GB2023816A - Improvements in or relating to bioassay - Google Patents

Abstract

A process is disclosed for determining the concentration of an unbound species, e.g., thyroxine, cortisol, testosterone, or digoxin, in a liquid sample containing the species and protein. The sample is incubated with antibody specific to the species to be detected and a distinguishable analogue of the species, the antibody being initially separated from the sample by semipermeable membrane capable of excluding the passage of natural protein and antibody but allowing passage of the species and its analogue. The antibody and analogue are contained in semipermeable microcapsules. Free species in the sample permeates the membranes and competes for sites of attachment to the antibody with the analogue. The antibody is then separated from residual unbound species and unbound analogue. The amount of analogue present either in association with the antibody or in the remainder of the reaction system is indicative of the level of unbound species originally present in the sample.

Description

SPECIFICATION Improvements in or relating to bioassay This invention relates to the detection of the presence and concentration of unbound species, e.g., hormones, in a liquid sample such as serum. The procedure well suited for the detection of free species of the type which are capable of reversibly binding with protein in samples containing the free species, the binding protein, and a concentration of species-protein complex.

The "competitive assay" technique for measuring the concentration of various biologically active substances for diagnostic purposes is now well established. The technique involves a reaction system wherein antibody specific to the species to be determined is incubated together with the test sample and an aliquot of a distinguishable analogue of the sample. The analogue and natural species compete for sites of attachment to the antibody, and after separation of the antibody from the remainder of the reaction system, the antibody or supernatant may be assayed for analogue. The amount of analogue associated with the antibody is an inverse function of the concentration of the test species in the sample.

Two common methods of performing such competitive determinations are known as "liquid phase radioimmunoassay" and "solid phase radioimmunoassay". In each system the immunogenic substance which is bound to an antibody must be removed from unbound immunogenic substance in order to make a measurement of the amount of labeled analogue on the antibody. From this measurement, the amount of immunogenic substance in the sample is determined.

In "liquid" phase radioimmunoassay systems, the antibody, labeled analogue, and immunogenic substance to be determined are incubated in solution. A number of antibody binding sites are available and reaction time is rapid. To separate the antibody complexed immunogenic substance from the uncomplexed immunogenic substance, the antibody immunogenic complex is precipitated, centrifuged, and the supernatant is decanted.

"Solid phase" radioimmunoassay systems avoid the precipitation step. The antibody is bound to a glass chip or inside surface of a tube. Centrifugation of the glass chips or decantation of the coated tubes will separate the antibody complexed immunogenic substance from the uncomplexed immunogenic substance.

However, the reaction time is relatively slow because a number of available active sites on the antibody is blocked by the chip ortube.

Hormones such as those excreted from the thyroid, testes, ovaries, adrenal cortex, and other mammalian glands are often transported through the circulatory system in association with hormone binding proteins or globulins. Often, it is of diagnostic significance to be able to determine the presence and concentration of free hormone, as opposed to protein-bound hormone or total hormone. For example, thyroxine (T4) in the bloodstream exists in a dynamic equilibrium as a species bound to a transport protein, (thyroxine-binding globulin, albumin, or prealbumin) and, in a small fraction (potentially 0.1% or less), as an unbound substance ("free" or unbound T4). Free T4 is thought to be the active species of the thyroid hormone system.

Unfortunately, free-T4 assay methodology has been tedious, complicated, expensive, and error-prone because of the difficulty in standardization of materials and in measuring minute quantites of unbound hormone in the presence of overwhelming amounts of hormone relatively loosely bound to protein. The competitive assay procedure noted above if not capable of determining free species concentrations.

Of the methods for measuring free T4 and other hormones which exist in vivo as protein complexes, the dialysis membrane method affords a high degree of accuracy. A dialysis bag containing known quantities of serum to be tested and labeled hormone is suspended in buffer. The labeled hormone distributes itself between the free and bound states in the same proportions as the serum hormone. The mass of labeled hormone added must be extremely small so as to avoid seriously perturbing the system, yet the count rate must be high so that detection of small amounts is feasible. In the case of radioactive labels, this requires that labeled hormone of very high specific activity be used.After a suitable period (about 24 hours), the tagged and natural hormone not bound to protein diffuse through the semipermeable dialysis bag while hormone bound to protein (molecular weight greater than 20,000) remains within the dialysis bag. To determine the quantity of free hormone present in the serum, one assays an aliquot of the buffer for labeled hormone. As a result of the assay, the fraction of labeled hormone that crossed into the buffer may be calculated, and the fraction of total hormone that exists in the free state inferred. To convert this fraction to mass units (e.g., ng/dl) of free hormone, a total hormone assay must also be run on the sample.

While this system can give meaningful results in free T4 and other free hormone assays, it is poorly suited for routine use. Two tests must be run, and the accuracy of the final result can be comprised by either procedure. Rather large amounts of radioactivity must be employed per test. Interfering impurities in the labeled hormone can cause special problems, and long incubation times are required. Only serum can be used, and it must be extremely fresh. Furthermore, the collection and standardization of all the materials necessary for the assay is not a routine matter. Thus, the application of the procedure has been limited due to the high cost and labor intensity of the method and the skilled personnel required.

Another method of free hormone assay involves reaction kinetics and requires two separate tests. Each test measures a kinetic curve related to how fast an antibody captures hormone, e.g.., T4, away from the opposing pull of the primary binder such as thyroxine binding globulin.

In such an assay system, inaccuracies result from several pathologic conditions. If more protein binding sites are present than usual, the effective attraction of the globulin for the hormone will be greater and the rate of binding to antibody will decrease. In the case of thyroxine analysis, this would give the appearance that the sample had little T4 when in fact there might be a high level of T4, but all bound.

It is one object of the present invention to provide a process, reagent, and test set for the direct assay of unbound hormones and other such species. The assay may be performed in the presence of serum proteins and the bound hormone.

