1 00:00:16,763 --> 00:00:17,430 ADAM MARTIN: OK. 2 00:00:17,430 --> 00:00:19,980 So I'm going to start out today's lecture 3 00:00:19,980 --> 00:00:23,190 on the wrong foot by quoting somebody 4 00:00:23,190 --> 00:00:25,800 that you guys probably don't know 5 00:00:25,800 --> 00:00:28,590 and who was a New York Yankee. 6 00:00:28,590 --> 00:00:34,580 So Yogi Berra, the famous Yankee catcher once said, 7 00:00:34,580 --> 00:00:38,130 "You can observe a lot by watching," OK? 8 00:00:38,130 --> 00:00:40,560 And that is very appropriate for biology 9 00:00:40,560 --> 00:00:46,350 because a lot of things in biology 10 00:00:46,350 --> 00:00:50,940 have been discovered simply by watching for them in cells 11 00:00:50,940 --> 00:00:55,620 or watching for them to happen at the molecular level. 12 00:00:55,620 --> 00:01:00,120 And so our ability to visualize and see 13 00:01:00,120 --> 00:01:04,050 what's going on inside cells and at the molecular level 14 00:01:04,050 --> 00:01:06,630 is really critical for the process 15 00:01:06,630 --> 00:01:09,510 of biological discovery. 16 00:01:09,510 --> 00:01:13,530 So today I'm going to tell you about tools, 17 00:01:13,530 --> 00:01:17,940 both sort of older tools but also kind of the cutting edge, 18 00:01:17,940 --> 00:01:21,210 for how biologists are really observing 19 00:01:21,210 --> 00:01:28,080 what's going on in living cells and in life in general. 20 00:01:28,080 --> 00:01:28,590 OK. 21 00:01:28,590 --> 00:01:34,290 So let me start by just having you guys think a little bit. 22 00:01:34,290 --> 00:01:40,530 What do you require of me to see what I write on the board? 23 00:01:40,530 --> 00:01:41,240 Yeah, Rachel. 24 00:01:41,240 --> 00:01:42,090 AUDIENCE: Light. 25 00:01:42,090 --> 00:01:42,450 ADAM MARTIN: What's that? 26 00:01:42,450 --> 00:01:43,470 AUDIENCE: Light. 27 00:01:43,470 --> 00:01:44,770 ADAM MARTIN: You need light. 28 00:01:44,770 --> 00:01:47,610 And what does the light help you to do? 29 00:01:52,470 --> 00:01:53,136 What's that? 30 00:01:53,136 --> 00:01:55,580 AUDIENCE: [INAUDIBLE] 31 00:01:55,580 --> 00:01:58,820 ADAM MARTIN: You need it to see the board. 32 00:01:58,820 --> 00:02:01,140 And so let's say the light's on, OK? 33 00:02:07,020 --> 00:02:11,150 Is that, can you read this? 34 00:02:11,150 --> 00:02:13,560 No, what's the problem? 35 00:02:13,560 --> 00:02:14,060 What's that? 36 00:02:14,060 --> 00:02:16,130 Size, right? 37 00:02:16,130 --> 00:02:18,440 So Natalie suggested size. 38 00:02:18,440 --> 00:02:21,020 Right, so one thing that you need 39 00:02:21,020 --> 00:02:24,230 is some amount of magnification, right? 40 00:02:28,700 --> 00:02:32,000 But let's take another-- 41 00:02:32,000 --> 00:02:34,550 let's say I do magnify this. 42 00:02:34,550 --> 00:02:36,610 What if I magnify it? 43 00:02:36,610 --> 00:02:40,970 And I'm going to start writing my notes on the board, right? 44 00:02:40,970 --> 00:02:41,700 How is this? 45 00:02:41,700 --> 00:02:42,200 Helpful? 46 00:02:46,310 --> 00:02:47,530 Jeremy, what's wrong? 47 00:02:47,530 --> 00:02:50,370 AUDIENCE: Differentiate [INAUDIBLE].. 48 00:02:50,370 --> 00:02:51,120 ADAM MARTIN: Yeah. 49 00:02:51,120 --> 00:02:52,830 You have to be able to distinguish 50 00:02:52,830 --> 00:02:56,400 different objects, in this case, these letters, right? 51 00:02:56,400 --> 00:03:00,600 So in addition to just magnifying it, 52 00:03:00,600 --> 00:03:04,410 you also need the structures to be far enough apart such 53 00:03:04,410 --> 00:03:06,190 that you can distinguish them. 54 00:03:06,190 --> 00:03:09,690 So you need what is known as resolution. 55 00:03:13,314 --> 00:03:14,220 OK. 56 00:03:14,220 --> 00:03:16,170 This was resolution. 57 00:03:16,170 --> 00:03:18,630 But this is resolution where the letters are actually 58 00:03:18,630 --> 00:03:20,370 resolved, OK? 59 00:03:20,370 --> 00:03:30,750 So structures need to be far enough apart 60 00:03:30,750 --> 00:03:31,980 so that you can resolve them. 61 00:03:36,960 --> 00:03:39,810 Now let's come back to Rachel's point, right? 62 00:03:39,810 --> 00:03:44,760 Why is it that we need light to see what's on the board? 63 00:03:47,650 --> 00:03:48,150 Right? 64 00:03:51,630 --> 00:03:53,145 What if I draw without pressing? 65 00:03:57,010 --> 00:03:57,510 Right? 66 00:03:57,510 --> 00:03:58,683 Is that-- yeah, Orey. 67 00:03:58,683 --> 00:03:59,850 AUDIENCE: You need contrast. 68 00:03:59,850 --> 00:04:01,142 ADAM MARTIN: You need contrast. 69 00:04:01,142 --> 00:04:02,090 Exactly, right? 70 00:04:02,090 --> 00:04:05,550 The light sort of gives you contrast between the chalk 71 00:04:05,550 --> 00:04:07,920 and the black part of the board. 72 00:04:07,920 --> 00:04:09,555 So you also need contrast. 73 00:04:13,440 --> 00:04:17,550 And contrast is the ability to-- 74 00:04:17,550 --> 00:04:21,750 the structures need to be differentiated 75 00:04:21,750 --> 00:04:23,090 from the background, OK? 76 00:04:23,090 --> 00:04:33,700 So structures need to be different from background. 77 00:04:40,700 --> 00:04:42,590 What else do you need to read my writing? 78 00:04:51,550 --> 00:04:52,050 Right? 79 00:04:52,050 --> 00:04:53,905 What if I were just to-- 80 00:05:04,110 --> 00:05:06,340 everyone can read that? 81 00:05:06,340 --> 00:05:07,780 What's wrong? 82 00:05:07,780 --> 00:05:08,290 Carlos? 83 00:05:08,290 --> 00:05:10,010 AUDIENCE: Needs to be clear and legible-- 84 00:05:10,010 --> 00:05:10,630 ADAM MARTIN: What's that? 85 00:05:10,630 --> 00:05:11,130 Yeah. 86 00:05:11,130 --> 00:05:13,780 I need to have, like, good handwriting, right? 87 00:05:13,780 --> 00:05:20,680 So I like to think of this as this is an aspect of sample 88 00:05:20,680 --> 00:05:22,720 preservation, OK? 89 00:05:22,720 --> 00:05:27,070 So there's a sample preservation issue. 90 00:05:27,070 --> 00:05:29,840 I can't butcher the letters and the words. 91 00:05:32,450 --> 00:05:32,950 OK. 92 00:05:32,950 --> 00:05:37,330 So in the process of doing all these other things, right, 93 00:05:37,330 --> 00:05:40,480 magnifying your image, resolving things in your image, 94 00:05:40,480 --> 00:05:45,130 and generating contrast, you can't destroy your sample such 95 00:05:45,130 --> 00:05:48,550 that it's illegible, basically, OK? 96 00:05:48,550 --> 00:06:00,880 So in this case, structure must be preserved while doing one 97 00:06:00,880 --> 00:06:02,620 through three on this list. 98 00:06:08,830 --> 00:06:10,120 OK. 99 00:06:10,120 --> 00:06:13,780 So I'm going to start with resolution. 100 00:06:13,780 --> 00:06:18,790 We'll talk about, what are the limits to resolving things 101 00:06:18,790 --> 00:06:20,500 in biological specimens. 