The Kinetochore and Chromosome Segregation

Part 1: Chromosome Segregation

[00:00:07.22] Hi. I'm Sue Biggins
[00:00:09.02] and I'm an investigator at the Fred Hutchinson Cancer Research Center
[00:00:11.23] in the Division of Basic Sciences.
[00:00:13.23] I'm also an investigator with the Howard Hughes Medical Institute.
[00:00:17.07] Today, I'm going to talk about the process of
[00:00:19.20] chromosome segregation,
[00:00:21.07] which ensures that your cells inherit
[00:00:22.13] the right chromosomes when they divide.
[00:00:25.02] Cells must inherit the right chromosomes.
[00:00:27.11] This is a karyotype of a normal cell
[00:00:29.28] -- these are chromosome spreads --
[00:00:31.20] and you can see that you have two copies
[00:00:33.21] of every chromosome,
[00:00:34.17] one from mom and one from dad.
[00:00:35.29] This looks really different in a cancer cell.
[00:00:38.29] Here is a cancer cell chromosome spread
[00:00:41.21] and you can see a really large variation in chromosome number
[00:00:44.16] -- there's entire gains and losses of chromosomes.
[00:00:46.29] In addition, there are chromosomal rearrangements
[00:00:49.05] that often occur.
[00:00:50.25] This condition where cells have the wrong number of chromosomes
[00:00:53.24] is called aneuploidy
[00:00:55.11] and it's one of the most common hallmarks of cancer cells,
[00:00:57.22] and you can learn a lot more about the consequences of aneuploidy
[00:01:00.12] in the lectures from Angelika Amon.
[00:01:04.08] To ensure that daughter cells
[00:01:06.14] inherit the right chromosomes,
[00:01:08.03] they undergo two major events during the cell cycle.
[00:01:10.14] After G1 of interphase,
[00:01:13.11] they enter S phase,
[00:01:15.09] and the chromosomes replicate, or duplicate,
[00:01:17.18] to form sister chromatids.
[00:01:19.24] At mitosis, the chromosomes segregate
[00:01:21.23] to opposite poles
[00:01:23.13] and then when the cells undergo cytokinesis
[00:01:25.23] to form daughter cells,
[00:01:27.04] this ensures that each daughter cell inherits
[00:01:28.28] one copy of each chromosome.
[00:01:31.20] Now, a chromosome missegregation event during mitosis,
[00:01:34.24] shown here,
[00:01:36.16] results in aneuploidy,
[00:01:38.26] as I just showed you on the previous slide,
[00:01:40.10] which is a common hallmark of cancer,
[00:01:42.03] and I should mention it's a common hallmark of other diseases,
[00:01:44.10] such as birth defects.
[00:01:45.28] And I want to remind you that you have
[00:01:47.25] greater than a trillion cells in your body,
[00:01:49.24] so this process has to happen correctly
[00:01:52.05] for every pair of chromosomes in every cell
[00:01:54.15] every time they divide.
[00:01:56.21] So, mitosis was first observed
[00:01:58.24] in the late 1800s by Walther Flemming,
[00:02:01.10] and these are just drawings he did.
[00:02:03.01] What he did is he stained
[00:02:05.05] grasshopper cells or salamander cells
[00:02:07.15] with a DNA dye,
[00:02:09.05] and then he was able to deduce that
[00:02:11.11] during the cell cycle chromosomes duplicate and partition.
[00:02:14.09] And you can see here that they partition...
[00:02:16.12] he even drew in fibers that we now know are microtubules.
[00:02:19.16] Due to major advances in cell biology and microscopy,
[00:02:23.12] we can actually visualize this process in real time.
[00:02:26.21] I'm going to show you two movies
[00:02:28.15] from Geert Kops' lab,
[00:02:29.27] where he visualizes mitosis.
[00:02:32.12] What he did, or what their lab did,
[00:02:34.25] was they expressed
[00:02:37.23] a fluorescent fusion to a histone
[00:02:39.26] so that they could visualize the chromosomes,
[00:02:42.13] and then they marked the microtubules
[00:02:44.13] with a fluorescently tagged component
[00:02:46.23] of the microtubules.
[00:02:48.06] So, I'm going to start the movie
[00:02:50.14] and what you're watching here are the microtubules
[00:02:53.05] forming a spindle,
[00:02:54.24] and those microtubules are capturing the chromosomes,
[00:02:57.05] aligning them, and, there,
[00:02:59.06] they're pulling them to opposite poles.
[00:03:01.11] Now, this is a really dynamic and complicated process,
[00:03:04.12] and it turns out it's actually also
[00:03:06.21] very error-prone.
[00:03:08.07] I'm going to show you an example of that in the next movie.
[00:03:11.29] Here, most of the chromosomes get properly attached
[00:03:16.18] to microtubules,
[00:03:18.11] but in the upper-left you can see an entire set of chromosomes
[00:03:21.18] doesn't get captured,
[00:03:23.22] and therefore it doesn't segregate to opposite poles,
[00:03:25.20] and that generates the aneuploidy
[00:03:27.18] that I just told you is a hallmark of many diseases.
[00:03:31.14] So, let's break down the events of chromosome segregation.
[00:03:34.12] As I said, during S phase
[00:03:36.13] chromosomes replicate to make sister chromatids,
[00:03:38.20] shown here,
[00:03:40.02] and then a complex called cohesin
[00:03:42.06] physically links those sister chromatids together,
[00:03:44.22] and that's important so that the cell
[00:03:46.29] can distinguish pairs of chromosomes.
[00:03:49.16] At metaphase, a bipolar spindle forms,
[00:03:52.13] and this is due to the microtubules
[00:03:54.23] that emanate from the spindle poles,
[00:03:56.16] and some of those capture
[00:03:58.25] a locus on the chromosome called the kinetochore,
[00:04:02.06] and we're going to talk about the kinetochore in depth today.
[00:04:06.03] Once every pair of chromosomes
[00:04:08.15] is properly captured by microtubules,
[00:04:10.17] a complex called the APC,
[00:04:12.05] or anaphase promoting complex, is activated,
[00:04:14.24] and it cleaves cohesin so that the linkage is destroyed,
[00:04:18.10] and that allows the microtubules to
[00:04:20.04] physically pull those chromosomes
[00:04:22.15] to opposite poles at anaphase.
[00:04:24.29] Now, the kinetochore, as I said,
[00:04:27.01] is the structure that directs segregation.
[00:04:29.13] This is a really enormous protein complex
[00:04:31.23] that assembles on centromeric DNA,
[00:04:34.17] and so this is just a cartoon here
[00:04:36.25] so that you can see the kinetochore.
[00:04:38.16] The centromere I'm going to define as
[00:04:40.19] the DNA on the chromosome
[00:04:42.14] where the kinetochore forms,
[00:04:44.02] and in this image you can see that
[00:04:46.12] the inner centromere is the underlying DNA
[00:04:49.14] and then there's an inner kinetochore,
[00:04:51.16] which are the proteins that directly bind to the centromeric DNA,
[00:04:54.22] and then there's a whole 'nother layer of proteins
[00:04:56.27] that assemble onto that
[00:04:58.24] and they form the outer kinetochore.
[00:05:00.10] And that is the microtubule attachment site.
[00:05:02.20] And we now know that the human kinetochore
[00:05:05.00] has more than 100 unique components,
[00:05:07.15] so it's a really complicated structure
[00:05:09.17] and most of these proteins are present
[00:05:11.19] in multiple copies,
[00:05:13.03] so there's hundreds and hundreds of proteins
[00:05:15.08] in just a single kinetochore,
[00:05:17.05] one of the most complicated protein structures in the cell.
[00:05:21.07] The centromere is defined as
[00:05:23.01] the chromosomal site of kinetochore assembly, as I said,
[00:05:26.22] and you can see it, it's the primary constriction in these images
[00:05:28.29] that you can see here.
[00:05:30.14] When we look at this by electron microscopy,
[00:05:33.06] the underlying chromatin is this
[00:05:36.13] sort of fibrous mat underneath,
[00:05:37.29] and then you can see the structure on top of that,
[00:05:39.26] where the white arrows are, is the kinetochore,
[00:05:42.10] and then you can actually see the microtubules
[00:05:44.10] are attaching to that kinetochore.
[00:05:47.02] And I think you can appreciate that
[00:05:49.04] it's very difficult to really understand
[00:05:51.04] the structure of this complex protein structure
[00:05:53.07] from looking at EM (electron microscopy) images in cells.
[00:05:57.08] There's three types of centromeres
[00:05:59.09] -- a point centromere, a regional, and a holocentric --
[00:06:02.03] and they're defined by the number of microtubules
[00:06:04.12] that the kinetochore can bind.
[00:06:06.15] So, point centromeres bind to just a single microtubule,
[00:06:09.13] regional kinetochores are the most common,
[00:06:11.18] and they consist of multiple microtubule binding sites
[00:06:14.09] in a single kinetochore on the centromere,
[00:06:17.10] and then holocentric chromosomes are really interesting
[00:06:21.18] -- they're microtubule attachment sites over the entire chromosome,
[00:06:24.14] and that's common in insects and worms.
[00:06:27.28] Now, the centromeres vary a lot in size
[00:06:30.03] and sequence as well,
[00:06:31.27] and this is just representative centromeres
[00:06:34.09] from a number of organisms
[00:06:36.14] to make this point.
[00:06:37.23] So, in budding yeast, at the top,
[00:06:39.11] there's a 125 base pair centromere,
[00:06:42.05] just 125 base pairs is sufficient
[00:06:44.24] to ensure that that's a centromere.
[00:06:47.17] And this is really different in other organisms,
[00:06:49.15] such as humans and Arabidopsis,
[00:06:51.13] that have megabases of DNA
[00:06:53.14] that form the centromere.
[00:06:55.00] They also vary a lot in sequence.
[00:06:57.07] Budding yeast is exceptional
[00:06:59.07] and it has a sequence-specific centromere
[00:07:01.09] that's sufficient to drive kinetochore formation.
[00:07:04.17] However, in most organisms,
[00:07:06.10] such as the rest of them shown here,
[00:07:08.03] the hallmark of the centromere
[00:07:09.24] is really just repetitive DNA,
[00:07:11.20] and so in humans it's just
[00:07:13.26] repeats of alpha-satellite arrays.
[00:07:16.26] So, the underlying sequence is highly repetitive
[00:07:19.17] and it's usually AT-rich
[00:07:21.14] and, because of this,
[00:07:23.06] the last region of the genome
[00:07:25.02] is usually the centromere to be sequenced,
[00:07:27.08] because it's such highly repetitive DNA.
[00:07:30.10] Now, because there's really no sequence specificity
[00:07:33.11] to centromeres other than budding yeast,
[00:07:35.01] it turns out that centromeres are normally
[00:07:37.08] epigenetically defined or specified.
[00:07:40.04] So, the site on the chromosome
[00:07:41.20] that's going to become the kinetochore
[00:07:43.19] is defined epigenetically
[00:07:46.03] and the most common hallmark is this nucleosome
[00:07:47.27] called CENP-A.
[00:07:49.12] So, as you know, the chromosome
