| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Departments of
1
Biochemistry,
2
Pathology, and
3
Urology, Queen’s University and Kingston General Hospital, Kingston, Ontario, K7L 3N6 Canada
aaddress correspondence to this author at: Department of Pathology, Richardson Laboratories, Queen’s University, Kingston, Ontario, K7L 3N6 Canada; fax 613-533-2907, e-mail collier{at}cliff.path.queensu.ca
Testosterone is the principal circulating androgen in men. A major fraction of testosterone is specifically bound with high affinity and low capacity to sex-hormone-binding globulin (SHBG), whereas most of the remaining testosterone is bound with low affinity to albumin (ALB), leaving only 1–2% to circulate as "free" testosterone (FT) not bound to protein in serum (1). Because the SHBG–testosterone complex dissociates very slowly in vitro (dissociation half-time, 20 s), SHBG-bound testosterone (SHBG-T) is not considered to be available for biologic action in tissues. According to the original "free hormone hypothesis", only FT is able to enter target cells. Tait and Burstein (2), however, suggested that because testosterone is only loosely bound to ALB, this fraction is also available for tissue uptake (dissociation half-time, 1 s). The concept of bioavailable testosterone (BAT), also known as "non-SHBG-bound testosterone" [or both FT and ALB-bound testosterone (ALB-T)], was supported by in vitro kinetic calculations (3), but others (4)(5) have challenged the assumptions used for these calculations. Indeed, if SHBG facilitates entry of testosterone into tissues, as reported by Pardridge (6), then SHBG-T would also be "available" for tissue uptake and action. Controversy remains about the best biochemical measurement for testosterone function in humans.
The initial clinical diagnostic evaluation for men with symptoms of hypogonadism involves measurement of the serum total testosterone. In situations such as hyper- and hypothyroidism, obesity, and the use of antiepileptic drugs, the correlation between clinical symptoms and the serum total testosterone concentration has been reported to be poor. This may be attributable in part to the inactivity of the SHBG-T fraction and to variation in the plasma concentration of SHBG in these situations. The fact that plasma SHBG changes during life and in several pathologic situations is well established (7)(8). It is currently common practice to also measure FT or BAT during these investigations (9)(10).
BAT assays are based on the selective precipitation of SHBG-T by ammonium sulfate (AS), leaving ALB-T and FT (i.e., BAT) in the supernatant. The original assays pretreated the serum with [3H]testosterone before AS precipitation. BAT was then calculated from the percentage of the tracer in the supernatant multiplied by the total testosterone concentration, which was usually measured by RIA (with or without an extraction step) (3)(9). A novel approach used extraction and chromatography of the precipitate before RIA (11). The development of a standard BAT assay is required because current reports suggest that BAT can range from 20% to 70% of total testosterone for men (3) and 10% to 34% for women (3)(12).
We investigated the measurement of BAT using the Bayer ImmunoTM 1 immunoassay system (an enzyme immunoassay; Bayer Corporation, Diagnostics Division) after precipitation with AS. Serum samples collected from 48 men during routine clinical investigation or monitoring of hypogonadism at the Kingston General Hospital Urology Clinic were stored at -20 °C until analysis. Total testosterone was measured on the Immuno 1 System with between-assay (total) imprecision (CVs) of 14% and 6.7% for 2.68 nmol/L and 53.39 nmol/L, respectively. This "competitive magnetic separation immunoassay" is performed by incubating (37 °C) the sample with anti-testosterone antibody, followed by addition of the "testosterone enzyme conjugate" (i.e., the analog, which competes with the testosterone in the sample for the binding sites on the antibody). After the addition and incubation with the monoclonal Immuno Magnetic particle (which binds to the antibody complex), the monoclonal Immuno Magnetic particle–antibody complex is washed, the enzyme substrate is added, and product formation is measured. Thus, the signal is inversely proportional to the testosterone concentration in this assay for total testosterone. The FT (Coat-A-Count Free Testosterone Assay; Diagnostic Products Corporation) was measured at Hospital In-Common Laboratory (Toronto, Canada). This study was approved by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board.
For the BAT assay, a saturated solution of enzyme-grade AS was prepared by dissolving 70 g in 100 mL of distilled water at 70–80 °C, followed by filtration at the same temperature. The solution was stored at room temperature. Serum samples were thawed, vortex-mixed, and allowed to warm to room temperature for 1 h. An equal volume of AS solution was slowly added to the serum samples during vortex-mixing. This was done to minimize any local excess concentrations of AS that might lead to the "microprecipitation" of ALB and SHBG (11). After refrigeration at 4 °C for 1 h, the samples were centrifuged at 1100g for 5 min. BAT was then measured as testosterone in the supernatant. Total testosterone was measured on a 1:1 dilution of serum with diluent supplied for Bayer Immuno 1 assays and used for the calculation of recovery and percentage of BAT. Results are presented as the mean ± SD.
Treatment of serum with saturated AS (final solution was 50% saturated AS) led to a mean recovery of 4.8% ± 1.6% testosterone in the supernatant. Because this was much lower than expected, the various reagents in the method were investigated with a protein-free solution. We found that the presence of either saline or AS in the final sample increased testosterone results by the Immuno 1 system. We speculate that AS in particular interferes with one of the reagents (possibly one of the antibodies) used in the immunoassay. Interference of AS in the final measurement of testosterone has not been reported previously in BAT assays because the previous reports used isotope detection in contrast to the enzyme immunoassay used in our method.
