| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 ARUP Institute for Clinical and Experimental Pathology, LLC, Salt Lake City, UT 84108
2 ARUP Laboratories, Special Chemistry Section, Salt Lake City, UT 84108
3 Department of Pathology, University of Utah Health Sciences Center, Salt Lake City, UT 84132
aaddress correspondence to this author at: ARUP Laboratories, 500 Chipeta Way, Salt Lake City, UT 84108; fax 801-584-5109, e-mail ericksja{at}aruplab.com
Angiotensin-converting enzyme (ACE; EC 3.4.15.1) catalyzes the formation of angiotensin II by cleaving the C-terminal histidylleucine dipeptide from angiotensin I (1). Indications are that ACE is affiliated with an autonomous renin-angiotensin system of the brain that participates in physiologic processes inside the brain (2)(3). In addition, studies suggest that changes in ACE concentrations in brain tissue, caused by various neurologic disorders, are reflected by alterations in ACE activity in cerebrospinal fluid (CSF) (4). For example, increased ACE concentrations in CSF are associated with neurosarcoidosis (4)(5)(6)(7), with affected patients generally having activities approximately twofold or more higher than those of healthy individuals (4)(6)(7). Increased CSF ACE has also been implicated in neurologic diseases, such as bacterial and viral meningitis and Behcet disease (4)(5)(6)(7). Decreased concentrations have been reported in patients with Alzheimer disease, Parkinson disease, and progressive supranuclear palsy (8)(9).
The spectrophotometric assays customarily used for serum ACE lack the sensitivity for measuring the concentrations typically found in CSF. Consequently, more sensitive and costly methodologies are routinely used, such as fluorometric assays or HPLC (4).
To provide a more economic assay to measure CSF ACE for the screening of neurosarcoidosis, several attempts were made at modifying commercially available reagents and protocols used specifically for spectrophotometric measurement of serum ACE. A successful assay was eventually established that utilizes the ability of ACE to hydrolyze the tripeptide N-[3-(2-furyl)acryloyl]-L-phenylalanylglycylglycine to furylacryloylphenylalanine and glycylglycine (10). However, the sample volume (0.6 mL) required to achieve acceptable sensitivity was impractical. The assay was therefore reformatted for a 96-well microplate. This dramatically decreased sample and reagent volumes and increased accuracy by making duplicate measurements viable for practically all specimens received in our laboratory.
ACE reagent, calibrator, bovine serum albumin (BSA), and buffer reagents were purchased from Sigma Diagnostics®. Immulon 1B Removawell® Strips, holders, and plate sealers were purchased from Dynex Technologies, Inc. The SPECTRAmax® PLUS plate reader was manufactured by Molecular Devices Corp. and was controlled with their ProMax software. Data analysis was performed using Microsoft Excel software.
CSF was collected in plastic tubes and stored at 2–8 °C for 7 days or less, or was frozen at -20 or -70 °C for longer storage. Severely hemolytic or xanthrochromic specimens were avoided.
ACE reagent was reconstituted at 0.6 times the manufacturer’s recommended volume. ACE calibrator was reconstituted as instructed; diluted with 10 g/L BSA in phosphate-buffered saline (PBS; 0.008 mol/L Na2HPO4, 0.003 mol/L KH2PO4, 0.150 mol/L NaCl, pH 7.2) to activity values of 5.0, 3.0, and 1.0 U/L; and aliquoted for the calibrator and high and low controls, respectively. (One unit is defined as the amount of ACE needed to produce 1.0 µmol of furylacryloylphenylalanine/min.)
The assay was performed by adding 135 µL of ACE reagent to each microwell, excluding one well for the blank. The plate was sealed and incubated in the plate reader for 10 min at 37 °C to warm the reagents. The seal was removed, and 115 µL of ACE calibrator, control, 10 g/L BSA in PBS (for background), or unknown was added in duplicate into the designated wells. The plate was resealed, shaken to mix the reagents, and incubated for 10 min to start the reaction. After incubation, the seal was removed, the plate was briefly shaken, and the initial absorbance was measured immediately at 340 nm. The plate was resealed, and the incubation was resumed for 1 h. After incubation, the final absorbance was measured as before.
CSF ACE activities were determined by first calculating the absorbance change (
A) for each well. The mean A of the background wells was then subtracted from the A of each remaining well, producing a background-corrected A value. The values were then inserted into the following equation:
 |
A" is the absorbance change of the unknown or control corrected for background, and the "Mean corrected calibrator A" is the mean of the background-corrected change of the calibrator. Specimens with ACE activity >5.0 U/L were diluted in 10 g/L BSA in PBS and reassayed. The assay was linear to 5.0 U/L, with a slope and y-intercept of 0.9684 and 0.1466 U/L, respectively (R2 = 0.997; n = 5). Mean (SD) recovery was 106 (5)% as calculated from CSF samples to which four concentrations of ACE had been added. Twelve replicates of the 10 g/L BSA in PBS diluent generated a detection limit of 0.4 U/L (mean of 0.24 U/L plus 2 SD of 0.06 U/L). Assay imprecision was as follows: for intraassay imprecision, mean (SD) activities of 1.2 (0.15), 2.2 (0.16), and 4.2 (0.14) U/L (n = 10) with CVs of 13%, 7.2%, and 3.3%, respectively; for interassay imprecision (n = 18), mean (SD) activities of 1.0 (0.27), 2.3 (0.27), and 4.2 (0.29) U/L with CVs of 26%, 12%, and 7.0%, respectively.
