OR WAIT 15 SECS
Kevin Altria is an associate director in the pharmaceutical development department of GlaxoSmithKline. He is editor of "CE Currents" and a member of LCGC Europe's editorial advisory board.
A review of the expanding range of consumables available to capillary electrophoresists and why they are useful
CE has become commonly used in routine analysis in recent years and there is now a large range of accessories and reagents available for routine CE operations (Table 1). The range of products available is quite extensive and analysts may not be aware of the diversity of offerings and why they are useful. This article is a guide to the range of supplies that are currently available and highlights some of their specific characteristics and background. A list of some of the main suppliers with some additional information can be found in references 1 and 2, and also by using the FindAnalytichem search engine on chromatographyonline.com Simply type in "Capillary Electrophoresis" and click the product section.
Table 1: Range of frequently used consumables and reagents in CE.
Pre-made buffers. CE users often use standard generic buffer compositions for their separations and this allows commercial suppliers to stock pre-made buffer solutions. For example, pH 2.5 phosphate buffer is commonly used for basic compounds while both pH 7 phosphate and sodium tetraborate (natural pH of 9.3) are commonly used for the analysis of acidic species. These approaches were discussed in a previous CE Currents article.3 Common buffer concentrations are 25, 50 and 100 mM. These standard CE buffers are available, along with others, from CE instrument suppliers and from HPLC/CE consumable supplier companies. Traditional reagent suppliers also supply pre-made CE buffers.
Pre-made buffers offer several convenient advantages in routine operation. They are generally prepared against a standard such as GLP/GMP in ISO9001 facilities with appropriate trained staff. This means that the purity and quality of the solutions are assured and consistent. The reagents used are generally Electrophoresis grade and the solutions are pre-filtered. Traceability can also be assured through the Lot number with buffer pH being tested against referenced solutions in accredited laboratories. The buffer solutions generally come with some sort of certificate of analysis verifying the pH and electrolyte concentration.
This helps particularly when working in a regulated environment, such as the pharmaceutical industry, where there is a strong need to demonstrate traceability and quality of materials used. Method transfer between sites is also aided if there is a possibility of using pre-made buffers because this reduces the potential variability of buffer preparation in the transfer exercise. Costs are typically in the order of $50 for 250 mL of buffer, which may seem expensive but this should be viewed against the time required to prepare buffers and the consequences of preparation errors. Buffer solutions are generally supplied in plastic bottles labelled with all the required information.
Reagent solutions typically include the rinsing and regenerating solutions used in CE. For example, 0.1 M and 1 M NaOH are frequently used in rinsing and regenerating fused-silica capillaries. The costs and benefits of these pre-made solutions are similar to pre-made buffers.
Pre-made buffers can form the basis for more complicated electrolyte solutions. For example, cyclodextrin can be added to provide solutions for chiral analysis (or SDS added to generate micellar solutions for MECC). The purity of the reagents used is critical and should be specified and used consistently in routine operations. For example, SDS is available in different grades and purity. Variable purity would have an effect on the SDS concentration which, in turn, would affect the separations achieved because it would alter the number of micelles present to perform chromatography.
Cyclodextrins such as sulphated cyclodextrin (Figure 1) are chemically modified to alter their chemical properties. This chemical derivatization may not be total and the cyclodextrin may be only partially derivatized. The extent of the substitution is technically referred to as the degree of substitution (DS). The DS for derivatized cyclodextrin is typically measured by NMR and is generally included on Certificates of Analysis. The variability in the purity of these cyclodextrins can have a huge impact on the separations achieved. For example. Figure 2 shows the difference obtained for the chiral separation of terbutaline using two batches of sulphated cyclodextrin.4 Consistent purity materials should be used or, as a last resort, it may be necessary to adopt a testing procedure for batches to ensure that they are suitable. It is also important to assess the effect of reagent purity when validating methods and appropriate controls should be written into methods.
Reagents have been devised for specific applications of CE. Some of these are highly proprietary, are used for specific analytes and used in conjunction with separation protocols, which specify buffer and sample preparation in great detail.
DNA is profiled using sieving gel matrices. Originally these matrices were fixed gels that were permanently bonded into the capillary. Replaceable gel buffer solutions have now become common, which allow gels to be pumped through the capillary and eliminates problems, such as fouling and gel deterioration, that occurred with the fixed gels. The gel buffer solutions are sold for particular DNA analyses and may cover specific DNA types and size ranges. Obviously, details of the buffer would need to be specified in a method and lot-to-lot variability assessed during validation studies. DNA standards are available for use as reference materials to calibrate methods.
