Storage conditions and stability of thyrotropin and thyroid hormones on filter paper.

We evaluated the effects of temperature and humidity on thyroid hormones (T4, T3) and thyrotropin (TSH) measured in blood spots dried on filter paper. RIAs for T4 and T3 blood spots were optimized to measure these analytes over the neonatal and euthyroid adult reference intervals. Sensitivities of the T3 and T4 assays were 0.5 and 10 nmol/L, respectively. A blood-spot immunoradiometric assay for TSH involving magnetizable particles was developed with a sensitivity of 6 milli-int. units/L. Control sera stored at -20 degrees C, 4 degrees C, room temperature, 37 degrees C, and at external ambient temperatures for 36 days showed no significant change in measured concentrations of TSH or T3 during 30 days or for T4 at -20 and 4 degrees C. T4 markedly declined in blood spots stored at room temperature (either high or low humidity), 37 degrees C, or ambient temperature. TSH and T3 in blood spots evidently are stable at temperatures likely to be encountered during storage or transport, but blood spots for T4 assay must be stored between -20 and 4 degrees C.

Immunoradiometric assay with use of magnetizable particles: measurement of thyrotropin in blood spots, to screen for neonatal hypothyroidism.

We adapted a commercial immunoradiometric assay (IRMA) to measure thyrotropin in filter-paper blood spots. Two 3-mm blood spots are used for each standard and sample. These are incubated for 2 h with radiolabeled antibody and for 30 min with magnetic antibody, followed by a 10-min separation procedure. Assay sensitivity is 6 milli-int. units/L. Coefficients of variation (precision profile of the standard curve) ranged from 4.3 to 9.6%. The coefficient of correlation (r) between thyrotropin concentrations in the blood spots and in serum was 0.93. Pre-elution of the blood spots is necessary for short incubation time. Short incubation time, little need for specialized equipment, the high precision and sensitivity characteristic of IRMA, and ease of collection, transport, and storage of the blood-spot samples make this assay suitable for neonatal hypothyroid screening.

A reappraisal of the binding characteristics of human thyroxine-binding globulin for 3,5,3'-triiodothyronine and thyroxine.

The binding characteristics of T4 and T3 to dilute plasma were studied separately in five normal euthyroid subjects with normal levels of thyroxine-binding globulin (TBG). Scatchard analyses of these data revealed similar mean affinity constants for T4 [2.0 +/- 0.7 (SD) X 10(9) M-1] and T3 (2.0 +/- 0.7 X 10(9) M-1), but a 5-fold higher capacity for T4 (0.75 +/- 0.18 mol T4/mol TBG) than for T3 (0.14 +/- 0.06 mol T3/mol TBG). Similar results were obtained using various assay buffers, pH concentrations, or separation methods. This characteristic pattern of T4 and T3 binding was retained by thyroid hormone free plasma, with the only difference being a slight parallel shift to the left of the Scatchard plots for both T4 and T3. The calculated affinities (Ka) for T4 and T3 were 5.2 X 10(9) M-1 and 5.2 X 10(9) M-1, respectively. High affinity T4 and T3 binding was abolished in plasma selectively depleted of TBG, but was retained after selective depletion of either prealbumin or albumin. Highly purified TBG, prepared from normal serum, demonstrated binding characteristics for T3 and T4 similar to dilute plasma. Displacement of [125I]T4 from dilute plasma by unlabeled T3 or T4 revealed a binding potency of T3 relative to T4 of 9%. Binding affinities derived from analog displacement studies appear invalid as these calculations assume equal binding capacities of TBG for T4 and T3. It seems clear from these studies, that the binding characteristics of human TBG are inconsistent with a single competitive binding site for thyroid hormones.

Deiodination of reverse 3, 3', 5'-triiodothyronine by hepatic nuclear protein preparations.

A reassessment of the binding characteristics of [125I]rT3 to putative receptors in nuclear protein extracts of rat and pig liver revealed that significant deiodination of radioligand occurred during incubation. When previously reported separation procedures are used, released radioiodine is included in the protein-bound [125I]rT3 fraction during separation of protein bound from free hormone by Sephadex G-25 chromatography. This misclassification produces artefacts in binding curves and Scatchard plots used to calculate binding affinity and capacity. Previously reported affinities and capacities derived by this methodology are therefore erroneous. Deiodination of rT3 in the nuclear protein extracts appears to be mediated by outer ring deiodinase. Whereas dithiothreitol markedly enhanced radioiodine generation, the enzyme inhibitors ipodate and salicylate reduced iodine production. These effects produced dramatic changes in apparent binding curves for the radioreceptor assay. When [125I]T3 was incubated with nuclear protein extract, no significant deiodination was detected. Whereas it is likely that the deiodinase is a microsomal contaminant of the nuclear preparation, as suggested by the presence of glucose-6-phosphatase in the nuclear protein preparation, the possibility of an intrinsic nuclear-linked deiodinase cannot be overlooked.