The invention provides a process for determining in a protein-containing sample the presence of a species unbound to protein, said process comprising the steps of; (a) incubating the sample with an antibody complementary to said species and a distinguishable analogue of said species, the sample and antibody being separated by membrane having a porosity sufficient to allow passage of said species and its distinguishable analogue, but insufficient to allow passage of said antibody or protein-bound species present in the sample; (b) allowing unbound species in the sample to traverse said membrane and to compete with its distinguishable analogue for binding sites on said antibody; (c) separating said antibody from residual unbound species and unbound analogue; and (d) determining the amount of said distinguishable analogue which is bound to said antibody or which is antibody free, said amount being indicative of the amount of species unbound to protein in the sample.

The invention also provides a test set for use in determining in a liquid sample the amount of an unbound species, said test set comprising: (a) a distinguishable analogue of said species; (b) a plurality of microcapsules adapted for use in absorbing unbound species from said sample, said microcapsules containing antibody complementary to said species, and comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody or natural proteins; and (c) a standard containing a predetermined amount of said species for comparison with a sample.

The invention also provides a reagent adapted for use in the determination of unbound species in a liquid sample comprising a plurality of microcapsules containing: (a) a distinguishable analogue of a species which species is capable of antibody formation; and (b) antibody complementary to said species, said microcapsules comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody and natural proteins.

The invention also provides a two part reagent adapted for use in the determination of unbound species in a liquid sample which reagent comprises a part A comprising a distinguishable analogue of a species which species is capable of antibody formation, and a part B comprising a plurality of microcapsules containing antibody complementary to said species, said microcapsules comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody and natural proteins.

The invention also provides microcapsules containing antibody complementary to a species which is capable of antibody formation.

Examples of species include L4hyrnxine, L-3,5,3' tri-iodothyronine, cortisol, testosterone, neonatal thyroxine and digoxin.

The standard comprises for sample aliquots of material containing known quantities of free species which, when tested in parallel with the sample, allow construction of a standard curve.

Bn embodiments of the invention, semipermeable microcapsules containing antibody complementary to the hormone or other species to be detected are incubated with both a distinguishable analogue of the hormone and the test sample. The analogue may be introduced into the microcapsules prior to the incubation in quantities such that the antibody is saturated at the hormone binding sites. The microcapsule walls comprise membranes separating the test sample from the antibody, and have a permeability sufficient to allow passage of free species and its analogue, but insufficient to permit passage of the antibody, natural proteins, or protein-bound hormone. When a test sample is added to an aliquot of the microcapsules, free hormone diffuses through the membranes and competes with its analogue for sites of attachment on the antibody.Thus, the distribution of analogue between the antibody and the remainder of the reaction system becomes indicative of the amount of free hormone originally present in the sample. This procedure combines the inherent accuracy of a conventional liquid phase radioimmunoassay and the simplicity and convenience of a solid phase system.

Next, the antibody together with its bound hormone (and bound analogue) is separated from the free species, and either the antibody or the remainder of the reaction system is assayed for analogue. The separation may be conducted by inducing an osmotic change so that the capsules collapse and unbound hormones migrate out of the capsules. This can be accomplished, for example, by adding serum albumin or polyethyleneimine to the system which promotes efflux of intracapsular liquid. Alternatively, free species may simply be washed from the microcapsules. Results are interpreted by comparing the assay OY analogue content to a standard, such as a curve of free hormone concentration vs. radioactive count, fluorescent intensity, or other marker characteristic used to indicate the presence of the analogue.

The assay may be used to detect the presence and/or concentration of free thyroxine, tri-iodothyronine, neonatal thyroxine, testosterone, cortisol, other steroid hormones, and other substances which reversibly bind with protein. The assay may also be used to detect species which do not bind to protein to any significant degree, e.g., drugs such as digoxin. Essentially any such material may be determined provided a complementary binding substance and a distinguishable analogue of the material is available or can be produced.

The assay may be routinely conducted using a test set comprising analogue, microcapsules containing antibody standards and blanks containing predetermined concentrations of the subject species, and a reagent for removing unbound species and analogue from the capsules after completion of the incubation.

For quality control, the solution in the microcapsule may be colored. A colored supernatant, after the capsules have been settied out or sedimented via centrifugation, would be indicative of microcapsule breakage and possible leakage of antibody. Unlike conventional assays, this assay avoids loss of clinical correlation with diagnosed conditions over a broad range of concentrations of protein, hormone, and interfering substances.

We believe we have provided a rapid, simple, and reproducible method of detecting the presence and concentration of species unbound to protein in a liquid sample, a reagent useful in such assays, and a test set suitable for rapidly and conveniently conducting such assays. We believe we have also provided an improved method of determining free T4 in samples extracted from human blood and containing protein bound4.

Further, we believe we have provided a competitive process for the determination of an immunogenic substance which combines the advantages of the solid radioimmunoassay system and the advantages of the liquid phase radioimmunoassay system; and reliable process for the determination of the amount of digoxin in serum.

There now follows a description, to be read with reference to the accompanying drawings of embodiments of the invention. This description, which is illustrative of process test set, and reagent aspects of the invention, is given by way of example only, and not by way of limitation of the invention.

In the accompanying drawings: Figure 7 illustrates a standard curve which is a plot of counts per minute (x 103) on a y-axis scale vs, free-T4 concentration in ng/dl on an x-axis log scale (see Table 1 for data); Figure 2 illustrates a second standard curve for a free-T4 assay in cpm (linear) v. ng/dl free-T4 (log scale).

Data for this figure is found in Table 2; Figure 3 is a graph of cpm from T4 (1 25i) bound toT4 antibody contained within semipermeable microcapsules immersed in standard solutions of free-T4 vs. time. As serum T4 replaces T4 (1251) at T4 antibody binding sites, the cpm remaining in association with the antibody decreases.Note that the rate of replacement is largely linear to two hours of incubation at 37"C; Figure 4 graphically illustrates the correlation of results between the dialysis assay technique and a microencapsulation system embodying the invention: Figure 5 graphically illustrates the normal range of free T4 as established by the microencapsulated assay system; Figure 6 graphically illustrates the correlation between kinetic radioimmunoassay and the microencapsulated type assay system; Figure 7shows a digoxin standard curve as counts per minute (cpm) bound to antibody, as plotted on the linear scale, vs. digoxin in nanograms/milliliter on the log scale of 2 cycle semilog paper (data from Table 5); Figure 8 shows digoxin as % bound (relative) vs. digoxin concentration.Percent bound (relative) is calculated as % bound (relative) = mean cpm bound/mean cpm bound of 0 ng/dl digoxin standard x 100 (data from Table 8); and Figure 9 is a diagram showing the retention of antibody and free passage of immunogenic substance in microcapsules.

