0021-972X/78/4704-0746$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society
Vol. 47, No. 4 Printed in U.S.A.
Radioimmunoassay of Reverse Triiodothyronine* Veterans Administration Hospital, Jefferson Barracks and Washington University, St. Louis, Missouri 63125 ABSTRACT. rT 3 RIA standard curves were generated using rT 3 standards from various sources. The rT 3 antiserum and the method of free and bound hormone separation used were the same in all assays. The slopes of rT3 RIA curves differed considerably; standard curves generated with Henning (Berlin) rT 3 standard had the steepest slope. Because of the differences in slopes of standard curves, bound to free estimates, when extrapolated from standard curves generated with various rT 3 standards, resulted in 2- to 3-fold differences in rT 3 assay values in the same sera. Circulating rT 3 concentration in human sera was 20 ng/100 ml when Henning rT3 standard was used in the assay, whereas with Jor-
gensen's rT 3 standard, a value of 42 ng/100 ml was noted. As the rT 3 antiserum and assay conditions used were identical in all comparisons (except rT 3 standards), the 2- to 3-fold differences in rT3 values noted in the same sera suggested variability in potency of rT3 standards. The differences in rT3 values in the same sera (resulting from extrapolation from standard curves using different rT3 standards) were of the same magnitude as that noted in normal human sera in various studies, suggesting that the reported rT 3 variations in sera of euthyroid individuals may also be due to differences in rT3 standards employed in the assay. (J Clin EndocrinolMetab 47: 746, 1978)
T
HE PHYSIOLOGY and metabolism of rT 3 is of current interest. This hormone is predominantly formed in extrathyroidal tissues as a result of tyrosyl or inner ring monodeiodination (5-deiodination) of T4, whereas the extrathyroidal generation of T 3 requires monodeiodination of the phenolic or outer ring of T4 (5' deiodination) (1). Recent investigations suggest that there may be separate deiodinases involved in the inner and outer ring monodeiodination of T4 (2). The measurements of rT 3 and T3 in the circulation and in tissues are of significance because in addition to elucidating the pathways of T4 deiodination, they have served to generate provocative concepts in regard to the peripheral regulation of T3. Clinically, the elevation of rT 3 seen in a variety of systemic illnesses (3), in starvation (4), surgery (5), and in neonates (6), with a concomitant decrease in T3, has given rise to the speculation that the diversion of T4 monodeiodination from the outer to the inner ring (to generate inert rT3) in these states may be protective by counteracting excessive calorigenic effects of T3.
Methods, Results, and Discussion
The reported concentrations of rT 3 in the serum of euthyroid individuals vary from laboratory to laboratory (Table 1). The lowest mean rT 3 concentration noted is 14 ng/100 ml, whereas the highest value of 60 ng/100 ml has been recorded in the studies of Burman et al. (18). The mean rT 3 concentration in the serum, therefore, is not known with sufficient accuracy, so that the daily rT 3 production rates derived from isotopic kinetic computations have been questioned. The variations in circulating rT 3 values in euthyroid individuals are attributed to a variety of factors, such as differences in technical aspects of rT 3 assay that include sensitivity and specificity of the rT 3 antibody, differences in iodine concentration in different geographic locales, age-associated changes in iodothyronine metabolism, nutritional variables, lack of uniformity among investigators in correcting rT 3 values for T4 cross-reaction, etc. A careful inspection of the rT 3 assay values from different studies (Table 1), however, reveals that most investigators who have used rT 3 standards from Henning (Berlin) have reported lower mean values Received November 17,1977. Address requests for reprints to: Dr. B. N. Premachan- than those who have used rT 3 standards from dra, Veterans Administration Hospital 657/151C, St. other sources. This raises the possibility that Louis, Missouri 63125. rT 3 variations in sera of euthyroid individuals * This work was supported in part by the Narveen may also be due to differences in rT 3 standards Medical Research Foundation, St. Louis, MO. 746
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
BHARTUR N. PREMACHANDRA
RIA OF rT 3 TABLE 1. Circulating rT3 concentration (nanograms per 100 ml) in euthyroid individuals, as reported by various investigators Mean
Range"
Visser et al. (7) Ratcliffe et al. (8) Kodding et al. (9) Meinhold et al. (10) Burrows et al. (11) Hufner and Grussendorf (12) Premachandra Griffiths et al. (13) Gavin et al. (14) Premachandra et al. (15) Fisher et al. (16) Vagenakis et al. (4) Nicod et al. (17) B u r m a n e t a / . (18) Chopra (19) Kaplan et al. (20)
14 17 16 18 16 20
9-20 10-27 0-32 2-46 10-28 6-45
20 44 32 42 41 36 45 60 41 23
8-26 28-58 17-47 25-60 19-88C 6-66 24-64 36-84 27-62 7-39
Source of L-rT 3 standards Henning*
Jorgensen
Cahnmann Meltzerrf
" Range as reported by the authors or derived from mean ± 2 SD.