[00:07:51.02] consists of chromatin,
[00:07:52.22] which is built up of nucleosomes.
[00:07:54.24] Normally, those nucleosomes
[00:07:57.17] contain H2-A, H2-B, H3, and H4.
[00:07:59.25] However, at the centromere
[00:08:01.25] there's an interesting nucleosome called CENP-A,
[00:08:03.23] where the two copies of H3
[00:08:06.08] are replaced with this histone variant
[00:08:08.08] called CENP-A.
[00:08:09.17] And so what this means is that at the centromere
[00:08:12.02] there's arrays of CENP-A nucleosomes
[00:08:14.11] interspersed with H3 nucleosomes.
[00:08:17.10] And this histone variant
[00:08:20.04] is only present at centromeres,
[00:08:22.18] so these nucleosomes are exclusively at the centromere
[00:08:25.10] and are clearly the epigenetic mark.
[00:08:29.23] Another hallmark of centromeres is that
[00:08:32.04] they're embedded in heterochromatin.
[00:08:34.04] In addition, there are specific modifications
[00:08:36.18] to the histones in these centromeres,
[00:08:38.09] so they have a lot of unique features,
[00:08:40.07] and the field is still trying to understand
[00:08:42.22] how this histone variant is deposited
[00:08:44.28] specifically at the centromere
[00:08:46.21] and how that's controlled,
[00:08:48.07] and then what are the mechanisms that make sure that
[00:08:50.16] it doesn't accidentally go into the rest of the chromosome,
[00:08:52.16] where it could form additional kinetochores.
[00:08:56.18] Okay, so let's turn to the kinetochore now.
[00:08:58.21] The first kinetochore proteins
[00:09:00.27] were identified by exploiting the fact that
[00:09:04.13] it turned out that patients
[00:09:06.09] who have some autoimmune diseases
[00:09:08.04] have antibodies that recognize the centromere
[00:09:10.20] or kinetochore locus.
[00:09:12.01] And you can see that in this image from Bill Earnshaw.
[00:09:14.21] In the top image of a cell,
[00:09:16.24] an immunofluorescence image,
[00:09:18.06] you can see tubulin, shown in blue,
[00:09:20.22] and at the tip of those kinetochore microtubules,
[00:09:23.19] you can see the red dots
[00:09:25.04] that are the antibodies from these patients,
[00:09:27.08] and that's called ACA -- anti-centromere antibodies.
[00:09:30.23] So, Earnshaw and, independently,
[00:09:33.26] another group, Goldner,
[00:09:35.11] realized that they could take advantage of the fact
[00:09:37.20] that these antibodies recognized kinetochores
[00:09:39.21] to clone the first kinetochore proteins.
[00:09:41.10] And actually, in this image below,
[00:09:42.28] you can see CENP-A staining
[00:09:44.21] -- it turned out that CENP-A,
[00:09:46.15] that I just talked about,
[00:09:48.08] was the first kinetochore protein that Earnshaw identified.
[00:09:51.09] Over the years, biochemistry and genetics
[00:09:54.01] have really identified the bulk of kinetochore proteins,
[00:09:57.06] and these are immunoprecipitations from Ian Cheeseman,
[00:10:01.00] where... you don't need to worry about what all these complexes are
[00:10:03.25] or what all these proteins are,
[00:10:05.12] the point I'm trying to make is one of the main approaches
[00:10:07.24] to identifying kinetochore proteins
[00:10:09.25] was to take one kinetochore protein
[00:10:12.01] and immunoprecipitate it to identify its co-purifying proteins.
[00:10:16.07] And over the years we learned that
[00:10:18.26] one of the hallmarks of the kinetochore
[00:10:20.18] is that it can be broken down into subcomplexes,
[00:10:23.02] stable subcomplexes that can be purified on their own
[00:10:26.08] and reconstituted on their own.
[00:10:28.08] In addition, I should comment that genetics, of course,
[00:10:31.26] contributed heavily to identifying these proteins as well.
[00:10:34.10] Model organisms, such as budding yeast,
[00:10:37.08] fission yeast,
[00:10:38.26] Drosophila, worms...
[00:10:40.05] people did genetic screens to look for things like
[00:10:42.10] chromosome loss mutants,
[00:10:44.00] and those screens additionally identified
[00:10:46.10] many kinetochore proteins.
[00:10:47.28] So, what we know now is that there are these...
[00:10:51.05] there are many subcomplexes
[00:10:52.27] and I won't talk about all of them,
[00:10:54.23] and the basic idea is that they form
[00:10:56.17] a hierarchical kinetochore structure,
[00:10:58.22] and that's just cartooned here.
[00:11:01.27] There are many, many copies of these subcomplexes,
[00:11:04.19] and so in this cartoon this is greatly simplified,
[00:11:07.10] but the things to remember are that,
[00:11:09.29] as I said, there's a centromeric chromatin structure
[00:11:12.12] at the bottom that's the foundation for the kinetochore
[00:11:15.15] and a hallmark is the CENP-A nucleosome,
[00:11:18.04] the inner kinetochore forms on top of that
[00:11:20.09] and this is often called the CCAN,
[00:11:22.07] for constitutive centromere-associated network.
[00:11:24.24] These are proteins that then bind to
[00:11:27.09] that centromeric chromatin
[00:11:28.27] and form the foundation of the kinetochore.
[00:11:30.21] They're present throughout the cell cycle.
[00:11:32.25] And then at mitosis a number of outer kinetochore proteins
[00:11:35.22] assemble onto that base,
[00:11:38.04] and as I said there's many subcomplexes,
[00:11:40.07] some of which directly bind to the microtubule.
[00:11:44.08] Okay, so now that I've told you about the structure of the kinetochore,
[00:11:46.23] let's talk about its function.
[00:11:48.18] And there's three major functions
[00:11:51.05] that kinetochores carry out:
[00:11:52.14] one is a mechanical role,
[00:11:53.24] where they maintain an attachment
[00:11:56.02] to the tip of the dynamic microtubule;
[00:11:58.15] a second is that they have
[00:12:00.13] interesting biophysical properties
[00:12:02.18] -- tension can stabilize these attachments,
[00:12:04.19] as we'll talk about that;
[00:12:06.09] and finally, they serve as signaling hubs
[00:12:08.09] -- kinetochores can completely halt the cell cycle
[00:12:11.08] if there's a defect in kinetochore microtubule attachments.
[00:12:14.05] And so I'm going to talk about each of these functions.
[00:12:16.22] Let's start with the mechanical role.
[00:12:20.01] So, to remind you,
[00:12:21.24] microtubules are the polymers that segregate the chromosomes,
[00:12:23.27] and these are hollow protofilaments
[00:12:25.27] that are assembled from alpha- and beta-tubulin dimers,
[00:12:29.14] and they exist in one of the four states shown here.
[00:12:32.05] They're either assembling due to the addition of tubulin subunits at the tip,
[00:12:36.09] or they're disassembly due to their removal,
[00:12:38.15] and when they switch from assembling to disassembling
[00:12:40.13] it's called a catastrophe event,
[00:12:42.13] and when they switch from disassembling back to assembling
[00:12:44.27] it's called a rescue event.
[00:12:46.10] And this process where microtubules in a population
[00:12:48.23] are constantly growing and shrinking
[00:12:50.16] is termed dynamic instability,
[00:12:51.26] and was originally discovered
[00:12:54.01] by Mitchison and Kirschner.
[00:12:55.16] Okay, so the kinetochore has a hard job:
[00:12:57.19] it has to stay attached to these microtubule tips.
[00:13:02.01] Kinetochores have to make load-bearing attachments.
[00:13:04.02] Now, the microtubules have an inherent polarity
[00:13:06.19] -- the - end is embedded in the poles
[00:13:08.11] and the + end is defined as the distal end.
[00:13:11.02] So, remember, microtubules
[00:13:13.05] are growing and shrinking in the population,
[00:13:14.27] and at some point
[00:13:17.02] a microtubule will capture a kinetochore.
[00:13:20.01] And most of the time this is a lateral attachment
[00:13:21.24] where the kinetochore initially is on the side of the microtubule.
[00:13:24.29] Now, if things stayed like this,
[00:13:27.02] this would not result in proper chromosome segregation.
[00:13:29.22] Instead, it has to get converted
[00:13:32.29] to a tip attachment, shown here,
[00:13:34.20] where the kinetochores attach to microtubules
[00:13:37.15] from opposite poles.
[00:13:38.25] And remember, I told you
[00:13:41.02] that thousands of tubulin subunits
[00:13:43.10] are being added and lost from this tip
[00:13:44.26] while that kinetochore has to stay bound,
[00:13:46.16] so it's almost like you're trying to climb a rope
[00:13:48.18] and someone is constantly pulling the rope out from under you.
[00:13:50.29] So the kinetochore has to stay bound
[00:13:52.17] to this really dynamic tip,
[00:13:54.00] and so a huge question is, how can that happen?
[00:13:56.27] So, over the years, we've identified proteins
[00:13:59.16] within the kinetochore
[00:14:01.02] that bind to the microtubule,
[00:14:02.17] and I'm only going to talk about one of those today,
[00:14:04.04] which is Ndc80,
[00:14:05.15] because it's really considered to be
[00:14:07.00] the major microtubule binding activity
[00:14:08.22] within the kinetochore.
[00:14:10.07] This is the structure of a domain
[00:14:12.05] of the Ndc80 protein itself,
[00:14:13.26] which was originally solved by the Harrison lab,
[00:14:17.01] and you can see it's a CH domain,
[00:14:19.04] which stands for calponin homology,
[00:14:21.14] and this is commonly found in other microtubule binding proteins,
[00:14:26.06] such as EB1.
[00:14:27.06] Now, Andrea Musacchio's lab solved an engineered structure...
[00:14:30.11] Ndc80 is a four-protein complex,
[00:14:32.21] they made this engineered complex called the "bonsai"
[00:14:35.12] and discovered that there's actually two calponin homology domains,
[00:14:38.02] one in Ndc80 and one in Nuf2,
[00:14:40.06] and then over the years we've realized that
[00:14:43.06] Ndc80 has the major microtubule binding activity.
[00:14:45.15] This is higher structural information
[00:14:49.10] from Eva Nogales' lab.
[00:14:50.24] The microtubule is shown in green
[00:14:52.21] and you can see Ndc80 is shown in blue,
[00:14:55.27] and you can see the calponin homology domain
[00:14:58.05] directly binding to that microtubule.
[00:15:00.07] The Nuf2 is shown in yellow
[00:15:02.10] and you can see that it doesn't directly bind,
[00:15:04.07] and I want to point out
[00:15:07.06] the red is another part of Ndc80,
[00:15:08.27] which is an N-terminal extension
[00:15:10.18] that also has microtubule binding activity.
[00:15:13.05] So, this is the complex that's really considered to be
[00:15:16.00] the major microtubule binding activity,
[00:15:17.14] however, I want to emphasize,
[00:15:19.09] there really are other microtubule binding proteins in the kinetochore
[00:15:21.21] and we're still trying to understand
