University Place

cc >>

Tom Zinnen

Welcome everybody to Wednesday Nite at the Lab. I'm Tom Zinnen and welcome to the Biotechnology Center here at UW-Madison. On behalf of our partners, the Wisconsin Alumni Association, Wisconsin Public Television, and the Science Alliance, thanks for coming here. We do this every Wednesday night, 50 times a year. Tonight I'm delighted to have Steve Paddock give a talk on "A

Brief History of Time Lapse

Picture the World Through Kaleidoscope Dyes." Which I think is from "Lucy in the Sky with Diamonds." I get to date myself tonight. Sean, excuse me, Steve works in Sean Carroll's lab here on campus and is employed by the Howard Hughes Medical Institute. He got his PhD at the University of Bristol in the UK and then did a post doc on Northwestern University studying cell behavior. He came to the University of Wisconsin Madison about 20 years ago to work in the microscopy facility and starting about 15 years ago, started working in the Carroll lab. I'm looking forward to this presentation very much, and please join me in welcoming Steven Paddock to Wednesday Nite at the Lab.

APPLAUSE

Brief History of Time Lapse

>>

Steve Paddock

Thanks, Tom. Thank you, everybody, for coming out this evening. The title is actually a little strange in that it is the title of two talks. I gave the first title, "A Brief History of Time Lapse," and then I switched to "Picture the World with Kaleidoscope Dyes." So when I saw it advertised on the website I threw a few time lapse images in. If you're for a Brief History of Time Lapse, the history is going to be incredibly brief.

LAUGHTER

Steve Paddock

So the whole talk is based around "Picture the World with Kaleidoscope Dyes." As Tom said, it's from "Lucy in the Sky with Diamonds." And I guess Paul McCartney insisted that "Lucy in the Sky with Diamonds" came from his daughter, Lucy, or his daughter came to him with a picture, a painting that she made. And he said, what's that? And she said, "That's Lucy in the sky with diamonds." So that was the story behind the song. Not LSD.

LAUGHTER

Steve Paddock

But I guess maybe. I don't know. So I'm just showing you a slide show right now up here. I don't know if any of you passed through the Dane County Regional Airport last year. I had the good fortune of helping with an exhibit entitled "Tiny: Art from Microscopes at UW-Madison." And it was up longer than we expected it to be, and it got a lot of interest from people going back and forth to the airport. And we got many great comments, one or two negative comments but not many. And what I thought I'd like to do this evening would be to try and explain to you how these images are produced, some of the history behind the scientific imaging that's led to these images and how we use some of these images in our work. Most of the images in the Tiny exhibit were actually produced using a relatively modern technique called laser scanning confocal microscopy. And we're lucky enough to have an instrument in the lab in molecular biology. So I'm going to stop the DVD and go to my PowerPoint show. And so here's just a collection of images. These are not all of my images. These are images produced by people on campus here that were part of this show. And kind of in an abstract way the bright colors, they do have kind of an artistic side to them, but they do also have scientific meaning. And several of the labs that I have been trained in most of the people that I have worked with have stressed the value of trying to make an image aesthetically appealing as well as scientifically kind of relevant. And I should probably stress from the outset that it's easy to view these images kind of as if they are pieces are artwork, but they're actually the result of many long hours of painstaking experiments, not always by me but by other people in the lab. And these are just kind of the tip of the iceberg. So it's easy just to see them and think oh, this is great, but they are the results of a lot of hard work and research. So, as Tom said, I work in Sean Carroll's lab, and over the years we work on developmental biology and evolution. So we're trying to figure out the actual mechanism of how evolution works at the molecular level. And the key to studying evolution is to study developmental biology. And it's a relatively new field, and Sean is kind of one of the leaders in this new field of evolutionary developmental biology or Evo Devo for short. And my nickname is actually Stevo Devo.

LAUGHTER

Steve Paddock

And it takes in from molecular biology through embryology to developmental biology, from comparative anatomy all the way out to paleontology. So we're used to studying genes that are millions of years old. And we talk about deep evolutionary time of 200 million years. A model organism that we use is mainly the fruit fly. I meant to bring some, I forgot to bring them, sorry. So but what I'd like to do is tell you a few stories about how this imaging developed. And what I'd like to start off with is kind of deep technological time. So I'm going to go way back in deep technological time to 1985.