* Berlin GMBH. Age, 10-15 yr. DL-rT 3 .
c
d
employed in the assay. Therefore, we carried out rT 3 assays, as described previously (15), in identical serum samples under the same assay conditions using rT3 from different sources. The same batch of charcoal-stripped iodothyronine-free human plasma was used to make up rT3* standards from various sources; similarly the same shipment of [125I]rT3 (Abbott Labs., North Chicago, IL) immediately after receipt was used in all assays. Antibody to rT3 was produced in rabbits by immunizing
them with rT 3 (Jorgensen)-albumin conjugates. The lowest rT 3 concentration consistently detectable in the assay was 5 ng/100 ml or 5 pg/assay tube. T4 cross-reaction with rT 3 antibody in our assay was 0.04%, which means that even at a serum T4 concentration of 12 /xg/100 ml (upper normal range), rT 3 values in serum would not be overestimated by more than 4-5 ng. The rT 3 standard curves and representative assay values in seven serum samples are shown in Fig. 1 and Table 2. An inspection of these data clearly makes it evident that the differences in rT 3 values noted in the same samples (when assayed with different rT 3 1
rT 3 standards were dried at 40 C for 48 h and stored over a dehydrant under vacuum and all solutions were made in double distilled deionized water.
standards) are of the same magnitude as those noted in various studies referred to previously (Table 1). At the level of 25 ng/100 ml rT 3 standard (Fig. 1), the amount of [125I]rT3 isotopic displacement2 from the antibody (i.e. difference between 0 and 25 ng/100 ml), as noted with Jorgensen, Washington Reference Laboratory, and Henning standards, were 11, 17, and 25 arbitrary U, respectively, whereas between 0 and 50 ng/100 ml, the corresponding displacements were 18,27, and 30 arbitrary U. At higher concentrations (0-200 ng/100 ml), the differences between standards, as manifested in units of isotopic displacement, narrowed with the approaching saturation of the antibody-binding sites. As marked differences in slopes of standard curves generated with different rT 3 standards were apparent, the extrapolation of rT 3 values from different curves resulted in appreciable differences in the same serum samples. The values obtained with Henning and Washington Reference Laboratory standards were closer than the values observed with Jorgensen standard. It should be noted, however, that the values obtained with Jorgensen standard in our assay as well as those of others (Table 1) are similar or even lower than those noted with Cahnmann rT 3 standards. In the one exception, Griffiths et al. (13), using Henning rT 3 standards, reported mean rT 3 values which were even higher than those noted with Cahnmann standards. Finally, Kaplan et al. (20), using DL-rT3 standard, obtained values in normal human serum that were markedly lower than that reported by Chopra (19) who also used DL-rT3; on the other hand, in the studies of Burman et al. (18), the standard curves generated with DL-rT3 had similar slopes, intercepts, and nonimmune binding as L-rT3, but the values reported in human sera were even 2
In our assay, the antiserum is prelabeled with [ I]rT3. rT 3 released from denatured binding sites in serum (effected by TCA-NaOH addition) competes with labeled rT3 for the same binding sites on the antibody and displaces [125I]rT3. This unbound labeled rT3 is separated from the bound hormone by means of a resin sponge which abstracts only the free [128I]rT3 (sponge [125I]rT3 uptake). Also, unextracted sera were used in our rT 3 assays, and previous studies (21) have shown no discrepancies in rT 3 values between ethanol-extracted and native sera. I25
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
Reference
747
PREMACHANDRA
748 80
200
RT3 CONC,(NG/100ML) FIG. 1. rT 3 RIA standard curves generated with Henning (top curve), Washington Reference Laboratory (middle curve), and Jorgensen (bottom curve) rT3 standards. Uptake values shown at each standard concentration represent mean of 10 determinations. Vertical bars at each point represent range of values noted. TABLE 2. rT;) concentration (nanograms per 100 ml) in sera of euthyroid individuals, as read from assay curves using various rT 3 standards (representative data) Normal human
Assay values from curves using 1standards from:
(euthyroid)
Jorgensen
1 2 3 4 5 6 7
20 25 30 40 50 60 65
Washington Ref. Lab." 11 15 17 22 28 33 38
Henning 8 9 11 15 18 20 22
" rT 3 prepared by Dr. M. Rubin of the Washington Reference Laboratory (P.O. Box 3645, Washington, DC).
