| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Laboratoire de Biochimie and
2
Service d'Hépato-Gastroentérologie, Hôpital Jean Verdier, Avenue du 14 juillet, 93143 Bondy, France;
a author for correspondence: fax 33 01 4802 6831, e-mail jenny.vaysse{at}jvr.ap-hop-paris.fr
The sex steroid-binding protein, also called sex hormone-binding globulin (SHBG), is an androgen- and estradiol-carrier protein in human serum. The two identical subunits of this glycoprotein both bear three oligosaccharide chains, two N-linked at Asn and Asn and one O-linked at Thr. The role of this carbohydrate moiety in the biological function of SHBG is not known; however, the complete enzymatic deglycosylation of human SHBG has very little effect on the steroid-binding activity (1). Because estrogens regulate SHBG synthesis, this protein is usually measured to evaluate estrogen status, especially in the biochemical assessment of hirsutism. Other indications for SHBG measurement are promising because high SHBG concentrations are associated with a decreased risk of breast cancer in postmenopausal women (2) and with an increased incidence of hepatocellular carcinoma in patients with cirrhosis (3). Furthermore, in the latter situation, it has been suggested that increased androgen transport through the hepatocyte membrane, via the asialoglycoprotein receptor, might be involved in hepatic carcinogenesis (3). Determination of desialylated SHBG (dSHBG) is needed to investigate this hypothesis and could be useful in tracking this disease.
Total SHBG is usually measured with a radiometric method using tritiated dihydrotestosterone or with immunoradiometric assay (4)(5). To our knowledge, procedures for dSHBG determination have never been described. The present study was thus designed to propose two easy ELISA methods for the parallel measurement of SHBG and dSHBG.
Total SHBG was determined with a classical ELISA sandwich assay. Calibration was performed with purified human SHBG obtained from Calbiochem or Eurodiagnostica. Controls were run with sera from Eurodiagnostica and with dSHBG prepared by treatment of SHBG obtained from Calbiochem with agarose-linked Clostridium perfringens neuraminidase (EC 3.2.1.18) from Sigma (6). The sera, collected from in-patients included in a screening program (3) and from healthy donors, were stored at -25 °C. Informed consent was obtained, and blood sampling was performed in accordance with ethical standards. The microtiter plates (Nunc) were coated overnight at 8 °C with 100 µL of a 1:500 dilution of a 3.4 g/L purified immunoglobulin fraction of rabbit anti-SHBG serum (Dako) in 17 mmol/L phosphate buffered saline, pH 7.4 (PBS). The contents of the wells were removed and saturation of each well was achieved with 400 µL of 10 g/L bovine serum albumin (BSA) in PBS (BSA/PBS) incubated for 1.5 h. After the wells were washed (for each wash session, the wells were washed six times with PBS containing 0.5 mL/L Tween 20), 100 µL of each specimen diluted in BSA/PBS was added. Calibrators in concentrations ranging from 0 to 60 µg/L were used. For serum samples, two dilutions were tested: 1:546 and 1:1066. The plates were incubated for 1 h, and then washed. After the addition of 100 µL of 1.3 g/L peroxidase-conjugated anti-SHBG immunoglobulins (Dako) diluted 1:2000 in BSA/PBS, the plates were incubated for 1 h. After the wells were washed, 100 µL of substrate (0.1 g/L o-phenylenediamine in sodium citrate, pH 4.5, with 30 mg/L H202) was added, and the plates were incubated for 30 min. The reaction was stopped with 100 µL of 0.5 mol/L sulfuric acid, and the absorbances were measured at 492 nm with a nonspecific estimate at 620 nm on a LP400 microplate reader (Sanofi Pasteur). Except for the coating step, all incubations were done at room temperature under gentle continuous shaking. Each determination was performed in duplicate. The absorbances of blanks without coating antibody run simultaneously were subtracted. The data were analyzed after fitting with the Levenberg-Marquardt algorithm and processed with the KaleidaGraph (Ver. 3.0) package from Synergy Software.
