Pathology Grand Rounds: October 5, 2023 - Jerry Chipuk, PhD

Name: Pathology Grand Rounds: October 5, 2023 - Jerry Chipuk, PhD
Uploaded: 2023-10-05T20:05:51.7066667Z
Duration: 1 h 1 min 40 s
Description: The Pathology Grand Rounds talk from Oct. 5, 2023, featuring Jerry Chipuk, PhD, Professor of Oncological Sciences and Dermatology at the Icahn School of Medicine at Mount Sinai. His topic was, Mitochondria Control of Melanomagenesis.

October 05, 2023

The Pathology Grand Rounds talk from Oct. 5, 2023, featuring Jerry Chipuk, PhD, Professor of Oncological Sciences and Dermatology at the Icahn School of Medicine at Mount Sinai. His topic was, Mitochondria Control of Melanomagenesis.

Information

ID: 10822
To Cite: DCA Citation Guide

Download Transcript

00:00It's really an honor and pleasure to
00:03introduce to speak at Pathology Grand Rounds.
00:06Doctor Jerry Chippic, Professor of
00:08Oncological Sciences and Dermatology,
00:10Associate Director of Basic Science
00:12Shared Research Sources Director
00:15of Cell Biology at the I Can
00:17School of Medicine, Mount Sinai.
00:19Doctor Chippic completed his PhD at Case
00:22Western Reserve with David Daniel Poor
00:24working on cell death and TGF Beta Singling.
00:27He then completed postdoctoral
00:29work with Doug Greene,
00:30where he taught Doug most of
00:33what Doug knows about cell death.
00:35While there, he made seminal contributions to
00:38our understanding of P53 induced cell death.
00:41Puma's mechanism of action involvement of
00:43single lipids with backs back activation
00:46then started his own lab at Mount Sinai,
00:48where he continued to work on
00:50how mitochondrial morphology and
00:51function influence cell death,
00:53rass transformation,
00:54and map kinase activity,
00:56particularly within the context of Melanoma.
00:59It's also expanded our technical
01:01repertoire with personal favorites
01:03like sparkle and flambe assays,
01:07particularly inspiring those the way
01:08that Jerry serves as a mentor and
01:10champion of diversity within science.
01:12He sits on multiple DEI initiatives,
01:15mentored over 40 scientists directly
01:16within the lab ranging from high
01:19school students all the way up,
01:20including multiple first generation
01:22degree holders that he's mentored
01:24from technicians into Graduate
01:26School and studies. His
01:28mentorship and collegiality
01:29extends beyond Mount Sinai,
01:31and I've been very fortunate to get
01:33to know him at multiple meetings
01:34over the past 15 years and see
01:36him really is this insightful,
01:38humorous and caring colleague.
01:40I've always appreciated very
01:42much learning from him,
01:44both about the way he uses precision
01:46within his science and also the
01:49artistry of both his experimental
01:50design and his presentation,
01:52which I'm sure we'll all be able to
01:54enjoy and appreciate today as he
01:56speaks to us on the mitochondrial
01:59control of Melano Melanoma Genesis.
02:01Got it out.
02:03Please.
02:08All right. Well, First off, I'd like to
02:10extend my warmest appreciation to you, Sam.
02:13It's always lovely to come and visit friends
02:16and explore some of the science that we do.
02:19And I also want to say thank you to
02:22everyone in the audience for joining
02:24me this afternoon and everyone who
02:25joined me for dinner last night.
02:27I had a really lovely time.
02:29It's an honor to speak in in front of such
02:32an esteemed faculty and in member in front
02:34of the members of the Melanoma spore.
02:36So I'm excited to share our biology with you.
02:40My laboratory investigates how the
02:42mitochondrial network impacts upon general
02:45biology and we often position our studies
02:48in the context of cancer mechanisms
02:51to explore cellular transformation,
02:53chemotherapeutic success and prognosis.
02:56So today, in the next 45 to 50 minutes,
02:59I hope to provide you with a broad
03:02perspective of one of my group's
03:04research interests that centers on how
03:06the mitochondrial network influences
03:09melanocyte biology and how this
03:11provides some unique perspectives
03:12into the process of Melanoma genesis
03:19here. OK, good.
03:23So this research program originated by
03:26observing mitochondrial shape changes
03:28in how these shape changes reveal
03:31insights into the cellular state.
03:33But if you Google mitochondrial shape,
03:36you'll get hundreds of
03:37images of mitochondria.
03:38That suggests mitochondrial shape
03:40is boring and uninteresting.
03:42However, I think it's time that
03:44we kind of abandon this notion,
03:45and because if you observe
03:47mitochondria in living cells,
03:49you will see just the opposite,
03:51phenotype, and that is mitochondria
03:53are dynamic organelles which
03:55present themselves as a multitude
03:57of sizes and a multitude of shapes.
04:00And in this video you're looking at a
04:03primary mouse embryonic fibroblast.
04:05Its mitochondria are labeled with YFP
04:08in the matrix and about 5 minutes of
04:10mitochondrial movement is condensed
04:12into approximately 3 seconds.
04:14And in this this is a beauty.
04:16In this video,
04:17it's a really beautiful example
04:19of looking at both intra and
04:22interorganellar communication,
04:23where metabolites,
04:25proteins,
04:26lipids and DNA are shared to
04:29preserve mitochondrial network
04:31homogeneity and efficiency.
04:33And starting just over a decade ago,
04:35I became interested in understanding
04:38if fundamental mitochondrial biology
04:40related to the dynamic nature of
04:42this network influence cancer
04:44mechanisms and offered any clinically
04:47relevant insights into cancer.
04:49And as such,
04:50we choose to explore mitochondrial
04:52function in the context of MAP kinase
04:55signaling because its mutations
04:57capture a majority of cellular tumors
05:00and it demonstrates significant
05:02therapeutic opportunities with
05:03the advent of targeted therapies.
05:06In this pathway,
05:07signaling is normally initiated at
05:09the level of the surface of the
05:11plasma membrane by via a series
05:13of receptor tyrosine kinases that
05:15respond to extracellular ligands like e.g.
05:17F and the signal is transmitted to
05:20the nucleus via a series of GGP aces
05:22such as Ras or kinases such as B
05:25ref that enable transcriptional programs
05:27that support survival and proliferation.
05:30Interestingly,
05:30pathway mutations also
05:32occur in multiple nodes,
05:34for example,
05:35the receptor or Ras or B ref that
05:38allow for constitutive signaling
05:40through this pathway in the
05:43absence of receptor ligation.
05:45This pathway has also received a lot of
05:47attention from drug companies of course,
05:48as they have developed these
05:50numerous specific and promising
05:52small molecules that target
05:54distinct nodes within the pathway.
05:56Now for this talk,
05:57I'm going to focus primarily on
05:59mitochondrial phenotypes that
06:00result from oncogenic MAP kinase
06:03signaling in the context of Melanoma.
06:05And a little bit later on I'm going
06:07to bring up a few of these inhibitors
06:09just throughout the experimental design.
06:10So just keep them in mind.
06:14So First off,
06:15Melanoma is a curious disease.
06:17Curious is always my word.
06:19I use it to describe everything.
06:20It can be good and bad.
06:23It accounts for, you know,
06:24only about 4% of skin cancers,
06:26but about 80% of skin cancer deaths
06:29incidence is increasing even in nations
06:32where Melanoma has been historically lower.
06:35I'm going to talk about maybe
06:36some of the reasons for this a
06:38little bit later in the talk.
06:40Secondly, I want to introduce the cell
06:41line or the cell of origin which is
06:43related to disease and it's the melanocyte.
