Carbohydrates II

Last time we finished discussing some of the derivatives monosaccharides can form and their chemical properties. Today we will start discussing how these compounds come together to form oligosaccharides and polysaccharides. We will also see different types of polysaccharides and their different functions.

Disaccharides

One of the reactions we left out the other day was the one between an hemiacetal or hemiketal with an alcohol. If we consider the reaction to be catalyzed by acid, we have:

This compound will be an acetal if the alcohol reacted with an hemiacetal, or a ketal if the aclohol reacted with an hemiketal. If the hemiacetal/hemiketal is a monosaccharide and the alcohol is another monosaccharide, what we get is the condensation of two monosaccharides to form a disaccharide, in this case known as maltose:

The new bond between the two sugars is known as the glycosidic bond. In the same way that the formation of an hemiacetal or hemiketal could give a or b anomers, the condensation of an hemiacetal or hemiketal with an alcohol can give us a or b linkages between the two monosaccharides in the disaccharide. However, this is not due to the free rotation of the aldehyde or ketone, but to the nature of the condensation reaction. The first step of the reaction is the loss of water on the formation of an oxonium (=O⁺-) species, which transforms the anomeric carbon into a flat sp² carbon, sort of like a carbonyl carbon:

Now the alcohol attacks this carbon because it is electropositive, to form the glycosidic bond. However, since it is flat, the attack can be from the top or from the bottom, which leads to a b or an a glycosidic linkage:

For two glucoses conected between the C1 of the first and the C4 of the other, and a b glycosidic bond, we get:

This disaccharide is known as cellobiose. Now we have to sit back and think of the following:

- First, just considering hexopyranoses, we have eight possible starting monosaccharides, and we can use any combination of the eight as the second monosaccharide. This gives us 8 x 8 = 64 possible disaccharides.

- Second, we use the anomeric (hemiacetal/hemiketal) carbon from the first sugar, but we can use any of the other five alcohols of the second sugar to make the disaccharide. This means, 8 x 8 x 5 = 320 possible dissaccharides.

- Finally, the glycosidic linkage can be either a or b, so we end up having 640 possible compouds using just paris from the starting eight disaccharides!!!

Now we look at the carbon atoms that are connected in the glycosidic bond. The one from the first saccharide will almost always be C1. We call the atoms on the first sugar 'primes' (that is, we put a ' to indicate this: C1'). The one on the second sugar can change. If it was cellobiose, the link will be between C1' and C4, so we add a (1T4): Now we have O-b-D-glucopyranosyl-(1T4).

For the second sugar, we do the same: We indicate the configuration of the anomeric carbon, the series (D or L), and if it is a furanose or pyranose. For a second D-glucose in a configuration, we end up with O-b-D-glucopyranosyl-(1T4)-a-D-glucopyranose. This is cellobiose, the molecule we saw above:

This can get pretty complicated if we look at polysaccharides. We use a shorthand notation, and assign a three letter code to each saccharide. The other thing we do is several assumptions:

-First, we assume that the sugars are all D, so we drop the 'D'.

-Second, we assume that all linkages (except when noted) are through and oxygem, so we drop the 'O'.

- Third, the pyranose form of aldohexoses predominates, so we drop that '-pyranosyl' and '-pyranose' terminations. Similarly, the furanose form of ketohexoses predominates, so we drop the '-furanosyl' and '-furanose' terminations.

- Since the configuration of the anomeric carbon of the second sugar (the one not forming a glycosidic bond) can metarotate, we drop that too.

Now, think of the number of possible dipeptides we can get using the 20 standard aminoacids: 20 x 20 = 400. Although we have a lot more monomers to pick from, we can only combine them one way, the peptide bond. In saccharides, we can use many combinations, which give us many different compouds from the same starting materials.

If you mentaly extend this, the number of trimers (three monomers) you can get using saccharides is unbeleivably bigger than the number of tripeptide we can get using the 20 standard amino acids.

So, considering this, and considering the monomers as letters in an alphabet that we can combine to make words, sugars will give us countless more possibilities to make 'words' and 'phrases'. Therefore, if you are a cell and want to communicate something to another cell (or you are a protein or membrane and want to communicate something to other protein or receptor), the choice will be to use sugars to encode you rmessage or signal. This is why sugars are so widespreadly used as 'markers' to pass information around cells.

Now, lets stop philosophaesing and get down to earth again. When we combine two sugars, the anomeric center of one of them will be blocked by formation of the glycosidic bond. Unless we used the hydroxyl from the hemiacetal or hemiketal of the other sugar (that is, the OH in C1') to form the glycosidic bond, the other sugar still has the possibility of opening into its linear form (hydrolysis of the hemiacetal/hemiketal):

This is important, because this end of the disaccharide can still be oxidized (i.e., can still reduce things), and it will be called the reducing end of the disaccaride. Bigger and longer polysaccharides will also have a reducing end (or several if they are branched, see below).

