Cellular Organization of Complex Cell Structures
Transcript of Part 1: How Does Complexity Arise from Molecular Interactions?
00:00:01.24 Hello. My name is Tony Hyman. 00:00:03.07 I'm a director of Max Planck Institute in Dresden in Germany, 00:00:06.27 and I'd like to talk to you today about organization of cytoplasm. 00:00:10.14 So, one of the key questions in biology that we're all interested in 00:00:15.15 is the following: How does complexity arise from molecular interactions? 00:00:19.18 The things we're interested in, in terms of cells, 00:00:23.03 are often 5 or 6 orders of magnitude bigger than the molecules that make them up. 00:00:28.25 So, if you have a molecule, over here, on this scale, 00:00:32.15 the cells, for instance, are 10^5... 10^6 orders of magnitude bigger 00:00:38.03 than the molecules that make them up. 00:00:39.24 So, what are the rules by which these molecules can interact 00:00:43.09 to create these very complex structures 00:00:45.13 which are so much bigger than they are themselves? 00:00:47.27 I've worked on this problem for most of my career 00:00:51.00 in a small nematode called C. elegans. 00:00:54.12 And, C. elegans makes embryos about 50 microns long, 00:00:59.23 with a very reproducible cell division. 00:01:02.11 There was a Nobel prize won for the study of various things in C. elegans, 00:01:07.27 in particular cell death in the adult. 00:01:09.28 One of the most important things about working on C. elegans 00:01:13.05 is it's completely invariant. 00:01:14.15 So, the cell lineage of the worm... the way in which one cell 00:01:18.25 goes all the way to the different cells of the adult 00:01:21.25 is completely invariant. 00:01:23.04 But, also, the cell division itself is invariant. 00:01:26.15 I've looked in many, many of these embryos over the years, 00:01:29.21 probably too many to think and calculate 00:01:31.20 But, each one of them looks the same as the one before. 00:01:35.17 So this reproducibility allows us to study problems of organization, 00:01:40.23 of cells after fertilization. 00:01:44.20 What I've got in the background here is a movie 00:01:46.26 of a C. elegans embryo, shortly after fertilization. 00:01:50.06 And what you can see is the cytoplasm of the embryo 00:01:54.16 with its 2 pronuclei, one here and the other one over here. 00:02:00.11 And these are the haploid genomes of the mother and the father 00:02:05.10 which are now in one egg, and I'm going to show you a movie of these 2 pronuclei 00:02:09.08 coming together. And when they come together, they then form a mitotic spindle, and they divide. 00:02:15.08 And this is using a technique known as Nomarski microscopy 00:02:18.04 which is a wide-field, bright-field microscopy technique. 00:02:22.00 So, what you can see now is the two pronuclei coming together 00:02:27.16 in a process known as pronuclear migration. 00:02:29.23 And, boom, they've hit each other, and now they're going to make a mitotic spindle, 00:02:34.24 which you can see in the middle here. 00:02:36.27 And they undergo anaphase. The cell divides into two. 00:02:40.28 Note how one cell is smaller than the other cell. 00:02:44.27 And now you're going to see the second cell divisions. 00:02:48.12 Notice how the divisions now are asynchronous, 00:02:53.19 and the cells are also different sizes, and now you have a 4-cell embryo. 00:02:58.11 Have a look at the timing down here, 00:03:01.11 which is about 30 minutes in this timer on the bottom right. 00:03:05.04 So, about 30 minutes after fertilization. 00:03:09.13 That means the cells are undergoing this very, very complex set of arrangements, 00:03:14.02 and they've built very, very complex structures 00:03:16.07 necessary for their correct cell division. 00:03:18.17 And, if you look at C. elegans embryos, not only is it interesting to study the cell division itself, 00:03:24.29 but an added nice feature of it is the cell division itself is asymmetric. 00:03:29.14 And what I mean by asymmetric is that the fate of the two cells 00:03:34.15 is different. So, this one cell divides into 2 cells, which are illustrated here. 00:03:39.27 The green cell is the germline, and the red cell is the soma. 00:03:43.22 And those two different cells go on to give cells of very different fates. 