Top
SCIENCE. STORIES. IMPACT.

iBiology

Bringing the World's Best Biology to You

Chemical Glycobiology: Study of Glycans and the Immune System

Part 1: Chemical Glycobiology

00:00:01.09 Hi, and welcome to the iBioLecture called Chemical Glycobiology.
00:00:06.16 My name is Carolyn Bertozzi, and I'm a professor of Chemistry
00:00:10.24 and Molecular and Cell Biology at the University of California at Berkeley
00:00:15.16 and also an investigator of the Howard Hughes Medical Institute.
00:00:18.16 I am a chemist by training, who became interested in the biology of sugars
00:00:24.04 which is what the term glycobiology means
00:00:27.07 when I was in graduate school, and then later as a postdoctoral fellow.
00:00:31.29 And my laboratory research at UC Berkeley
00:00:35.23 seeks to combine chemistry and biology together to understand
00:00:39.22 what sugars are doing in the human body.
00:00:43.08 So, let me begin by giving you an introduction
00:00:46.23 as to why I am so interested in the biology of sugars
00:00:50.12 along with all the students and postdoctoral fellows
00:00:53.12 that work with me in the laboratory.
00:00:55.25 It turns out that around the turn of the millennium
00:00:58.10 there was a very exciting sequence of breakthroughs in biology
00:01:01.14 having to do with sequencing genomes.
00:01:05.14 So in the early days of genome sequencing
00:01:07.27 the first eukaryotic organism to be characterized in this way
00:01:12.11 was the budding yeast. And one of the great surprises
00:01:16.07 from the sequence of the yeast genome
00:01:18.14 was that fact that it only contains about 6,000 genes
00:01:22.29 and that's not a very large number. In fact,
00:01:26.13 before the genome was completed there were some estimates
00:01:29.05 that there would be far more genes required
00:01:32.14 to encode all of the interesting functions that eukaryotes perform.
00:01:36.18 So the number 6000 seemed very small.
00:01:39.08 Of course, the budding yeast
00:01:40.26 is a relatively simple eukaryotic organism.
00:01:44.11 It's a single celled organism.
00:01:46.14 And there was this idea that more cells would require more genes.
00:01:50.23 And that seemed to be the case as
00:01:52.14 more and more genomes were decoded.
00:01:55.11 So, when the C. elegans genome was finally sequenced,
00:01:58.26 that genome had about 15,000 genes,
00:02:02.11 which is a larger number of genes
00:02:04.07 and it seemed to make sense.
00:02:05.12 And then the next major organism to have its genome sequenced
00:02:09.05 was Drosophila, or the simple fruit fly.
00:02:11.23 which had about 20,000 genes,
00:02:13.23 and all the while scientists were working on the human genome
00:02:17.14 operating under the assumption that the human genome
00:02:20.05 would be much larger than any of these other model organisms.
00:02:24.28 In fact, some early scientists estimated that
00:02:27.15 the human genome might have over 100,000 genes.
00:02:30.20 Well, around the turn of the millennium there was a big surprise
00:02:34.04 yet again in the world of genome sequencing
00:02:36.06 when the human genome was finally completed
00:02:38.29 and it turns out that it's not that much larger than the fly genome
00:02:44.03 or the C. elegans genome, and really not even that
00:02:46.11 much larger than the yeast genome.
00:02:47.28 There were only around 25,000 genes encoded in the human genome.
00:02:53.00 These are the genes that encode for proteins, and by the way,
00:02:55.20 I think that nobody would be more surprised than Charles Darwin himself
00:02:59.22 to discover that we really don't have that many genes.
00:03:03.14 So then the next question became,
00:03:05.15 given that we only have about 25,000 genes,
00:03:08.23 how is it that human beings can be so biologically complicated
00:03:13.20 compared to some of these other simple model organisms.
00:03:17.13 Well, in one of the major publications in 2001
00:03:21.24 that put forth the sequence of the human genome
00:03:24.09 Craig Venter and his colleagues made this statement
00:03:27.27 which is that the finding that the human genome
00:03:30.04 contains fewer genes than previously predicted
00:03:32.22 might be compensated for by combinatorial diversity
00:03:36.13 generated at the level of post-translational modification of proteins.
00:03:42.02 So put another way, what happens in the higher,
00:03:46.18 more complicated organisms is that
00:03:48.29 the proteins encoded by the genome
00:03:51.14 are modified in very complicated and diverse ways
00:03:56.01 to create a lot more biological complexity
00:03:58.20 than one might predict just by simply counting the genes in the genome.
00:04:02.15 And modifications of those proteins are what we call
00:04:05.16 the post-translational modifications.
00:04:08.02 Well, it turns out that one of the most
00:04:11.08 complicated of those post-translational modifications
00:04:14.23 is the attachment of sugars to those proteins.
00:04:18.07 And that's the process that we call glycosylation
00:04:21.26 and by the way, you'll see that prefix glyco again and again and again
00:04:26.07 throughout these lectures.
00:04:27.00 That is the Greek prefix for sugar.
00:04:29.05 It turns out that many of the proteins that are glycosylated
00:04:34.20 are the proteins attached to the membranes of cells
00:04:38.12 In fact, they reside on the surface of cells,
00:04:42.00 and we call those sugar modified proteins, glycoproteins.
00:04:46.22 And an example of one of those glycoproteins is shown here
00:04:49.16 in cartoon form, where the protein is this blue structure
00:04:52.11 that is anchored into the membrane of the cell.
00:04:54.16 We also have lipids in our cell membranes
00:04:58.21 that have sugars attached to them,
00:05:00.00 so for example this cartoon illustrates what we call a glycolipid.
00:05:04.18 And the sugar part of the glycoprotein or the glycolipid
00:05:08.14 is what we might refer to scientifically as a glycan.
00:05:12.16 Glycan is just another word for a complex sugar molecule.
00:05:16.19 So, because of all of these glycoproteins and glycolipids,
00:05:21.12 on the surface of our cells, we basically can think
00:05:24.14 of our cells as having a sugar coating.
00:05:27.01 And in fact, that's why on the very first slide
00:05:29.17 of this lecture, I had a cartoon of M&Ms
00:05:32.22 because in many ways you can think of our cells as being like M&Ms.
00:05:37.13 They are coated with sugar molecules.
00:05:40.20 Now, that to me is fascinating.
00:05:43.19 and of course it begs the question,
00:05:45.11 what is the function of all of these sugar molecules
00:05:48.06 that are attached to the proteins and the lipids
00:05:50.27 and decorating the surfaces of our cells?
00:05:53.24 And that question, what are the functions of the sugars?
00:05:56.27 That is the question that is embodied by the field of glycobiology.
00:06:02.00 Now, it turns out that unlike M&Ms,
00:06:07.01 which have a fixed sugar coating, that never changes,
00:06:10.14 of course, until you eat the M&M,
00:06:13.04 the sugar coating on the surface of our cells does change.
00:06:16.12 And it can change dramatically as our cells change their state.
00:06:20.16 Now, we have terms that we use to describe
00:06:23.15 the collections of these sugars.
00:06:25.11 In fact, we have this term glycome.
00:06:27.28 You can think of this term glycome as being analogous
00:06:31.26 to other terms that you might be more familiar with.
00:06:34.09 Genome, for example.
00:06:36.07 The genome is the complete collection
00:06:38.17 of all of the genes in our cells.
00:06:40.18 And maybe you have heard the term proteome.
00:06:43.22 The proteome is the complete collection of the proteins that our cells are making.
00:06:48.15 Well, in the field of glycobiology,
00:06:51.01 we use the term glycome to describe
00:06:54.09 the totality of glycans produced by a cell.
00:06:58.08 And as I am showing here in this cartoon, the glycome is dynamic.
00:07:02.13 So, in other words, when a cell has one particular state
00:07:06.04 it will have a particular collection of glycans
00:07:09.08 which are shown by these various cartoon structures.
00:07:11.21 But if the cell undergoes a physiological change,
00:07:15.01 the collection of glycans can change.
00:07:18.21 Some of the structures might become more abundant
00:07:20.29 or less abundant, and there might be entirely new
00:07:24.03 glycan structures that weren't present in the original state.
00:07:27.21 So it's the dynamic nature of the glycome
00:07:31.00 that is so interesting from the perspective of
00:07:33.15 understanding what these sugars are doing in biology.
00:07:37.08 Just to give you some examples
00:07:39.03 of situations in which the glycome changes,
00:07:42.17 if you look at the complete collection of glycans
00:07:46.14 when a cell is in an embryonic state,
00:07:49.04 so it's a cell that has just formed, you'll see
00:07:51.21 a certain collection that is quite different
00:07:54.05 from the collection of glycans when
00:07:56.00 that cell is in what we call a differentiated state.
00:07:59.06 In other words, when the cell has chosen to become a muscle cell,
00:08:03.13 or a neuron, or a skin cell.
00:08:05.12 Each of those cell types has its own distinct glycome,
00:08:10.03 which is different from the embryonic cell that it came from.
00:08:13.00 It turns out that the glycome also changes during diseases.
00:08:18.20 So if you look at the complete collection
00:08:21.00 of glycans when a cell is in a healthy, normal state,
00:08:24.23 it is different from the collection of glycans
00:08:27.18 when that cell becomes a cancer cell.
00:08:29.13 Now this is a very interesting discovery
00:08:32.25 from the perspective of clinical medicine.
00:08:36.09 Because if we could actually see how the glycome
00:08:40.16 changes on cells in the human body,
00:08:42.17 we might be able to detect cancers.
00:08:46.00 And for that reason, as you'll see later in my second lecture
00:08:50.07 we are very interesting in developing tools to image the glycome,
00:08:54.25 to see the glycome inside a living body.
00:08:58.01 So, keep that in mind.
00:09:00.19 Now let me tell you a little bit about where
00:09:03.11 these glycans come from inside the cell.
00:09:06.03 Because they are the products
00:09:07.14 of fairly complicated metabolic pathways.
00:09:11.06 They are products of metabolism,
00:09:13.15 and all of that begins with the uptake of simple sugars into the cells.
00:09:18.13 And these simple sugars we call monosaccharides,
00:09:21.27 and they are denoted by these little colored balls.
00:09:24.09 These are simple sugar molecules.
00:09:26.19 So you eat food. The food has sugars in it,
00:09:30.01 and your cells take up those simple sugars. Now inside the cell,
00:09:35.01 the monosaccharide building blocks are processed by enzymes.
00:09:39.03 Eventually, those building blocks are sent into
00:09:42.17 subcellular compartments that we call
00:09:44.25 the endoplasmic reticulum, or the ER.
00:09:46.06 and the Golgi compartment.
00:09:49.01 And these membrane bound organelles are basically
00:09:53.24 an assembly line for the construction of complex glycans
00:09:58.25 from simple monosaccharide building blocks.
00:10:01.02 So, the glycans are built inside the ER and the Golgi,
00:10:04.29 attached to either proteins or lipids,
00:10:07.10 and then eventually those proteins and lipids,
00:10:10.08 or glycoproteins and glycolipids
00:10:12.04 are delivered to the plasma membrane where the cells
00:10:15.06 are now coated with these sugar molecules.
00:10:17.22 Let me tell you about the monosaccharide building blocks.
00:10:22.26 And first of all, I should say that there are many of these sugars in nature,
00:10:26.17 and different organisms have different collections.
00:10:30.08 So I am only showing you the monosaccharides
00:10:32.23 that you would find in vertebrate glycans,
00:10:35.22 which are distinct from the monosaccharides you would find
00:10:38.15 in bacteria, or even plants, but these are the ones that we have inside our bodies.
00:10:44.02 And there are nine of them.
00:10:45.29 So, this is a good number to know,
00:10:47.17 there are 9 monosaccharide building blocks.
00:10:50.05 Just like there are 4 nucleotides in your DNA,
00:10:53.00 or 20 amino acids in your proteins.
00:10:56.06 And each of these monosaccharide building blocks
00:10:59.20 goes by a different name, and we also have abbreviations
00:11:03.00 that we use to denote them very quickly.
00:11:05.06 So for example, many of you are familiar with this sugar, called glucose.
00:11:09.25 Glucose is really the parent of all of the other monosaccharide units.
00:11:15.18 In fact, your cells can build any of these other building blocks
00:11:19.02 starting from glucose, if it had to.
00:11:21.10 And glucose goes by the abbreviation, Glc,
00:11:25.04 and we often just say "Glick",
00:11:26.28 a simple little word to denote glucose.
00:11:29.11 And then some of these sugars are perhaps more exotic
00:11:32.17 in their structures, for example, this one.
00:11:34.05 This monosaccharide is called sialic acid.
00:11:37.16 It has more carbon atoms than the other sugars.
00:11:40.27 It also has this carboxylate. It carries a negative charge,
00:11:44.26 and I am going to come back to sialic acid later on
00:11:47.28 because it occurs in some interesting biological circumstances.
00:11:51.21 Now we have terminology that we use to
00:11:55.28 describe the structures of higher order glycans.
00:11:59.03 These are structures that are made up of multiple monosaccharide building blocks.
00:12:03.01 So for example, glucose is simply a monosaccharide,
00:12:07.23 and we often think of this as a metabolic sugar.
00:12:10.11 But if you take glucose and link another sugar to it,
00:12:14.00 and this one is galactose, these two together make a disaccharide
00:12:19.13 that's known as lactose, which you might have heard of also
00:12:22.13 because it's abundant in milk.
00:12:24.09 It's a milk disaccharide.
00:12:26.00 It's DI-saccharide because it has two monosaccharide units.
00:12:30.09 And here's a structure of what we would call either an oligosaccharide
00:12:36.00 or a polysaccharide, which are terms that we use interchangeably.
00:12:39.27 This is a much larger structure which has many copies
00:12:44.16 of glucose all linked together in a long polymer.
00:12:47.19 This structure is cellulose.
00:12:50.06 It's the major component of plant cell walls,
00:12:53.13 and in fact it's the most abundant organic material on earth.
00:12:57.25 It's very important to understand the structure of cellulose.
00:13:02.00 Now when we link monosaccharides
00:13:06.13 together to make these larger glycans,
00:13:08.27 we need terminology to describe the nature of those linkages.
00:13:13.20 In this way the glycans are more difficult
00:13:17.04 and more structurally complicated
00:13:18.28 than other biopolymers like DNA or RNA or proteins.
00:13:23.12 The difference is that those other biopolymers are linear,
00:13:28.00 and all of the linkages, whether they are amide bonds in the proteins,
00:13:31.15 phosphodiesters in the nucleotides,
00:13:34.13 all those linkages are the same.
00:13:36.03 But for glycans, each linkage can be different
00:13:39.18 and rather than simply being linear,
00:13:41.25 the glycans can also be branched.
00:13:43.28 And we also have issues of what we call stereochemistry
00:13:47.25 which has to do with the orientation of linkages.
00:13:51.05 So, it's more complicated. So just to give you
00:13:53.09 a sense of how complicated it can be,
00:13:55.17 what I am showing here is the structure of a trisaccharide.
00:13:59.25 So there are only three monosaccharide
00:14:02.20 building blocks linked together.
00:14:04.05 It's a fairly simple structure,
00:14:05.23 compared to some other glycans in nature.
00:14:07.28 But even with this trisaccharide,
00:14:11.01 we have to describe not only the orientation
00:14:14.05 of how each of these sugars is linked to the next one,
00:14:16.23 but also the position on each sugar to which a sugar is linked.
00:14:20.12 Because as you'll see, each of these sugars has multiple hydroxy groups.
00:14:25.09 And each hydroxy group could potentially be
00:14:28.00 a site of linkage to another sugar.
00:14:29.17 So, we need to understand for each sugar,
00:14:33.17 both the regiochemistry of its linkage,
00:14:36.11 which is the orientation, or I should say the position,
00:14:39.23 as well as the stereochemistry,
00:14:41.28 which is the orientation.
00:14:44.06 So for example, over at the end here, there is a galactose,
00:14:47.28 and this galactose is linked to the hydroxy group at the 4 position
00:14:53.01 of N-acetylglucosamine. That 1-4 linkage is the regiochemistry.
00:14:59.25 And this orientation of a bond
00:15:02.11 is what we call the stereochemistry,
00:15:04.00 and we define this as beta.
00:15:06.29 In contrast, this fucose residue
00:15:09.07 is linked to what we call an alpha linkage
00:15:12.23 to the 3-hydroxy group of N-acetylglucosamine
00:15:16.05 So if we take all of that information together,
00:15:18.27 we would then describe the trisaccharide
00:15:21.09 as galactose beta one four
00:15:24.25 to N-acetylglucosamine with at the same time in parentheses
00:15:29.08 a fucose linked alpha one three to the same N-acetylglucosamine.
00:15:34.08 So as you can see it gets pretty difficult.
00:15:37.03 but there are some simple elements of the structure
00:15:40.11 that are easy to remember.
00:15:41.23 So with DNA we think of a 5 prime end and a 3 prime end,
00:15:45.26 with proteins we think of an N terminus and a C terminus
00:15:50.19 Well, with glycans there are also two distinct ends.
00:15:54.27 We call them the non-reducing terminus and the reducing terminus.
00:16:01.19 So at the very least, you can think of a glycan as having two ends.
00:16:04.15 And then if you need more details about the structure
