the audience.
11:20What we,
11:21utilize is a model where
11:23we take
11:24fibrils,
11:25so artificially generated aggregates of
11:28this disease linked protein alpha
11:29synuclein,
11:31and we generate them under
11:32highly controlled conditions. They have
11:33to have a very specific
11:35size to be bioactive,
11:36and we are also very
11:37careful handling them in the
11:39cell culture hood as you
11:40probably can imagine. And,
11:42so we then inject these
11:44into animals.
11:46Specifically, we target the dorsal
11:47lateral striatum.
11:49And the reason that we
11:50target that area is because
11:52here we have
11:53very precise
11:55regional control
11:56because that means that the
11:58cortical neurons in layer five
12:00that projected the striatum, these
12:02are the first ones who
12:03will be exposed,
12:05to the synuclein pathology. And
12:07indeed, we can then show
12:08that these, neurons,
12:11build on themselves, develop aggregates.
12:13So in a sense, it's,
12:15reminiscent of the prior hypothesis
12:17that the proteins synuclein that
12:20we inject in its aggregated
12:21form can template
12:22the endogenous synuclein
12:24and stem cause pathology. And
12:26this is very likely what
12:27happens in the in the
12:29patients,
12:30because, for example, the Braque
12:31staging,
12:33really,
12:34tracks all of these changes
12:35in the progression across synaptically
12:38connected
12:39neural populations.
12:40So we use this approach
12:41to specifically set,
12:43pathology in the cortex of
12:45temporal and spatial control.
12:47And, the way these data
12:49look like in this mouse
12:50model, here's data from the
12:51cortex. And I would ask
12:53you just focus on the
12:54bottom row because here, you
12:56see the synuclein
12:58aggregates
12:59easier because they are marked
13:00in red.
13:02And, when quantifies
13:04the progression of pathology, you
13:05see a significant increase
13:08between
13:09the,
13:10starting, like, one month post
13:12injection to later time point.
13:13So that's a model to
13:14analyze progression,
13:16and we heavily utilize it
13:17for that reason.
13:21Now when we analyze how
13:23this model then replicates,
13:25synapse changes or synapse loss
13:27in the patients,
13:29we can find that,
13:31also here, just like the
13:32synuclein pathology,
13:34the synapse loss is progressive.
13:36I don't show the data
13:37here, but in the first
13:38stages, we do not yet
13:39see synapse loss. But then
13:41as soon as the pathology
13:42has, been around for, let's
13:44say, four weeks, we then
13:46see,
13:48a a gradual increase since,
13:49gradual loss of synapses. Synapses.
13:51The way we quantify this
13:52is, we take three d
13:55reconstructions.
13:56These are images that are
13:58obtained,
13:59using a super resolved imaging
14:01approach.
14:02So these allow us to
14:03separately analyze pre and postsynaptic
14:06sites. I mean, analyze synapses
14:08as structures that has,
14:10appeared in space. So this
14:11is a highly, computationally,
14:14driven analysis of of synaptic
14:17connectivity.
14:18And we see that both
14:19at three months and then
14:20later at six months, so
14:21left and right side,
14:22that synopsis are lost.
14:25Now importantly, it depends on
14:27what markers you're analyzing.
14:28These are the synopsis that
14:30are marked by the protein
14:31big load one. These are
14:33short range local excitatory connections
14:35in the cortex.
14:36However, when we analyze long
14:38range connections, so inputs, for
14:39example, from the thalamus
14:41to the cortex, those inputs
14:43are spared. So synaptic vulnerability
14:46is not across the board.
14:47Certain synapse types are more
14:49vulnerable than others. And that's
14:50actually very interesting for therapeutic
14:52interventions because it means there
14:54are factors that protect the
14:55other synapses. So we just
14:56have to figure out what
14:58makes these long range inputs
15:00stable,
15:01versus, what makes the short
15:03range inputs vulnerable. But that's,
15:05this is one of the
15:06reasons why we analyze,
15:08synapse specific vulnerabilities.
15:12Now
15:13we not only have synapse
15:15loss. So as you could
15:16see, it's, a relatively small
15:18portion of synapses that is,
15:20removed.
15:22We also have broad effects
15:23on the remaining synopsis. And
15:25among those effects are that,
15:27the synopsis, even if they
15:28stay,
15:29show ultra structural aberrations.
15:31These involve include a separation
15:33of pre and postsynaptic sites.
15:35And as you can imagine,
15:36that must,
15:37massively affect, the ability to
15:39release and detect neurotransmitters.
15:42But we also see changes
15:43in the presynaptic organization, and
15:45these are data
15:47obtained in the bottom
15:48by, our collaborator,
15:50Laura Volpicelli, at the University
15:52of Alabama in Birmingham.
15:55And so even though these
15:56are relatively modest changes in
15:58synaptic vascular size, they will
16:01also impact synaptic transmission.
16:05So PD in the cortex
16:07involves both the loss of
16:08synopsis and structural changes of
16:10the remaining synopsis, both of
16:12which will impact,
16:14connectivity.
16:15Now we try to get
16:17a little bit more on
16:17the molecular underpinnings of that.
16:20And here we have,
16:22both in our own lab
16:23and also in collaboration with
16:25the lab of Terry Kumar,
16:27at WashU,
16:28performed a number of different,
16:30analysis.
16:31And it goes back to
16:32the role of synuclein in
16:34this type of pathology. As
16:35I mentioned, we use synuclein
16:37aggregates to trigger this pathology
16:39and to trigger the synapse
16:40loss, but we can also
16:41detect whether synuclein itself has
16:44a, a local role at
16:46these, these, vulnerable synapses.
16:48The way we test this
16:49is, again, like high resolution
16:51imaging, and we track what
16:53happens in the vicinity of
16:55these synuclein aggregates.
16:57What we can find is,
16:58as shown here, that the
17:00synapses that are very close
17:02to these aggregates,
17:03and these are,
17:05shown here in magenta on
17:07the left. So the synapses
17:08that are very close to
17:09these aggregates are the first
17:10ones to be lost. So
17:12even at the earliest stages
17:14when we can't see broad
17:15synapse loss, when you look
17:17very closely at the sites
17:18of these aggregations,
17:20that's where synapse,
17:21number,
17:22goes down. Oh, sorry. I
17:24got the order wrong. What
17:25I show you here is
17:26actually the data that synaptics
17:28synuclein accumulates
17:30at these aggregate sites. And
17:32then here, these are the
17:32data that synapses are lost
17:34close to these aggregates. But
17:36there's a
17:37a correlation between,
17:40synuclein aggregation
17:42and synapse loss.
17:43And,
17:44recently has found that the
17:46vulnerable synapses
17:47actually have ten times more
17:48synuclein than the non than
17:50the protected ones. So synuclein
17:52certainly is a risk factor.