LAUGHTER

Steve Paddock

My first story starts with two Englishmen sitting in a Bedouin tent in the dark. And I'm not talking about the Sahara or the Arctic or the Antarctic. I'm talking about this place, which is Cambridge in England. And you can tell it's Cambridge because this is the King College's chapel, the famous chapel at King's College. This is the River Cam. And one other clue to it being Cambridge rather than Oxford is that the people in the boat, which is called a punt, in Cambridge when you're punting you stand on the deck, which is the sensible way of doing it, but in Oxford they stand actually inside the boat on the other end. So there's a great kind of rivalry between the two towns in cricket and rugby and there's a boat race in London and they put their milk in their tea in different ways. So anyway, it goes back hundreds of years, thousands. But anyway, getting back to the two English scientists sitting in a Bedouin tent. They were actually developing a microscope in the MRC labs in Cambridge. And the two scientists were John White and Brad Amos. And John White just retired as the chair of our department here. And John had a biological problem to solve. In growing up in England, English science, a lot of the problems you solve you have to start off with paper and string and try and build the equipment yourself. And so John was a great equipment builder. And, oops, oh, I see. So John was a great tinkerer and an equipment builder. The biological problem he was trying to solve was back in 1985 many people were looking at cells and embryos using a technique called immunofluorescence microscopy. So they were using the power of immunology to label specific structures within cells so that people were able to raise specific antibodies to purified proteins and then tag a fluorescent dye to the antibody which was binding to the protein in the cell. This is an image that I took way back then of some cells growing in tissue culture. And here's my first reference to time lapse microscopy. A and B is two frames from a time lapse movie that I was making. So what I was doing I was filming some cells in tissue culture, and what actually is happening is I was filming the nucleus rotating. Sometimes the nucleus will actually up and rotate. And then I was trying to figure out what the cytoskeleton was doing during nuclear rotation. So I would film it and then fix it, kill it and then label it with a fluorescently labeled antibody. And then I would look at the distribution of the cytoskeleton within the cells and try and figure out how the cytoskeleton was related to the nuclear rotations. So with cells growing in tissue culture they're kind of very flat. So you can actually resolve structures in the fluorescent structures in cells growing in tissue culture. This is a schematic of the microscope that you'd use, a fluorescence microscope. It would have a very bright lamp, the HBO lamp house, which would supply a bright light. And it would go through an excitation filter which would filter out all of the wave lengths that you don't need. So a common dye that we would use would be something called fluorescein, which is excited at 480 nanometers. So that excitation filter would take everything else out except for 480, and then the light would bounce off of the dichroic beam splitter, go through the objective lens which would focus it on the specimen. Any labeled structures within the cell would fluoresce because they would be excited by the light. And then a longer wavelength light of less energy would be emitted. And then that would come back up along the green path. And the dichroic beam splitter would let that longer wavelength light through. And then the barrier filter would kind of clean up the signal so you'd only just get what wavelength that was coming off the fluorescein. And then you could see it by eye and that would take the photograph of it. Okay, so it worked well for cells growing in tissue culture, flattened cells. There's a fluorescence microscope. There's the light source at the back, and then you look at it by eye on the stage and everything. In such a microscope, the specimen is actually flooded with light. You can see there that the light is kind of shining off the specimen. So the specimen with significant 3-D structure, this is a HeLa cell. I don't know if you've read, many cancer cells, not all of them, but many cancer cells have a significant 3-D structure. That's one of their properties, they tend to round up in culture. This is a HeLa cell labeled with anti-tubulin, fluorescently labeled. You could actually focus it up and down. Looking by eye, you could see microtubules but taking a photograph was impossible. And then with an embryo, this was a nematode embryo and this was John White's specific problem. He was trying to follow the cell divisions within a nematode embryo. And the first 24 hours or so of embryogenesis there's many, many, many cell divisions. And so the cells kind of don't increase in size. They get smaller and smaller, but the structures get brighter and brighter because there's more cells. And you just can't see any structure. By the way, John had just spent the previous 14 years trying to figure out the neuro-connections of the 302 nerve cells in the nematode using electron microscopy. So John kind of knew about this technique called confocal microscopy. This is a pattern from 1957 from Marvin Minsky. And so rather than kind of flooding the whole specimen with light, the idea of confocal microscopy is to