00:03:48.09 So that means that you can study, in this 30 minute time period, 00:03:52.11 the division of the cell... the basic cell division processes necessary to divide a cell 00:03:57.04 which are likely common to all cells that are trying to divide. 00:04:01.00 And, in addition, you can ask, how is it the cell division can be asymmetric? 00:04:06.02 Now, when you look at a division like this, one very important thing to realize, 00:04:12.07 of course, is that the fertilized egg comes from the fertilization of the oocyte 00:04:17.03 and the sperm. And the majority of the cytoplasm comes 00:04:23.22 from the oocyte, and it's a gift from the mother. 00:04:27.23 And the sperm comes in, provides some components, 00:04:30.25 but it mainly triggers the fertilization process and adds some of the DNA. 00:04:35.27 And so the cytoplasm is a gift from the mother, 00:04:38.24 and in the C. elegans embryos, the oocyte is relatively undifferentiated. 00:04:44.25 That's not true for oocytes in all organisms, 00:04:48.03 but in our particular nematode, the oocyte itself is relatively undifferentiated. 00:04:52.13 It's in cytoplasm that's a gift from the mother. It's like a bag of cytoplasm components. 00:04:57.26 And then, that gets fertilized, at about half an hour afterwards, 00:05:02.02 you organize this rather undifferentiated bag of cytoplasm 00:05:06.21 into this complex choreographed set of events necessary for cell division. 00:05:11.15 One of the key events in trying to understand the division of this nematode, 00:05:20.10 was the sequencing of the DNA. So, the DNA sequence 00:05:25.11 of C. elegans was the first multi-eukaryotic system in which the DNA was sequenced. 00:05:29.26 And, the sequence of the organism defines the potential catalog 00:05:35.28 of genes required for its life. 00:05:39.00 And C. elegans has, depending on the time of the month, 00:05:43.01 about 20,000 different genes. 00:05:47.29 So, what you can ask then is, "Here are my genes..." 00:05:52.09 So, the yellow represents the chromosome, and the black bars represent 00:05:56.03 distribution of genes along the chromosome. 00:05:59.11 And you can ask which one of these genes here 00:06:03.10 is responsible for cell division? 00:06:06.07 Of these 20,000 genes, which one of those are involved in this process of making the embryos 00:06:13.01 and dividing them? 00:06:15.03 And we decided to look at this problem, 00:06:17.21 because of the development of a process... a technique called RNA interference. 00:06:24.28 And this discovery of RNA interference 00:06:26.21 by Andy Fire and Craig Mello also won the Nobel Prize a few years ago. 00:06:32.26 Now, the key thing you want to do, as I said, 00:06:36.13 is to create a catalog. The genes themselves 00:06:39.13 when you sequence the genome, 00:06:43.09 you want to be able to assign a function to each one of those genes, 00:06:47.27 So let's say you've got 20,000... 25,000 genes. 00:06:51.00 You'd like to go to each one of them and catalog its function. 00:06:54.26 And that has been something that most of biomedical research 00:06:59.14 has been trying to do over the last 20 years 00:07:01.09 since the sequencing of the human genome 00:07:03.14 is trying to assign function to each one of those genes. 00:07:07.00 As I said, the key to step forward in that cataloging process 00:07:13.15 of assigning functions is the development of RNA interference. 00:07:16.00 And here I have a slide which outlines how we think RNA interference works 00:07:20.04 and how we can use it to silence the function of genes. 00:07:23.29 So, the idea is, you want to take each gene, one after the other. 00:07:27.03 You want to silence its function. And you want to ask, after we silence its function, 00:07:31.26 what effect does that have on the development or cell division of the organism? 00:07:38.14 So, in our particular case, we'd like to ask, after we've silenced the gene, 00:07:41.19 what effect does that have on the cell division? 00:07:44.26 Now, what we do is we make an RNA in the test tube 00:07:50.11 called a trigger double strand RNA. 00:07:52.06 So, we make a double-stranded RNA, and then we introduce it into the mother. 