17:54So the accumulation of synuclein
17:55is probably,
17:57a good candidate
17:58as, for causal,
18:00to cause the loss.
18:02And as,
18:03we wanted to test,
18:05however, this,
18:07impacts function,
18:09I could, rely on our,
18:12excellent collaboration that we have
18:13with Mike Higley in the
18:14neuroscience department
18:16where, Mike, in this case,
18:18for example, analyzed slice physiology
18:20in this peer in this,
18:22fibril mouse model.
18:24And, consistent with our, histochemical
18:27data, you could show that,
18:29exhalatory transmission
18:31is strongly reduced
18:32in these animals.
18:34What you see what you
18:35see here is actually plotted
18:36the time it takes between
18:38excitatory transmission events.
18:41So as you see here
18:42in the second graph,
18:45that timing gets longer and
18:46longer.
18:47That means,
18:48fewer and fewer events. So
18:50that's how we quantify
18:51the,
18:52frequency of, synaptic transmission events.
18:56Now
19:02analyzing the vulnerability of synopsis
19:04is one thing,
19:05but we also want to
19:06analyze
19:08how different cell types
19:10are vulnerable to PD pathology.
19:14And,
19:15we here rely on or
19:17can rely on,
19:18information about neuroanatomical
19:20connectivity of the cortex.
19:22As we know, layer five,
19:24where these projection neurons live
19:25that project to the striatum,
19:27is the first layer that
19:28shows, significant pathology.
19:31And, so we also have
19:33information that the neurons that
19:35are in layer five
19:36is, particularly these intertendencephalic,
19:39projection neurons that are shown
19:41here in gray,
19:42that they project
19:44bilaterally. So they project,
19:47if the ipsio contralateral
19:49to the same side of
19:50the striatum, they will inject
19:51it, project to,
19:53the same,
19:55area of the striatum.
19:57But that allows us to
19:59test this in our model,
20:01is what happens if we
20:03use this unilateral injection to
20:05the striatum
20:06and then see how pathology
20:08progresses? Do we see pathology
20:10only in the side of
20:11the cortex for the injected
20:12striatum that sits, or do
20:14we see,
20:15pathology on both sides?
20:18And we actually do see
20:19Sanukian pathology
20:21both ipsy and contralaterally to
20:23the injected striatum.
20:24Here again in layer five.
20:26And so that's consistent
20:28with the layer five projection
20:30neurons,
20:32to be the primary targets
20:33of pathology
20:34in the cortex,
20:36because these are the only
20:37ones that really project dorsolateral,
20:40both ipsi and contralaterally to
20:41the striatum.
20:43And this is accompanied also
20:45in the contralateral side with
20:47synapse loss, just what I
20:48showed you for the ipsilateral
20:49side before.
20:51So it's not just that
20:52synuclein pathology appears, but also
20:53then connectivity is impaired even
20:56though this is a hemisphere
20:57that has been spared by
20:59our injection.
21:01Now
21:03this is all data in
21:04the mouse model.
21:07The question now is how
21:08can we validate this in
21:09the human tissue?
21:11And here, again, our collaborators,
21:14so Terry Kumar and Rousalbek
21:16stepped in
21:17and have pulled off some,
21:19what I find, very impressive
21:21approaches. What you see here
21:22on the left is, human
21:24PD tissue,
21:26analyzed
21:27by synaptic marker staining.
21:29And this is not your
21:30regular confocal. This is high
21:32resolution meso scans
21:34at synaptic resolution. So wherever
21:36you go in this image,
21:38you can zoom in and
21:39have this at a subsynaptic
21:41resolution where you see the
21:42pre and the postsynaptic sides.
21:44So this is incredibly
21:46rich in detail.
21:47So you can look at
21:48the whole,
21:49across all cortical layers
21:52from the meso scan down
21:53to the synaptic level and
21:55analyze synaptic aberrations in the
21:57in the patients.
21:59This is a really
22:01impressive,
22:02advance that they have. So
22:04we're now in the position
22:05to test all of our
22:06approaches
22:08in the patients.
22:10And I wanted to add,
22:12two slides
22:13because,
22:15I
22:17am trained as a biochemist,
22:18so I want to have,
22:20mechanisms.
22:21And, here again, the transcriptional
22:24changes,
22:25helped us to,
22:26identify target pathways.
22:29As I mentioned,
22:30earlier on,
22:31among the genes that are
22:32downregulated
22:34in the, PD in the
22:35pathological
22:36neurons in human PD tissue,
22:38as well as now a
22:39mouse model, are these trans
22:41synaptic proteins that are here
22:43marked with this red circle
22:44on the left.
22:45These are called neurexins.
22:47These are on the presynaptic
22:48side, and they are down
22:49regulated both in the mouse
22:51model and in the human
22:52tissue.
22:54So we asked whether they
22:56have a role, and that's
22:58driven by the fact that
22:59pathology progresses
23:01between synaptically
23:02connected populations. So we want
23:04to know whether synapses themselves
23:06are sites of pathological
23:07transmission.
23:10And for that,
23:11we have established,
23:13both in vitro and in
23:14vivo approach. I'll show you
23:15what today the data for
23:17the in vitro approach.
23:18And these experiments are performed
23:20by Maria Juliano, who's also
23:22here in the audience.
23:24So and I just keep
23:26this, very straightforward.
23:27And in a nutshell, we
23:29could show that,
23:31cultured neurons that we use
23:33in this model
23:34show strong pathological responses to
23:36the addition to the of
23:37this, synuclein fibrils, just what
23:39I have shown you in
23:40vivo.
23:41Let's see on the top,
23:43where you see these,
23:44aggregates marked in green.
23:46What you see in blue
23:47is just a neural morphology,
23:48specifically the dendrites.
23:50And so we quantify the
23:52typical pathology as you see
23:54here on the graph on
23:54the left. But if you
23:56knock down Eurexin, you see
23:57a drastic,
23:59reduction in how the pathology
24:02appears.
24:03And,
24:04there is, at least to
24:05my knowledge, no other intervention
24:07that can reduce the synuclein
24:09pathology in any comparable way.
24:11So, these new reactions are
24:14critical
24:15for
24:16the pathology
24:17appearance in these excitatory neurons.
24:20And,
24:21I just, wanted to,
24:23end with this. We have
24:24done structure function analysis on
24:26this. We do this in
24:27the in the neuroexon knockout
24:29mouse model right now. But,
24:30I think this is one
24:32way to communicate how we
24:33try to get at molecular
24:34approaches.
24:35And just to,
24:37conclude,
24:39what we know is,
24:41synapses are not only vulnerable,
24:44but also show synapse type
24:46specific vulnerability patterns.
24:49We have, not just loss,
24:50but also changes. So the
24:52structure of synapses in the
24:53PD pathology, so the pathology
24:55will have broad effects on
24:57cortical function.
25:00In agreement, we see altered,
25:02synaptic transmission
25:03in this mouse model.
