| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
-L-Fucosidase Activity with a Centrifugal Analyzer: Reference Values in a Healthy French Adult Population
1
Biochemical Laboratory A, Bichat-Claude Bernard Hospital, 75877 Paris Cedex 18, France
2
Department of Medical Biostatistics, Saint Louis Hospital, 75475 Paris Cedex 10, France
a Address correspondence to this author at: Laboratoire de Biochimie A, Hôpital Bichat-Claude Bernard, 46, Rue Henri Huchard, 75877 Paris Cédex 18, France; fax 33-1-40-25-88-21, e-mail labo.bioa{at}bch.ap-hop-paris.fr)
Release of 
-L-fucosidase (AFU; EC 3.2.1.51) from
the cells to the extracellular fluid can reflect an inherited or
acquired disease involving the lysosome apparatus or a nonspecific
lesion leading to cell lysis. Increased activities of AFU have been
observed in the sera of patients with diabetes mellitus, alcoholic
cirrhosis, and gastric carcinoma (1). Deugnier et al.
(2) first proposed that serum AFU activity is a useful
marker in the diagnosis of hepatocellular carcinoma (3). Its
use as a diagnostic test, however, is limited by the absence of
well-defined reference values and the lack of a sensitive and rapid
automated enzyme assay.
In most cases, AFU activity is determined by a modification of the
method of Zielke et al. (4). This method is limited by its
long incubation time (30–60 min), which reflects the poor affinity of
the enzyme for the colorimetric substrate used
(4-nitrophenyl--L-fucoside; 4-NPF). Wood
(5) observed that specific activities obtained with
fluorescent 4-methyl-umbelliferyl--L-fucoside
(4-MUF) substrate are more than twice those obtained with 4-NPF.
However, adaptation of this method as described by Lombardo et al.
(6) on the CLA 1500 continuous-flow analyzer (Carlo
Erba) leads to a large expense for substrate. Here we describe
the optimization of the fluorometric assay and its micromethod
adaptation to the Monarch 2000 centrifugal automated analyzer
(Instrumentation Laboratory) fitted with a spectrofluorimeter. We
further describe the distribution of plasma AFU activity on a healthy
French adult population.
All chemicals were purchased from Sigma except the
4-methylumbelliferone (4-MU) calibrator, which was obtained from Sebia.
The working substrate solution, 0.25 mmol/L 4-MUF, was obtained by
dilution of a 5 mmol/L 4-MUF solution (in methanol-water; 1:2, by
volume) in an acidic buffer (0.1 mol/L citrate-phosphate
buffer, pH 5.0). The working 50 and 100 µmol/L 4-MU calibrator
solutions were obtained by dilution of a 10 mmol/L 4-MU methanolic
solution in the acidic buffer and confirmed by measurement of their
absorbance at 321 nm (
321 = 16 720). The
present investigation was carried out with plasma samples, according to
the recommendations of Lombardo et al. (7). Plasma samples
(n = 274,) were obtained from healthy blood donors (age range,
18–60 years) with written informed consent. We used only
noninfectious (negative for hepatitis B, hepatitis C, and HIV virus)
samples and excluded plasma with alanine aminotransferase 
56
U/L and/or 
-glutamyltransferase activities 35 U/L (measured on a
Hitachi 747 analyzer; Boehringer-Mannheim). Some studies
(8)(9) have reported differences in enzymatic
properties between low- and high-activity plasma variants.
Rather than study conditions for maximal AFU activity on a plasma pool
(7), we performed optimization studies using plasma selected
from two healthy adult variants expressing low and high activity. For
the determination of precision, plasma pools were obtained from samples
with low, medium, and high AFU activity. Aliquots were stored at
-20 °C for at least 6 months. For the determination of linearity,
plasma samples (n = 21) were obtained from patients with
hepatocellular carcinoma.
Before AFU measurement, all samples (plasma, calibrators, and controls)
were first diluted automatically (1:10) with 0.15 mol/L NaCl containing
10 g/L bovine serum albumin (BSA) to prevent saturation of the
fluorescence intensity. Diluted plasma and calibrators (10 µL) were
mixed with 75 µL of acidic buffer with or without 4-MUF. After
a 15-min incubation at 37 °C, enzymatic activity was
stopped by the addition of 75 µL of a basic buffer (0.3
mol/L glycine-NaOH, pH 9.75). The fluorescence
(
excitation, 360 nm;
emission, 450 nm) of the reaction product
(4-MU) was recorded in the linear region delimited by the reference (10
µmol/L 4-MU in the basic buffer). A two-point calibration procedure
was used to measure the corrected fluorescence intensity
(reaction-blank) of each sample and to evaluate the enzymatic activity
(1 U = 1
µmol · L-1 · min-1). The
colorimetric method of Zielke et al. (4) was also adapted to
the Monarch 2000 centrifugal analyzer. In that case, undiluted plasma
samples and calibrators were mixed with the acidic buffer containing
the 4-NPF substrate (reaction) or not (blank), and the absorbance of
the reaction product (4-nitrophenol) was recorded at 405 nm.
Correlation analysis and linear regression analysis were performed with
the Statview software for Macintosh. Distribution of the plasma AFU
activity was analyzed with the FASTCLUST procedure of SAS software (SAS
Institute).
The slope obtained after serial dilution of a high-activity plasma sample in a diluent of 0.15 mol/L NaCl containing 10 g/L BSA was 1.25-fold higher than the slope with 0.15 mol/L NaCl as diluent. Citrate-phosphate buffer was preferred to acetate-phosphate buffer to prevent the formation of a cloudy mixture obtained with some plasma samples. The effects of pH (4.0–6.3) and 4-MUF concentration (0.01–1 mmol/L) on AFU activity were examined using low- and high-activity plasma variants. We measured slight differences in the pH activity profile and in Km values (low variant, pHmax = 5.0; Km = 0.035 mmol/L; high variant, pHmax = 4.8; Km = 0.022 mmol/L) as reported by Willems et al. (8). Working conditions (pH 5.0, 0.25 mmol/L 4-MUF) were chosen in favor of the low-activity variant.