In a process embodying the invention containing the species to be detected is incubated with antibody complementary to the species (or other substance capable of reversibly binding with the species) and a distinguishable analogue of the species, and that the sample and antibody be separated by a semipermeable membrane or membranes that exclude the passage of high molecular weight proteins.

Antibodies to selected species may be produced in accordance with well-known techniques involving injection of the hormone into laboratory animals, possibly together with an adjuvant, and subsequently extracting and purifying the produced antibodies. Many such antibodies are available commercially. Many distinguishable analogues of detectable species also commercially available or can be made using known techniques. The distinguishable analogue has a molecule having antibody binding properties similar to and preferably identical to the species sought to be detected, and is characterized by a property which allows a measure of its concentration to be readily obtained.Preferred analogues comprise a sample of the species to be detected tagged with a radioactive atom; for example, thyroid hormones may be conveniently tagged with 1251 and can then be quantitated by measuring gamma radiation. However, it is contemplated that other types of analogs may be employed, so long as the analogue has a molecular weight and resulting dimensions well below those of naturatproteins and the antibody used. Thus, analogues may be produced by tagging a sample of the species to be detected with a relatively low molecular weight enzyme, fluorescent moiety, or other moiety which enables quantitative measurement of the concentration of the analogue by physical or chemical means.

To practice the assay, the antibody and analogue are separated from the sample by one or more membranes having a permeability sufficient to preclude the passage of antibodies and natural proteins (which uniformly have a molecular weight in excess of 20,000 daltons) but sufficient to allow free passage of the species to be detected and its analogue. In a preferred embodiment of the invention, the membranes take the form of semi-permeable microcapsules containing the antibody and analogue.

It is noteworthythatthe semipermeable microcapsules used have a permeability similar to the dialysis membranes described above. However, the semipermeable microcapsule wall is hundreds of times thinner than conventional dialysis membranes, and its available surface area is orders of magnitude larger per unit weight. Free-T4 or other free hormones freely enter the microcapsule and displace, on proportion to their concentration, labeled T4 from the T4 antibody. Thus, an equilibrium of free T4 and labeled T4 is approached within the capsule during the incubation period.

Suitable methods of encapsulating biological materials in membranes having the foregoing permeability properties are disclosed in detail in U.S. patent application serial numbers 606,166, (Aug.20, 1975) (equivalent U.il. patent specification No. 1,540,461 of Damon Corp.), 931,177 (Aug. 4, 1978), and 30,847 (Aprii 17, 1979), all to F. Lim et al., and in U.S. application serial number 24,600 (March 1979) to F. Lim. The presently preferred method of producing such microcapsules produces semi-permeable polyamide membranes by an intenacial polycondensation technique.Mutually immiscible solvents or solvent systems are selected, e.g. water and a cyclohexane based solvent, and one monomer of a complementary pair while form a copolymer is dissolved in the water together with the material to be encapsulated. The aqueous solvent containing the material to be encapsulated is then emulsified within the other solvent to form a plurality of discrete droplets. The second, complementary monomer is next added to the continuous phase of the emulsion to initiate polymerization about the droplets at the phase boundary. Membrane permeability and uniformity of polymer depositions are controlled by varying the affinity of the continuous phase of the emulsion for the encapsulated during the course of polymerization and by controlling the concentration of the reacting monomers and the duration of the polymerization.

In one approach, the continuous phase at the outset is a solvent or solvent system having a relatively high affinity for the encapsulated monomer so that, in a first stage of polymerization, a relatively thick polymer network is produced about the droplets. Thereafter, the continuous phase is altered such that its affinity for the first monomer is decreased, e.g., by diluting the continuous phase with a second solvent or by replacing the continuous phase with a fresh solvent. Upon the addition of second monomer, further polymerization occurs preferentially within the initially deposited polymer network, patching macroporous defects and resulting in uniform capsule membranes which allow diffusion of solutes below a certain molecular weight.

In another approach, the continuous phase at the outset is selected to have a low affinity for the encapsulated monomer so that thin, relatively dense membranes form in a first stage of polymerization.

Thereafter, the affinity of the continuous phase for the encapsulated monomer is increased to draw further quantities of monomer th rough the membrane and to deposit a second outer layer of insoluble polymer.

When the discontinuous aqueous droplet phase is buffered to provide a compatible environment for liable biological materials such as an antibody, the encapsulation can be conducted in a manner to preserve a large percentage of the liable material's biological activity. The operability of the encapsulated material is also preserved by adding second monomer to the continuous phase in increments over the duration of the polymerization so that is concentration at any given time is relatively low and the antibody is not exposed to high concentrations of potentially destructive substances.

In a preferred reaction system, aqueous droplets containing a diamine, a high molecular weight filler material, and the antibody are produced in a continuous phase of cyclohexane whose affinity for the monomer dissolved in the droplet phase is modified by the addition of chloroform as a diluent. The addition of a diacid halide to the system results in the formation of semipermeable polyamide microcapsules. This microencapsulation approach is effective for producing membranes having an upper limit of permeability in the 200G-30,000 dalton molecular weight range. Thus, the capsules can easily be engineered to permit the diffusion of many hormones.

To conduct the assay, the test sample is mixed with microcapsules of the type described and incubated, preferably at about 37"C. Prior to the incubation, the capsules may be suspended in a solution of the species analogue of sufficient concentration to load the antibody with a detectably amount of the analogue concentration. However, the assay can be conducted by adding the analogue to the reaction system during or after the incubation with serum. Protein in the sample and protein-species complexes cannot traverse the microcapsule wall.

