Capillary electrophoresis (CE) is an analytical separation method. Being an analytical method, it uses very low amounts of sample. One would therefore not normally use CE if one was trying to purify a significant amount of a particular compound for later use. You use it to determine what is in a sample.
Like all other electrophoretic techniques, CE separates molecules in the presence of an electric field. This separation is done inside a capillary. The capillary is a long and very thin tube made of fused silica (glass) and coated on the outside with polyimide (for strength). There are several different sized capillaries commercially available. Capillaries with internal diameters of 10 to 100 micrometers are commonly used. Lengths commonly vary from 20 to 100 cm. Molecules separate based on their charge-to-mass ratio (or more correctly their charge-to-frictional coefficient ratio where frictional coefficient is often approximated to molecular size which is further approximated to molecular mass). There are two basic types of CE. CE where the capillary is filled with solution containing a gel (sieving matrix) and CE where the capillary is filled with a solution not containing a gel.
No gel: A capillary is filled with a conducting solution, usually an aqueous buffer. Both ends of the capillary are placed in their own vial containing buffer. One of the buffer vials is gounded (detection end of the capillary) and the other is hooked up to a high voltage power supply (injection end of the capillary). When you turn on the voltage current is carried through solution in the capillary from the power supply to the ground.
Being made of fused silica, the surface of the capillary is covered with exposed silanol groups. Since the inside of the capillary is filled with buffer, the silanol groups on the inner surface the can loose protons, giving them a negative charge. It is important to note that these charges are attached to the capillary wall. They do not migrate. Negative charges are countered by positive charges. These positive ions are components of the buffer solution within the capillary. These cations are not fixed in place and can migrate. When a positive voltage is placed on the injection end of the capillary these cations will migrate towards the detection end (which is grounded = negative relative to the injection end). Note, the negative charges on the capillary surface are immobile and will not migrate. As the cations migrate they will pull the buffer solution with them. Therefore there is a net flow of buffer from the positive to the negative end of the capillary in the presence of an electric field. This is called electroosmotic flow (EOF).
What happens when you inject a sample into the capillary? All molecules in the sample will have a tendency to move towards the negative (detector) end of the capillary due to EOF. Neutral molecules will move at the same rate as EOF. However, negative ions will also have a tendency to migrate back towards the positive (injector) end. This is because opposite charges attract (an of course like charges repel). This is referred to as electrophoretic mobility. Therefore with respect to anions, EOF will pull them towards the detection end and electrophoretic mobility will pull them back towards the injector end. Who wins? The stronger one. EOF is usually (unless you chemically treat your capillary or use a low pH to reduce EOF- something which we have not discussed) stronger than electrophoretic mobility so the anions have a net mobility towards the detector. However, the net mobility will be slower than EOF since electrophoretic mobility is taking away from it. Thus, anions will move slower than neutral molecules and reach the detector later. It is the opposite for cations. Again, they will move due to EOF and electrophoretic mobility. However, both these mobilities are towards the detector. Therefore net mobility will be greater than EOF and cations will reach the detection end of the capillary first. Note: in CE you can separate and detect cations, anions and neutrals in a single run.
In the end what allows you to separate molecules by CE is differences in their electrophoretic mobilities. So what determines electrophoretic mobility? Charge-to-mass ratio (or more correctly charge-to-frictional coefficient ratio). The higher the ratio the larger the electrophoretic mobility. Since different ions have different charge-to-mass rations they can be separated due to differences in their electrophoretic mobilities. In a CE run with an untreated capillary and the injection end positive, the order at which molecules pass the detector (from first to last) is: high charge-to-mass ratio cations, low high charge-to-mass ratio cations, all neutral species regardless of mass, low high charge-to-mass ratio anions, high charge-to-mass ratio cations.
Up until know it appears that neutral species will all migrate at the same speed and therefore cannot be separated by CE. This is not completely true. Under the conditions described thus far they cannot be separated but there are many tricks one can use. One can add charged things to the buffer that form complexes with the neutral molecules you want to separate. These complexes will be charged and therefore one will be able to separate them. An example of this is adding borate to sugars. Borate will complex with the sugars forming anionic complexes which will have different charge-to-mass ratios and can be separated by CE.