25:05And we have evidence that
25:07the that, certain proteins, including,
25:10these transsynaptic proteins have a
25:12role in pathology progression. So
25:13this could be another way
25:15of slowing down progression
25:17in the patients.
25:18And the things we we
25:19don't know,
25:21really have to understand
25:22how in our mouse model,
25:25cortical circuits are impacted.
25:27We have to move beyond
25:28recordings to really look at
25:29activity patterns in the whole
25:31cortex, and,
25:32that's currently something that we
25:34test.
25:35We also need to know,
25:37to what extent
25:38does,
25:39synaptic,
25:41biology or synaptic,
25:42proteins, to what extent do
25:44they
25:44modulate
25:45or even, drive progressions? Or
25:47are they really targets?
25:50And then, the last question,
25:52and that's a question that
25:53I've already
25:54discussed at
25:55length with Jaime,
25:57It's,
25:58okay. So we have this
25:59model. It replicates one aspect
26:01of PD, which is the
26:03synuclein,
26:03aggregates,
26:04but that's not what happens
26:05in the patients. You often
26:07have copathology,
26:08whether it's tau, or whether
26:09it's amyloid copathology.
26:12Are they just living separately
26:13and doing the individual damage,
26:15or do they have synergistic
26:16effects? And we have already
26:17started the first experiments,
26:19and it, is, quite,
26:21promising that,
26:23amyloid pathology and synutin pathology
26:26are not just additive, but
26:28could actually,
26:29at least that the amyloid
26:30pathology could drive the synuclein
26:32pathology.
26:33So that's a active area
26:34of investigation that I think
26:36is also very important for
26:37the patients.
26:38After that, I would like
26:39to,
26:40acknowledge,
26:41my lab,
26:44who are here. And I
26:45also wanted to acknowledge the
26:47ASAP team. This is a
26:48team picture that we,
26:50took at one of our,
26:52retreats,
26:53which really brings together
26:55a large range of interdisciplinary,
26:57researchers
26:58that, some of us are
27:00new to the PD field,
27:01others are experiencing it. So
27:03it's a a very enjoyable
27:05and,
27:06great environment to learn about
27:08PD and to apply our
27:09our research interests.
27:11That, I would like to
27:12thank you for your attention.
27:22Right. Good.
27:23Kumar. Yeah.
27:25So if you were to
27:26superimpose
27:27the g g b the
27:28GWAS data onto your synaptic
27:30protein, you know, that slide
27:32that you had, is there
27:34a map? Do they match?
27:37It's not that straightforward. So
27:38the GBOS data are less
27:41informative
27:42than the the the expression
27:43data. The expression data have
27:45one advantage. We specifically
27:47and I didn't I I
27:48explained this in detail. We
27:49specifically analyzed gene expression changes
27:52in pathology bearing neurons.
27:54But at some point, it
27:55becomes super streamlined. So we
27:56look we look at that
27:58subpopulation of neurons that have
27:59the synuclein aggregates so we've
28:01had much better resolution.
28:02And then does neurons seem
28:04to bind? I didn't get
28:05that. Did you say it
28:06binds to I didn't. Reform?
28:09I didn't say it, but,
28:10because I didn't want to
28:12cover too much of the
28:12mechanism. But, yes, your your
28:14hunch is absolutely right. That's
28:15what we observe, and,
28:17that's what we are currently
28:18quantifying. But there is evidence,
28:21including also from other labs
28:22that synuclein and fibrils can
28:24bind to the surface of
28:25neurexin. And we have also
28:27with the structure functional identified
28:30modifications in neurexins
28:32that are required for
28:35them. It's a precedent. Right?
28:36So some of the toxins
28:37like botulism and stuff like
28:38that bind to synaptic proteins.
28:41Yeah. Protein.
28:45Okay. Go ahead. David, please.
28:47Do
28:48the advocates of the alpha
28:49SNP induced any microglia activation?
28:53We
28:54have
28:55only,
28:57we are planning these experiments
28:58right now.
29:00So we have the tissue
29:01in our mouse model. We
29:02are starting to analyze this,
29:04and we also want to
29:05analyze the role of microglia
29:06and the loss of synapses
29:08that we have here. But,
29:09that's that's the next step
29:11for us, leaving our neuron
29:13centric view of the world.
29:14And,
29:15and that's,
29:17yeah. We have started to
29:19to to,
29:20these experiments.
29:23Please.
29:24The role of dopamine and
29:26GABA. So
29:28I mean, in clinical practice,
29:30those are the main things,
29:31and dopamine interaction without this
29:33nuclei. And then in terms
29:35of synapse loss, have you
29:37quantitated whether
29:39it is more,
29:40GABA type of
29:42synopsis, or is it a
29:44excitatory synopsis? We have quantified
29:46this. I just didn't go
29:47into the details. We have
29:48analyzed different synapse types. So
29:50if I just ask answer
29:51first about the GABAergic synopsis.
29:53Inhibitory synapses are among those
29:55synapse types that are completely
29:57protected. So even at the
29:58latest time points that we've
29:59analyzed so far, we have
30:00not seen a loss of
30:02GABAergic synapse. So it seems
30:04to be
30:05pretty restricted to local inhibitory
30:07to local acceptory synapse.
30:09Interplay with, dopamine, we have
30:12not yet tested. So we,
30:16at this point, we don't
30:17have evidence,
30:18about, dopaminergic
30:20dopa dopamine release
30:22loss of of dopamine release
30:23sites,
30:24in the cortex,
30:26in our model. But then
30:28we,
30:29we do not
30:31aim to replicate that aspect
30:33of PPE at all.
30:38That's in clinical practice, dopamine
30:40is dopamine. I mean, that's,
30:41like, ninety nine percent of
30:43Oh, yeah. I'm I'm very
30:45used to having to justify
30:46my data.
30:50But,
30:52I personally
30:53I I would, like to
30:55suggest
30:56that these are parallel processes,
30:58that they do not actually
31:00directly impact each other. So
31:02that PD pathology could progress
31:04in different regions using in
31:05in different ways. So
31:08That's not I think that
31:10dopamine,
31:11because it's highly oxidative,
31:14it actually
31:16accentuates the role of albasyanuclide
31:19to map.
31:20And especially in the synapses
31:22with dopamine, that's why substantia
31:24nigra dies first.
31:27We have no evidence for
31:28synapse for neuron loss in
31:30the cortex. So I think
31:31it's really a different mechanism.
31:36If, what you're saying is
31:38true that the binding of
31:39alpha synuclein with the norexin
31:40Yep. Person in vivo,
31:43are you thinking about ways
31:45to,
31:45interfere with that binding? I
31:47mean, would that be a
31:48good feasible
31:50or translatable mechanism? Yeah. So
31:52we
31:53hear what it works. Yeah.
31:54That's really worked.