Next, the antibody is separated from the remainder of the reaction system (optionally exclusive of the microcapsule membranes) and either the antibody or the remainder of the system is assayed for analogue.

In a preferred method of making the separation, a high molecularweight hydrophilic material such as a solution of polyethylenimine or serum albumin is added to the extracapsular volume. This induces an osmolality increase resulting in collapse of the microcapsules and migration of intracapsular unbound hormone and its analogue into the supernatant. The result is a packed pellet of encapsulated antibody which, after an optional centrifuge treatment, may be isolated by aspirating or decanting the supernatant.

Alternatively, the separation may be conducted by washing free species and analogue from the capsule.

Tables 1 and 2, and corresponding Figures 1 and 2 of the drawing disclose the results of assays embodying the invention designed to detect the presence of free T4 using free T4 antibody, 1251 labled thyroxine as an analogue, and solutions of known free T4 concentration. Tests of unknowns run in parallel with the procedure used to gather this data may be interpreted by reference to the standard curves.

Table 1 Free T4 Assay Standard Curve Free T4 CPM MEAN CPM 0.5 ng/dl 57097 57,285 57472 1.3 ng/dl 54256 54,839 54717 3.0 ng/dl 50960 50,953 50946 5.0 ng/dl 48487 48,302 48117 7.7 ng/dl 45982 46,025 46067 TOTAL COUNT T4 (1251) ADDED 86,403 incubation 2 hours, 37"C TABLE 2 Free T4 Assay Standard Curve A VG.CPM 0.1 ng/dl 61219 60398 60933 61183 0.5 ng/dl 58020 55618 56675 56389 1.0 ng/dl 52096 52671 52824 53705 2.0 ng/dl 48483 50126 49536 49971 4.0 ng/dl 44853 45108 44732 44235 6.0 ng/dl 42008 41344 41456 41015 TOTAL COUNT T4 (125I) ADDED 86,903 2 hour incubation, 37 C Example I - Thyroxine Assay Preparation of Microcapsules Hexanediamine carbonate (pH = 8.5 + 0.1) solution is prepared by mixing 17.7 ml 1,6 hexanediamine with 32 ml of water, and bubbling CO2 through the solution for about 1 hour or until the pH level is reached.

Terephthaloyl chloride (TCl) solution is prepared by adding 20 g TCI in 200 ml of organic solvent consisting of 4 parts cyclohexane and 1 part chloroform. TCI is dissolved by stirring vigorously, and the solution is then centrifuged for 10 minutes at 2600 rpm. Any precipitate is discarded.

750 ml cyclohexane are mixed with 125 ml SPAN-85 (emulsifier, fatty acid ester or sorbitan) in a 2-liter mixter equipped with a magnetic stirring bar. While stirring, a mixed solution made from one ml antiserum to thyroxine (4% in phosphate buffered saline, R.F. Laboratories, Houston, Texas, or Radioassay Systems Laboratories, Carson, Calfornia), 25 ml of polyvinyl pyrrolidone - 4% bovine serum albumin, and 30 ml of hexanediamine carbonate solution is added to the cyclohexane. When droplets of the desired size have been produced, 70 ml TCI solution are added. Thirty seconds later, 37.5 ml of TCI are added. Sixty seconds later, 25 ml of chloroform are added. Three additional 25 ml aliquots of chloroform are added at 30 second intervals.

The microcapsules are recovered by centrifuging the two-phase reaction system, decanting the supernatant, and mixing the capsules with TWEEN-20 (polyoxyethylene derivative of fatty acid partial ester of sorbitol anhydride - emulsifier buffered with NaHCO3) and phosphate buffered saline. The capsules retain the polyvinylpyrrolidone and bovine serum albumin filler materials, as well as the thyroxine antibody.

Saturation of Antibody with Analogue Microcapsules made in accordance with the foregoing procedure may be loaded with 1251 labeled thyroxine (Cambridge Nuclear Corporation, Billerica, Massachusetts) by the following steps.

1. Add to each of 100 standard tubes 0.5 ml of microcapsul suspension and 1.0 microCurie of T4 (1251) (high specific acitivity of 5-6000 micro Ci per microgram). Allow to incubate at 37"C for at least thirty (30) minutes.

2. Wash the microcapsules with twice their volume of phosphate buffered saline (0.15 M NaC1, pH = 7.5, 0.015 M phosphate buffer).

3. Centrifuge at 2000 xg for 15 minutes and decant supernatant.

4. Repeat steps 2 and 3 twice.

5. Dilute microcapsule suspension with 1.6 times their volume of the phosphate buffered saline disclosed above. Total volume equals 80 ml. 0.8 ml of microcapsule suspension are used per test; thus 100 tests may be conducted with the capsules.

Test Procedure 1) Place 25 microliter test samples and 5 samples of known free T4 concentration in separate tubes. In the standard curve from Table 1, concentrations of 0.5, 1.3,3.0, 5.0 and 7.7 ng% of free T4 were used, but any series of free T4 concentrations may be adapted according to well-known experimental techniques. A control tube containing 25 microliters of saline may also be included as a further check on assay accuracy.

2) Pipette 800 microliter of T4 (1/81/45/81) pre-saturated microcapsules (supplied as such) into each tube.

3) Vortex each tube and incubate for 120 minutes at 37or.

4) After incubation, add 1.0 ml of 2.0% polyethyleneimine (m.w. 40-50 thousand) in phosphate buffered salineto each tube.

5) Incubate for an additional 20 minutes.

6) Decant supernatants.

7) Count each tube for one minute in a gamma counter.

Calculation of Results 1) Each time an assay is run for determination of unknown free-T4 concentration in a sample(s), standards to prepare the standard curve shouid be run.

2) Upon completion of the assay, a standard curve such as shown in Figure 2 or 3 is prepared using values obtained from the standards which were assayed concurrently with unknown samples.

3) Counts per munite (cpm) for each value can be plotted in the linear scale of 2 cycle semilog graph paper versus free-T4 concentration in nanogram percent on the log scale.

