NOTE 
Annals of Nuclear Medicine Vol.4, No.1, 39-42, 1990 
Methodological aspects for free hormone estimation using microencapsulated antibody method - The effects of hormone binding protein on permeability of microcapsule membrane 
Takuma HASHIMOTO Kaichiro ISHIBASHI** and Fujitsugu MATSUBARA* 
* Department of Laboratory Medicine, Kanazawa University School of Medicine, Kanazawa, 920, Japan **Eiken Immunochemical Laboratories Tokyo 114, Japan 
The effect of T4 non-carrying thyroxine binding globulin (TBG) on free thyroxine determination using the microencapsulated antibody method was studied, to investigate the precise reliability of the membrane and to find possible applications for estimating other free steroid hormones. When increased amounts of purified TBG were added to a test tube containing microcapsule suspension, it affected the accuracy of the results. We found that with higher amounts, 125I-T4 leaked through the membrane into the medium, thereby giving a falsely increased free T4 result. Our finding indicates that further improvements in the microcapsule membrane are necessary; or alternatively, it may also be possible to balance the binding affinity inside and outside the membrane by adding a suitable amount of carrier protein, to the contents of the capsule, so that both successful FT4 determination and other free steroid hormone assays may be undertaken. 
Key words : Free thyroxine, Microencapsulated antibody method, Free hormones 
INTRODUCTION 
THE ESTIMATION of free hormone levels in the blood, when the hormone has its binding protein, is remarkably accurate when compared to the total hormone estimation. In fact, free thyroxine (FT4) determination is now a routine procedure for the evaluation of thyroid function.1-4 In a report by Chopra and Tulchinsky5 on the pathogenesis of gynecomastia in hyperthyroid men, the estimation of free estrogen and free testosterone was the most important tool in clinical investigation. If a sirnple and appropriate method, similar to FT4 determination in the thyroid disease, were available, it would be very beneficial in determining other endocrine disorders caused by free hormone imbalance. Buehler6 has reported using the microencapsulated antibody method in determining FT4 with radio-immunoassay (RIA) and showed it to be applicable for assaying free cortisol and free testosterone. This 
Received June 7, 1989; revision accepted November 10, 1989. For reprints contact : Takuma Hashimoto, Department of Laboratory Medicine, Kanazawa University, School of Medicine, 13-1, Takara-machi, Kanazawa, 920, JAPAN. 
unique microencapsulated antibody method seemed to be attractive, practicable and promising for several free hormone estimations, but it did not become a commercial success. In order to make further improvements in technology, at first we considered it important to examine the reliability of the microcapsule membrane. In this paper we describe our assessment of some methodological problems in the microencapsulated antibody method used in the FT4 RIA kit, which will have important significance when a successful assay method for free testosterone or free cortisol is developed in the future. 
MATERIALS AND METHODS 
Experimental procedures : 
Preparatron o TBG 
Highly purified TBG was prepared from pooled human serum by affinity chromatography as previously described.7 The technique employed was similar to the method described by Pensky and Marshall.8 1251 thyroxine was added to the purified TBG and the mixture was subjected to disc electrophoresis. A single protein band coincided with the radioactivity. The TBG did not contain any T4 after purification. The purified TBG zero serum contains a concentration of 4.0 g/dl albumin, 10.1 mg/dl pre-albumin and 0.54 mmol/l of nonesterified fatty acids (NEFA). 
Assay procedures 
The assay for FT4 using the microcapsule method (Liquisol FT4 kit Damon Diagnostics, USA) was performed in accordance with the manufacturer's instructions. Briefly, twenty-five pl of saline or purified TBG (O, 13, 52 ,ug/ml) was added to test tubes containing 500 pl of microcapsule suspension which contained anti-T4 combined with 125I-T4. Twenty-five pd of standard solution (0, 0.15, 0.38, 1.10, 2.00, 3.10 and 5.00 ng/dl) was pipetted into each tube and the tubes were vortexed for 4 seconds. After the tubes were incubated in a water bath at 37 deg.C for one hour, they were vortexed for 4 seconds, and then incubated for an additional one hour. After that, one ml of washing solution (polyethyleneimine) was added to each tube followed by 4 seconds vortex. The tubes were incubated at room temperature for 20 minutes, and subsequently centrifuged at 1,400 g for 10 minutes. After decanting the supernatant, the sedimenting microcapsules were counted in a gamma counter for 1 minute. The intra- and inter-assay coefficients of variation were 7.2-8.5 % between 0.40 and 4.50 ng/dl and 14.7-15.9 % between 1.00 and 3.00 ng/dl respectively. FT4 was also measured with an Amerlex kit (Amersham, U.K.); T3, T4, TSH, thyroglobulin (TG), TBG with an EIKEN kit (EIKEN ICL, Japan), and T3 RSU with a Dainabot kit (Dainabot Co. Japan). The accuracy of these kits and the analytical procedures for measurement of these hormones have previously been described in detail.9,10 All samples were measured simultaneously in order to avoid inter-assay variations. 