31:56The mic. They're actually
32:00So we,
32:02we are currently testing this
32:03again, led by Maria. So
32:05we have evidence that
32:07the,
32:11the role of neurexin in
32:12pathology progression involves an extracellular
32:14modification of heparan sulfate proteoglycans.
32:17And we
32:18are collaborating with a colleague,
32:22who has,
32:24glycan arrays, and he will
32:25then map to what type
32:27of glycans
32:28so subtypes of glycans these
32:30synuclein propels can bind. And
32:32that could pros could give
32:34us a specific molecular lead
32:36for what types of heparan
32:38sulfate glycans are targeted by
32:40the synuclein,
32:41and that, can,
32:43that could be then possible
32:45a site of intervention.
32:47For example, we're targeting the
32:48specific enzymes that may need
32:50this particular type of glycan
32:52modification because these are so
32:53complex. There are hundreds of
32:55different options
32:57that should be possible to
32:58target them if they are
32:59specific.
33:01Yes.
33:02Please. I initially had a
33:03dopamine related question, but you
33:05answered that. My second question
33:06would be,
33:07have you look have you
33:09done any behavioral tasks in
33:11the mice
33:12to determine what,
33:14abilities are affected?
33:16We have stayed clear of
33:18behavioral tests because they can
33:21be,
33:23not robust, and so we
33:25wanted to focus on the
33:27physiology
33:28instead.
33:30We are considering now to
33:31do go into behavioral essays,
33:34but,
33:35I think for us, the
33:36physiology comes first as a
33:38function of really.
33:41Alright?
33:54Hello, everybody.
33:55Thank you to Jaime for
33:57sort of spearheading this, and
33:59thank you for Puja to
34:00doing this.
34:02Okay.
34:04So today, I'm gonna tell
34:05you about GBA link Parkinson
34:07and dementia with Lewy bodies.
34:09And a lot of the
34:10themes that Thomas has sort
34:12of already reiterated
34:13will sort of come up
34:15again in my talk, so
34:16I don't have to give
34:17you lots of introductions.
34:19So what the outline of
34:21my talk will be sort
34:22of threefold. Let's see if
34:23this pointer
34:25works.
34:26No.
34:27There's no pointer. Sorry.
34:29So I'm gonna introduce to
34:30you what GB a is.
34:32It's links to Gaucher disease,
34:34which are very well established,
34:35and the more recent links
34:37to Parkinson and dementia with
34:39Lewy bodies.
34:40And then tell you because
34:41I'm a mouse and iPSC
34:43lab, I'm gonna tell you
34:44about our mouse studies on
34:45how to model GBL link
34:47Parkinson and dementia,
34:49with Lewy bodies in using
34:51mice, and then also what
34:52we've gained as insights
34:54by doing these mice studies.
34:56Okay.
34:57So just to begin with,
34:59GBA is a lysosomal
35:01enzyme called glucocerabrosidase.
35:03It's abbreviated commonly as GKS,
35:05so you may have heard
35:06that. And it's an enzyme
35:08that sits in the lysosome
35:10and is part of the
35:11sphingolipid,
35:12salvage pathway.
35:14And it's a glucosidase. So
35:15it'll take complex
35:16ceramides, glucosal ceramides,
35:19and hydrolyze the glucose of
35:21it to release ceramide.
35:23And so what happens when
35:24you have loss of function
35:25mutations in GBA,
35:27you get a disease called
35:28Gaucher disease, which all of
35:29you are more familiar than
35:31I am.
35:32So you this is a
35:35primarily a childhood disease of
35:37liver and spleen and bone.
35:39And in in this Gaucher
35:40disease, you have two things
35:42that are happening.
35:43One is that the complex
35:45glucosophagolipids
35:47accumulate,
35:48primarily peripherally in macrophages,
35:51and this is how you
35:52stain for them,
35:53as Gaucher cells.
35:55And also you have a
35:56ceramide deficiency because there's not
35:58ceramide not being released. And
36:00ceramide,
36:01as you all know, is
36:03necessary for barrier function of
36:05skin. So they also have
36:06skin issues.
36:07And Gaucher's in the US
36:09is largely treated. You there's
36:11very good enzyme replacement
36:13therapy
36:14options. And there are also
36:16options
36:17to what are called substrate
36:19reduction therapies. That meaning that
36:21you try and reduce the
36:23amount of glucosylceramides
36:25that are actually synthesized.
36:27And both these strategies are
36:28very efficacious.
36:30And so patients can manage
36:32Gaucher disease
36:33all the while
36:34all the way through adulthood.
36:37So GBA,
36:39the reason GBA and Gaucher
36:41is very well established, and
36:43GBA mutations due to founder
36:45effects are very common in
36:47the population. So for instance,
36:49if you're asked a Nazi
36:50Jewish or for instance Norwegian,
36:53they're huge founder effects. And
36:54so these mutations are very,
36:56very common. And especially two
36:58mutations I'm gonna talk to
37:00you about is just called
37:01the n three seventy s
37:02mutation
37:03and the l triple four
37:04p mutation.
37:06So even though the link
37:08between Gaucher and GBA was
37:10very well established, the link
37:12between,
37:14GBA and Parkinson and Gaucher
37:16is very recent. It's actually
37:18in two thousand nine it
37:19actually came only at a
37:20clinical practice
37:21even though we have GWAS
37:23and,
37:24GWAS and, you know, familial
37:26genetics.
37:27So this was a study
37:29that came from Ellen Sidransky's
37:30lab at the NIH,
37:32and she led a multicenter
37:35study which showed that GBA
37:37mutations
37:38are actually causing Parkinson.
37:40And in fact, they showed
37:41that,
37:42Gaucher patients, which are homozygous
37:44GBA,
37:46patients, have homozygous mutations,
37:49are at a twenty four
37:50risk for developing Parkinson. And
37:52the parents of these patients,
37:54who are heterozygous
37:55carriers, are at a five
37:57fold risk for developing Parkinson.
37:59And the GBA link part
38:01of Parkinson is actually associated
38:03with synuclein pathology,
38:05the Lewy body pathology. That's
38:07the defining feature of sarisporadic
38:09disease.
38:11So sort of flash forward,
38:13this is the genetic landscape
38:14of Parkinson today. So in
38:16the top cluster are all
38:18the, familial genes. We have
38:20twenty three PARC genes, PARC
38:22one through twenty three.
38:23And then in the middle
38:25are the medium to high
38:27risk genes, which are fairly
38:28common in the population, and
38:30I've circled GBA,
38:31and the other being LERP
38:33two, which you also heard
38:34of. And in the very
38:35this sort of corner here
38:38are common variants that are
38:39all in the sporadic population.
38:42So you can see that
38:43GBA is, turns out to
38:46be one of the most
38:47common genetic risk factor for
38:49Parkinson.