[00:15:24.03] how they all act together and what their relative contributions are.
[00:15:28.01] So now, let's talk about the mechanism
[00:15:29.26] that the kinetochore can use to
[00:15:31.25] actually maintain an attachment to these dynamic tips.
[00:15:34.17] So, there are two major models right now.
[00:15:36.20] The first is the conformational wave model,
[00:15:38.22] and the idea here is that the microtubule
[00:15:41.23] stores a lot of energy
[00:15:43.11] and when it switches from assembling into disassembling,
[00:15:45.20] the tip structure peels back,
[00:15:48.04] the protofilament curves back,
[00:15:50.21] and as long as the kinetochore can maintain an attachment
[00:15:53.25] to this curved protofilament,
[00:15:55.15] it can harness the disassembling energy
[00:15:58.09] of the microtubule to move the chromosome.
[00:16:00.11] And there's two ways that you can imagine that
[00:16:02.25] the chromosome can stay attached:
[00:16:04.16] one is a ring model
[00:16:06.23] where part of the kinetochore
[00:16:08.08] forms a ring around the microtubule,
[00:16:09.22] and as long...
[00:16:11.11] and there's evidence in budding yeast that the Dam1 complex
[00:16:13.03] does form a ring;
[00:16:14.16] there's also a fibril-based mechanism,
[00:16:16.00] where individual components of the kinetochore
[00:16:17.26] can attach to the microtubule tip
[00:16:19.25] and then maintain an attachment
[00:16:22.10] through this conformational wave.
[00:16:23.27] Now, Hill proposed a different model
[00:16:27.16] called biased diffusion.
[00:16:28.21] In this model, the idea is that the kinetochore
[00:16:30.18] contains multiple weak microtubule binding elements
[00:16:32.16] that rapidly bind and unbind to the microtubule tip,
[00:16:36.05] and as long as this diffusion
[00:16:38.05] is biased towards the tip,
[00:16:39.29] which can occur due to thermal fluctuations
[00:16:41.29] of the chromosome binding near the tip,
[00:16:44.12] that then the kinetochore would stay attached
[00:16:46.09] to the growing and the shrinking tip.
[00:16:48.19] And there's evidence that this model might be true
[00:16:51.21] because the Asbury and Davis lab
[00:16:53.19] showed that the Ndc80 complex itself
[00:16:55.11] can undergo biased diffusion.
[00:16:57.11] Finally, there's hybrid models
[00:16:59.05] that would take both of these into account,
[00:17:01.00] and so we're still actively trying to understand
[00:17:02.24] what the underlying mechanism for attachment really is.
[00:17:06.28] Okay, so now let's talk about
[00:17:08.22] the interesting biophysical properties
[00:17:10.14] of the kinetochore,
[00:17:11.28] where tension can stabilize, rather than destabilize,
[00:17:14.02] an attachment.
[00:17:16.07] Kinetochores have to make bioriented attachments,
[00:17:19.01] so each sister kinetochore
[00:17:20.22] has to bind to microtubules from opposite poles.
[00:17:22.15] However, as I said at the beginning,
[00:17:24.09] the process of making attachments
[00:17:26.03] is actually very error-prone
[00:17:28.13] and there's a number of incorrect attachments that can be made,
[00:17:30.21] such as mono-oriented attachments,
[00:17:33.09] where one or both sister kinetochores
[00:17:35.11] attach to microtubules from a single pole,
[00:17:37.09] there's even merotelic, shown here on the bottom,
[00:17:40.07] where one kinetochore can attach to microtubules from both poles.
[00:17:43.14] And it's essential that the cell can detect
[00:17:46.04] and correct these errors.
[00:17:48.02] And so how can cells detect and correct these errors?
[00:17:50.19] It turns out that tension is the key.
[00:17:53.11] Now, Dietz originally noticed that
[00:17:56.18] chromosomes would continually reorient on the spindle
[00:17:59.04] until they became stably bioriented,
[00:18:02.02] and then Bruce Nicklas really did
[00:18:04.03] the pioneering experiment
[00:18:05.25] to show that that's because tension is stabilizing the attachments.
[00:18:08.19] So, what he did is take advantage of the fact that
[00:18:12.19] grasshopper spindles are really large
[00:18:14.07] and that you can use micromanipulation techniques.
[00:18:17.04] So, he used meiosis I spindles,
[00:18:18.26] where now the homologues
[00:18:20.29] are attached to opposite poles,
[00:18:22.06] rather than the sister chromatids,
[00:18:23.29] and he took a micromanipulator
[00:18:27.12] and was able to convert a bioriented attachment,
[00:18:31.03] where the homologues are attached to opposite poles,
[00:18:33.20] into one where the homologues
[00:18:35.23] are attached to the same pole.
[00:18:37.10] And he noticed that this is highly unstable
[00:18:39.16] and what would happen, then,
[00:18:41.01] is that the kinetochore would detach.
[00:18:43.04] And there were one of two fates:
[00:18:45.06] if it went back and created the wrong attachment again,
[00:18:47.18] that was unstable;
[00:18:49.23] however, if it created the correct attachment,
[00:18:51.27] shown in the top,
[00:18:53.10] it was now stabilized.
[00:18:54.21] And then to demonstrate that this is because of tension,
[00:18:57.16] he actually used a needle
[00:18:59.23] to artificially apply an opposing force.
[00:19:03.12] So, he took that incorrect attachment
[00:19:05.28] where both homologues are attached to the same pole,
[00:19:08.24] applied an opposing force,
[00:19:10.25] and now the incorrect attachment was stable.
[00:19:13.17] And this really demonstrated
[00:19:16.06] that that's because tension is stabilizing that attachment.
[00:19:18.21] So, tension stabilizes attachments
[00:19:22.02] and they're in vivo very stable
[00:19:24.11] once they come under tension,
[00:19:25.14] and in contrast any attachment
[00:19:27.22] that's lacking tension is highly unstable.
[00:19:29.19] So, how can tension stabilize attachments?
[00:19:32.05] And we know that there's at least two mechanisms
[00:19:34.05] that contribute to this.
[00:19:35.24] There's work from a number of labs
[00:19:37.24] that have shown there's an error correction pathway
[00:19:39.25] that uses phosphorylation
[00:19:42.20] by the Aurora B protein kinase.
[00:19:44.12] So, the idea here is that
[00:19:46.21] when there's an incorrect attachment that lacks tension,
[00:19:48.15] the Aurora B kinase recognizes that
[00:19:50.24] and can phosphorylate kinetochore proteins,
[00:19:53.17] and I want to mention that Ndc80
[00:19:55.19] is one of the major targets of the Aurora B kinase.
[00:19:58.06] When these proteins are phosphorylated,
[00:20:00.22] it destabilizes that incorrect attachment,
[00:20:03.24] and then that gives the cell a chance
[00:20:05.25] to go back and make that correct attachment again.
[00:20:08.04] Now, there's a second mechanism,
[00:20:10.03] and that's one where tension
[00:20:11.26] directly stabilizes attachments,
[00:20:13.12] and the idea here is that tension itself
[00:20:15.29] promotes more microtubule binding elements
[00:20:18.04] to bind to the tip of the microtubule.
[00:20:19.16] And this is like those finger traps,
[00:20:21.18] if you remember,
[00:20:22.26] where as you pull on them things get even tighter.
[00:20:25.02] And this is really just an intrinsic property
[00:20:27.24] of the kinetochore-microtubule interaction
[00:20:29.08] -- there is no Aurora B involved in this --
[00:20:31.18] and in Part 2 of my talk
[00:20:33.08] I'll tell you how we discovered this mechanism.
[00:20:36.20] Now, let's turn to the signaling functions of the kinetochore
[00:20:41.01] and, as I said, the kinetochore
[00:20:43.10] can transmit a signal to completely halt the cell cycle.
[00:20:46.15] Cell cycle progression has to be halted
[00:20:49.17] when kinetochore-microtubule interactions are incorrect,
[00:20:52.14] so when there's a lack of tension
[00:20:53.27] or an unattached kinetochore,
[00:20:55.27] the checkpoint has to recognize that and halt the cell cycle.
[00:20:58.14] And the spindle assembly checkpoint
[00:21:00.01] -- I'm going to call it the spindle checkpoint, for short --
[00:21:02.07] is the signal transduction system
[00:21:04.19] that can do that.
[00:21:06.09] Genetic screens from Andy Hoyt
[00:21:08.26] and Andrew Murray's labs
[00:21:10.11] originally identified the checkpoint genes,
[00:21:12.16] and the idea here is they used budding yeast
[00:21:15.01] as a model organism.
[00:21:16.09] As you know, the yeast will divide
[00:21:18.07] and make a colony when they're grown under normal conditions.
[00:21:21.17] If you add a microtubule depolymerizing drug,
[00:21:23.20] such as benomyl,
[00:21:25.19] this creates a difficulty for the cell
[00:21:28.02] in making kinetochore-microtubule attachments,
[00:21:30.08] because now the microtubules are less stable.
[00:21:33.00] And then there's a mitotic delay
[00:21:35.18] while the cell tries to make the proper attachments,
[00:21:38.02] and that's due to the spindle checkpoint.
[00:21:39.27] So, the Hoyt and the Murray labs
[00:21:41.28] reasoned that mutants in the checkpoint
[00:21:45.00] would not be able to delay,
[00:21:46.23] would therefore go through the cell cycle with incorrect attachments,
[00:21:49.10] and that would lead to cell death.
[00:21:51.28] So mad2 is one of the checkpoint mutants,
[00:21:54.04] and when they treated that with benomyl
[00:21:56.06] there was no delay, and then they just
[00:21:58.14] get a microcolony of dead cells.
[00:22:00.26] So by looking for mutants that cannot grow on benomyl,
[00:22:03.00] they were able to isolate, really,
[00:22:05.04] the majority of spindle checkpoint genes.
[00:22:07.18] We now know those spindle checkpoint genes
[00:22:09.19] that were originally identified in budding yeast
[00:22:11.11] are conserved,
[00:22:12.28] and there are additional ones that they didn't identify,
[00:22:14.22] but the bulk really were identified in these two classic screens
[00:22:17.28] in the early '90s.
[00:22:20.16] Okay, so where is checkpoint signal generated?
[00:22:23.17] The first homolog of the yeast genes identified
[00:22:26.02] was the Xenopus MAD2 gene,
[00:22:28.08] and the Murray lab showed that
[00:22:31.02] it's required for the checkpoint using frog egg extracts,
[00:22:33.01] and then the made the striking finding that
[00:22:35.21] Mad2 binds specifically to unattached kinetochores.
[00:22:38.15] So you can see that in this image, here,
[00:22:40.07] from their paper.
[00:22:42.03] This is a newt lung cell,
[00:22:44.10] and this was a collaboration with Ted Salmon's lab,
[00:22:48.04] and you can see on the left the chromosomes that...
[00:22:51.01] most of them have aligned at the metaphase plate
[00:22:53.09] and are properly attached to microtubules,
[00:22:55.24] and then on the right is