06:45You know,
06:46cell type is really important
06:48when we consider mitochondrial
06:50function as melanocytes are derived
06:52from the neural tube and often
06:54straddle the interface between
06:56epithelial and neuronal phenotypes
06:58from the organelle perspective.
07:00And these cells are just normally
07:02kind of dotted along the skin.
07:05They synthesize melanin,
07:06and they transfer it in the form of
07:08these tiny parcels from the epidermal
07:10basal layer to keratinocytes,
07:12where melanin then is responsible
07:14for dissipating UV energy to
07:16prevent macromolecular damage.
07:18Here I'm showing a very simplified version,
07:21our process of melanomagenesis,
07:23which emphasizes a series of
07:25histopathological features associated
07:27with primary and invasive disease.
07:30But what happens between normal
07:32skin and primary Melanoma in C2?
07:35And, you know,
07:36the answer to this is often
07:38associated with benign skin lesions
07:40known as nevi or common moles,
07:42which are kind of minimally described as
07:44these clusters of senescent melanocytes.
07:47And if you look at your arms or legs,
07:48you find a mole.
07:49And more likely than that,
07:50there's a mutation there for
07:52B Rocky 600 E that drives the
07:54proliferation of those melanocytes.
07:58So in the laboratory, we can model
08:00Melanoma genesis by isolating primary
08:03melanocytes from healthy patients.
08:06And then we infect them with an
08:08UNCLE gene like B ref E600E or
08:10an UNCLE gene and an RNAI against
08:12a tumor suppressor such as P10.
08:15And this combination of treatments
08:17allows us to compare the biology of
08:20primary melanocytes to senescent cells,
08:23and how overriding senescence may
08:25eventually lead to a transform phenotype.
08:28And when you observe mitochondrial
08:30networks in these representative cells,
08:32you can easily determine that there are
08:35abundant changes as we transition from
08:38normal to senescent to transform phenotypes,
08:41which together represents multiple
08:43research programs within my group.
08:46And looking at these mitochondrial
08:49networks a little bit in more detail,
08:51now here you see this normal milanocyte,
08:54it has a beautiful normal
08:57mitochondrial architecture.
08:58And then when you introduce the raffin
09:00to them and you undergo the process
09:02of oncogene induced senescence,
09:03of course you see the cell getting
09:05larger because it's senescence.
09:06Everybody calls it the flat egg
09:08phenotype or the fried egg phenotype.
09:10But what you also notice is this marked
09:13expansion of the mitochondrial network.
09:15And here we're standing for HSP 60,
09:18which is a mitochondrial matrix marker.
09:20And you can see that there's also
09:22a lot greater intensity of H SP60.
09:24So we have more mitochondria in these cells.
09:28But why generate all these mitochondria
09:31if these cells are destined to
09:34never divide again?
09:35Why build 3 powerhouses in a
09:37community of homes that's never
09:39going to increase in population?
09:41And do these changes impact on metabolism
09:44signaling or the fate of these cells?
09:48And using this as a platform,
09:49we're going to investigate some of
09:51the roles of these mitochondrial
09:53changes in these processes.
09:55So to explore these questions,
09:57I already mentioned earlier that we
09:59model the early stages of Melanoma
10:01genesis and here you're just looking at
10:04some bright filled images of primary
10:06melanocytes and how they change
10:08morphology during Aqua gene induced in
10:10essence that's initiated by B raft signaling.
10:12And we see this really beautiful.
10:14You can see melanocytes are not the
10:16most attractive cells when in just the
10:19control conditions they're spindly.
10:20You look at them on the microscope,
10:21they're not gorgeous,
10:22but you induce opportunity to
10:24senescence and now a majority of
10:26them become really beautiful,
10:27amazing architectures and phenotypes.
10:29We can see that of course these cells
10:32express the B rat P600E Aqua gene.
10:35The majority of them have senescence
10:37associated beta galactocytase activity.
10:39We can detect P21P16 expression in
10:42these cells and they also secrete
10:44all of the classical markers of the
10:47senescence associated secretory phenotype.
10:49And here we're just measuring the
10:50RN A's from them,
10:54data I'm not showing you.
10:55Of course, we've also looked at Saps
10:57and we've also looked at a few other
10:59morphological features to confirm
11:00that this process is what we expect.
11:02So now let's investigate a little bit
11:05about how BR FP600 signaling actually
11:08alters mitochondrial biology to connect
11:10some of these questions I asked earlier.
11:13So we first measure in these
11:15situations E CAR and O CAR.
11:17This is looking at the extracellular
11:19acidification rate or the oxygen
11:21consumption rates in these cells.
11:23And we do this by using the
11:25Agilent Seahorse technology,
11:26where we do everything in real time.
11:28And measuring E CAR and O CAR
11:30allows us to look at glycolysis and
11:34mitochondrial respiration respectively.
11:35And here what you can see is when you
11:38introduce B ref into these cells,
11:40they're normally quite glycolytic.
11:41And when you introduce B ref into them,
11:44their glycolysis completely collapses
11:46and they have a coordinated increase
11:49in their mitochondrial function.
11:51They have a coordinated increase
11:53in their basal respiration,
11:55their maximal respiration and
11:57their spare respiratory capacity.
12:00And you can see this is a kind of
12:02amazing re metabolic programming
12:03of these cells when you put the
12:05B ref on the gene into them.
12:09So digging a bit deeper into the
12:12bioenergetics of these cells to explain
12:15the enhanced mitochondrial respiration,
12:18we look to see who's actually
12:20responsible for providing electrons
12:22to the electron transport chain.
12:23If we remember NADH supplies to complex 1F,
12:27ADH supplies to complex 2 and that's
12:30how you stimulate respiration to
12:31drive proton pumping to create what we
12:34call mitochondrial delta psi or the
12:37bioenergetic differences between the
12:39intermembrane and the rest of the cell.
12:42And if you look at this here,
12:43you'll see that both Complex
12:45One is perfectly fine,
12:47Complex 2 is perfectly fine
12:48in the control cells.
12:50But in both situations when you
12:52introduce the rapid 600 E oncogen,
12:54both Complex one and Complex 2
12:57markedly increase in their activity.
13:00And so the way that we study
13:04this is normally driven by State
13:053 respiration measurements,
13:07which basically just means if you remember
13:09the concepts of respiratory control,
13:11whenever the level of ADP is markedly
13:14higher compared to the levels of a TP,
13:17you can stimulate respiration
13:18within these cells.
13:19And this is what this is basically
13:21looking at State 3 driven respiration
13:24to either Complex one or Complex 2.
13:26So B,
13:28roughly 600 E markedly renovates
13:32Melanicites mitochondrial network
13:33to become more productive.
13:36And we can see here are some just some
13:39nice examples of this basal respiration
13:42changes By about eightfold the spare
13:45capacity increases by about fourfold
13:48and it tells us that wild B ref.
13:52E 600 E is building more
13:54powerhouses within the cell.
13:56It's not in a manner that necessarily
13:58allows them to be efficient,
14:00because if you measure the consequences
14:03of these mitochondrial expansions,
14:04you'll see something that's very
14:07curious and that it's easy to
14:09determine that cellular fitness was
14:11not considered when these melanocytes
14:13were generating mitochondria when
14:15they encounter B RAF signaling.
14:17And here we're just looking at
14:18a few assays to quantify this.
14:20In A and B, we're looking at mitosox.
14:22This is kind of a very standard
14:24assay looking for reactive oxygen
14:26species generation and it basically
14:29represents when electrons aren't
14:31properly moving between iron,
14:33sulfur cluster groups within
14:34individual complexes within
14:36the electron transport chain,
14:37or when these electrons are not
14:40flowing efficiently between the
14:41complexes that drive proton pumping.
