Sorting of Small RNAs Into EVs Secreted by Human Cells
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 340 | |
| Autor | ||
| Lizenz | CC-Namensnennung - keine kommerzielle Nutzung - keine Bearbeitung 4.0 International: Sie dürfen das Werk bzw. den Inhalt in unveränderter Form zu jedem legalen und nicht-kommerziellen Zweck nutzen, vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. | |
| Identifikatoren | 10.5446/45077 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
00:27
Vorlesung/Konferenz
01:05
Vorlesung/Konferenz
01:38
Vorlesung/Konferenz
02:10
Vorlesung/Konferenz
02:45
Vorlesung/Konferenz
04:39
Vorlesung/Konferenz
05:52
Vorlesung/KonferenzBesprechung/Interview
06:23
Vorlesung/Konferenz
06:58
Vorlesung/Konferenz
07:35
Vorlesung/Konferenz
08:53
Vorlesung/Konferenz
09:29
Vorlesung/Konferenz
10:21
Vorlesung/Konferenz
17:50
Besprechung/Interview
25:08
Vorlesung/KonferenzBesprechung/Interview
25:36
Vorlesung/Konferenz
27:36
Vorlesung/Konferenz
28:50
Vorlesung/KonferenzBesprechung/Interview
29:21
Vorlesung/Konferenz
30:04
Vorlesung/Konferenz
31:14
Vorlesung/Konferenz
31:45
Vorlesung/KonferenzComputeranimation
Transkript: Englisch(automatisch erzeugt)
00:14
Well, I'd like to start by crediting the Lindau Foundation for providing a forum of such obvious interest and value
00:22
to so many students and postdoctoral fellows around the world, only a small fraction of whom those of us in the front rows will be able to visit with, but it is a wonderful opportunity. Today I'm gonna talk about how cells export not only proteins, but RNA molecules.
00:44
Of the thousands of protein molecules that all cells manufacture, a surprising number, roughly 30%, are exported outside of the cell. These are, very often, water-soluble protein molecules
01:01
like insulin, and yet, made in the cytoplasm, they have to somehow cross the barrier of at least one biological membrane. One of the great triumphs of 20th century cell biology was the elucidation of the organelles in cells that are responsible for the encapsulation
01:21
of protein molecules like insulin for their export outside of the cell. Let me give you a couple of examples of the organelles that we now understand and mediate all protein traffic in eukaryotic cells. So this is a cell, quite dramatic, beautiful cell,
01:42
that represents the range of organelles that one finds in a differentiated mammalian tissue. And for those who haven't seen this, let me draw your attention to these encapsulated forms of almost crystalline insulin, which are introduced into a conveyor belt-like mechanism
02:02
of membranes, ultimately producing a granule that delivers insulin by fusion at the cell surface, secreting it to the cell exterior. This process of secretion of molecules is recapitulated in virtually all of the cells of our body
02:22
using mechanisms that are conserved over two billion years of evolution, including the capture and secretion of chemical neurotransmitters at the synapse. In the presynaptic terminal, you see collections of thousands of small vesicles responsible for the release of neurotransmitters
02:42
in response to an action potential. Much of what we now know to explain how these membranes are generated and conveyed within mammalian cells came from the brilliant pioneering work of a great cell biologist by the name of George Palade,
03:02
shown here, sitting adjacent to his favorite tool, the electron microscope. Palade and his colleagues at the Rockefeller University understood through a very elegance and laborious series of experiments how protein molecules move from one station to the next,
03:21
largely in pancreatic tissue. I had the opportunity when I was a beginning postdoctoral fellow to hear Dr. Palade describe his pioneering work in a version of his Nobel lecture that he delivered at a meeting of the American Society for Cell Biology in December of 1974,
03:41
just after having returned from Stockholm. And the results were stunning and beautiful. His morphology was unsurpassed. But it struck me, having been trained as a biochemist, that in that era, 1974, in spite of his achievement, we knew not a single protein molecule or gene
04:02
involved in the organization and execution of this elaborate pathway. I resolved as I began my career in 1976 at the University of California at Berkeley to try to devise a scheme to understand the molecular mechanism of this process by applying a combination of genetics and biochemistry,
04:23
and at the same time, a brilliant young biochemist by the name of James Rothman embarked on a similar effort using biochemistry to understand the means by which protein molecules are conveyed within mammalian cells. The choice of organism in my case
04:41
was a simple microorganism, baker's yeast, that you heard about this morning from Yoshinori Osumi. We knew from the pioneering work of Hartwell and others, including Paul Nurse, that a genetic approach could be successful in elaborating and understanding
05:01
a complex biochemical process as complex as what Palade had described in mammalian cells. And our attention was drawn by thin-section electron microscopy to the interior of the yeast cell, which displays under the bud surface of a dividing cell
05:22
a collection of small vesicles looking rather similar to the synaptic vesicles you saw a moment ago. And we guessed that these vesicles may be responsible not only for secretion, but that the membrane fusing at the bud plasma membrane would allow the cell to enlarge.