38:50In fact, five percent of
38:51all GBA,
38:53all Parkinson
38:54patients have GBA mutations.
38:56And because of this link
38:58of this founder effects, there
38:59are certain ethnicities
39:01that are actually very at
39:02high risk for developing GBA
39:04link pockets.
39:06So since this two thousand
39:08nine,
39:09landmark paper,
39:11there have been a lot
39:12of follow-up clinical studies to
39:14understand
39:15whether sporadic Parkinson,
39:17which is the eighty five
39:18percent of population,
39:20and
39:21peep, patients that have GBM
39:23mutations, is there anything clinically
39:25different between them? And what
39:27is sort of a clear
39:29consensus
39:29is that if you have
39:31GBM mutations that is driving
39:33your disease, you have a
39:34faster decline and you have
39:36much more cognitive
39:37deficits. So there is it
39:39it is actually predisposing
39:41you to having,
39:43dementia.
39:45So when in,
39:47in two thousand twenty three,
39:49the
39:50GWAS studies for Lewy body
39:52dementia, the related
39:54synucleonopathy
39:55was actually published. And you
39:57can see this is a
39:58Manhattan bot. And you can
40:00see that GBA is one
40:01of the genes that is
40:03actually driving,
40:05dementia with Lewy bodies. The
40:06other being SNCA, which is
40:08the gene for synuclein.
40:09And this being a dementia
40:12sort of across between Parkinson
40:14and Alzheimer's, you can see
40:15APOE and BIN1 are also,
40:18genes that are linked to
40:19Alzheimer are also,
40:21genes linked to,
40:23dementia with Lewy bodies.
40:25So if you just look
40:26at the familial forms of
40:28Lewy body dementia, you can
40:30see that GBA is the
40:31only gene linked to this,
40:34disorder.
40:35So
40:36because
40:37the GBA is linked to
40:38two synucleopathies,
40:40you know, Parkinson and Lewy
40:42body dementia,
40:43you know, there's a lot
40:44of interest farmers you know,
40:46in pharma to sort of
40:47develop GBA therapies. And because
40:49of all the
40:51therapeutic
40:52advances already for Gaucher, they
40:54can actually use those leads
40:56to actually treat this. I
40:58should say at this point,
40:59the reason why Gaucher patients,
41:02even though they are on
41:03drugs and can manage their
41:05Gaucher symptoms,
41:07still develop Parkinson or Lew
41:09body dementia is because those
41:10drugs don't cross the blood
41:12brain barrier.
41:13So then right now, there's
41:14a huge impetus
41:16to try and convert, you
41:17know, convert the Gaucher drugs
41:20to cross the blood brain
41:21barrier because you know that
41:22they're actually efficacious
41:24because they can treat Gaucher.
41:25So I I mean, I
41:26won't,
41:27belabor this, but there are
41:29very interesting,
41:30therapeutic strategies in the pipeline.
41:33And so far, they all
41:35look very promising.
41:36One is AAV based therapies,
41:38which they're actually inject injecting
41:40intrathecally,
41:41and the other is using
41:43small molecules because a lot
41:45of the GBM mutations
41:47are actually,
41:49trafficking mutants. So they don't
41:50go to the lysosome properly.
41:52And so if you can
41:53make it go to the
41:55right place within the neuron,
41:56then you can actually treat
41:58the cell. Okay.
41:59And the last thing that
42:00I wanna sort of plug
42:01is ambroxol,
42:03which is a, over the
42:04counter,
42:06drug,
42:07cough medicine that's available worldwide,
42:10but that is now actually
42:11in phase two trial,
42:13for GBL link park ins.
42:15It's not available here, but
42:16it's available
42:17elsewhere in the world.
42:19Okay. Okay. So okay. So
42:21the question really is
42:23why do GBA mutations
42:24predispose
42:25you to both, PD and
42:28DLB?
42:29So this,
42:30there has been sort of
42:31a very synuclein centric model
42:34because it causes Lewy bodies
42:35in this. And one of
42:37the ideas that was actually,
42:39put forward by Dimitri Cronk,
42:41who's the head of neurology
42:42in Northwestern,
42:44is that,
42:45a, that the lipids that
42:47accumulate can actually template synuclein
42:50aggregation, which is we we
42:51could also show this. And
42:53because the lysosome, which is
42:55where, you know, proteins are
42:57degraded is not functioning,
42:59sort of this two hit
43:00model that, you know, alpha
43:02synuclein is no longer degraded
43:04plus aggregating causing Lewy bodies.
43:07So this was in two
43:08thousand eleven, but subsequently,
43:10it's we have on as
43:12a field have understood that
43:14it's much more complex. It's
43:16just not about synuclein
43:17alone.
43:18And therefore, it's now realized
43:20that there because,
43:22the proteins are misfolded, there's
43:24ER stress.
43:25And what I will come
43:26back to later is that
43:27there are a lot of
43:28lipid imbalances
43:30with because this is an
43:31enzyme that's involved in lipid
43:33homeostasis.
43:34Yeah.
43:36Okay. So at this point,
43:37I wanna just sort of
43:38make a plug for our
43:40story that came out last
43:41week.
43:42So we we are using
43:44mouse models to understand
43:46GBLINK Parkinson,
43:48and we have sort of,
43:51used
43:52the, what we learned from
43:53human genetics that the l
43:55triple four p mutation in
43:57particular
43:58is actually predisposed
43:59to cognitive
44:00deficits. So we've made mice,
44:03that have this mutation on
44:05a null background to model
44:07basically Gaucher patients.
44:09And because,
44:11you need ceramide for barrier
44:13function of skin, if you
44:14just make these mutations, the
44:16mice get skin phenotypes and
44:17they die. So we've rescued
44:20the Gaucher, GBA expression in
44:22skin. Okay.
44:23But what we can show
44:24is these are really good
44:26models. They can, in the
44:27previously we before this, before
44:30we made these mice, most
44:31of the mice would only
44:32live three weeks, so we
44:34could not because of the
44:35skin issues.
44:36So we could not study,
44:37like, age related phenotypes.
44:40So in this case, these
44:41these mice, you know, accumulate
44:44the
44:45the lipids.
44:46They're like you would expect,
44:47like patients do. They have
44:49very little protein because they
44:51have the mutations,
44:52and they have no enzymatic
44:54activity because the because it's
44:56a mutant protein. So they're
44:57really a very good model.
44:59And what we've done is
45:00taken these mice and crossed
45:02them to synuclein transgenic. So
45:04we can compare,
45:06the black mice, which is
45:07the wild type mice, to
45:09the GBA, the amber.
45:11And then we can cross,
45:12compare the same
45:14impact of this GBA mutation
45:15on a wild type background
45:17or a synuclein background. So
45:19red versus green.
45:21So what we've done in
45:23using these mice is sort
45:24of behavioral assays to sort
45:26of understand how,
45:29GBA impacts motor functions
45:31as well as cognitive functions.