[00:22:57.29] immunofluorescence against Mad2,
[00:23:00.03] and you can see there's almost no Mad2 staining.
[00:23:02.11] So the kinetochores that are properly attached
[00:23:05.00] do not have any Mad2.
[00:23:07.07] But if you look on the right-hand side...
[00:23:09.21] or, on the left-hand side,
[00:23:11.14] you'll see a mono-oriented chromosome.
[00:23:14.03] One kinetochore is attached to the pole
[00:23:16.04] and the other one is not,
[00:23:17.14] and then if you look at the staining of Mad 2,
[00:23:19.29] you can see if specifically
[00:23:22.25] stains the unattached kinetochore
[00:23:24.18] in that pair of chromosomes.
[00:23:26.17] So this really was the evidence that Mad2
[00:23:28.17] specifically binds to unattached kinetochores,
[00:23:30.20] and over the years a lot of work
[00:23:32.16] has shown that kinetochores
[00:23:34.12] clearly generate the checkpoint signal.
[00:23:35.27] And we now know that there's a phosphorylation cascade
[00:23:38.03] that recruits checkpoint proteins
[00:23:40.13] to unattached kinetochores to trigger the checkpoint,
[00:23:42.22] and I'm not going to talk about all those details.
[00:23:45.17] Now, what does the checkpoint monitor?
[00:23:47.05] I've already told you that defects in tension
[00:23:49.13] or attachment
[00:23:50.29] will lead to aneuploidy and chromosome missegregation,
[00:23:53.19] so it's critical that both of these defects are monitored.
[00:23:57.24] And it's been really controversial
[00:23:59.15] whether tension or attachment
[00:24:01.10] is the key to triggering the checkpoint.
[00:24:04.06] Work from Conly Rieder suggested that attachment is monitored.
[00:24:08.11] So, he did a classic experiment
[00:24:10.11] where there was a cell that had a mono-oriented,
[00:24:12.07] unattached chromosome,
[00:24:14.20] so there's a single kinetochore attached to the microtubules
[00:24:17.08] and the other is unattached.
[00:24:19.20] He went...
[00:24:21.01] and those cells are arrested because the checkpoint is on
[00:24:23.05] because there's a defect.
[00:24:25.02] Now, he went in and he laser ablated
[00:24:27.23] that unattached kinetochore,
[00:24:30.00] and that satisfied the checkpoint
[00:24:32.03] and the cell cycle continued,
[00:24:33.17] even though the remaining kinetochore is attached
[00:24:35.26] and not under tension.
[00:24:37.14] And this was the work that really suggested that
[00:24:39.15] unattached kinetochores were what the checkpoint was monitoring.
[00:24:42.17] However, remember that classic experiment
[00:24:45.12] I told you about from Bruce Nicklas,
[00:24:47.15] where there was a mono-oriented chromosome attached...
[00:24:50.26] where both sister kinetochores are attached to the same pole,
[00:24:53.21] and in that situation,
[00:24:55.09] where there's no tension but there is attachment,
[00:24:57.06] it turns out that the cell cycle was stopped
[00:24:59.27] due to the checkpoint.
[00:25:01.19] And when he went in and micromanipulated
[00:25:05.01] and applied the opposing tension,
[00:25:07.02] it turned out that the cell cycle went ahead,
[00:25:09.00] even though the kinetochores
[00:25:10.29] were attached to the wrong pole.
[00:25:13.27] And so this showed that tension
[00:25:15.27] seemed to be sufficient
[00:25:17.11] to tell the cell that everything was ready to go,
[00:25:19.02] even though those attachments were incorrect.
[00:25:21.16] So it's still really confusing
[00:25:23.18] whether tension or attachment
[00:25:25.05] is the primary signal,
[00:25:26.14] and I just want to remind you that defects in tension
[00:25:28.13] are converted to unattached kinetochores.
[00:25:30.10] So remember, when kinetochores lack tension,
[00:25:32.22] these attachments are released
[00:25:34.17] due to error correction mechanisms,
[00:25:36.01] and that generates unattached kinetochores.
[00:25:38.05] So, the reason it's been so hard to understand
[00:25:40.24] might just be because these two issues
[00:25:43.04] are so interrelated.
[00:25:45.21] Now, let's talk about how the checkpoint halts the cell cycle.
[00:25:48.27] A clue to this came from
[00:25:51.14] work from Ted Salmon's lab
[00:25:53.08] that looked at the Mad2 protein
[00:25:54.19] and found that it dynamically cycles on and off
[00:25:56.12] the kinetochore.
[00:25:57.14] So, what they did here is they took
[00:26:00.17] cells expressing a Mad2-GFP fusion
[00:26:02.17] so that the kinetochores...
[00:26:04.06] they could see the kinetochores...
[00:26:05.22] Mad2 on the kinetochores,
[00:26:07.02] and they treated the cells with nocodazole
[00:26:08.18] to generate lots of unattached kinetochores,
[00:26:10.06] which is why all of these kinetochores
[00:26:12.09] are lighting up with Mad2.
[00:26:14.04] They then did what we call a photobleaching experiment,
[00:26:16.14] where they ablated,
[00:26:18.20] in a single kinetochore, shown there,
[00:26:20.08] the Mad2 fluorescence,
[00:26:23.01] and then they watched,
[00:26:25.00] how long does it take for new Mad2
[00:26:27.06] to bind to the kinetochore?
[00:26:28.18] And what they learned from this experiment is that
[00:26:30.22] a population of Mad2 rebinds within 25 seconds,
[00:26:34.07] which is extremely rapid,
[00:26:35.28] and I want to point out,
[00:26:37.21] there was also a more stably bound population as well.
[00:26:41.20] So, this dynamic localization
[00:26:44.05] of Mad2 suggested that, perhaps,
[00:26:46.07] it forms an inhibitory complex
[00:26:48.05] when it goes to the kinetochore that can quickly diffuse
[00:26:50.04] and then generate a cell cycle arrest.
[00:26:53.26] We now know that the APC
[00:26:55.25] is the target of the checkpoint.
[00:26:57.13] I told you earlier that
[00:26:59.07] this anaphase promoting complex
[00:27:00.20] promotes the metaphase-to-anaphase transition,
[00:27:03.09] and it turns out there's an activator called Cdc20
[00:27:06.02] that triggers APC activity.
[00:27:08.24] The spindle checkpoint directly inhibits this protein.
[00:27:11.29] Mad2 binds to Cdc20,
[00:27:15.15] and additional proteins do as well,
[00:27:17.09] and this complex is called the MCC,
[00:27:19.05] or the mitotic checkpoint complex,
[00:27:21.06] and when that complex binds to Cdc20,
[00:27:23.12] it prevents it from activating the APC.
[00:27:26.15] So, Hong Tau Hu in Andrea Musacchio's lab
[00:27:29.06] solved the structure of Mad2
[00:27:30.20] and they found out there's actually two forms.
[00:27:32.00] There's a closed form,
[00:27:33.23] and that directly binds to Cdc20 and inhibits it,
[00:27:36.24] and then there's an open form.
[00:27:38.16] And so the idea here,
[00:27:40.17] in this template model that Andrea proposed,
[00:27:42.04] is that there's a cytoplasmic pool of Mad2
[00:27:44.18] that's in the open form
[00:27:46.11] and then there's actually a closed form of Mad2
[00:27:48.09] sitting on the kinetochore.
[00:27:49.23] When the open form binds to the closed form
[00:27:52.07] on the kinetochore,
[00:27:53.27] it gets converted to the closed form,
[00:27:55.17] that then diffuses away and can bind to Cdc20 to inhibit it.
[00:27:58.28] I just want to summarize the general model
[00:28:01.23] for the checkpoint right now
[00:28:03.13] -- a lot of details are missing,
[00:28:04.24] but you'll get the overall picture.
[00:28:06.02] The idea here is there's a Mad2 complex,
[00:28:08.14] and Mad1 is its receptor.
[00:28:10.01] That complex binds to the kinetochore,
[00:28:12.17] such as an unattached kinetochore,
[00:28:14.29] and then open mad Mad2
[00:28:17.25] binds to the Mad2 that's closed on the kinetochore,
[00:28:20.06] that converts it to closed Mad2,
[00:28:22.24] that closed Mad2 binds to Cdc20,
[00:28:25.13] and then that assembles into a mitotic checkpoint complex
[00:28:29.07] that then can go inhibit the APC.
[00:28:31.20] And so that keeps the cell cycle off
[00:28:33.17] while the proper attachments are made.
[00:28:35.21] Once everything is okay,
[00:28:37.10] that checkpoint complex is relieved
[00:28:39.13] and Cdc20 is then activated
[00:28:42.16] to then go bind to the APC
[00:28:45.03] and promote cell cycle progression.
[00:28:47.01] And so we're still working out a lot of details of this model,
[00:28:49.17] but that gives you a general picture
[00:28:51.06] of how the checkpoint is thought to work.
[00:28:53.22] Okay, so what I've told you about today are
[00:28:56.06] the complex functions of kinetochores.
[00:28:58.03] They play mechanical roles in directly coupling the kinetochore
[00:29:01.00] and chromosome to microtubules
[00:29:02.28] to move the chromosomes during the cell cycle.
[00:29:05.28] We've identified some of the major players
[00:29:08.11] that bind to microtubules, such as Ndc80,
[00:29:10.23] but we're still trying to identify
[00:29:12.22] what other microtubule binding proteins
[00:29:14.12] are in the kinetochore,
[00:29:15.26] what are their relative contributions
[00:29:17.19] to maintaining an attachment,
[00:29:19.14] and, really, what is the underlying mechanism
[00:29:21.09] -- conformational wave, biased diffusion --
[00:29:23.12] by which the kinetochore can maintain an attachment?
[00:29:26.12] Kinetochores have these really interesting biophysical properties
[00:29:29.03] where tension actually stabilizes attachments,
[00:29:31.07] and we know that there's at least two mechanisms
[00:29:33.00] that contribute to that
[00:29:34.21] -- an Aurora B mechanism,
[00:29:36.15] we still don't know how Aurora B
[00:29:39.01] selectively works on kinetochores lacking tension
[00:29:41.03] but doesn't affect those that have tension,
[00:29:43.17] and then as I've said, we've identified a second mechanism,
[00:29:46.09] an intrinsic mechanism,
[00:29:48.08] that we're still trying to understand
[00:29:50.11] the underlying mechanism for.
[00:29:51.28] And then, finally, they can act as signaling hubs,
[00:29:53.20] where a single unattached kinetochore
[00:29:55.11] can completely halt the cell cycle
[00:29:57.07] through the spindle checkpoint.
[00:29:58.22] And we're still trying to understand
[00:30:00.21] what the checkpoint is monitoring,
[00:30:02.24] what state of kinetochore-microtubule attachment,
[00:30:04.29] and then there's a lot of work to be done
[00:30:07.08] to understand how that signal is amplified
[00:30:09.13] to halt the cell cycle
[00:30:11.08] and then, how,
[00:30:13.04] when everything is set up correctly,
[00:30:14.12] is that checkpoint silenced so that then the cell cycle can continue?
[00:30:18.03] So, I'm going to talk, specifically, in my next talk
[00:30:21.05] about the first two functions,
[00:30:23.11] and I thank you for your time.