higher than those reported by Chopra (19). Kaplan et al. (20) strongly emphasized the importance of specificity of antiserum as they obtained 2- to 10-fold differences in values in the same samples when assayed with different antisera, especially the less specific ones. Meinhold and Wenzel (personal communication) observed 2-fold differences in rT 3 concentration in the same serum samples when assayed with rT 3 antisera from several European investigators. In our comparisons, the same antiserum under identical conditions was used in all assays so that the differences in rT 3 values we have recorded in normal human sera (which roughly equal those noted between assays; Table 1) can at least partly be attributed to a nonuniformity in rT 3 standard preparations. Ultraviolet spectrophotometry3 of various rT 3 standards (1 mg/ml in 0.1 N NaOH) did not exhibit significant discrepancies, and at 320 nM, virtually identical readings were obtained with all rT 3 preparations. The differences in rT 3 standards (which were reproducible; Fig. 1) were apparently not due to measurable T 3 or T4 contamination of rT3, as noted from the failure of various rT 3 standards, even at a concentration of 600 ng/100 ml, to react with either T3 or T4 antibody as examined in T 3 and T4 RIAs. It would seem that differences in techniques for the synthesis and purification of rT 3 may contribute to nonuniformity in rT 3 preparations from various sources. It is quite likely that impurities or inert byproduct material in the synthesized rT 3 standards may artifactually give rise to high values, as a larger amount of rT 3 would be needed to effect a certain degree of isotopic displacement from the antibody in comparison to a purer preparation. A common assertion in respect to RIA of hormones is that although RIA values may 3
Dr. Cahnmann (NIH) suggests that determination of absorption profile of rT 3 standards (filtered through 0.22fim filter) from 280 nM and up, instead of measuring just at absorption maximum, is preferable to obtain maximum/minimum (peak/valley) responses for comparison to remove doubts of the presence of soluble inert material in rT 3 standards despite their established purity by thin layer chromatography and gas-liquid partition chromatography.
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
100
JCE&M • 1978 Vol47 • No 4
RIA OF rT 3
While this manuscript was under review, Burrows et al. (11) also reported similar observations on the variability of rT 3 standards. In their studies, rT 3 from Warner Lambert Co. had approximately half of the immunological potency as Henning rT3. In receptor assays, Jorgensen (personal communication) noted that rT 3 prepared by the Shiba-Cahnmann method resulted in relatively high values, about 1% that of L-T 3 , while a sample made essentially by the Henning procedure and purified further by thin layer chromatography gave a value of about 0.2% L-T 3 .
Acknowledgments Grateful acknowledgment is made for the courtesies of Drs. Jorgensen and Cahnmann for the supply of iodothyronine analogs and other valuable comments; Abbott Laboratories, North Chicago, IL, for gifts of radioactive material and other supplies; and Henning, Berlin, for gifts of L-rT3 and T2. It is also a pleasure to acknowledge Dr. Wenzel's critical review of this manuscript and his suggestions and valuable comments.
References 1. CAVALIERI, R. R., AND B. RAPOPORT, Impaired peripheral
conversion of thyroxine to triiodothyronine, Annu Rev Med 28: 57, 1977. 2. CAVALIERI, R. R., L. A. GAVIN, F. BUI, F. MCMAHON, AND M.
HAMMOND, Conversion of thyroxine to reverse-T3 by a soluble enzyme system in rat liver, Abstracts of the 53rd Meeting of the American Thyroid Association, 1977, p. T-ll. 3. CHOPRA, I. J., U. CHOPRA, S. R. SMITH, M. REZA, AND D. H.
SOLOMON, Reciprocal changes in serum concentrations of 3,3',5'-triiodothyronine (reverse T3) and 3,3'5-triiodothyronine (T3) in systemic illnesses, J Clin Endocrinol Metab 41: 1043, 1975. 4. VAGENAKIS, A. G., A. BURGER, G. I. PORTNAY, M. RUDOLPH, J. T. O'BRIAN, F. AZIZI, R. A. ARKY, P. NICOD, S. H. INGBAR,
AND L. E. BRAVERMAN, Diversion of peripheral thyroxine metabolism from activating to inactivating pathways during complete fasting, J Clin Endocrinol Metab 41: 191, 1975. 5. BURR, W. A., E. G. BLACK, R. S. GRIFFITHS, R. HOFFENBERG,
H. MEINHOLD, AND K. W. WENZEL, Serum triiodothyronine
and reverse triiodothyronine concentrations after surgical operation, Lancet 2: 1277, 1975. 6. CHOPRA, I. J., J. SACK, AND D. A. FISHER, 3,3',5'-Triiodothy-
ronine (reverse T3) and 3,3',5-triiodothyronine (T3) in fetal and adult sheep: studies of metabolic clearance rates, production rates, serum binding, and thyroidal content relative to thyroxine, Endocrinology 97: 1080, 1975. 7. VISSER, T. J., R. DOCTER, AND G. HENNEMANN, Radioim-
munoassay of reverse tri-iodothyronine, J Endocrinol 73: 395, 1977. 8. RATCLIFFE, W. A., J. MARSHALL, AND J. G. RATCLIFFE, The
radioimmunoassay of 3,3',5'-triiodothyronine (Reverse T.-)) in unextracted human serum, Clin Endocrinol 5: 631, 1976. 9. KODDING, R., J. JANZEN, E. SCHMIDT, AND R. D. HESCH,
Reverse triiodothyronine in liver disease, Lancet 2: 314,1976. 10. MEINHOLD, H., K. W. WENZEL, AND P. SCHURNBRAND, Ra-
dioimmunoassay of 3,3',5'-triiodo-L-thyronine (reverse T3) in human serum and its application in different thyroid states, Z Klin Chem Klin Biochem 13: 571, 1975. 11. BURROWS, A. W., E. COOPER, R. A. SHAKESPEAR, C. M.