To eliminate this interference by AS, we decided to determine BAT indirectly by measuring testosterone in the precipitate after AS pretreatment. The precipitate was reconstituted in Bayer Diluent-B Reconstitution Buffer and then assayed for testosterone. A control sample for total testosterone was measured by 1:1 dilution of serum with diluent to control for any minimal interference in measurement of testosterone by the diluent. With the saturated AS solution, testosterone in the precipitate was >90% of the total testosterone, indicating that ALB-T was also precipitating or that testosterone was being displaced from ALB and then binding to SHBG. Indeed, ALB was detected in the precipitate by the Synchron CX7 Clinical System (Beckman Coulter, Inc.).
Confirmation that the concentration of AS being used was also precipitating ALB led to a study to optimize the AS concentration. Various concentrations of AS were added to equal volumes of serum, and the precipitate was dissolved in the diluent. Both testosterone and ALB were then measured. The recovery of testosterone in the precipitate increased as a function of AS concentration (Table 1
). In particular, large increments in the recovery of testosterone were evident at AS saturations of 65% and 75%. However, experiments at 73% and 75% AS saturation, revealed that 
6 g/L ALB was typically precipitated, whereas ALB was undetectable (<10 µg/L) at 68% and 65% AS saturation. It appears that 70% AS saturation was a threshold because of the greater degree of variation in results for samples treated with this concentration (within-assay CVs, 13–21%; Table 1
). Decreasing the AS concentration by only 2%, from 70% to 68% saturation, produced results with moderate within-assay CVs (5–7%) and no precipitation of ALB. A similar approach of reconstituting the AS precipitate was used by Ratajczak et al. (12) in 1981, followed by dichloromethane extraction and measurement of testosterone by RIA. They reported that at a final AS saturation of 41%, the supernatant was completely free of SHBG, whereas no ALB was found in the precipitate. A clear conclusion of the work reported here is the sensitivity of the BAT assay to AS concentration and ALB precipitation. Each laboratory should establish the critical concentration of AS that does not precipitate ALB.
|
In 48 samples from men undergoing routine clinical investigation or monitoring of hypogonadism, the mean percentage of testosterone recovered as BAT (68% AS saturation) was 62.1% ± 8.6%, with the percentage of BAT being independent of the total testosterone concentration. In 20 assay precipitates, no ALB was detected. Previous reports have indicated BAT percentages of 20–70% (3), 30–80% (13), and 45% (9) in healthy men, whereas older men have slightly lower percentages of 30%, and impotent men have percentages of 20% (9). Although we did not perform a method comparison of our BAT method with another non-SHBG-bound-T method in this current study, we did include this in a subsequent study of 119 men and 21 women who were being treated or investigated for hypogonadism (unpublished data). Our BAT method demonstrated good correlation not only with total testosterone and FT, but also with FT as calculated by Vermeulen et al. (14), which was based on the measurement of testosterone, ALB, and SHBG.
The high correlation between BAT and testosterone and the good correlation between FT and testosterone (Fig. 1
) are consistent with data reported by Winters et al. (13) in 28 thin or obese, but otherwise healthy, men in which the correlation between FT and testosterone was slightly better than the correlation between BAT and testosterone. Several studies have noted good correlations between BAT, FT, calculated FT, and total testosterone (7)(12)(13). If BAT continues to demonstrate good correlation with total testosterone for other patient groups and in studies with substantially larger study groups, the question must be raised whether BAT adds any useful information to the total testosterone concentration or if it is a redundant test. The same issue has recently been raised for FT (13).
|
In summary, with the Immuno 1, BAT cannot be measured accurately on the supernatant, which contains AS, but can be measured indirectly in the precipitate after reconstitution with the manufacturer’s diluent. The optimal concentration of AS for the selective precipitation of SHBG was 34% AS saturation. AS solution was added slowly during vortex-mixing to minimize the microprecipitation of ALB, and samples were kept at 4 °C for 1 h after AS precipitation before the final centrifugation was performed. The analytical reliability of selective SHBG-T precipitation with AS needs to be carefully investigated when establishing BAT assays. The very good correlation between BAT and total testosterone (and between FT and total testosterone) raises the question of whether these newer tests add any useful clinical information to total testosterone results. To define the usefulness of these new tests, large studies on a variety of well-characterized patient populations are needed.
Acknowledgments
We thank Dr. Ekins for insight into the historical development of free hormone and BAT assays. Support and encouragement from Bayer Corporation in Canada were very much appreciated, as was the assistance of Mary Waugh and the staff of the Kingston General Hospital Core Laboratory.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  W. de Ronde, Y. T. van der Schouw, H. A.P. Pols, L. J.G. Gooren, M. Muller, D. E. Grobbee, and F. H. de Jong Calculation of Bioavailable and Free Testosterone in Men: A Comparison of 5 Published Algorithms Clin. Chem., September 1, 2006; 52(9): 1777 - 1784. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