Part 2: Investigating Kinetochore Function

[00:00:07.17] Hi. I'm Sue Biggins and I'm an investigator
[00:00:09.22] at the Fred Hutchinson Cancer Research Center
[00:00:11.18] in the Division of Basic Sciences
[00:00:13.11] and I'm also an investigator
[00:00:14.21] with the Howard Hughes Medical Institute.
[00:00:16.16] And for part two of my talk,
[00:00:17.26] I'm going to discuss our work
[00:00:19.28] investigating kinetochore function.
[00:00:22.09] Just to remind you,
[00:00:23.24] chromosome segregation is mediated
[00:00:25.22] by kinetochore-microtubule attachments.
[00:00:27.17] At metaphase, microtubules emanating from spindle poles
[00:00:31.10] attach to the chromosomes
[00:00:33.12] through structures called kinetochores
[00:00:34.26] that I discussed in detail in the first part of my talk.
[00:00:38.01] Once every pair of chromosomes is properly aligned,
[00:00:41.14] the cell enters anaphase
[00:00:43.04] and the microtubules physically pull chromosomes
[00:00:44.24] to opposite poles.
[00:00:47.13] Now, as I described in the first part of my talk,
[00:00:49.13] kinetochores carry out really complex functions.
[00:00:52.09] They have a mechanical role,
[00:00:53.25] where the kinetochore has to stay attached
[00:00:55.26] to the dynamic tip of a microtubule
[00:00:57.19] that's constantly growing and shrinking from under it,
[00:01:00.22] they have interesting biophysical properties
[00:01:02.06] where tension stabilizes attachments,
[00:01:05.19] and they serve as signaling hubs,
[00:01:07.17] where a single kinetochore
[00:01:09.19] can completely halt the cell cycle
[00:01:11.16] if there's a defect in kinetochore-microtubule attachments.
[00:01:14.20] And to try to understand more about how kinetochores work,
[00:01:17.25] it turns out there's a number of challenges to studying them.
[00:01:22.08] First, kinetochores contain hundreds of proteins,
[00:01:25.02] and this has made it difficult
[00:01:27.05] to understand what the individual contributions
[00:01:28.21] of proteins are to all of these function,
[00:01:30.25] especially because, often, when you mutate one,
[00:01:33.11] a lot of other kinetochore proteins
[00:01:34.29] are no longer localized to the kinetochore.
[00:01:37.25] Attachment and tension
[00:01:40.03] are completely coupled in vivo:
[00:01:41.29] as soon as proper attachments are made,
[00:01:43.25] they quickly come under tension,
[00:01:45.11] and that's made it difficult to understand
[00:01:47.02] the contributions of attachment and tension.
[00:01:49.20] We can't manipulate tension in vivo,
[00:01:51.28] so it's really difficult to directly probe
[00:01:54.10] and put kinetochores under tension in cells
[00:01:56.13] to ask how that affects their properties.
[00:01:59.21] So, many years ago,
[00:02:01.09] I decided that to really understand
[00:02:03.03] some of the functions of kinetochores,
[00:02:04.18] we should try to reconstitute kinetochore-microtubule interactions
[00:02:07.04] to do direct studies in vitro,
[00:02:09.06] where we could probe them and put them under tension
[00:02:12.01] and ask how they behave.
[00:02:14.24] Before I tell you how we did that,
[00:02:16.08] I want to introduce you to the model organism
[00:02:18.05] we've been using for these studies,
[00:02:19.16] which is the budding yeast kinetochore.
[00:02:21.22] This is just a schematic cartoon
[00:02:23.23] of what the budding yeast kinetochore looks like.
[00:02:26.04] It's the simplest kinetochore,
[00:02:28.16] there's just a single microtubule binding site
[00:02:30.14] that binds to a single centromeric nucleosome
[00:02:34.11] in the kinetochore.
[00:02:35.26] The inner kinetochore contains the proteins
[00:02:37.29] that directly bind to that centromere,
[00:02:40.03] and then the outer kinetochore
[00:02:41.26] are the proteins and complexes
[00:02:44.04] that directly mediate attachment to the microtubule.
[00:02:46.18] So, multiple subcomplexes
[00:02:48.11] link the chromosome to the microtubule
[00:02:50.08] and these subcomplexes
[00:02:52.20] are present in multiple copies,
[00:02:54.00] as I mentioned in my first talk.
[00:02:55.29] Now, despite the relative simplicity
[00:02:58.13] of the budding yeast kinetochore,
[00:03:00.09] the bulk of the functions and subcomplexes
[00:03:02.04] are conserved.
[00:03:05.17] The major microtubule binding activity
[00:03:07.08] in the yeast kinetochore
[00:03:09.00] has been attributed to two complexes,
[00:03:10.12] called Ndc80 and Dam1,
[00:03:11.25] and now I've just simplified the cartoon
[00:03:14.15] and I'm just showing a single Dam1 complex
[00:03:18.01] and two Ndc80 complexes.
[00:03:20.18] In green is the Ndc80 complex.
[00:03:22.28] This is a 4-component complex
[00:03:25.02] and there's about 8 of them
[00:03:27.04] in a single yeast kinetochore.
[00:03:29.02] And mutants in vivo h
[00:03:30.29] ave the most severe attachment defects,
[00:03:32.18] which is one of the reasons this has been considered to be
[00:03:35.27] the major microtubule binding component of the kinetochore.
[00:03:39.11] Now, the Dam1 complex
[00:03:41.02] is shown as this ring, here.
[00:03:42.22] It's a 10 component complex
[00:03:44.14] and we think there's about 16 of them in a kinetochore
[00:03:47.24] that can form a ring around the microtubule,
[00:03:49.14] and that was nicely shown by the Nogales and Harrison labs.
[00:03:53.01] And mutants in vivo are defective in end-on attachments.
[00:03:57.17] So, in thinking about how to reconstitute kinetochores,
[00:04:00.27] the question is,
[00:04:02.08] how can we obtain kinetochores to work with?
[00:04:04.15] And there are two major approaches.
[00:04:06.09] One would be to reconstitute them
[00:04:08.08] using recombinant subcomplexes,
[00:04:10.03] so the idea here would be to
[00:04:12.02] make each of those subcomplexes in bacteria
[00:04:13.18] or in other organisms
[00:04:15.26] and then try to put them together.
[00:04:16.28] And the other obvious approach was
[00:04:19.03] we could try to purify native kinetochores from cells.
[00:04:22.10] And there are issues with both of these approaches.
[00:04:25.04] First, the complexity of the kinetochore...
[00:04:27.10] as I said, it contains many, many subcomplexes,
[00:04:29.21] and even in budding yeast
[00:04:31.18] there's at least 9 subcomplexes
[00:04:33.04] that are present in multiple copies.
[00:04:35.16] So it's really complex.
[00:04:38.02] The template...
[00:04:40.08] as I described in the first part of my talk,
[00:04:41.24] the kinetochore assembles on a specializes chromatin structure
[00:04:44.19] and, at the time,
[00:04:46.05] no one had reconstituted that chromatin structure,
[00:04:48.09] so we didn't know if we really could make the template
[00:04:50.22] for the kinetochore to assemble.
[00:04:52.27] And finally the regulation.
[00:04:54.15] There's really not much know about
[00:04:56.18] any sort of post-translational regulation
[00:04:58.13] or cell cycle requirements
[00:05:00.04] for the assembly process.
[00:05:02.01] So, at the time, reconstituting the kinetochore
[00:05:04.18] seemed like a big challenge and very difficult.
[00:05:07.11] Now, there were also challenges to purifying native kinetochores.
[00:05:10.17] First, purity.
[00:05:12.20] As I mentioned, the kinetochore
[00:05:15.02] is on the chromosome
[00:05:16.22] and it's just a single locus on the chromosome.
[00:05:18.05] So we didn't know if we could come up with
[00:05:20.08] a purification strategy that would be clean enough
[00:05:23.00] and not have contaminants
[00:05:24.28] from other chromosomal proteins.
[00:05:26.19] Abundance was a huge issue.
[00:05:28.13] One kinetochore on a chromosome...
[00:05:30.03] we're completely limited by the number of chromosomes
[00:05:32.15] in a cell.
[00:05:34.03] And then, finally, the size.
[00:05:35.19] A single, even yeast, kinetochore
[00:05:38.15] is estimated to be greater that 5 megaDaltons,
[00:05:40.19] which is bigger than a ribosome,
[00:05:43.13] and so we really didn't know
[00:05:46.08] if we could come up with a purification strategy
[00:05:47.26] that would maintain such a large complex
[00:05:49.20] in its native form.
[00:05:51.28] So, in thinking about these things,
[00:05:54.09] we decided to take the approach of trying to purify kinetochores,
[00:05:56.07] and what I'm going to tell you today is
[00:05:57.28] we were able to achieve that.
[00:05:59.25] Bungo Akiyoshi, a really talented grad student in my lab,
[00:06:02.18] developed a purification technique.
[00:06:04.27] So, it turns out...
[00:06:06.17] this is again a schematic of the yeast kinetochore
[00:06:09.07] -- that he Flag-epitope tagged
[00:06:12.22] one component of a complex called Mis12.
[00:06:15.02] This is a 4 component subcomplex.
[00:06:17.28] When he epitope tagged the Dsn1 protein
[00:06:20.22] and purified it,
[00:06:22.15] he developed conditions where he could purify most of the kinetochore.
[00:06:25.08] And that's shown on this silver stain here.
[00:06:27.19] On the left is a negative control
[00:06:30.25] and on the right is the Dsn1-Flag purification,
[00:06:33.03] and you can see that he purified
[00:06:35.03] many more than just the four components
[00:06:37.19] of this subcomplex
[00:06:39.02] when he purified Dsn1.
[00:06:42.04] So, we did mass spec to determine
[00:06:44.21] what those proteins were in that preparation
[00:06:47.24] and it turns out that mass spec
[00:06:49.20] identified greater than 95% of kinetochore proteins.
[00:06:53.26] So, we had proteins from
[00:06:56.11] every single kinetochore subcomplex
[00:06:58.12] with the exception of the CBF3 complex,
[00:07:00.17] which is a yeast-specific DNA binding complex.
[00:07:03.25] So we had achieved purification
[00:07:05.07] of most of the kinetochore in this single step,
[00:07:07.18] and on that silver stained gel I showed you earlier,
[00:07:11.00] almost all of those proteins are kinetochore proteins
[00:07:13.03] and Bungo showed that by epitope tagging them
[00:07:15.17] and showing that when he repurified the kinetochores,
[00:07:17.26] that the epitope-tagged protein would then move on the silver stained gel.
[00:07:23.03] There's also sub-stoichiometric proteins
[00:07:25.07] that we can see by immunoblotting.
[00:07:27.25] We also have regulatory and spindle checkpoint proteins
[00:07:30.29] present in this prep.
[00:07:32.08] However, it turns out that
[00:07:34.28] we have no Aurora B or motor proteins,
[00:07:37.27] and there's no Aurora B because in budding yeast
[00:07:39.27] CBF3 is the receptor for the Aurora B kinase.
[00:07:44.11] We then performed gel filtration to see
[00:07:47.00] if all those kinetochore proteins
[00:07:49.07] were really present in a single complex.
[00:07:50.27] This is fractions from a gel filtration column
