| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Oak Ridge Poster Sessions |
a author for correspondence: fax (409) 772-9231, e-mail aamohamm{at}utmb.edu
Measurement of 24-h urinary free cortisol (UFC) provides the most sensitive and specific diagnostic information for adrenal malfunctions, especially for Cushing syndrome (1)(2). The existing methodologies include immunoassays (3)(4)(5) and HPLC, with and without solid-phase extraction and derivatization (6)(7). Immunoassays for UFC are precise but overestimate the concentration because of antibodies' cross-reactivity with various metabolites and with synthetic corticoids having a chemical configuration similar to that of cortisol (8). HPLCs, on the other hand, although very specific, require large volumes of mobile phase, as well as sample pretreatment. Capillary electrophoresis (CE) overcomes many of these problems through the use of an open-tubular format, thus avoiding interaction of analytes with the solid-phase resin. Earlier, we reported the feasibility of rapid UFC detection in solid-phase extraction CE with a neutral capillary (9). Using a neutral capillary, we could demonstrate a detection limit of 552 nmol/L for UFC in human urine.
However, neutral capillaries have several disadvantages: Not only are they expensive, they also are unstable and lack a uniform capillary coating. This makes the neutral capillary unattractive for the development of a routine clinical laboratory assay. In the present study, we developed a solid-phase microparticle extraction (SPME) coupled micellar electrokinetic capillary chromatography (MEKC) with use of a fused-silica capillary as a potential analytical method for rapid separation and detection of UFC. The overall performance, feasibility, linearity, recovery, and lower limit of detection of UFC in human urine were evaluated.
To isotonic saline or urine (10 mL) we added 276 nmol/L corticosterone (Sigma Chemicals) as internal standard (IS) and various concentrations (138–4416 nmol/L) of cortisol (hydrocortisone; Sigma Chemicals) and its conjugated metabolites (purchased from Steraloids): 4-pregnen-11ß,17,21-triol-3,20-dione-21-acetate (cortisol acetate), -21-sulfate, and -21-glucosiduronate. Steroid stock solutions (1 g/L) were made by dissolving the respective steroids in absolute ethanol.
The samples are rapidly reconcentrated under reduced pressure by using 3 M Empore SPME disc cartridges (Fisher Scientific). The discs have C18 modified 12-µm (o.d.) fused silica particles immobilized on an inert matrix of polytetrafluoroethylene fibrils (7 mm in diameter, 0.5 mm thick) secured in 3-mL polypropylene columns. These discs are preconditioned with 250 µL of methanol and washed with 1 mL of deionized water. After passing the respective supplemented urine/saline samples under suction, the discs are washed twice with 1 mL of acetone:water (10:90 by vol) and then once with 1 mL of distilled water. The final elution of steroids was done with 80 µL of acetonitrile followed by 320 µL of 10 mmol/L sodium dodecyl sulfate (SDS). The effluent was used for CE analysis without any further processing. The extraction efficiencies achieved under similar conditions (9) were between 89% and 94%.
MEKC analysis was done with a Beckman P/ACE 5010 capillary electrophoresis system equipped with system Gold software for data analysis. The run buffer contained 15 mmol/L phosphate, pH 2.5, 100 mmol/L SDS, and 200 mL/L acetonitrile. We used a 57-cm [75 µm (i.d.) x 375 µm (o.d.)] fused-silica capillary (Supelco). Samples were injected under high pressure for 20 s, and capillary temperature was maintained at 16 ± 0.1 °C. Fixed-wavelength detection at 254 nm and 10 kV voltage under reversed polarity were used for all separations.
Because steroids are neutral and lipophilic molecules, separation in a fused-silica capillary is not possible without addition of an organic modifier, such as acetonitrile, to allow partitioning of the steroids between the SDS (micellar phase) and the acetonitrile/water (aqueous organic phase). To achieve high resolution, it is critical that the electrosmotic flow be negligible. With the fused-silica capillary this was achieved by using an acidic pH (2.5), which protonates the silica and allows the capillary to behave analogously to a neutral capillary. At this pH, the migration of steroids across the detection window is dictated by electrophoretic mobility of the SDS micelles and partitioning of the individual steroids into the micellar phase. Thus, the nonpolar hydrophobic steroids, being more soluble in SDS micelles, are eluted first, followed by the polar steroids.
Figure 1
A shows the separation of IS and cortisol (138 nmol/L) in
steroid-supplemented human urine. Baseline separation was achieved for
both free cortisol and the IS corticosterone at 9.4 and 8.2 min,
respectively. No interferences from added bovine serum albumin or other
urinary steroids affected the free cortisol determinations
(9). In the present study and in others (7)
corticosterone was used as the IS because other currently recommended
compounds such as dexamethasone and fluodrocortisone are in wide use
therapeutically. Moreover, urinary concentrations of free
corticosterone are extremely low and do not fluctuate widely
(10)(11). Also, correlation was excellent for
peak area ratios (peak area of cortisol/peak area of IS) calculated for
supplemented isotonic saline (138–4416 nmol/L; r =
0.99; data not shown) and urine samples (Fig. 1B
; 138-4416 nmol/L,
r = 0.99).
|
These observations show the feasibility of a rapid UFC detection in
SPME-CE with a fused-silica capillary. Very rapid preconcentration
(<15 min) by SPME disc cartridges and separation by CE can
substantially reduce labor requirements and turnaround times. The
detection limit of 138 nmol/L (signal/noise 
3) achieved with a
fused-silica capillary was improved over that obtained with a neutral
capillary under similar conditions. Additionally, endogenous
interfering compounds in urine migrated earlier than IS and cortisol
than when separated in a neutral capillary, thereby making the
detection more efficient. Although adequate for the detection of UFC in
Cushing patients, the present detection limit for cortisol of 138
nmol/L (276 nmol/day, assuming 2 L of urine excreted in 24 h) is
higher than the reference range for healthy persons (30–150 nmol/day).
Improving the detection limit by at least 4–5 times is required to
make this a clinically acceptable assay to detect cortisol within the
reference range. Currently, we are considering several approaches to
increasing the analytical sensitivity, e.g., better detector systems,
increasing urine sample volume for extraction, electrokinetic injection
of sample, and field-amplified sample stacking. The speed and high
resolving power of CE show definite advantages for reagent consumption,
simplified assay methodologies, and potential multianalyte and
multilane capabilities.
Footnotes
Clin. Chem. Div., Dept. of Pathol., Univ. of Texas Med. Branch, Galveston, TX 77555-0551
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