00:07:57.08 And what happens after that is that the Dicer, 00:08:03.07 which is an enzyme, chops up the double stranded RNA 00:08:06.10 into smaller components that I've shown here. 00:08:10.19 They're called little siRNAs. The Dicer chopped them up into 21-25 base pair 00:08:16.19 little pieces. Then what happens 00:08:20.01 is that the RISC complex binds these siRNAs, 00:08:25.07 unwinds them, and individual unwound RNAs 00:08:33.14 are then bound to the endogenous mRNA 00:08:38.02 which is single stranded, and that triggers its cleavage. 00:08:42.19 So, over here, I've shown that the mRNA has been cleaved 00:08:47.05 by the binding of the single stranded siRNA. 00:08:52.14 So, that's a very powerful technique because now what we can do 00:08:55.23 is we can remove the RNA for any gene we're interested in 00:08:59.27 because RNA makes the protein, you therefore don't make any more protein, 00:09:04.22 and that means that you can then study what happens to your organism 00:09:09.26 when you've removed the function of that gene. 00:09:13.14 Now, it's not quite as simple as that, because there's a problem. 00:09:18.19 which is that, of course, RNA interference removes the messenger RNA very quickly. 00:09:23.21 It's a catalytic process that destroys the mRNA. 00:09:26.22 But, the protein levels then decline at a different rate, 00:09:32.01 and that's because once the message has been destroyed, 00:09:35.20 there's no more message to make any more protein, 00:09:39.06 but the protein that was already existing, 00:09:40.26 first of all has to disappear. 00:09:43.15 And that's one of the difficulties of RNA interference 00:09:46.26 is waiting for the protein to disappear. 00:09:49.01 So, what you're normally doing is worrying about what's known to biologists... called a run-down, 00:09:53.20 which is the protein levels are slowing running down, 00:09:56.17 and the difficulty in RNA interference is deciding, 00:09:59.15 what does it mean if I look at cell division when there's this much protein left, 00:10:04.15 and later on, I look at the same cell division with this much protein left. 00:10:07.09 How do I interpret that phenotype? 00:10:09.06 And it turns out that C. elegans, 00:10:11.24 has a particular advantage which allows us to get around that problem. 00:10:16.08 Now, here what I've done is I've shown from the Wormbook and WormAtlas 00:10:21.15 a picture of a C. elegans with its gonad and its gut. 00:10:26.05 And what you can see from this is C. elegans is basically an egg-laying machine. 00:10:30.07 It makes about 200 embryos after its made, 00:10:34.01 and it makes them rapidly and then it goes on and dies. 00:10:37.28 What I've shown here are the gonads, which here is where the cells of the future oocytes 00:10:44.07 are being made. Then, as they come around the arm of the mother, 00:10:47.14 they're made into these different oocytes, 00:10:49.12 and then they go through the spermatheca here, 00:10:51.28 and then they go on to be fertilized eggs. 00:10:55.19 And here I've got a blow-up of the gonad from the Wormbook, 00:10:59.05 where you can see what's happening is the gonad is what's called syncytial here. 00:11:04.12 (In other words, it's got no membranes.) 00:11:07.06 And, as you come around the bend of the gonad, you make oocytes. 00:11:11.02 And the key thing about this process is shown in the next slide, 00:11:15.27 which is, first of all, you inject the double strand RNA, or you can feed it. 00:11:21.26 And then what happens is that the RNA is then degraded. 00:11:27.05 Then, the protein flushes out because the gonad is constantly making oocytes. 00:11:37.01 So, I've shown in the arrow this constant flushing out process of protein 00:11:40.14 as it makes new oocytes. 00:11:42.20 Now, of course, it then makes new protein to make more oocytes, 00:11:46.27 but because the messenger RNA is now missing because you silenced it, 00:11:49.29 it now means you're making these oocytes in the absence of the protein that you're interested in. 00:11:56.12 And that's particularly because we can access what's known as the maternal cytoplasm. 