An alternative to plotting cpm v. free T4 concentration is to plot percent bound (relative) v. free T4 concentration. This can be accomplished by calculating the percent bound (relative) for each standard, control, or unknown and plotting these values on two cycle semilog paper in a manner similar to that described previously for cpm. Percent bound (relative) is calculated as follows: percent bound (reiative) = cpm bound/mean cpm bound of standard of lowest free T4 concentration.

Figure 4 is a graph of free T4 concentration in ng% of about 200 test samples, each of which were assayed by the method of this invention and the dialysis method. As shown, there is a high degree of correlation between the two test methods.

Figure 5 is a graph of frequency of a given free T4 concentration vs. free T4 concentration based on some 200 test samples assayed in accordance with the procedure set forth above.

Figure 6 is a graph of free T4 concentration in ng% of about 100 test samples, each of which were assayed by the method of this invention and the kinetic radioassay technique. As shown, there is a high degree of correlation between the two test methods.

Table 3 shows the consistency of the results intra assay.

TABLE 3 Intra Assay Variation (Values in ngidlJ Level A (1.2) Level B (2.0) Level C {5.3) 1.5 2.3 3.6 1.3 2.1 4.2 1.1 2.2 4.2 1.1 2.1 3.3 1.2 2.1 3.3 1.3 2.0 3.9 1.3 1.7 4.1 1.4 2.1 3.7 1.2 2.5 3.5 1.3 1.8 3.5 1.3 2.2 4.4 1.3 2.1 3.9 1.6 2.3 4.6 1.2 2.5 4.6 1.3 2.3 4.4 1.2 2.1 4.2 1.2 2.1 3.4 1.3 2.3 3.8 1.5 2.1 4.2 1.4 2.2 4.6 1.4 2.1 3.7 1.4 2.1 4.0 1.4 2.1 4.0 1.4 2.3 4.3 1.4 2.3 4.0 1.6 2.4 4.0 1.3 - 2.0 4.2 1.2 2.1 4.2 2.4 4.5 Coefficient of Variation Numberx + S.D. C. V.

A 28 1.34i.13 9.6% B 29 2.2 i .17 7.7% C 29 4.1 i .38 9.3% Table 4 shows the consistency of results, interassay.

TABLE 4 Interassay Variation: (values in ngidl) Test# LevelA (1.2) Level B F2.0) Level C(5.3) 1 1.3 2.4 4.6 1.2 1.8 4.4 2 1.4 1.5 4.6 1.5 2.6 4.7 1.2 2.0 4.0 1.3 2.0 3.7 1.4 2.5 3.8 1.2 1.9 4.0 3 1.3 2.2 3.9 4 1.4 2.0 3.2 1.2 2.1 4.0 5 1.5 1.8 4.2 6 1.3 2.0 4.2 1.2 2.0 4.5 7 1.3 2.1 3.5 1.35 2.5 3.7 1.1 1.7 4.0 1.2 2.0 3.5 1.4 4.2 1.4 4.0 Coefficient of Variation Number x + S.D. C.V.

A 20 1.32 #.11 8.41% B 18 2.1 +.27 13% C 20 4.0 #.4 10% Example Il - Dicoxin Assay A procedure for encapsulating antibody specific to digoxin is set forth below.

The reagents used and their suppliers are listed below: Sodium Bicarbonate Cyclohexane Chloroform All Fisher Chemical Sodium Chloride Sodium Phosphate, monobasic Sodium Phosphate, dibasic 1 ,6-Hexanediamine tSpan-85 Tween-20 Ruger Chemical, New Jersey Anti-Digoxin Rabbit Antiserum Arnel Products, Brooklyn, New York Bovine Serum Albumin Comassie Brilliant Blue R Sigma Chemical Polyvinylpyrrolidone-40 Aldrich Chemical Terephthaloyl Chloride Eastman Kodak The quantity of each ingredient is as follows: 1,6 Hexanediamine Carbonate 30 ml Phosphate Buffered Saline 40 ml Polyvinylpyrrolidone 40/Comassie Blue 25 m Anti-Digoxin Rabbit Antiserum 5 ml Cyclohexane, ACS 750 ml SPAN-85 125 ml Terephthaloyl Chloride Solution 107.5 ml Chloroform 100 ml Tween-20 Wash Solution Q.S. 200 ml Phosphate Buffered Saline Q.S. 81 Procedure: Rinse all glassware in distilled water prior to use.

1. Place 2 liter glass mixer in hood on magnetic stirrer.

2. On bench adjacent to hood, setup microscope.

3. In 250 ml glass graduated cylinder, carefully measure 125 ml SPAN-85.

4. Pour measured SPAN-85 into glass mixer in hood.

5. Measure 750 ml cyclohexane in a 1000 ml glass graduated cylinder. Pour into glass mixer in hood.

6. Place cover on mixer.

7. Turn on magnetic stirrer.

8. Measure: 30 ml Hexanedianmine Carbonate; 30 ml PBS: 7/45/8 ML1/85/82/3 PVP/Comassie Blue/4% BSA; 5 ml Antibody; mix the 40 ml PBS with the 5 ml Antibody in a 50 graduated cylinder.

9. Put a 2-inch stir bar with spin ring in a 400 ml beaker. Place on magnetic stirrer.

10. To beaker add 30 ml hexanedianmine. Start magnetic stirrer.

11. To Hexanediamine in beaker add in the following specified order: 25 ml 15% PVP/Comassie Blue/4% BSA; 35 ml PBS/Antibody solution. Let mix for 2 min. Let solution mix for three (3) min.

12. While solution is mixing, measure one 70 ml portion and one 37.5 ml portion of TCL in separate 100 ml glass graduated cylinders. Cover with watch glass and set in hood. Measure four (4) 25 ml portions of chloroform in separate 25 ml glass graduated cylinders. Cover with watch glasses and set inside hood.

13. Through side arm of vial, add contents of 400 ml beaker to glass mixer in hood.

14. As rapidly as possible, with a disposable glass 1 ml pipette, take a sample of the solution in the mixer and put it on the microscope slide. Check to determine that the droplets are of an acceptable size. (10-80 microns in diameter).