Proteins are frequently characterized using capillary iso-electric focusing (CIEF) and this involves use of ampholytes to create a pH gradient (Figure 3). The capillary is filled with a solution of the ampholytes. The ampholyte contains a mixture of components that are amphoteric (i.e., contain both positively and negatively chargeable groups). These components migrate under the influence of the voltage and by doing so generate a pH gradient. Amphoteric analtyes have an iso-electric point that is the pH value at which their positive charge exactly equals their negative charge. At the iso-electric point (pI) the analyte has a net zero charge and stops migrating. If the protein drifts away from its pI it acquires a charge that makes it migrate back to its pI — hence the term focussing. The Ampholytes can be bought as reagents or in pre-made application solutions. Ampholytes are sold to cover specific pH ranges, which may span several pH units or a 1 or 2 pH units. It is important to specify the exact Ampholyte reagents or buffers used. Proteins standards are also frequently used in CIEF as reference compounds to calibrate the pH gradient.
Application-specific multiplexed (24 or 64 capillaries) capillary array instruments are based around one assay type. All consumables include all the required solutions, capillaries and standards/reagents required are detailed in method protocols. Applications of the multiplex instruments include analysis of both proteins and DNA. A large percentage of the data used in the Human Genome Project was generated by multiplex CE.
Capillaries are composed of fused silica and are coated with polyimide. Typical internal diameters that are routinely employed are 50, 75 or 100 μm with a typical outer diameter of 375 μm. The internal surface of the capillary can be uncoated, chemically modified or coated by using additives in the buffer.
Uncoated fused silica capillaries. Uncoated capillary can be purchased in lengths that can then be cut to length and a window for detection created. Lengths of capillary, for example, 5 m can be purchased from CE instrument and consumable suppliers or from specialist capillary manufacturers. Capillaries are provided on spools for ease of use and handling. Capillaries are generally round but square and rectangular capillaries are becoming available. These help with improved sensitivity as they potentially reduce stray light noise and have a wider pathlength but they require careful alignment in the cartridge.
Bubble cell capillaries. Bubble cell capillaries (Agilent, Waldbronn, Germany) increase sensitivity, according to the company. A swollen area of the capillary is used as the detection area. For example, a 50 μm capillary may have a detection area swollen to 150 μm, which gives an S/N increase of around 2.7 as the noise is slightly increased. Figure 4 shows the separation of a mixture of basic compounds using both a standard 50 μm capillary and with a 50 μm capillary that had a 150 μm bubble cell. The advantage of the bubble cells is that they give increased sensitivity through having a wider detection diameter but without any noticeable increase in operating current. The bubble cell capillaries cannot normally be fitted onto other systems.
Capillary cartridges. The capillary is housed in a cartridge device that provides mechanical support. The cartridge also ensures that the capillary is aligned into the optical path of the UV detector and is held firmly to reduce any baseline noise due to mechanical movement of the capillary. The cartridges also form part of the temperature regulation system as they may be filled with a temperature controlled coolant liquid or temperature controlled air is forced through the cartridge. If the temperature is controlled through use of a coolant then consumables include gaskets and o-rings, which are needed to prevent fluid leakage.
The cartridge interfaces into the optical path of the UV detector. The amount of UV light that shines onto the capillary is regulated either through an interface device or through an aperture slit. The interface device used is secured to the capillary before being placed in the cartridge. The interface devices are optimized for use with specific capillary internal diameters (or bubble cells) to give the maximum signal-to-noise (S/N) ratio. Alternatively, the amount of light is regulated using an aperture slit in the cartridge itself. For example, a 200 μm slit would be used for maximum peak resolution and efficiency or a 1000 μm slit for improved sensitivity. The use of the correct aperture or interface device is essential for reproducible results and should be specified in methods.
The cartridge devices can be labelled with the capillary/method details and logged for traceability.
The polyimide of the capillaries increases the mechanical strength of the capillary. However, the coating must be removed in the detection area of the capillary to allow the UV light to pass through. Capillaries can be purchased pre-cut to length and with a pre-made detection area from CE instrument suppliers. It is considerably less expensive and more flexible to purchase spools of capillary and prepare capillaries yourself.
Cutting the capillary correctly is vitally important as a poorly cut capillary can give significant peak tailing and reduced performance. A recent paper discussed the various approaches to successfully cleaving capillaries.5 Capillaries can be cut with a laser to give the best finish (typically used by CE instrument manufacturers) — but these are not typically available in standard labs. Devices such as crystal cutter, a ceramic stone, or a capillary-cutting device incorporating a sprung-loaded industrial diamond, are employed to generate a score through the polyimide which allows the capillary to be "pulled" apart to produce a clean cut.6
The polyimide can be burnt off to form the detection window using a device that contains an electrical filament. The capillary is pushed through the filament which glows red hot when the switched on. The polyimide is charred and any residue removed with a methanol-moistened tissue. Figure 5 shows that consistent detection windows are generated using this approach. The polyimide should not be scrapped off as this scratches the surface and increases noise, the use of a match or flame should also be avoided as this can cause the capillary to bend and become distorted.