102 00:06:26,440 --> 00:06:30,040 And in biology, the one instrument 103 00:06:30,040 --> 00:06:33,100 we use a lot is a microscope, OK? 104 00:06:33,100 --> 00:06:35,590 And a microscope is basically a collection 105 00:06:35,590 --> 00:06:40,210 of lenses that allow you to do many of the things I just 106 00:06:40,210 --> 00:06:43,240 drew on the board. 107 00:06:43,240 --> 00:06:46,510 I'll point out a couple sort of broad sort 108 00:06:46,510 --> 00:06:48,890 of types of microscopy. 109 00:06:48,890 --> 00:06:52,660 So the human eye up here can resolve up 110 00:06:52,660 --> 00:06:55,810 to about 100 to 200 microns, if you're 111 00:06:55,810 --> 00:06:59,230 looking at something at reading distant distance, right? 112 00:06:59,230 --> 00:07:02,200 But cells are like way smaller than that, right? 113 00:07:02,200 --> 00:07:05,060 So we need some sort of instrument 114 00:07:05,060 --> 00:07:07,630 that allows us to see things that 115 00:07:07,630 --> 00:07:12,340 is lower than the resolution limit of a human eye. 116 00:07:12,340 --> 00:07:15,760 And so one way is to use a light microscope 117 00:07:15,760 --> 00:07:20,140 where you're using visible light to observe your sample. 118 00:07:20,140 --> 00:07:21,730 And many of the images that you're 119 00:07:21,730 --> 00:07:23,740 seeing that we're showing you are 120 00:07:23,740 --> 00:07:28,210 from visible light microscopes. 121 00:07:28,210 --> 00:07:32,170 To see smaller things, type of microscope that's often used 122 00:07:32,170 --> 00:07:34,500 is an electron microscope. 123 00:07:34,500 --> 00:07:38,650 And electron microscopy allows us to observe structures 124 00:07:38,650 --> 00:07:42,880 all the way down to the sub nanometer level of resolution, 125 00:07:42,880 --> 00:07:44,020 OK? 126 00:07:44,020 --> 00:07:47,680 Now one limitation to the electron microscope 127 00:07:47,680 --> 00:07:50,560 is, you have to kill the sample, OK? 128 00:07:50,560 --> 00:07:55,150 So that can lead to artifacts and problems. 129 00:07:55,150 --> 00:08:00,100 And we'll discuss away at the end where light microscopy is 130 00:08:00,100 --> 00:08:02,590 being extended down to the limits 131 00:08:02,590 --> 00:08:07,430 that approximate that of an electron microscope. 132 00:08:07,430 --> 00:08:07,930 OK. 133 00:08:07,930 --> 00:08:12,500 So what determines then, the resolution of a microscope? 134 00:08:12,500 --> 00:08:16,510 And so I'm going to sort of define 135 00:08:16,510 --> 00:08:21,680 a measurement of resolution which I'll call the d-min, 136 00:08:21,680 --> 00:08:23,870 or minimum distance. 137 00:08:23,870 --> 00:08:33,410 And this will be the minimum distance between two points 138 00:08:33,410 --> 00:08:34,285 that can be resolved. 139 00:08:48,450 --> 00:08:48,950 OK. 140 00:08:48,950 --> 00:08:52,910 And what I showed you on that past slide is basically 141 00:08:52,910 --> 00:08:56,330 the limit on the right here is the d-min 142 00:08:56,330 --> 00:09:00,710 for these different types of microscopy techniques, OK? 143 00:09:00,710 --> 00:09:03,890 And what that means is, so the minimum distance would be, 144 00:09:03,890 --> 00:09:08,180 if your minimum distance is 200 nanometers, if two points are 145 00:09:08,180 --> 00:09:11,540 greater than 200 nanometers apart from each other, 146 00:09:11,540 --> 00:09:15,760 then you can distinguish them as two different objects. 147 00:09:15,760 --> 00:09:19,880 However, if they are closer than 200 nanometers together, 148 00:09:19,880 --> 00:09:21,710 you wouldn't be able to see that these 149 00:09:21,710 --> 00:09:23,180 are two different objects. 150 00:09:23,180 --> 00:09:26,750 They would be overlapping each other, OK? 151 00:09:26,750 --> 00:09:31,160 And typically, the d-min for a light microscope 152 00:09:31,160 --> 00:09:34,670 is around 200 nanometers, OK? 153 00:09:34,670 --> 00:09:41,075 And the d-min results, if we are to determine-- 154 00:09:41,075 --> 00:09:44,330 if I'm to tell you what determines 155 00:09:44,330 --> 00:09:48,350 this minimum distance, we have to think about a microscope. 156 00:09:48,350 --> 00:09:52,220 So here, I'm drawing a specimen here. 157 00:09:52,220 --> 00:09:54,140 I've just drawn my specimen. 158 00:09:54,140 --> 00:09:56,540 It's on a slide or a cover slip. 159 00:09:56,540 --> 00:09:57,830 Here's your specimen here. 160 00:10:00,590 --> 00:10:02,990 And you might have a light source 161 00:10:02,990 --> 00:10:04,070 to generate the contrast. 162 00:10:06,770 --> 00:10:11,150 And then there'd be some sort of objective lens 163 00:10:11,150 --> 00:10:15,710 underneath the slide and the specimen. 164 00:10:15,710 --> 00:10:17,390 So this would be an objective lens. 165 00:10:22,670 --> 00:10:25,790 Sorry about my sample preservation here. 166 00:10:25,790 --> 00:10:28,700 And so the light is going to be hitting the sample. 167 00:10:28,700 --> 00:10:32,120 And the objective lens will be collecting 168 00:10:32,120 --> 00:10:37,240 a cone of light that's going into the lens here, OK? 169 00:10:37,240 --> 00:10:41,430 And maybe I'll magnify this a bit so you can see it better. 170 00:10:41,430 --> 00:10:45,540 So I'm just going to magnify this region over here. 171 00:10:45,540 --> 00:10:47,840 So if this is my specimen, I'm going 172 00:10:47,840 --> 00:10:51,410 to draw the objective a little farther away this time. 173 00:10:51,410 --> 00:10:52,730 This is the objective. 174 00:10:55,330 --> 00:10:57,320 And the objective is able to capture 175 00:10:57,320 --> 00:11:00,020 a range of different angles of light 176 00:11:00,020 --> 00:11:02,690 that come from the specimen, OK? 177 00:11:02,690 --> 00:11:06,230 So it's collecting angles. 178 00:11:06,230 --> 00:11:09,030 And I'll just define here an angle theta, 179 00:11:09,030 --> 00:11:14,240 which is like the 1/2 angle of this whole cone of light, OK? 180 00:11:14,240 --> 00:11:16,880 So what determines the resolution limit 181 00:11:16,880 --> 00:11:27,470 in this type of system is first of all, the wavelength 182 00:11:27,470 --> 00:11:30,950 of the light that's used. 183 00:11:30,950 --> 00:11:35,150 So if you're using white light, that might be from 400 to 800. 184 00:11:35,150 --> 00:11:37,430 If you're exciting GFP, you're going 185 00:11:37,430 --> 00:11:41,780 to excite with a wavelength that's 488 nanometers. 186 00:11:41,780 --> 00:11:46,310 So it's usually around maybe between 450 and 550 nanometers 187 00:11:46,310 --> 00:11:48,780 for many different fluorescent proteins. 188 00:11:48,780 --> 00:11:49,280 OK. 189 00:11:49,280 --> 00:12:01,580 So lambda here is the wavelength of the excitation 190 00:12:01,580 --> 00:12:02,885 of the light you're using. 