15. When the droplets are of acceptable size T=0, add measured 70 ml of TCL through side arm of mixer.

AtT=30, exactly 30 seconds later, add second 37.5 ml portion of TCL to mixer through side arm. Let mix exactly 60 seconds.

16. At 60 seconds (T=90) add first 25 ml portion of chloroform. Mix 30 seconds. (T=120) Add second measured 25 ml chloroform, mix 30 seconds (T=150). Add third measured 25 ml chloroform. Mix 30 seconds (T=180). Add fourth measured 25 ml chloroform. Mix exactly 30 seconds (T=-210"). Stop mixer.

17. Pour contents of mixer into two 1 liter plastic centrifuge bottles which have been rinsed in distilled water 3 times. Centrifuge at 500 RPM for three minutes.

18. Carefully decant supernatant.

19. To each bottle add approximately 50 ml of Tween-20 solution (i.e. approximately the same volume Tween-20 as capsules). Mix well with stir bar retriever for about 5 min.

20. Add approximately 10-15 mls PBS to each bottle. Stir well.

21. Repeat step 204-5 times.

22. Add 400 mls PBS. Mix well.

23. Balance and centrifuge bottles for 20 min. at 3000 RPM. Aspirate supernatant. Add approximately 800 ml PBS to each bottle. Stir well and cap.

24. Repeat step 23, 10 times. After final aspiration add 100 ml of PBS to each bottle and combine contents of both bottles. Shake well. Pour into 500 ml glass graduated cylinder and Q.S. to 500 ml with PBS.

25. Store in 1 litre glass reagent bottle at 4 C.

The Tween-20 wash solution is prepared as follows: Procedure: Can be made day before use 1. Weigh 6.06 g sodium bicarbonte on triple beam balance.

2. Carefully put 6.06 g sodium bicarbonate into 250 ml volumetric flask containing stir bar.

3. Add approximately 200 ml purified water and stir on magnetic stirrer until sodium bicarbonate is dissolved.

4. Remove stir bar and Q.S. to 250 ml with purified water.

5. Carefully pour into 500 ml Erlenmeyer flask containing stir bar.

6. Carefully measure 250 ml Tween-20 in a 250 ml graduated cylinder.

7.. Add to Erlenmeyer flask containing sodium bicarbonte solution.

8. Mix on magnetic stirrer until completely mixed (approximately 1 hour).

9. Store tightly sealed in polyethylene 1 L bottle with screw cap at approximately 25"C.

Theterephthaloyl chloride solution is made as follows: Should be made day of use. DO NOT REFRIGERATE.

1. Weight of terephthaloyl chloride (TCl( IS RECORDED ON BOTTLE CONTAINING TCI. Multiply weight of TCI in grams by ten to calculate volume in milliters cyclohexane/chloroform solution to add to TCI.

Calculations: g. TCI x 10 = ml total volume cyclohexane/chloroform solution. Be sure total volume is sufficient for procedure being followed.

2. The cyclohexane/chloroform solution is four parts cyclohexane and, part chloroform. Divide the total volume cyclohexane/chloroform solution (Step 1 above) by five to calculate the volume chloroform. Multiply the chloroform volume by four to determine the cyclohexane volume.

(a) ml total volume - 5 = ml chloroform.

cyclohexane/chloroform (b) ml chloroform x 4 = ml cyclohexane.

3. Carefully measure the calculated volmes cyclohexane and chloroform in graduated cylinders.

Combine in Erlenmeyer flask. Swirl gently to mix. Cover with watch glass. Put in fume hood.

4. Put magnetic stirrer in fume hood.

5. Put bottle containing TCI on magnetic stirrer. Open bottle and as quickly as possible add a magnetic stir bar and cyclohexane/chloroform solution. Replace cap on bottle.

6. Stir on magnetic stirrer until al TCI is dissolved. It may be necessary to tip bottle to dissolve any TCI around top of bottle.

7. As quickly as possible, equally fill as many 200 ml glass centrifuge bottles as necessary with TCI solution and cap. Centrifuge 10 min. at 2600 rpm in room temperature centrifuge.

8. Pour supernatant in 500 ml amber bottles and seal well.

9. Store, tightly sealed, at approximately 25"C.

The procedure for preparing 15% PVP 40/Comassie Blue with 4% BSA is as follows: Procedure: Make day of use. DO NOT REFRIGERATE.

1. Accurately weigh on triple beam balance 7.5 gm Polyvinylpyrrolidone 40; 2 g BSA; and .1 g (100 mg) Comassie Blue.

2. Put polyvinylpyrrolidone 40 into 50 ml glass beaker with stir bar.

3. Add approximately 20 ml PBS. Put glass cover plate on beaker.

4. Stir on magnetic stirrer until dissolved.

5. Put .1 g Comassie Blue and 2 g BSA into another 50 ml glass beaker.

6. Add approximately 20 ml PBS. Put glass cover plate on beaker.

7. Stir and heat slightly using magnetic hot plate/stirrer #10 with heating unit set on 2. Let stir until dissolved (approximately 10 minutes).

8. Carefully add entire contents of beakers containing polyvinylpyrrolidone 40 solution and Comassie Blue solution to a 50 ml glass volumetric flask with stir bar.

9. Mix combined solutions for 10 minutes on magnetic stirrer. Remove stir bar.

10. Q.S. to 50 ml with PBS.

11. adjust to pH = 7.5 + 0.5 with 1 N sodium hydroxide.

Orig. pH final pH Amt. 1 N NaOH used 12. Filter solution with Nalgene disposable membrane filter unit (0.45 u).

13. Store sealed in 60 ml polyethylene bottle with screw cap at approximately 25"C.

The procedure for preparing phosphate buffered saline is as follows: Procedure: Can be made day before use.

1. In 1000 ml glass graduated cylinder, carefully measure 1000 ml phosphate buffered saline stock.

2. Pour into 20 L polyethylene carboy.

3. Measure, in 1000 ml glass graduated cylinder, 9000 ml deionized water and add to carboy containing phosphate buffered saline stock.

