Development of laser-based detection methods for capillary electrophoresis (CE)

---

The short optical pathlengths (≤ 50 µm) encountered in CE for on-column UV absorbance result in poor concentration detection limits (≥ 10^-6 M), an inconvenience that has plagued CE since its commercialization in the early 1990s. On the other hand, superb detection sensitivity for CE can be achieved using laser induced fluorescence¹; however, this method is limited by the minimum analyte concentration (10-100 nM) that can be effectively conjugated to an appropriate fluorophore unless a pre-concentration step is first incorporated². Our group is interested in laser-based methods to a) improve UV absorbance-based detection based on the thermo-optical effect^{3, 4} and b) optimize “native” fluorescence detection^{5, 6}, thereby eliminating the need for fluorescent labelling.

Our laser-based detection development for CE has been carried out using an inexpensive KrF excimer laser (λ=248 nm) for a) thermo-optical absorbance (TOA) detection of organic acids^{7, 8} and native or derivatized peptides^9-11, and b) laser-induced native fluorescence (LINF) detection of proteins and smaller organic acids in biofluids¹², pharmaceuticals¹³ and tryptophan-containing peptides¹⁴. The TOA detector, which is a pump-probe beam technique that is essentially pathlength independent, has demonstrated detection limits 10-100 times lower than by traditional absorbance with CE particularly for weakly absorbing species¹¹. The KrF excimer laser as a UV excitation source provides the latitude to study species with wide ranging absorption maxima (from 220 to 270 nm) using this single-wavelength source.

For tryptophan-containing polypeptides, the KrF laser is a cost-effective alternative for native protein analysis by LINF, which we demonstrated by profiling different body fluids¹². The KrF excimer laser, operated at high repetition rates without gated integration, provides an LOD for Trp of 3 nM. The CE-LINF system has been used occasionally to look at protein digests collected from our enzyme microreactor because of its higher sensitivity and simplified maps (i.e., only trp- and tyr-containing peptides are detected). Besides LINF, the commonly used protecting agent for solid-phase peptide synthesis, FMOC (9-Fluorenylmethoxycarbonyl), fluoresces well when excited at 248 nm and we showed by CE-LIF the diastereomeric separation and purity analysis of synthetic dipeptides protected with FMOC¹⁵.

↑ Back to Top

 

Immobilized enzyme microreactors for rapid, sensitive peptide mapping by CE

---

Enzymatic digestion of protein is typically carried out either in a homogeneous solution (liquid-phase) followed by separation and identification of peptide fragments, or in a non-homogeneous fashion “in-gel” on trapped proteins that have been separated by polyacrylamide gel electrophoresis. This is generally followed by peptide extraction and mass mapping. However, enzyme autolysis, non-reusability of enzyme and losses due to sample handling all contribute to reduced detection sensitivity when it comes to the determination of low abundance proteins. These limitations apply to peptide sequencing as well, where enzymatic digestion is an integral step. To this end, we have been developing and characterising enzyme microreactors (µ-IMERs) employing immobilized trypsin^{16, 17} and chymotrypsin. We have explored several enzyme immobilization techniques thus far: trypsin coupled to diisothiocyanate-derivatized controlled pore gass particles (trypsin-DITC-CPG) packed into a fused silica column (30 cm × 530 µm I.D.); trypsin coupled to aminopropyl-activated CPG using either EDC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide) or glutaraldehyde (GA)^18-21 coupling chemistry; trypsin or chymotrypsin directly cross-linked using GA, with out a solid support^22-26. Peptide mapping is achieved by CE-UV, CE-LIF, HPLC-UV and/or MALDI-TOF MS.

The challenge of mapping very small amounts of protein is exacerbated by dilution of peptide fragments within the microreactor, leading to an overall loss in sensitivity of the mapping process. Therefore, we have also characterized a µSPE device coupled to the CE-UV system for quantitative transfer of the tryptic fragments into the separation capillary^{17, 27}. To circumvent this detection sensitivity limitation during the characterization of our immobilized enzymes, we are preparing fluorescently-labelled protein substrates so that the collected peptides can be mapped by CE-LIF^{24, 25}. This approach will allow us to digest sub-nanomolar quantities of protein without being limited by post-digestion labelling reactions. Our lab is equipped with LIF detectors operating at 488 nm, 410 nm, 325 nm and 248 nm.

↑ Back to Top

 

CE and capillary electrokinetic chromatography (EKC): fundamental studies and applications

---

The separation of peptides by CE can, in theory, be improved by adding a mechanism of partition to that of electromigration through the use of pseudo-stationary phases. We have carried out several studies of peptide separations by micellar electrokinetic chromatography (MEKC) and by cyclodextrin-modified CE, probing the structure–dependent binding of peptide analogs to additives, determining association constants, and developing separation methods for a variety of applications. For example, we showed that electrostatic interactions between charged micelles and oppositely charged analyte was strong enough to overcome partition-based selectivity in the MEKC separation of enkephalin neuropeptide analogs⁹. We quantified peptide-micelle complexation and compared the results to association constants determined by absorption spectroscopy to investigate the extent of electrostatic versus hydrophobic analyte inclusion²⁸. We used cationic surfactant to fine tune the electroosmotic flow to determine trans, trans-muconic acid—a biomarker of benzene exposure—in urine⁷. Cyclodextrin EKC was used to separate underivatized dipeptide stereoisomers and determine their binding constants with -cyclodextrin²⁹. Similar projects involving the development of CE and MEKC methods for various applications are ongoing in our research group.