191 00:12:09,920 --> 00:12:18,080 And this is all then divided by 2 times the NA, which is 192 00:12:18,080 --> 00:12:20,510 a property of the subjective. 193 00:12:20,510 --> 00:12:27,890 And what NA is, NA stands for numerical aperture. 194 00:12:35,090 --> 00:12:38,810 And what that is, is basically the range of angles 195 00:12:38,810 --> 00:12:42,800 that this objective can collect, OK? 196 00:12:42,800 --> 00:12:50,330 So it's N sine theta, where theta is this angle here. 197 00:12:50,330 --> 00:12:53,690 So you get the best performance if the objective 198 00:12:53,690 --> 00:12:56,720 can collect all of the light that comes 199 00:12:56,720 --> 00:12:59,530 from this side of the sample. 200 00:12:59,530 --> 00:13:13,580 OK, and then N refers to the refractive index of the media 201 00:13:13,580 --> 00:13:15,020 that this light is going through. 202 00:13:18,300 --> 00:13:18,800 OK. 203 00:13:18,800 --> 00:13:21,860 And so if you have an objective and you 204 00:13:21,860 --> 00:13:24,920 have your sample in here and there's 205 00:13:24,920 --> 00:13:27,390 a slide and a cover slip-- 206 00:13:27,390 --> 00:13:29,300 I'll extend this out-- 207 00:13:29,300 --> 00:13:32,850 you often have immersion oil. 208 00:13:32,850 --> 00:13:35,030 There'd be some sort of immersion media here. 209 00:13:41,510 --> 00:13:44,000 And I don't know if you've ever used a microscope that's 210 00:13:44,000 --> 00:13:46,190 meant to be used with immersion media 211 00:13:46,190 --> 00:13:47,950 and you don't add that immersion media. 212 00:13:47,950 --> 00:13:51,350 But your image quality, if you don't add that media, 213 00:13:51,350 --> 00:13:53,510 is like really bad, right? 214 00:13:53,510 --> 00:13:57,800 And that's because you're affecting 215 00:13:57,800 --> 00:14:03,830 the numerical aperture of what this lens can collect. 216 00:14:03,830 --> 00:14:07,760 And therefore, you degrade your image quality, OK? 217 00:14:07,760 --> 00:14:11,120 But basically, the more light, the more angles of 218 00:14:11,120 --> 00:14:16,280 light that you collect, the higher the numerical aperture. 219 00:14:16,280 --> 00:14:19,560 And therefore, the lower this d-min is going to be. 220 00:14:19,560 --> 00:14:23,990 And so the greater you'll be able to resolve objects that 221 00:14:23,990 --> 00:14:26,045 are near each other in space. 222 00:14:29,850 --> 00:14:30,690 OK. 223 00:14:30,690 --> 00:14:35,190 So the take home message from all of this 224 00:14:35,190 --> 00:14:38,430 is that you notice that magnification is not 225 00:14:38,430 --> 00:14:39,600 a part of this. 226 00:14:39,600 --> 00:14:43,590 But the wavelength of the light is really critical, OK? 227 00:14:43,590 --> 00:14:46,470 So usually, this minimum distance 228 00:14:46,470 --> 00:14:48,870 ends up being the wavelength of light 229 00:14:48,870 --> 00:14:51,870 that you're using divided by 2. 230 00:14:51,870 --> 00:14:56,940 And this usually ends up being about 200 nanometers. 231 00:14:56,940 --> 00:15:00,720 So that's the diffraction limit of a light microscope, 232 00:15:00,720 --> 00:15:02,370 as you see up there. 233 00:15:02,370 --> 00:15:09,240 And this resolution is basically limited by the diffraction 234 00:15:09,240 --> 00:15:11,490 or behavior of light, OK? 235 00:15:11,490 --> 00:15:18,330 So light microscopy is limited by the diffraction of light. 236 00:15:28,200 --> 00:15:29,705 And it was thought for a long time 237 00:15:29,705 --> 00:15:31,080 that no matter what you do, you'd 238 00:15:31,080 --> 00:15:34,782 never be able to break this limit of about 200 nanometers. 239 00:15:34,782 --> 00:15:36,240 But at the end of the lecture, I'll 240 00:15:36,240 --> 00:15:39,630 tell you about some very smart people who figured out a way 241 00:15:39,630 --> 00:15:41,340 to actually break this limit. 242 00:15:41,340 --> 00:15:44,280 And we'll talk about how they were able to do that. 243 00:15:46,890 --> 00:15:50,100 Now I want to talk about a few other limitations 244 00:15:50,100 --> 00:15:54,180 of microscopy. 245 00:15:54,180 --> 00:15:57,510 And I'm starting by showing you this electron micrograph 246 00:15:57,510 --> 00:16:00,810 of the endoplasmic reticulum. 247 00:16:00,810 --> 00:16:04,080 And one important consideration you have to make 248 00:16:04,080 --> 00:16:10,230 is two dimensional versus three dimensional structure. 249 00:16:10,230 --> 00:16:13,440 So for electron microscopy, you basically 250 00:16:13,440 --> 00:16:16,770 cut the sample so you have a very thin slice of it. 251 00:16:16,770 --> 00:16:20,010 It's like slicing bread except these slices are 252 00:16:20,010 --> 00:16:24,270 on the order of 30 to 60 nanometers in height. 253 00:16:24,270 --> 00:16:27,990 And then you pass an electron beam through the sample 254 00:16:27,990 --> 00:16:32,400 after it's stained in order to visualize your specimen. 255 00:16:32,400 --> 00:16:34,380 And one thing you have to keep in mind 256 00:16:34,380 --> 00:16:37,890 is that, you're looking at a slice through this. 257 00:16:37,890 --> 00:16:41,400 And it doesn't give you three dimensional information, OK? 258 00:16:41,400 --> 00:16:46,110 So if we were to think about the endoplasmic reticulum, 259 00:16:46,110 --> 00:16:48,811 you might have an endoplasmic reticulum. 260 00:16:52,180 --> 00:16:55,900 And if you take an optical slice through this, 261 00:16:55,900 --> 00:16:58,360 then you would see something like this where you see each 262 00:16:58,360 --> 00:16:59,920 of the stacks individually. 263 00:17:09,230 --> 00:17:12,349 And so this might make you to conclude 264 00:17:12,349 --> 00:17:15,200 that the way that the endoplasmic reticulum is 265 00:17:15,200 --> 00:17:19,310 structured is it's kind of like a stack of pancakes, where each 266 00:17:19,310 --> 00:17:22,940 of these, you have a lipid bilayer 267 00:17:22,940 --> 00:17:25,210 surrounding a lumen of the ER. 268 00:17:25,210 --> 00:17:26,810 So right, the lumen would be inside 269 00:17:26,810 --> 00:17:29,303 like this for each of these. 270 00:17:29,303 --> 00:17:31,220 And they're just stacked on top of each other. 271 00:17:34,020 --> 00:17:38,000 And this is the textbook model for endoplasmic reticulum 272 00:17:38,000 --> 00:17:38,990 structure. 273 00:17:38,990 --> 00:17:44,330 But it was actually, if you don't consider this in 3D, 274 00:17:44,330 --> 00:17:45,410 you might miss something. 275 00:17:45,410 --> 00:17:49,850 And what was missed was reported in 2013 276 00:17:49,850 --> 00:17:52,490 in this paper, where rather than just taking 277 00:17:52,490 --> 00:17:57,140 a single slice, what they did is they made lots of slices. 