00:12:02.00 And therefore, we can ask the embryos to be laid 00:12:05.03 i n almost the complete absence of the protein we're interested in. 00:12:08.19 And that is a great advantage of doing RNA interference in 00:12:11.24 C. elegans and why it's been so successful 00:12:14.07 in looking at particular phenotypes. 00:12:16.19 The genome-wide screen was primarily the work of two people, Pierre Gonczy and Chris Echeverri. 00:12:24.04 Pierre Gonczy and Chris pioneered the techniques for doing this in high throughput. 00:12:31.04 And, they designed the following screen, 00:12:34.22 where you took a genome-wide set of double stranded RNAs, 00:12:37.17 and we synthesized them, microinjected a pair of worms, 00:12:42.05 and we looked for cell division embryonic defects. 00:12:43.21 So, the idea was to try and film the cell division for after RNAi 00:12:51.03 of each one of the genes in the genome. 00:12:53.06 And one of the problems you have, in any experiment like this, 00:12:58.09 is we made about 45,000 different movies of these cell divisions. 00:13:03.05 And how do you quantify this problem? 00:13:05.11 What you want to get away from is this idea of saying, 00:13:09.13 well this seems to have a cell division defect, 00:13:11.08 that may have a defect in dividing the cell in two. 00:13:14.09 You want to be quantitative. 00:13:15.28 By being quantitative, you can then understand a lot more about the particular phenotype. 00:13:20.23 We came up with the following method to do it, 00:13:23.11 which is, we divided the division into 47 different things that we thought 00:13:30.13 could go wrong, 00:13:32.09 and then we looked through the movies manually, 00:13:35.06 and if there was a defect, we gave it a 1, and if there was no defect, we gave it a 0. 00:13:39.13 And then we did each movie 5 times, 00:13:42.17 and then we did 5 movies of each embryo. 00:13:46.16 And then, from that, we were able to get a heat map, 00:13:49.09 where you can see that the colors represent the penetrance of the phenotype. 00:13:53.07 So, like 100% means that every movie that we looked at had a particular phenotype. 00:14:00.27 And so, by doing that for all the movies, we then ended up with what's known 00:14:05.02 as a heatmap here, which is the defect categories on this axis, 00:14:09.02 and the genes on this axis, and then you can see you have clusters of genes 00:14:14.09 with certain phenotypes. 00:14:15.20 So these ones, for instance, are required for chromosomes to form and segregate. 00:14:21.11 What that screen did, is it defined a catalog of genes required for the first cell division 00:14:29.14 of a C. elegans. And for those of you who are interested, 00:14:32.11 we put all the movies online at the website on the bottom of the slide. 00:14:36.22 And, of course, we were also able to divide up the genes into functional categories, 00:14:41.04 of genes that are required for particular, different purposes. 00:14:45.12 But one of the questions that biologists now face 00:14:48.25 is, in this huge amount of information, how do we make sense of it? 00:14:51.17 How do we make sense of this massive complexity... 00:14:54.04 800 genes is an enormous amount of data to work with. 00:14:59.05 How can we make this simpler in a way that we can actually try and study particular problems? 00:15:05.08 Because all we've done here is really created the catalog 00:15:09.04 of genes... what you might call Gene X is required for process Y, 00:15:13.12 and that, of course, is powerful, but it doesn't tell us 00:15:18.05 how the embryo itself is organizing and dividing. 00:15:22.01 And that's the problem that I've always been interested in. 00:15:24.13 Now, one thing to remember... if you think about this question... 00:15:28.23 How is the function of 800 genes coordinated? 00:15:32.05 It's important to remember the following fact, 00:15:34.15 which is cytoplasm is not an undifferentiated soup of genes. 00:15:39.23 You can't think about the cytoplasm as a set of 800 proteins, 00:15:44.02 floating around, and somehow organizing a cell division. 00:15:49.27 That's not how we think about the cell. That would be much too complicated. 00:15:53.24 Rather, what we do, as cell biologists, 00:15:56.06 and generally as biologists, is we think about biology as a hierarchy of organization. 