05:41
Thus the prediction that genes and proteins required for the production, movement, and fusion of these vesicles would be essential for cell viability, thus requiring the use of the genetic approach that Hartwell and Nurse perfected in their dissection of the yeast division cycle.
06:01
My efforts were aided enormously by a most remarkably talented graduate student, one of my first graduate students, Peter Novick, shown here at his bench. We all had rather long hair back then. You can see. I too had long hair. I have much less, but I am pleased to say that I have even more than Novick does now.
06:24
But it's amazing how things come full circle. Novick has gone on to a remarkably successful career, and he is now the George Palade Chair of Cell Biology at UC San Diego. Well, we embarked on a screen for temperature-sensitive lethal mutants that block secretion.
06:41
And of course we were doing this in the context of the University of California Berkeley, the birthplace of the student protest movement. There were naturally concerns about our proposal to kill living beings, even yeast cells, and we were the subject of a protest movement and the torture in the laboratories.
07:02
Yeast have feelings too. We overcame these objections, and in May of 1978, just as Dr. Palade came to Berkeley to deliver two special lectures, we had our first mutant called SEC1, short for secretion one.
07:20
Palade encouraged Novick to inspect this cell by thin-section electron microscopy. And one of the great memories I have in my career came when Peter called me excitedly down to the electron microscope room to see the following image of a cell literally chock full of mature secretory vesicles that fill the entire cytoplasmic volume
07:42
as a result of a lesion in a gene that we now know to be evolutionarily conserved, which is required at this stage in the secretion of proteins and of neurotransmitters in all cells. Now we then embarked on a year and a half search for more genes. We now have probably close to 100 such genes.
08:03
We were able to map these genes on a pathway remarkably similar to the one that Palade had demonstrated. And yet on cloning most of these genes, they showed no homology to anything. Now I wasn't surprised by this. The genes turned up in databases in subsequent years.
08:21
They are conserved, as I said, over evolution. But having been trained by one of the great biochemists of the 20th century, Arthur Kornberg, I knew that what we would have to do was to devise cell-free reconstituted reactions that would allow us to recapitulate in the test tube significant events in the movement of proteins
08:42
into and along with vesicles. Here my work was aided once again by a fantastically brilliant graduate student by the name of David Baker, who has gone on to his own remarkably successful career in a very different research area. Baker devised within two weeks
09:01
of beginning his graduate career in my lab, a cell-free reaction that allowed us to reconstitute in the test tube the events of the first half of the secretory pathway. And over the years since then, we exploited his discovery to isolate many of the protein molecules in functional form that are responsible for the production of vesicles
09:23
that bud from an intracellular membrane. And I'll just show you one example. Subsequent to Baker's contribution, we purified all of the proteins that are necessary to pinch a vesicle from the endoplasmic reticulum to capture protein molecules en route
09:41
to the Golgi apparatus. And we discovered that these protein molecules that are encoded by the sec genes create a novel coat protein that literally envelops a piece of membrane pinching it from the ER, producing a uniform field of so-called coated vesicles. Now, we've used this logic of a combined genetic
10:03
and biochemical approach repeatedly over the past several decades. And I'd like to tell you now about our most recent efforts to understand a very different process involving the production of vesicles not confined to the interior of a cell,
10:20
but those that bud from the surface of a cell or are produced and ejected outside of the cell, so-called extracellular vesicles or exosomes. We, beginning about 20 years ago, people discovered that cultured mammalian cells, indeed likely all metazoan cells, manufacture vesicles for export.
