Abstract
Capillary Electrophoresis (CE) is a family of related techniques that use narrow-bore capillaries to perform high efficiency separations of both large and small molecules. The new technology of CE applies existing theory with new methodology. For example, taking the technique for slab gel separations of proteins and applying it to CE produces precise, quantitative separations. The CE produces very predictable and reproducible results with minimal sample sizes. The theory of CE, three separation methods, and results of these separations are presented here.
Introduction into Capillary Electrophoresis
Capillary Electrophoresis (CE) is the newest member into the family of chromatography. Chromatography permits scientists to separate diverse mixtures of closely related compounds. CE harnesses the power of electophoretic and electroosmotic flow in order to produce separations of samples that consist of only a few nanoliters. The CE machine has a very simple Shematic.
The capillary and the electrodes of the variable power supply (0 to 30 kV) both meet in the buffer vials. The detector is located on the outlet side of the capillary (depicted here on the right side). The capillary then acts as a salt bridge between the inlet and the outlet vials, continuing the electrical loop. The inner walls of the capillary are lined with a negative charge. This is achieved by the dissociation of functional groups making up the fused-silica surface.1 The walls of the capillay can have any pH above 3. The movement of the buffer (composed of cations) will create flow along the capillary. This flow is known as the electroosmotic flow (EOF). Electrophoretic flow (EPF) is created when the voltage is applied. The anions are pulled towards the anode, and conversely, the cations are pulled towards the cathode. The analyte velocity is determined by the effects of the electroosmotic and electrophoretic flows. UEEO= Electroosmotic Velocity UEP= Electrophoretic Velocity
For Anions: UT= UEEO-UEP Cations: UT= UEEO+UEP Neutrals: UT= UEEO
In the electropherogram (absorbance vs. time), the cation peaks will be followed by a single peak for the neutral molecules, and the followed by the anion peaks. Notice that for all of the neutral molecules a single peak is observed. This is because the velocity for all of the neutrals is equal the electroosmotic flow. A specific technique for the separations of neutral molecules is detailed in following sections. The electroosmotic flow is stronger than the electrophoretic flow. For neutral molecules the analyte velocity is equal to the electroosmotic flow. The electroosmotic flow also creates a type of flow that is comparatively better that the flow found on HPLC. In CE there is a plug flow, while in HPLC the solvent line has a concave flow. The plug flow in CE reduces peak broadening, and exhibits higher efficiencies than HPLC. CE uses a UV/Vis lamp. The Beckmann instrument is equipped with a series of filters that produces the desired wavelength, selection of the filter depends the species being studied. This instrument can hold up to eight different filters. The Beckman instrument is capable of two different types of detection. The most common method of detection is direct detection, the other being indirect detection. In direct detection, the buffer is transparent at the wavelength selected. For example, when detecting an analyte that is conjugated, the preferred wavelength would be 254 cm-1. The results will be tainted if the buffer absorbs at this wavelength. Indirect detection is directly opposite. In indirect detection the buffer is absorbing the light and the analyte is transparent. This produces a negative peak in the electropherogram. The computer software actually reverses this process to correspond to the direct detection method.