45:33And so if you can
45:34just look at the colors,
45:36if you look at the
45:37amber colors versus
45:39the black colors, these are
45:40longitudinal
45:41motor behavior tests. So at
45:43the, in panel a is
45:45balanced beam. So you make
45:46the mice cross from one
45:48end of the very narrow
45:50beam to the other and
45:51quantitate how many runs they
45:53can do in a given
45:54time.
45:55And here on the panel
45:56b is,
45:58the grip strength. How long
45:59a mouse can hold on
46:00to a grip strength meter.
46:02And you can see in
46:03this as this is the
46:05same cohort of mice analyzed
46:07over one year. And you
46:09can see in wild type
46:10that they are fine in
46:12black. In amber,
46:13you can see the GBA
46:14mice are also fine. They
46:16have no motor deficits. But
46:18the synuclein transgenic, which are
46:19in red, can show progressive
46:22phenotypes.
46:22But if you cross in
46:24GBA,
46:24which is in green, they
46:25get much worse.
46:27Yeah? So it says that
46:29GBM mice on their own
46:30don't get any motor phenotypes,
46:32but on a on a
46:33synuclein background, they can make
46:35things worse.
46:36We've done this using variety
46:38of other motor behaviors. In
46:39this case, hyaline clasp and
46:41open field. And the same
46:43story holds. As as you
46:46on their own, they're fine.
46:47But in a on a
46:48synthetic background, they make things
46:50worse.
46:51So what about cognition?
46:53Because, you know, GBA is
46:54linked in particular to cognitive
46:57deficits, we were very interested
46:58in doing this. And so
47:00the way we've done this
47:01is using fear conditioning.
47:03So if you put a
47:04mouse in a in a
47:06box and then pair a
47:07tone with the shock, then
47:09they remember that this was
47:11a very aversive
47:12environment. So the next day,
47:13if you bring them and
47:15just give them the tone,
47:16no shock, then they will
47:18freeze, you know. So we
47:19can calculate how long they
47:21freeze as a measure of
47:23how much they remember this
47:24fear fear memory. And you
47:26can see on the panels
47:27on the on the, e
47:29that if you look at
47:31the amber bars between the
47:33training day and the testing
47:34day, both at three and
47:35twelve months, that they are
47:37showing no significant differences
47:40in so they don't remember
47:42this fear, fear condition.
47:44So because there were some
47:46differences in already on the
47:48training day, we did another
47:50essay called novel object recognition.
47:52This is what what you
47:54do is you put a
47:54mouse with,
47:56two objects
47:57on one day, and you
47:59calculate how much time they
48:00spend with each
48:02object. And then the next
48:03day, you swap out one
48:05of the objects, and you
48:06put a new object, a
48:07novel object. And then normally,
48:09mice are curious, so they
48:10will go to the novel
48:12object if they remembered,
48:13these were the one was
48:14an old object. And you
48:16can see in this very
48:17clearly
48:18that GBM mice, even at
48:20three months, have a cognitive
48:21deficits.
48:23But on a synuclein
48:24transgenic
48:25background, they still have the
48:26same,
48:27effect. So it says that
48:29very clearly that we can
48:31dissociate
48:32both motor and cognitive,
48:35phenotypes with genotype.
48:37So a wild type are
48:38normal. GBA only have cognitive
48:41deficits.
48:42Synuclein transgenic
48:43only motor deficits. And when
48:45you have a double cross,
48:46you have
48:47cognitive deficits that are driven
48:49by GBA
48:50and then worsen motor deficits.
48:53So what about how is
48:55this related to synuclein pathology?
48:57Because in Parkinson's field or
48:59Lewy body dementia, you can't
49:00get away from, synuclein pathology.
49:03So what we did is
49:04if you look at the
49:05column, the third column, just
49:07the green column,
49:08we stain the cortex and
49:10many other areas, but I'm
49:12just gonna show you the
49:13cortex,
49:14for phosphocinuclein,
49:16which is phosphorylated
49:18at a particular site at
49:19position one twenty nine. And
49:21this is the, is used,
49:23as Thomas said, to sort
49:25of distinguish normal synuclein
49:27from aggregated synuclein. And this
49:29is also used in the
49:31path labs to score for
49:32Lewy bodies.
49:34And you can see very
49:35clearly if you just go
49:36vertically down that
49:38wild type and GBMI,
49:40even at twelve months of
49:41age in which these pictures
49:42are taken,
49:43do not show any synuclein
49:45pathology. The synuclein transgenic does,
49:45but when you have a
49:45double transgenic, you get much,
49:47transgenic
49:48does,
49:49but when you have a
49:50double transgenic, you get much
49:51more pathology.
49:53And this is sort of
49:53quantified in these graphs
49:56here. And what is interesting
49:57is it's not the number
49:59of cells that are positive,
50:01but the cells that are
50:02actually positive become much more
50:04intense. And this is shown
50:05in sort of these histograms
50:07on this side.
50:09So the take home message
50:10is that GBM mice don't
50:12show, phosphocinogrelin
50:14pathology.
50:16And we've done this other
50:17ways. We've done this by
50:19western blotting, and we get
50:20a similar result. You can
50:22see that on their own
50:23wild type and GBM mice
50:25don't have any pathology, but
50:27synuclein transgenics do, and then
50:30many of the double transgenic,
50:31you have more pathology.
50:34So this tells us something
50:36rather surprising for us that
50:38the cognitive symptoms are actually
50:41occurring independent
50:42in this mouse of alpha
50:43synuclein pathology.
50:45So this is sort of
50:46counterintuitive
50:47to to what Thomas told
50:48you, like,
50:50a a few slides ago.
50:52But the motor symptoms
50:55are very tightly linked to
50:57alpha synuclein
50:58pathology
50:59load. So the more the
51:01pathology,
51:01more the motor symptoms. This
51:03is just for GBLINK Parkinson.
51:05This is not for
51:07linked
51:08models.
51:08Okay.
51:10Okay.
51:10Okay. So while we were
51:12doing this, we had an
51:14an a competing lab, Gomiteng's
51:16lab at Columbia,
51:18published this study,
51:20and they used l triple
51:21four p but heterozygous
51:23mice.
51:24And they also analyzed,
51:26cognitive,
51:27impairment in this mice.
51:29I won't belabor this, but
51:31they could show that
51:33they could in their mice
51:34also that they have issues
51:36in the panel a with
51:38fear conditioning.
51:39And they also did,
51:41Morris water mace if you
51:42whether you could figure out
51:44where the hidden platform was.
51:46And they could show that
51:47in the GBA mice that
51:49they do have a deficits
51:50at seven months a day.
51:52But most importantly, what they
51:54did is they crossed their
51:56mice
51:57to synuclein knockout. So they
51:59where in which synuclein has
52:00been deleted.
52:01And they still get these
52:03deficits.