14:46We also can see that mitochondrial Ross
14:49leads to protein oxidation versus the
14:52generation or due to the generation
14:54of reactive oxygen species and radical
14:57species such as superoxide and hydroxyls
15:00and also these non radical species such
15:03as hydrogen peroxide and cichlid oxygens.
15:06All of these target prolines, arginines,
15:09lysines, threonine residues within your
15:12proteins and compromised protein function,
15:14assembly and quality.
15:16And then finally,
15:17we also measured melon D aldehyde,
15:20which is the final step of
15:23polyunsaturated fatty acid peroxidation.
15:25And you can see it's also increased
15:27in the presence of B raft.
15:29So what this basically tells us is
15:31increasing B raft V600 signaling
15:33stimulates Mart mitochondrial Ross
15:36production protein oxidation and
15:38also oxidation of the of the membrane
15:40compartments within the cell.
15:42And what's important to note in
15:44this situation is we're not looking
15:46at very small changes,
15:47we're actually looking at
15:48quite significant changes.
15:49If you notice here,
15:51the control in these cells are always FCCP.
15:53And this is a,
15:55you know,
15:55it's the classical uncoupler
15:57of mitochondrial biology.
15:58It basically is in the cytosol of cells,
16:01it takes protons and drags them
16:03inside the mitochondrial matrix.
16:04And when that happens,
16:06the mitochondrial respond by increasing
16:09their ability to pump or to move
16:11electrons and to pump protons to
16:13compensate for that change in Delta psi.
16:16FCCP is the classical maximal ability
16:19for mitochondria to do something
16:21which you can see with Ross protein
16:24oxidation and lipid peroxidation.
16:26B ref.
16:27E600E signaling almost equals the
16:30same level as FCCP treatment,
16:32so it's a significant amount of damage
16:33that's happening to these cells.
16:37So after identifying these phenotypes,
16:39we see marked mitochondrial expansion,
16:42we see enhanced respiration, we see
16:46consequential macromolecular oxidation.
16:48We next questioned if the mitochondrial
16:51network itself started to signal for help
16:54to assist with their own quality control.
16:57And this is basically driven
16:59here because you can see again,
17:01and I showed you these earlier,
17:03these mitochondria are getting much larger.
17:05And what drives this expansion
17:06of the mitochondrial network?
17:07Well, normally it's proteins like PGC 1A,
17:10which is a key protein involved in
17:13the transcription of metabolism genes
17:16and also mitochondrial bioenergetics
17:18and mitochondrial Biogenesis.
17:20But we also see TPM,
17:23which is one of the major transcription
17:25factors of the mitochondrial network that's
17:27responsible for driving mitochondrial RNA
17:29polymerase to the mitochondrial genome.
17:30And we also look at another marker,
17:32Tom 20, which is just a classical
17:35marker of mitochondrial mass.
17:37So all of these markers of
17:39mitochondrial function are going up.
17:40We see there's increased mitochondrial
17:43function and marked inefficiency.
17:44So it wasn't a surprise to us when
17:47we started to screen for stress
17:49within the network that these
17:51signaling pathways were turned on.
17:53And in particular we looked at a TF5ATF4
17:56and D TI-3 which is also called Chopped.
18:00And these are the three transcription factors
18:03that drive the canonical mitochondrial
18:05unfolded protein response in cells.
18:08And these transcription
18:10factors are activated.
18:13Here we're looking at the level of RNA,
18:14but they're also each one of them have
18:17a cohort of transcriptional targets that
18:19they induce to then restore the quality
18:22of the mitochondrial network back to normal.
18:24And all of these classical transcription
18:26targets of the pathway are also activated.
18:29And you can see that indeed.
18:31So let's take a little bit deeper into the
18:35mitochondrial unfolded protein response.
18:37You know the the MiTo UPR activates
18:40following a host of stressors
18:42commonly in the literature.
18:43It can be metabolic distress,
18:45it can be overexpressing mutant proteins
18:47in the matrix or the inner membrane
18:50space of mitochondria that are just,
18:52you know,
18:53overexpressed in recombinant mutants.
18:56But it can also be induced by infecting
18:58cells and a lot of the literature
19:00looks at infections in C elegans.
19:03Or if you infect helis cells
19:05with pseudomonas,
19:06you can activate this pathway and
19:09what it does is basically exists as
19:12a beautiful example of mitonuclear
19:14communication.
19:15Those are the three transcription
19:16factors that are responsible for
19:18the ones I mentioned earlier,
19:19and they induce A transcriptional
19:21and epigenetic program to restore
19:23the quality of these mitochondria.
19:25And the way they do this is by
19:27increasing protein folding capacity,
19:28Ross protection and protease
19:31activation within the network.
19:33And it's just a really beautiful
19:36example of how mitochondria and
19:38the nucleus can actually interact.
19:41And the reason why is the miter.
19:44UPR depends primarily on a TF5,
19:48and this is a basic lucine zipper
19:51transcription factor containing an
19:52n-terminal mitochondrial targeting
19:54sequence and A/C terminal nuclear
19:57localization signal.
19:58And upon normal conditions ATF5IS
20:01constitutively expressed.
20:03And because the mitochondrial
20:04targeting sequence is there,
20:05it's always read by the
20:07mitochondrial network,
20:08imported and degraded.
20:09And when there's an aberration
20:11in mitochondrial function,
20:12because delta psi starts to decrease
20:15because efficiency is changing,
20:16proton pumping is changing,
20:18the nuclear localization signal starts
20:20to get detected more frequently
20:22within the cytosol and you shift from
20:25mitochondrial import to nuclear import
20:26to then drive a transcriptional program.
20:29So it's a nice way where one protein can
20:32communicate the quality of mitochondria.
20:35So in a broader context,
20:37ATF5 normally functions with TOP and a TF4,
20:42and it's important to note this because
20:45almost all the literature suggests
20:47that they always cooperate together
20:48and one of the things we're going
20:50to see here is a unique dependency
20:52almost exclusively on a TF5 biology.
20:58I also need to suggest that or to
21:00say there's a tremendous literature
21:03on a TF5 already in cancer,
21:05but the biology of a TF5I
21:07think needs to be reevaluated.
21:09Because once we recognize it
21:11as the major transcription
21:13factor for the MiTo UPR pathway,
21:15I think we have to kind of look at the
21:17literature and say what makes sense,
21:19what doesn't make sense from the
21:21mitochondrial perspective and reintegrate
21:23this literature and reexamine some
21:25of the phenotypes that are generated
21:27when we regulate ATF5 in both normal
21:31cells and also transformed states.
21:34So this is just showing a quick example of
21:37what ATF 5 looks like if you stay in for it.
21:39Some of the antibodies are terrible.
21:41And there's also another issue and
21:43that is you'll see if you put your
21:45FP600E when it's spelled wrong.
21:47And how do you figure that
21:48out when you're at Yale?
21:49That I can't spell.
21:50All right, this is what I learned at Yale.
21:53You'll see little speckles of ATF5
21:54and the image there is terrible.
21:56I guess everybody always says
21:57that it looks beautiful here.
21:58ATF 5 accumulates in the nucleus
22:00when you add an Oncogen.
22:02And unfortunately,
22:02the antibodies always pick up
22:05the degraded peptides as well.
22:07So the cytosis kind of flooded
22:08with these A TF5 peptides that
22:10are degraded by the mitochondrial
22:11network and it also picks up that.
22:13So there's always the cytosol,
22:15that contribution that you see,
22:17but all of the cells that
22:19you infect with B ref,
22:20We've also done work with Ras, H Ras,
22:22N Ras all activate the same pathway.
22:24But for this talk I'm just
22:26going to talk about B ref.
22:28So what connects these pathways?
22:30How do you go from B rough to a TF5?
22:33And what makes a TF5 unique and
22:35interesting from this perspective?