10:40
These are produced by mechanisms similar to that used by enveloped RNA viruses for their encapsulation and budding at the cell surface, or their accumulation within a structure called the multivesicular body. That ejects particles from the body by fusion at the cell surface.
11:00
There are a variety of these extracellular vesicles, and we're only now just beginning to understand how they're produced, and the diversity of them that are produced by a variety of cultured mammalian cells, including human tumor cells. And the interest in these vesicles is quite pronounced because human tumors are known to produce these vesicles in abundance, and the content,
11:22
the molecular content of these vesicles appears to change during the development of metastatic cancer. So a great deal of interest is focused on using these vesicles for diagnostic purposes, and also possibly for the development of novel therapies. Well, my laboratory has been interested
11:40
strictly in basic issues, and one basic issue that I'd like to tell you about this afternoon is how these vesicles, quite surprisingly, capture small RNA molecules. Jan Latval in Gothenburg, Sweden, now about 10 years ago,
12:00
discovered in a fairly crude preparation of such extracellular vesicles that they not only contain a typical biological membrane and integral membrane proteins, but quite unexpectedly, they contain a population of small RNA molecules, roughly 100 nucleotide long.
12:21
A diverse population of such small RNA molecules that appear to differ when vesicles are isolated from different cultured cell lines. So we're very interested in how it is that these RNA molecules can become packaged into such a vesicle. Well, now about just over six years ago,
12:40
a wonderful graduate student named Matt Shurtleff, who is now a post-doctoral fellow and is here in the audience today, having been invited to this meeting, embarked on an effort to try to isolate a particular small vesicle exported by a standard cultured human cell line, HEC-293 cells.
13:01
And unlike much of the literature in this field, we decided to devote considerable energies to the purification over several steps of isolation to obtain as homogeneous a population of vesicles as possible with simple techniques. Here is a summary of what Matt did.
13:20
He sedimented vesicles at low speed to remove debris. He collected vesicles by high-speed sedimentation. He then used buoyant density sedimentation to achieve a buoyant equilibrium position of vesicles that are enriched in a tetraspanin membrane protein called CD63.
13:43
At an interface between 20 and 40% sucrose, Matt was able to further purify these vesicles by absorption to an immobilized form of an antibody against one of the tetraspanin proteins, CD63. He then did a deep-seq microRNA analysis
14:03
of the 600 or so major microRNAs found in HEC-293 cells and he found only a fraction of these microRNAs are enriched in extracellular vesicles and of those enriched, only a very small number are hundreds of fold enriched.