00:16:06.20 you have to understand what we call
00:16:08.15 regiochemistry and stereochemistry.
00:16:10.28 Okay. Now as I said, this is a simple structure.
00:16:15.11 In nature these structures can be far more complicated.
00:16:17.29 And I've taken it up a notch in this slide just to show you examples
00:16:21.27 of actual glycan structures that have been found on human glycoproteins.
00:16:28.14 And these are examples of two varieties,
00:16:31.19 one we call an N-glycan.
00:16:34.02 We call this an N-glycan because it is attached
00:16:36.29 to a nitrogen atom on the side chain
00:16:39.26 of an asparagine residue within the protein scaffold.
00:16:44.08 This variety is called an O-glycan.
00:16:47.03 It's an O-glycan because it is linked to the oxygen atom
00:16:50.13 on the side chain of either serine or threonine
00:16:54.02 that's within the protein that is the scaffold.
00:16:56.19 And as you can see this particular N-glycan is branched.
00:17:00.09 It has these two arms. We call these antenna.
00:17:02.24 It turns out these N-glycans
00:17:05.04 can have three antenna or four antenna.
00:17:07.10 They can be much more complicated than this.
00:17:09.17 And here, this is an O-glycan that also has a
00:17:12.01 branch point and then another branch point.
00:17:13.29 It's a pretty complicated structure, but one thing we've learned
00:17:17.25 by looking at all of the different glycans in the glycome
00:17:21.21 is that those structures are not random.
00:17:24.13 In fact, elements of those structures are highly conserved
00:17:27.28 in particular organisms. So invertebrates, for example,
00:17:31.14 the N-glycans can be very diverse in the parts
00:17:35.00 that are out here in the structures of the antenna.
00:17:37.02 However, this part here that is close to the protein scaffold
00:17:42.04 is generally highly conserved,
00:17:45.04 and similar from glycan to glycan to glycan.
00:17:47.15 Likewise, in the O-glycan family, there's a lot of diversity out here
00:17:51.20 but this sugar is always conserved.
00:17:54.07 It is always the same sugar that is linked
00:17:55.21 in the same way to the protein backbone.
00:17:58.00 So there are conserved and variable parts of these glycans.
00:18:01.04 Alright, now I mentioned that the glycans
00:18:05.10 are assembled inside the Golgi and the endoplasmic reticulum.
00:18:09.11 And there are enzymes that reside in those
00:18:12.03 compartments that do this enzymatic chemistry.
00:18:16.24 We call those enzymes glycosyltransferases
00:18:18.25 Now, I thought I would mention a point of historical interest,
00:18:23.01 which is that the discovery of this mechanism of biosynthesis
00:18:26.22 is largely attributed to Luis Leloir who back in the 1950's
00:18:31.09 discovered that glycogen, which is a storage form of glucose
00:18:36.13 in vertebrate systems, is built biosynthetically from a precursor
00:18:41.20 in which the glucose is linked to a nucleotide diphosphate
00:18:46.11 and we call this nucleotide sugar, UDP-glucose.
00:18:51.09 Here's the UDP part, uridine diphosphate,
00:18:54.02 and there's the glucose.
00:18:56.03 Now this was an important discovery because it suggested
00:18:59.17 a mechanism by which glycans in general
00:19:02.04 might be synthesized, and in fact
00:19:04.05 the importance of Leloir's discovery was recognized with a Nobel Prize.
00:19:08.09 In the forward sense, the way that glycogen
00:19:11.18 is assembled is through the action
00:19:14.05 of an enzyme that one would classify as a glucosyltransferase.
00:19:18.16 It transfers a glucose onto the growing polysaccharide.
00:19:23.06 And the substrate it uses is, again, the UDP-glucose.
00:19:27.12 Now, it turns out that all of the glycosyltransferases,
00:19:32.18 or I shouldn't say all, but most of the glycosyltransferases
00:19:35.26 use substrates that are similar to this nucleoside diphospho sugar.
00:19:41.11 And I'll just show you examples from, again, vertebrate biology.
00:19:45.19 So many of the sugars can be found in this UDP form, not just glucose,
00:19:50.22 but also galactose, N-acetylgalactosamine,
00:19:54.16 and N-acetylglucosamine.
00:19:56.05 Whereas some of the sugars are found linked
00:20:00.07 in the form of a GDP-nucleoside
00:20:04.21 for example GDP-mannose, and GDP-fucose.
00:20:08.09 And these are the substrates
00:20:09.20 for their respective glycosyltransferases.
00:20:12.21 And then sialic acid kind of stands alone in vertebrate biology
00:20:17.04 in that it's activated form is to a cytidine
00:20:21.29 monophosphate, or CMP-sialic acid.
00:20:25.29 And there are a family of sialyl transferases
00:20:28.19 that all use this as what we call a glycosyl donor.
00:20:32.27 So these are the substrates that are made
00:20:35.05 inside your cells and used by your enzymes.
00:20:37.17 Just to give you a sense of how enzymes
00:20:40.14 might assemble a tetrasaccharide,
00:20:43.00 this is a pathway that is found in vertebrate systems
00:20:46.15 so this disaccharide is synthesized,
00:20:49.10 and then a sialyl-transferase will take
00:20:52.20 the sialic acid from CMP-sialic acid
00:20:55.25 and transfer it onto this sugar,
00:20:58.00 converting the disaccharide to a trisaccharide.
00:21:02.08 Then, along comes a fucosyltransferase,
00:21:05.15 that will transfer fucose from GDP-fucose
00:21:08.23 and convert the trisaccharide to a tetrasaccharide.
00:21:11.24 This particular tetrasaccharide has some very
00:21:15.06 interesting biological properties that
00:21:17.22 I'll be coming back to later in this lecture.
00:21:21.09 In the history of glycobiology,
00:21:23.08 probably one of the most important discoveries that
00:21:26.14 really started to attract a lot of interest from outside the field
00:21:31.11 was the discovery in the middle of the last century
00:21:34.07 of the human blood groups.
00:21:36.14 Now, this is a discovery that has had huge implications
00:21:41.01 in respect to understanding immunology and the human immune system.
00:21:44.28 and also it's a discovery that was central
00:21:47.20 to the development of blood transfusions.
00:21:50.20 Of course the blood transfusion
00:21:52.07 is one of the most important clinical procedures.
00:21:55.07 It turns out that your blood type
00:21:58.16 is determined by sugars.
00:22:00.04 So hopefully all of you know your blood type,
00:22:03.23 I can tell you mine is O positive.
00:22:06.09 Some of you might be blood type A. Some of you will be blood type B,
00:22:10.20 and some of you might be blood type AB.
00:22:12.18 Well, what it means to be O, or A, or B,
00:22:16.11 or AB is simply what are the structures of the sugars on your blood cells.
00:22:21.12 So for example, as someone who is blood type O,
00:22:25.04 what that means is that my blood cells
00:22:27.17 have this trisaccharide structure
00:22:30.16 on the surface, on the glycoproteins and some of the glycolipids.
00:22:34.14 That defines me as blood type O.
00:22:38.22 Now some of you are blood type A.
00:22:41.01 What that means is that you
00:22:43.12 also have this sugar biosynthesized in your cells
00:22:47.11 but you have an enzyme that I don't have.
00:22:50.05 That enzyme transfers this new sugar
00:22:53.11 onto the trisaccharide to build a tetrasaccharide.
00:22:57.13 And if you have this particular tetrasaccharide on your blood cells,
00:23:01.02 you're blood type A, by definition.
00:23:04.10 Now those of you who are blood type B
00:23:06.15 have a slightly different enzyme.
00:23:08.05 Instead of transferring this red sugar,
00:23:11.09 which is N-acetylgalactosamine,
00:23:14.04 your enzyme transfers the green sugar,
00:23:17.12 which is galactose. So when galactose,
00:23:19.24 is added to this trisaccharide,
00:23:21.23 you get a tetrasaccharide, which is slightly different.
00:23:24.05 And this is the B tetrasaccharide.
00:23:25.28 So those people are blood type B.
00:23:28.15 For those of you who are really into chemical detail,
00:23:33.00 if you look closely at the structure of A and the structure of B,
00:23:36.08 you'll notice that there is a single chemical functional group
00:23:40.23 that is different between these two structures.
00:23:43.03 It's very subtle. So right here in blood type A
00:23:45.28 there's an N-acetylamido group,
00:23:48.13 an N-acetyl group. Over here in blood type B, it's a hydroxy group.
00:23:53.10 That's the only difference.
00:23:54.10 And yet, the human immune system
00:23:56.27 is so exquisitely sensitive to structural differences
00:24:00.25 that your immune system can detect
00:24:03.02 the difference between these two instantly.
00:24:05.04 And that's why if you have blood type A, and by accident,
00:24:09.22 you receive a blood donation from a blood type B donor,
00:24:13.18 your immune system will react against this and reject the blood.
00:24:17.07 And that's a disaster.
00:24:18.22 So understanding the structures of the human blood types
00:24:22.02 and what the means to the immune system,
00:24:23.13 was absolutely critical for blood transfusions to occur.
00:24:28.15 And by the way, those of you who are the AB blood type,
00:24:32.06 what you have is this enzyme and this enzyme.
00:24:35.21 You got one enzyme from your mother, and the other from your father
00:24:38.13 and you can make a 50/50 mixture of these two structures.
00:24:41.29 That's what you've got on your blood cells.
00:24:43.19 So that is considered a real classic discovery in the field of glycobiology.
00:24:49.09 Again, it dates back to the mid-late-1900s,
00:24:52.24 but nowadays there is a lot going on in the field.
00:24:57.00 And major discoveries have been made that
00:24:58.27 have now created opportunities
00:25:00.25 to treat very serious human diseases.
00:25:04.02 And I thought I would take a moment to give you
00:25:06.22 a little history with respect to two discoveries in the field
00:25:09.26 that are attracting a lot of attention in the clinical world today.
00:25:13.19 The first of those has to do with the mechanism
00:25:16.27 of influenza virus infection,
00:25:19.17 which is also what we call the flu.
00:25:21.19 And the second has to do with the way
00:25:24.16 that white blood cells, also called leukocytes,
00:25:27.14 attach to endothelial cells.
00:25:29.26 which are the cells that line your blood vessels.
00:25:32.09 It turns out that when your white blood cells
00:25:34.00 start to stick to the side of your blood vessels,
00:25:36.01 that can lead to inflammation,
00:25:37.27 which is involved in a variety of different diseases.
00:25:41.04 So let's start by talking a little bit about the flu.
00:25:44.04 Now influenza has been a major global health problem
00:25:50.27 dating back really, you know, hundreds and hundreds of years.
00:25:54.26 But one of the first documented pandemics of influenza
00:25:58.04 was the famous pandemic of 1918
00:26:01.10 which wiped out a huge portion of the population
00:26:05.05 over 70 million deaths have been attributed
00:26:07.24 to this particular flu pandemic
00:26:09.18 which, by the way, is more deaths than were associated with
00:26:13.11 World War One and World War Two combined.
00:26:16.13 So this was a major killer back in the early part of the 1900s,
00:26:20.20 and in fact, scenes like this one in which huge warehouses
00:26:24.14 or even airplane hangars were cleared out
00:26:26.23 and just lined wall to wall with beds with infirm patients
00:26:30.15 who were trying to survive their bout with the flu.
00:26:33.00 This was a common image during that period in history.
00:26:35.28 And movies have been made about this crisis,
00:26:38.06 and certainly many books have been written about this crisis.
00:26:41.03 And this was a lesson to humanity that the influenza virus,
00:26:45.16 although many of us get the flu and we recover just fine,
00:26:49.03 that this should not be taken lightly.
00:26:51.17 Influenza can be very deadly,
00:26:53.07 particularly for elderly people and for very small children.
00:26:56.27 So, for this reason over the last decade there has been a lot of work
00:27:01.13 in developing influenza vaccines.
00:27:04.16 And it has been a very difficult problem,
00:27:06.10 because, as many of you know,
00:27:08.03 the flu is a highly shifting and changing virus.
00:27:13.06 It can mutate very rapidly
00:27:15.08 so that the strain of flu that we get vaccinated against this year,
00:27:19.16 that strain morphs and changes
00:27:21.14 and next year it's different enough that the vaccine no longer works.
00:27:24.26 For that reason, scientists and physicians are always trying
00:27:28.22 to stay one step ahead of the flu.
00:27:31.09 And every year, you go in for another flu shot,
00:27:33.25 which hopefully will protect you from that year's influenza strain,
00:27:37.22 although it might not do much for the subsequent year and so on.
00:27:39.28 But nonetheless, with the advances of the last decade or two,
00:27:44.00 we now have fairly reliable flu shots
00:27:46.29 that we can get every year, and hopefully you go in
00:27:49.15 and get your flu shot, and I pulled this image off of the web
00:27:52.29 because I thought you might be interested to hear
00:27:54.28 that the flu vaccine is actually generated in chicken eggs.
00:27:59.07 We use those eggs as little factories to make these vaccines
00:28:02.24 and what you are looking at here is a scientist
00:28:04.29 who is basically injecting eggs with a flu strain
00:28:08.01 that will then propagate within those eggs.
00:28:09.11 So try to get your flu shot if you can.
00:28:12.15 However, as many of you know who've been paying attention to the news
00:28:16.06 in recent months, just because you get protected against
00:28:20.16 what we think will be next year's flu strain
00:28:22.19 doesn't mean that you are protected against all forms
00:28:25.05 of influenza. And one of the most scary features of
00:28:28.09 influenza is that sometimes it has the ability
00:28:31.04 to move from one organism to another.
00:28:34.16 So, there are strains of influenza that normally make birds sick,
00:28:40.22 and some of those have been so catastrophic to poultry industries
00:28:45.18 that there's interest in vaccinating chickens against the flu
00:28:48.27 the same way that we vaccinate ourselves against the flu.
00:28:51.15 But, if bird flu gets into a human, it can make that human very sick.
00:28:56.12 And so, many of you have probably heard
00:28:58.25 about these local incidents of bird flu
00:29:03.03 in humans, and this is a map showing you where
00:29:05.07 some of those bird flu cases have been identified.
00:29:09.03 So far, the good news about bird flu is
00:29:10.11 that while we might catch it from a bird
00:29:12.28 and get very sick, it doesn't look like
00:29:15.00 we then can transmit it to another human.
00:29:17.02 Now, this is not the case with the most recent scary outbreak
00:29:23.01 of influenza, which has been called the swine flu.
00:29:26.03 This is a flu that's thought to have come from pigs
00:29:28.28 and then moved into humans.
00:29:30.11 It's also called H1N1 influenza,
00:29:34.24 and I'll show you in a minute where those terms come from.
00:29:37.24 But basically, the swine flu can go from pigs to humans, but
00:29:42.21 now it can also go from humans to humans,
00:29:45.05 which means that the swine flu is a much bigger risk for a pandemic
00:29:49.09 because of human-to-human transmission.
00:29:51.16 Fortunately, so far, it looks like it is a fairly mild form of the flu,
00:29:55.29 but there have been many cases reported,
00:29:57.11 most of them in North America
00:29:59.25 and many of them here in the United States,
00:30:01.16 as you can see from this map.
00:30:03.24 So there are many reasons why we want to understand
00:30:07.14 at the molecular level how influenza works.
00:30:11.21 So we can develop better vaccines and also generate drugs
00:30:15.15 to help treat people who have contracted the flu.
00:30:19.02 where a vaccine is really, you know, not relevant anymore.
00:30:21.29 So there's been a lot of research on the influenza virus,
00:30:25.02 right down to the individual molecules
00:30:26.24 that are involved in the infection cycle.
00:30:29.18 And what was discovered, starting back in the 1970s and 1980s
00:30:34.00 is that the very early stages of influenza virus infection
00:30:38.16 involves sugars. And in fact, that stage is the stage
00:30:42.21 at which the influenza virus particle,
00:30:44.22 which is shown here in this electron micrograph,
00:30:47.10 attaches to a human host cell that it is destined to infect.
00:30:51.07 There are sugars involved in that very first interaction
00:30:53.28 between the particle and the host.
00:30:55.19 Now, also the host generates new viral particles,
00:31:00.00 which then bud and leave the host
00:31:02.14 and it turns out that there are sugars involved in that step as well.
00:31:05.07 And let me show you how. Okay.
00:31:08.18 So, here's a cartoon that illustrates the anatomy
00:31:11.07 of the influenza virus. It's a membrane-enclosed virus
00:31:15.05 that has a core that has both RNA and proteins.
00:31:19.02 But there are two proteins that sit on the membrane envelope
00:31:23.02 of the virus, and those proteins go by the name
00:31:26.03 hemagglutinin, which I abbreviate "H",
00:31:28.25 and neuraminidase, which I abbreviate "N".
00:31:32.02 And remember the swine flu is more scientifically termed H1N1.
00:31:38.26 Well the H1 is a certain form of hemagglutinin,
00:31:42.18 and the N1 is a certain form of neuraminidase.
00:31:46.00 Now we know quite a bit about what these two proteins do.
00:31:50.20 In fact, we even know their molecular structures in very great detail.
00:31:54.16 Hemagglutinin is a receptor.
00:31:57.07 It's a protein that binds to a sugar,
00:32:00.12 and that sugar happens to be sialic acid,
00:32:03.03 which I mentioned before.
00:32:05.01 Neuraminidase is an enzyme.
00:32:07.21 and what neuraminidase does is it catalyzes