22:37Well the A TF5 dependent minor UPR
22:40literature doesn't implicate any post
22:42translational modifications within this
22:44pathway to allow for organelle repair.
22:47But there are descriptions of P300
22:49or CBP which is a transcriptional
22:51coactivator that acts in part through the
22:54accetalation of his stones and non his
22:57stone substrates that satellites ATF5.
22:59And there are roles in the serum responses,
23:03there are roles in infection,
23:05there are roles in adepogenesis
23:07and longevity and C elegans.
23:10And these observations let us
23:12to investigate if onpogenic map
23:14kinase signaling also could change
23:16a TF5 settlation patterns.
23:17So what we did was we took linocytes,
23:20infected them with B RP600E purified
23:24ATF5 and you can see that we see a
23:26nice acetylation bands on a TF5.
23:28And if you do this in the presence of
23:30Matt kinase inhibitors or P300 inhibitors,
23:32you lose the acetylation.
23:33We then take that purified protein
23:36and then subject it to mass spec to
23:38identify where the acetylation takes
23:40place and it takes place on lysine 29.
23:42And what's curious is the acetylation
23:46is not there for the Matt kinase
23:48inhibitor or P300 inhibitor and if
23:49you also stress the mitochondrial
23:51network, it's not there.
23:52This is the FCCP result, and also,
23:54if you activate the mitochondria,
23:56unfolded protein response from
23:58within the mitochondria by
23:59inhibiting one of the chaperones.
24:01That's the small molecule GTPP.
24:03It also is not there suggesting that
24:06this is a specific modification to
24:08this pathway that's dependent upon
24:11B ref and not necessarily conserved
24:13throughout all MiTo UPR responses.
24:16We can make mutants of this
24:18a TF5 wild type K29 Q,
24:20which is the acetylation mimic,
24:21or K29R which is an acetylation
24:24null mutant of a TF5.
24:26You can put it into melanocytes,
24:28you can stimulate them with
24:30a MiTo UPR inducer,
24:31and all of them will stabilize as normal.
24:35And what's curious is looking at the
24:37impact of a TF5 on actually senescence.
24:40So here we're looking at melanocytes,
24:43we infect them with B ref with 600
24:45E looking at 3 days and 21 days.
24:47You can see you get a nice time
24:49dependent increase in beta galactosidase
24:51and senescence in these cells.
24:54If you silence ATF5,
24:56all the cells almost immediately sines.
24:59In the presence of B ref, they're fine.
25:02In the absence of of B ref, they will.
25:04They will persist.
25:06We introduced a combination within
25:07three days the majority of the
25:09population is already senescent,
25:11suggesting that the MiTo UPR
25:13pathway actually puts the brakes
25:15on the senescence pathway.
25:17And then if you reconstitute with wild type,
25:19you see a similar situation
25:21to the wild type state.
25:22And what's curious is if you
25:24put in the acetalation mimic,
25:26this is the form of a TF5 that looks
25:29constitutively acetalated by B ref.
25:31It completely blocks B ref.
25:33Induced on the induced senescence
25:36and if you prevent acetylation,
25:38all the cells again become
25:40senescent immediately suggesting
25:42that this pathway really uniquely
25:44integrates BRF signaling with Uncle
25:46Jeannie Duce senescence via ATF5.
25:49And these are just some of the
25:51different transcriptional programs
25:52that are regulated by these mutants
25:54looking at the minor UPR perspective.
25:55And these are some of the outcomes
25:57of these cells, just some images.
26:00So I mentioned earlier that
26:02there's a unite a unique MiTo UPR
26:06comparing B ref A600E signaling
26:07to this small molecule that the
26:09majority of the literature uses
26:11to activate this pathway GTPP.
26:13And here you can see B ref A600
26:15deactivates the SASS P The normal MiTo
26:17UPR pathway or at least the small
26:20molecule induced pathway doesn't.
26:22There's differential regulation
26:23of the minor UPR targets to
26:25correct mitochondrial dysfunction.
26:27This differential regulation of
26:29mitochondrial included genes When you
26:31do this and just one way to summarize
26:33this as well is just looking at T fam.
26:36T fam is not induced.
26:37This is the transcription factor
26:39again that's responsible for
26:41creating the mRNA transcripts
26:43from the mitochondrial genome.
26:44Also not induced from the canonical
26:47MiTo UPR always specific to B ref.
26:49And then finally we can study
26:51some of these metabolic changes.
26:53You may remember I said when you
26:55put B ref into a Melano site,
26:56they go from becoming glycolytic
26:58to mitochondrial and that is
27:00conserved in these data here.
27:02But when you treat with GTPP
27:03they do just the opposite.
27:05They collapse their mitochondrial
27:07function and they increase their
27:09glycolytic activity just again showing
27:11that this is a very unique UPR.
27:13And perhaps due to where the
27:15nature of the stress originates,
27:18small molecule stress within the
27:20mitochondria creates one flavor of
27:22the UPR and then a more physiological
27:25potentially you know potentially
27:26physiological regulator of MiTo UPR
27:28coming from the outside into the
27:31mitochondria changes the way that the
27:33cells respond to this pathway Okay.
27:35So what happens if we eliminate
27:38ATF5 in throughout the system?
27:41I won't go through all of the data.
27:43We've also done this in the a TF5
27:45deficient mouse and if you have no
27:47oncogene in the a TF5 deficient mouse,
27:49the skin is perfectly fine.
27:51There might be a slight
27:54inflammatory component in the skin,
27:55but it's pretty minimal.
27:56You get normal Melania type function.
27:58No issues at all here.
28:01If we take melanocytes and
28:02then we introduce B RAF,
28:03we silence ATF5 or we do the combination,
28:06you get some interesting responses.
28:08Of course we can silence ATF5ATF4 and chop.
28:11Remember I mentioned earlier that
28:13it's usual that they work together?
28:15There's always a marked compensation
28:16when one or the other is eliminated
28:19and you can see that a TF5IS induced
28:21a bit when we silence ATF5 or ATF4
28:24is induced when you silence ATF5.
28:26Chop doesn't move too much and importantly
28:29B REF signaling doesn't change.
28:31We needed to show that when you
28:33silence the mitochondrial unfolded
28:34protein response pathway that you
28:36don't now lose a BRB ref or six energy
28:39signaling and we don't lose it at the
28:40level of the expression and we also
28:42don't lose the downstream targets.
28:45What's curious is you do lose
28:47mitochondrial expansion,
28:48suggesting that when you put B
28:50ref into a cell and you get all
28:52those beautiful mitochondria.
28:53If you don't have a TF5 that doesn't happen.
28:57The reason why is you lose PGC
28:591A and these mitochondria also
29:01don't engage an increase in their
29:04delta PSI and TMRE staining is
29:06a nice surrogate to look at.
29:07Oxygen consumption or proton pumping
29:10within this network and B REF
29:12V600E will increase TMRE staining.
29:15And if you do this in the presence
29:16of a TF5 you basically just
29:18flatline and there's no change.
29:22Same thing happens with some of
29:23these data sets that I showed
29:25you earlier before where you have
29:27this marked shift in glycolysis.
29:29Loss in glycolysis and a marked
29:32increase in mitochondrial function
29:33doesn't happen in the absence of a TFIB
29:36REF cannot remodel the mitochondrial
29:39network to erode glycolysis or
29:41increase mitochondrial function if
29:43the MiTo UPR pathway is not active.
29:47And also looking at the state 3 driven
29:50respirations, complex one changes,
29:52complex 2 changes,
29:53all of those are also minimized
29:56in this presence of no ATFI,
29:58suggesting that in order for
30:00B ref to expand mitochondria,
30:03remodel them,
30:04increase complex one and complex
30:072 activity erode glycolysis,
30:09it must activate this organelle
30:11quality control program.