Acid Analysis
The simplest of the three different methods is the acid analysis. This is the simplest because the capillary and buffers are low matinance. This method was used to seperate similar acids. The test mixture is composed of benzoic acid, p-hydroxybenzoic acid, and p-hydroxyphenylacetic acid. These acids are very similar in structure (see attached structures). These similarities make them hard to distinguish. Using CE though, separation is simple. This separation uses a borate buffer with at pH 8.35. The capillary is a regular silica capillary. The capillary is very durable and the results are easily reproducible. The capillary can be regenerated using 1 molar hydrchloric acid, followed by a rinse with deoinized water. The table summerizes the materials needed and the position of the vials on the sample tray. The hydrochloric acid and water are used in the prerinse. The empty vial is used to collect the waste materials during the prerinse. The buffer is used during the seperation and must have vials on the inlet and the outlet sides. The procedure is similar in all cases, usually containing three parts. The first is the prerinse. This is used to clean and regenerate the column. Next, the test mixture is injected in to the capillary. Finally, the voltage is applied and the separation occurs. In this specific case a 2 minute rinse with the hydrocloric acid is used to clean the column. This is then followed by a 2 minute rinse with the deionized water is used to flush any excess acid. A pressure injection is needed to obtain a sample. This injection is 5 seconds. The separation time will depend on the length of the capillary. This analysis used a 27 cm capillary that required a seperation time of 5 min. The voltage should be applied to the full amount as quick as possible to achieve reproducible results. In the method, the voltage was applied in a matter of 0.17 min. This is the fastest setting without overloading the capillary. The source filter is fixed to produce a wavlength of 214 cm-1. The difference of time for the separation of the test acids is only 1 minute. The first to reach the detector is the p-hydroxyphenylacetic acid. The retention time was 4.00 minutes. This is followed by p-hydroxy benzoic acid with a retention time of 4.33 minutes, and benzoic acid at 4.50 minutes (see attached electropherogram). The next analysis might seem unorthdox. Samples of diet sodas were prepared for separation using the same methodology. Preparation of the sodas was trivial. The only necessary preparation needed was to degas the soda. Air bubbles in the capillary will cause a break in the electrical loop. This will cause a current error. It is essential to use diet sodas because the sugars in regular sodas will solidify when the voltage is applied. The components of the sodas of interest are caffeine, citric acid, and phosphoric acid. Again the exact same method is used as outlined for the test mix analysis. The only difference is that the separation time is now 10 minutes instead of 5 minutes. The sodas that were analysized are Diet Coke, Diet Caffeine Free Coke, Diet Sprite, and Diet Big Red. The diet sodas (except Diet Caffeine Free Coke) produced the same electropherograms. This raises an interesting question, what is the retention time of caffeine? The electropherogram of Diet Coke shows large peaks at 2.25, 2.83. and 3.73 minutes. Inspection of the Diet Caffeine Free Coke shows peaks at 2.70 and 3.55 minutes. The only difference in the two samples is caffeine, therefore the retention time of caffeine is 2.25 minutes. The other two peaks belong to citric acid and phosphoric acid. The samples were run numerous times to check for reproducibility. Two electropherograms of Diet Caffeine Free Coke as well as one electropherogram of Diet Coke follow. The citric acid peaks in the two Diet Caffeine Free Cokes differ by .09 min (2.79 and 2.70 minutes). The difference in the phophoric acid peaks is .17 min. (3.72 and 3.55 minutes). The results are very reproducible. This type of analysis is extremely reliable for multiple separations. This capillary is very durable. The soda was not treated in any way. The capillary and buffer does not need any special treatment. This is one of the characteristics that makes this method very simple.