00:32:11.02 the cleavage of sialic acid off of the host cell.
00:32:15.18 So, this protein attaches to sialic acid,
00:32:18.16 and this protein cuts the sialic acid off and throws it away.
00:32:22.09 Now when that discovery was made,
00:32:24.25 it struck many scientists as a paradox.
00:32:27.15 Why would the virus have a protein that attaches to sialic acid,
00:32:31.25 and yet another protein that just
00:32:32.29 cuts off that sialic acid and tosses it away?
00:32:35.07 Well, I'll show you what those two proteins do.
00:32:38.07 It turns out that hemagglutinin
00:32:40.28 is important in the very first stage of infection
00:32:43.04 where the virus lands on a cell, a human host cell.
00:32:47.17 The hemagglutinin attaches to the sialic acid.
00:32:50.14 and basically allows the protein, or I should say the virus particle,
00:32:54.15 to dock on the cell surface. Once that occurs,
00:32:57.27 it triggers an endocytosis event,
00:33:01.04 where the host cell inadvertently engulfs the viral particle into a vesicle.
00:33:06.26 The membrane of the virus fuses with the membrane of the vesicle,
00:33:10.16 and releases the nucleic acid into the cell.
00:33:14.20 And now, that viral nucleic acid takes over the machinery
00:33:18.07 of the cell and forces the cell to generate more viral particles.
00:33:22.27 Those viral particles assemble around the membrane
00:33:26.13 of the host cell, and eventually a new viral particle
00:33:29.13 buds off of the cell surface,
00:33:31.01 as I showed you in that previous electron micrograph.
00:33:33.06 But remember that with all that hemagglutinin around
00:33:36.15 that viral particle might get stuck
00:33:38.22 on the cell surface where the sialic acids are.
00:33:41.25 And so the job of neuraminidase is to cut
00:33:44.10 those sialic acids off at that point
00:33:46.13 so that the virus can release itself
00:33:48.18 from the cell and go find another host cell to infect
00:33:52.28 and complete the cycle.
00:33:54.21 So that's why we need these two proteins that act on sialic acid.
00:33:58.14 Now knowing the importance of neuraminidase in the viral lifecycle
00:34:04.13 many scientist thought that if one inhibits that enzyme,
00:34:08.14 and prevents this very last step in the cycle, one might
00:34:12.02 be able to shut down the propagation of the influenza virus.
00:34:16.10 And so a large drug discovery effort was underway back in the 1990s,
00:34:23.07 even the late 1980s, to develop inhibitors of the neuraminidase enzyme.
00:34:26.25 And that was done by understanding
00:34:29.21 the mechanism of that enzymatic reaction.
00:34:33.21 So, the mechanism is shown here.
00:34:36.20 Here is a sialic acid, and picture it bound to the surface of a cell
00:34:40.26 through a glycan on a glycoprotein or a glycolipid.
00:34:44.18 So the R group is the rest of the glycan, or the rest of the glycoprotein.
00:34:48.05 What happens during the neuraminidase catalyzed reaction,
00:34:53.10 is that there is a cleavage of the bond
00:34:55.29 right here between the sugar ring carbon
00:34:59.14 and this oxygen that's called the glycosidic bond.
00:35:02.06 And what the enzyme does is that it finds a way to make this bond
00:35:06.12 reactive, so this bond is cleaved.
00:35:09.00 And there is a transition state for this reaction
00:35:11.28 in which there's basically a change in
00:35:15.02 the hybridization of the carbon atom
00:35:17.08 at this position, so that it goes from being what we call
00:35:20.03 sp3 hybridized to sp2 hybridized.
00:35:23.08 It becomes planar, and also a positive charge develops on the ring.
00:35:27.15 And then that leads to the formation of this intermediate,
00:35:30.07 and then water from the environment reacts with the intermediate
00:35:33.12 to form a free sialic acid molecule,
00:35:36.19 which then floats away.
00:35:37.26 Well, what several pharmaceutical companies did
00:35:42.06 is to look at the structure of this presumed transition state
00:35:45.17 and try and mimic that structure with these synthetic molecules
00:35:50.03 that are somewhat reminiscent of sialic acid.
00:35:52.27 For example, this compound has the sp2 hybridization at this carbon
00:35:58.07 similar to the transition state
00:35:59.29 and so does this compound. This compound
00:36:02.12 has a positive charge in the form of this guanidino group
00:36:05.22 and this compound has a positive charge in the form of this amino group.
00:36:09.04 These two molecules are actually now on the market
00:36:12.28 as flu drugs. This compound goes by the trade name Relenza,
00:36:17.29 and this compound goes by the name Tamiflu.
00:36:20.24 So if you feel the very, very early symptoms of the flu coming on,
00:36:25.08 you can go to the doctor get a prescription for one or the other
00:36:28.00 of these and try and prevent a full-blown onset of the flu.
00:36:32.25 Or if someone in your family has been diagnosed with the flu,
00:36:36.17 and you are worried that you might catch it,
00:36:38.16 once again, you might take one of these two drugs
00:36:41.25 as a preventative measure,
00:36:43.11 as a prophylactic against the flu.
00:36:46.06 So this is a nice example where understanding the glycobiology
00:36:50.21 of influenza led ultimately to the development of drugs to treat the flu.
00:36:56.11 It's very nice story.
00:36:57.07 Okay. The other story I thought I would tell you has to do
00:37:01.03 with inflammation. So, as I mentioned before,
00:37:03.27 sometimes it happens that the white blood cells, which normally
00:37:08.09 flow freely through your bloodstream, find themselves
00:37:11.29 sticking to the endothelial cells that line the blood vessel wall.
00:37:15.08 When that occurs, its usually bad news
00:37:18.21 because it means that you might be
00:37:20.02 in the throes of an inflammatory disease.
00:37:21.28 So, during inflammation this endothelium gets activated,
00:37:26.17 and molecules appear on the endothelium
00:37:29.01 that normally wouldn't be there.
00:37:30.08 And as a consequence, those molecules can bind
00:37:33.14 to other molecules on leukocytes
00:37:35.10 and now the cells attach to each other.
00:37:37.09 Because the blood is flowing, the initial attachment
00:37:40.16 is what we consider a weak attachment
00:37:42.00 where the cells are kind of rolling along the blood vessel wall.
00:37:45.10 Because the blood is pushing them along,
00:37:47.09 they are only loosely attached.
00:37:48.02 But eventually, they will become firmly attached,
00:37:51.08 and in fact, they can even become migratory,
00:37:53.15 burrow their way through the endothelial cells
00:37:56.12 and enter the surrounding tissue.
00:37:57.18 And if your leukocytes leave the bloodstream,
00:38:00.13 and enter the tissue, which is a process called extravasation,
00:38:04.19 those leukocytes can damage the tissue,
00:38:07.21 and basically cause the pain
00:38:08.29 and the swelling associated with inflammation.
00:38:11.27 This is a picture, not of an inflamed tissue,
00:38:16.13 but a picture of a blood vessel in the lymph node,
00:38:18.19 where it turns out that white blood cells
00:38:21.14 are normally found attached to the blood vessel walls.
00:38:24.14 This is because your lymph node is constantly collecting
00:38:27.24 leukocytes out of the bloodstream,
00:38:29.23 and collecting them in the lymph nodes
00:38:31.03 is part of the lymph node's job.
00:38:33.01 But it's a nice picture because it illustrates
00:38:35.04 that what is normal in the lymph node,
00:38:36.21 would be very abnormal outside of the lymph node.
00:38:39.12 And if you saw this situation outside of the lymph node,
00:38:42.00 chances are you are having an inflammatory reaction,
00:38:45.05 and maybe an inflammatory disease.
00:38:46.28 And it's a pretty striking process.
00:38:49.04 So what do we know about how the leukocytes interact
00:38:53.01 with the endothelial cells?
00:38:54.10 It turns out that many proteins are involved in this cell-cell adhesion event,
00:38:58.23 but sugars are involved as well,
00:39:01.18 particularly in that very early stage of rolling.
00:39:04.27 So back in the late 1980s and early 1990s,
00:39:08.06 a family of glycan binding proteins was discovered to be involved
00:39:12.15 in leukocyte rolling, and we call that family the "selectin" family
00:39:17.09 of adhesion molecules. There are three members of that family:
00:39:20.20 two of the members reside on activated endothelial cells.
00:39:24.14 They come up when the endothelial cells are stimulated
00:39:28.26 with an inflammatory signal,
00:39:30.04 and those two are called P-selectin and E-selectin.
00:39:33.26 There's a third selectin, which is found on leukocytes.
00:39:36.27 And it goes by the name L-selectin.
00:39:40.07 L-selectin is hanging around on leukocytes most of the time,
00:39:43.11 but it needs to bind to a sugar which appears on
00:39:47.11 the endothelial cells, and that sugar is usually not present,
00:39:50.05 unless there's inflammation.
00:39:51.17 Likewise, P-selectin and E-selectin,
00:39:54.18 they bind sugars that are found on the leukocytes,
00:39:56.27 and sometimes two selectins with their two sugars can interact
00:40:01.05 at the same time to help the leukocyte roll
00:40:03.27 on the endothelium.
00:40:05.17 Now scientists became very interested
00:40:07.22 in this system because they realized
00:40:08.29 that if you could prevent the binding of L or E or P-selectin
00:40:13.13 to these various sugar molecules,
00:40:16.00 you might be able to block leukocyte recruitment into the tissue
00:40:21.03 during an inflammatory disease.
00:40:23.01 and basically make an anti-inflammatory drug.
00:40:25.18 And if you could do that, maybe you could treat
00:40:28.03 a lot of different diseases that were known
00:40:29.27 to involve the extravasation
00:40:33.03 of leukocytes into the tissue.
00:40:34.23 And these include rheumatoid arthritis,
00:40:36.27 which is inflammation of the joints,
00:40:38.13 chronic asthma, inflammation of the bronchial passages in the lungs.
00:40:43.12 One might be able to prevent
00:40:45.05 the rejection of transplanted organs,
00:40:47.19 that are recognized as foreign by the immune system.
00:40:50.08 Psoriasis, which is an inflammation of the skin.
00:40:54.24 Inflammatory bowel disease,
00:40:55.22 which is inflammation of the colon,
00:40:57.08 and many, many other indications
00:41:00.05 that many people suffer from.
00:41:01.26 So the bottom line is that inhibitors of selectin mediated
00:41:05.23 cell adhesion could potentially be used to treat all of
00:41:09.09 these illnesses. A very broad spectrum anti-inflammatory drugs.
00:41:13.17 Now it's turned out to be a difficult challenge.
00:41:16.18 In part because the way that the selectins
00:41:19.14 bind to sugars doesn't really lend itself to making inhibitors,
00:41:24.26 the way we were able to make
00:41:26.07 inhibitors for neuraminidase of influenza.
00:41:29.28 What we do know is that all
00:41:31.13 three selectins bind this tetrasaccharide,
00:41:34.11 which goes by the common name, sialyl Lewis x.
00:41:38.24 And for those of you who are focusing on chemical detail
00:41:42.14 you might recognize this is the same structure
00:41:45.00 that I showed in a previous slide
00:41:46.28 when I was illustrating how glycosyltransferases
00:41:50.02 build complex structures from simple building blocks.
00:41:53.10 Sialyl Lewis x has sialic acid at its non-reducing end,
00:41:59.13 linked to galactose, linked to N-acetylglucosamine,
00:42:03.18 and then branched from that same sugar is fucose.
00:42:06.27 These are the four sugars.
00:42:08.23 So the three selectins will all bind to this structure,
00:42:12.09 but it turns out they bind this structure rather weakly.
00:42:15.23 So, the dissociation constant, which is a measure of binding affinity
00:42:20.09 is only around 1 millimolar,
00:42:23.15 so that's considered a very weak interaction.
00:42:26.11 Now you might wonder if the interaction is that weak,
00:42:30.02 how is it that the selectins
00:42:31.28 can allow two cells to bind to one another at all?
00:42:36.00 Well it turns out that in nature,
00:42:38.01 that sugar does not just stand alone.
00:42:40.04 It's displayed in a multivalent manner on glycoprotein scaffolds.
00:42:47.07 And so, the selectins have professional ligands
00:42:49.23 in the body that are glycoproteins with many, many copies
00:42:54.18 of that sugar, sialyl Lewis x.
00:42:56.13 For example, there are three of these
00:42:58.10 glycoproteins that are known to bind L-selectin.
00:43:00.26 They go by these three scientific names,
00:43:04.04 but basically what they all share
00:43:05.16 is a long protein stalk with many copies
00:43:09.00 of the sugar which is illustrated by this hairbrush like structure
00:43:12.29 where you can picture each bristle is a different sugar molecule
00:43:16.10 and they're all displayed on this one long stalk.
00:43:19.17 Incidentally, it turns out that the sugar is not alone in
00:43:23.24 this structure. There are sulfate groups on the sugar molecule
00:43:27.07 and the sulfate groups also contribute to the binding affinity.
00:43:30.16 P-selectin also has a professional ligand
00:43:35.03 that is known as PSGL-1
00:43:37.12 that just stands for P-selectin ligand, or glycoprotein ligand one.
00:43:42.09 And again, there are many, many sugars
00:43:44.22 that are like bristles on a long hairbrush
00:43:46.04 and also some sulfate groups that are involved
00:43:48.25 in binding. So in vivo, the situation is very complicated.
00:43:52.10 The sugars are involved,
00:43:53.19 but they are involved in a multivalent manner.
00:43:57.07 Well, scientists over the years
00:43:59.18 have realized that if you want to inhibit
00:44:03.01 multivalent binding between two objects
00:44:06.28 whether they are two cells, or a virus and a cell,
00:44:08.28 or a bacterium and a cell, the best way to do that is
00:44:11.23 not with a monomeric inhibitor,
00:44:14.15 but rather with a multivalent inhibitor
00:44:16.28 So in other words, if two cells interact
00:44:20.21 through multiple weak receptor-ligand interactions,
00:44:23.28 and each of these interactions could be a selectin and a sugar,
00:44:28.06 then you are much better off competing with this situation
00:44:31.10 using an inhibitor that also has multiple copies
00:44:34.21 of the ligand. So, the inhibitor, in other words,
00:44:38.21 should mimic the cell, it shouldn't just be a simple monomer.
00:44:41.27 And this kind of inhibition can be much more effective.
00:44:44.05 As an example of what is going on in the field,
00:44:47.09 it turns out that you can achieve
00:44:49.10 that kind of multivalent ligand display
00:44:52.01 using a variety of different architectures.
00:44:54.07 One of those is to use a liposome.
00:44:57.01 A liposome is just a small mimic of a membrane enclosed cells.
00:45:03.17 It's basically a lipid bilayer in a little circle with nothing inside necessarily.
00:45:09.13 Okay, and we can make these by synthesis.
00:45:12.16 And the way these are made is by taking lipids and mixing them together
00:45:16.09 in such a way that they form this bilayer like structure
00:45:19.10 usually these liposomes have nanometer dimensions,
00:45:22.22 ten to one hundred nanometers,
00:45:23.21 much smaller than cells.
00:45:25.03 And if one of the lipids has a sugar on the end
00:45:28.24 that is able to bind to the selectins,
00:45:31.01 then basically you end up with a liposome
00:45:34.02 that's got sugars on it and basically serves
00:45:37.02 to display those sugars in a multivalent manner.
00:45:39.03 And these kinds of sugar coated liposomes
00:45:41.23 this is just one example of a multivalent architecture
00:45:45.22 that's been used to inhibit selectin mediated cell adhesion
00:45:50.00 with very high affinity. Very high potency.
00:45:52.13 Much more potent than individual sugars.
00:45:55.07 I should also point out that the liposome is just
00:45:57.16 one example. Many groups have made polymers with sugars on them
00:46:02.14 so they have multivalent sugar displayed on a polymer.
00:46:04.29 Groups have made dendrimers, which are kind of star-like structures
00:46:09.18 and there are all kinds of structures you can envision
00:46:12.15 in which there are many, many sugars displayed
00:46:14.22 on a scaffold, rather than just one.
00:46:16.26 So, that's an area of interest in the field,
00:46:19.22 but I think we still have a long way to go
00:46:21.24 before these multivalent selectin inhibitors
00:46:26.00 make it into clinical practice.
00:46:27.24 But there are exciting roads ahead.
00:46:29.29 Okay. So, let me just wrap up this lecture
00:46:33.05 with three take-home messages
00:46:35.18 that you should try to remember.
00:46:37.06 First, remember that glycans have complex structures.
00:46:41.03 And those structures change as a cell undergoes physiological
00:46:45.12 changes. The glycome of a healthy cell is different
00:46:49.09 from the glycome of a cancer cell.
00:46:51.04 And this is going to be important in the next lecture
00:46:56.03 as I'll mention shortly.
00:46:57.28 Also, glycans can contribute
00:47:00.08 directly to important physiological processes
00:47:03.16 that are associated with human disease.
00:47:06.06 Sugars can be ligands for viruses as well as bacteria.
00:47:09.23 And sometimes when sugars
00:47:12.03 on one cell interact with receptors on another cell
00:47:14.12 that cell-cell interaction can be detrimental,
00:47:18.12 as in the case of chronic inflammation.
00:47:20.01 And then finally, if we can understand at the molecular level
00:47:24.10 how the sugars contribute to the disease
00:47:26.07 then we might be able to develop new therapeutic agents
00:47:29.13 to help treat these diseases.
00:47:31.04 And I hope that you have found this as interesting
00:47:33.04 as I have and also the students
00:47:35.18 and postdocs that work in my laboratory. Thank you.