↑ Back to Top

 

Fractionation, chemical characterization and bio-toxicity studies of the combustion products of tobacco and one of its major components, chlorogenic acid

---

The goal of this project is to fractionate and determine the chemical composition of the most biotoxic extracts associated with the combustion products of chlorogenic acid and of whole tobacco smoke. Chlorogenic acid is the most abundant naturally occurring polyphenol found in leaf tobacco. The biological effects of its combustion products remain largely unknown. The first part of this project involves the concurrent analysis and comparison of the combustion products of chlorogenic acid and its bio-toxicity³⁰. Preparative scale LC and analytical LC/MS are used, respectively, for the fractionation and chemical characterization of extracts of the combustion products. Toxicity is assessed using the in-vitro micronucleus test (IVMNT). For example, a sample of chlorogenic acid was burned at 640 °C for 2 min and the particulate matter of the smoke was collected onto Cambridge filter pads followed by selective extraction in several different solvents. Various fractions of the chlorogenic acid combustion products were tested for induction of micronuclei in V79 Chinese hamster fibroblast cells. We identified over forty compounds in the dimethyl sulfoxide (DMSO) extract by HPLC coupled to electrospray time-of-flight MS. The DMSO extract was then fractionated and sub-fractionated, with toxicity studies carried out on all fractions. The most toxic response was determined to contain catechol.

In the second phase of the project, dosing of the whole smoke exposure system using a Borgwaldt RM-20S instrument is being evaluated using LC/UV and a hydrocarbon analyser equipped with a flame ionization detector (FID), by evaluating markers of both the particulate and vapour phases, as well as determine the biological effects of each phase (particulate and vapour) on a specific biological system. The methodology for bio-toxicity studies involves use of the neutral red uptake (NRU) assay to compare the exposure of human alveolar basal epithelial cells (A549 cells) to whole smoke versus the vapour phase only. This can be achieved by comparing bioassay results with and without the Cambridge filter pad inline. Future objectives include: i) developing an analytical method for the dose assessment and ii) characterization of the vapour phase of the smoke delivered and adsorbed/absorbed by cell culture media. The latter would allow for the correlation of toxicological information to accurate dose measurements.

↑ Back to Top

 

Development of coupled methods to fingerprint oligosaccharides

---

Chitin, polyß-(1→4)-N-acetyl-D-glucosamine, is the insoluble substance comprising the exoskeleton of crustaceans and thus a waste fisheries product. Conversion of chitin to chitosan by deacetylation gives a soluble, biocompatible derivative that can be used for vitamin, drug and vaccine delivery and in biomedical devices. This conversion is typically achieved by heating chitin with concentrated NaOH. In collaboration with biochemists, we are using the “green” route of directed evolution to engineer the deacetylase enzyme AxeA_tr (Acetyl xylan esterase A [EC 3.1.1.72], truncated) to produce highly deacetylated chitosan oligosaccharides (COS). We have been developing CE- and HPLC-based fingerprinting methods to separate COS having high charge- and poly-dispersity as a means to characterize the AxeA mutants. We developed a modified conjugation method using the fluorophore 8-aminopyrene-1,3,6-trisulfonic acid (APTS) to derivatize chitin (fully acetylated) and chitosan (fully deacetylated) oligosaccharide standards (dimer to hexamer) for analysis by CE-LIF at λ=488 nm^{31, 32}. To investigate the extent of AxeA mutants’ deacetylation of non-fluorescent substrate we use HILIC (Hydrophilic Interaction Liquid Chromatography) separation using a cyano column coupled to ion trap-MS detection. We demonstrated that AxeA, and its variants, could deacetylate up to 5 residues of the acetylated hexamer substrate leading to various patterns of deacetylation³³.

In a separate project, we are using soluble cellulose enzyme to investigate whether a highly reproducible oligosaccharide fingerprinting method can be developed using CE-LIF and HPLC-LIF for carboxymethylated and hydroxymethylated cellulose derivatives.

↑ Back to Top

 

Fractionation, chemical characterization and bio-toxicity studies of the combustion products of tobacco and one of its major components, chlorogenic acid

---

Microencapsulation as a means of enzyme immobilization provides an interesting alternative to enzyme attachment on solid phases because the interior cavity of the capsules creates an aqueous microenvironment that protects the enzyme from the external medium to thereby maintain enzyme activity and stability. In collaboration with Prof. Dominic Rochefort’s group, we have been studying microencapsulated laccase, an oxidase enzyme, using o-phenylenediamine (OPD) as a model substrate³⁴. A capillary electrophoretic method with UV absorbance detection (CE-UV) is used to separate the substrate and products and to quantify the enzymatic reaction for both free and immobilized enzyme. Microcapsules are then packed into a capillary-sized microreactor and conversion of substrate to product can be followed by CE-UV in an off-line manner. Coupling the microreactor to the CE-UV system on-line will be achieved using a tee and micro-injection valve. Our long-term goal is to develop an integrated system based on CE that can be used to evaluate the efficiency of microencapsulated enzymes in biosensor applications.

↑ Back to Top

 

Footnotes:

---