Patient materials 
Seventeen patients receiving T3 replacement therap: after total thyroidectomy (T4 less than 1 ,ug/dl) wen studied. Sera were aliquoted and stored at -20 deg.C until the assay was undertaken. 
Statistical Analysis 
The results were expressed as the mean+-standard deviation (SD). The significance of differences between mean values was evaluated by the paired t-test. 
RESULTS AND DISCUSSION 
Figure 1 . illustrates the effect of varying doses of purified TBG from O to 52 ,ug/ml on a Liquisol FT4 
kit. The results show that the FT4 values determined by this method are strongly influenced by the concentration of TBG in the samples : when the added purified TBG concentration outside of the capsules was high, 125I-T4 was extracted from the microcapsules. Because of this leakage, the FT4 standard curve can no longer be distinguished at a concentration above 13 ug/ml of the purified TBG. Figure 2. is a schematic diagram of the effect of uncomplexed TBG on FT4 determination using the microencapsulated antibody method. The difference between the 125I radioactivity in the microcapsules dissolved in saline and those dissolved in the purified zero TBG solution might be mostly due to the fact that the TBG solution contains human albumin and pre-albumin, which affected the leakage of the encapsulated 125I-T4.n Each standard solution in the Liquisol FT4 kit contains approximately 3.0 g/dl of human albumin and 17.8 ug/dl of TBG. The reason for the distinctive and exceptional decline in Liquisol FT4 zero serum is inexplicable. Whatever the reason, the difference between the results in the saline and the purified zero TBG solution indicate, that saline should not be used for dilution. As we10 and Melmed et al.12 reported previously, the values obtained with the Liquisol kit were significantly higher than those expected in patients with primary hypothyroidism. In this study, as shown in Table I ., this kit gave erroneously higher concentrations, in contrast with the Amerlex kit, in patients undergoing T3 treatment after total thyroidectomy. Previously, the microencapsulated antibody method used in FT4 RIA was reported to be applicable for the assay of free fractions of cortisol or testosterone by Buehler.6 However, plasma free cortisol or testosterone determined by this principle could also be affected by the binding protein for each hormone in a similar manner. It has been reported by Key and Moorel3 and Slaats et al.14 that sex-hormone binding globulin (SHBG) has interfered with estradiol and testosterone estimation in a non-extraction RIA method. Although many investigators have reported various methods for measuring free aldosterone,15 free cortisol,16 free deoxycorticosterone,17 free estrogens,18 and free testosterone,19 the methods for their assays are complicated, time consurning and not practical in clinical use. Despite the methodological problems which are not completely solved yet, the microencapsulated dialysis method appears to have the potential to be a sensitive, flexible and practicable method for the determination of various free hormones. In addition, this method is principally a miniature dialysis system and there is a consensus that free hormones determined by equilibrium dialysis should be a reference standard against which other measurements are compared.20 In conclusion, further improvements of the microcapsule membrane and/or addition of carrier protein to the capsule, to balance the binding affinity inside and outside the membrane, are necessary for the determination of FT4 and other free steroid hormones. If further improvements in this area of technology are able to be made, we will have an easier and more precise method of estimating all free hormones. 
ACKNOWLEDGEMENTS 
We are indebted to Dr. Ellen Ingham for encouragement, helpful discussions, and comments. We also thank Miss Misako Yagi and Miss Yasuko Kamei for their excellent secretarial assistance and to Miss Hermione Elliot for her help in English in the preparation of this manuscript. This work was supported in part by the Japan Clinical Pathology Foundation for International Exchange and Clinical Pathology Foundation of Japan. 
