Carbohydrates

Again, we change the type of molecules under study pretty drastically. So far, we have seen amino acids (peptides and proteins) and lipids (fatty acids, phospholipids, and membranes). Now we will start analyzing the most diverse family of biomolecules, the carbohydrates, sugars, or saccharides.

Until the early 60's, carbohydrates where thought to only serve for energetical and structural purposes. However, it was later found that carbohydrates play a very important role in molecular recognition. We can summarize the functions of carbohydrates as follows:

- Energy storage. Oxidation of saccharides to CO₂ and H₂O is the central energy yelding process in nonphosynthetic organisms.

- Structural functions. Polymers of saccharides, polysaccharides, are used to provide structural strenght to cells. The best example is the cell wall of plant cells, formed by the polysaccharide cellulose, which coats the cell and makes it rigid. There are many other polysaccharides that provide dynamic skeletons to algea and bacteria, many of which are used by humans for the same purpose (gum arabic, xantan gum, carrageenans, etc.). Other polysaccharides are used as lubricants in joints and as adhesives between cells forming colonies.

- Molecular recognition and signaling. Saccharides associated with membrane lipids or proteins (glycolipids and glycoproteins, or glycoconjugates in general) are involved in siganling processes, are involved in cell-cell interactions, and inmunological responses. Saccharides covalently attached to proteins determine their location or destination in the cell and the metabolic routes they are supposed to follow.

Saccharides are polyhydroxy aldehydes or ketones, or simple derivatives of these compounds. The basic ones are characterized as 'carbon hydrates', in which we have an empirical formula C_n(H₂O)_n - This means, we have one water molecule per carbon atom. As we will see, some carbohydrates have nitrogen replacing oxygen, as well as sulfur, in the form of sulfates (-O-SO₃^-).

Carbohydrates can be found as single units, or monosaccharides, as short polymers composed of several monosaccharides, or oligosaccharides, or as polysaccharides. The only difference between the last two is the number of monosaccharides. The name polysaccharide is reserved for carbohydrate polymers with more than 20 monosaccharides. When we have two or more monosaccharides linked together, the covalent linkage is usually called a glycosidic bond, and as we will see later, it determines both the identity, properties, and mobility of the polymer.

Monosaccharides

We will start our discussion with monosaccharides of three to seven carbons. As we said above, monosaccharides are polyalcohols that have an aldehyde or ketone. Therefore, we have two flavors: aldoses or ketoses. Here we see another common naming convention in carbohydrates: generally, monosaccharides (and disaccharides) are named using the '-ose' termination. The simples monosaccharide has three carbon, and as we said comes in aldose and ketose versions:

In this case, we don't use the '-ose' nomenclature - We just call them D-glycerladehyde (an aldose) and dihydroxyacetone (a ketose). Which enzyme could convert one into the other? Another thing that we note here is that (except for dihydroxyacetone) we will have stereocenters. Thus, glyceraldehyde can be D- or L-glyceraldehyde. Remember that this molecule was used to determine the configuration of the Ca of amino acids:

We use the Fisher notation (which we learned already) to represent extended linear sugars easily. Since we are trigger-happy with conventions, here goes another one: Carbon 1 (C1) is the one on top. We will start with the aldose series, in which C1 is an aldehyde. By adding a 'C' and a 'H₂O' to glyceraldehyde, an aldotriose, we get aldotetroses:

Since we added another stereocenter, we can now have diastereomers. We have to remember the differences between diastereomers, enantiomers, and epimers. Whenever we have more than one stereocenter, we can have several diastereomers of the compound. Diastereomers have different configurations in their stereocenters, and may or may not be enantiomers. Enantiomers, irrespective of the number of chiral centers, have to be specular images of each other. An epimer, on the other hand, is a compound in which we invert just one of the chiral centers. For example, we get D-threose by epimerization of the C2 stereocenter of D-erythrose.

Now, what about the D- and L- configurations? If you saw above, we changed the configuration of C2 from D-erythrose to D-threose, but we kept the 'D-'. We use the chiral carbon further away from the carbonyl carbon (either the aldehyde or the ketone) as the reference, and we compare its configuration to that of glycerladehyde. If the bottom part matches D-glycerladehyde, it is a D-aldose, and if it matches L-glyceraldehyde, we have an L-aldose.

If we consider all the possible stereoisomers (or diastereomers) we can have in tetroses, we can have 2², meaning 4. Two will correspond to D-aldoses, and two to L-aldoses. In this case, we will have D-erythrose, D-threose, L-erythrose, and L-threose.

Now, if we add one more 'C' and 'H₂O' units to the aldotetroses, we get the aldopentoses:

Now we have 2³, or 8 possible distereomers. Four correspond to the D-series, and four to the L-series. Finally, if we add one more 'H-C-OH' group to the mix, we get the aldohexoses:

Here we can have 2⁴, or 16 diastereomers, 8 of which will be D-series. and 8 of which will be L-series.

If we hade added the 'H-C-OH' fragment to dihydroxyacetone instead of D-glycerladelhyde, we would have gotten the ketoses. Since we have one stereocenter less in these ones, we will have half of the possible distereomers in each series. So we have the ketotetroses and ketopentoses:

And if we add one more carbon, we get the ketohexoses:

As for naming all this stuff, there are some rules, but they are broken all the time. To name a ketose, we usually add the '-ul' extension to the name of the corresponding aldose. So ribulose in the ketose series corresponds with ribose in the aldose series. However, there are expensive books (like the one you bought), tables, the Web, and other places to go to find the names. I will require you to know only the most common ones, like D-glucose, D-mannose, D-galacotse, D-ribose, and D-fructose...