[00:07:53.02] and these are just some representative fractions,
[00:07:56.07] and then what we did is immunoblot with representative antibodies
[00:07:57.29] spanning the inner to the outer kinetochore,
[00:08:00.21] and you can see that we have components
[00:08:02.27] throughout the whole kinetochore
[00:08:04.09] present in a few fractions,
[00:08:05.25] and that told us that we have at least some
[00:08:07.27] full kinetochores in our preps.
[00:08:10.20] Okay, so now once we had purified these things,
[00:08:13.05] the next big questions is,
[00:08:14.26] well, are they functional?
[00:08:16.08] Can they do any of the functions
[00:08:18.08] that kinetochores do in the cell,
[00:08:20.10] outside of the cell?
[00:08:22.08] The first question is, can they just bind to a microtubule?
[00:08:25.19] So, kinetochores initially just bind a
[00:08:27.27] nywhere on the lattice of a microtubule
[00:08:29.09] and that was our first question.
[00:08:31.13] Can they stay bound to a growing and shrinking tip?
[00:08:33.22] If you remember from my first part of my talk,
[00:08:35.21] the kinetochore has to maintain
[00:08:38.02] an attachment to a dynamic tip,
[00:08:39.23] and so that was the second question.
[00:08:41.12] And then, they have to be able to
[00:08:44.10] form load-bearing attachments
[00:08:45.21] where they come under tension,
[00:08:47.00] because kinetochores come under tension in the cell,
[00:08:49.05] and so that was our next question.
[00:08:50.20] We set up a really great collaboration
[00:08:53.13] with Chip Asbury's lab, he's a biophysicist
[00:08:55.11] at the University of Washington,
[00:08:56.28] and with his lab we took biophysical approaches
[00:08:59.23] to address these three questions,
[00:09:01.18] and I'll tell you what we did.
[00:09:03.29] First, we made the kinetochores fluorescent,
[00:09:06.05] so that we could see them under the microscope.
[00:09:08.12] And so you can GFP tag any kinetochore protein,
[00:09:11.20] but for the work I'll show you we GFP tagged
[00:09:13.19] the centromeric histone called CENP-A.
[00:09:16.01] We then purified those kinetochores
[00:09:17.20] so that they were fluorescent green
[00:09:20.06] due to the presence of CENP-A.
[00:09:22.19] We then made microtubules,
[00:09:24.08] and this is easy to do in vitro
[00:09:26.05] with purified tubulin,
[00:09:27.13] and we treated them with taxol,
[00:09:29.11] which is a drug that stabilizes the microtubules,
[00:09:31.21] so these are not dynamic,
[00:09:33.14] and then we used rhodamine as a fluorescent tag
[00:09:36.07] to make them fluorescent
[00:09:37.24] so that we could see them under the microscope.
[00:09:39.15] So, we attached rhodamine-labeled microtubules
[00:09:42.02] to a cover slip
[00:09:43.11] and then we just flowed the purified kinetochores in
[00:09:47.12] and asked, can they bind to the microtubule?
[00:09:49.08] So, this is an image of our work
[00:09:51.01] and you can see, in red are the microtubules
[00:09:53.02] and in green are the kinetochores,
[00:09:54.22] and you can see that the kinetochores
[00:09:56.09] can bind all over the microtubules.
[00:09:58.10] So this confirmed that the kinetochores we purified
[00:10:01.08] are functional to bind to microtubules.
[00:10:04.02] Bungo went through as showed that
[00:10:06.14] the association depends on the microtubule binding proteins in the kinetochore
[00:10:09.05] such as Ndc80,
[00:10:10.20] and he did that by just purifying kinetochores
[00:10:12.16] from various mutants,
[00:10:14.16] where then they were lacking that complex.
[00:10:18.05] Okay, so the next question then is,
[00:10:19.23] well, can they maintain an attachment to a dynamic tip?
[00:10:23.19] And to do that we developed the following assay:
[00:10:25.01] we take a microtubule seed
[00:10:26.25] and we attach it to a cover slip.
[00:10:28.24] We then grow a dynamic microtubule
[00:10:31.24] off of the cover slip,
[00:10:33.03] and we do this in the presence of the kinetochores
[00:10:34.25] so that they bind all along the lattice of the microtubule.
[00:10:38.03] Then, to induce microtubule disassembly,
[00:10:40.28] we flow out the free tubulin
[00:10:42.21] and that causes the tips of the microtubules to disassemble.
[00:10:46.06] And I'm going to show you this in the next slide in a movie.
[00:10:50.03] So, what you're looking at here is the microtubule,
[00:10:51.26] and you can see a number of kinetochores
[00:10:54.19] bound to the side of the microtubule.
[00:10:56.25] And when I start the movie,
[00:10:59.16] if you watch the bottom of the microtubule,
[00:11:01.21] when it starts to disassemble,
[00:11:03.22] you'll see that it hits that first green kinetochore
[00:11:06.00] and that kinetochore stays attached.
[00:11:08.19] So, I'll start the movie for you.
[00:11:10.02] So, you can see now,
[00:11:11.20] the microtubule is disassembling
[00:11:13.04] and the kinetochore is staying attached to that microtubule
[00:11:15.05] as it disassembles right out from under it.
[00:11:17.15] So that showed that our kinetochore
[00:11:19.13] can bind to a microtubule tip.
[00:11:20.29] And I want to emphasize, as I said earlier,
[00:11:22.29] we have no motor proteins in these preps
[00:11:25.22] that could drive kinetochore movement.
[00:11:28.29] So what's happening here is that
[00:11:30.18] the ability of the kinetochore to stay bound,
[00:11:32.11] just to the microtubule tip,
[00:11:34.18] is sufficient to have it move the kinetochore.
[00:11:39.05] Next, we wanted to ask,
[00:11:40.16] can we put these under tension,
[00:11:42.02] and to do this we developed yet another assay,
[00:11:43.27] and this is a laser trapping
[00:11:46.01] or optical trapping assay.
[00:11:47.09] And what we do here is...
[00:11:49.03] the setup is exactly the same,
[00:11:51.11] we grow a dynamic microtubule off of a cover slip,
[00:11:54.12] but now what we do is we take large polystyrene beads
[00:11:58.08] and we attach kinetochores to those beads,
[00:11:59.26] and so the beads serve as cargo that we can hold
[00:12:03.21] -- they sort of take the place of the chromosome.
[00:12:05.24] We use a laser trap to capture that bead
[00:12:08.23] and then we can position it at the tip of a microtubule
[00:12:11.13] so that a kinetochore binds to the tip of the microtubule.
[00:12:14.23] And then we can do different things:
[00:12:16.19] we can apply tension
[00:12:18.22] using a feedback-controlled stage,
[00:12:20.23] we can vary the tension,
[00:12:22.07] or we can keep the tension constant.
[00:12:25.16] So, using this assay,
[00:12:27.09] we were able to analyze a kinetochore attachmen
[00:12:30.10] t to a microtubule tip under tension.
[00:12:32.06] So, what you're seeing here is a trace of a single experiment
[00:12:35.10] where we attached a kinetochore to the tip of a microtubule
[00:12:38.12] and, in this experiment,
[00:12:40.13] we maintained the tension at 1.7 picoNewtons.
[00:12:43.25] And in these experiments we're filming them,
[00:12:47.14] so we can see what's happening to the bead and the microtubule
[00:12:50.09] and, on the left,
[00:12:52.03] if you look at the bead position,
[00:12:53.15] the relative bead position tells us what's happening to the microtubule.
[00:12:56.07] So, if the bead is moving in one direction,
[00:12:58.00] we can see that the microtubule is assembling,
[00:13:01.00] and then you can see it suddenly switches
[00:13:03.03] to moving in a different direction,
[00:13:05.02] and that's the microtubule
[00:13:07.12] switching into a state of disassembly,
[00:13:09.18] and then if it switches between these phases
[00:13:11.15] it's either a rescue or a catastrophe event.
[00:13:14.24] At some point in the experiment,
[00:13:16.20] the kinetochore will just let go --
[00:13:18.15] it will detach from the tip of the microtubule --
[00:13:21.04] and at that point we call that the total lifetime of the attachment,
[00:13:24.25] so that's the length of time that the kinetochore stayed attached
[00:13:28.20] to that dynamic tip.
[00:13:30.01] And so you can see in this experiment
[00:13:31.21] that we learned a lot in this single experiment.
[00:13:33.23] The kinetochores can stay bound to the tip of a microtubule
[00:13:36.29] under tension,
[00:13:38.12] and we found that these particles
[00:13:41.12] can remain bound through all of the dynamic phases
[00:13:43.09] that microtubules undergo.
[00:13:44.24] So at this point we were really convinced
[00:13:46.14] that we had reconstituted dynamic kinetochore-microtubule attachments
[00:13:49.10] under tension.
[00:13:52.06] These purified kinetochores
[00:13:54.02] maintain attachments much better than any individual subcomplex
[00:13:57.19] has been seen to maintain an attachment,
[00:13:59.29] and so, in the past,
[00:14:01.22] work from various labs
[00:14:04.08] have reconstituted attachments between
[00:14:06.21] the Ndc80 complex
[00:14:09.09] and the Ndc80+Dam1 complexes to the microtubules.
[00:14:13.14] And that's shown here, this is work from Chip Asbury's lab
[00:14:15.10] and Trisha Davis,
[00:14:16.18] and you can see that, with the Ndc80 complex,
[00:14:19.01] if they coat beads with Ndc80
[00:14:20.20] and then put it on the tip of a microtubule,
[00:14:22.26] what you're looking at here is the probability
[00:14:24.25] that an attachment will survive a given time.
[00:14:27.20] And so, when the add Dam1 to Ndc80 beads,
[00:14:30.29] it definitely improves the performance,
[00:14:33.29] but if you look at our kinetochores,
[00:14:36.02] they greatly outperform either of these single microtubule binding complexes.
[00:14:40.22] And I don't have time to go through it, but we were able to show that,
[00:14:43.26] in our experiments,
[00:14:44.27] single kinetochores suffice for this coupling activity,
[00:14:47.04] whereas in the experiments with Ndc80 and Dam1,
[00:14:50.13] multiple subcomplexes were involved
[00:14:52.18] in the attachment.
[00:14:54.00] So the kinetochores...
[00:14:55.16] native kinetochores really do outperform
[00:14:57.17] any individual microtubule binding complex
[00:14:59.10] that's been seen before.
[00:15:02.08] So, having purified kinetochores, then,
[00:15:06.12] let us really address these functions that I talked about
[00:15:08.27] in the first talk.
[00:15:10.07] And so, today, I'm going to tell you
[00:15:11.27] some of the work that we did to look at the mechanical role
[00:15:13.20] of the kinetochore binding to the microtubule,
[00:15:16.06] and then we'll talk about the biophysical properties.
[00:15:18.28] So, as I discussed in the first part of the talk,
[00:15:20.25] there's various models for how the kinetochore
[00:15:23.06] might maintain an attachment.
[00:15:25.06] Now, the original model was that
[00:15:27.02] motor proteins at the kinetochore