30:14And if we look at some of the
30:18reasons as to why we see this marked
30:20increase in mitochondrial function,
30:23what you'll notice is that B RAF will
30:25induce expression of or replication of
30:28the mitochondrial genome most of the time.
30:30You know, people who study the nucleus,
30:32there's like 2 copies of most genes.
30:34People who study the mitochondrial genome,
30:36you can have anywhere between 0 copies
30:38of the genes because your red blood cells
30:40don't have mitochondria or certain muscles,
30:42and neurons have thousands of copies of
30:45mitochondrial genome and then melanocytes.
30:47They kind of are in the middle.
30:49And you see the V ref can induce
30:51the levels of mitochondrial DNA.
30:53And this doesn't happen in the
30:55absence of a TF5, the T fan,
30:57the PGC 1A results are held consistent
30:59and if you also look at the expression
31:01of all of these different genes that
31:04arise from the mitochondrial genome,
31:06they're all lost when a TF5 isn't there.
31:10So we picked these genes to represent.
31:12It's not comprehensive.
31:13You know there are 13 proteins,
31:16there are rivals on the RN A's and
31:17all of the TRN A's are encoded by
31:19the mitochondrial genome and this
31:21is just representations of those
31:22so that it looks nice.
31:26And then finally, what we can see
31:30is all of these negative outcomes
31:33that occur because of B ref induced
31:36mitochondrial remodeling also don't
31:38occur if we don't have this orbinal
31:41quality control program activated.
31:43You can see that the MiTo
31:44Sox intensity goes away.
31:45This is looking at the comparing
31:47the Gray and the green lines.
31:49The same thing is there for the
31:52protein oxidation and the same thing
31:54is there for for the lipid oxidation.
31:56And if you look at the activation of
31:59the MiTo UPR pathway in this situation,
32:02you can also see that the combination
32:05completely ablates the activation
32:07of this pathway.
32:08So it's a really interesting
32:10phenomenon that you think your FP6
32:12under these signaling remodels,
32:13mitochondrial function,
32:14you get all of these mitochondrial,
32:16all these activities are engaged.
32:19The mitochondrial network tries
32:20to keep itself happy,
32:21tries to repair and restore
32:23and replenish the network,
32:25but it actually has a deleterious
32:26effect to the cell,
32:28suggesting that all quality control
32:31programs are not always necessarily
32:33to the benefit of the cell.
32:35So let's move away a little bit
32:40from human epidermal melanocytes.
32:41All of that work I just showed
32:43you was in primary melanocytes.
32:44And, you know, they're expensive,
32:47they're difficult to culture,
32:48they're slow growing.
32:49If you ever have to do mitochondrial
32:51metabolism research with melanocytes,
32:52you have to wait two months
32:54to get all the dishes grown,
32:55and then they're always finessed.
32:56In the absence of the Aqua gene,
32:58You know,
32:59we can move to some different models.
33:01And these are a series of Melanoma
33:04sun lines that are derived
33:07from primary Melanoma lesions.
33:10They're all be wrapped in 600D positive.
33:12And what you'll notice is all
33:15of them have statistically
33:16significant high levels of a TF5,
33:19both at the level of mRNA
33:21and also with protein.
33:23And if you look at the other markers,
33:25a TF4 and chop,
33:26they're kind of all over the place.
33:27Some of them have high levels,
33:29some of them have no change.
33:31And so this started to suggest
33:33to us that once you get through
33:37the escape from opportunities,
33:38in essence you have this addiction to the
33:42mitochondrial UPR to maintain survival.
33:45So what we did was we looked at all
33:49these cell lines we could silence
33:51ATF5ATF4 and chopping these cell lines.
33:52And what you'll notice is in contrast
33:56to the Melania site populations that
33:58are primary here this the levels
34:01of these cooperating transcription
34:03factors are quite consistent.
34:05So in almost all of the literature,
34:07when you silence ATF5ATF4 and chop skyrocket,
34:11if you silence chop a TF5 and
34:14ATF4 skyrocket here,
34:15you silence one the other stay
34:18pretty pretty consistent.
34:19And that's the same for all of these.
34:21And this slowly led us to believe
34:23that there was a unique dependency
34:25on the ATF5 pathway,
34:27and that these transcription factors,
34:29while throughout the classical literature,
34:31usually cooperate,
34:32there can be instances where
34:34they don't have to.
34:36And
34:39I promise, this is the
34:40worst slide I've ever made.
34:41I hope you agree with me.
34:42All right, it's busy, it's dense,
34:45and every time I present,
34:46I think to myself, I need to rebuild it.
34:47And then I just can't be bothered.
34:49I don't know why. So anyway,
34:51this is just one of those cell lines.
34:54I had a master's student who
34:55worked with me for two years.
34:57He had the seven cell lines that
34:59are derived from primary Melanoma
35:01lesions and I made him go through
35:03all of these assays 7 times over.
35:05OK and all the cell lines.
35:06And he categorized all of them
35:07and cataloged all of them.
35:08This is just one slide that
35:10shows one of the cell lines.
35:11This is the UP model.
35:13It's another V Rep P600E driven system.
35:15Of course they have basal ATF5 levels.
35:19They don't have that much chop or a TF4.
35:22And if you silence ATF5,
35:24chop and a TF4 don't change very much.
35:26What's interesting is if you
35:28silence ATF5 in these cells,
35:31P21P16 levels increase very
35:33quickly and you get this really
35:35beautiful expansion of the cell and
35:38of cellular senescent phenotype.
35:39You're going to see the
35:41mitochondria look different here,
35:42there's a different architecture
35:42to these than I showed you earlier.
35:44In the primary melanosite you get those
35:47beautiful long connected mitochondria
35:48and here it's slightly different and
35:50we're going to talk about in the
35:52second-half of this why that's the case.
35:55You can also see that they
35:57have an opposite effect.
35:58You sign on to a TF5 and now
35:59you get mitochondrial expansion,
36:01you get T FM,
36:02you get PGC 1A,
36:03you get mitochondrial enhancement
36:05of function.
36:06And you also see this increase in Okar.
36:09So these cells are consuming more oxygen
36:11through the mitochondrial network.
36:15It's kind of a fun control.
36:16What we did was we also stimulate
36:18these cells with GTPP which is
36:20again is that small molecule.
36:21Take a cancer cell that we think is
36:23addicted to a TF5 signaling and now
36:25hyperactivate the pathway and what happens?
36:27The exact opposite is mitochondria
36:29collapse and function so it gives you
36:32this concept of minor hormesis where this
36:35tonic mitochondrial function is necessary.
36:38Tonic mitochondrial stress signaling
36:40is necessary to maintain survival
36:42and if you hyperactivate the pathway
36:45now the cell crashes.
36:47So we've done this in seven cell lines
36:50and we've also started to move into some
36:51of the metastatic cell lines as well.
36:53Everyone knows STML series,
36:55A370, Fives, all of these.
36:57Once you enter the metastatic lesion and
37:00you're no longer at the primary site.
37:02This requirement really goes by,
37:08this is starting to look at some of
37:10the ATF5 expression levels in patient
37:14samples comparing here normal skin,
37:17dysplastic Nevis and intradermal nevi.
37:19We're still in the process of going
37:22through all of the T123 and four
37:25samples from the Melanoma patients.
37:27They are also statistically increased,
37:30but I'm not showing them here because
37:31we don't have the study done yet.
37:33But what you can see is as you
37:34start to enter the progression,
37:36you start to see this enhancement
37:38of the levels of a TF5 that
37:41are detected within the lesion.
37:43And what we did then was we cooperate.