52:03So saying that you do
52:05not need,
52:06in this mouse model any
52:09synuclein to drive the cognitive
52:11deficits. The GBA alone, by
52:14its function in the lysosome,
52:15can cause these deficits.
52:18Okay. So we are interested
52:19to try and figure out
52:20what the mechanisms
52:21are. And so to get
52:23some insights into this, we
52:25did single cell RNA sequencing
52:27that you're all familiar. This
52:28is just to get a
52:29transcriptional,
52:31definition,
52:32if it were, of the
52:33brain as a as at
52:35a single cell resolution in
52:37these four mice.
52:39So this is what what
52:40you you compress all this
52:42complex data into two dimensions
52:44called UMAPs,
52:45and this is what the
52:46data look like. And if
52:48we look in and sort
52:49of dive in only into
52:51neurons comparing wild type versus
52:53GBA and these transgenics.
52:55And if you look in
52:56the bottom here,
52:58if you look at what's
52:58happening in neuronal types,
53:00I was expecting because GBA
53:03is a lysosomal
53:04enzyme,
53:05it controls lipid homeostasis,
53:07that those would be the
53:08pathways that are most prominent.
53:11But,
53:12surprised when you do unbiased
53:13things, you get surprises.
53:15And the hits you get
53:17are all related to synaptic
53:18vesicle trafficking,
53:20synapse structure,
53:22and so forth. So we
53:23are now sort of,
53:26and if you can do
53:27more analysis, then this is
53:29exactly what you get. You
53:30get more and more synaptic
53:31pathways that are
53:33dysregulated
53:34when you actually have a
53:35lysosomal
53:36enzyme that is actually missing.
53:39So we've done some preliminary
53:41experiments to confirm that this
53:43indeed impacts synapses by doing
53:45electron microscopy.
53:47Like,
53:48Thomas showed you, we can
53:49show there are fewer synaptic
53:51connections in the cortex of
53:52GBMIs.
53:54We can also look at
53:55very detailed outer structure as
53:58a proxy for synaptic vesicle
54:00cycling
54:01to look at how many
54:02vesicles there are in each
54:04terminal and how many clathrin
54:06coated vesicles. These are vesicles
54:07that are sort of functionally
54:09turning over. And we can
54:11see very profound changes in
54:13these, GBA mutant synapses.
54:15But we're still left with
54:17a lot of questions. And
54:18this is what we're currently
54:20trying to tackle.
54:21The problem for us is
54:23that
54:25lysosomes and synapse synaptic vesicles
54:28are two organelles that have
54:30distinct identities in the neuron.
54:32So the lysosome is mainly
54:34in the cell body, and
54:36synaptic vesicles are at the
54:38terminal.
54:38And we don't understand how
54:40you can impact
54:41if you remove an enzyme
54:43that's sitting in the soma,
54:45how you're impacting this,
54:47you know, distinct organelle.
54:49And one idea is that
54:51it's controlling lipids. It's controlling
54:54key lipids
54:55that are necessary for synaptic
54:57function. And we have some
54:59beginning to get some answers
55:00into this. And if you
55:02look at the we've looked
55:03for some lipids called PIP2,
55:05the phosphoinositides,
55:06the essential for synaptic cycling.
55:09And you can see that
55:10in the GBM mice, there's
55:11less,
55:12PIP2. So now we're sort
55:14of on a quest to
55:15do lipidomics,
55:16sort of organella lipidomics to
55:18understand
55:19the basis for this, sort
55:21of fine.
55:22Okay. I how am I
55:23doing? Okay. So I'm just
55:25gonna wrap up and tell
55:26you that, you know, if
55:27there's one take home message
55:29that you can say that
55:30GBA can,
55:32cognitive deficits in GBA,
55:35occur independent of alpha synuclein
55:37pathology and that these are
55:39related to actually
55:41synaptic signatures.
55:43And
55:43that there are actually treatments
55:45that are in the pipeline,
55:46and I'm actually very hopeful
55:48for this sort of five,
55:50ten percent of the population
55:52that there will be actually
55:54effective treatments for that. And
55:56that we should actually look
55:58not just from the from
55:59the perspective
56:00of PD or motor symptoms,
56:02but also from this, you
56:03know, perspective of cognitive symptoms.
56:06Okay. With that, I'd like
56:07to thank the people who
56:09did this. All the people
56:10who did this have gone
56:11to brighter and greener pastures.
56:14So Vidya Dara, who's now
56:15an assistant professor in Rosalind
56:17Franklin University,
56:18David
56:19Backstrom, who's now back in
56:21Sweden in Umea University,
56:23Risha, who started as a
56:24MD PhD.
56:26I could she only wanted
56:27to be an MD, and
56:28I convinced her to be
56:29an MD PhD.
56:31Okay.
56:33And who's at Columbia, and
56:34then Jevin, who's a who's
56:36a PhD student. So and
56:37funding sources. Thank you.
56:44Yeah. Yeah. So
56:47thank you for a really
56:48nice talk. Thank you. Yeah.
56:50For those of us who
56:51study Alzheimer's disease and model
56:52that in mice, there is
56:54I mean, there is just
56:54so many thematic similarities in
56:56where you guys are and
56:57where we are. I mean,
56:59we have the APP,
57:00rare genetic variant for mutation
57:02that causes AD
57:04early onset, and then APOE,
57:05which is the strongest genetic
57:07risk factor. And now you
57:08have SNCA or synuclein alpha,
57:10which is your rare cause,
57:12then you've got GBA as
57:13your most common, most impactful.
57:15And in the APP APOE
57:17story, there's some biochemical
57:20functional relationship between those two
57:22proteins.
57:23Do you think you're gonna
57:24find some similar similarities across,
57:27you know, how it impacts
57:28ATP?
57:29Do you think GBA may
57:31you know, synuclein similar ways?
57:33So in the motor aspects
57:35of this, this is entirely
57:36true because,
57:38GBA makes synuclein
57:40aggregates
57:41much worse because the lipids
57:43themselves can template the the
57:45seeding of synuclein.
57:47And because in GBA mutations,
57:50lysosomes
57:51are not functional, the pH
57:53is off. So in that
57:55sense,
57:56yes. So there is very
57:58interesting data that even in
57:59sporadic disease,
58:01GBA levels are down even
58:03though there's no mutation in
58:05those patients,
58:06and that glucosal sphingosine, which
58:09is the biomarker of GBA
58:10deficiency, is high.
58:12So there is
58:14talk of trying these therapies
58:16also on sporadic.
58:19So and I think I
58:20would support that. I think
58:21the data to date would
58:23suggest that it would or
58:25maybe, Clemens, you think so?
58:27Is that yeah. Right? Yeah.
58:28That it would be to
58:29treat increase GK's activity even
58:32in sporadic would be would
58:34be my thought. Yeah.