scan the specimen with a discrete beam of light. And in Minsky's microscopy apparatus, his aim was to look at cells in living brains, living nerve cells in brains. But the idea, all of the confocal microscopes of today are based on this basic design. But his idea was way before its time. The technology wasn't really available to him. So the technique is the light path, this would be a transmitted light microscope. The light source, a very bright light source on this side, the light passes through a pin hole, number 14, it's focused by a lens into the specimen, and then any light going through the specimen is again focused by another lens on a second pin hole, number 24, and then that is some kind of a detector that detects the point of light. So in that kind of setup we would just get a point of light. So in Minsky's apparatus the actual specimen moved. It was called a stage scanning mechanism. And he used a primitive tuning fork device. It worked but his visual display was in a telescope so the images were not very kind of impressive, especially to biologists who were using microscopes with stained specimens and could get really good images. In his pattern there is also an epifluorescent, a reflected light version which is incredibly similar to the first diagram I showed you of the fluorescence microscope. And in this is setup, the lens, you only use one lens, number 11, which got rid of a lot of alignment problems which you had with the transmitted system. So John knew about the stage scanning systems but they were kind of not good for biology because biological specimens tend to be wet and wobbly. So any wobble in stuff you just lose your resolution. If you move your camera taking a picture, then goodbye clarity. So what John and Brad would do is they were trying to figure out how they could actually scan the beam itself. And they, after a lot of kind of time in the Bedouin tent and the workshop and coming out and making kind of modifications to that apparatus, they hit upon a two-galvanometer scanning scanner. I'm not going to go into the details of it because I don't really understand them, but basically they were able to figure out a way of scanning the beam in a pretty efficient way. And it was about this time I moved to Madison to the microscopy facility, and we actually took delivery of one of the first four prototype microscopes. They hooked up with Bio-Rad. And this was the actual instrument. You can see the fluorescence microscope on the bottom here, the regular epifluorescence microscope. And then this was their scan head which produced a scanned beamed of the specimen. The detectors were photomultiplier detectors and the output was sent into a computer frame store, which at that time computer frame stores were kind of new too. So not only did they have to solve the problem of scanning the beam, but they also had to tame a computer programmer to write them the program and get the computer hooked up with a scanning beam. So you could imagine that the mirrors have to, the computer has to keep track of where the mirrors are in order to build the image on the screen. So as the beam is scanning across the specimen then the computer kind of puts the image onto the screen. And this system would produce an image of a fluorescent specimen in about one second. And I'd never heard of confocal microscopy when this thing came but everybody really wanted to use it. It was really amazing and fun technique to use. So John and Brad and Mike Fordham, who I guess was a computer programmer, published a paper in the Journal of Cell Biology which is an evaluation of confocal versus conventional imaging of biological structures by fluorescence microscopy. And this paper literally took the cell biological community by storm because what they did is they just went around and collected specimens and took a picture in the regular fluorescence microscope and then took a picture in the confocal microscope. So here's or HeLa cell which is how it looked in the regular microscope. You could actually see structures by eye because your eye has a better dynamic range than anything else. But looking in the confocal mode you start to see spindle microtubules. And if you move the fine focus just a little tad, you've got a different view. So the actual out-of-focus fluorescence was just gone. So these images were actually called optical sections because you were actually sectioned the structure by light rather than a physical means. So the paper really was just amazing because people just saw a real proof of the art. Here's John's nematode embryo in the regular epifluorescence microscope, and suddenly you're seeing structures. This is a sea urchin embryo. Again, what people used to see, and then what they say with the confocal microscope. One of the things they started to worry about was that the microtubules looked kind of wiggly and wavy in the confocal. And it turned out that their specimen preparation was not up to scratch. Once they kind of figured that out, the microtubules looked more normal. And so even just cells growing in tissue cultures the images were improved. You could start to see microtubules kind of going about the nucleus. I probably would have benefited from that with my nuclear rotation studies, my time lapse. So you could optically section, because as long as the specimen stayed still, it didn't move, you could take an image, the computer would move the stage by a certain set amount and then take another image, and so you could take a Z series