United States Patent 3659104

A new and improved in vitro concept in measuring serum thyroxine (T-4) is sclosed which employs an alkaline crosslinked dextran gel (Sephadex) column to dissociate and separate the T-4 from the serum protein in a single operation. An isotope dilution technique combined with saturation analysis is used to estimate the T-4 content in serum.

Gross, Jack (Jerusalem, IL)
Gordon, Amirav (Jerusalem, IL)
Schick, Lloyd Alan (Elkhart, IN)
Application Number:
Publication Date:
Filing Date:
Primary Class:
Other Classes:
250/303, 250/304, 436/529, 436/545, 436/804, 436/825, 436/826
International Classes:
G01N33/487; G01N33/78; (IPC1-7): G21H5/02
Field of Search:
250/83R,83SA,16T 23
View Patent Images:

Primary Examiner:
Lawrence, James W.
Assistant Examiner:
Willis, Davis L.
What is claimed is

1. A process for the in vitro determination of thyroxine in serum comprising:

2. A process as in claim 1 wherein the radioactivity of the gel column is determined before and after elution to determine the ratio of radioactive thyroxine retained by the column to that originally added.

3. A process as in claim 1 wherein the aqueous alkaline solution is buffered to a pH of about 8 to 10.

4. A process as in claim 3 wherein the aqueous alkaline solution is a barbital buffer having a pH of 8.6.

5. A process as in claim 1 wherein the pH of the column is maintained at about 11 to 13.

6. A process as in claim 1 wherein the eluting agent is serum.

7. A process as in claim 1 wherein the eluting agent is human α-globulin.


The thyroid gland is extremely important in the animal body because of its effect upon the basal metabolic rate. This effect is regulated by the thyroxine hormone which is released in response to nervous or hormonal stimuli. Thyroxine enters the circulatory system and acts either directly upon the cell or indirectly upon other hormonal systems. Abnormal activity of the thyroid is a common malady in humans. In hypothyroidism, the body has decreased thyroid activity which is manifest by such diseases as cretinism and myxedema. Hyperthyroidism is a state of excessive thyroid activity in which one becomes nervous, develops an increased pulse rate and sometimes goiter.

Thyroid deficiency was associated with a reduced metabolic rate as early as 1895 and several systems based on basal metabolic rate were devised for estimating thyroid activity. However, such systems were not reliable, so more direct and precise methods were sought. In 1896, iodine was discovered in thyroid extract but the relationship between blood iodine level and thyroid function was not firmly established until 1933. This led to the use of protein-bound iodine as a means of estimating thyroid function, and by 1955 the PBI test was standard for checking thyroid activity. It soon became apparent that this test was influenced by the administration of other iodine containing compounds to the patient. Thus, the butanol-extractable iodine (BEI) procedure was devised which gave better correlation between serum iodine levels and the clinical findings but was likewise non-specific for T-4.

A major advance in T-4 analysis occurred in 1959 when Galton et al., Biochem. J. 72, 310 (1959) liberated T-4 from serum protein by hydrolysis and separated it from other iodine compounds by passage through a resin column. Later, Pileggi et al., J. Clin. Endocr. Met. 21, 1272 (1961) developed this column chromatographic procedure into a clinically usable method.

In 1964, Murphy et al., J. Clin. Endocr. Met. 24, 187 (1964) developed a T-4 assay based upon competitive protein binding and isotopic dilution which required the initial extraction of T-4 from the serum with alcohol followed by centrifugation and drying. Although this test was highly specific, only 80 percent of the T-4 could be recovered from the serum.

More recently, U.S. Pat. No. 3,471,553 set forth still another column chromatography T-4 assay in which an anion-exchange resin is adjusted to an alkaline pH of about 12 to dissociate the thyroxine from its albumin and globulin carriers. The diluted serum solution is then poured onto the resin wherein proteins, amino acids, thyroxine, iodotyrosine and inorganic iodine are adsorbed. Successive washes with acetate buffer isopropyl alcohol and acetic acid remove serum proteins, iodothyronines and some organic iodine compounds. Further treatment of the resin with 50 percent acetic acid at a pH of 2 quantitatively removes T-4. Even though many modifications have been made in the original T-4 by column assay of Galton et al., the procedure is substantially the same i.e., a diluted serum sample is applied to an ion exchange resin. The unwanted contaminants are then eluted from the column and discarded. Following this, the hormones to be measured are eluted, collected and the T-4 determined by iodine analysis.


The present invention for determining T-4 is based on the competitive protein binding principle which is a modified form of saturation analysis. The T-4 to be determined is mixed with a determinate sample of T-4 labeled with a trace amount of radioactive isotope. A binding agent is added which will bind a definite number of molecules. Since the binding agent cannot distinguish between the labeled and unlabeled molecules, they compete with each other on an equal basis for the binding sites. These molecules are uniformly mixed so that the binding agent will bind them in the same ratio as that existing in the free or unbound state. As the concentration of the unlabeled molecules increases, the ratio becomes smaller and fewer labeled molecules are bound by the binding agent, leaving more of the labeled molecules in the free state. By calibrating the binding of labeled molecules in the presence of a known amount of unlabeled molecules, a quantitative procedure can be established.