37:46We collaborated with A
37:50and I'm a dermatologist at NYU,
37:54Julie Chile Solavy and we
37:57sequenced 51 Melanoma patients,
38:00all of which these are primary lesions.
38:02And then using the MiTo UPR signatures,
38:05we look to see how they clustered.
38:07And what you can see is this,
38:08this nice clustering.
38:10Looking at chopped a TF4,
38:11all of the targets are in blue and
38:14green respectively and then all of
38:15the targets of a TF5 are in red.
38:17We don't necessarily have to look at a
38:19TF5M RNA changes because remember this
38:21protein is constitutively expressed
38:23and goes between the organelles,
38:24so it doesn't have to be increased
38:26in terms of levels.
38:28But what you can see really nicely is this,
38:30this clustering arrangement.
38:31And what's curious is all the patients
38:34who are in green are the ones who
38:36developed high risk metastatic
38:37disease as well suggesting that this
38:40tonic activation of the meta UPR
38:43pathway could have some predictive
38:44ability to look at patients who
38:46have the highest risk of disease.
38:53OK. So, so far I showed you the
38:58transition between primary melanosite
39:00and B rap driven melanosite.
39:01These are stead images.
39:03I probably should have
39:04described this 40 slides ago.
39:06Looking at these are fixed cells,
39:08same with HP60.
39:09And then we take the images deconvolute
39:12and then make these beautiful movies
39:14of what the network looks like.
39:16Normal Milano sites,
39:17nice connected network,
39:19OIS Milano sites, still connected,
39:21but a lot more mitochondria.
39:23And then when you transform these cells,
39:25what do you see?
39:26All the mitochondria are kind of unconnected,
39:28disconnected, undergoing fragmentation,
39:30and they don't look very helpful.
39:34And I was always interested in knowing,
39:38you know, when my lad first started,
39:40we studied this transition from primary
39:42to transform because it was the easiest.
39:44We could do this in self culture very easily.
39:47This part here from primary to OIS was
39:49much more complicated and that's what
39:51we're just figuring out right now.
39:52In order to share it with you,
39:54let me tell you a little bit
39:55about this next step.
39:56So I mentioned earlier that money
39:59country are dynamic organelles and
40:01they undergo rounds of fusion.
40:04They undergo rounds of fission.
40:06And the reason why they do
40:07this is to exchange material.
40:09And if you look at a mitochondrial network,
40:12the network within a single cell
40:14tends to be quite homogeneous.
40:16They're always exchanging proteins,
40:18lipids, DNA, metabolites,
40:19and that keeps the network equal and happy.
40:22So when you image them,
40:24they all look the same.
40:26And what's responsible for this,
40:29of course,
40:29is the mitochondrial dynamics machinery.
40:31We're not going to talk about fusion.
40:33We're just going to talk about division.
40:35And the main protein that's responsible
40:37for mitochondrial division is
40:38dynamically related to protein one.
40:40It's normally cytosolic,
40:42you phosphorylate it to activate it,
40:45it as a monitor,
40:46goes to mitochondria,
40:47accumulates,
40:48generates decision machinery and then
40:51divides the mitochondrial network.
40:54And I became very interested in
40:56knowing how the Met Klinase pathway
40:58could alter mitochondrial dynamics
41:00because of this phenotype here.
41:02And that is, if you look at wild type maps,
41:05you see beautiful connected mitochondria
41:07and then when you fragment them,
41:09mitochondrial division stays
41:12much longer in cells.
41:15These cells always display a chronically
41:18divided mitochondrial network,
41:19the majority of cancer cells,
41:21and then compare those to
41:23the actual tissue of origin,
41:24you will see chronically divided
41:26mitochondrial networks in
41:28the majority of cases.
41:29You can see that the consequence
41:32of this of course is less oak car,
41:35less oxygen consumption at the basal level,
41:38the maximum level,
41:39and also a decrease in the ability for
41:42these mitochondria to generate a TP.
41:44And what we discovered in this paper was
41:47that when you put grass into a cell,
41:50you get activation of course
41:51of the math kinase pathway.
41:53And we identified that there's a
41:55phosphorylation site on D RP1AT
41:57series 616 that's activated by
42:00on the genuine kinase signaling.
42:02And this is what was responsible
42:04for actively dividing that network.
42:06If you looked into literature,
42:07616 in the human,
42:08and I think it's 5:30 or 592 in the mouse,
42:12this is the activating phosphorylation
42:13for the decision machinery.
42:15And if you add these targeted
42:17therapies to cells, you can turn
42:20off the phosphorylation in meths.
42:21And you can also see how you have
42:24the phosphorylation of D RP1 in green
42:27here and then it completely goes away
42:29when you treat with GSK from 120212
42:31or and you can see the difference
42:34in the money comes from network.
42:35Those are fragmented to beautifully
42:37connected within just a few hours actually
42:42and I'm not going to take you through all
42:44of this because this portion is published.
42:46What you can see is you can screen all of
42:49these, all of these targeted therapies.
42:51Here we're just using a 375 cells.
42:53The B ref is 600 positive.
42:55You can also do SKML 28,
42:57so you can hit them with Plexicon.
42:58You can hit them with the, you know,
43:00very old PD mech inhibitor GSK.
43:03You can also use or lot NIM to hit or B2.
43:06You get really beautiful rapid fusion of the
43:10mitochondria networks in all of these cases.
43:12And what they basically do is,
43:13of course they reduce the European
43:161 phosphorylation to turn it off,
43:17but they also you lose D RP1
43:20protein and you lose message.
43:22And what happens is, guess what,
43:24all those chronically divided
43:25mitochondria now fuse.
43:26And when they fuse,
43:28they exchange materials and now all the
43:30mitochondrial function goes back to normal.
43:33So it's a really dynamic process
43:35and it's really amazing to look at.
43:37And if you go into patient sections
43:40here we're just looking at melanomas
43:42and C2 using the V E1 antibody against
43:44V ref P600E and then screening
43:46using a commercially available mouse
43:48antibody that also picks up the human
43:50phosphorylation adjacent sections
43:51and you can see really beautiful
43:54relationships between the two.
43:55Any lesion that's V ref P600E wild
43:57type also doesn't have brass mutation,
43:59almost never has V reps or never
44:02has V RP1 phosphorylation positive.
44:05And the majority of tumors that are
44:08B reference 600D positive always
44:10have or approximately 70% of them
44:13have D RP1616 positivity.
44:14I'm going to visit this again in a
44:17few minutes because this doesn't tell
44:18you so much information because I mean
44:20you can just screen and you can say,
44:21well this person has a tumor,
44:23it's B roughly 600 positive,
44:24they already have a tumor.
44:25It's not that interesting,
44:26but just kind of a proof of
44:28principle that the two are related.
44:29So here we're now going into Malana
44:34Sites again looking at B ref,
44:37P600E and P10.
44:38You can of course create,
44:40transform Milan of sites and you
44:42can see this really nice change and
44:44might have come to architecture.
44:46And if you eliminate D RP1 in
44:50Milan of sites and transform,
44:52you now lose the ability for B ref
44:56and P10 to transform these steps.
44:58What do you think they do instead?
45:02Undergo senescence.
45:02But I'm not going to talk about it
45:04here in the first molecular cell
45:05paper showed the senescence pathway
45:07being activated in meths and now
45:08we're showing the same thing here in
45:10the in the melanocyte populations.
45:12So you have to be able to chronically
45:14divide your mitocundial network in order
45:17to generate these transformed cells.
45:19Seems to be established in meths,
45:21in melanocytes.
45:22And other people have now done work
45:24in the pancreas and in the lung in
45:26different brain tumor initiating
45:27cells showing the same pathways.