278 00:17:57,140 --> 00:17:59,400 And they kept track of where they are. 279 00:17:59,400 --> 00:18:03,770 So they basically did a three dimensional reconstruction 280 00:18:03,770 --> 00:18:06,830 of the endoplasmic reticulum. 281 00:18:06,830 --> 00:18:10,040 And by imaging this other dimension, 282 00:18:10,040 --> 00:18:12,350 they came to a very different conclusion 283 00:18:12,350 --> 00:18:15,860 about how the endoplasmic reticulum is organized. 284 00:18:15,860 --> 00:18:19,460 And instead of being stacks of membranes 285 00:18:19,460 --> 00:18:22,430 on top of each other like pancakes, 286 00:18:22,430 --> 00:18:24,140 instead, it's a helicoid. 287 00:18:24,140 --> 00:18:28,940 So this is an ER from a professional secretory cell, 288 00:18:28,940 --> 00:18:30,990 like a salivary gland cell. 289 00:18:30,990 --> 00:18:34,130 And you can see in 3D, you get a very different picture 290 00:18:34,130 --> 00:18:36,830 of the organization of this organelle. 291 00:18:36,830 --> 00:18:42,920 It's actually wrapped around and spiraling membrane stacks, OK? 292 00:18:42,920 --> 00:18:45,530 So their model is that, basically 293 00:18:45,530 --> 00:18:48,890 the endoplasmic reticulum in some cell types 294 00:18:48,890 --> 00:18:54,290 basically has a parking garage like structure, OK? 295 00:18:54,290 --> 00:18:56,150 So in this case, in these cells, it 296 00:18:56,150 --> 00:19:01,950 seems like the ER Is basically a parking garage for ribosomes. 297 00:19:01,950 --> 00:19:02,450 OK. 298 00:19:02,450 --> 00:19:04,010 And you don't get that information 299 00:19:04,010 --> 00:19:06,110 unless you consider the three dimensional 300 00:19:06,110 --> 00:19:11,940 structure of the thing that you're looking at. 301 00:19:11,940 --> 00:19:15,170 So in addition to electron microscopy, 302 00:19:15,170 --> 00:19:19,220 there are techniques that involve light microscopy that 303 00:19:19,220 --> 00:19:21,590 involve optical sectioning. 304 00:19:21,590 --> 00:19:24,680 And so normally, if you're looking at fluorescence, 305 00:19:24,680 --> 00:19:26,930 if you're doing fluorescence microscopy, 306 00:19:26,930 --> 00:19:30,290 you'd be exciting the whole volume of your sample 307 00:19:30,290 --> 00:19:32,660 and exciting all of the fluorophores 308 00:19:32,660 --> 00:19:35,630 such that fluorescence from out of the focal plane 309 00:19:35,630 --> 00:19:37,610 would be getting into your image. 310 00:19:37,610 --> 00:19:41,360 And that would give you a much more hazy, unclear image. 311 00:19:41,360 --> 00:19:44,360 But there are techniques such as confocal fluorescence 312 00:19:44,360 --> 00:19:48,920 microscopy that allow you to exclude the out of focus light 313 00:19:48,920 --> 00:19:52,130 such that you're basically getting an optical section 314 00:19:52,130 --> 00:19:53,270 through your sample. 315 00:19:53,270 --> 00:19:56,570 And that can give you a much cleaner and better resolved 316 00:19:56,570 --> 00:19:57,700 image, OK? 317 00:20:00,800 --> 00:20:06,470 Now I want to talk a little bit about another dimension, which 318 00:20:06,470 --> 00:20:08,870 is time. 319 00:20:08,870 --> 00:20:12,950 And again, you're seeing images in textbooks. 320 00:20:12,950 --> 00:20:15,500 And usually, you're just seeing a single image. 321 00:20:15,500 --> 00:20:17,720 And whenever you see a single image, 322 00:20:17,720 --> 00:20:19,190 you have to think about how things 323 00:20:19,190 --> 00:20:23,820 might be changing in time in order to understand the system. 324 00:20:23,820 --> 00:20:27,800 So one example that I like here is shown here, 325 00:20:27,800 --> 00:20:29,750 where these are different proteins that 326 00:20:29,750 --> 00:20:32,300 are labeled in a yeast cell. 327 00:20:32,300 --> 00:20:34,970 And you see that these proteins form patches 328 00:20:34,970 --> 00:20:37,010 at the edge of the yeast cell. 329 00:20:37,010 --> 00:20:40,490 And some of these patches just contain the green protein, 330 00:20:40,490 --> 00:20:42,260 which is SLA1. 331 00:20:42,260 --> 00:20:47,300 And other patches contain just the red protein which is ABP1. 332 00:20:47,300 --> 00:20:52,280 And there's another class of patch which contains both, OK? 333 00:20:52,280 --> 00:20:58,573 So how might you interpret this fixed image over here? 334 00:20:58,573 --> 00:21:00,365 What might be one model you would conclude? 335 00:21:07,180 --> 00:21:10,270 Well, what was initially concluded 336 00:21:10,270 --> 00:21:14,320 from this type of experiment is you have three different types 337 00:21:14,320 --> 00:21:17,950 of patches that are distinct from each other in the cell 338 00:21:17,950 --> 00:21:21,790 because they have different molecular compositions, OK? 339 00:21:21,790 --> 00:21:23,830 And that was what was initially thought. 340 00:21:23,830 --> 00:21:26,320 But it was wrong because researchers 341 00:21:26,320 --> 00:21:31,710 had to really consider the aspect of time in this problem. 342 00:21:31,710 --> 00:21:34,900 And I'm going to show you a movie now over here 343 00:21:34,900 --> 00:21:37,930 where you're going to see this yeast cell. 344 00:21:37,930 --> 00:21:40,360 And now you have these different proteins 345 00:21:40,360 --> 00:21:44,830 tagged with different fluorescent proteins. 346 00:21:44,830 --> 00:21:49,000 And we can watch them in time as they progress 347 00:21:49,000 --> 00:21:53,110 through a stereotypic cycle. 348 00:21:53,110 --> 00:21:55,480 So what you're going to notice in this movie 349 00:21:55,480 --> 00:21:59,440 is that you see these green patches appear. 350 00:21:59,440 --> 00:22:04,240 And every single green patch at some point is joined by red. 351 00:22:04,240 --> 00:22:07,330 And then the green disappears and the red stays around. 352 00:22:07,330 --> 00:22:10,570 And then the thing disassembles, OK? 353 00:22:10,570 --> 00:22:14,980 So what was initially thought to be three different structures 354 00:22:14,980 --> 00:22:17,410 in the cell, eventually, it was found out 355 00:22:17,410 --> 00:22:19,900 that there was a dynamic process where 356 00:22:19,900 --> 00:22:22,870 this patch sort of matured over time 357 00:22:22,870 --> 00:22:24,970 and eventually disappeared into the cell. 358 00:22:24,970 --> 00:22:29,170 And what this process is, is actually endocytosis in yeast. 359 00:22:29,170 --> 00:22:30,910 And you're seeing different proteins 360 00:22:30,910 --> 00:22:33,670 getting recruited to endocytic vesicles 361 00:22:33,670 --> 00:22:37,910 as they bud from the plasma membrane of this yeast. 362 00:22:37,910 --> 00:22:38,410 OK. 363 00:22:38,410 --> 00:22:42,970 So that's just my caution in interpreting fixed images, 364 00:22:42,970 --> 00:22:45,040 because you have to think about how 365 00:22:45,040 --> 00:22:46,750 they might be changing in time. 366 00:22:49,280 --> 00:22:49,780 All right. 