Cyclic sugars

As you probably noticed, I many times write the sugars as cyclic structures. This is because this is the natural form in which sugars exists. Until now we have drawn them as extended, linear molecules. However, we have an aldehyde (or ketone) and several alcohols lying around very close to the carbonyl center. From organic chemistry, we know that alcohols can react with aldehydes and ketones to form hemiacetals and hemiketals, respectivelly:

Depending on the sugar we have (aldose or ketose) and the alcohol involved in the formation of the hemiacetal/hemiketal, we will get cyclic structures of different sizes. If you remember from organic chemistry, cyclic strcutures with six or five atoms in the ring are more stable than, say, cyclic structures with seven of eight atoms. Thus, we get mainly six or five-membered rings. How do we represent this? We can use the good'ol Fisher projection, and make a loooooooong bond. For glucose going to its six- and five-membered cyclic forms, we would get:

Yaky! Several problems. First, it is hard to imagine an electron traveling half a mile to get were it has to go. Then, we have to add long lines, with 'corners' that do not represent atoms, which is a really bad thing to do if you are not used to the notation. It is a lot better to use some sort of projection to see what the heck we are doing. These drawings are called Haworth projections, and make life a loooot easier:

However, we have to remember that in these systems all the carbons and oxygens are sp³ centers, and therefore the ring is not planar as in the haworth projection. Cyclic structures with sp³ centers will adopt chair or boat conformations. In most cases, the chair conformations are more stable than the boat conformation due to releif of sterica clashes. Now we can see clearly the formation of the two possible ring structures of glucose:

In all these projections we put the oxygen from the ring looking back, and the carbon of the hemiacetal/hemiketal looking to the right. Now, we have to consider some other things before we go any further. First is the names that we assign to rings of different sizes. We name the cyclic sugars by comparing them to simple cyclic ethers, furan and pyran:

Therefore, sugars with five-membered rings will be called furanoses (D-glucofuranose, D-mannofuranose, etc.) and those with six-membered rings pyranoses (D-glucopyranose, D-mannopyranose, etc.). .

The second thing we overlooked is that when we form the hemiacetal/hemiketal, we form a new chiral center, which is called the anomeric center. We call these two diastereomers anomers. We can have glucopyranose with the resulting hydroxyl from the hemiacetal 'up' or 'down'. These are two different compounds, with different physical properties. Depending on where the hydroxyl ends up we will have an a-anomer (down) or a b-anomer (up). The two forms of D-glucopyranose are therefore a-D-glucopyranose and b-D-glucopyranose.

In aqueous solution, there is an equilibrium between the two anomers, because we have a continuous opening and closing of the ring. The ratios are 1/3 a-D-glucopyranose to 2/3 b-D-glucopyranose, and almost no linear forms:

This process of equilibrating the ratios of a and b anomers is known as mutarotation. The final a/b anomeric ratio will depend on the environment, namelly, the solvent, monosaccharide concentration, and temperature. Since most of the sugars are present in solution as cyclic molecules, you have to keep this in mind throughout the study of ploysaccharides.

Hexose derivatives

Most of the times, the sugars that we see in biochemistry are slightly different from what we have seen so far. Sugars are many times modifies by enzymes before they go on to do what they ought to do. In most cases this involves the capping of an hydroxyl by certain chemical group, most notably phosphate and sulfate, the change of one of the hydroxyl for a nitrogen, usually the C2 hydroxyl for an amine or acetoamido group, or the oxidation of one of the aldehydes or alcohols to an acid:

The oxidation can be of the C1 aldehyde or of the C6 alcohol. In the first case, we get aldonic acids. Instead of forming hemiacetals, this molecules form very stable lactones:. For D-glucose (or D-gluconic acid) we have formation of D-glucono-d-lactone:

In case of oxidation of the C6 alcohol, we get uronic acids, which for the normal hemiacetal or hemiketals and the anomeric center. For D-glucose, this would be D-glucoronic acid:

Many times the formation of these derivatives is required for the formation of a particular covalent bond with a protein or lipid. In other cases, the precense of phosphate of sulfate groups affects the solubility or ability of the sugar to travel though membranes.

Two other important derivatives of aldoses and ketoses are reduced and the deoxygented derivatives. In the first ones, the carbonyl from the aldehyde or ketone is reduced to an alcohol, to give polyhydroxylated alcohols known as alditols. We saw one of them, inositol, which was a component of certain shpingolipids. Inosoitol can be considered as a reduced homolog (one carbon less) of glucose:

The deoxysugars are also very important. In these, one of the hydroxyl groups is completely reduced to '-H':

The best example is deoxyribose, which is an important component of deoxyribonucleic acid (DNA)

Reducing properties

In aldoses we have an aldehyde in one end. As such, the aldehyde behaves as a reducing agent. In the precense of mild oxidizing agents, such as Cu²⁺ or Fe³⁺, we get an oxidation of the aldehyde to a carboxilic acid, and the sugars are called reducing sugars. This reaction was used for quantitating the amounts of glucose in blood in patients with diabetes:

Since cyclic sugars are continuously in equilibrium with their open forms, even pyranoses can act as reducing agents. Furthermore, long chains of polysaccharides will usually have a reducing end.

Next time we will start studying, disaccharides, larger olygosaccharides, and polysaccharides.

Prepared by Guillermo Moyna, 1999.