Radioimmunoassay and related methods - Analytical Chemistry (ACS

---

Radioimmunoassay and Related Methods Research in endocrinology owes much to the development of biological assays for many hormones in the first half of this century. Despite numerous improvements, most of these bioassays still have deficiencies. They are usually too insensitive to permit direct, accurate detection of hormone levels in blood or other physiologic fluids. They require large groups of animals, with factors such as age, sex, weight, strain, and diet being carefully controlled. Frequently, because of inter-assay and intra-assay variations, even larger animal groups and statistical analyses are required to detect significant responses. Finally, bioassays are generally not amenable to the assay of large numbers of specimens. With the onset of subcellular fractionation methods, numerous in vitro bioassays for hormones were developed which employed responses of certain metabolic pathways to these agents. However, they often had the disadvantage of nonspecific effects and the difficulty of equating the in vitro and in vivo responses. Although more sensitive than the animal bioassays, they were generally not useful for blood hormone determinations. For similar reasons instrumental methods such as gas chromatography, although capable of detecting steroid and thyroid hormones, have generally not been amenable to assaying large numbers of biological specimens and have not been applicable to protein hormones. In 1959 Rosalyn S. Yalow and the late Solomon A. Berson announced a radioimmunoassay procedure for the detection of insulin in human plasma. This pioneering work initiated the rise of a new era in endocrinology, an era which has yet to reach its zenith. One only has to glance a t the endocrine journals, or papers presented a t endocrine meetings, to immediately appreciate the impact of Berson and Yalow’s accomplishment. Today, the 878 A

applications of the radioimmunoassay (RIA), or related techniques such as the competitive protein-binding (CPB) assay, radioenzymatic assay, radioreceptor assay, and immunoradiometric assay, are by no means restricted to endocrinology. Any substance which is of sufficiently low concentration to make detection by biological, chemical, or instrumental means difficult, if not impossible, is a candidate for this type of assay. The materials being assayed by these procedures include protein and nonprotein hormones, vitamins, nucleic acids, enzymes, drugs, metabolites, cancer antigens, viral antigens, antibodies, and structural proteins. A computerized literature search revealed over 3000 English language publications on this subject since 1964. This area has been the subject of numerous reviews, symposia, and workshops. The limited space of this report will be used to consider the basic principles and methodologies and to give examples of the numerous applications.

Principle The RIA and related assays depend on the same principle. As shown in Figure 1,a compound to be measured (C) and the same compound tagged with a radioactive tracer (C*) compete for binding to or reaction with a reactive agent. In the RIA this agent is an antibody and is depicted in Figure 1 as AB, In the following discussion of the principle, only AB will be used, although it is interchangeable with the reactive agents of the other related assays. The competitive principle is quantitated by determining the amount of radioactivity (C*) in two pools: that which has bound to AB and is called Bound (B), and that which has not bound and is called Free (F). Increasing concentrations of C result in concomitant decreases in B and increases

ANALYTICAL CHEMISTRY, VOL. 45, NO. 11, SEPTEMBER 1973

Charles D. Hawker Laboratory Procedure@ Division of The Upjohn Co. Kalamazoo, Mich. 49001

in F. Therefore, the B / F ratio (alternately, the percent of B in the B F total) decreases as the concentration of C increases. Figure 2 shows the RIA principle more schematically. Test tubes are depicted with constant amounts of C* and AB, and the arrows indicate the incubation and separation of B and F pools: the tubes on the left show the initial assay components, and the corresponding tubes on the right show the separated B and F pools after reaction. In the top example, no standard or test compound (C) is added; all of the C* reacts and is found in the B pool. In the other two cases, addition of different amounts of C results in binding of C and C* to AB in proportion to the C/C* ratio. Since the C/C* ratio is the same in both the B and F pools, the C*-AB/C* ratio (or B/F) is dependent on the amount of C present. Figure 3 shows a typical RIA standard curve which displays this doseresponse relationship. The percent bound is plotted as a function of the quantity of the hormone testosterone (C) per assay tube. In this type of plot, the B pool for a tube with no C is set equal to 100%. As increasing amounts of C are added, the % bound decreases. A similar curve is obtained with the B/F ratio or the B/(B + F) ratio instead of YObound on the ordinate. This curve displays the doseresponse relationship suggested in Figure 2. By comparison of the % bound for an unknown specimen with the curve, the concentration of C in

+

Report

e*

+

C*-A

AB

+

(FREE)

(BOUND)

C

TI C-AB Figure 1. Schematic diagram of radioimmunoassay principle

the unknown can be determined. However, it is not always easy to fit the curve to the experimental data, and reading the lower concentrations can be troublesome because they are closer together. An alternate method of plotting the standard curve which avoids these difficulties has been described. The logit of the 70bound is plotted on the ordinate, and the log concentration is plotted on the abscissa. The logit function is described by the equation: logit % B

=

In

%B (100 - % B)

Figure 4 is a plot of logit % B vs. log C for the same data used in Figure 3. In this example and in most cases, the standard curve obtained with this method is linear. With a standard line, it is easier to fit the line to the experimental data and to read unknown concentrations with accuracy. Further, both of these steps can be computerized for automation of the RIA procedure, and numerous statistical analyses can be incorporated to derive confidence limits or other statistics for both the standard line and the unknown results. Graph paper which plots these functions directly from the 70bound and standard concentration is commercially available (Codex Book Co., Norwood, Mass.).