14:21
One in particular, miR223, is enriched about 1,000 fold. So the cell has gone to considerable efforts to define, sort and capture a small number of microRNAs. Now, to what purpose is this sorting performed? It could be that the cell is devoted
14:42
to getting rid of microRNAs that it doesn't want, so this could be another way of disposal, or it could be that these microRNAs are being packaged into vesicles to be delivered to a distal tissue, perhaps to alter gene expression. The jury is out on that, but we decided to focus our efforts on trying to understand the sorting event
15:02
because in inspecting just the few microRNAs that are so highly enriched, we found no primary sequence RNA sequence similarity, suggesting that there were some other rules that were being used to achieve this sorting event. Well, of course, the way we pursue projects of this sort,
15:23
as I've indicated before, is by trying to biochemically reconstitute a significant event using membranes and cytosol, in this case, from broken HEC293 cells. Matt was up to the task, and he devised a very simple, elegant reaction
15:40
with membranes and cytosol from broken cells that we believe has allowed us to reconstitute a high-fidelity RNA sorting event. The procedure he used is highlighted on the top. Membranes from broken cells are mixed with cytosol and ATP in the presence of a chemically synthetic form of a microRNA.
16:03
I'm gonna tell you about two of them, one that Matt worked on called MIR223, and another one called MIR122 that another graduate student, Moraima Tomoshdiaz, has isolated in high buoyant density vesicles produced by a triple-negative breast cancer cell line. After a 20-minute incubation at 30 degrees,
16:22
membranes are sedimented, washed, and treated with ribonuclease to digest unincorporated RNAs. These membranes are then further collected. They are ruptured by detergent, and the protected RNA is amplified and quantified by QTPCR.
16:44
Here are some representative data. Matt found in a typical experiment that about 9% of exogenous MIR223 was protected. He showed in other experiments that that protection depends upon the addition of cytosol and on the addition of membranes.
17:01
They must be present together, and that the reaction is suppressed if the incubation is simply held on ice. Now that this is a RNA-selective sorting event is demonstrated by a reaction, a parallel reaction, conducted with a purely cytoplasmic microRNA called MIR190, which you see here is very poorly incorporated,
17:21
just above the background obtained in a parallel incubation held on ice. Matt then used this same reaction to try to fish out an important RNA-binding protein that may be necessary to recognize MIR223, to allow it to be captured into vesicles formed in this reaction.
17:40
He did this by using a three-prime biotinylated derivative of synthetic pure MIR223, which is introduced, just as before, into a reaction, but after the reaction is complete, the RNA, exogenous RNA is degraded, the membranes are dissolved,
18:02
and a protein that adheres to the microRNA is captured on streptavidin beads, which are washed and then eluted with high salt. He found one major protein in his mass spec analysis of these proteins that bound to the streptavidin beads that had already been described
18:21
to be secreted in exosomes by cultured mammalian cells. And he made a CRISPR knockout, a homozygous knockout of the HEC293 cells in order to examine whether this protein, called YBX1 or Y-box1 protein, is essential for the sorting of MIR223
18:41
in our cell-free reaction. And here is that experiment. In two controls, with wild-type membranes and cytosol, about 10% of the exogenous MIR223 is packaged in vitro, but if the cytosol is obtained from a YBX1 homozygous null, the efficiency of packaging of MIR223
19:02
is reduced about five-fold. It is largely, though not completely restored, if the cytosol is obtained from a null cell into which the YBX wild-type gene has been reintroduced. And in other experiments, Matt's shown that extracellular vesicles produced by these null cells have much less MIR223 packaged.
19:24
So both in vivo and in vitro, we have evidence that the YBX1 protein is required. Now in further studies in collaboration with Alan Lambowitz at the University of Texas, Matt found that other major RNAs fail to be packaged or greatly reduced
19:42
in cells devoid of the YBX1 protein. For example, full-length tRNAs, YRNAs, and vault RNAs are reduced three to four-fold in exosomes secreted by the YBX1 null cell. So we believe that the YBX1 protein is important for a variety of RNAs
20:02
to be captured into these extracellular vesicles. Now surprisingly, in the experiments I'll show you now, Morima has found something rather different in requirements for the packaging of the microRNA that she found in vesicles isolated from a triple negative breast cancer cell line, MIR122.