[00:15:28.21] could maintain an attachment
[00:15:30.03] and walk along the microtubule.
[00:15:32.02] And this was actually the most popular model
[00:15:34.02] until we found out that
[00:15:38.11] motor proteins are actually not essential for chromosome movement
[00:15:40.29] in cells.
[00:15:42.04] So, they're extremely important,
[00:15:43.20] but they're not driving kinetochore attachment
[00:15:45.25] and movement on the microtubule.
[00:15:48.09] There's ring models that take into account
[00:15:51.03] the idea that the kinetochore can form a ring
[00:15:54.10] around the microtubule and that,
[00:15:56.00] as long as that can stay attached,
[00:15:57.28] the peeling protofilaments,
[00:15:59.29] as the microtubule disassembles,
[00:16:01.11] will maintain an attachment
[00:16:02.24] -- they'll harness the energy of depolymerizing microtubules.
[00:16:06.01] And thi model has been very popular
[00:16:07.24] because work from the Nogales, Barnes, and Harrison labs,
[00:16:10.12] as you can see in these EM (electron microscopy) images,
[00:16:12.23] showed that the Dam1 complex
[00:16:15.06] can form a ring around the microtubule.
[00:16:18.07] Biased diffusion models are also very popular,
[00:16:20.12] and that idea here is that multiple microtubule binding elements
[00:16:23.04] within the kinetochore
[00:16:24.23] rapidly bind and unbind to the tip of the microtubule,
[00:16:27.11] and as long as this occurs in a biased manner
[00:16:29.05] it would maintain an attachment near the tip.
[00:16:31.14] And this model,
[00:16:33.13] these are just EM images of the Ndc80 complex
[00:16:35.16] to show you that they have...
[00:16:38.08] they're long coiled-coil proteins
[00:16:40.14] that have globular domains that bind to the microtubule head,
[00:16:42.28] and the Asbury and Davis labs
[00:16:46.11] showed that this complex can undergo biased diffusion.
[00:16:48.11] So, we still don't know what the actual underlying mechanism is.
[00:16:51.24] And part of the difficulty has been that,
[00:16:54.17] in vivo, it's very difficult to visualize
[00:16:57.09] the structure of a kinetochore-microtubule attachment.
[00:16:59.25] So, this is an EM image
[00:17:01.28] from Bruce McEwen's lab,
[00:17:03.09] and you can see the underlying kinetochore structure
[00:17:05.12] and then you can see microtubules attaching to it,
[00:17:08.12] and it's very difficult to look at this structure
[00:17:10.16] and really understand
[00:17:12.08] how it's attaching.
[00:17:13.21] So we thought, perhaps,
[00:17:15.07] if we did electron microscopy
[00:17:17.05] on our purified kinetochores attached to microtubules,
[00:17:20.02] we might learn more about the structure.
[00:17:22.04] We set up a collaboration with Tamir Gonen's lab
[00:17:24.15] and we were able to see
[00:17:27.11] in EM, in negative stain EM images,
[00:17:29.26] our purified kinetochores attached to microtubules.
[00:17:32.21] And so this is just one of the images that we obtained,
[00:17:35.24] you can see it's an extremely large structure,
[00:17:37.28] it's greater than 100 nanometers,
[00:17:40.01] and the really exciting this is
[00:17:42.02] we were able to see a ring encircling the microtubule
[00:17:45.11] that we believe is the Dam1 ring,
[00:17:47.07] and then there's other microtubule attachment sites.
[00:17:50.22] And now we're working with Eva Nogales' lab
[00:17:52.16] to really understand how these different complexes
[00:17:55.00] are binding to the microtubule and, of course,
[00:17:58.10] now we can do immuno-EM and other techniques
[00:18:00.06] to try to elucidate
[00:18:02.14] which proteins we're looking at in this structure
[00:18:04.04] and how they correlate with, sort of,
[00:18:06.04] the models that have been put forth for the kinetochore architecture.
[00:18:11.06] Next, I'm going to talk about
[00:18:14.09] the biophysical properties of the kinetochore,
[00:18:16.18] where tension can stabilize attachments.
[00:18:19.25] So, as I said in the first talk,
[00:18:22.10] tension is important to ensure that
[00:18:25.19] cells have made the proper attachments,
[00:18:27.04] so once kinetochores make bioriented attachments to microtubules
[00:18:30.25] from opposite poles,
[00:18:32.09] they come under tension and they're highly stable in vivo.
[00:18:34.12] In contrast, when cells make
[00:18:36.07] incorrect attachments that lack tension,
[00:18:38.15] there are error correction mechanisms
[00:18:40.12] that ensure that these are turned over
[00:18:42.15] to give the cell a chance to make the proper attachment again.
[00:18:47.12] When we started this work, the major mechanism know of error correction
[00:18:49.10] is via Aurora B phosphorylation,
[00:18:51.28] so the idea here is that the Aurora B kinase
[00:18:54.05] recognizes incorrect attachments,
[00:18:57.03] it then phosphorylates kinetochore proteins
[00:18:59.26] such as the Ndc80 and Dam1 proteins
[00:19:02.09] that I told you at the beginning
[00:19:04.08] are major microtubule binding elements in the kinetochore,
[00:19:06.03] and that gives the cell a chance to make the correct attachment again.
[00:19:09.03] However, as I told you,
[00:19:11.03] our kinetochores don't have any Aurora B,
[00:19:13.05] so we weren't sure how they would behave under tension.
[00:19:16.23] So we decided to ask
[00:19:19.12] what the effect of tension on kinetochore-microtubule attachments is,
[00:19:22.06] and so the idea here is that
[00:19:24.21] you know that tension usually destabilizes attachments,
[00:19:27.15] and so the idea is,
[00:19:29.12] if we monitor
[00:19:32.03] the lifetime of the attachment at a range of different forces,
[00:19:36.24] how do the kinetochores behave?
[00:19:38.25] Now, for a typical protein-protein interaction,
[00:19:40.19] where tension usually destabilizes it,
[00:19:42.10] the curve would look like this
[00:19:44.10] -- as you increase the force,
[00:19:47.07] the lifetime of the attachment would decrease --
[00:19:49.19] and that's what we expected would happen,
[00:19:51.16] because we were missing the Aurora B kinase
[00:19:53.17] on our kinetochores.
[00:19:56.00] So, we at first looked at 2 picoNewtons
[00:19:58.03] and we found that the kinetochores
[00:19:59.23] would stay attached to the microtubule tips
[00:20:01.14] for about 20 minutes.
[00:20:03.05] And then the big surprise was as we increased the force,
[00:20:06.24] the lifetime increased.
[00:20:09.08] So by 5 picoNewtons of force,
[00:20:11.09] the kinetochores were maintaining attachments
[00:20:13.14] for almost an hour,
[00:20:14.25] and this is longer than mitosis
[00:20:17.01] even takes in budding yeast.
[00:20:18.16] Now, as continued to increase the force,
[00:20:21.16] the kinetochores started to just detach
[00:20:24.08] or rupture from the tip of the microtubule.
[00:20:27.23] So, what we learned from this experiment is that
[00:20:31.04] something about high tension
[00:20:33.04] is really promoting attachments,
[00:20:34.20] and this is an intrinsic property
[00:20:36.22] of the kinetochore-microtubule interaction
[00:20:39.02] -- there is no Aurora B kinase involved.
[00:20:41.18] Somehow, tension is just stabilizing an attachment,
[00:20:44.14] and this is like those finger traps that you remember,
[00:20:46.13] as you pull on them they come under more tension.
[00:20:49.10] And so we think that kinetochores
[00:20:51.11] have that property when they bind to microtubules:
[00:20:54.02] as we put them under tension,
[00:20:55.23] they are stabilized.
[00:20:57.10] This is reminiscent of two-state interactions
[00:21:00.08] such as receptors and ligands,
[00:21:01.16] where there's a strong and a weak state,
[00:21:03.06] and something about tension promotes the stronger state,
[00:21:05.15] and so we're actively working now,
[00:21:07.22] and trying to understand the underlying mechanism where tension
[00:21:11.00] can really stabilize these attachments intrinsically.
[00:21:15.12] So, in the future,
[00:21:18.17] we're using these purified kinetochores
[00:21:21.00] to continue to study the three major roles of kinetochores.
[00:21:23.15] For the mechanical role,
[00:21:25.15] we're going to continue the EM work
[00:21:26.27] to try to really understand more about the structure of the kinetochore
[00:21:29.24] and how it physically attaches to the microtubule.
[00:21:31.23] I've showed you today that
[00:21:33.21] we've confirmed that there's a ring structure at the kinetochore
[00:21:36.09] and we're hoping to get higher resolution images
[00:21:38.01] of how the kinetochore otherwise is attaching.
[00:21:40.27] We're also probing the various functions of kinetochore proteins.
[00:21:44.18] We're going to continue to work on the biophysical properties.
[00:21:46.25] We discovered, using these purified reconstituted kinetochore-microtubule attachments that
[00:21:53.01] tension really directly stabilizes attachments,
[00:21:54.22] completely independently of Aurora B,
[00:21:57.22] so we're going to try to identify the proteins involved in that,
[00:21:59.18] and then how does that integrate with the Aurora B error correction pathway
[00:22:02.24] in the long term.
[00:22:04.04] And then, finally, I didn't have time to talk about it today,
[00:22:06.19] but we've been able to reconstitute
[00:22:08.16] major events of the spindle checkpoint
[00:22:10.15] using our purified kinetochores
[00:22:12.16] and identify some of the key phosphorylation events
[00:22:15.15] that recruit checkpoint proteins to the kinetochores.
[00:22:18.20] And in the future we're developing fluorescent assays
[00:22:20.02] to look how tension and attachment
[00:22:23.14] affect the occupancy of checkpoint proteins,
[00:22:27.03] and we hope to eventually reconstitute the checkpoint in vitro
[00:22:28.29] to really understand the individual events
[00:22:30.25] that cause the kinetochore-microtubule attachment defects
[00:22:34.15] to then halt the entire cell cycle.
[00:22:36.25] So, thank you for your time
[00:22:38.08] and I'd just like to acknowledge
[00:22:39.24] the people involved in this work.
[00:22:41.10] Bungo Akiyoshi pioneered the
[00:22:43.20] entire kinetochore purification.
[00:22:45.06] I've had help from a number of other people in the lab.
[00:22:48.11] Krishna Sarangapani and Andy Powers
[00:22:51.29] in Chip's lab
[00:22:53.05] did the biophysical work,
[00:22:55.00] and this has been a really long-term, fantastic collaboration
[00:22:56.27] with the Asbury lab.
[00:22:58.07] And Shane Gonen and Tamir Gonen w
[00:23:00.03] ere our EM collaborators,
[00:23:01.26] who really helped us see the first structure of a kinetochore.
[00:23:03.16] And of course I'm very grateful to the funding agencies,
[00:23:05.26] over the years,
[00:23:07.18] that have supported this work.
[00:23:09.06] Thank you.