00:16:01.01 So, here I've shown you 1 hierarchy of organization, 00:16:05.13 which is, you can see that proteins themselves form protein complexes 00:16:11.14 as an example. And that appears to be one of the first levels of organization 00:16:17.09 under which proteins function in the cell. 00:16:19.29 Most proteins appear to function in protein complexes with other proteins, 00:16:24.23 and in fact, some of the protein complexes are very, very sophisticated 00:16:28.04 with many, many proteins. The ribosome, for instance, has over 100 proteins. 00:16:33.02 And so, here are some crystal structures of ribosomes and nuclear pore complexes. 00:16:40.17 And, molecular machines tend to be protein complexes that are very complex. 00:16:45.19 And that's one way that we can think about how to organize the cytoplasm, 00:16:50.04 which is to say, not what the function of the individual proteins are, 00:16:53.20 but what are the functions of different protein complexes. 00:16:56.28 And, because of the power of that approach, 00:17:00.13 there's been a huge amount of work done using mass spectrometry. 00:17:05.02 Mass spectrometry is a way of defining individual proteins 00:17:09.19 which has been very powerful for working out which proteins are in which protein complexes. 00:17:14.26 And here I have a diagram of how we do these experiments. 00:17:19.19 We take an antibody, and we immunoprecipitate one protein. 00:17:25.00 Now, if the protein is in a protein complex, it brings down not only the protein that you're 00:17:28.23 immunoprecipitating, but also all the other proteins in the complex. 00:17:33.00 So, if there are 10 proteins in the complex, you'll immunoprecipitate 1 member of that complex, 00:17:37.22 and then you'll bring down all the other 9 proteins. 00:17:41.25 Then, what you do, is you take that mixture and you hit it with trypsin. 00:17:47.15 You add trypsin or another proteolytic enzyme, 00:17:51.09 and that cleaves the different proteins into their component peptides. 00:17:56.06 The key thing with mass spectrometry, which has been known for many decades, 00:18:02.11 is that the mass accuracy of the measurements are exquisitely sensitive. 00:18:07.03 I'm not going to go into the details -- there are many places you can read into that problem. 00:18:13.03 But, the key thing to remember is that mass spectrometry can, with exquisite precision, 00:18:19.03 measure the mass of different peptides. 00:18:22.05 And then, what you can do, is you can ask which peptides -- from the masses -- 00:18:27.18 are like to be part of different proteins, and that's turned out to be very accurate 00:18:32.05 with modern mass spectrometers, and then that allows you 00:18:35.04 to define the different protein complexes and the proteins in those complexes. 00:18:39.22 And that's what been known as proteomics. 00:18:41.16 And, for instance, here's one study where we can look at the different proteins 00:18:46.01 in the different complexes and define different protein complexes required for different purposes. 00:18:51.05 And, proteomics is a big industry now. 00:18:56.06 Many labs are trying to define the sets of protein complexes 00:19:01.13 involved in various different processes. 00:19:04.05 For instance, we might like to understand not what are the proteins required for cell division, 00:19:07.03 but what are the protein complexes required for the division of a cell. 00:19:11.09 Now, some protein complexes are more interesting in the way they actually function. 00:19:20.11 And I want to show you a couple of examples of those. 00:19:22.23 One of those are polymers. 00:19:24.03 And so, here I'm showing you a microtubule growing and shrinking. 00:19:28.10 And, for instance, there are things like centrioles, which are also very complex structures. 00:19:34.16 They're protein complexes that are also put together in very complex ways. 00:19:40.06 And, in the next parts of my talk today, 00:19:43.04 I'm going to talk to yo more about microtubules and centrioles 00:19:46.21 and how they're put together. 00:19:48.11 If we go back to our scale axis that I introduced at the beginning, 00:19:54.20 If you remember I talked about the fact that we have proteins over here, 00:19:59.07 and how are they organized to make cells, 00:20:02.00 which are many orders of magnitude bigger than the proteins themselves. 