4. Stir using magnetic stir bar retriever.

5. Check pH. If necessary, adjust pH to 7.5 + 0.05. Final pH= 6. Store phosphate buffered saline in tightly sealed 20 liter polyethylene carboy. Store at approximately 25"C until used.

The procedure for making 1,6 Hexanediamine Carbonate is as follows: Procedure: Can be made day before use.

1. Place bottle of 1,6 hexanediamine in 3 liter beaker. Add enough tap water to the beaker to reach level of hexanediamine in bottle.

2. Loosen cap of hexanediamine bottle.

3. Place beaker on magnetic stirrer/hot plate with heat setting at 2 until hexanediamine is completely melted.

4. In graduated glass 25 ml cylinder, accurately measure 17.7 ml hexanediamine. Pour carefully into 500 ml amber bottle.

5. Accurately measure 32 ml purified water in 50 ml glass graduated cylinder. Add to hexanediamine in amber bottle.

6. Bubble CO2 through solution for approximately 1 hr. until pH = 8.5 + 0.1. Final pH 7. Seal amber vial and store at approximately 25"C.

The microcapsule and principle of its operation are set forth diagramatically in Figure 9. As is shown in Figure 9, antibody 9 which is specific to digoxin is trapped within wall 10 of microcapsule 12. Walls 10 of the microcapsule 12, however, have openings 14 which are small enough to permit the free passage of digoxin.

Thus, digoxin 16 from the sample and labeled digoxin 18 are free to pass through the walls of the microcapsules and compete with each other for sites on antibody 9. Any unbound digoxin 16 or 18 can be washed from within the microcapsule after incubation is completed.

As is set forth above, some of the digoxin is labeled and it is the labeled digoxin that competes with the unlabeled digoxin from the sample for sites on the antibody specific to digoxin. The preferred labeled digoxin is (1251) digoxin which may be obtained from New England Nuclear Corporation, Billerica, Massachusetts.

Test Procedure 1. A sample or samples of serum to be tested is obtained through procedure well known to the art. In this example controls are used.

2. Pipet 0.1 ml (100 R i) of each standard, control or unknown patient serum sample into correspondingly labeled tubes containing 0.5 ml of digoxin antibody microcapsule suspension as prepared above.

3. Pipet 0.5 (500 it I) of 1251 labeled digoxin (in phosphate-buffered saline) into each tube. Vortex for 3-4 seconds minimum.

4. Incubate all reaction tubes in a water bath (37"C) for at least fifteen (15) minutes. [Alternatively incubate all reaction tubes at room temperature (20-25 C) for at least thirty (30) minutes 5. Add 0.5 ml (500 , I) of wash solution to each tube. Vortex for 3-4 seconds minimum.

6. Incubate all reaction tubes at room temperature (20-250C) for five (5) minutes.

7. Centrifuge alltubes at 1600 xg minimum, for ten (10) minutes, room temperature (20-25"C), and decant supernatant into an appropriate waste container catching the last drop on blotter paper.

8. Count all tubes in a gamma counter, adjusted for 125, for one (1) minute each.

9. Plot standard curve of known amounts of digoxin v. cpm and read unknown digoxin amounts from this curve.

Reagents Used The digoxin-antibody is encapsulated in semi-permeable nylon microcapsules suspended in phosphate buffered saline as prepared above. For quality control, the antibody microcapsules contain blue dye (Comassie Blue R250). A blue supernatant, after the microcapsules have settled out or sedimented via centrifugation, would be indicative of microcapsule breakage and possible antibody leakage.

725, Labeled Digoxin Each ml of solution contains 10 g of 1251 digoxin of less than 4.2 , Ci.

Buffer Solution 0.015 M phosphate buffer, pH 7.5 containing 0.15 M sodium chloride 0.5% Bovine Serum Albumin (BSA), 0.1% sodium azide.

Wash Solution Polyethlene glycol 6000,20% solution in 0.5 M barbital buffer, pH 8.9 containing 0.15 M sodium chloride.

Calculation of the Results The results of this experiment a typical test are shown in Table 5 and depicted graphically in Figures 7 and 8.

In Figure 7 CPM is plotted on the linear scale of 2 cycle semilog graph paper against the concentration of Digoxin in ng/ml (nanograms/milliliter) on the log scale. An alternative to plotting CPM vs. digoxin concentration is to plot % bound (relative) vs. digoxin concentration (see Figure 8). This can be accomplished by calculating the % bound (relative) for each standard, control, or unknown, and plotting these values on two cycle semilog paper in a manner similar to that described previously for CPM. Per cent bound (relative) is calculated as follows: % bound (relative) = (Mean CPM bound of 0 ng/ml digoxin standard) x 100.

Table 5 ngiml CPM Bound Digoxin 1 2 Standards 0 16,888 17,136 0.5 14,464 14,473 1.0 12,279 12,192 2.0 9,690 9,717 4.0 7,344 7,445 Controls 1 10,360 10,486 2 8,226 8,380 CPM Total = 26,780 Digoxin Relative % Bound (CPM Value Mean Bound/CPM Bound)* x 100 ngiml 17,012 100.0 14,469 85.0 12,236 72.0 9,704 57.0 7,395 43.0 19,423 61.3 1.7 8,303 48.8 3.1 *Bound = CPMBound at 0 ngimi digoxin The assay system of the instant process has been found to be quite precise. The intra assay variation is less than 7% and the interassay variation is 1055 than 9%.

Intra Assay Variation The coefficients of variation, CV, for two control samples, each assayed twenty4ive (25) times within one experiment were found to be Mean Digoxin ngiml CV Control 1 1.54 4.9% Control 2 1.95 6.2% InterAssa y Variation The coefficients of variation for two control samples, each assayed three times in eighteen separate experiments were found to be: Mean digoxin ngiml CV Control 1 1.50 7.3% Control 2 3.04 88% Table 6 shows the extremely high specificity of this assay for digoxin and to the exclusion of other potentially interfering substances.