Blank absorbances were <0.02. Within-batch imprecision was established
on 70 samples analyzed in duplicate. The CV in the concentration range
0–75 µg/L was 4.9%. The between-batch CV, determined with controls
over 10 assays, was 8%. The range of values found for the healthy
donors was similar to that reported by other authors
(4)(7). A comparison of the concentrations
obtained with the described ELISA-SHBG and with RIA-SHBG
(Eurodiagnostica) showed a linear regression: y =
1.17x 0.165, where y is the ELISA value and
x is the RIA value; r = 0.994, n = 30,
Sy|x=0.81 µg/L. The sialylation rate of SHBG
did not affect the immunological reactivity of this protein because no
substantial differences could be seen in the calibration curves
obtained either with native SHBG from both origins or with dSHBG (Fig. 1
A.).
|
The determination of dSHBG, or more accurately of sialic acid-deficient SHBG, was initiated according to Tertov et al. who recently described a procedure for desialylated LDL determination (8). Because removal of sialic acid leads to the exposure of galactosyl residues, these authors first captured asialo-apolipoprotein B with Ricinus communis (RCA120; Sigma), an immobilized lectin that has affinity for galactose, then measured the protein with a peroxidase-conjugated apolipoprotein B antibody. In our first attempt, we applied this procedure to the determination of dSHBG. Experimental conditions were similar to those described for SHBG, except that (a) the plates were coated with a 18 mg/L RCA120 solution instead of an anti-SHBG immune serum, (b) 17 mmol/L phosphate buffer without saline (PB), pH 7.4, was used, and (c) in vitro neuraminidase-treated SHBG was used as a standard. This low ionic strength PB was suitable because of the relatively low lectin-binding affinity. When applied to dSHBG calibrators, this method gave absorbances that correlated well with dSHBG concentrations. However, when the method was used with sera, the absorbances were too low to give reliable results, probably because dSHBG is only a minor component among the bulk of serum compounds able to bind to RCA120. Although this procedure is not suitable for dSHBG determination in serum, it can help evaluate the efficiency of the neuraminidase treatment on SHBG standard solutions; complete desialylation is obtained when the difference of absorbances between SHBG and dSHBG reaches a maximum, i. e., after 30–60 min incubation at 30 °C. Thus, a 1-h treatment with neuraminidase was carried out to prepare the dSHBG standard.
Because the Tertov et al. procedure proved to be unsuitable for the
dSHBG measurement in serum, a reverse sandwich assay was performed that
involved coating the wells with 100 µL of 1:500 anti-SHBG in PB and
additional detection of the bound dSHBG with 100 µL of 0.5 mg/L
peroxidase-linked RCA120 in BSA/PB. Because this
lectin can bind to immobilized immunoglobulins through their
carbohydrate moiety, we ran a blank test in which no serum was added;
the absorbances of these blanks, always <0.15, were acceptable.
Concentrations of the dSHBG standard solutions ranged from 0 to 500
µg/L (Fig. 1B
). Serum samples were usually diluted 1:286 and 1:546 in
BSA/PB; however, for some patients, higher dilutions had to be tested.
All determinations were done in duplicate, and data were analyzed as
previously described for total SHBG.
The within-run CV for duplicates was ~7% (n = 173), and mean analytical recovery (overload = 30 µg/L) was 105% (SD = 17%, n = 40). Because the in vivo desialylation process may affect the glycan chains differently, serum probably contains several SHBG isoforms with various sialic acid or terminal galactosyl residues contents and thus with different reactivities toward RCA120. Because of this heterogeneity in the serum isoform profile, the use of in vitro neuraminidase-treated SHBG as a standard is questionable. Serum dSHBG concentrations were thus expressed as arbitrary units, one unit being defined as the absorbance obtained with a 1 µg/L dSHBG calibrator. In healthy donors, values range from 9 to 12 U/L, whereas in some patients with hepatic diseases, concentrations may rise to 100 U/L.
SHBG and dSHBG can be determined in parallel easily using the described procedures because most of the steps are similar: coating with anti-SHBG antibody, BSA saturation, conditions for washing and incubation, enzymatic amplification, photometric measurement, and data analysis. Because total SHBG determination can be measured in a low ionic strength buffer, i.e. PB, the two assays differ only by (a) the nature of the peroxidase-conjugate (anti-SHBG for total SHBG and RCA120 for dSHBG), (b) the serum dilution, and (c) the standard for calibration. The convenience of this coupled procedure allows extensive studies of various pathological conditions, such as hepatocellular carcinoma, and may thus be useful for a better knowledge of the physiopathology of this disease.
Acknowledgments
This research was supported in part by the ARC (Paris) project no. ARC 6492–94.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), Eurogenetics (
).
The desialylation of SHBG (
) did not modify the signals.
(B) dSHBG prepared from Calbiochem (