00:20:05.01 What you can see of course is that one of the ways that we think about that 00:20:09.00 is how the proteins are organized into these complexes. 00:20:12.24 And you can see that helps to bridge the scale gap. 00:20:17.05 But what about this big area here? 00:20:20.01 It turns out, in this area, there are also a lot of complexes, 00:20:23.09 about which we understand very little. And these are called compartments, or else organelles. 00:20:28.20 And, how protein complexes organize into compartments is much less well understood 00:20:35.21 than how proteins are organized into protein complexes. 00:20:38.23 And, these various different structures are things like centrosomes, 00:20:43.27 kinetochores, proteins, nuclear bodies 00:20:48.23 and these are large, non-membrane-bound compartments, 00:20:53.22 where many, many different protein complexes live together 00:20:56.18 and undergo particular aspects of cell physiology. 00:21:01.14 So, for instance, there are centrosomes, shown here. 00:21:04.29 And here I have kinetochores, illustrated by electron microscopy. 00:21:09.16 And these are all large, non-membrane-bound complexes... 00:21:14.03 compartments, which contain many, many different protein complexes. 00:21:19.15 And, let me show you movies... some other ones, such as the cortex, 00:21:24.05 which is the outside layer of the cell. 00:21:26.06 That, as an example, is also a compartment. 00:21:31.18 A very interesting compartment. 00:21:34.06 Then, for instance, we can look at centrosomes and chromosomes. 00:21:38.13 These are other compartments in the cells. 00:21:39.28 And, you can see here, a movie of mitosis, 00:21:45.10 where we've labeled the centrosomes and the chromosomes with GFP, 00:21:48.26 and you can see them going through mitosis. 00:21:51.13 These are also large compartments, 00:21:53.21 which are organized from many, many different protein complexes. 00:21:56.28 The key thing about these compartments is, first, they have no obvious structure, 00:22:05.16 when you look by electron microscopy. 00:22:07.13 A second important thing about these compartments 00:22:10.22 is they turn over quickly. 00:22:13.07 So, in other words, if you do a photobleaching experiment, 00:22:15.21 then the components in these compartments will exchange with the cytoplasm very quickly. 00:22:22.08 And, I'll show you some examples of that in subsequent slides. 00:22:25.12 But, the protein complexes that make up these compartments tend to turn over very slowly -- 00:22:30.18 on the order of hours. 00:22:32.00 So, the protein complexes I discussed with you earlier in this segment are very, very stable. 00:22:37.26 You can isolate them in test tubes. Some of them you can just leave sitting around in a test tube 00:22:41.18 for many, many days, and they'll be stable. 00:22:45.18 So, they don't turn over very fast. 00:22:47.18 But, the protein complexes in these compartments exchange very, very rapidly. 00:22:53.02 Just to give you an example of that, let's come back to our favorite topic, the centrosome. 00:23:00.05 What I'm showing you here is the centrosome by electron microscopy. 00:23:03.16 You can see very little because it's a very fuzzy, electron dense material, 00:23:08.07 with no obvious structure. What you can see is the centriole in the middle of the centrosome. 00:23:12.07 Now, let's take a component of the centrosome, the gamma-tubulin, 00:23:17.26 and we label it with GFP. Now gamma-tubulin is a big complex of proteins. 00:23:22.08 It depends, but it varies between 3 and 8 proteins in different systems. 00:23:26.17 If we label one of those components, 00:23:28.19 we're not labeling or following the activity of the whole complex. 00:23:32.07 If we photobleach that particular GFP... in other words, we hit it with very bright laser light 00:23:40.28 so that the fluorescence of the complexes in the centrosome is now bleached, 00:23:46.05 we can look at the recovery of fluorescence. 00:23:48.22 And the fluorescence can only recover by new gamma-tubulin complexes 00:23:53.20 diffusing in from the cytoplasm and into the centrosome. 