Part 2: Imaging the Glycome

00:00:06.26 So welcome to the second lecture in the series
00:00:09.00 called Chemical Glycobiology.
00:00:11.07 My name is Professor Carolyn Bertozzi.
00:00:13.26 I am in the department of Chemistry at UC Berkeley,
00:00:17.21 with also an appointment in Molecular and Cell Biology Department
00:00:21.11 and I am also an investigator of the Howard Hughes Medical Institute.
00:00:25.06 I.. This is the second part of a two part series.
00:00:28.23 The first part focused on the background and history
00:00:32.09 about the field of glycobiology and the structures of sugars.
00:00:36.18 And this part, which is part two, is going to focus more
00:00:40.26 on some recent research from my own laboratory at Berkeley
00:00:43.25 with the goal towards imaging the glycome.
00:00:46.18 And in fact, what you are looking at here is
00:00:49.07 a fluorescence image of a zebrafish
00:00:53.06 that's five days old in which some of the sugars have been tagged
00:00:57.24 with fluorescent molecules and this allows us to literally visualize those sugars
00:01:02.18 in a live organism. And so let me tell you about
00:01:06.15 why we are interested in that.
00:01:08.03 Okay, now the field of molecular imaging is certainly large
00:01:13.15 and expanding rapidly, and it's very important
00:01:16.28 both from the perspective of clinical medicine
00:01:19.15 as well as in the research laboratory.
00:01:22.15 So many of you might have found yourself lying on a table like this
00:01:26.14 at one point in your lives, as have I.
00:01:29.05 This is a magnetic resonance imaging scanner, or a MRI scanner
00:01:34.24 and what this scanner can do
00:01:36.26 is literally take a picture of molecules
00:01:39.29 inside our bodies without having to cut us open.
00:01:42.24 Which is very convenient for many of us.
00:01:44.26 And there are devices very similar to this one which allow
00:01:49.13 clinicians to look at other molecules in our bodies,
00:01:52.00 molecules that are radiolabeled for example.
00:01:54.15 And that is what this image is here. This is called a PET scan.
00:01:59.00 of a human body that has been injected with a radiolabeled version of glucose.
00:02:04.02 Now these are important tools for looking at molecules in clinical settings.
00:02:09.28 but we also like to look at molecules in research settings
00:02:12.16 particularly to understand what molecules are doing inside
00:02:16.02 living cells and nowadays we have some wonderfully advanced
00:02:20.06 fluorescence microscopy tools
00:02:22.10 that allow us not just to see the collections of molecules in cells,
00:02:26.12 but individual fluorescent molecules as well.
00:02:29.12 So it's a very exciting time to be doing molecular imaging in living systems.
00:02:34.24 Now, I mentioned in my first lecture that one of the most
00:02:38.29 intriguing aspects of glycobiology is the fact that
00:02:43.15 the glycome or the complete collection of glycans that a cell makes
00:02:48.09 is dynamic. It's not fixed.
00:02:50.19 So, the glycome, or the collection of sugars on the surface of a cell
00:02:55.21 when it's in one particular state
00:02:57.12 is different from the collection of sugars on that same cell
00:03:00.27 when the cell has transformed its state.
00:03:04.02 This is certainly the case when embryonic stem cells
00:03:07.11 choose a differentiation fate
00:03:10.00 and then become a differentiated cell,
00:03:12.09 like a muscle cell, or a neuron. If you look at the collection of
00:03:16.02 the glycans on the cell surface glycolipids and glycoproteins,
00:03:20.03 you'll find that the nature of that collection has changed
00:03:23.01 as the cell has undergone a differentiation process.
00:03:26.12 From the perspective of clinical medicine,
00:03:30.07 it's interesting to note that the glycome changes when healthy cells
00:03:34.18 transform and become cancer cells.
00:03:37.28 So, you can understand if the sugars on the surface of the cell
00:03:42.02 are changing in a manner that is characteristic of a cancerous state,
00:03:46.06 it would be very convenient if we could actually see
00:03:49.19 those sugars change in vivo, in living human beings.
00:03:54.10 Because that would offer the possibility of imaging those changes
00:03:58.15 and detecting cancers when they are at hopefully very early stages.
00:04:02.17 So there are many reasons why one would want to image the glycome,
00:04:07.11 to take a look at how those sugars are changing inside living subjects.
00:04:12.00 So we can understand how do the glycans relate to differentiation of cells
00:04:17.06 or to organogenesis. And can we detect changes in the glycome
00:04:22.20 that correlate with disease states.
00:04:24.23 So that we can detect those diseases at very early stages.
00:04:29.11 The challenge, of course, is to figure out how to tag these sugar molecules with
00:04:35.14 probes that we can actually visualize using imaging technologies.
00:04:40.16 So even in the research laboratory, this is very challenging.
00:04:43.22 The most commonly used imaging technique in research laboratories
00:04:48.05 is fluorescence imaging, or optical imaging.
00:04:51.02 And we don't even have a way to put fluorescent tags
00:04:55.02 on these sugars. So that is a challenge
00:04:57.24 that students and postdoctoral fellows
00:05:00.02 in my laboratory have taken on.
00:05:03.09 And it's a project we started working on back in the late 1990's.
00:05:07.08 So it's been progressing now for over ten years.
00:05:09.21 Just to give you a little bit of counterpoint,
00:05:12.17 people who study proteins using molecular imaging techniques,
00:05:16.22 have a wonderful arsenal of tools available to them,
00:05:20.18 which is why images like the one I showed on the previous slide
00:05:24.06 are so common and so familiar.
00:05:26.23 to most cell biologists. In fact,
00:05:28.15 most cell biologists probably spend a good portion of their day
00:05:31.22 looking at images similar to this one.
00:05:34.06 In this cell two different proteins have been labeled
00:05:37.14 with two different fluorescent dyes,
00:05:39.28 one green and one red.
00:05:40.24 And the protein that appears green has been labeled using a genetically encoded reporter.
00:05:47.03 So let me just show you what that reporter is.
00:05:49.15 It is called the green fluorescent protein,
00:05:52.19 abbreviated GFP, and it's one of the most important tools for fluorescence imaging
00:05:59.21 of proteins. The green fluorescent protein, because it's a protein that is encoded by a gene,
00:06:05.01 and it turns out that one can fuse the gene that encodes GFP
00:06:08.20 to the gene that encodes the protein that you are interested in imaging.
00:06:12.19 And now, when the GFP protein fusion is expressed,
00:06:17.14 basically your protein of interest is labeled with a fluorescent molecule.
00:06:21.08 The GFP has a chromophore inside that is fluorescent.
00:06:25.09 And this is such an important tool. It's revolutionized our abilities
00:06:31.08 to study proteins, not just in live cells, but even in transgenic organisms.
00:06:35.19 which we express these GFP-fusions by manipulating the genome
00:06:40.20 in the embryonic stem cell stage.
00:06:42.22 This is so important that it was recognized with the Nobel Prize in Chemistry
00:06:47.03 in the year 2008, and that prize was granted to these three scientists.
00:06:52.00 Very well deserved. Very exciting development
00:06:54.23 in the field of molecular imaging.
00:06:56.22 Now those of us who study glycobiology, of course,
00:07:01.09 would love it if we had a tool that was comparable
00:07:04.12 to the green fluorescent protein.
00:07:05.28 So that we could label our sugars with a fluorescent tag,
00:07:09.05 the way that people label proteins with this fluorescent tag.
00:07:12.16 But unfortunately the biosynthetic machinery,
00:07:17.17 by which sugars are built, doesn't really lend itself
00:07:21.02 to these genetically encoded reporters.
00:07:23.09 That's because sugars, unlike proteins, are not primary gene products.
00:07:28.21 Each of these complex glycans is not encoded by a single gene.
00:07:33.17 Rather, they are the products of a complex series of
00:07:38.26 metabolic steps inside the cell.
00:07:41.02 They are products of metabolism.
00:07:43.21 And so for that reason, we cannot use genetically encoded reporters like GFP
00:07:48.02 to introduce labels onto sugars.
00:07:50.17 So I drew this cartoon as kind of some comedy, because we would love
00:07:54.24 to be able to put a tag like GFP on the sugars,
00:07:57.18 but there is really no mechanism to do that using genetics.
00:08:01.11 So, in my laboratory, we have been focusing
00:08:04.21 on using the metabolic pathways of the cell.
00:08:08.22 to introduce fluorescent probes into the sugars
00:08:12.12 that are not the GFP but rather small molecule fluorescent reporters.
00:08:17.18 And let me show you schematically how we go about doing this.
00:08:21.22 It's a two-step process, and we call this metabolic labeling of glycans
00:08:27.14 with chemical reporters.
00:08:29.15 The first step is to feed cells a synthetic
00:08:33.14 monosaccharide building block
00:08:35.17 Now this synthetic sugar looks very much like one of the natural
00:08:40.08 monosaccharide building blocks that you might find
00:08:43.06 in foods that you eat, and I introduced those structures in my first lecture.
00:08:47.26 But the difference is that we have altered the structure
00:08:51.07 of this monosaccharide building block to introduce
00:08:54.22 a chemical functional group "X"
00:08:57.14 and "X" is just the generic for now,
00:08:59.08 and in a minute we will get to what that is.
00:09:00.23 But "X" is what we call the chemical reporter.
00:09:04.17 It's a reactive functional group
00:09:07.05 so that when the cell takes this monosaccharide
00:09:10.18 building block, and when the enzymes process it and integrate it into the glycans,
00:09:15.29 when the glycan appears on the cell surface, that reactive functional group
00:09:20.07 is now on display and can be used in a second step
00:09:24.09 which is a chemical reaction with a probe molecule.
00:09:27.25 Now the probe is a fluorescent dye, for example.
00:09:31.18 And that probe has its own chemical functional group, let's call it "Y".
00:09:35.10 just as a generic label. But "X" and "Y" have to be designed
00:09:40.15 so they react with each other to form a bond.
00:09:43.29 And once that bond is formed
00:09:45.17 now there is a fluorescent probe molecule
00:09:49.20 bound to a cell surface glycan.
00:09:53.07 And that probe now allows us to visualize all of the glycans
00:09:57.01 that possess this red monosaccharide
00:09:59.17 building block as one of their constituents.
00:10:01.29 So, this is a very straightforward schematic
00:10:05.19 in that what we need to do is feed the cell a modified sugar
00:10:09.15 and then do a chemical reaction on that sugar
00:10:12.27 once it's been displayed on the cell surface with a probe.
00:10:15.21 But it turned out that there was a very significant chemical challenge.
00:10:20.08 embedded in this problem because we have to select
00:10:24.03 the chemical reporter "X" and a complementary functional group "Y",
00:10:29.09 so that these two functional groups react
00:10:33.00 only with each other and not with anything else inside this cell.
00:10:38.25 And let me tell you, there is a lot of functionality inside this cell.
00:10:43.21 All of your proteins and your lipids, your nucleic acids,
00:10:47.15 metabolites, there's water, you know, there are a lot of chemical entities inside this cell
00:10:53.27 and "X" and "Y" have to be designed to avoid any side reaction
00:10:58.26 with those biological groups and still react only with each other.
00:11:03.14 So that was a pretty significant challenge
00:11:06.11 and we embodied that concept of that challenge
00:11:09.12 with this term "bioorthogonal".
00:11:12.29 So, what this means literally is not interacting with biology.
00:11:18.07 "X" and "Y" have to react mutually and selectively with each other.
00:11:22.10 They cannot react with anything
00:11:24.07 that's found in nature. And of course, they certainly cannot be harmful
00:11:28.09 to the biological system under study
00:11:30.25 otherwise that would be a disaster from the perspective of
00:11:34.10 imaging sugars in live animals.
00:11:36.15 So we spent many years trying to craft reactions that
00:11:42.24 basically fit this description where the two
00:11:45.27 reactive counterparts, X and Y, are bioorthogonal.
00:11:50.03 And to make a long story short, what we discovered is that
00:11:55.14 an ideal chemical reporter, the group X that we put inside the sugar
00:12:00.16 is the azide, which is defined as having
00:12:04.12 three nitrogen atoms linked to each other
00:12:06.10 in a manner that is shown here.
00:12:08.26 Now the azide has a number of properties that make it very well suited
00:12:14.04 for this particular application as a chemical reporter.
00:12:17.04 First and foremost, azides are not found in biological systems.
00:12:22.03 As far as we know chemists invented the azide.
00:12:26.06 We have not found it in nature.
00:12:27.25 Maybe nature overlooked this functional group
00:12:29.27 when she was creating all of the chemical
00:12:32.21 diversity that's found on our planet.
00:12:34.22 Furthermore, azides are essentially inert in biological systems.
00:12:39.29 There's really not much that they can react with
00:12:41.22 that is normally found in a biological environment.
00:12:45.22 Azides are very easy to attach to sugar molecules.
00:12:50.14 The chemistry is very simple.
00:12:52.05 And importantly, azides are small.
00:12:54.12 So the van der Wahl's radius of an azide is only about 2.4 Angstroms.
00:12:59.22 Which means that it is just a little bit larger than a methyl group.
00:13:03.08 And that's important because we are attaching an azide to a sugar
00:13:08.11 and then asking all of the enzymes
00:13:09.23 inside the cell to metabolize that sugar as if
00:13:12.17 it were a normal metabolic substrate.
00:13:15.14 And so we can't change the structure too much,
00:13:18.12 otherwise those enzymes might reject
00:13:21.06 that sugar as being too foreign, too unnatural.
00:13:23.25 But the azide is a pretty small modification.
00:13:26.19 We thought maybe we could sneak it through the door.
00:13:28.18 of some of those enzymes.
00:13:31.01 Let me just make a point because
00:13:33.08 I am asked this question frequently
00:13:34.20 amongst biologists. Many biologists are familiar with the word azide
00:13:40.08 in the context of sodium azide
00:13:42.13 and they know sodium azide to be a metabolic poison.
00:13:46.27 They often put a little bit of sodium azide into their buffers
00:13:51.01 to prevent bacterial growth. So they know that sodium azide is cytotoxic.
00:13:55.08 and therefore they ask me, "Aren't azides cytotoxic?
00:13:59.04 How can you put an azide on a sugar
00:14:00.18 and feed it to cells and feed it to organisms? Isn't that toxic?"
00:14:03.24 The answer is no. Sodium azide is a very different entity
00:14:08.11 from azides attached to a molecule.
00:14:11.14 So once the azide is attached to a molecule like a sugar,
00:14:14.19 it's totally harmless. By contrast, if azide is a salt like sodium azide,
00:14:20.02 you are right, then it is very toxic.
00:14:21.25 But we don't actually have to worry about that.
00:14:24.01 because there is no way in your body to convert a sugar azide
00:14:28.03 to something like sodium azide.
00:14:30.14 So, don't worry about that.
00:14:31.24 Okay. Now what's most important about the azide
00:14:35.16 is that the azide can undergo chemical
00:14:38.20 reactions with other functional groups
00:14:41.12 which remember in the previous slide I called those Y.