section through something like a mitotic spindle and then send the images into another program to make a 3-D reconstruction. And a lot of these programs already existed for MRI and other data. So it's kind of like an MRI at the microscopic level. One of the advantages, I mean I'm just telling the story as if John White and Brad Amos were the only people doing this, but there were other confocal microscopes coming out. There was another, Brakenhoff in the Netherlands had one. Once the companies saw that these microscopes were so popular then Zeiss and Olympus and all the different companies got on board. But the great advantage of the White and Amos microscope, because of the way they were scanning the beam, was that you could adjust the pin hole in front of the detector so you could increase or decrease the optical section and thickness that you were taking. You could also, so this is a tissue taken at 4X where you can take kind of a map view of it. And this is taken at 40X where you can actually see cellular structures. But then you can decrease the area of specimen that you scan and increase magnifications but not changing the actual lens. So you can impart different magnifications to the same lens. And this is a little schematic that I kind of drew. A human egg is at 130 micrometers. Using a 4X with a pin hole open you can get the whole egg. With the pin hole kind of closed down you get a smaller optical section. So these are the kinds of structures that you can kind of visualize using the confocal system. The other thing you could do was you could label with two different immunofluorescent dyes. So the green one would be fluorescein and the red one would be rhodamine. And as long as the specimen stayed steady and your dichroic mirror angle was consistent and you used a lens which was corrected for chromatic aberration, then you could look at two different proteins in the same cell and then color them green and red. Another thing you could do would be to take an optical section, this is a nuclei in an epithelium and then there's another layer of nuclei at a different level and colorize them. So that's kind of the basic kind of story behind the development of the MRC microscope. And when I moved up to Sean Carroll's lab, I was hit with some more kind of interesting kind of problems. At that time we were working exclusively on fruit fly development. And this is a fruit fly embryo. And he were interested in the genes that control the setting up of the different segments of the body early on in embryogenesis. And I was working with a graduate student then, Jim Langeland, and he was working on the gene Harry which is expressed in seven stripes in the early embryo. And he was wanting to look at different genes that related to Harry at the same time. And at the time we could do double labeling so we could look at Harry and Kruppel, or Giant and Harry. And Jim said to me, well, Steve, do you think we could do three at the same time? And we had a krypton/argon laser on our microscope at the time which had three lines in it, 568, 488, remember that was the fluorescein line, and 647. And there was a knew dye that had just come out called cyanine 5, and Jim figured out the immunological staining protocol of staining three different immunofluorescently labeled proteins in one cell. And at that time it was really hard to get an image from a PC to a Mac. So we had to get another computer guy trained up and he wrote this program called RGB Merge where we took the three images and then it clunked through combining them and then it turned it into a TIF file. Most of the microscope companies have their own propriety formats too which then they didn't used to give out the code for that. And then we put it all together in Photoshop. And at that time it came up with this triple-labeled embryo. And we presented this at a fruit fly meeting, I don't know, 15 years ago, and it was really kind of quite well received. And one of the things that you can immediately see is the little round circles are actually nuclei and where the same gene is being expressed in a nucleus then you get an additive color. So where the green and the red gene are being expressed then you get yellow. So you get these kind of rainbow effects. And it also has biological meaning. And then a couple of years later Bio-Rad came out with this system that had three detectors on a light path whereby you could actually excite three fluorochromes at the same time simultaneously. And then on the screen you'd get the red, the green, and the blue image and the merged image at the touch of a button within a second. Whereas RGB Merge took about an hour to figure out one image but technology moves on. So the idea is that we can look at development in a series of fixed and stained cells and figure out which genes are being turned on. This is an early embryo where you kind of get a few domains coming up and then the seven stripes of Harry and then gastrulation occurs and then engrailed comes on and you get 14 stripes and these cells are starting to know where they're going to be in the embryo and they're getting their fates kind of worked out and some are dying because they're not in the right place. So people work on this kind of thing to figure out how a single cell develops into a multi, and we're using microscopy to figure these things out. So with fruit flies, the appendages develop in the, for want of a better word, in the larva, in the maggot, from these structures called imaginal discs, which are really