Coated capillaries. Capillaries are available with a variety of coatings. The coatings are generally used to alter the level of electro-endosmotic flow (EOF) that occurs in the capillary when the voltage is applied. In applications such as CIEF and using gel-filled capillaries there is preference to eliminate EOF completely because this complicates the separation and can lead to zone-mixing. An example coating is polyvinyl alcohol-coated (PVA) which is covalently bonded onto the capillary surface and eliminates EOF. PVA and similar coated capillaries also shield the capillary surface and prevent/reduce peak tailing that may have occurred due to analyte adsorption.
Some capillaries are available that are coated with substances (e.g., sulphonic acids) that enhance the level of EOF and/or make it pH-dependent. Alternatively, it is possible to purchase positively charged capillaries that have a strong cationic charge on the surface — this eliminates tailing of positively charged analytes and also reverses the EOF direction.
There are a variety of kits available for specific applications. These kits generally contain both the capillary and all the buffer and regeneration solutions required to operate the method. The buffer systems are developed and optimized for specific analytes and have pre-determined operating conditions such as detection wavelengths, operating voltages, temperature and so on. The commonly used kits include those for small inorganic anions (e.g., chloride, phosphate etc. or metal ions such as sodium, calcium etc.). Other kits include nucleic acids, DNA, oligonucleotides, sugars etc. Carbohydrate kits that may also contain the required dyes to allow derivatization and fluorescence detection. DNA application kits may also contain reference materials to calibrate and/or assess system performance.
Method development kits are also available. For example, chiral method development kits contain a range of cyclodextrins that can be assessed as chiral selectors for a specific analyte.
One of the operating difficulties associated with CE is the variability of EOF generated by different capillaries (due to their variable charge densities) and the pH dependence of the EOF flow rate. Dynamic coating systems (CElixir buffers) have been developed (Microsolv, Eatontown, New Jersey, USA) that overcome these problems,7 according to the company. When using these systems the surface of a new uncoated fused silica capillary is initially extensively rinsed with NaOH solution to activate the surface silanols. The capillary is then rinsed with a solution containing a cationic surfactant, which forms a highly positive and consistent surface coating. The capillary is then rinsed with an anionic surfactant that binds onto the positively charged surface (Figure 6). This generates a consistently negatively charged surface that produces a pH-independent consistent EOF flow.
Use of this dynamic coating system has been shown to greatly improve method reproducibility.7 Figure 7 shows the repeatability of a separation of basic compounds on two freshly prepared capillaries.7 This level of repeatability assists in method transfer and improves precision both within-run and across runs.
The dynamic coating systems are available for conventional CE separations covering a range of pH values and also as application-specific, such as metal ion analysis.
Platinum electrodes are used as inert materials to complete the electrical current. These can easily get bent or broken if they are misaligned or catch on the vial caps. If this occurs then it is well worth keeping one or two spares.
Occupational qualification (OQ) and performance qualification (PQ) kits are also available for use in routine operations to demonstrate equipment suitability — especially after maintenance or if the equipment has been repaired/relocated. These contain capillaries, reagents and traceable reference sample solutions. The reference solutions are analysed and the response obtained, for example, migration times, peak heights and so on are compared with the expected values and the data archived. This approach is needed in CE because it is not possible to perform tests, such as measuring the flow rate etc., that are used in HPLC.8
Filters are used for both reagents and sample solutions. The filtration step for both reagents and samples should be validated and consistent procedures and filter types should be used in routine operation. Vials and caps are unique to the CE instrument supplier. Sampling from very small sample volumes is possible in CE using small sprung-loaded inserts into the vials.
Oddities include a bench that can be purchased to raise a CE system to the same height as a MS system to avoid siphoning flow. Another very specific consumable are the capillary arrays used in multiplex CE systems — these are fragile and expensive.
CE is now well established and routine for many applications such as protein analysis using IEF, DNA analysis and metal/inorganic anion analysis. Standard conditions have been established for these analyses which have led to the marketing of application kits and buffers. The use of common buffers has also led to these becoming commercially available. Reagents such as suphated cyclodextrin have become standard and the need for known and constent purities has been recognized. Method complexity has led to the need for consist capillary coatings. Analysis-specific instruments have been developed as "turnkey" systems where there standard conditions and reagents/capillaries are employed. The need for greater robustness and method repeatability has led to the design of approaches such as dynamic coatings. Undoubtably as further applications become standard then additional application kits and reagents will become available.
Kevin Altria is an associate director in the pharmaceutical development department at GlaxoSmithKline. He is editor of CE Currents in LCGC Europe.
3. K.D. Altria, LCGC Eur., 13(3), 320–330 (2000).
4. K.D. Altria and F. Campi, LCGC Int., 12(6), 358–362 (1999).
5. J. Macomber, P. Lui and R. Acuña, LCGC N. Am., The Applications Notebook, Sep 1, (2009).
6. K.D. Altria et al., LCGC Int., 10(3) 157–162 (1997).
7. K.D. Altria, J. Pharm. Biomed. Analysis, 31, 447–453 (2003).
9. K.D. Altria and D. Hindocha, LCGC Int., 11(11), 712–718 (1998).