45:29And what we decided to do was to
45:32take Melano sites CRISPR 616 A,
45:36which is basically a mutant form of
45:38D RP1 that can't divide the network.
45:40And then we compared it to a CRISPR
45:43version of 6/16 D chronically
45:45activated D RP1.
45:46And you can see what happens
45:47to these Melano sites.
45:48I mean they have crazy mitochondria in
45:50the middle, right? Proof of principle.
45:51If you can't divide your
45:53mitochondria network, what happens?
45:54This might.
45:55This Melano site just expands up like
45:58crazy and makes lots of medic country.
46:01And here you're just simply looking
46:02at the ability to transform these
46:04cells can't divide your network,
46:06You'll never transform these cells,
46:07and if you're already divided,
46:09you transform more easily.
46:13So this is a really fun experiment
46:14and I had to convince people
46:15to do this and I love it.
46:17So what we can do instead is you can
46:21take the line of sites and you can of
46:23course introduce an awkward unit to them.
46:24And you can activate here at the
46:27600B to generate senescent lesions.
46:30What I wanted to do is put in a
46:33pharmacologically activated form
46:34of D RP1 where we can induce
46:37its activity with doxycycline.
46:39And here what we're using is the 616
46:41D So this is the moment it's on,
46:43it divides the network.
46:45OK And what we did was we took D
46:50RP600E generated senescent and lab
46:52sites living culture for two months,
46:54completely senescent population.
46:57And the D RP1 which is pharmacologically
47:00activatable selected for,
47:02but we never turned it on,
47:04but we did instead was once we knew
47:06that the populations were senescent,
47:07we turn it on and guess what happens,
47:10a subset of these cells now
47:12reenter cell cycle,
47:13divide their mitochondrial network
47:15and escape the senescence program.
47:17So simply dividing your mitochondrial
47:19network is sufficient to drive
47:22oncogene into senescence escape.
47:24But I mean,
47:25who's got the system in their cells?
47:27All right, it's not important, it's just fun.
47:31What is relevant is if you start
47:34to look at what could these
47:38cells potentially encounter,
47:39Heavy metals, Nicotine.
47:44Electronic cigarette residues.
47:45The things I'm showing you here
47:47and if you make these senescent
47:50melanocytes with B raffy 600 E
47:53and now you culture them once
47:55you know they're senescence.
47:56And there were,
47:57I should mention the reason why
47:59we selected these two reasons.
48:00Funding #2.
48:03They're all published to make
48:06mitochondrial shape changes.
48:08You take the cells and you expose
48:11them to cadmium chloride or nicotine
48:13or .25 puffs per mil of cell culture
48:16media of electronic cigarette residue.
48:18Banana flavor is the one.
48:20It will induce medicantial division and
48:23you'll start to get escape from Hawaiias.
48:26So you can envision riverbeds
48:29that have contaminations,
48:32populations,
48:33people where they live along rivers
48:34and and bodies of water which
48:37historically have been quite clean.
48:39Places like the Ganga River,
48:42different places where there's lots
48:44of industry flowing into these rivers
48:46and we're starting to see increases
48:48in Melanoma genesis along populations.
48:50This is environmental toxicity and
48:51all of these things by themselves
48:53never transform cells and you're
48:55never going to take cadmium
48:56chloride and do anything.
48:58But if you do it in the presence of an
48:59Aqua genus cell that's already got it,
49:01then you could potentially make an impact.
49:05All right. And then I'll
49:06be wrapped up very soon.
49:08I know we're almost ending.
49:09If you look at some of the consequences
49:13of transforming melanocytes
49:14plus or minus D RP1, you know,
49:16here's looking at complex one
49:17activity or complex 2 activity and
49:19you'll see the complex one activity.
49:20When you transform, A lanocyte drops,
49:23and it doesn't drop if you
49:27inhibit D RP1 beforehand.
49:28So this tells us that if you can't
49:30divide your multicultural network,
49:31you'll never change the
49:33metabolism of these cells.
49:35And is it important?
49:36And this is the experiment that
49:37tells you that it's actually
49:39important and we can use ND I-1.
49:41You know, humans are efficient
49:43animals in the most part.
49:45You know,
49:46our out of our complex one protein
49:48is our complex one structure is
49:50about 45 proteins and another
49:51hundred 150 factors that regulate
49:53the assembly and vocalization of
49:55this complex yeasts are amazing.
49:57They only have one protein
49:59that replaces all of this.
50:00It is called ND I-1.
50:02And if you express it in melanocytes,
50:03you get of course increased
50:05complex one activity.
50:07But what's amazing is if you
50:08sustain complex one activity,
50:10when you try to transform,
50:11you lose the transformation
50:13potential of these cells.
50:14So they have the graph,
50:16they silence tumor suppressor gene,
50:18they're dividing the mitochondrial network,
50:20but now they're not losing complex
50:22one activity and they don't transform.
50:25And if we generated a series,
50:28oh,
50:28I skipped all the data and
50:30good luck getting it back.
50:32If you look at the reason why when
50:33you divide your mitochondrial network,
50:34you select for all of these
50:36mutations in the mitochondrial gene
50:37and that encode for complex one,
50:39all right,
50:40Kind of a proof of principle
50:41of that is we take melanocytes,
50:43and you know when you do,
50:45when you transform melanocytes,
50:46you get a heterogeneous population
50:48of mitochondrial genomes,
50:49mutations throughout the genome.
50:50Some are important, some are not.
50:52So what we do is we subject them to ethidium,
50:54bromide, pyruvate and uranine selection.
50:58That allows us to basically kick
50:59out all of the mitochondrial
51:01genomes in that cell and we reduce
51:04it to 1 genome per cell.
51:05You take the pressure off and that
51:07one genome repopulates the cells.
51:08So now you create these clonal
51:11melanocytes that are transformed.
51:13You implant them into animals and
51:16what you see is controlled cells of
51:19course have normal complex one activity,
51:22no tumor volume.
51:23You transform,
51:24they lose complex one activity.
51:25I really showed you that.
51:26And they generate A tumor and growth.
51:29Zero cells have no complex one activity
51:31of course the Pentium mitochondrial
51:32genome and they don't form a tumor.
51:35Row 0 is this process of eliminating
51:38the mitochondrial genome.
51:39You take 10 clones that have
51:42decreasing ability to have complex one,
51:44which you see is this sweet spot of
51:47tumor formation occurring in this 30
51:49to 40% deletion of of complex one.
51:53And this exactly mirrors the levels
51:55of complex one that you see when you
51:57transform these these cells in culture.
52:01Small mitochondria are bad.
52:03Another reason why they're bad is
52:05paper that we published a long time
52:06ago was it's really hard to kill
52:09cells with small mitochondria.
52:10The reason why we got on to this
52:11is in the cell death field,
52:13it's knowing that if cells
52:14are undergoing mitosis,
52:15you can't induce mitochondria out
52:18of membrane reverbalization of that
52:20process of releasing cytochrome C
52:22to activate the cell death pathway.
52:24And you also whenever you have
52:28hyperframeants of mitochondria,
52:29it's really complicated to treat
52:31them with chemo,
52:32the the cell lines themselves,
52:33they just don't respond very well.
52:35And we figured out the mechanism here,
52:36basically the Bax protein,
52:38everybody knows the BCL two family
52:40with Doctor Katz in the audience.
52:41These provototic proteins can't
52:43link into the outer mitochondria
52:45membrane to drive them up.
52:47And furthermore,
52:48when you have many mitochondria
52:50in the cell and you have a fixed
52:53repertoire of provototic BCL 2 proteins,
52:55it's also harder to drive cytochrome
52:57cereals and apoptosis.
52:59Mindy, you know,
52:59this is a really beautiful paper
53:01that basically showed the
53:02more mitochondria you have,
53:03the more mitochondrial mass you have,
53:05the more drug it takes to kill the cell.