great for confocal imaging. They're relatively flat sheets of cells. And this one here is going to be the wing. This is a wing imaginal disc. You can dissect them out and then stain them, fix and stain them, and then look at them in the microscope. This is the wing. This is the haltere. Fruit flies just have one pair of wings and then they have these reduced wings called halteres which are balance organs that flap along behind the wings. And then this is the antenna, antenna imaginal disc. So the study of this is called pattern formation where you could look at the gene expression patterns and figure out what's going on in development. There's an awful lot of genes turned on at the same time. And these are wing haltere imaginal discs. And Scott Weatherbee in the lab many years ago came up with these beautiful images of different genes being expressed all at the same time. And these are all fixed and stained different discs. So with the multiple labeling you can get an idea of how these fields of cells are being set up. So some of the biology here. This is called the wing pouch, this whole area. And the red gene is called vestigial, and that labels all of the cells that are going to form the wing. And then the blue gene, which kind of looks purple here, overlapping the red, is called apterous, and those are going to be the cells on the underside of the wing. And later on in development this whole structure is going to fold along this boundary here. And then the green gene is something called CI, and that's actually not a nuclear, one that's outside of the cell. You can see it's around all the little circles. All the circles nucleus. Many of these control genes are what are called transcription factors, but CI is on the outside. And again here's another wing imaginal disc. There apterous in green, and then these yellow dots here are going to form the sensory bristles. These cells are being told by a gene that they're going to be the sensory bristles. So then you can play around with the colors for kind of presentation purposes too, which is a lot of fun within Photoshop. And this purple/green image is generally recommended because a lot of people with red/green color blindness have trouble with these red and green ones. So if you're giving a presentation you want everybody to see it. And the purple/green ones are the best ones to use. Okay, so Sean gives a lot of talks, he gives a lot more talk now than he did, but he was at the University of North Carolina giving a talk, and he had just given his talk and he was being walked across campus by this really great biologist, Fred Nijhout, who is kind of the world's expert in butterfly biology. And he said, gee, Sean do you think all this fruit fly stuff, what do you think happens with butterflies? And to cut a long story short we started up a whole big project with butterfly development. Butterflies have a lot more colorful patterns on their wings than fruit flies. And this is a butterfly embryo. And just to pick a gene, this gene is called distalis in red. And it stains the distal portions, it tells the cells in the distal portions of the developing limbs that they're going to be the ends of limbs. And distalis is expressed in distal portions of fruit fly appendages too. Distalis is also expressed in centipede limbs and something very basic to the insects called -- limbs. So this kind of gives you a flavor of how we study evolution at the molecular level. We kind of look at the comparative gene expression patterns and find them. Evolution doesn't use different genes for different animals, it uses the same genes in different ways. So it's not the gene itself, it's what regulates the gene that we're finding at least for limb development. And then one of the great surprises is when we looked at imaginal discs. These butterfly imaginal discs, you dissect them out of a caterpillar. They have five instars. This is the fifth instar and you can actually see the shape of the wing starting to develop within this beautiful wing disc. The peripheral tissue actually dies off in the end. So it's kind of like a cookie cutter. The shape kind of comes out with a cookie cutter, and the cutter part just dies away. But you can see the venation developing here, and then the engrailed, the genes that set up the territories in the fruit fly wing imaginal disc are exactly the same as the ones that set up territories in butterfly imaginal discs. One of the great surprises was that when we looked at distalis in imaginal discs, it was actually setting up the positions of where the eye spot field is. So another theme of this Evo Devo is that genes actually recycle during development. They're used for different things, for controlling different things. Distalis was used back in the embryo for setting up the limb fields and now it's setting up the eye spot fields. And you can see it a little more clearly here. This is a 4X confocal image that I showed you before and. Then if I go to 60X, you can actually see the cells within a single eye spot field. And it's probably kind of a gradient of gene expression that's set up. And when Julie Gates in the lab looked at different wing discs from different species, only the ones with eye spots had the hot spots of distalis where the eye spot fields were. And if we looked at mutant cells this kind of finding held up. And then Craig Brunetti in the lab looked at some other genes, and then the actual color seemed to be related to other gene expression. So engrailed is in purple with