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
not compare favorably from one laboratory to another, it is the relative changes in hormone "* levels (during therapy or other clinical and experimental maneuvers) that are of importance and significance. While this is undoubtedly true in many clinical and other situations, our experience suggests that total reliance on this basis may not always be reassuring and can even be incorrect and misleading. For example, a test serum with rT 3 value of 50 ng/100 ml as read from Henning rT 3 standard curve would read 120 ng/100 ml in Jorgensen's standard curve (Fig. 1). When these levels are compared with the mean values in euthyroid normals, as obtained with these two standards (which is 20 ng/100 ml for Henning and 42 ng/100 ml for Jorgensen; Table 1), the test serum rT3 concentration is 2.5 times (50/20) greater than the mean value in Henning comparison, whereas it is almost 3 times (120/42) y greater than the mean value in Jorgensen's standard comparison. The two relative statements are not exactly comparable. Greater distortions in relative comparisons of rT 3 standards become apparent when they are incorporated in iodothyronine-free plasma from dif7 ferent species. For instance, use of rT 3 from Henning resulted in values in human sera (20 ng/100 ml) which were approximately half of those noted with Jorgensen rT 3 (42 ng/100 ml; Table 1; in assays with human sera all standards were incorporated in iodothyronine-free human plasma), whereas in dog sera, the use of Henning standards resulted in values (50 ng/100 ml) which were approximately onefourth of those noted with Jorgensen's standard (187 ng/100 ml; both Henning and Jorgensen's standards were made in iodothyronine-free dog plasma ) (21). Even more inaccurate and misleading conclusions might become evident if one is measuring rT 3 in sera containing little or no rT3, as in the case of the chicken (15). It is also obvious that data relating to quantitative changes in peripheral conversion of T4 to rT 3 under clinical and experimental conditions may not favorably compare between investigators if they are using rT 3 standards which lead to a 2- to 3-fold difference in rT 3 values in the same serum samples, as can be realized from an inspection of data in Table 2.
749
750
PREMACHANDRA AICKIN, S. FRASER, R. D. HESCH, AND C. W. BURKE, LOW
serum L-T3 levels in the elderly sick; protein binding, thyroid and pituitary responsiveness, and reverse T3 concentrations, Clin Endocrinol 7: 289, 1977. 12. HUFNER, M., AND M. GRUSSENDORF, Radioimmunoassay for
3,3',5'-triiodothyronine (reverse T3) R-T3) in unextracted human serum, Clin Chim Ada 69: 497, 1976. surement of serum 3,3',5'-(reverse) T3, with comments on its derivation, Clin Endocrinol 5: 679, 1976. 14. GAVIN, L., J. CASTLE, F. MCMAHON, P. MARTIN, M. HAMMOND, AND R. R. CAVALIERI, Extrathyroidal conversion of
thyroxine to 3,3',5'-triiodothyronine (reverse-T3) and to 3,5,3'triiodothyronine (T3) in humans, J Clin Endocrinol Metab 44: 733,1977.
i 1978 No 4
children 1 to 15 years of age, J Clin Endocrinol Metab 45: 191, 1977. 17. NICOD, P., A. BURGER, V. STAEHELI, AND M. B. VALLOTTON,
A radioimmunoassay for 3,3',5'-triiodo-L-thyronine in unextracted serum: method and clinical results, J Clin Endocrinol Metab 42: 823, 1976. 18. BURMAN, K. D., R. C. DIMOND, F. D. WRIGHT, J. M. EARLL, J. BRUTON, AND L. WARTOFSKY, A radioimmunoassay for
3,3',5'-L-triiodothyronine (reverse T3): assessment of thyroid gland content and serum measurements in conditions of normal and altered thyroidal economy and following administration of thyrotropin releasing hormone (TRH) and thyrotropin (TSH), J Clin Endocrinol Metab 44: 660, 1977. 19. CHOPRA, I. J., A radioimmunoassay for measurement of 3,3',5'triiodothyronine (reverse T3), J Clin Invest 54: 583, 1974.
15. PREMACHANDRA, B . N . , S. LANG, J. A . ANDRADA, AND J. H.
20. KAPLAN, M. M., M. SCHIMMEL, AND R. D. UTIGER, Changes
KITE, JR., Reverse triiodothyronine in the chicken, Life Sci 21: 205, 1977.
in serum 3,3',5'-triiodothyronine (reverse T3) concentrations with altered thyroid hormone secretion and metabolism, J Clin Endocrinol Metab 45: 447, 1977.
16. FISHER, D. A., J. SACK, T. H. ODDIE, A. E. PEKARY, J. M. HERSHMAN, R. W. LAM, AND M. E. PARSLOW, Serum T
Vol. 47, No. 4 Printed in U.S.A.