REFERENCES 
1. Hashimoto T: Tests of thyroid function. Labmedica 3: 13-20, 1986
2. Amino N, Nishi K, Nakatani K, et al : Effect of albumin concentration on the assay of serum free thyroxin by equilibrium radioimmunoassay with labeled thyroxin analog (Amerlex free T4). Clin Chem 29: 321-325, 1983
3. Wang, YS, Hershman JM, and Pekary AE: Improved ultrafiltration method for simultaneous measurement of free thyroxin and free triiodothyronine in serum. Clin Chem 31: 517-522 1985
4. Kaptein EM, Macintyre SS, Weiner JM, et al: Free thyroxine estimates in nonthyroidal illness: Comparison of eight methods. J Clin Endocrinol Metab 52: 1073-l077, 1981
5. Chopra IJ, and Tulchinsky D : Status of estrogenandrogen balance in hyperthyroid men with Graves' desease. J Clin Endocrinol Metab 38 : 269-277, 1974
6. Buehler RJ: Applications of microencapsulated anti-body in free hormone radioimmunoassays. In Free Hormones in Blood, Albertini A, and Ekins RP, (eds), Amsterdam, Elsevier Biomedical Press, pp 121-127 1982
7. Miyai K, Ito M, and Hata N : Enzyme immunoassay of thyroxin-binding globulin. Clin Chem 28 : 2408-2411 1982
8. Pensky J, and Marshall JS : Studies on thyroxine-binding globulin (TBG): Separation from human serum by affinity chromatography. Arch Biochem Biophys 1 35 : 304-310, 1969
9. Hashimoto T, Migita S. Matsubara F: Response of thyrotropin and free thyroid hormones to graded exercise in normal male subjects. Endocrinol Japon 33: 735-741 1986
10. Hashimoto T, Okada T, Nagahara M, et al : The effect of thyroxine (T4) unattached thyroxine binding globulin (TBG) on free T4 determination. In Current Problems in Thyroid Research, Ui N, Torizuka K, Nagataki S, and Miyai K, (eds), Amsterdam, Excerpta Medica, pp 279-282, 1983
11. Robins J, Edelhoch H : Hormone transport in blood. Chapter 6. Thyroid hormone transport proteins : their nature, biosynthesis, and metabolism. In Werner's The Thyroid (5th ed), Ingbar SH, Braverman LE, (eds), Philadelphia, JB, Lippincott Co., pp 116-127, 1986
12. Melmed S, Geola FL, Reed AW, et al : A comparison of methods for assessing thyroid function in non-thyroidal illness. J Clin Endocrinol Metab 54 : 300-306 1982
13. Key TJ, and Moore JW: Interference of sex-hormone binding globulin in a non-extraction double-antibody radioimmunoassay for estradiol. Clin Chem 34: 1357-1358, 1988
14. Slaats EH, Kennedy JC, Kruijswijk H : Interference of sex hormone binding globulin in the "Coat-a-Count" testosterone non-extraction radioimmunoassay. Clin Chem 33 : 300-302, 1987
l5. Vecsei P, Haack D, Karl-Heins K, et al : Comparative diagnostic evaluation of the radioimmunoassays of zona glomerulosa hormones : Aldosterone and 18-OH-corticosterone in urine. Basic Clin Endocrinol 1 : 279-292, 1981
16. Serio M, Bassi F, Salerno R: The diagnostic value of urinary free corticosterroids. In Free Hormones in Blood, Albertini A, and Ekins RP, (eds), Amsterdam, Elsevier Biomedical Press, pp 363-368, 1982
17. Cope CL, Loisou S : Deoxycorticosterone excretion in normal, hypertensive and hypokalemic subjects. Clin Sci Mol Med 48 : 97-105, 1975
l8. Oakey RE : Urinary excretion of non-conjugated oestrogens during the menstrual cycle. J Obstet Gynec 80: 794-797, 1973
19. Ghielmi S, Gastaldi C, Temponi M, et al : Free testosterone assay by equilibrium dialysis and column adsorption methods. In Free Hormones in Blood, Albertini A, Ekins RP, (eds), Amsterdam, Elsevier Biomedical Press, pp 369-381, 1982
20. Surks MI, Hupart KH, Pan C, et al : Normal free thyroxine in critical nonthyroidal illnesses measured by ultrafiltration of undiluted serum and equilibrium dialysis. J Clin Endocrinol Metab 67 : 1031-1039, 1988