In practicing the present invention, the crosslinked dextran gel column acts as the secondary binding agent for the unbound or free T-4, whereas the primary binding agent is a thyroxine (T-4) binding protein. First, a measured amount of serum is mixed with some T-4 labeled with radioactive iodine on top of the column. Both the column and T-4 mixture are at pH12 to 13. At this pH, the T-4 binding serum proteins such as prealbumin, albumin and thyroxine binding globulin are completely dissociated from T-4. As the mixture flows down the column, the T-4 is bound by the dextran gel. The serum proteins are washed away with a barbital buffer at a pH of 8.6 which automatically adjusts the pH of the column to that of the buffer. The pH is such that the transfer of the T-4 from the column to the primary binding agent is facilitated. The primary binding agent dissolved in barbital buffer is then added. Equilibrium is quickly established between the two binding agents and the primary binder is washed away carrying a portion of the T-4 with it. By measuring the amount of radioactivity on the column before and after treatment with the primary binding agent and comparing the percent retained by the column with a standard curve in which percent retained is plotted against T-4 concentration, the amount of T-4 in the serum can be determined.


This invention is predicated upon the discovery that a dextran gel column retains radioactive T-4 very well up to a pH of 9. Between pH 9 and 11, radioactive T-4 is rapidly eluted from the column whereas above pH 11 the retention of T-4 by the column is greatly enhanced. Thus, dextran gel binds T-4 but not protein, whereas anion exchange resins bind both T-4 and protein which are removed nonspecifically and only under acidic conditions.

The dextran gels employed in the column are crosslinked with various amounts of epichlorohydrin as described in U.S. Pat. No. 3,042,667 and have a water regain of from 1 to 5 grams per gram of dry gel product. Such gels are produced commercially by Pharmacia of Uppsala, Sweden and sold under the trade name of Sephadex in various molecular weight ranges and sieve sizes. Thus, Sephadex G-10 has a water regain of one gram per gram of dry gel, Sephadex G-15 has a water regain of 1.5 grams per gram of dry gel, Sephadex G-25 has a water regain of 2.5 grams per gram of dry gel and Sephadex G-50 has a water regain of 5 grams per gram of dry gel. Of these gels, Sephadex G-25 is preferred.

The dextran gel column employed in this invention is prepared by suspending 500 grams of dry gel in two liters of distilled water and allowing it to hydrate overnight. Fines are removed by slurrying the gel in 0.1 N sodium hydroxide for about 5 minutes, allowing it to settle for 15 minutes and then drawing off the supernatant by suction. The process is repeated three times and the gel is suspended in 4,400 milliliters of 0.1 N sodium hydroxide. Four milliliters of this suspension is placed in a six milliliter plastic syringe barrel having a diameter of 13 millimeters and a length of 66 millimeters. Each barrel is prefitted with a bottom closure means, for example, a removable cap, and a detergent treated sintered polyethylene retaining disc about 1.5 millimeters thick and having a diameter of 13 millimeters is pressed coaxially to the bottom of the plastic barrel. After placement of the suspension in the syringe barrel, the suspension is stirred and permitted to settle free of air bubbles after which another detergent-treated sintered polyethylene disc, like the first-mentioned disc, is inserted into the syringe barrel and pushed coaxially into firm contact with the gel. About 1.5 milliliters of sodium hydroxide solution remains above the upper disc. The upper end of the syringe is closed with a new polyethylene cap. This procedure provides a column containing about 450 milligrams of gel.

The T-4 test herein contemplated is performed by utilizing the gel column thus prepared as follows:

1. Remove the top cap from the column, decant the supernatant and place the column in an upright position.

2. Add 0.45 milliliter of a 0.1 N sodium hydroxide solution containing about 0.10 microcuries of radioactive T-4 to provide from 60,000 to 120,000 counts per minute of gamma radioactivity.

3. Add 0.1 milliliter of serum sample and mix with the radioactive T-4. If a standard is to be determined, nonradioactive T-4 is added.

4. Remove the bottom cap and allow the serum mixed with the radioactive T-4 to flow into the column.

5. Wash the column with 4 milliliters of a 0.075 molar aqueous barbital solution buffered to a pH of 8.6.

6. Replace the bottom cap and determine the radioactivity of the column in counts per minute with a gamma counter.

7. Remove the bottom cap and add one milliliter of 0.15% human α-globulin dissolved in 0.075 molar barbital buffer.

8. Add 4 milliliters of 0.075 molar barbital buffer at pH 8.6 and allow to flow through the column.

9. Replace the bottom cap and again determine the radioactivity of the column.

10. Calculate the percentage of radioactivity retained on the column and determine the T-4 content of the serum sample by comparing the percent retention to a standard curve prepared with T-4 solutions of known concentrations.

Although a specific embodiment of the test method comprising this invention has been described, it should be understood that several variations are possible. Thus, the gel column can be made alkaline by potassium hydroxide or ammonium hydroxide if desired. Human α-globulin can be replaced with an equivalent amount of human serum, bovine serum or bovine gamma globulin as the primary binding agent. Aqueous alkaline solutions buffered to a pH of from 8 to 10 with sodium phosphate or tris (hydroxymethyl) amino methane can be used at concentrations from 0.01 to 0.2 molar, but an aqueous barbital solution is preferred, since it facilitates better quantitation when used to dissolve the human α-globulin or other binding agents.

Another variation of the present invention involves determining the radioactivity of the solutions before addition to the gel column and after elution therefrom rather than determining the radioactivity of the column itself. Percent retention is then determined by a difference calculation. However, for the sake of efficiency, it is preferable to determine the radioactivity of the column before and after elution.

Assays for T-4 by the column methods of the prior art should not be confused with the present method which involves saturation analysis using a radioactive tracer. Previously, a column was utilized only to separate contaminating iodines prior to analysis of iodine. In certain cases, this was done by using an ion exchange resin column at an alkaline pH, and iodine analysis was performed on the T-4 recovered from the column by measuring the effect of iodine on the ceric-arsenious acid reaction. The saturation analysis method herein disclosed is a direct determination of thyroxine, rather than the indirect measurement of thyroxine as iodine.