367 00:22:49,780 --> 00:22:51,715 So now we have to consider contrast. 368 00:23:00,748 --> 00:23:08,170 And in bright field microscopy, and bright field microscopy 369 00:23:08,170 --> 00:23:13,390 basically involves white light as your light source. 370 00:23:16,080 --> 00:23:22,090 And so you'd have a microscope that has a white light source. 371 00:23:22,090 --> 00:23:24,095 You might have your specimen here. 372 00:23:24,095 --> 00:23:24,970 Here's your specimen. 373 00:23:29,660 --> 00:23:32,020 Then you'd have some sort of detector 374 00:23:32,020 --> 00:23:33,150 at the end of your system. 375 00:23:38,070 --> 00:23:41,250 And there would be some objective lens in between, 376 00:23:41,250 --> 00:23:43,680 which I'm going to ignore for now. 377 00:23:43,680 --> 00:23:47,010 And so, for bright field microscopy, 378 00:23:47,010 --> 00:23:51,030 you're taking a sample and shining it with light. 379 00:23:51,030 --> 00:23:53,150 The light that doesn't go through your sample 380 00:23:53,150 --> 00:23:55,330 will go right through to the detector. 381 00:23:55,330 --> 00:23:56,760 And that's your background. 382 00:23:56,760 --> 00:24:00,390 But some of this light, the light that's going and hitting 383 00:24:00,390 --> 00:24:06,890 your sample, could be absorbed or it could be refracted. 384 00:24:06,890 --> 00:24:09,690 And it's the refraction or absorption 385 00:24:09,690 --> 00:24:13,020 of this light which generates the contrast for bright field 386 00:24:13,020 --> 00:24:14,380 microscopy. 387 00:24:14,380 --> 00:24:24,270 So in bright field, native structures in the cell 388 00:24:24,270 --> 00:24:34,290 absorb or refract light. 389 00:24:34,290 --> 00:24:36,390 And this is what generates the contrast. 390 00:24:40,300 --> 00:24:40,800 OK. 391 00:24:40,800 --> 00:24:44,910 So the images shown up here are bright field images of cells. 392 00:24:44,910 --> 00:24:47,490 And in each of these cases, there's no dye. 393 00:24:47,490 --> 00:24:49,200 There's no fluorescent protein. 394 00:24:49,200 --> 00:24:52,020 But you're able to see the outline of the cell. 395 00:24:52,020 --> 00:24:56,340 And you're able to see even individual organelles 396 00:24:56,340 --> 00:24:59,160 or structures within the cell that 397 00:24:59,160 --> 00:25:03,790 are interacting with the white light and generating contrast, 398 00:25:03,790 --> 00:25:05,040 OK? 399 00:25:05,040 --> 00:25:07,760 So that's one way to generate contrast 400 00:25:07,760 --> 00:25:12,600 is just hope that whatever is native in your cell generates 401 00:25:12,600 --> 00:25:13,500 the contrast. 402 00:25:13,500 --> 00:25:15,690 But there are also-- 403 00:25:15,690 --> 00:25:20,415 you can increase contrast in specimens by adding dyes. 404 00:25:25,380 --> 00:25:28,500 And if these dyes bind to specific structures 405 00:25:28,500 --> 00:25:32,940 like a membrane, then that will increase your contrast. 406 00:25:32,940 --> 00:25:37,450 So the electron microscopy images that I showed you-- 407 00:25:37,450 --> 00:25:41,460 so for electron microscopy, you generate contrast 408 00:25:41,460 --> 00:25:45,540 by adding a dye that is an electron-dense dye, which 409 00:25:45,540 --> 00:25:47,560 will bend the electron beam. 410 00:25:47,560 --> 00:25:53,340 And that's what allows you to get an image from an EM. 411 00:25:53,340 --> 00:26:04,230 So an EM contrast is from an electron-dense dye 412 00:26:04,230 --> 00:26:11,880 such as uranyl acetate or some other type of dye. 413 00:26:11,880 --> 00:26:19,050 Now, fluorescence microscopy, as Professor Imperiali showed you, 414 00:26:19,050 --> 00:26:21,300 involves taking a fluorescent molecule 415 00:26:21,300 --> 00:26:25,230 and attaching it to your protein of interest. 416 00:26:25,230 --> 00:26:29,410 So you're actually getting protein-specific contrast, 417 00:26:29,410 --> 00:26:30,285 which is very useful. 418 00:26:37,220 --> 00:26:38,250 OK? 419 00:26:38,250 --> 00:26:40,530 And the way a fluorescence microscope works 420 00:26:40,530 --> 00:26:44,550 is just shown up here where you might have a light source that 421 00:26:44,550 --> 00:26:46,650 has a range of wavelengths. 422 00:26:46,650 --> 00:26:50,220 And you can use a filter to select one. 423 00:26:50,220 --> 00:26:54,180 In the case of GFP, it would be blue light or 488 nanometer 424 00:26:54,180 --> 00:26:55,140 light. 425 00:26:55,140 --> 00:26:59,190 And that would then be shined onto your specimen. 426 00:26:59,190 --> 00:27:02,190 And then the light is absorbed by fluorophores 427 00:27:02,190 --> 00:27:04,050 in your specimen. 428 00:27:04,050 --> 00:27:07,320 Some energy is lost, such that the light that's 429 00:27:07,320 --> 00:27:12,570 emitted from GFP is a longer wavelength, in this case, 430 00:27:12,570 --> 00:27:14,010 green. 431 00:27:14,010 --> 00:27:16,290 And then you can filter again to make 432 00:27:16,290 --> 00:27:19,830 sure only the green light is what goes to the detector. 433 00:27:19,830 --> 00:27:23,490 So this is a very efficient way of generating contrast 434 00:27:23,490 --> 00:27:26,220 because you can use filters to select 435 00:27:26,220 --> 00:27:29,070 only the wavelength of light that is emitted 436 00:27:29,070 --> 00:27:32,410 from your fluorescent molecule. 437 00:27:32,410 --> 00:27:32,910 OK. 438 00:27:32,910 --> 00:27:37,470 Any questions about that and about my very short version 439 00:27:37,470 --> 00:27:41,120 of how fluorescence microscopy works? 440 00:27:41,120 --> 00:27:41,770 Yeah, Rachel? 441 00:27:41,770 --> 00:27:45,460 AUDIENCE: [INAUDIBLE] dichroic mirror? 442 00:27:45,460 --> 00:27:49,200 ADAM MARTIN: The dichroic mirror reflects certain wavelengths 443 00:27:49,200 --> 00:27:51,120 that are below a certain wavelength. 444 00:27:51,120 --> 00:27:54,600 And will pass wavelengths that are above a wavelength. 445 00:27:54,600 --> 00:27:58,800 So it will basically reflect the excitation light. 446 00:27:58,800 --> 00:28:04,140 But it will pass the emitted light, OK? 447 00:28:04,140 --> 00:28:08,310 And so, there are tons of these mirrors. 448 00:28:08,310 --> 00:28:12,180 Some are not dichroic, but they can 449 00:28:12,180 --> 00:28:13,920 reflect four different wavelengths 450 00:28:13,920 --> 00:28:16,200 and pass all other wavelengths. 451 00:28:16,200 --> 00:28:19,350 And so, this allows you to image multiple fluorophores 452 00:28:19,350 --> 00:28:21,810 at the same time, OK? 453 00:28:21,810 --> 00:28:24,000 The specifics aren't as important 454 00:28:24,000 --> 00:28:28,450 as the general concept of how this works. 455 00:28:28,450 --> 00:28:28,950 OK. 456 00:28:28,950 --> 00:28:32,880 Now I want to come back to the resolution limit. 