There is one more thing that we have to mention before we go into larger saccharide, and it has to do with the conformational mobility of disaccharides. After we form the glycosidic bond, we have two new bonds to worry about. One between the first sugar to the left and the oxygen, and the other one between the oxygen and the second sugar to the right. Both bonds are formed between sp³ centers, and are single bonds. They are therefore rotatable, and they make the glycosidic linkage extremely flexible.

The two dihedral angles related with the two bonds are called Phi and Psi (F and Y - rings a bell?). The first one is defined as the the dihedral formed between H1', C1', OX, and CX, being H1' and C1' the anomeric proton and anomeric carbon of the sugar to the left, and OX and CX the oxygen and carbon at position 'X' of the second sugar (the oxygen of the glycosidic bond 'belongs' to the sugar to the right...). In the case of maltose (O-a-D-Glcp-(1T4)-b-D-Glcp), we have:

Since we have single bonds, disaccharides (and polysaccharides in general) will be a lot more flexible than, for example, polypeptides. Despite this, in the same way that we can analyze the possible (or allowed) combinations of F and Y angles in a dipeptide by use of a plot of energy versus <F,Y> (the Ramachandran plot) we can use a similar plot to analyze the possible conformations a glycosidic bond can take, their energies, and the likelyhood of having those combinations of dihedral angles. Again, if we look at maltose, such a plot will look like this:

Again, we use the -180 to 180 convention for the angles as we did for peptides. Here, the blueish, violet, and the red inside the violet are high energy contours, and the greenish, yellow, orange, and red inside the orange are low energy contorus. This means that the likely conformers of maltose will be around <F = 5, Y = -10>, because this cobmination of angles gives the lowest energy conformers for the disaccharide.

These plots are, however, not as useful as the Ramachandran plots for polypeptides, because we can have way to many combinations of monosaccharides forming disaccharides, and it is therefore not very practical to have books stuffed with these data. In polypeptides we also had a lot less possible combinations (those that gave a-helixes and b-sheets, basically). In any event, people like me puttzes around with these, because we still think that something good can come of them...

Common disaccharides

Now that we have the tools, lets look at some common disaccharides. The one that everyone knows about is sucrose, which is composed of glucose and fructose, linked together through C1' and C2. In this case, the linakge is between both anomeric carbons, so sucrose has no reducing end, and won't mutarotate. C1' is in a configuration, and C2 is in b configuration:

Note that here fructose is flipped around in the drawing. The full name for sucrose is O-a-D-glucopyranosyl-(1T2)-b-D-fructofuranose, or a-Glc-(1T2)-Fru. Sucrose is an important source of glucose (D-glucose), mainly because it is readily converted to glucose by the action of enzymes called glucosidases, which cleave the (1T2) bond between the two units. As we will see below, there is pretty much one glucosidase for every particular

Another important disaccharide is lactose, which is found in large concentrations in the milk of mammals, and is formed by linking galactose and glucose in a b 1T4 linkage:

Using all the bells and whistles, lactose is named O-b-D-galactopyranosyl-(1T4)-b-D-glucopyranoside, or b-Gal-(1T4)-Glu. Lactose may ring a bell for those of you who may suffer from lactose intolerance. The problem of lactose intolerance arises from our continued use of dairy products after early infancy. In newborn mammals have enzymes that process lactose. A b-D-galactosidase, which chops up the b-linkage between Gal and Glu, giving galactose and glucose, and an epimerase, which epimerizes the C4 of galactose to give glucose. Therefore, we get two glucoses that we use as fuel.

In those who suffer from lactose intolearance, the level of the b-D-galactosidase is very low after they are supposed to stop feeding from their mother's milk. If they keep ingesting diary products after this, the lactose cannot be processed into glucose, and it is not absorbed. It continues through the digestive tract, and it is fermented by bacteria thar live in our gut, producing acid and gas - Lots of gas. So it hurts you, and you are certainly not to become a crowd pleaser.

For infants, it can be far worst than a digestive disturbance, because they cannot get the glucose from their mother's milk, and they need to take supplements for this, or actually milk treated with b-D-galactosidase.

Polysaccharides and proteoglycans

Common to all polysaccharides is the definition of homopolysaccharide and heteropolysaccharide. In homopolysaccharides, all the sugar monomers are identical. For example, all glucose. In heteropolysaccharides, we can have a variety of sugar monomers forming the polysaccharide In both cases, we can have unbranched (or linear) polysaccharides, or branched polysaccharides. Assuming all aldohexhoses:

Although they are polymers and have some of the same propertties than polypeptides, the synthesis of polysaccharides is not encoded as the synthesis of preoteins - In proteins we have a DNA/RNA template that gives us always the same proteins. In polysaccharides, each unit elongating the chain is attached by a different enzyme. Therefore, the number of monomers in a polysaccharide may vary from molecule to molecule, depending on the stage at which we are in the synthesis, as well as how active the enzymes making them are, etc., etc. Thus, a sample of cellulose, for example, will have thousands of molecules of different lenghts, and are pretty hard to analyze by analytical and structural methods.

Now, that we saw all the definitions and properties pertaining disaccharides and polysaccharides, we will move into some classes of polysaccharides. We will divide our disucssion into two main group, the storage polysaccharides (those used for energy), and the structural polysaccharides.