00:14:44.06 The other group is Y. This group is X.
00:14:46.18 There are several different Y groups that will react with azides.
00:14:50.27 And the first such group that we explored
00:14:54.03 is a phosphine. So a phosphine is what you get when you take a phosphorus atom
00:14:59.18 and you link three substituents to it like I've shown here.
00:15:03.25 Now back in the early part of the 20th century.
00:15:07.27 We are talking around 1915 to 1920,
00:15:11.12 a German chemist by the name of Hermann Staudinger
00:15:14.28 discovered that phosphines will react with azides to generate
00:15:20.17 a product in which there is a phosphorus-nitrogen bond
00:15:24.08 and he called this kind of an intermediate an aza-ylide.
00:15:27.19 It's a very interesting reaction in that these
00:15:30.22 two chemical groups join together and make this covalent
00:15:34.22 adduct. We were intrigued by this Staudinger chemistry
00:15:38.29 because both phosphines and azides are bioorthogonal
00:15:43.08 and their reaction is very selective. They are very non-toxic.
00:15:47.27 And it seemed to fulfill many of these criteria for bioorthogonality.
00:15:52.07 With one exception. It turns out that aza-ylides
00:15:55.16 are not particularly stable in water.
00:15:58.00 They tend to hydrolyze.
00:16:01.05 So the P-N bond gets cleaved.
00:16:03.00 And of course, if our goal is to use the chemistry
00:16:05.22 to attach a probe to a sugar in the living body
00:16:09.10 where we are surrounded by water
00:16:12.09 then that hydrolytic sensitivity would be a major problem for an aza-ylide.
00:16:17.01 So what we did was we modified this reaction in a fairly subtle way
00:16:22.19 where we attached to the phosphorus atom a benzene ring
00:16:26.18 that had this methoxycarbonyl group.
00:16:29.01 So now, once the aza-ylide was formed
00:16:32.18 faster than it could hydrolyze, this nitrogen atom
00:16:35.16 cyclized to this carbonyl, cleaved the ester bond,
00:16:39.18 and formed an amide.
00:16:41.29 And this cyclization was so fast that it didn't
00:16:45.04 even matter whether there was water around.
00:16:47.09 Now once this product was formed,
00:16:49.08 water could come along at its leisure
00:16:51.15 hydrolyzed the P-N bond and form this product, but now, because an amide
00:16:56.28 bond had formed during this intramolecular cyclization step
00:17:01.04 these two parts, one that came from the azide
00:17:04.00 the other attached to the phosphine
00:17:06.14 were still linked together through a very stable linkage.
00:17:09.09 So this is what we call the Staudinger ligation.
00:17:13.18 It's a reaction in which we take advantage of the classic Staudinger chemistry
00:17:19.02 in the first step but then we alter it to form a product
00:17:22.20 in which the two components are permanently
00:17:24.14 ligated together in a manner that is
00:17:26.08 very stable in a biological system.
00:17:29.03 And the Staudinger ligation was one of the first
00:17:32.06 chemical reactions that we used to try to image sugars in living systems.
00:17:38.01 So which sugars would you like to image?
00:17:41.24 Of course there are many interesting choices.
00:17:44.08 And as I introduced in my first lecture,
00:17:46.24 in vertebrates there are nine
00:17:49.07 fundamental monosaccharide building blocks
00:17:52.01 that are the constituents of our various glycans.
00:17:55.04 They are all interesting in their own right,
00:17:58.01 but when we were thinking about what sugar would we like to image,
00:18:01.10 we were naturally drawn to this monosaccharide,
00:18:05.24 called sialic acid. And as I mentioned in my first lecture,
00:18:09.21 sialic acid tends to appear in some very interesting circumstances
00:18:15.08 in vertebrate biological systems.
00:18:18.12 I mentioned for example, that sialic acid
00:18:21.01 is a component of the ligand for the selectins.
00:18:24.27 I also mentioned that sialic acid
00:18:27.01 is the structure that the influenza virus binds to.
00:18:30.18 Well, it turns out that sialic acid is a pretty busy monosaccharide.
00:18:36.20 because it also seems to be related to embryonic development
00:18:40.19 and to cancer. In fact, the glycomes of embryonic cells,
00:18:46.01 differentiating tissue, as well as cancer cells
00:18:49.14 have been studied in much detail
00:18:52.15 and it turns out that many of the glycans
00:18:54.25 found both embryonically and on cancers
00:18:57.22 have sialic acid as one of their constituents.
00:19:00.19 So what I have shown here are just three examples from a much larger set.
00:19:05.03 These are glycan structures that have been found to be highly elevated
00:19:10.25 on cancers compared to the normal healthy tissue counterpart.
00:19:14.14 And many of them are also developmentally regulated.
00:19:17.29 which is probably not an accident, because we know that often cancers
00:19:22.14 start to acquire properties that we
00:19:25.12 normally associate with embryonic tissue.
00:19:27.19 They've sort of lost their identity
00:19:30.11 and they are reverting to an embryonic state.
00:19:32.23 and that's reflected in their glycan structures.
00:19:35.27 So polysialic acid which is a homo-polymer
00:19:38.28 made up of repeating units of sialic acid
00:19:41.16 is highly abundant on a variety of tumors
00:19:44.13 that are derived from neural crest tissue.
00:19:46.16 Sialyl Lewis X, which I mentioned before
00:19:50.02 is part of the ligand for the selectins,
00:19:54.02 is also found in elevated levels on a variety of different epithelial cancers.
00:19:57.08 and blood cancers. And Sialyl Tn, a very simple disaccharide,
00:20:03.12 is actually not normally found on any healthy human tissue
00:20:07.12 but it appears on a variety of prostate cancers.
00:20:11.03 and we don't really know why that is,
00:20:12.29 but the fact that these structures appear in the context of malignancies
00:20:18.06 has suggested to many people that there might be
00:20:21.03 elevated levels of sialic acids on these tissues.
00:20:25.25 So if we could image the sialic acids,
00:20:28.05 maybe we could detect these tumors in living systems.
00:20:31.25 That was one of the motivations to study the sialic acids.
00:20:35.25 Okay. So then the question became, in our minds,
00:20:39.24 how do we put an azide into the sialic acids?
00:20:43.06 Well, we have to understand the fundamental metabolism
00:20:46.22 that produces sialic acid in our body.
00:20:50.05 And that's what is shown in this slide.
00:20:52.15 It all begins with this simple monosaccharide
00:20:56.23 building block called N-acetylmannosamine.
00:21:00.08 and we abbreviate that ManNAc for short.
00:21:03.14 So you eat ManNAc and in fact if you eat bread or drink beer,
00:21:09.13 any yeast product will have some ManNAc.
00:21:11.21 If you don't eat ManNAc, that's okay
00:21:14.15 because you are able to biosynthesize it from glucose
00:21:17.21 through other enzymatic steps that I haven't shown.
00:21:20.27 But in any event, once ManNAc enters your system,
00:21:23.19 it has basically one metabolic fate.
00:21:26.15 It is destined to be converted to sialic acid.
00:21:28.27 in your cells. And that happen through a series of enzymatic steps
00:21:34.18 and I am not going to go through all these details.
00:21:37.18 You can certainly look at the slide in detail if you are interested.
00:21:40.17 But it suffices to say, that you eat this sugar
00:21:44.00 and after all of these enzymes have acted on it
00:21:46.29 at the end of the day a sialic acid is made,
00:21:50.05 and it appears usually at the end of complex glycan chain
00:21:54.19 on the surface of your cell, either in a glycoproteins or glycolipids.
00:21:58.13 So we knew from the work of many labs
00:22:01.17 that this sugar gets converted to this sugar.
00:22:04.07 And I've shown the acetyl group in blue.
00:22:08.04 So that it is clear where that ends up in the biosynthetic product.
00:22:11.28 And the reason that I highlight that acetyl group
00:22:14.12 is because we know from the work of several groups
00:22:17.27 that you can make subtle modifications
00:22:20.22 to the structure at this position without
00:22:23.14 significantly harming the efficiency of
00:22:26.13 any of these enzymatic steps,
00:22:28.06 which is really quite remarkable if you think about it.
00:22:30.23 Normally we think of enzymes as being very particular
00:22:34.18 for their substrates, but in this pathway
00:22:37.26 there is a little bit of leeway and you
00:22:39.00 can modify this side chain a little bit.
00:22:41.26 So knowing that, it was clear to us that this was a site
00:22:46.08 that we might be able to attach an azide
00:22:48.16 without confusing the enzymes in the cell.
00:22:52.01 And that's exactly what we did.
00:22:54.27 So by chemical synthesis we generate this unnatural version
00:22:59.04 of ManNAc, which has an azide.
00:23:01.21 So as an abbreviation we call this compound ManNAz,
00:23:06.04 where it is the azide version of ManNAc.
00:23:09.10 And we wanted to get ManNAz into cells, so to do that we
00:23:15.12 block each of the hydroxyl groups, which are characteristic of sugar molecules
00:23:19.29 with acetyl esters. Those are protecting groups.
00:23:24.13 So once the acetyl groups are in place the sugar is now sufficiently lipophilic
00:23:29.21 that it will penetrate through the cell's membrane, and once it is inside the cell,
00:23:35.09 we have non-specific esterases that simply cleave
00:23:38.25 these acetyl groups off and throw them away.
00:23:40.22 And then the sugar is ready for metabolism.
00:23:43.05 So the enzymes start processing the sugar
00:23:46.11 and after several hours what we found is that,
00:23:48.15 lo and behold, some fraction of the sialic acid residues
00:23:52.07 on these cells had been replaced with an azido analog.
00:23:56.23 So in other words, all we do is add this compound to cell culture media,
00:24:01.16 and the cells do all the hard work
00:24:03.05 and they produce this azido-sialic acid on their cell surface glycans.
00:24:08.01 Remarkably the cells don't seem particularly
00:24:11.28 offended by this transformation.
00:24:13.26 Some fraction of their sialic acids
00:24:16.18 have been replaced with this unnatural variant
00:24:18.29 and that seems fine with them.
00:24:21.16 But it's very convenient for us
00:24:22.28 because now we can use the azide to do chemistry.
00:24:26.18 And so by introducing a phosphine reagent,
00:24:29.05 that's been linked to a fluorescent probe, we
00:24:32.13 do a Staudinger ligation between the phosphine and the azide.
00:24:35.10 now there's a covalent bond between the probe molecule and the sugar.
00:24:40.22 So wherever there is a sialic acid,
00:24:42.16 we can now visualize it
00:24:44.06 using fluorescence microscopy focusing on the probe.
00:24:47.12 And so in this manner we have been able to image
00:24:50.17 sialic acids on a variety of interesting cell types.
00:24:54.01 And sialic acid is not the only sugar
00:24:56.29 that we can see using the azide as a chemical reporter.
00:25:00.20 So, just as we can introduce the azide into sialic acid,
00:25:05.07 by feeding the cells ManNAz.
00:25:08.09 It turns out we can introduce an azide into fucose
00:25:11.22 simply by feeding the cells this 6-azido fucose derivative
00:25:16.23 that has acetyl esters protecting its hydroxy groups.
00:25:19.26 And likewise, we can introduce azides into N-acetylgalactosamine
00:25:25.18 or GalNAc, by feeding cells this modified form of GalNAc,
00:25:30.18 that has an azide on its acetyl group. We call this GalNAz.
00:25:34.13 analogous to ManNAz, and I actually will be coming back to this sugar
00:25:38.20 GalNAz towards the end of the lecture.
00:25:41.10 We also put quite a bit of time into developing probes
00:25:45.16 that we can use to visualize these sugars.
00:25:49.10 Now the Staudinger ligation is wonderfully selective
00:25:51.24 in that it allows us to form a covalent bond between the probe and the sugar.
00:25:57.06 But more than that we can exploit elements
00:25:59.24 of that ligation chemistry in order to design what we call smart probes.
00:26:05.11 So these are probes that are actually invisible
00:26:08.06 until they find an azido sugar
00:26:11.28 form that bond through the Staudinger ligation,
00:26:14.13 and then they become fluorescent.
00:26:16.04 So they basically swim around in the system,
00:26:18.11 they look for azides and as soon as they find them,
00:26:20.26 they make a bond and light up, and you don't see them otherwise.
00:26:23.29 This allows us to do fluorescence imaging
00:26:26.01 with very good signal above background.
00:26:28.23 The way we do this is to exploit the fact that
00:26:32.19 during the course of the Staudinger ligation,
00:26:34.11 this ester bond will be cleaved.
00:26:36.18 In the previous slides I had simply a methyl ester at this position.
00:26:41.18 where methanol was released during the reaction
00:26:44.28 which is not particularly remarkable.
00:26:46.25 But, one could replace that methyl ester
00:26:49.00 with an ester linkage to a fluorescence quencher.
00:26:52.23 And if there is a fluorescent molecule bound elsewhere,
00:26:55.16 that fluorophore will be quenched
00:26:57.19 by the quencher in this initial molecule.
00:27:01.01 But as soon as this probe finds an azide,
00:27:03.21 located within a sialic acid, let's say, or another sugar on the cell,
00:27:08.02 the phosphine reacts with the azide
00:27:10.14 the Staudinger ligation unfolds,
00:27:13.01 the ester is cleaved, the quencher is released,
00:27:16.12 now that fluorophore is activated.
00:27:19.18 So only once the fluorophore finds its target
00:27:22.09 and reacts, only then do you see the fluorescence.
00:27:25.14 Otherwise it's invisible.
00:27:26.27 Let me show you an actual chemical structure of a probe
00:27:29.25 that we prepared that follows precisely this mechanism.
00:27:35.08 This is a molecule that we call affectionately "QPhos",
00:27:40.06 which stands for quenched phosphine.
00:27:41.22 It's got three parts to it:
00:27:44.20 so over here, this green part is a very well known
00:27:48.08 fluorescent molecule called fluorescein.
00:27:50.07 It's a green fluorescent dye commonly used in the research laboratory.
00:27:55.22 In the middle in black is a phosphine that's all set up for a Staudinger ligation.
00:28:01.29 You can see that on this benzene ring there is an ester group.
00:28:06.05 That's positioned for that intramolecular reaction
00:28:09.14 of the aza-ylide intermediate.
00:28:11.06 And then finally over here, the blue part
00:28:14.03 is a well-known fluorescence quencher,
00:28:16.23 called Disperse Red 1.
00:28:18.28 And as long as this Disperse Red 1 is linked to this ester
00:28:22.15 it will quench the fluorescence of the fluorescein molecule.
00:28:25.26 So let me show you some images that we took using QPhos
00:28:30.07 to detect sialic acid on cancer cells in culture.
00:28:35.12 So in this experiment what we did was we grew some cells
00:28:40.12 called HeLa cells. These are human cervical epithelial cancer cells
00:28:45.22 that you can grow in a test tube. And while those cells were growing,
00:28:49.13 we fed them ManNAz in the media.
00:28:53.03 We just added it to the cell culture media,
00:28:55.08 they took it up, they digested it, and they put azides into their sialic acids.
00:29:00.13 Then, after a few days of treatment with ManNAz,
00:29:03.09 we reacted those cells with QPhos.
00:29:06.22 And then after a couple of hours we took a fluorescence image of those cells.
00:29:11.13 The cells were also stained with a blue fluorescent dye
00:29:15.03 that is specific for the nucleus.
00:29:16.17 That just helps you to orient exactly where the cells are positioned
00:29:20.27 since each one has one nucleus.
00:29:22.29 So, what you can see is that each cell