Basic R I A Requirements Pure Antigen. The antigen used for radioactive labeling and as the assay standard must be pure for the assay to be specific. When used for

immunizations, pure antigen may result in easier assay development, but it is not an absolute requirement, because the assay depends on competition between standard (or test) antigen and labeled antigen, and any antibody binding sites directed against other moieties do not enter into the reaction. Antiserum Preparation. Many species have been used to produce antisera; rabbits have been used most frequently, because they generally best combine the small animal advantages of a minimal immunogen requirement and inexpensive procurement and care with the large animal advantage of providing ample quantities of serum. It is important to distinguish between an antigen (a substance capable of binding to a specific antibody) and an immunogen ( a substance capable of provoking an immune response). Proteins of greater than 5000 mol wt can usually be both antigens and immunogens. However, smaller peptides and compounds such as drugs and nonprotein hormones, although antigenic, must be coupled to a large protein carrier such as albumin to be immunogenic. When coupled, the small compounds are called haptens, and the carrier-hapten complexes are called conjugates. Usually, the immunogen is mixed with an adjuvant, such as Freund’s complete adjuvant, which when injected serves to both enhance and prolong the immune response. A simultaneous injection with pertussis

vaccine also aids the response by acting as a nonspecific stimulator. Usually, for booster immunizations the pertussis vaccine is omitted, and incomplete adjuvant is substituted for complete adjuvant. In rabbits, intradermal injections in the groin region along the lymph system are an effective means of immunization. A total of 1ml of immunogen suspended in adjuvant is injected in as many as 50 sites, each receiving 20 ~1 of the suspension. The initial immunization yields “primary response” antibodies which are usually less specific. In a typical protocol, booster immunizations are given a t two-month intervals following the first immunization, and serum is collected every two weeks after the first booster. This serum will contain “secondary response” antibodies which will generally be more specific and have a higher titer (dilution at which serum can be used). Antisera must be evaluated for titer, sensitivity, and specificity; this requires the use of labeled pure antigen. The usual working titer is that dilution which binds 50% of the total labeled antigen; antisera with titers of less than 1 : l O O are generally not useful. High titers do not necessarily correlate with sensitivity. This is evaluated by running standard lines with unlabeled pure antigen at known concentrations. Different antisera will give parallel lines similar to the one shown in Figure 4,and the line farthest to the left will indicate the most sensitive antiserum. Sensitivity can vary among bleedings from the same animal; each bleeding should be checked. Specificity is evaluated by examining the degree of cross-reactivity with other compounds similar to the one to be measured. Standard lines are run for each compound, and the concentration of the test compound at 50% bound is divided by the concentration of the standard antigen a t 50% bound. The result is the frac-

ANALYTICAL CHEMISTRY, VOL. 45, NO. 11, SEPTEMBER 1973

079 A

Why gamble on FlDs when Gow-Mac is your best dollar value ?

tion of cross-reactivity for the test compound on a scale on which the standard antigen is 1.00. The concentration (if any) of the test compound in the specimen to be assayed determines whether this observed crossreactivity will significantly interfere in the assay. Radiolabeled Antigen. The original hormone RIA methods used radioiodine for labeling, because iodine could be substituted onto the protein tyrosine residues with relative ease and high specific activity. The chloramine T method, which has been used nearly exclusively since its introduction, made possible high specific activities while using smaller, safer amounts of isotope. High specific activities are important, because to achieve maximum sensitivity the concentration of labeled antigen must

be kept below the minimum concentration of antigen to be measured. In this method, radioiodide is oxidized with chloramine T to iodine which in turn reacts with the protein tyrosyl residues a t p H 7.5. The unreacted radioiodine is reduced to iodide with sodium metabisulfite, and the labeled protein is immediately separated from unreacted iodide and damaged protein fractions to protect it from further radiation damage. Originally, lS1I was employed for these iodinations, but it has largely been replaced by 1251,because the commercial preparations of 1251 are now of excellent quality, giving this isotope several advantages over 1311. The preparations have an isotopic abundance of about 99% compared to 1520% for 1s1I, and the longer half-life (60 days) of I25I compared t o 1311.