SDS Gel Seperations
The SDS gel technique is comparable to slab gel techniques. The CE method is much faster though. A typical slab gel separation takes many days to complete. After staining, the gel the separation must be determined by human eyes. The results can be easily misread and misinterpreted. On the other hand. the CE takes anywhere from 20 minutes to an hour. The results are then analysized by a computer. This produces faster and more accurate results. This method, compared to the acid analysis, is very similar, while the buffers and the capillary are more sensitive and require more care in storage and handling. The capillary and buffer must be stored at 20oC. The capillary must also be stored with the exposed ends submerged in the buffer. Outside of the buffer, the capillary will dry within ten minutes when exposed to the air. The buffer must also be de-gased before use. Proteins are naturally neutral molecules. The theory section stated that neutral molecules could not be separated. The gel buffer in effect gives the molecules a sense of charge. This enables them to be separated. The buffer is composed of micells. The micells have a negatively charged head and a hydrophobic tail. The hydrophobic cores of the micells gather around the protein with the polar exterior (or head) extending outwards from the portein. This will provide the protein with a charge to mass ratio. This is the basis of the separation. The lager the protein the longer the retention time. This method will result in the determination of protein molecular weights. In order to do so, a test mix of known molecular weights is needed to obtain a standard curve. For the best results, the approximated molecular weight of the protein being studied should be similar to the known samples. In addition to the known samples, a reference standard must also be used. The reference standard is Orange G. This will be used to calculate a realtive migration time,or RMT. The RMT is equal to the migration time of the protein divided by the migration time of the reference standard (Orange G). A plot, log of molecular weight vs. 1/RMT, will produce a standard curve. The slope function will give rise to an equation used for determining the unknown molecular weight. Once the migration time of the unknown protein is known a simple calculation is all that is needed to determine the molecular weight of the protein. The sample preparation time for this method is longer than the acid analysis method. The protein must be prepared so that it has a linear structure. The disulfide bonds must be broken and the protein must be denatured. In a 400 micro-L vial the following ingredients must be added: - 0.1 to 1 mg protein mix (or unknown sample) - 100 micro-L Sample Buffer - 10 micro-L Orange G - 5 micro-L 2-mercaptoethanol (to cleave disulfide bonds) - 85 micro-L deionized water filtered through 0.2 micro-m filter This mixture should then be stirred on a vortex mixer for 2 minutes or until the proteins are dissolved. Next, the mixture is placed into a water bath at 100oC for ten minutes. Finally the mixture is cooled in an ice bath for 3 minutes. Five vials are needed for this method. Three of the vials are filled with the SDS gel buffer. One vial is used for the sample mixture and the last is used for 1 M HCl. The sample mix is placed inside of a 400 micro-L sample vial. The injection only draws nanoliters of the sample. Because of the small amounts of test mix prepared, the micro-vial limits the amount of test mix wasted. Two of the gel buffer vials are used in the separation. The third vial of gel buffer is used for a prerinse to ensure that the inside of the capillary is entirely coated with gel. The 1 M HCl is used for a prerinse to clean out the capillary again. This method will consist of two prerinses, injection, and a separation. This method uses a 27 cm capillary. A longer capillary will require longer times than seen here. The first rinse is with the HCl for 1 minute, followed by a 3 minute rinse with the gel buffer. For a 47 cm capillary these times should be 2 minutes and 5 minutes respectively. The injection time is a range from 1 to 99 seconds, an average time is usually 30 seconds. This ensures that plenty of protein is injected with out causing peak broadening. Because the differneces in separation times is small, peak broadening lead to peaks blending into another. The separation process for a 27 cm capillary lasts 16 minutes. The voltage for this capillary is set for 8.1 kV. Longer capillaries require higher voltages. The voltage should be 300 volts per 1 centimeter of capillary. The time as well as the voltage should be adjusted for different lengths of capillaries. The length of the capillary will effect the resolution of the separation. If two proteins of similar molecular weight are studied, a longer capillary will resolve the peaks with a higher accuracy. The test mix is composed of seven different proteins. These proteins range from 14.2 KDa to 205 KDa. The relative migration times (RMT) of the proteins will providethe standard curve for the determination of unknown molecular weights. The slope and intercept of the line will then provide a simple formula for determining the unknown molecular weights of proteins.
Capillary Electrophoresis has many uses, and many more that have not yet been determined. A study on the separation of single stranded DNA is under way. This is phenominal, the CE has the potential to separate strands of DNA that only differ by one nucleic acid. CE is the only technique that has the power to acheive such an accomplishment. A sample mix consisting of 21 different lenghts (40 - 60 pD(A) ) are easily distinuishable. Further research on how different base pairs compare is currently being persued. This new technology has the potential to revolutionize forensic science. In fact, capillary electrophoresis has gained the attention of scientist in the fields of organic chemistry, biology, and pharmocology. The ease of use and the wide range of possibilities enable capillary electrophoresis to change the face of technology as we know it today.