00:29:25.17 is surrounded by a nice bright green outline.
00:29:29.15 Those are the sialic acids on the
00:29:32.05 membrane associated glycoproteins and glycolipids.
00:29:35.09 of the cell and they are nicely lit up.
00:29:37.25 We are actually imaging the sugars.
00:29:39.26 You might see a little bit of fluorescence inside the cells as well.
00:29:43.16 That's basically part of the secretory pathway,
00:29:47.23 the Golgi compartment for example.
00:29:49.01 And then there's a little bit of staining of
00:29:51.14 little vesicles throughout the cells as well.
00:29:53.05 That's because after we react sialic acids on the membrane
00:29:57.17 some of those sialic acids end up internalized
00:30:00.21 into vesicles because the membranes are constantly being
00:30:03.26 engulfed and recycled and turned over.
00:30:05.29 They are going back into the late part of the Golgi compartment
00:30:08.23 and then coming back to the membrane.
00:30:10.09 There's a lot going on in fact during the course of reaction with QPhos.
00:30:14.26 Now we had some issues once we started
00:30:17.29 doing these kinds of experiments.
00:30:19.07 For example, we realized that in order to get this nice bright staining
00:30:23.24 of the sialic acids, we had to react the cells with QPhos for several hours.
00:30:28.27 Now that's not a problem for imaging the sugars on cultured cells,
00:30:34.17 because the cells are in a flask with the reagents
00:30:37.26 and they can sit in the incubator for several hours.
00:30:40.12 and just let the reaction unfold.
00:30:42.09 The reason we had to let the reaction go for several hours
00:30:45.15 as opposed to several minutes
00:30:47.02 or several seconds is that it turns out the Staudinger ligation reaction
00:30:51.02 is intrinsically rather slow. It just takes that long for the reaction to proceed.
00:30:57.24 And although that was not too problematic
00:31:00.21 for imaging experiments on cultured cells,
00:31:03.25 we worried that that could be a problem
00:31:06.23 for imaging sugars in living animals.
00:31:10.09 because of course a living animal is not at equilibrium.
00:31:14.10 It is not cells in a flask sitting in an incubator for many hours.
00:31:17.26 Animals have an active metabolism.
00:31:21.04 As soon as you introduce reagents into their bodies,
00:31:24.19 their bodies are working on clearing those reagents right out again.
00:31:27.29 And that clearance process can be very rapid for certain molecules
00:31:32.18 and certain animals. In fact, when we started looking into the Staudinger ligation
00:31:38.06 as a tool for imaging sugars in laboratory mice
00:31:41.23 we immediately ran into this problem.
00:31:44.29 So we injected mice with ManNAz
00:31:48.08 and we found that in the animal
00:31:52.03 the cells were perfectly willing to take up this sugar
00:31:54.23 and convert it to the azido-sialic acid,
00:31:58.01 that was no problem.
00:31:59.21 And it seemed harmless to the mice.
00:32:01.17 So there was no obvious detriment to replacing
00:32:04.09 some fraction of the sialic acids with the azido version.
00:32:08.00 The problem came when we then injected those same mice
00:32:11.26 with probe molecules bound to a
00:32:15.09 phosphine for the Staudinger ligation.
00:32:17.04 Because what we discovered is that the
00:32:19.24 reaction between the phosphine and the azide
00:32:21.26 was just too slow compared to the rate of metabolic clearance
00:32:25.23 of the probe out of the animal's body.
00:32:28.02 So we could only detect a little bit of this product
00:32:31.26 just on certain organs and tissues
00:32:34.00 The organs and tissues that were the most accessible
00:32:36.22 to the reagents after we injected them
00:32:38.14 or the organs and tissues that had really, really high levels
00:32:41.23 of the sialic acid we were trying to image.
00:32:45.07 And that was a little frustrating,
00:32:47.00 but it basically pointed to a very important element
00:32:50.12 of this experiment, which we really hadn't considered at the very outset.
00:32:55.07 You know, ten years ago.
00:32:56.04 Which is that the kinetics of the reaction
00:32:59.05 between the azide and the probe molecule
00:33:01.16 are very, very important for imaging in living animals.
00:33:06.02 That reaction has got to be fast enough
00:33:08.29 so that the probe can find the azide and react
00:33:12.11 more quickly than it is cleared out of the body.
00:33:14.21 And after many years of experimentation,
00:33:17.21 we finally concluded that the Staudinger ligation was just too slow.
00:33:23.11 Its kinetics were too slow compared to its rate of metabolism.
00:33:27.22 And it probably didn't help the situation that phosphines, in general,
00:33:32.25 can be oxidized in the mammalian liver
00:33:35.15 by cytochrome P450 enzymes.
00:33:37.05 which convert the phosphine to a phosphine oxide
00:33:42.08 and once that occurs, now this product
00:33:45.08 is no longer active in a Staudinger ligation.
00:33:47.08 So, undoubtedly, liver metabolism of the phosphine
00:33:51.19 contributed to its rapid clearance rate
00:33:54.23 which just out competed the intrinsically slow kinetics of the reaction.
00:33:58.23 So the bottom-line is that the Staudinger ligation
00:34:02.04 was great for in vitro imaging on cells in culture.
00:34:06.00 Not so great for in vivo imaging in live organisms
00:34:10.00 that have the ability to clear reagents rapidly out of the system.
00:34:13.20 So at that point we shifted our attention
00:34:17.19 to a different kind of chemistry that azides can undergo.
00:34:21.16 It's another chemistry that has its origins in the last century
00:34:26.03 in the hands of a German chemist.
00:34:27.28 In this case, Rolf Huisgen, Professor at University of Munich.
00:34:31.26 who back in the 1950s discovered that azides
00:34:35.26 can react with alkynes, undergo a 1,3-dipolar
00:34:41.07 cycloaddition reaction whose mechanism is shown here,
00:34:44.19 to form a cyclo adduct product called a triazole.
00:34:49.00 Now like the Staudinger chemistry,
00:34:52.04 this Huisgen cycloaddition chemistry
00:34:55.02 is bioorthogonal in that alkynes and azides
00:35:00.08 both are not found in biological systems,
00:35:03.04 at least not in most biological systems
00:35:05.06 and they react with each other very selectively
00:35:08.25 without any unwanted side reactions in the living system.
00:35:13.19 So we thought this would be another very promising avenue to explore.
00:35:16.25 The problem is that the classic Huisgen reaction
00:35:21.09 as described with a standard linear alkyne
00:35:24.27 is also very slow. In fact, even slower
00:35:28.11 than the Staudinger chemistry with phosphines.
00:35:30.28 So slow, that when people perform
00:35:34.22 these reactions in a research laboratory,
00:35:37.04 they usually have to apply elevated temperatures
00:35:40.03 or elevated pressures in order to get the reaction
00:35:43.08 to proceed to completion at a reasonable rate.
00:35:45.26 And when I say elevated temperatures,
00:35:47.22 I mean above 100 degrees centigrade for example.
00:35:51.19 So refluxing toluene. Hotter than the boiling temperature of water.
00:35:56.27 Obviously we can't do that to cells
00:36:00.14 and certainly not to living organisms.
00:36:02.14 But we started thinking about ways that we might be able to
00:36:05.18 accelerate the rate of this chemistry.
00:36:08.21 It turns out, back in the 1960s,
00:36:11.19 there was a publication by Wittig and Krebs
00:36:14.25 in which they reported that phenylazide reacts with
00:36:20.08 cyclooctyne which has a triple bond crimped into an eight-member ring
00:36:26.03 to form the triazole product very rapidly.
00:36:30.16 Now my graduate students were reading this paper.
00:36:33.25 It was published in German,
00:36:35.29 and unfortunately at that time we did not have
00:36:38.25 any coworkers in the laboratory whose German was good enough
00:36:41.18 to really understand the paper in its fine detail.
00:36:44.06 but all of us read this sentence
00:36:46.08 and recognized three words: phenylazide, cyclooctyne, and explosion.
00:36:53.02 So as far as we could see, Wittig and Krebs
00:36:57.14 had combined these two reagents
00:36:59.12 and they formed a product like an explosion.
00:37:02.11 which suggested to us that this must be a pretty fast reaction.
00:37:05.17 Keep in mind however, that what Wittig and Krebs did
00:37:09.05 is to react neat cyclooctyne,
00:37:12.03 which is a liquid, with neat phenylazide with no solvent.
00:37:17.05 So these were very concentrated reagents that reacted like an explosion.
00:37:20.09 And we figured, if we dissolve these reagents at lower concentrations
00:37:25.08 and prototypical solvents, maybe the reaction
00:37:28.07 will be very nicely controlled,
00:37:29.19 hopefully still very rapid at room temperature.
00:37:32.13 Now why you might ask does cyclooctyne
00:37:36.06 react with an azide like an explosion whereas
00:37:39.00 linear alkynes are so slow that you have to
00:37:42.17 heat them up above 100 degrees?
00:37:44.05 That can be understood by looking at what happens
00:37:47.27 to an alkyne when you force it into an eight membered ring.
00:37:51.13 Now normally, alkynes are linear.
00:37:54.12 That is their preferred geometry.
00:37:56.08 And when an alkyne reacts with an azide, this bond angle
00:38:00.28 is going from 180 degrees down to 120 degrees in the triazole product.
00:38:07.02 So that bond deformation costs a lot of energy
00:38:11.01 in the transition state for the reaction.
00:38:12.29 That's why we have to heat it up.
00:38:14.24 To give some of that energy to get over that activation barrier.
00:38:17.23 By contrast, if the alkyne is embedded in an eight membered ring
00:38:23.04 now the bond angle is bent to 160 degrees
00:38:27.25 It's not ideal, in fact, it costs a lot of strain
00:38:31.19 to bend a triple bond to 160 degrees
00:38:35.16 But the upshot is that now the difference
00:38:38.13 between 160 degrees and 120 degrees is quite a bit smaller.
00:38:42.06 It doesn't cost as much energy to go
00:38:44.09 from this structure through a transition state
00:38:46.22 to get to this product.
00:38:48.03 So our hope was that by straining the starting material
00:38:52.12 and bending the bond angle so it's closer
00:38:54.25 to the structure of the transition state
00:38:56.17 we would lower the activation barrier
00:38:58.24 so that the reaction can proceed at room temperature.
00:39:01.13 And that's exactly what we observed when we chemically synthesized
00:39:05.17 a whole family of cyclooctynes in the laboratory.
00:39:09.16 So let me just quickly summarize some kinetic comparisons
00:39:14.10 that we performed using a variety of synthetic reagents.
00:39:19.10 And so each of these compounds is a cyclooctyne
00:39:22.16 of some sort in some cases with a heteroatom in the ring
00:39:26.20 with various different substituents at different sites
00:39:29.12 and we made these compounds for a variety of reasons
00:39:32.21 that I won't go into.
00:39:33.13 but you'll see in some cases there were sp2 hybridized carbons
00:39:37.11 elsewhere in the ring. Our hope was that we could increase the strain
00:39:41.15 energy a little bit more by putting
00:39:43.16 sp2 hybridized carbons in there.
00:39:46.01 And some of these reagents
00:39:47.03 have electronegative fluorine substituents.
00:39:49.27 which we also though might accelerate the reaction
00:39:53.01 based on some frontier molecular orbital arguments.
00:39:56.12 Also shown among this collection is a phosphine reagent
00:40:00.17 which is set up to a Staudinger ligation.
00:40:03.03 What we did is we took each of these compounds.
00:40:06.07 We reacted each compound with benzyl azide
00:40:09.27 in a test tube. And then we measured the
00:40:12.12 second order rate constant of the reaction.
00:40:14.26 And in blue are shown basically the relative values that we measured.
00:40:19.24 So the bottom-line is that we found a compound
00:40:24.03 that reacts very rapidly with azides in a test tube.
00:40:28.15 This is a di-fluorinated cyclooctyne
00:40:32.00 that we have given the abbreviation DIFO.
00:40:34.22 So that's stands for Di-Fluoro Cyclo Octyne.
00:40:39.00 DIFO. And the relative rate of reaction of DIFO
00:40:42.26 compared to some other compounds you'll see,
00:40:45.17 for example a compound without the fluorine atoms.
00:40:48.02 is very high. So we can accelerate the rate 75 fold
00:40:52.10 just by putting some fluorine atoms here
00:40:54.29 compared to some parent compounds.
00:40:57.08 Importantly, if you compare the relative rate of reactivity
00:41:00.08 between DIFO compared to a Staudinger ligation phosphine,
00:41:03.06 you will see that it is a good 15 times faster.
00:41:06.20 And so we felt that where the Staudinger ligation
00:41:09.26 failed in vivo because of its slow kinetics.
00:41:13.02 we might find success now with DIFO
00:41:16.11 as a reactive counterpart to the azide.
00:41:20.01 So we tested that concept in a variety of imaging experiments
00:41:25.15 starting simply with cultured HeLa cells.
00:41:28.15 So remember, in a previous slide, I showed you
00:41:31.02 some images of the sialic acids on HeLa cells
00:41:34.26 where we reacted the azides with QPhos.
00:41:37.19 This is now a very similar experiment but
00:41:41.10 once we put the azides into the sialic acids
00:41:43.24 this time we react the azides with this compound.
00:41:48.05 So we call this DIFO-488.
00:41:51.00 Down here is the DIFO part, the di-fluoro cyclooctyne,
00:41:55.23 The green part of the molecule is a commercial
00:41:58.22 fluorescent dye called Alexafluor 488.
00:42:01.23 It's a bit like fluorescein in its spectral properties.
00:42:05.02 That is why we call this DIFO-488.
00:42:07.22 So what you are looking at here are the cultured HeLa cells.
00:42:11.19 They have been fed ManNAz to put azides into the sialic acids.
00:42:16.05 Then they were reacted with DIFO-488
00:42:19.03 and you can see that the sialic acids are shown in these nice
00:42:22.02 bright green halos around each cell.
00:42:24.10 The membranes are nicely lit up.
00:42:26.20 Those are the sialic acids.
00:42:28.06 This is a control experiment
00:42:30.12 in which instead of feeding the cells ManNAz,
00:42:33.27 we now fed them the natural sugar, ManNAc.
00:42:36.28 There are no azides in these cells.
00:42:39.20 We also reacted them with DIFO-488.
00:42:43.08 And now you don't see any green labeling of the membranes.
00:42:47.01 That is because there are no azides for DIFO to react with.
00:42:50.09 That's a testament to the selectivity of DIFO.
00:42:54.06 It only reacts with azides.
00:42:55.26 If there are no azides around, there is nothing for it to react with.
00:42:58.14 You don't see the sugars.
00:43:00.11 But the most important number on this slide
00:43:03.02 is this one right here.
00:43:05.09 Now we can see a beautiful image like this
00:43:07.17 by reacting the cells for less than 1 minute with DIFO-488.
00:43:13.02 Whereas in the previous experiment with QPhos,
00:43:16.10 the Staudinger ligation reagent,
00:43:18.20 we had to react the cells for over two hours.
00:43:22.02 If we can get this kind of an image within a minute,
00:43:25.22 I think now we can image the sugars in living animals.
00:43:29.28 This reaction is fast enough, so we hoped.
00:43:32.24 So, what animal model would one want to use to image the sugars?
00:43:38.17 Mice are obviously a very attractive model because one can
00:43:42.17 study a variety of human diseases in that model organism
00:43:46.04 including cancer. However, mice are not ideal for optical imaging,
00:43:50.15 because, as you know they are not transparent.
00:43:53.18 They are opaque.
00:43:54.20 One can do fluorescence imaging inside a mouse,
00:43:57.15 but it requires a specific kind of dye, generally a near infrared dye.