00:23:56.28 So that gives you an idea of the turnover of these complexes at the centrosome. 00:24:01.29 If we do that, we find that they recover very quickly. 00:24:06.11 So, if you notice in that movie, I photobleached the centrosome, 00:24:10.24 and within 60 seconds, the fluorescence is already beginning to recover, 00:24:14.11 where the gamma-tubulin complex itself turns over very slowly. 00:24:18.29 Many hundreds of hours, probably. 00:24:21.15 So, that is a conundrum which we don't really understand, which is, 00:24:28.08 what are the rules by which the protein complexes, which themselves are so stable, 00:24:32.21 interact together in things such as a centrosome or the kinetochore, 00:24:37.03 or other things such as the nuclear body to organize themselves? 00:24:41.29 Just to give you another example of the cortex, which is the outside layer of the cell. 00:24:48.16 I've shown you that's divided into 2 different domains, 00:24:51.21 a posterior and an anterior domain, 00:24:54.17 and those domains are defined by different protein complexes. 00:24:57.25 And, I've shown you an example of one complex over here, the so-called anterior PAR complex. 00:25:03.00 Now, that, of course, is a relatively stable protein complex. 00:25:07.16 But, let's photobleach a component of the other compartment here 00:25:11.01 and watch it's recovery rate. 00:25:12.29 And see how fast it... BAM! 00:25:15.14 We've photobleached it, and look, a few seconds later, it's almost completely recovered. 00:25:19.06 So, that's exactly the same sort of problem, 00:25:21.03 which is what are the rules which keep these protein complexes in these compartments, 00:25:27.03 despite the fact they're turning over so quickly. 00:25:29.10 So, we come back then, to our scale here. What I've done to introduce this problem 00:25:39.12 is to show you how we can go from the proteins all the way to the cells 00:25:46.06 and bridge that scale by understanding the organization of different compartments. 00:25:52.29 And, one of the ways we think about that is emergent properties, 00:25:58.08 which is something you may have been hearing about, 00:26:00.28 which is emergent properties of collections of individuals. 00:26:04.22 So, the proteins themselves are individuals, 00:26:07.21 and the collective is an example of protein complexes. 00:26:10.20 So we can try and understand what properties of protein complexes emerge 00:26:14.10 from the collection of proteins that make up those protein complexes. 00:26:18.21 The same we can answer with protein complexes. We take all those protein complexes, 00:26:23.15 we put them together in like a centrosome or a nuclear body and can ask 00:26:26.27 what activities emerge from the combined activity of all those protein complexes? 00:26:32.08 And we can go further, of course, to ask, now we have all the compartments, 00:26:36.13 and we've put them together in a cell. 00:26:38.05 How does the property of the cell emerge from the organization of those compartments? 00:26:42.28 How do they actually work together? 00:26:44.10 And then, of course, we could ask the same thing in tissues. 00:26:47.17 Cells collectively work together to make tissues. 00:26:52.04 And the important thing about this is that at each level, at each scale, 00:26:56.08 we have to use different ways to think about it. 00:26:58.10 We need different microscopy, we need different theory 00:27:02.26 to understand each one of these different scales. 00:27:06.08 And that's what's made biology so interesting is that each one of these scales, 00:27:09.29 we had to think of new ways to do it, 00:27:11.24 and try and develop new microscopy techniques, 00:27:13.22 and try and pick up new theoretical ways 00:27:16.24 to think about how these different organizational structures work. 00:27:21.05 And, in the other sections of my talk, 00:27:23.22 what I'm going to come back to is talk about some of these levels of organization 00:27:28.22 in a bit more detail... 00:27:30.18 microtubules, centrioles, and something called p-granules, 00:27:34.13 and illustrate from these talks the different sorts of microscopy 00:27:38.10 and the different sorts of problems that occur 00:27:40.27 in trying to understand how these complexes are made up.