457 00:28:32,880 --> 00:28:36,600 And I want to tell you about how we can beat it. 458 00:28:36,600 --> 00:28:39,240 So, beating. 459 00:28:39,240 --> 00:28:40,650 We all like winning. 460 00:28:40,650 --> 00:28:43,640 So beating the diffraction limit. 461 00:28:49,530 --> 00:28:56,940 And this is going to involve a type of microscopy that's 462 00:28:56,940 --> 00:29:00,870 really sort of been developed in the past decade, which is known 463 00:29:00,870 --> 00:29:02,910 as super resolution microscopy. 464 00:29:05,820 --> 00:29:11,070 So super resolution microscopy. 465 00:29:11,070 --> 00:29:12,690 And remember, I mentioned for you 466 00:29:12,690 --> 00:29:17,340 before that yes, electron microscopy can 467 00:29:17,340 --> 00:29:20,310 get you nanometer resolution. 468 00:29:20,310 --> 00:29:24,000 But you have to kill the cell. 469 00:29:24,000 --> 00:29:28,080 And also, it's hard to get protein specific contrast, 470 00:29:28,080 --> 00:29:28,830 right? 471 00:29:28,830 --> 00:29:31,050 So that kind of sucks because as biologists, 472 00:29:31,050 --> 00:29:33,630 usually we're interested in how things are 473 00:29:33,630 --> 00:29:36,660 functioning to stay, to live. 474 00:29:36,660 --> 00:29:39,570 So wouldn't it be great if we could somehow 475 00:29:39,570 --> 00:29:45,000 use light microscopy to get down into this nanometer range 476 00:29:45,000 --> 00:29:48,690 so that we can see how individual proteins are 477 00:29:48,690 --> 00:29:51,420 interacting with each other and organized 478 00:29:51,420 --> 00:29:53,310 at the nanometer level, OK? 479 00:29:53,310 --> 00:29:54,990 And so in the past decade, there's 480 00:29:54,990 --> 00:29:58,080 really been a revolution that's enabled 481 00:29:58,080 --> 00:30:02,700 us to do light microscopy with a resolution that gets down 482 00:30:02,700 --> 00:30:06,120 to the 10 or even single nanometer resolution. 483 00:30:10,970 --> 00:30:14,330 And there's a number of different super resolution 484 00:30:14,330 --> 00:30:15,530 techniques. 485 00:30:15,530 --> 00:30:19,680 I'm going to talk about just one of them. 486 00:30:19,680 --> 00:30:22,130 But both these techniques basically 487 00:30:22,130 --> 00:30:27,770 use the same concept, which is that they enable whoever's 488 00:30:27,770 --> 00:30:31,910 doing it to identify single molecules 489 00:30:31,910 --> 00:30:35,990 and define where those molecules are very precisely. 490 00:30:35,990 --> 00:30:39,410 And to turn fluorescent molecules on and off 491 00:30:39,410 --> 00:30:42,170 so that you can select individual molecules such 492 00:30:42,170 --> 00:30:44,360 that you can see them. 493 00:30:44,360 --> 00:30:48,410 So these are two different techniques. 494 00:30:48,410 --> 00:30:50,790 They're conceptually very similar. 495 00:30:50,790 --> 00:30:53,030 I'm going to focus on this one here. 496 00:30:53,030 --> 00:30:57,100 But it's pretty much similar to this one up here. 497 00:30:59,830 --> 00:31:05,300 And I just want to point out that one of our colleagues 498 00:31:05,300 --> 00:31:10,250 here at MIT, Ibrahim Cisse who's in the physics department, 499 00:31:10,250 --> 00:31:14,870 his lab builds these super resolution microscopes. 500 00:31:14,870 --> 00:31:17,810 And they're using super resolution microscopy 501 00:31:17,810 --> 00:31:21,470 to study the collective behaviors of proteins, 502 00:31:21,470 --> 00:31:26,650 in his case, during the function of gene expression. 503 00:31:26,650 --> 00:31:30,665 OK, so this is research that's actively being pursued at MIT. 504 00:31:33,740 --> 00:31:36,710 So let's just do a thought experiment again. 505 00:31:43,710 --> 00:31:44,210 OK. 506 00:31:44,210 --> 00:31:48,530 I'm drawing a single molecule or what you would see in an image 507 00:31:48,530 --> 00:31:50,840 if you were looking at a single molecule GFP. 508 00:31:54,550 --> 00:31:55,050 Great. 509 00:31:55,050 --> 00:31:56,070 Where is GFP here? 510 00:32:01,460 --> 00:32:02,293 Carmen? 511 00:32:02,293 --> 00:32:03,960 AUDIENCE: It's right there on the board. 512 00:32:03,960 --> 00:32:06,043 ADAM MARTIN: It's right there on the board, right? 513 00:32:06,043 --> 00:32:08,540 Is it here? 514 00:32:08,540 --> 00:32:09,664 What's that? 515 00:32:09,664 --> 00:32:11,110 AUDIENCE: I don't know. 516 00:32:11,110 --> 00:32:15,160 ADAM MARTIN: Who thinks GFP is right here? 517 00:32:15,160 --> 00:32:16,990 Who does not think GFP is right there? 518 00:32:21,550 --> 00:32:23,640 You have to be thinking one or the other. 519 00:32:23,640 --> 00:32:24,722 Yeah, Rachel. 520 00:32:24,722 --> 00:32:26,410 AUDIENCE: [INAUDIBLE] 521 00:32:26,410 --> 00:32:27,340 ADAM MARTIN: OK. 522 00:32:27,340 --> 00:32:30,920 So what Rachel says is that it's probably not right here. 523 00:32:30,920 --> 00:32:34,180 It's probably in the middle of this thing, right? 524 00:32:34,180 --> 00:32:38,140 And so if you're seeing a diffraction limited spot, 525 00:32:38,140 --> 00:32:40,760 you're going to get some sort of Gaussian of intensity, 526 00:32:40,760 --> 00:32:42,010 which I didn't draw well here. 527 00:32:42,010 --> 00:32:44,230 But it might be a little bit brighter in the center 528 00:32:44,230 --> 00:32:48,010 and drop off as you go towards the edge, right? 529 00:32:48,010 --> 00:32:50,440 So if I were to take a image intensity 530 00:32:50,440 --> 00:32:53,920 profile along the line here, you'd 531 00:32:53,920 --> 00:32:57,520 see something that kind of looked like a Gaussian, OK? 532 00:32:57,520 --> 00:33:02,500 And GFP, if there's a single molecule that you're imaging, 533 00:33:02,500 --> 00:33:06,520 should be right in the center of this Gaussian, OK? 534 00:33:06,520 --> 00:33:10,930 And so even though we're not seeing a spot, 535 00:33:10,930 --> 00:33:14,440 we're seeing a spot that its width here 536 00:33:14,440 --> 00:33:17,170 is diffraction limited. 537 00:33:17,170 --> 00:33:19,900 So this width is 200 nanometers. 538 00:33:19,900 --> 00:33:24,370 But if we can estimate where the molecule is 539 00:33:24,370 --> 00:33:27,250 in this region with nanometer precision, 540 00:33:27,250 --> 00:33:29,560 we could get a very accurate view 541 00:33:29,560 --> 00:33:33,190 of where this fluorescent molecule is, OK? 542 00:33:37,890 --> 00:33:40,720 So it relies on certain assumptions. 543 00:33:40,720 --> 00:33:42,210 The first assumption is that you're 544 00:33:42,210 --> 00:33:47,820 assuming we can see single fluorescent molecules. 545 00:33:47,820 --> 00:33:55,230 So that we visualize single fluorescent molecules. 