Radioimmunoassay of Reverse Triiodothyronine* Veterans Administration Hospital, Jefferson Barracks and Washington University, St. Louis, Missouri 63125 ABSTRACT. rT 3 RIA standard curves were generated using rT 3 standards from various sources. The rT 3 antiserum and the method of free and bound hormone separation used were the same in all assays. The slopes of rT3 RIA curves differed considerably; standard curves generated with Henning (Berlin) rT 3 standard had the steepest slope. Because of the differences in slopes of standard curves, bound to free estimates, when extrapolated from standard curves generated with various rT 3 standards, resulted in 2- to 3-fold differences in rT 3 assay values in the same sera. Circulating rT 3 concentration in human sera was 20 ng/100 ml when Henning rT3 standard was used in the assay, whereas with Jor-
gensen's rT 3 standard, a value of 42 ng/100 ml was noted. As the rT 3 antiserum and assay conditions used were identical in all comparisons (except rT 3 standards), the 2- to 3-fold differences in rT3 values noted in the same sera suggested variability in potency of rT3 standards. The differences in rT3 values in the same sera (resulting from extrapolation from standard curves using different rT3 standards) were of the same magnitude as that noted in normal human sera in various studies, suggesting that the reported rT 3 variations in sera of euthyroid individuals may also be due to differences in rT3 standards employed in the assay. (J Clin EndocrinolMetab 47: 746, 1978)
T
HE PHYSIOLOGY and metabolism of rT 3 is of current interest. This hormone is predominantly formed in extrathyroidal tissues as a result of tyrosyl or inner ring monodeiodination (5-deiodination) of T4, whereas the extrathyroidal generation of T 3 requires monodeiodination of the phenolic or outer ring of T4 (5' deiodination) (1). Recent investigations suggest that there may be separate deiodinases involved in the inner and outer ring monodeiodination of T4 (2). The measurements of rT 3 and T3 in the circulation and in tissues are of significance because in addition to elucidating the pathways of T4 deiodination, they have served to generate provocative concepts in regard to the peripheral regulation of T3. Clinically, the elevation of rT 3 seen in a variety of systemic illnesses (3), in starvation (4), surgery (5), and in neonates (6), with a concomitant decrease in T3, has given rise to the speculation that the diversion of T4 monodeiodination from the outer to the inner ring (to generate inert rT3) in these states may be protective by counteracting excessive calorigenic effects of T3.
Methods, Results, and Discussion
The reported concentrations of rT 3 in the serum of euthyroid individuals vary from laboratory to laboratory (Table 1). The lowest mean rT 3 concentration noted is 14 ng/100 ml, whereas the highest value of 60 ng/100 ml has been recorded in the studies of Burman et al. (18). The mean rT 3 concentration in the serum, therefore, is not known with sufficient accuracy, so that the daily rT 3 production rates derived from isotopic kinetic computations have been questioned. The variations in circulating rT 3 values in euthyroid individuals are attributed to a variety of factors, such as differences in technical aspects of rT 3 assay that include sensitivity and specificity of the rT 3 antibody, differences in iodine concentration in different geographic locales, age-associated changes in iodothyronine metabolism, nutritional variables, lack of uniformity among investigators in correcting rT 3 values for T4 cross-reaction, etc. A careful inspection of the rT 3 assay values from different studies (Table 1), however, reveals that most investigators who have used rT 3 standards from Henning (Berlin) have reported lower mean values Received November 17,1977. Address requests for reprints to: Dr. B. N. Premachan- than those who have used rT 3 standards from dra, Veterans Administration Hospital 657/151C, St. other sources. This raises the possibility that Louis, Missouri 63125. rT 3 variations in sera of euthyroid individuals * This work was supported in part by the Narveen may also be due to differences in rT 3 standards Medical Research Foundation, St. Louis, MO. 746
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
BHARTUR N. PREMACHANDRA
RIA OF rT 3 TABLE 1. Circulating rT3 concentration (nanograms per 100 ml) in euthyroid individuals, as reported by various investigators Mean
Range"
Visser et al. (7) Ratcliffe et al. (8) Kodding et al. (9) Meinhold et al. (10) Burrows et al. (11) Hufner and Grussendorf (12) Premachandra Griffiths et al. (13) Gavin et al. (14) Premachandra et al. (15) Fisher et al. (16) Vagenakis et al. (4) Nicod et al. (17) B u r m a n e t a / . (18) Chopra (19) Kaplan et al. (20)
14 17 16 18 16 20
9-20 10-27 0-32 2-46 10-28 6-45
20 44 32 42 41 36 45 60 41 23
8-26 28-58 17-47 25-60 19-88C 6-66 24-64 36-84 27-62 7-39
Source of L-rT 3 standards Henning*
Jorgensen
Cahnmann Meltzerrf
" Range as reported by the authors or derived from mean ± 2 SD.