Videos in this Talk

Part 1: Chromosome Segregation
Audience:
- Student
- Researcher
- Educators of H. School / Intro Undergrad
- Educators of Adv. Undergrad / Grad
Part 2: Investigating Kinetochore Function
Audience:
- Student
- Researcher
- Educators of H. School / Intro Undergrad
- Educators of Adv. Undergrad / Grad

Speaker: Sue Biggins

Total Duration: 0:54:11

Recorded: April 2016

All Talks in Cell Biology

Talk Overview

Proper chromosome segregation during cell division is critical to ensure that daughter cells inherit the correct number of chromosomes. Microtubules emanating from the spindle poles pull on sister chromatids to move one chromosome to each pole. The kinetochore, a protein complex on the chromosome, is key to regulating chromosome segregation. Kinetochores form attachments to microtubule ends (no easy feat since microtubules are constantly growing and shrinking), they sense tension to ensure that sister chromatids are connected to microtubules from opposite poles, and they signal the cell to stop cell division if attachment is not correct. Biggins gives an excellent overview of kinetochore structure and its critical functions in chromosome segregation.

When Biggins began working on kinetochores, the experiments that she could do were limited by the lack of a method to purify intact kinetochores. In Part 2 of her talk, Biggins explains how her lab purified kinetochores from yeast (for the first time ever!). They showed that the purified protein complex functioned in the same manner in vitro as endogenous kinetochores. Using electron microscopy and other techniques, Biggins and her collaborators were able to visualize the structure of the kinetochore-microtubule attachment and demonstrate, surprisingly, that tension directly stabilizes the attachment.

Speaker Bio

Sue Biggins

Dr. Sue Biggins studied biology as an undergraduate at Stanford University and initially thought she would apply to medical school after receiving her degree. However, after a summer working in a research lab, she changed her mind and decided to apply to graduate school. Biggins received her PhD in molecular biology from Princeton and was… Continue Reading

Comments

pankaj rathour says

September 28, 2020 at 8:12 am

thank you, so much.

Reply

Part 1: Chromosome Segregation

Part 2: Investigating Kinetochore Function

Talk Overview

Speaker Bio

Sue Biggins

Comments

Leave a Reply Cancel reply

Funders

Start Here

Explore

Talks By Topic

Talks by Series

Career

Courses

Educator

About Us

The Kinetochore and Chromosome Segregation

Talk Overview

Speaker Bio

Sue Biggins

Playlist: Cell Division

Related Resources

Reader Interactions

Comments

Leave a Reply Cancel reply

Footer

Funders

Start Here

Explore

Talks By Topic

Talks by Series

Career

Courses

Educator

About Us