546 00:33:58,530 --> 00:34:01,680 And that we can then estimate with some amount of precision 547 00:34:01,680 --> 00:34:05,610 the location of the molecule based 548 00:34:05,610 --> 00:34:09,389 on this diffraction limited sort of image that we get. 549 00:34:09,389 --> 00:34:20,190 So then we have to estimate the location based on the image. 550 00:34:24,360 --> 00:34:25,199 OK. 551 00:34:25,199 --> 00:34:28,500 And then our resolution is basically the error 552 00:34:28,500 --> 00:34:31,830 in fitting this curve, OK? 553 00:34:31,830 --> 00:34:42,810 So the error in the fitted position 554 00:34:42,810 --> 00:34:47,649 is equal to the standard deviation of this Gaussian. 555 00:34:47,649 --> 00:34:51,989 The standard deviation divided by the square root 556 00:34:51,989 --> 00:34:53,850 of the number of photons that you 557 00:34:53,850 --> 00:34:56,280 collected to get that image. 558 00:34:56,280 --> 00:34:59,445 So the square root of the number of photons. 559 00:35:02,420 --> 00:35:03,150 OK. 560 00:35:03,150 --> 00:35:06,330 And I just told you in the beginning of the lecture 561 00:35:06,330 --> 00:35:08,700 that this standard deviation is limited 562 00:35:08,700 --> 00:35:10,500 by the diffraction of light. 563 00:35:10,500 --> 00:35:12,870 So the standard deviation is going 564 00:35:12,870 --> 00:35:16,050 to be around 200 nanometers, right? 565 00:35:16,050 --> 00:35:18,330 But if you collect a lot of photons, 566 00:35:18,330 --> 00:35:23,760 you can accurately figure out where the fluorescent molecule 567 00:35:23,760 --> 00:35:27,110 is here if you know that it's a single molecule. 568 00:35:27,110 --> 00:35:27,610 OK? 569 00:35:27,610 --> 00:35:34,440 So the number of photons in a typical experiment 570 00:35:34,440 --> 00:35:37,990 is going to be around 10 to the fourth, OK? 571 00:35:37,990 --> 00:35:41,880 And so if 200 nanometers by 10 to the fourth, 572 00:35:41,880 --> 00:35:45,270 you're going to have sub nanometer resolution 573 00:35:45,270 --> 00:35:48,010 if you do the experiment right. 574 00:35:48,010 --> 00:35:49,350 OK? 575 00:35:49,350 --> 00:35:52,230 So you really need to see fluorescent molecules, however, 576 00:35:52,230 --> 00:35:54,420 in order to do this, OK? 577 00:35:54,420 --> 00:35:58,140 And the real breakthrough came with the realization 578 00:35:58,140 --> 00:36:01,530 that you could combine this type of fitting 579 00:36:01,530 --> 00:36:05,160 to estimate the position of single molecules 580 00:36:05,160 --> 00:36:08,700 with a certain type of fluorescent protein 581 00:36:08,700 --> 00:36:13,690 where you can turn the protein on and off stochastically, OK? 582 00:36:13,690 --> 00:36:14,190 OK. 583 00:36:14,190 --> 00:36:16,710 So we need one more component which 584 00:36:16,710 --> 00:36:25,470 is a photo-activatable fluorescent protein, 585 00:36:25,470 --> 00:36:31,170 in this case, the first one was photo-activatable GFP PA-GFP. 586 00:36:35,730 --> 00:36:39,900 And PA-GFP is a fluorescent protein like GPF. 587 00:36:39,900 --> 00:36:41,760 It's genetically encoded. 588 00:36:41,760 --> 00:36:44,370 But when it matures, it's not fluorescent. 589 00:36:44,370 --> 00:36:46,530 It's in a dark state, OK? 590 00:36:46,530 --> 00:36:50,240 So when it matures, it's dark. 591 00:36:50,240 --> 00:36:53,280 It has a dark state. 592 00:36:53,280 --> 00:36:55,920 And it starts out in this dark state. 593 00:36:55,920 --> 00:36:57,810 But you can turn it on. 594 00:36:57,810 --> 00:37:00,850 And you can turn it on with light. 595 00:37:00,850 --> 00:37:02,850 So that's where the photo activation 596 00:37:02,850 --> 00:37:04,830 is, because you're able to photo activate 597 00:37:04,830 --> 00:37:06,810 this fluorescent molecule. 598 00:37:06,810 --> 00:37:13,110 And you can photo activate with sort of UV light or 405 599 00:37:13,110 --> 00:37:16,260 nanometer light. 600 00:37:16,260 --> 00:37:20,190 And so that's not normally the excitation wavelength. 601 00:37:20,190 --> 00:37:26,250 But if you shine your sample with 405 nanometer light, 602 00:37:26,250 --> 00:37:29,910 it will convert some set of your molecules 603 00:37:29,910 --> 00:37:33,440 into the now fluorescent state, OK? 604 00:37:33,440 --> 00:37:39,060 So this then causes it to be fluorescent. 605 00:37:42,288 --> 00:37:43,830 And now it's going to be lighting up. 606 00:37:51,160 --> 00:37:51,660 OK. 607 00:37:51,660 --> 00:37:54,690 And I want to thank Professor Cisse because he gave me 608 00:37:54,690 --> 00:37:56,640 the next slide which I think nicely 609 00:37:56,640 --> 00:37:58,535 shows how this technique works. 610 00:38:01,050 --> 00:38:06,270 So the way you can get super resolution 611 00:38:06,270 --> 00:38:08,580 is you can't be looking at all your fluorophores 612 00:38:08,580 --> 00:38:11,130 at once because they're not far enough apart 613 00:38:11,130 --> 00:38:13,170 and they'll all bleed together so that you 614 00:38:13,170 --> 00:38:15,390 get a bad image, right? 615 00:38:15,390 --> 00:38:17,670 So this would be your conventional diffraction 616 00:38:17,670 --> 00:38:20,280 limited image where all the fluorophores-- there's 617 00:38:20,280 --> 00:38:21,790 about 20 fluorophores here. 618 00:38:21,790 --> 00:38:24,270 And you can see, you can't see individual fluorophores 619 00:38:24,270 --> 00:38:27,930 and you can't see what that says, OK? 620 00:38:27,930 --> 00:38:31,290 But if you take a divide and conquer approach 621 00:38:31,290 --> 00:38:34,320 with this, if you have a photo-activatable GFP, 622 00:38:34,320 --> 00:38:36,910 you don't need to look at them all at once. 623 00:38:36,910 --> 00:38:40,710 You can just look at three to start, OK? 624 00:38:40,710 --> 00:38:43,860 So now if you only activate a small subset 625 00:38:43,860 --> 00:38:47,580 and you ensure that you're activating it at a frequency 626 00:38:47,580 --> 00:38:50,370 such that they're well resolved from each other, 627 00:38:50,370 --> 00:38:52,920 then you can distinguish that there are 628 00:38:52,920 --> 00:38:55,150 three single molecules here. 629 00:38:55,150 --> 00:38:57,330 You can fit where they are. 630 00:38:57,330 --> 00:39:00,330 And now you know where they are with nanometer precision. 631 00:39:00,330 --> 00:39:02,820 So you know where those are. 632 00:39:02,820 --> 00:39:05,400 And then you want to look at other molecules. 633 00:39:05,400 --> 00:39:07,050 And so you have to get rid of these. 634 00:39:07,050 --> 00:39:11,190 And so what you would do is to bleach them. 635 00:39:11,190 --> 00:39:14,820 And bleaching is to use light to basically damage 636 00:39:14,820 --> 00:39:20,190 the fluorophores and get it to no longer fluoresce, OK? 637 00:39:20,190 --> 00:39:25,230 So this process is going to involve an iterative photo 638 00:39:25,230 --> 00:39:42,270 activation followed by measuring and fitting the image 639 00:39:42,270 --> 00:39:44,790 so that you can basically determine 640 00:39:44,790 --> 00:39:48,720 where each single molecule is in your image. 