* Berlin GMBH. Age, 10-15 yr. DL-rT 3 .
c
d
employed in the assay. Therefore, we carried out rT 3 assays, as described previously (15), in identical serum samples under the same assay conditions using rT3 from different sources. The same batch of charcoal-stripped iodothyronine-free human plasma was used to make up rT3* standards from various sources; similarly the same shipment of [125I]rT3 (Abbott Labs., North Chicago, IL) immediately after receipt was used in all assays. Antibody to rT3 was produced in rabbits by immunizing
them with rT 3 (Jorgensen)-albumin conjugates. The lowest rT 3 concentration consistently detectable in the assay was 5 ng/100 ml or 5 pg/assay tube. T4 cross-reaction with rT 3 antibody in our assay was 0.04%, which means that even at a serum T4 concentration of 12 /xg/100 ml (upper normal range), rT 3 values in serum would not be overestimated by more than 4-5 ng. The rT 3 standard curves and representative assay values in seven serum samples are shown in Fig. 1 and Table 2. An inspection of these data clearly makes it evident that the differences in rT 3 values noted in the same samples (when assayed with different rT 3 1
rT 3 standards were dried at 40 C for 48 h and stored over a dehydrant under vacuum and all solutions were made in double distilled deionized water.
standards) are of the same magnitude as those noted in various studies referred to previously (Table 1). At the level of 25 ng/100 ml rT 3 standard (Fig. 1), the amount of [125I]rT3 isotopic displacement2 from the antibody (i.e. difference between 0 and 25 ng/100 ml), as noted with Jorgensen, Washington Reference Laboratory, and Henning standards, were 11, 17, and 25 arbitrary U, respectively, whereas between 0 and 50 ng/100 ml, the corresponding displacements were 18,27, and 30 arbitrary U. At higher concentrations (0-200 ng/100 ml), the differences between standards, as manifested in units of isotopic displacement, narrowed with the approaching saturation of the antibody-binding sites. As marked differences in slopes of standard curves generated with different rT 3 standards were apparent, the extrapolation of rT 3 values from different curves resulted in appreciable differences in the same serum samples. The values obtained with Henning and Washington Reference Laboratory standards were closer than the values observed with Jorgensen standard. It should be noted, however, that the values obtained with Jorgensen standard in our assay as well as those of others (Table 1) are similar or even lower than those noted with Cahnmann rT 3 standards. In the one exception, Griffiths et al. (13), using Henning rT 3 standards, reported mean rT 3 values which were even higher than those noted with Cahnmann standards. Finally, Kaplan et al. (20), using DL-rT3 standard, obtained values in normal human serum that were markedly lower than that reported by Chopra (19) who also used DL-rT3; on the other hand, in the studies of Burman et al. (18), the standard curves generated with DL-rT3 had similar slopes, intercepts, and nonimmune binding as L-rT3, but the values reported in human sera were even 2
In our assay, the antiserum is prelabeled with [ I]rT3. rT 3 released from denatured binding sites in serum (effected by TCA-NaOH addition) competes with labeled rT3 for the same binding sites on the antibody and displaces [125I]rT3. This unbound labeled rT3 is separated from the bound hormone by means of a resin sponge which abstracts only the free [128I]rT3 (sponge [125I]rT3 uptake). Also, unextracted sera were used in our rT 3 assays, and previous studies (21) have shown no discrepancies in rT 3 values between ethanol-extracted and native sera. I25
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
Reference
747
PREMACHANDRA
748 80
200
RT3 CONC,(NG/100ML) FIG. 1. rT 3 RIA standard curves generated with Henning (top curve), Washington Reference Laboratory (middle curve), and Jorgensen (bottom curve) rT3 standards. Uptake values shown at each standard concentration represent mean of 10 determinations. Vertical bars at each point represent range of values noted. TABLE 2. rT;) concentration (nanograms per 100 ml) in sera of euthyroid individuals, as read from assay curves using various rT 3 standards (representative data) Normal human
Assay values from curves using 1standards from:
(euthyroid)
Jorgensen
1 2 3 4 5 6 7
20 25 30 40 50 60 65
Washington Ref. Lab." 11 15 17 22 28 33 38
Henning 8 9 11 15 18 20 22
" rT 3 prepared by Dr. M. Rubin of the Washington Reference Laboratory (P.O. Box 3645, Washington, DC).