Another trick is to use detergents such as sodium dodecyl sulfate (SDS). SDS is anionic. Above a concentration of about 5-10 mM SDS will self-assemble into micelles. These micelles will be anionic and therefore will move at a slower rate than EOF. Neutral molecules will partition between these micelles and the surrounding buffer. Neutral molecules that spend more time in the micelles will move slower than neutral molecules that spend less time in them. This is analogous to chromatography where the micelles are a pseudo-stationary phase and the surrounding buffer is the mobile phase. In fact, this method is called micellar electrokinetic capillary chromatography (MECC).
CE can be done using a sieving matrix in the capillary. A commonly used sieving matrix is polyacrylamide. A sieving matrix is an entanglement of polymers. This it like a molecular fishnet. Molecules migrating down the capillary must squeeze through the spaces between the entangled polymer molecules. Larger molecules have a more difficult time squeezing through these spaces and thus are slowed down. The use of a sieving matrix adds a separation-by-size component to the method. When separating different molecules that all have the same charge-to- mass (ie different sized DNA fragments) or molecules that are treated to induce them all to have near identical charge-to-mass ratios (ie completed complexation of different proteins with SDS) one can separate molecules approximately based on their size alone. Note: separation by size using CE requires all the molecules to have identical or near identical charge-to-mass ratios. If not, one is separating them based on both size and charge-to-mass.
After separating molecules one must detect them. There are 2 detection methods commonly used in CE: absorbance and fluorescence. When using absorbance detection the capillary is stripped of its polyimide coating forming a small ‘window' near the detection end. Molecules passing by this window are detected by their absorbance of a beam of light passing through this window. Signal obtained using absorbance is proportional to the pathlength, which in this case is the internal diameter of the capillary. This is commonly from 10 to 100 micrometers. Due to this short pathlength signal and sensitivity are relatively low. Fluorescence detection is much more sensitive however few molecules fluoresce. As molecules pass the window they are excited by a beam of light, often from a laser, and their fluorescence detected at right angles. Although pathlength is still short, this can be compensated for by using higher excitation power. Fluorescence sensitivity is usually far superior to absorbance.
The fluorescence detector system described has one major weak point. One must shine the laser through the capillary window. The capillary surface is not flat nor is it made from polished high quality quartz. As the laser beam enters and exits the capillary there will be a large amount of scatter. Since excitation and emission light is of different colors it is possible to filter out the scattered excitation light from the emission light assuming a sufficiently large Stoke's shift (difference in the wavelength of the excitation and emission light). However, you will not be able to filter all of the scatter and will get an increased background due to this. Background noise increases with background signal. Sensitivity decreases with background noise. Therefore sensitivity will suffer when you have a high background signal. The trick is to reduce the background by reducing the scatter. You do this by not passing the laser through the capillary.
A polished square surface cuvette is constructed from high quality quartz. This cuvette contains a square hole running through it. The detection end of the capillary is placed in this hole. The hole is wider than the capillary and there is room for a buffer to flow through the cuvette. The laser beam passes through the cuvette just below the tip of the capillary. This avoids scatter caused by passing the beam through the capillary. As molecules exit the capillary they are swept up by the flowing buffer and hydrodynamically focused into a cone. The laser shines through this cone and fluorecsence is collected at right angles. This detector is exquisitely sensitive, capable of detecting as little as 1 molecule of fluorescent dye. This type of detector is not currently commercially available. If you want one you have to build one yourself. It is well worth your effort. It just so happens I have one in my laboratory (a sales pitch for potential graduate students).
How does one get the sample into the capillary? There are 2 basic methods: Pressure injection and electrokinetic injection. In pressure injection positive pressure is applied for a given period of time to push sample into the capillary. Alternatively, negative pressure can be used to suck sample in. In electrokinetic injection, sample is electrophoresed into the capillary using a given electric field for a given period of time.
Why would you want to use a separation method such as CE to assay enzymes?