00:44:02.06 and it's a more complicated experiment.
00:44:04.19 And for a first attempt at in vivo optical imaging,
00:44:08.23 we thought we would use a model organism
00:44:11.02 that is a little friendlier to an optical microscope.
00:44:14.04 And the obvious choice was the zebrafish.
00:44:18.03 So zebrafish are translucent.
00:44:20.09 You can see right through them
00:44:21.24 and monitor their organs in vivo.
00:44:24.10 in the live animal. Also, zebrafish
00:44:27.16 are a very attractive model for vertebrate developmental studies.
00:44:31.29 They have all the same organs and parts that we have.
00:44:35.23 They are vertebrates. And their embryonic
00:44:38.04 developmental program has been very well characterized.
00:44:42.10 So we know what zebrafish embryos
00:44:43.28 should look like at virtually all stages of their development.
00:44:48.00 We also know that there are changes in the glycome.
00:44:51.29 that correspond with embryonic development.
00:44:55.05 And so we though this was a very nice organism
00:44:57.06 to first of all test our chemical tools,
00:44:59.25 but second of all we might be able to address
00:45:02.22 some very interesting fundamental questions
00:45:04.25 of how the glycome changes during development.
00:45:07.15 And learn something about the field of glycobiology.
00:45:10.06 So, we started with some proof of concept experiments.
00:45:14.09 in which we focused not on sialic acid
00:45:17.06 but rather on a sugar called N-acetylgalactosamine.
00:45:22.05 abbreviated GalNAc for short.
00:45:24.12 The reason we were so interested
00:45:26.11 in GalNAc is because, as I mentioned in my first lecture,
00:45:30.17 it's the conserved core residue in a family
00:45:34.09 of O-linked glycans that are associated
00:45:37.14 with glycoproteins from the mucin family.
00:45:41.10 Mucins are a general class of glycoproteins
00:45:44.29 that are characterized by having dense clusters
00:45:47.29 of these O-linked glycans along the polypeptide backbone.
00:45:52.10 They're involved in cell adhesion,
00:45:54.10 regulating cell-cell interactions, and they are known
00:45:57.24 to be both developmentally regulated
00:45:59.15 and sometimes upregulated during malignant transformation.
00:46:03.13 So the mucins are an interesting class of glycoproteins.
00:46:07.13 They all have a GalNAc residue at their core position.
00:46:10.28 So we thought if we could introduce an azide
00:46:13.03 into this GalNAc residue perhaps we could image the mucins
00:46:18.20 in vivo. And the way we do that is to
00:46:21.23 simply feed cells, or in this case, zebrafish embryos,
00:46:25.23 This azido-acetyl derivative that we call GalNAz
00:46:30.14 once again, the hydroxy groups are protected with acetyl esters.
00:46:34.13 So that the compound can penetrate cell membranes
00:46:37.08 and inside the cell the acetyl groups are removed by esterases.
00:46:41.07 in vivo. Okay. So this was one of our first experiments
00:46:46.00 we performed to simply ask the question will zebrafish embryos
00:46:50.02 metabolize GalNAz and introduce it into the mucins,
00:46:54.09 and if so, can we then tag the azides
00:46:57.05 with DIFO reagents and image that sugar in vivo.
00:47:01.27 So in this experiment we fertilized zebrafish embryos in a test tube
00:47:06.05 and we added GalNAz to the culture media
00:47:09.27 and just let them develop over the course
00:47:11.28 of several days and then after about five days
00:47:15.11 we took the five day old zebrafish larvae
00:47:19.23 and we reacted it with DIFO linked to a fluorescent dye,
00:47:23.23 in this case it was DIFO-488,
00:47:25.24 the same compound I showed on the previous slide
00:47:28.06 and then we put those labeled zebrafish in a fluorescence microscope
00:47:33.00 and we took a picture. So what you are
00:47:35.14 looking at here is the labeled zebrafish
00:47:39.03 five days old, treated just as shown
00:47:41.14 in this schematic. And wherever you see any brightness
00:47:44.27 basically you are looking at the DIFO label
00:47:48.09 which has been covalently attached to the azido sugar.
00:47:52.00 There's also a five day old zebrafish in this panel.
00:47:55.01 You can't see it, I realize.
00:47:56.25 The reason is that the only difference
00:47:58.24 is that this zebrafish embryo developed in the presence of
00:48:01.28 the natural sugar GalNAc, which has no azides.
00:48:05.16 So when we reacted this five day old zebrafish with DIFO-488,
00:48:10.06 there was nothing for the dye to react with
00:48:12.08 so you don't see any fluorescence.
00:48:13.14 Again it testifies to the exquisite selectivity of DIFO
00:48:18.13 for the azide. So basically this experiment
00:48:21.28 demonstrated to us that we can in fact image
00:48:25.03 those GalNAz residues in a live zebrafish.
00:48:29.20 And I should point out that these zebrafish
00:48:32.10 look perfectly normal, perfectly healthy
00:48:34.16 and happy, despite that fact that some fraction
00:48:37.11 of their GalNAc residues were replaced with GalNAz,
00:48:40.14 and that some fraction of those GalNAz residues
00:48:43.29 were chemically reacted with the DIFO reagent.
00:48:47.10 These zebrafish were then released back into their tanks,
00:48:50.17 and they go on to live a normal life as far as we can see.
00:48:53.22 So it doesn't appear to be harmful to the organism
00:48:56.06 which is important if the goal is to learn something
00:48:58.16 about the organism's fundamental biology.
00:49:01.16 We can also do experiments
00:49:03.18 where we probe for changes in the glycome,
00:49:06.03 as a function of time
00:49:07.27 which again, is one of the most interesting
00:49:10.01 aspects of the field of glycobiology.
00:49:11.27 For example, here is an experiment in which we took
00:49:16.02 a zebrafish embryo fertilized in culture
00:49:18.15 We add to the culture media GalNAz,
00:49:21.22 and we let those embryos develop in the presence of the azido sugar
00:49:25.17 Now, at a given time point,
00:49:28.01 we will take those embryos out of the media
00:49:30.15 rinse them off, and react them
00:49:33.11 with DIFO conjugated to a red fluorescent dye.
00:49:36.26 In this case it was Alexafluor 647.
00:49:40.09 Now some glycoproteins, those that were labeled
00:49:44.08 with azides now appear red,
00:49:46.16 but we can take those same embryos
00:49:47.25 that are carrying a fluorescent probe
00:49:49.21 and just put them back in media
00:49:51.09 and let them continue their developmental program.
00:49:54.04 What we do first is a 10 minute reaction
00:49:56.19 with a reagent called TCEP,
00:49:58.16 this is tricarboxydiethylphosphine.
00:50:01.05 What this reagent does is it reduces any
00:50:05.01 unreacted azides to the corresponding amine.
00:50:07.11 And incidentally the chemistry by which TCEP
00:50:10.17 reduces those unreacted azides
00:50:12.16 is classic Staudinger chemistry.
00:50:15.10 We are able to take advantage of that.
00:50:17.25 So we just destroy any unreacted azides.
00:50:20.08 We take these red labeled embryos,
00:50:22.16 put them back into the media containing GalNAz,
00:50:26.03 They continue to develop.
00:50:27.06 They take up more GalNAz as they are doing so.
00:50:29.21 And they'll integrate GalNAz into the next
00:50:32.26 wave of mucin glycoproteins.
00:50:35.06 And now we can take the embryos out
00:50:37.18 and label them again, this time with DIFO
00:50:40.03 conjugated to a green fluorescent dye.
00:50:42.16 So now, at the end of all of this,
00:50:44.29 there are two fluorescent dyes on the zebrafish.
00:50:48.13 The green dye reflect these azides that are
00:50:52.13 in glycoproteins biosynthesized most recently.
00:50:55.11 Whereas the red dye reflects the older population
00:50:58.18 of glycoproteins that were introduced earlier in development.
00:51:02.17 So this allows us to separate old glycoproteins from new glycoproteins
00:51:07.18 based on the color that they appear in the fluorescence microscope.
00:51:11.07 So that kind of an experiment is what led to images like this one
00:51:16.13 which I showed on the very first slide of this second lecture.
00:51:19.13 So this is the head of a 5 day old zebrafish
00:51:24.19 that has been labeled with GalNAz,
00:51:26.24 followed by three different colored fluorescent dyes
00:51:30.11 at three different time points.
00:51:31.17 In this particular image the newest glycoproteins
00:51:35.16 are appearing red, and you can see there's
00:51:37.11 a concentration of those glycoproteins
00:51:39.16 here in the olfactory organs,
00:51:41.22 which are basically the nostrils of the zebrafish.
00:51:45.05 Whereas older populations of glycoproteins appear either blue or green.
00:51:51.03 And you can see that the blue and the green
00:51:52.21 have quite a different distribution from the red.
00:51:55.10 And by looking at how these different colors
00:51:57.14 have moved around as a function of time,
00:51:59.26 we can learn something about the dynamics of the glycome.
00:52:04.02 At least from the perspective of these mucin glycoproteins
00:52:06.27 that we have labeled with GalNAz.
00:52:09.00 In fact one can use a confocal microscope to walk around
00:52:14.05 the labeled organism cell by cell
00:52:15.29 and take a very close look at what has happened
00:52:19.04 to the mucins as a function of time during the development.
00:52:22.21 Just for example, here is a panel of epithelial cells
00:52:26.16 and you can see that each epithelial cell looks as if it has
00:52:29.12 a red labeled membrane with blue and green puncta inside the cell.
00:52:34.26 Which is exactly what we would expect,
00:52:36.27 because in this experiment the red labeling
00:52:40.13 reflects the newest population of glycoproteins.
00:52:43.23 Those are the glycoproteins that just
00:52:45.18 emerged on the plasma membrane
00:52:47.08 with azides ready for us to label.
00:52:49.13 By contrast, older populations of glycoproteins
00:52:52.29 that were biosynthesized many hours before
00:52:55.15 we took this picture, they've already been recycled.
00:52:58.08 by endocytic vesicles in the membrane.
00:53:01.18 And that is why they appear inside the cell.
00:53:04.09 This is a structure that was rather dramatic
00:53:07.09 that appeared at around the 72 hour mark post fertilization.
00:53:12.11 What you are looking at here is a red projection
00:53:14.12 that is basically coming out of the screen
00:53:16.15 at you from the junction of three epithelial cells.
00:53:20.25 That red structure is a mechanosensory hair structure.
00:53:24.22 These are structures that protrude from the side
00:53:27.23 of the head of the embryo,
00:53:29.09 as you can see in this larger scale view.
00:53:32.06 These are structures that
00:53:33.19 the zebrafish can use to sense the flow of current in its environment.
00:53:36.28 for example.
00:53:38.20 And the fact that we captured this structure so dramatically in red
00:53:41.21 tells us first of all that it must contain these mucin like glycoproteins
00:53:47.29 otherwise we wouldn't be able to see it
00:53:49.23 because that is what we are imaging.
00:53:51.29 Secondly, that the majority of those glycoproteins
00:53:54.24 were formed within the window between
00:53:56.23 72 and 73 hours post-fertilization.
00:53:59.27 Which would be a very difficult window to capture
00:54:02.18 using other classic imaging technologies.
00:54:06.01 So these are the kinds of studies that are basically opening the window
00:54:10.01 through the vision of sugars into developmental biology.
00:54:14.10 And I think this is a very interesting
00:54:16.04 future application of the technology.
00:54:17.27 So what should you take home from this second lecture
00:54:23.02 in this series. First of all, there is a chemical lesson here,
00:54:26.11 which really has nothing to do with sugars.
00:54:28.15 It's the fact that, you know, we've taken advantage
00:54:31.13 of some very classic chemical tools.
00:54:34.08 These are chemical reaction that were first reported
00:54:37.20 as early as 1915. That is almost one hundred years ago.
00:54:41.15 And these old classic chemistries: the Staudinger chemistry
00:54:44.29 and the Huisgen cycloaddition chemistry
00:54:46.23 we have found to be real gems for modern applications in biology.
00:54:52.22 So by digging into the old chemical literature,
00:54:55.29 one can find some incredible tools
00:54:58.17 to bring forth into the modern era of biological research.
00:55:02.09 And that's very exciting for someone who was trained as a chemist
00:55:05.15 and at heart considers herself a chemist.
00:55:08.17 Secondly, you should remember the azide.
00:55:11.11 The azide has phenomenal properties
00:55:15.00 that make it so well suited for these applications
00:55:18.15 as chemical reporters.
00:55:19.25 And we think that the azide
00:55:21.13 has a lot of potential not just for imaging sugars,
00:55:24.09 the way that we do it, but for imaging
00:55:26.11 all kinds of interesting biomolecules.
00:55:28.26 including proteins, lipids, nucleic acids,
00:55:32.12 metabolites, things that are not necessarily encoded in the genome,
00:55:37.08 for which the GFP and other genetic reporters are really not well suited.
00:55:43.11 Third we now have shown that metabolic labeling with azido sugars
00:55:48.22 allows you to image the glycans in living systems.
00:55:51.23 And we think this will allow scientists
00:55:53.29 to study glycobiology using all of the hardware of molecular imaging
00:55:59.28 that so far has focused primarily on the study of proteins.
00:56:04.11 And finally, where we are going with this technology in my laboratory
00:56:10.04 is to continue our studies of developmental biologies
00:56:12.19 until we can understand how the glycome contributes
00:56:15.11 to some of the early cell fate decision making processes
00:56:18.12 during the embryogenic program.
00:56:20.10 For example, how does the glycome change
00:56:22.18 when pluripotent embryonic stem cells are just starting
00:56:27.12 to choose their ultimate differentiation fate.
00:56:30.06 We are hoping that we can capture a snapshot
00:56:32.22 of that moment from the perspective of the glycome.
00:56:35.20 And then finally, you know, one of our original motivations
00:56:38.27 was to develop these tools in a manner that
00:56:41.22 could be clinically useful. So we are hoping that
00:56:44.29 we can develop the chemical reagents
00:56:47.09 the azido-sugars in such a manner that you could
00:56:50.18 actually introduce these reagents into human subjects.
00:56:54.01 And use them to detect tumors at early stages
00:56:57.06 by virtue of their changing glycome.
00:57:00.01 So stay tuned, and hopefully we will have a lot more to say
00:57:03.02 about that in the future.
00:57:04.20 Finally I'd like to acknowledge the students
00:57:08.26 and postdoctoral fellows in my laboratory.
00:57:10.21 What you are seeing on this slide
00:57:12.17 is a snapshot of my laboratory at this particular moment in time
00:57:16.21 where I am showing you all of the students
00:57:19.02 and postdocs who are presently in the lab
00:57:20.25 although I have listed just a small list of alumni
00:57:24.10 whose work I alluded to during this lecture.
00:57:27.27 Most of their names were actually mentioned on the slides
00:57:30.19 along with reference to publications if you are interested in
00:57:33.10 digging into this in more gory detail.
00:57:36.07 But it should be said that all of the work that I presented
00:57:39.04 from my laboratory is in fact the hands-on research
00:57:43.10 of my students and postdoctoral fellows and I am
00:57:46.27 largely the mouthpiece that has the privilege of sharing this with the world.
00:57:50.29 So, thank you very much for tuning in, and I hope you enjoyed learning a little bit
00:57:55.04 about chemical glycobiology.