Table 6 Specificity % Cross-reactivity of various biochemical compounds with microencapsulated antidigoxin antisera Compound % Cross Reactivity Digoxin 100.00 Digitoxin 0.80 Progesterone 0.16 Cortisol 0.10 Testosterone 0.08 Dehydroandrosterone sulfate 0.08 Cholesterol 0.06 Quabain 0.02 Quantative Recovery The instant process has been found to be highly quantative, reflecting no less than 96% recovery of the digoxin samples added.

TABLE 7 Recovery Study Initial Total Measured Digoxin Level Digoxin Digoxin Digoxin nglml Added ngiml ngiml %Recovery 0.64 0.5 1.14 1.13 99 0.64 1.0 1.64 1.72 105 0.64 2.0 2.64 2.70 102 0.64 4.0 4.64 J 4.45 96 It is of particular importance to note the high degree of correlation of the results of the instant process with digoxin determinations as made by other assay methods.

Table 8 expresses comparative test results obtained by the process of this invention versus those of other assays.

The systems A and C represent liquid assay systems with precipitating reagent separation, while system B utilizes solid phase separation. As shown, there is a 0.97 - 0.98 correlation coefficient.

TABLE 8 AGREEMENT FOR VARIOUS ASSAY PROCEDURES Coefficients of correlation for 135 samples assayed by System A, System B, System C, and the Present System were: n r* y System A 135 0.98 1.12 x -0.1 System B 135 0.97 0.90x-0.02 System C 135 0.97 0.96 x +0 Present System 135 0.97 0.95 x +0.14 * r = coefficient of correlation In view of the foregoing, it is apparent that the assay technique disclosed herein may be used to determine the presence and concentration of any free species, provided that a complementary substance capable of specific binding with the species and a distinguishable analogue of the species are available.

The molecular weight of the species and of its analogue be sufficiently low so that it is feasible to provide microcapsule membranes which selectively allow diffusion of these substances while preventing passage of high molecular weight materials such as natural proteins. Fortunately, steroid hormones, thyroid hormones, many drugs and other classes of substances of clinical importance are characterized by molecular dimensions far smaller than natural proteins.

Claims (23)

1. A process for determining in a protein-containing sample the presence of a species unbound to protein, said process comprising the steps of: (a) incubating the sample with an antibody complementary to said species and a distinguishable analogue of said species, the sample and antibody being separated by membrane having a porosity sufficient to allow passage of said species and its distinguishable analogue, but insufficient to allow passage of said antibody or protein-bound species present in the sample; (b) allowing unbound species in the sample to traverse said membrane and to compete with its distinguishable analogue for binding sites on said antibody; (c) separating said antibody from residual unbound species and unbound analogue; and (d) determining the amount of said distinguishable analogue which is bound to said antibody or which is antibody free, said amount being indicative of the amount of species unbound to protein in the sample.

2. A process according to claim 1, wherein said antibody and distinguishable analogue are contained within microcapsules comprising said membrane; said separation step is effected by removing unbound species and said analogue from said microcapsules and separating said microcapsules from the remainder of the reaction system; and the amount of said analogue which is bound to said antibody is determined.

3. A process according to claim 2, wherein said separation step comprises inducing an osmolality change in the reaction system which collapses the microcapsules.

4. A process according to claim 2, wherein said separation step comprises washing unbound species and unbound analogue from said microcapsules.

5. A process according to any one of the preceding claims, wherein said distinguishable analogue comprises said species tagged with a radioactive atom.

6. A process according to any one of claims 1 to 4, wherein said species is L-thyroxine.

7. A process according to any one of claims 1 to 4, wherein said species is L-3,5,3' tri-iodothyronine.

8. A process according to any one of claims 1 to 4, wherein said species is cortisol, testosterone, or neonatal thyroxine.

9. A process according to any one of claims 1 to 4, wherein said species is digoxin.

10. A test set for use in determining in a liquid sample the amount of an unbound species, said test set comprising; (a) a distinguishable analogue of said species; (b) a plurality of microcapsules adapted for use in absorbing unbound species from said sample, said microcapsules containing antibody complementary to said species, and comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody or natural proteins; and (c) a standard containing a predetermined amount of said species for comparison with a sample.

11. A test set according to claim 10, wherein said species is thyroxine, said distinguishable analogue is radioactively labeled thyroxine, and said standard comprises aliquots of material containing known quantities of free-thyroxine which, when tested in parallel with the sample, allow the construction of a standard curve.

12. A test set according to claim 10, wherein said microcapsules contain antibody to digoxin and said distinguishable analogue is 125j labeled digoxin.

13. Atest set according to any one of claims 10, 11 and 12, also comprising a reagent capable of removing unbound species and its unbound distinguishable analogue from said microcapsules.

14. A test set according to claim 13, wherein said reagent comprises a serum albumin or polyethyleneimine.

15. A reagent adapted for use in the determination of unbound species in a liquid sample comprising a plurality of microcapsules containing; (a) a distinguishable analogue of a species which species is capable of antibody formation, and (b) antibody complementary to said species, said microcapsules comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody and natural proteins.

16. A two-part reagent adapted for use in the determination of unbound species in a liquid sample which reagent comprises a part A comprising a distinguishable analogue of a species which species is capable of antibody formation and a part B comprising a plurality of microcapsules containing antibody complemen tary to said species, said microcapsules comprising membranes of a permeability sufficient to allow passage of said species and its distinguishable analogue but insufficient to allow passage of said antibody and natural proteins.

17. A reagent according to claim 15 or claim 16, wherein said distinguishable analogue is an analogue of a species capable of reversibly binding with protein in a liquid sample.

18. A reagent according to claim 15 or claim 16, wherein said distinguishable analogue is radioactively labeled thyroxine and said antibody is complementary to thyroxine.

19. A reagent according to claim 15 or claim 16, wherein said distinguishable analogue is radioactively labeled digoxin and said antibody is complementary to digoxin.

20. Microcapsules containing antibody complementary to a species which is capable of antibody formation.

21. An assay process substantially as hereinbefore described with reference to Example I or Example II.

22. A micro-encapsulated antibody substantially as hereinbefore described with reference to Example I or Example II.

23. A process, test set reagent or microcapsules according to any one of the preceding claims modified in that the antibody is replaced by another substance capable of reversibly binding with the species.