53:07The reason is really simple.
53:08If you have 100 molecules of facts
53:10and you have 10 mitochondria,
53:11every mitochondria gets 10
53:13molecules and you create a four.
53:15Now you have 100 molecules of facts
53:16and you have 100 mitochondria.
53:18Each mitochondria gets one molecule of
53:20facts and you'll never kill that cell.
53:28I think I'm going to end with this
53:31slide here where what we decided to
53:33do is we use this mouse antibody that
53:37detects the phosphorylation event
53:39on D RP1 and we screen through many
53:42different types of skin lesions.
53:44Normal skin never positive dysplastic
53:47nevi almost always positive Melanoma in
53:50C2 about 70% and then any non Melanoma,
53:53any non B ref Melanoma 7%.
53:55Again that's not that interesting
53:57because you know you're just kind
53:59of following the process along.
54:00What's curious is you know the
54:02vast majority of these lesions are
54:04being referenced to be positive,
54:06but it doesn't necessarily tell you if
54:08you're going to eventually develop disease.
54:10So what we did instead was we did
54:12something called the discovery cohort
54:14where we took 300 and 53150 biopsies
54:18and within those 350 biopsies were
54:2335 patients that developed disease.
54:26What we're able to do is we could
54:28screen with the D RP1 antibody and we
54:30could pull out 31 of the 35 patients,
54:33suggesting that if you know your
54:35status of D RP1 in that lesion,
54:37it could potentially inform you on
54:38the likelihood of getting disease and
54:40maybe inform you on the number of times
54:42you should get these lesions screened.
54:44And if you're super keen in the audience,
54:48it creates quite of a really interesting
54:50biological question because here the lesion
54:52is gone and you screened it for B RAF
54:54or you screened it for both B RAF and D RP1.
54:57How is it you still develop disease
54:59and it you develop disease,
55:01but the original lesion is gone,
55:02but D RP1 still tells you
55:04if you're going to get it.
55:05It's a really amazing biology.
55:06And I have some ideas in terms
55:08of why I think that's the case.
55:10We developed a human specific antibody
55:11that's recombinant and we're doing
55:13that rather than using the mouse.
55:14We made one that's say human specific
55:16and it also works beautifully.
55:18So what I showed you today was a bunch
55:23of programs looking at mitochondrial
55:25control of linosite biology,
55:27Melanoma, genesis.
55:28I don't need to read everything
55:29that's going on here.
55:30This is just a summary of what
55:32I've been doing.
55:33At the moment we're looking at a TF5
55:36in the MiTo UPR in skin development.
55:39We're looking at some of the MiTo
55:41UPR regulation of spatial metabolism,
55:44gene expression,
55:45changes in skin and different
55:47types of tumor types.
55:49And we're of course looking at all of these
55:52different ways of affecting this process.
55:54And one of the things we're doing
55:56is making a proof of concept protact
55:58against ATF5 protein because it's
56:00really a hard protein to target.
56:02So these are all the people who do the work,
56:05collaborators,
56:05external collaborators,
56:06some of the people who support the group
56:09because they provide us with young
56:11scientists and the people who pay the bills.
56:14And I think for being a great audience.
56:15And I saw lots of heads shaking and
56:16smiling and stuff when I was talking.
56:18And I like that because it means
56:19I have people hear me. Thank you.
56:32Hi.
56:42Yeah, it is the classic and
56:45I think if anyone knows me,
56:48I'm not classic and so I'm traditional.
56:52We haven't done anything yet and that's where
56:54I want to start with the UV in Lenosites.
56:57I want to use things like the Yummers,
56:58like all of these types of models.
57:00We just haven't gone there yet.
57:01But this the, the movement from normal
57:03to the senescence and those phenotypes,
57:06all of this data are fresh.
57:09And you'll see some data are like really nice
57:11blue and red and some other data are pastel.
57:13The pastel is the good data and
57:14the blue and red is the stuff
57:16that's being replaced, right.
57:17So it's, it's active by your question.
57:20Yes, we will. I don't have a Gray,
57:23you know, telenova program at Sinai.
57:25It's not huge.
57:26It's like 2 investigators and this is like
57:47the experiment. Yeah,
57:48this is we don't have that information.
57:51I always wonder how you can take
57:53you know these are just these are
57:55nations that were taken off biopsy
57:57and then we score for D RP1 and
57:59tell the language that's about most
58:03of the time at a different size.
58:06We don't have a complete data set there.
58:08How molds are communicating with each
58:11other about me by are communicating with
58:13each other I predict will of course
58:16be to be soluble factors like and I
58:20predict that my world is all about
58:23what I study So you know understanding
58:25how mitocrontial division informs
58:27mitocrantial quality control which then
58:29impacts on how these cells secrete.
58:32We already know that mitocrantry controls.
58:35I don't think it's a surprise now that you
58:37have a major option regulating the form,
58:40the composition and the quality control
58:42of the organome that you can secrete that
58:45will form another place or it's just,
58:48you know,
58:48disease that already disseminated.
58:50We took it off and it just happens
58:52to be residual either one.
58:53But I still think it's interesting.
58:55And if you can take the lesion off
58:57and predict who's going to get,
58:58it might be a good way of screening.
59:06All
59:08right. Some
59:14trap application.
59:22Yeah.
59:27Yeah. So, yeah, exactly the question.
59:31We don't know. So, you know,
59:33we've done the fibroblast work.
59:35We've done the melanicite work
59:37looking at inducible forms of Ras,
59:40inducible forms of H, Ras, G12, the Q1,
59:44the Q61 series and it's not clear.
59:48So to add a more lack of
59:51clarity to the situation,
59:52I have a student now who's screening also
59:55in Drosophila try to help figure this out.
59:57I mean I I assume it's because of
59:59the way rats affects multiple arms
01:00:00of the pathway versus just if the
01:00:03rats mutation is a bit more focused.
01:00:05And if that's the case,
01:00:06I'm hoping we can do some of the genetics
01:00:07and your software to figure it out.
01:00:09So we are, you know, we are,
01:00:10we're regulating the might of UPR pathway,
01:00:12we're regulating muddy Contra division
01:00:14machinery there because you can
01:00:15look at developments of the wing,
01:00:17you could look at development
01:00:18of stress signaling in the eye,
01:00:19you can look at a bunch of.
01:00:20Wing development, phenotypes.
01:00:21You can parse out a lot of things
01:00:23because there's such a huge literature
01:00:25there on how each one of those
01:00:27phenotypes is genetically controlled
01:00:29by distinct portions of the pathway.
01:00:31I guess that's the best answer.
01:00:33We don't know,
01:00:33but we're kind of starting to work on it.
01:00:35But maybe not the most elegant thing.
01:00:46Yeah,
01:00:48yeah, minutes, actually.
01:00:50So if you treat with vemurafenib,
01:00:53you can basically put
01:00:54it under the microscope.
01:00:55You can take the plate of cells,
01:00:56treat with vemurafenib, a 370.
01:00:59Fives will be fully fused within two hours.
01:01:01You'll see increases in TMRA.
01:01:03You can do it like flow is super, super fast.
01:01:06They're extremely dynamic,
01:01:08suggesting that mitochondria in
01:01:09these tumors are just kind of
01:01:11primed and ready to go.
01:01:12And you refuse the exchange
01:01:14and they're fixed.
01:01:15So we actually have a really
01:01:17interesting aim and a grant looking
01:01:19at how the diversity of mitochondria
01:01:22within a particular cancer cell,
01:01:23the heterogeneity can be overcome by
01:01:26fusing them and allowing for repair.
01:01:32Maybe we'll take the additional
01:01:34questions up front. Thank you.