Storage polysaccharides

In most living organisms, glucose is stored as a long/large ploymers instead as monomers. Why? Osmotic pressure is proportional to the number of solute molecules, so you would get 1000 times the osmotic pressure if you stored 1000 glucose monomers than if you stored 1 polymer of glucose with 1000 units. As we saw before, osmotic pressure is a really bad thing to have in cells, unless you are a tree.

Second, we would have a large concentration of glucose inside the cell, and many of the glucose pumps that work using the glucose chemical gradient or passive transport would stop altogether.

In plants, the main form of storage of glucose is starch, which is composed of two types of glucose homopolymers, amylose and amylopectin. It is found in large quantities in potatoes and other tubers. Amylose is an un-branched polysaccharide formed exclusivelly by D-glucose, in which all the monomers are linked through a-(1T4) glycosidic bonds, forming chains that are basically ...a-D-Glu-(1T4)-a-D-Glu-(1T4)-a-D-Glu-(1T4)-a-D-Glu... :

An amylose strand can have anywhere from 1000 to 500,000 D-glucose units.

Amylopectin is basically amylose, but it has branching points, so it is a branched homopolysaccharide. The branching is at the C6 hydroxyl of some of the saccharide units (a 1T6 glycosidic bond), with branches coming out from every strand every 25 to 30 monomers:

In animals, glucose is stored as glycogen, which is basically the same as amylopectin (that is, 1-4 chains with 1-6 branches), but it branches out more than amylopectin, with chains coming out every 10 to 12 residues.

Amylose, amylopectin, and glycogen have a single reducing end. This is because all the chains and braches finish in a sugar whose anomeric carbon is involved in the glycosidic bond. So how are these things used to make glucose? There are enzymes in the cell, the a-D-glucosidases or a-amylases, that upon request (a signal of some sort), will start chomping out D-glucose units at all the a-glycosidic bonds. BTW, you have these in your saliva, and that's why when you chew on a piece of bread for a while it starts tasting sweet.

Amylases act on the non-reducing ends of the chains. Thus, in amylose, the degradation is linear, in amylopectin, several ends can be chewed up at the same time, and in glycogen, even more degradations can take place simultaneously by picking different branches from the molecule. This makes sense: A tree does not 'springs' into action fast, so it won't need large concentrations of glucose for short burst of energy. A person, on the other hand, needs to convert glycogen into glucose fast, and since there are more branches on which glucosidases can act simultaneously, more glucose can be generated at the same time.

The a glycosidic bond of starch and glycogen has important consequences on the three dimenssional structure these polymers adopt in solution. The preffered conformation of the glycosidic bonds is F ~ 0 and Y ~ 0. Therefore, since we have all-a anomers, the sugars will tend to curl up into a helical structure:

A consequence of this 3D structure is the amount of water that surrounds these glucose-generating molecules. Since they are curled up, most of the hydroxyls from the monomers will be exposed to solvent. This means that water will interact strongly with them, and solvate this molecules. This makes them soluble in water (even when they are huge) and increases their relative volume and size. Since they are soluble in water, they are easier to degrade by the amylases.

Structural polysaccharides

The other big class of homopolysaccharides are the structural polysaccharides cellulose and chitin. Cellulose is the component of the cell wall of plant cells, and forms part of all the fibrous materials from plants. Its structure looks a lot like starch, except that all the glycosidic bonds are b, and there are no 1T6 branches:

As oposed to amylose, cellulose forms elongated fibers that stretch out, not curl. It therefore forms a large network of h-bonds that stabilize the strand itslef, as well as clusters of strands that make up fibers. This gives a cellulose fiber several important structural properties: First, it is tough as nails (have you ever chewed wood lately?). Second, it is water impermeable, because as opposed to amylose, amylopectin, and glycogen, there are numerous h-bonds formed within the cellulose strands, and they thus exclude water:

Cellulose is pretty cool. You are at this moment wearing clothes made of cellulose, writing one paper, made of cellulose, living in a house that has insulation, most likely made of cellulose. So why, if it is so widespread, can't we get our glucose from cellulose? It would be pretty cool - You are hungry and you chew on some cardboard box. The reason is the type of glycosidic bond we have in cellulose. It is a b glycosidic bond, and most animals don't have b-D-glucosidases (b-amylases).

What about cows and termites? They don't have the enzyme either, but a) they chew the stuff forever, breaking it into very small pieces, and b) they have bacteria in their gut (and rumen in cows) that has a b-D-glucosidase, cellulase. The enzyme from this simbiotic bacteria breaks down the cellulose, and the cows get their sugar that way.

The second structural homopolysaccharide we will study is Chitin. It is the main component of the exoskeleton of bugs and crustaceans. It is is 99.95% the same as cellulose, but instead of having OH's in C2, it has a acetamido group (NH-COCH₃). This makes them even more impermeable.

Next time we will discuss cell wall polysaccharides, glycolipids, glycoproteins, and wrap up with carbohydrates.

Prepared by Guillermo Moyna, 1999.