higher than those reported by Chopra (19). Kaplan et al. (20) strongly emphasized the importance of specificity of antiserum as they obtained 2- to 10-fold differences in values in the same samples when assayed with different antisera, especially the less specific ones. Meinhold and Wenzel (personal communication) observed 2-fold differences in rT 3 concentration in the same serum samples when assayed with rT 3 antisera from several European investigators. In our comparisons, the same antiserum under identical conditions was used in all assays so that the differences in rT 3 values we have recorded in normal human sera (which roughly equal those noted between assays; Table 1) can at least partly be attributed to a nonuniformity in rT 3 standard preparations. Ultraviolet spectrophotometry3 of various rT 3 standards (1 mg/ml in 0.1 N NaOH) did not exhibit significant discrepancies, and at 320 nM, virtually identical readings were obtained with all rT 3 preparations. The differences in rT 3 standards (which were reproducible; Fig. 1) were apparently not due to measurable T 3 or T4 contamination of rT3, as noted from the failure of various rT 3 standards, even at a concentration of 600 ng/100 ml, to react with either T3 or T4 antibody as examined in T 3 and T4 RIAs. It would seem that differences in techniques for the synthesis and purification of rT 3 may contribute to nonuniformity in rT 3 preparations from various sources. It is quite likely that impurities or inert byproduct material in the synthesized rT 3 standards may artifactually give rise to high values, as a larger amount of rT 3 would be needed to effect a certain degree of isotopic displacement from the antibody in comparison to a purer preparation. A common assertion in respect to RIA of hormones is that although RIA values may 3
Dr. Cahnmann (NIH) suggests that determination of absorption profile of rT 3 standards (filtered through 0.22fim filter) from 280 nM and up, instead of measuring just at absorption maximum, is preferable to obtain maximum/minimum (peak/valley) responses for comparison to remove doubts of the presence of soluble inert material in rT 3 standards despite their established purity by thin layer chromatography and gas-liquid partition chromatography.
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
100
JCE&M • 1978 Vol47 • No 4
RIA OF rT 3
While this manuscript was under review, Burrows et al. (11) also reported similar observations on the variability of rT 3 standards. In their studies, rT 3 from Warner Lambert Co. had approximately half of the immunological potency as Henning rT3. In receptor assays, Jorgensen (personal communication) noted that rT 3 prepared by the Shiba-Cahnmann method resulted in relatively high values, about 1% that of L-T 3 , while a sample made essentially by the Henning procedure and purified further by thin layer chromatography gave a value of about 0.2% L-T 3 .
Acknowledgments Grateful acknowledgment is made for the courtesies of Drs. Jorgensen and Cahnmann for the supply of iodothyronine analogs and other valuable comments; Abbott Laboratories, North Chicago, IL, for gifts of radioactive material and other supplies; and Henning, Berlin, for gifts of L-rT3 and T2. It is also a pleasure to acknowledge Dr. Wenzel's critical review of this manuscript and his suggestions and valuable comments.
References 1. CAVALIERI, R. R., AND B. RAPOPORT, Impaired peripheral
conversion of thyroxine to triiodothyronine, Annu Rev Med 28: 57, 1977. 2. CAVALIERI, R. R., L. A. GAVIN, F. BUI, F. MCMAHON, AND M.
HAMMOND, Conversion of thyroxine to reverse-T3 by a soluble enzyme system in rat liver, Abstracts of the 53rd Meeting of the American Thyroid Association, 1977, p. T-ll. 3. CHOPRA, I. J., U. CHOPRA, S. R. SMITH, M. REZA, AND D. H.
SOLOMON, Reciprocal changes in serum concentrations of 3,3',5'-triiodothyronine (reverse T3) and 3,3'5-triiodothyronine (T3) in systemic illnesses, J Clin Endocrinol Metab 41: 1043, 1975. 4. VAGENAKIS, A. G., A. BURGER, G. I. PORTNAY, M. RUDOLPH, J. T. O'BRIAN, F. AZIZI, R. A. ARKY, P. NICOD, S. H. INGBAR,
AND L. E. BRAVERMAN, Diversion of peripheral thyroxine metabolism from activating to inactivating pathways during complete fasting, J Clin Endocrinol Metab 41: 191, 1975. 5. BURR, W. A., E. G. BLACK, R. S. GRIFFITHS, R. HOFFENBERG,
H. MEINHOLD, AND K. W. WENZEL, Serum triiodothyronine
and reverse triiodothyronine concentrations after surgical operation, Lancet 2: 1277, 1975. 6. CHOPRA, I. J., J. SACK, AND D. A. FISHER, 3,3',5'-Triiodothy-
ronine (reverse T3) and 3,3',5-triiodothyronine (T3) in fetal and adult sheep: studies of metabolic clearance rates, production rates, serum binding, and thyroidal content relative to thyroxine, Endocrinology 97: 1080, 1975. 7. VISSER, T. J., R. DOCTER, AND G. HENNEMANN, Radioim-
munoassay of reverse tri-iodothyronine, J Endocrinol 73: 395, 1977. 8. RATCLIFFE, W. A., J. MARSHALL, AND J. G. RATCLIFFE, The
radioimmunoassay of 3,3',5'-triiodothyronine (Reverse T.-)) in unextracted human serum, Clin Endocrinol 5: 631, 1976. 9. KODDING, R., J. JANZEN, E. SCHMIDT, AND R. D. HESCH,
Reverse triiodothyronine in liver disease, Lancet 2: 314,1976. 10. MEINHOLD, H., K. W. WENZEL, AND P. SCHURNBRAND, Ra-
dioimmunoassay of 3,3',5'-triiodo-L-thyronine (reverse T3) in human serum and its application in different thyroid states, Z Klin Chem Klin Biochem 13: 571, 1975. 11. BURROWS, A. W., E. COOPER, R. A. SHAKESPEAR, C. M.