641 00:39:48,720 --> 00:39:53,400 And then ending with bleaching to get rid of the fluorophores 642 00:39:53,400 --> 00:39:57,060 you just turned on so that everything is now dark again. 643 00:39:57,060 --> 00:40:00,390 And then you repeat this process iteratively 644 00:40:00,390 --> 00:40:06,030 to collect all of the single molecules that you can, OK? 645 00:40:06,030 --> 00:40:10,140 So in this case, we just got these three molecules. 646 00:40:10,140 --> 00:40:12,180 We would then want to bleach them 647 00:40:12,180 --> 00:40:14,940 so that we're now going to look at different fluorescent 648 00:40:14,940 --> 00:40:16,013 molecules. 649 00:40:16,013 --> 00:40:17,430 And we'll turn on a certain number 650 00:40:17,430 --> 00:40:19,920 of other fluorescent molecules. 651 00:40:19,920 --> 00:40:21,570 Here you see four. 652 00:40:21,570 --> 00:40:22,200 Here are two. 653 00:40:22,200 --> 00:40:23,867 They're a little close together, but you 654 00:40:23,867 --> 00:40:25,050 can see that there are two. 655 00:40:25,050 --> 00:40:25,980 Here are another two. 656 00:40:25,980 --> 00:40:30,360 They're close together, but you see two clear intensity peaks. 657 00:40:30,360 --> 00:40:31,850 And so you can fit those four. 658 00:40:31,850 --> 00:40:35,880 Now you have four more molecules to make up your image. 659 00:40:35,880 --> 00:40:39,480 You bleach them, excite or activate five more. 660 00:40:39,480 --> 00:40:41,640 Here are five fluorescent molecules. 661 00:40:41,640 --> 00:40:42,870 You can fit those. 662 00:40:42,870 --> 00:40:44,700 Determine their positions. 663 00:40:44,700 --> 00:40:47,130 And you just do this iteratively over and over 664 00:40:47,130 --> 00:40:51,130 again till you get as many molecules as you can. 665 00:40:51,130 --> 00:40:55,440 And at the end, you basically add all these images together 666 00:40:55,440 --> 00:41:00,840 to get the final super resolution image, OK? 667 00:41:00,840 --> 00:41:04,560 So this is an iterative process where the photo activation 668 00:41:04,560 --> 00:41:08,220 allows you to image single molecules such 669 00:41:08,220 --> 00:41:13,043 that you can see where they are with nanometer precision. 670 00:41:13,043 --> 00:41:14,460 And then you add them all together 671 00:41:14,460 --> 00:41:17,680 to get a super resolution image. 672 00:41:17,680 --> 00:41:20,890 Here is an example of this in practice. 673 00:41:20,890 --> 00:41:22,680 And this is the storm technique which 674 00:41:22,680 --> 00:41:25,170 doesn't involve a photo activatable fluorescent 675 00:41:25,170 --> 00:41:28,110 protein, but involves organic dyes blinking. 676 00:41:28,110 --> 00:41:32,160 And the concept is basically the same. 677 00:41:32,160 --> 00:41:35,430 And here you see a conventional image of an axon. 678 00:41:35,430 --> 00:41:37,830 And it's labeled with this beta-II spectrin. 679 00:41:37,830 --> 00:41:41,430 And you see beta-II spectrin is continuous. 680 00:41:41,430 --> 00:41:43,620 And it's staining in this axon. 681 00:41:43,620 --> 00:41:46,700 But if you look at the super resolution image, what 682 00:41:46,700 --> 00:41:49,080 you see is the beta spectrin actually 683 00:41:49,080 --> 00:41:53,190 has this repeated periodic pattern along the axon. 684 00:41:53,190 --> 00:41:55,770 And this is a cytoskeletal element 685 00:41:55,770 --> 00:41:58,260 that's basically present in rings up 686 00:41:58,260 --> 00:42:01,595 and down the axons of neurons. 687 00:42:01,595 --> 00:42:02,970 And you can kind of think of this 688 00:42:02,970 --> 00:42:05,760 as like the axon is a vacuum hose, where 689 00:42:05,760 --> 00:42:08,130 you have these rigid sort of rings that 690 00:42:08,130 --> 00:42:10,710 are aligned all along the axon. 691 00:42:10,710 --> 00:42:13,980 But because it's repeated and you have intervening areas 692 00:42:13,980 --> 00:42:16,680 without much cytoskeleton, you can kind of 693 00:42:16,680 --> 00:42:18,300 think of it as a way for the axon 694 00:42:18,300 --> 00:42:24,150 to be both rigid but also flexible and maneuverable. 695 00:42:24,150 --> 00:42:29,010 I just wanted to point out that several super resolution 696 00:42:29,010 --> 00:42:34,350 techniques were recognized in the 2014 Nobel 697 00:42:34,350 --> 00:42:36,750 Prize in chemistry. 698 00:42:36,750 --> 00:42:41,490 Eric Betzig on the left here developed the approach 699 00:42:41,490 --> 00:42:46,200 using the photo activatable GFP that I described to you. 700 00:42:46,200 --> 00:42:50,000 And two others, Stefan Hell and W.E. Moerner 701 00:42:50,000 --> 00:42:54,050 were awarded it for other types of super resolution technique. 702 00:42:54,050 --> 00:42:57,920 If you get a chance, you should go to the Nobel Prize website 703 00:42:57,920 --> 00:43:04,160 and listen to Eric Betzig's Nobel lecture. 704 00:43:04,160 --> 00:43:06,170 He has a very interesting story. 705 00:43:06,170 --> 00:43:09,230 And part of it involved how he managed 706 00:43:09,230 --> 00:43:11,280 to develop this technique. 707 00:43:11,280 --> 00:43:15,360 And he actually developed it in the living room of his friend. 708 00:43:15,360 --> 00:43:17,810 So this is actually one of the first super resolution 709 00:43:17,810 --> 00:43:19,520 microscopes. 710 00:43:19,520 --> 00:43:21,240 Here's the microscope and here you see-- 711 00:43:21,240 --> 00:43:22,820 I love this chair. 712 00:43:22,820 --> 00:43:25,790 But you see, you basically have this microscope 713 00:43:25,790 --> 00:43:28,410 in this guy's living room. 714 00:43:28,410 --> 00:43:30,290 So if you want to hear more about this story, 715 00:43:30,290 --> 00:43:31,770 listen to his Nobel lecture. 716 00:43:31,770 --> 00:43:33,950 He's a really funny guy and you get 717 00:43:33,950 --> 00:43:36,440 a sense of how science really works 718 00:43:36,440 --> 00:43:39,080 where you get this unemployed guy like building 719 00:43:39,080 --> 00:43:41,180 a microscope in his friend's living room 720 00:43:41,180 --> 00:43:42,500 and then wins a Nobel Prize. 721 00:43:45,230 --> 00:43:49,640 And just one reminder to end today, remember your news brief 722 00:43:49,640 --> 00:43:52,670 is due this Friday, November 30th. 723 00:43:52,670 --> 00:43:54,830 If you need help on selecting a topic, 724 00:43:54,830 --> 00:43:58,020 please see a member of our staff, 725 00:43:58,020 --> 00:44:01,400 including Professor Imperiali or myself. 726 00:44:01,400 --> 00:44:03,380 And so, good luck with that. 727 00:44:03,380 --> 00:44:04,040 Thank you. 728 00:44:04,040 --> 00:44:05,920 I'm all set.