Videos in this Talk
  • Part 1: Chemical Glycobiology
    Part 1: Chemical Glycobiology
    Audience:
    • Student
    • Researcher
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
  • Part 2: Imaging the Glycome
    Part 2: Imaging the Glycome
    Audience:
    • Student
    • Researcher
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
Speaker: Carolyn Bertozzi
Total Duration: 1:45:44
Recorded: June 2009
Educator Resources for this Talk
All Talks in Biochemistry

Talk Overview

Part 1
A large part of an organism’s complexity is not encoded by its genome but results from post-translational modification. Glycosylation, or the addition of sugar molecules to a protein is an example of such a modification. These sugars, or glycans, are often complex, branched molecules specific to particular cells. Cell surface glycans determine human blood types, allow viral infections and play a key role in tissue inflammation.

Part 2
Since glycans cannot be labeled with genetically-encoded reporters such as GFP, bioorthoganal reactions have been developed to allow their labeling and imaging. In this lecture, Bertozzi describes the chemistry and imaging methodology used to view glycoproteins in cells and whole organisms.

Speaker Bio

Carolyn Bertozzi

Carolyn Bertozzi

Carolyn Bertozzi is a professor of Chemistry and Molecular and Cell Biology at the University of California, Berkeley, Director of the Molecular Foundry at the Lawrence Berkeley National Lab and a Howard Hughes Medical Institute Investigator. She received an A.B. degree in chemistry from Harvard and a Ph.D. from UC Berkeley. Her postdoctoral work was… Continue Reading

Playlist: Membranes and Organelles

Related Resources

Basic Reading

Varki, A; Cumming, R.D.; Esko, J.D.; Freeze, H.H.; Hart, G.W.; Etzler, M.E. (2009) Essentials of Glycobiology 2nd ed. Nelson, D.L.; Cox, M.M. (2005) Lehninger Principles of Biochemistry 4th ed. Chapter 7: Carbohydrates and Glycobiology p. 238.

Additional Reading
Sletten, E.M.; Bertozzi, C.R.; Bioorthogonal chemistry: fishing for selectivity in a sea of functionality. Angew Chem Int Ed Engl. 2009, 48, 6974-98.

Chang, P.V.; Prescher, J.A.; Hangauer, M.J.; Bertozzi, C.R.; Imaging cell surface glycans with bioorthogonal chemical reporters. J Am Chem Soc. 2007, 129, 8400-1.

Baskin, J.M.; Prescher, J.A., Laughlin, S.T.; Agard, N.J.; Chang, P.V.; Miller, I.A.; Lo, A.; Codelli, J.A.; Bertozzi, C.R.; Copper-free click chemistry for dynamic in vivo imaging. Proc Natl Acad Sci USA. 2007, 104, 16793-7.

Reader Interactions

Leave a Reply

Your email address will not be published. Required fields are marked *

iBiology and iBiology Courses are part of the Science Communication Lab (SCL). Our mission remains the same, to connect people to science. However, our focus has shifted to producing and evaluating cinematic films for education and the public, which you can find on the Science Communication Lab website. For more information, please see this blog post!