Downloaded from https://academic.oup.com/jcem/article-abstract/47/4/746/2678797 by Leiden University / LUMC user on 16 January 2019
not compare favorably from one laboratory to another, it is the relative changes in hormone "* levels (during therapy or other clinical and experimental maneuvers) that are of importance and significance. While this is undoubtedly true in many clinical and other situations, our experience suggests that total reliance on this basis may not always be reassuring and can even be incorrect and misleading. For example, a test serum with rT 3 value of 50 ng/100 ml as read from Henning rT 3 standard curve would read 120 ng/100 ml in Jorgensen's standard curve (Fig. 1). When these levels are compared with the mean values in euthyroid normals, as obtained with these two standards (which is 20 ng/100 ml for Henning and 42 ng/100 ml for Jorgensen; Table 1), the test serum rT3 concentration is 2.5 times (50/20) greater than the mean value in Henning comparison, whereas it is almost 3 times (120/42) y greater than the mean value in Jorgensen's standard comparison. The two relative statements are not exactly comparable. Greater distortions in relative comparisons of rT 3 standards become apparent when they are incorporated in iodothyronine-free plasma from dif7 ferent species. For instance, use of rT 3 from Henning resulted in values in human sera (20 ng/100 ml) which were approximately half of those noted with Jorgensen rT 3 (42 ng/100 ml; Table 1; in assays with human sera all standards were incorporated in iodothyronine-free human plasma), whereas in dog sera, the use of Henning standards resulted in values (50 ng/100 ml) which were approximately onefourth of those noted with Jorgensen's standard (187 ng/100 ml; both Henning and Jorgensen's standards were made in iodothyronine-free dog plasma ) (21). Even more inaccurate and misleading conclusions might become evident if one is measuring rT 3 in sera containing little or no rT3, as in the case of the chicken (15). It is also obvious that data relating to quantitative changes in peripheral conversion of T4 to rT 3 under clinical and experimental conditions may not favorably compare between investigators if they are using rT 3 standards which lead to a 2- to 3-fold difference in rT 3 values in the same serum samples, as can be realized from an inspection of data in Table 2.
749
750
PREMACHANDRA AICKIN, S. FRASER, R. D. HESCH, AND C. W. BURKE, LOW
serum L-T3 levels in the elderly sick; protein binding, thyroid and pituitary responsiveness, and reverse T3 concentrations, Clin Endocrinol 7: 289, 1977. 12. HUFNER, M., AND M. GRUSSENDORF, Radioimmunoassay for
3,3',5'-triiodothyronine (reverse T3) R-T3) in unextracted human serum, Clin Chim Ada 69: 497, 1976. surement of serum 3,3',5'-(reverse) T3, with comments on its derivation, Clin Endocrinol 5: 679, 1976. 14. GAVIN, L., J. CASTLE, F. MCMAHON, P. MARTIN, M. HAMMOND, AND R. R. CAVALIERI, Extrathyroidal conversion of
thyroxine to 3,3',5'-triiodothyronine (reverse-T3) and to 3,5,3'triiodothyronine (T3) in humans, J Clin Endocrinol Metab 44: 733,1977.
i 1978 No 4
children 1 to 15 years of age, J Clin Endocrinol Metab 45: 191, 1977. 17. NICOD, P., A. BURGER, V. STAEHELI, AND M. B. VALLOTTON,
A radioimmunoassay for 3,3',5'-triiodo-L-thyronine in unextracted serum: method and clinical results, J Clin Endocrinol Metab 42: 823, 1976. 18. BURMAN, K. D., R. C. DIMOND, F. D. WRIGHT, J. M. EARLL, J. BRUTON, AND L. WARTOFSKY, A radioimmunoassay for
3,3',5'-L-triiodothyronine (reverse T3): assessment of thyroid gland content and serum measurements in conditions of normal and altered thyroidal economy and following administration of thyrotropin releasing hormone (TRH) and thyrotropin (TSH), J Clin Endocrinol Metab 44: 660, 1977. 19. CHOPRA, I. J., A radioimmunoassay for measurement of 3,3',5'triiodothyronine (reverse T3), J Clin Invest 54: 583, 1974.
15. PREMACHANDRA, B . N . , S. LANG, J. A . ANDRADA, AND J. H.
20. KAPLAN, M. M., M. SCHIMMEL, AND R. D. UTIGER, Changes
KITE, JR., Reverse triiodothyronine in the chicken, Life Sci 21: 205, 1977.
in serum 3,3',5'-triiodothyronine (reverse T3) concentrations with altered thyroid hormone secretion and metabolism, J Clin Endocrinol Metab 45: 447, 1977.
16. FISHER, D. A., J. SACK, T. H. ODDIE, A. E. PEKARY, J. M. HERSHMAN, R. W. LAM, AND M. E. PARSLOW, Serum T