distalis in, again engrailed is being recycled. So that kind of gives you a flavor of this is all fixed and stained tissues. There's another engrailed. So around 1990 or so people started to figure out that the green fluorescent, has anybody heard of green fluorescent protein? So many jellyfish have this protein that fluoresces. It happens to fluoresce, excited by the same wavelength as fluorescein at 488. And it's turned out, to cut a long story short and one Nobel Prize later, that figuring out the gene sequence to green fluorescent protein you can actually genetically engineer animals so that they, you can tag the gene for GFP to a gene of interest. So where the gene of interest is being expressed in a cell then the GFP will also be expressed. So if you shine a 488 nanometer light on a transgenic animal that has GFP, then anywhere that's green will be where that protein is being expressed. So here's John White's nematode embryo circa 2009. This is taken by Kevin O'Connell. And it's a transgenic nematode. This is GFP, a nuclear GFP, I think, or maybe it's a microtubule one, but anyway what Kevin is able to do is follow the cell divisions in exquisite detail. His microscope is fast enough to take Z series over time of the GFP being expressed within the mitotic spindles. This is called multi-dimensional imaging. So you can actually look at gene expression in living cells using GFP. Here's a fruit fly labeled with engrailed GFP. If you shine a 488 nanometer light on it, anywhere the engrailed gene is being expressed fluoresces green. And there's been much work in kind of engineering GFP, genetic mutants of GFP. So now there are many, many color variants available of these called fluorescent reporter proteins. In the Tsien lab in California what they did was streak a bacterial plate with different GFPs, different colored GFPs and came up with a nice little California scene. And Roger Tsien just recently won a Nobel Prize for his work with GFP, along with Marty Chalfie and a Japanese worker whose name I don't remember. So the modern confocal microscopes from the days of White and Amos, technology has moved on and the companies have incorporated, lasers have become a lot cheaper to put in. So rather than one laser, we have many different smaller, kind of diode lasers. But one of the great advances is what's called an AOTF which is an acousto optical filter, which allows you to kind of, rather than using a glass filter, you use an electronic filter that filters the light really, really fast, kind of really high rates of speed. So you can kind of collect images a lot more quickly. You can filter the light a lot more quickly. And this is kind of the light path within the system that we have in the lab. What I'm leading up to is kind of a neat new technique that I think is so cool in that now you can genetically engineer three different colors of reporter proteins into a single mouse. And so the cells in the mouse will randomly incorporate the three different reporters. So if one cell has three copies of CFP, it will fluoresce blue, in two CFPs one YFP will be fluoresced light blue and so on and so on. You can't control where the reporters are going to be expressed. But using the new kind of microscopes with the AOTFs and filter switching and kind of very sensitive detectors, scientists in Jeff Lichtman's lab have been able to delineate up to 99 different colors. So this is how the brain looks in such a mouse, which is called a brain-bow mouse. So rather than two or three you can look at 99 different. So this is really kind of incredibly artistically beautiful. But it's also used for kind of mapping, it's much easier to map a whole bunch of different colored cables than all gray cables. So people are using it to map out kind of neuro-networks. One thing Jeff Lichtman is finding out is it's incredibly hard to do this, but also that young animals start off complicated and the nerve becomes simpler as you get older rather than the other way around. But even using this beautiful brain-bow technique, they're calling these things connectomes. And they're actually having to go back to electron microscopy to actually figure out the whole, oh yeah, I did put the gray one in. So I've told you a little bit about how the confocal microscope was developed. How we're using it to study the mechanisms of evolution. And some of the new kind of techniques of imaging living cells. Microscopy is an incredibly fast moving and exciting kind of field right now. Not to short change you on the time lapse aspect of the thing, I want to just show you just how complicated development actually is. And there's a relatively new microscope that's come out called a swept field microscope where the field sweeps this way and that way. And Ernst Stelzer in Germany has come up with some amazing images of zebrafish development. These are GFP labeled zebrafish embryos. And what he's done is he's interested in where the cells actually go, and so he's able to add color to the actual direction of where the cells are going. So with that I'll just leave you with this. So it's a whole jumble of cell movements out of which comes some order. >> Is that a single embryo developing? >> This is a single zebrafish embryo developing. That's time going by at the top there. So this technique is incredibly kind to cells. They don't die. One of the problems with the early confocal microscopes was that they were very hard on living cells. And that's it. Well thank you very much for coming out tonight.

APPLAUSE