The detection limit (mean + 3 SD) of the method was 0.026 U/L (n = 35), as determined by measuring the fluorescence intensity of the diluent solution. The upper limit of linearity of the assay, assessed by serial dilutions (from 1:2 to 1:200 in 0.15 mol/L NaCl containing 10 g/L BSA) of high-activity plasma samples was 35 U/L. None of abnormal samples (n = 21; AFU = 10.2 ± 5.5 U/L, mean ± SD) included in this study exhibited AFU values >35 U/L. Within-batch CVs ranged from 2.5% (low-activity samples, n = 20) to 3.0% (high-activity samples, n = 20), and the between-batch CV was 5.1% (n = 41). Plasma AFU activity was not affected by storage at 4 °C for 48 h and at -20 °C for 6 months. Analytical recovery was studied by the addition of serial dilutions (0.15–15 U/L) of human placental AFU to a plasma pool. The mean recovery of AFU activity was 101% (97–107%). We found no interference with bilirubin up to 500 µmol/L or with hemoglobin up to 1.5 g/L. Beyond these concentrations, bilirubin and hemoglobin reduced fluorescence intensities. Assay of AFU activity was unaffected by opalescent solutions (triglycerides >10 mmol/L). Plasma AFU activities measured in 41 samples from different healthy individuals correlated with those determined with the adapted colorimetric method of Zielke et al. (4): y = 3.003x + 0.240; r = 0.914.
The frequency distribution of plasma AFU values exhibited positive
skewness and suggested a trimodality (Fig. 1
). Taken all together, the mean (± SD) plasma AFU value for
healthy adults was estimated at 4.73 ± 2.42 U/L (0.66–10.53
U/L). Using the FASTCLUST procedure, we partitioned plasma AFU into the
three activity ranges described in Fig. 1
. Fifteen percent of
individuals expressed low activity (1.23 ± 0.47 U/L), 45%
expressed medium activity (3.69 ± 0.82 U/L), and 40% expressed
high activity (7.27 ± 1.20 U/L).
|
The use of AFU as a routine clinical enzymology marker requires optimization of the assay reaction conditions and improvement of its reliability and practicability by automation. The conditions that provided maximal activity (0.1 mol/L citrate-phosphate, pH 5.0, 0.25 mmol/L 4-MUF, 15-min incubation time at 37 °C) were close to those used by Lombardo et al. (6) on a continuous flow analyzer. The sensitivity of the fluorimeter of the Monarch 2000 centrifugal analyzer allowed us to reduce incubation time and plasma sample volume (1:10 dilution). The use of rotors equipped with microcups (280 µL total volume) allowed us to reduce the volume of the expensive selected substrate (4-MUF) 10-fold. Automation by Monarch 2000 yielded precision near that obtained by the AFU colorimetric measurement with a Cobas Bio centrifugal analyzer (10). In the physiologic range, the fluorometric assay correlated with the colorimetric assay, but the high specific activities obtained with the 4-MUF substrate yielded AFU values threefold higher than those obtained with the 4-NPF substrate. The fluorometric detection also yielded a good accuracy without the interferences commonly observed with colorimetric determinations. The micromethod we describe can be applied to other discrete automated fluorometric analyzers and fluorescence microplate automated systems.
In this study, we confirm the existence of a plasma AFU activity polymorphism with a trimodal distribution (9)(11). Median values calculated for low (variant), medium, and high activities were close to those obtained by Wood (9) from a healthy student population. However, because of the strict selection of the French adult population studied, the upper limit of "normal" values in the present study was 1.4-fold lower. The anonymous collection of plasma did not allow us to study age or gender as variables. However, previous studies (6)(12) detected no gender differences or significant age variations in subjects >25 years of age.
We conclude that automation of the fluorometric AFU method greatly improved sensitivity, specificity, and practicability of the assay. This test, therefore, can be used in large-scale screening for fucosidosis where values are lower than the low variant (13). However, the trimodal distribution of plasma AFU activity limits the diagnostic value of this test for acquired diseases. Nevertheless, as suggested by Giardina et al. (3), determination of plasma AFU activity may be very useful in the follow-up of cirrhotic patients for the early detection of hepatocellular carcinoma.
References
-L-fucosidase and ß-N-acetylglucosaminidase activities in diabetic, cirrhotic and gastric cancer patients. Clin Chim Acta 1980;103:155-158.
-L-fucosidase: a new marker for the diagnosis of primary hepatic carcinoma?. Hepatology 1984;4:889-892.
-L-fucosidase activity and early detection of hepatocellular carcinoma. Cancer 1998;83:2468-2474.
-L-fucosidase. J Lab Clin
Med 1972;79:164–9..
-L-fucosidase. Clin Chim Acta 1975;58:251-256.
-L-fucosidase. Hum Genet 1981;59:115-118.
-L-fucosidase: a common polymorphic variant for low serum enzyme activity, studies of serum and leukocyte enzyme. Hum Hered 1979;29:226-229.
-L-fucosidase with a centrifugal analyzer. Clin Chem 1986;32:1588.
-L-fucosidase (FUCA2) and the plasminogen (PLG) system. Clin Genet 1984;26:23-29.
-fucosidase activity during childhood and pregnancy. Clin Chim Acta 1990;187:273-280.
-L-fucosidase. Am J Hum Genet 1976;28:42-50.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