20:23
She showed that this microRNA is also enriched a thousand-fold or more in high buoyant density vesicles. She was able to repeat Matt's experiment, Matt's cell-free reaction experiment, showing that a very efficient packaging occurs when a full incubation is conducted,
20:42
but with cytosol from the YBX1 null cell, packaging is greatly reduced, and the packaging is very deficient when the incubation is held on ice. But in contrast to the behavior of MIR223, Morima reports that MIR122 is much less dependent
21:02
on the YBX1 protein. It's reduced only a few percent, whereas, again, the reaction is essentially null when the incubation is conducted on ice, and in other experiments she showed that the normal packaging of MIR122 is dependent on cytosol and membranes.
21:21
Now, she then repeated Matt's experiment with a biotinylated derivative of MIR122, just the same experiment as before, but with a biotinylated derivative, and she found in her mass spec analysis a different set of RNA binding proteins, nucleolin, the lupus law antigen, and nucleophosmin.
21:44
She systematically depleted the expression of each of these proteins by a technique that had been reported by Jonathan Weissman's laboratory called CRISPRi that allows the depletion of, in this case, essential proteins to produce a cell
22:01
that is essentially devoid of the lupus law antigen compared to other control cytoplasmic proteins. So she then evaluated reactions programmed with either wild-type cytosol or law-depleted cytosol to detect a role for the law protein in the packaging of MIR122,
22:24
and in the final data slide that I'll show, she sees a quite efficient packaging with wild-type cytosol, approximately three to four-fold depletion, reduction in the efficiency of packaging with law-depleted cytosol,
22:41
but on purification from E. coli, a recombinant form of lupus law, she could restore packaging in an incubation that contained membranes and law-depleted cytosol. In important controls, she showed that membranes are required for that reaction,
23:02
and she showed, even more importantly, that cytosol from the depleted cell, when omitted from the incubation, is not restored to normal packaging on addition of the recombinant protein. That means that law is only one of probably many different proteins
23:21
that are required for this sorting event. Well, I've given you just a flavor of the approach that we now have at hand, which I'm quite excited about, because this offers us the opportunity to purify all of the different protein molecules that cooperate to sort RNA into vesicles, and we think this will provide us with the tools to understand what these sorting events may mean
23:42
in the physiologic context of a cell. So let me summarize what I've said in the form of just a series of cartoons. We believe that the YBX1 protein functions in the cell in part to convey small RNA molecules
24:00
into a vesicle that may form by butting into the interior of this endosome called a multivesicular body. We'd love to know how this RNA is recognized. Matt, at the end of his thesis work, discovered a sequence of a run of three, a four C residues near the three prime end
24:22
of miR223 required for sorting, and so we're pursuing that. We'd love to know how these RNP's are captured on this membrane or possibly on a membrane that buds from the cell surface. We think the machinery involved in conveying these RNP's into vesicles may be shared by a normal set of proteins
24:43
first described by Scott Emmer in yeast called the escort proteins that are responsible for the down regulation and internalization, ultimately the destruction of ubiquitylated cell surface receptors, and we think these two pathways may intersect. We'd love to know if the primary source of membranes
25:02
for this reaction in vitro are either these endosomes or the plasma membrane of the cell. Many, many questions remain. Well, let me show you the people in the group. This was a shot taken several years ago, and let me highlight for you Matt who's in the audience and will be happy to field questions afterwards,
25:22
initiated this work and has really changed the theme of my lab for the next years, and Moraima who's followed with equally important experiments that we're eager to pursue. Now what have I been doing all these years? Beginning about 12 years ago, I became very concerned about how we decide
25:42
where and what to publish in biomedical science. I was able to have some influence on this in my role as the editor-in-chief of the proceedings of the National Academy of Sciences. During that five-year term, I became increasingly concerned with the influence
26:01
of a number called the journal impact factor, which I believe has exerted a toxic effect on how we choose where to publish. Beginning now eight years ago, I was given an opportunity to take that concern to a greater effect with my appointment
26:20
as the editor-in-chief of an open access online journal called eLife. eLife was from its inception unusual in that it had the support of major funding agencies, specifically the Wellcome Trust in Britain, the Max Planck Society in Germany, and in the U.S., the Howard Hughes Medical Institute.