58:37Congratulations
58:38to the wonderful dog and
58:40the awesome new paper. That's
58:41really great.
58:43I'd be curious if you
58:44could hypothesize
58:46how on a molecular,
58:48you know, precise level do
58:50you think,
58:52reduced g, GK's function
58:55impacts endocytosis?
58:57Yeah. So, I mean, I
58:59don't have a good answer.
59:00I mean, I think, currently,
59:02we're gonna see whether,
59:04if we're gonna purify vesicles
59:06from these mice
59:07and look at
59:09sort of unbiased lipidomics
59:11to see if that's the
59:12case.
59:13If that's we don't get
59:15clean answers there, then we
59:16have to really think. We
59:18we can get some insights
59:20from the transcriptional
59:21signatures. But right now, I
59:23don't have a clear answer.
59:25I wish I had a
59:26clear answer. I don't know
59:27that. Yeah. Yeah. Sorry.
59:29I don't The APOE I
59:31mean, it seems like APOE
59:32is also quite strong.
59:34Yes. Yes. Yes. You realize
59:36how strong about it.
59:38But you you think it's
59:38also lipid
59:40mediated
59:40or because in Alzheimer's, it
59:42is still unclear what APOE
59:44is it affecting amyloid plaques?
59:46Is it affecting
59:47inflammation
59:48or directly on the synapses?
59:51So it's very confusing.
59:52I I You think there's
59:54any more clarity there? No.
59:55It's I don't think we
59:56have many because, you know,
59:58it's a two thing. One,
60:00you know, there is
01:00:01increased certain lipids, but there's
01:00:03a deficiency of other lipids
01:00:05and the lysosome down function.
01:00:07It's a it's not a
01:00:08and also I didn't talk
01:00:10about this, but GBA
01:00:11in
01:00:13immune cells does all kinds
01:00:14of things. So it's,
01:00:17I'm sort of looking very
01:00:18neurocentric, but I don't think
01:00:20that is actually the truth.
01:00:22Yes, sir. Yeah. Okay. You
01:00:24you just mentioned it, but
01:00:25the
01:00:26like, if the main cell
01:00:27type is the main cell
01:00:28type in Gaucher disease that's
01:00:30affected to some macrophages Yep.
01:00:32So, like,
01:00:34what's the root of micro,
01:00:37microglia
01:00:37in the brain and, like,
01:00:40Yeah. We're we're, we are
01:00:43just beginning to look at
01:00:44that. There are other you
01:00:45know, Prime Mysteries Lab has
01:00:47looked at that, and they
01:00:48find that,
01:00:50microglia
01:00:51inflamed.
01:00:52There's
01:00:54there's you know, it's a
01:00:55complex one. David David wouldn't
01:00:57have access to those criteria.
01:00:59Does she behave a nutrition?
01:01:00Yes. Yes. Very much so.
01:01:02And that's why this you
01:01:03know, that's how they stain
01:01:04for them. They do they
01:01:06are called Gaucher cell. They're
01:01:07staining for
01:01:09macrophages that have full of
01:01:10lipid in them. So it's
01:01:12such a prominent phenylalanine. That
01:01:14they can use it. And
01:01:16also they have, you know,
01:01:17liver enlargement as Yes.
01:01:21One common thing among all
01:01:22of these is the autophagy.
01:01:24Yes. That's correct.
01:01:26Where you have
01:01:27proteins and the lipids and
01:01:29everything, and that's that's overwhelmed,
01:01:31then the
01:01:32secondary thing would be that
01:01:34the people tend to
01:01:35this is this is correct.
01:01:37There's a lot of autophagy
01:01:38finis
01:01:40phenotypes or in other, like,
01:01:41stem cell derived neuronal models.
01:01:44And you could manipulate those
01:01:46to to get some, at
01:01:48least in a in in
01:01:49vitro system, some benefit.
01:01:52Whether that will work
01:01:54in vivo, I don't know.
01:01:56No. It's not it's it's
01:01:58I should say that there
01:02:00is some data
01:02:01that all lysosomal
01:02:03storage diseases
01:02:04actually increase the risk for
01:02:06Parkinson's.
01:02:07It just negotiated a very
01:02:09prevalent
01:02:10lysosomal
01:02:11storage disease. And in other
01:02:13diseases,
01:02:14it's also much more fatal
01:02:16early on that they don't
01:02:17have. Yes. So yes.
01:02:20Yeah. No. Congratulation.
01:02:22But it's still being a
01:02:24career for GPA is a
01:02:25is a predisposition.
01:02:27Yeah. And it's, do you
01:02:29have any insight about genetic
01:02:31modifiers
01:02:32either in human or brain?
01:02:33Yeah. So the I didn't
01:02:35clarify this, but that what
01:02:36she brings up is a
01:02:37very important question. Is that,
01:02:41not everybody
01:02:42who has GBM mutations will
01:02:45develop Parkinson.
01:02:47There's very strong genetic and
01:02:49even within the Gaucher, forget
01:02:51Parkinson, there's a huge phenotypic
01:02:54variability that you could be
01:02:55homozygous
01:02:56for l triple four p,
01:02:57but not actually only come
01:02:59to the clinic for Parkinson,
01:03:01you know, that you made
01:03:02it all through childhood just
01:03:04fine. In partly because the
01:03:06sphingolipid
01:03:08metabolism pathway is highly hairy
01:03:10and complex
01:03:11and that there are many
01:03:12modifiers.
01:03:13So there have been a
01:03:14huge quest to find modifiers,
01:03:17because they would make the
01:03:19perfect therapeutic
01:03:20target because in humans, they're
01:03:22actually
01:03:23modifying the disease.
01:03:25And
01:03:26we are also looking for
01:03:27those.
01:03:28We have a couple.
01:03:30We're not ready for a
01:03:31short time. But, you know,
01:03:32Dimitri Crunk's lab has done
01:03:34a lot on that front,
01:03:35and they found recently a
01:03:37command Khan's labs.
01:03:39Controls GK's activity in duets.
01:03:43And I think that
01:03:44therein will be
01:03:46new leads to
01:03:52This
01:03:53is exclusively
01:03:54seen in non neuropathic
01:03:57forms.
01:03:58That's it. Oh, yeah. Yeah.
01:03:59You're right. Okay. I didn't
01:04:00go into this either.
01:04:02But then yes. Because the
01:04:04neuropathic
01:04:05form, there are three forms
01:04:06of negotiate.
01:04:07Like, type one, type two,
01:04:09and type three, which is
01:04:10the neuropathic
01:04:11form. And after the four
01:04:13imputations
01:04:14are associated
01:04:15with neuropathic function. Yes. Because
01:04:17the neuropathic forms are by
01:04:19large,
01:04:20early
01:04:21infantile child and much more
01:04:23severe.
01:04:24So they don't live long
01:04:26enough. Yeah.
01:04:33Okay. Thank you.