We investigated blood and/or serum interference by adding a serum (final concentrations, 5, 10, and 20 µL of serum per mL of CSF) with an ACE activity of 181 U/L (upper limit of the reference interval for serum is 64 U/L) to two CSF specimens with ACE activities of 1.0 and 1.6 U/L, respectively. All samples showed marked interference, producing activities of 2.0, 3.0, and 4.8 U/L and 2.9, 3.9, and 5.6 U/L, respectively, for the two specimens at the three added volumes.
A split-sample study (n = 35) against a reputable assay (Roche FARA) used by the University of Washington was completed. Deming regression generated a slope and a y-intercept of 0.960 and 0.175 U/L, respectively (R2 = 0.938). Note, however, that the results from the university are reported in whole numbers with a detection limit of 1 U/L. Activities from our assay are to the nearest 0.1 U/L because of the higher sensitivity.
Measurement of 167 CSF specimens, deemed normal from negative oligoclonal banding and myelin basic protein results, produced an upper limit of the reference interval of 2.5 U/L at the 95% confidence limit [mean (SD), 1.2 (0.63) U/L]. Only four (2.4%) of the results exceeded the calculated reference interval (Fig. 1
). Analyses were also made vs patient age (8–79 years) and gender. As shown in Fig. 1
, no correlation between normal CSF ACE activity and patient age was evident. The same was true when activities were segregated by gender, with males (n = 63) and females (n = 104) generating upper limits of the respective reference intervals of 2.4 and 2.5 U/L.
|
As described previously, severe hemolysis may artificially increase CSF ACE measurements. This especially becomes relevant if the patient has highly increased blood ACE. Clean taps are therefore preferred. However, only two specimens were immediately excluded from our reference interval study, and only one was excluded from the correlation study because of visibly severe contamination. No major discrepancies or excessive outliers were evident in the data derived from the remaining samples, some of which did exhibit minor contamination. In other words, the majority of the CSF specimens received in our laboratory are sufficiently clean for analysis using the assay. Nevertheless, we recommend that any visible contamination be noted in case of questionable results and that grossly hemolyzed specimens be avoided altogether. In addition, hemolyzed CSF that has been centrifuged may contain residual serum that could possibly influence results.
The use of our assay and reference interval for the screening of neurosarcoidosis parallels other published studies addressing the disease. As stated previously, affected patients frequently have CSF ACE activities approximately twofold (or more) higher than those in healthy individuals. For example, in a study by Schweisfurth and Schiöberg-Schiegnitz (4), utilizing a fluorometric assay, patients with acute sarcoidosis of the brain had a median CSF ACE activity of 88.5 U/mL compared with a median control patient value of 39.5 U/mL. Oksanen et al. (6), using a radioactive inhibitor binding assay, showed that patients with active neurosarcoidosis had a mean CSF ACE activity of 1.42 U/mL compared with a control value of 0.78 U/mL. Another study by Jones et al. (7), involving two patients diagnosed with neurosarcoidosis, listed CSF ACE activities of 1.8 and 5.4 µmol · L-1 · min-1 for two affected patients vs a mean value of 0.59 · L-1 · min-1 for 38 control individuals. The upper limit of our reference interval (2.5 U/L) and the mean for our reference population (1.2 U/L) retain an analogous relationship, mirroring the approximate twofold or greater increases disclosed above.
In summary, the spectrophotometric microplate-formatted CSF ACE assay described here has adequate sensitivity, accuracy, precision, and reliability for the screening of neurosarcoidosis. The assay incorporates economical and available commercial reagents and common buffers in a configuration that allows for a reasonable CSF sample volume in addition to use of instrumentation typically found in the clinical laboratory. By measuring 167 CSF specimens from healthy controls, we have established an upper reference value of 2.5 U/L. A single reference interval is reasonable because no relationship between ACE activities and patient age or gender is evident. In addition, the assay adequately correlates with an established assay used by the University of Washington despite that assay’s limitations. Satisfactory correlation is also supported by the fact that our upper reference value varies by only 0.5 U/L (2.5 vs 2 U/L), a difference that is irrelevant considering the sensitivity issues addressed previously. It does, however, provide additional assurance that our assay is working within the parameters of its intended purpose. Furthermore, our results are analogous to those obtained in other studies addressing CSF ACE activity in neurosarcoidosis screening.
Acknowledgments
We gratefully acknowledge the ARUP Institute for Clinical and Experimental Pathology for financial support.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