26:42
These groups and their presidents convened a meeting at Janelia Farm, now almost eight years ago, where the question was posed what to do about the problems that the investigators of these organizations face when they try to publish their most important work. And I wanna reflect on an important comment
27:01
at the outset of this meeting that was made by the then president of the Wellcome Trust, Mark Wolpert, who said it was time to wrest control of the biomedical literature back into the hands of the active scholars who engage in this research. And he was implicitly saying that it should be removed from control of commercial interests
27:21
and specifically of journals where all the decisions are made, or all the important final decisions are made by professional editors. As a result of this meeting and subsequent discussions, I began as the editor-in-chief of this journal eLife, which I think many of you have heard of. We developed a novel way of evaluating research
27:41
that was submitted to us, a consultative approach where referees and board members convene a consultation session online to confer on whether the paper has achieved some novel insight and whether the experiments that have been conducted were appropriately conducted.
28:00
We've been publishing now for about six years and we remain kind of an experimental journal trying new things all the time. Now, we started this journal with the intention immediately of avoiding the use of impact factor in the evaluation of scholarship.
28:21
The members of my board and I have very strong opinions on this and those of you in the audience know how pervasive this number has been in what decisions you, students, and postdocs make about where to publish. Indeed, many journals, commercial and non-commercial, advertise this impact factor irrespective of the number
28:42
in promoting your choice for publication. Nature magazine was perhaps most responsible, I would say, for promoting this number and leading, I would say, to a problem in the way that we evaluate science,
29:01
the exclusive nature of the selection of papers to be published in the three most prominent journals, Cell, Nature, and Science, I believe lead to misbehavior on the part of some scientists. And I believe this misbehavior is often very strongly influenced by the use of this journal impact factor.
29:21
Nature, until very recently, promoted their impact factor in big lights. All of us received this, 42.351, a remarkable level of accuracy for this calculation, far beyond the value of the data on which the calculation is made. Well, this was only one example. Many journals were advertising this number.
29:42
Postdocs in the room could probably recite these numbers and unfortunately, people were making decisions about where to publish on the basis of very small differences in this number. Fortunately, the editors of the flagship Nature resolved to convince the marketing people at Nature Publishing Group to eschew the use of this number
30:02
and two years ago, the then editor-in-chief of Nature, Phil Campbell, published a hard-hitting editorial that I wanna highlight for you. Time to remodel the journal impact factor. He said, metrics are intrinsically reductive and as such, can be dangerous. Relying on them as a yardstick of performance,
30:23
rather than as a pointer to underlying achievements and challenges, usually leads to pathological behavior. The journal impact factor is just such a metric. Well, many of us applaud this decision of the flagship journal. We gathered together in San Francisco some years ago,
30:42
a group of journal editors, including the then editor of Science Magazine, Bruce Alberts, to promote a document called the Declaration of Research Assessment that I'd like to leave you with, Dora San Francisco. In this declaration, we describe other means by which scholarship could be evaluated.
31:00
For example, one could read the papers instead of relying on a name of a journal or a surrogate number. And we ask that all involved in this enterprise move away from this toxic number in evaluating scholarship. Now, I have to leave you with a surprising note and that is that not everyone agrees with us about the impact factor. And it came as a shock to me
31:21
when I received the following tweet three years ago before the US national election, which alerted me to the problem that Donald Trump actually believes in impact factor. Here is an imaginary tweet, one based on the concern that he might control the National Institutes of Health.
31:41
Let me share this with you. Huge news, we won negotiations with McMillan, all nature journals to be rebranded Trump, all with an impact factor greater than 50 in 2016. Suck it, Checkman. Thank you for your attention.
Empfehlungen
Serie mit 34 Medien