FREE

BOUND

The GOW-MAC 750 F I D Gas Chromatograph performs very well and it is economically priced. Sensitivity is as high as instruments costing many times more. You should know how this effect is achieved because nothing is sacrificed. Rather t h a n looking at the 750 as a lowpriced chromatograph, look at it as a high-priced flame ionization detector with high sensitivity, wide temperature range, good thermal stability, excellent reproducibility and a large column oven capacity. These features are achieved by optimizing the F I D performance. Temperature programming and multiple detector systems are omitted, but you don't need them for routine analyses. GOW-MAC'S application testing laboratory will r u n your samples and report back to you - no obligation.

GOW-MAC INSTRUMENT CO.

3c 6 C"

...OO

0.0.

nnn

nnn

I082

6C 6 C" 000.0. 0000..

FREE

nnn nnn

BOUND

Figure 2. Schematic representation of radioimmunoassay

CIRCLE 85 O N READER SERVICE CARD

880 A

BOUND

3

100 Kings Road, Madison, N.J. 07940 Telephone: 201 /377-3450 Telex: 136331

Shannon Free Airport, Co. Clare, Ireland Telephone: 61 632 Telex: 6254

FREE

ANALYTICAL C H E M I S T R Y , VOL. 45, NO. 1 1 , S E P T E M B E R 1973

EPA evaluated!* total organic carbon volatile organic carbon total carbon

Dohrmann’s DC-50 organic analyzer makes aii of these measurements accurately and rapidly. Based on proven methods, it avoids interferences and undesireable pyrolysis reactions that historically have resulted in significant errors.

R E L I A B L E M E A S U R E M E N T S . Determines important, lightweight volatiles such as low molecular weight alcohols and ketones, normally lost by acidification and sparging. FAST: 5 minutes per determination A C C U R A T E : Repeatability of f 1 mg/liter or f 2 % W I D E R A N G E : 1 to 2,000 mg/liter ( p p m ) without dilution P R I C E : $7,500, including start-up assistance and operator training

* “. .

,

D O H R M A N N Division, Envirotech Corporation, 1 0 6 2 Linda Vista Avenue, Mountain View, CA 94040 (415) 968-9710.

E N VIROTEC H

CIRCLE 54 ON READER SERVICE CARD

, Overall the unit performed in a most satisfactory manner. Detection limit was 0.5 to 1.0 mg C/1, working range was 1 to 2,000 mg C/1, with precision a n d accuracy comparable to approved m’ethodology . . . Excellent comparisons were obtained in practically all cases. Thus, t h e Dohrmann DC-50 appears to be a n acceptable alternative for measuring total organic carbon in water a n d wastes.”

“Newsletter, Analytical Quality Control” No. 15, October, 1972, AQCL, NERC, EPA, Cincinnati, Ohio 45268.

Figure 3. Standard curve for radioimmunoassay of testosterone

l'

m

;a 3b ;e

167 3i3 ti25 In58 TESTOSTERONE, w/TUBE

ZBOO so00 1a.00

Figure 4. Standard line for radioimmunoassay of testosterone

(eight days) results in less frequent iodinations. Nonprotein compounds assayed by RIA or CPB methods originally used 3H-labeled antigens which were usually commercially available. More recently, the use of iodine isotopes has been extended to these nonprotein compounds, such as digoxin, cyclic AMP, and steroid hormones by iodinating tyrosine methyl ester (TME) derivatives of the compounds. The CIRCLE 145

ON READER SERVICE CARD

specific activities of the iodinated compounds are much greater, and they can be counted with greater efficiency. Further, gamma counting is less complicated, faster, and more economical than liquid scintillation counting, although 3H does have advantages of being a less expensive isotope and not requiring laboratory licensing for its purchase a t the levels needed for an RIA. Purification of the iodinated anti-

Philips' n e w modular automation systems bring y o u quantitative x-ray diffraction & spectrometry: " a d d - o n " o r "start-from-scratch"

1.Add Stepping motor to control goniometer

Get Scan with slew return , . , Automatic data print out and command to step to next increment to initiate new measurement.

Modular options leading to c o m pletely automated systems for X-Ray Diffractometry and X-Ray Spectrometry are readily obtained using Philips' n e w QXD-QXS 300 Automation Systems. A d d o n to your existing equiDment, or start from scratch. Take these steps to quantitative analysis, a n d direct concentration readout:

2.Add Automatic Angle Programming

Get (With above) profile or peak mode programming. Profile mode: slew to specified lower 28,step to specified upper 28;you get a net integrated intensity print out; slew to next profile lower 2 8 . Peak mode: slew to 2 0 , take data, print out, slew to next 28. Print out data. etc. Store 64 addresses, random 2 0 and sector 28 collection.

3.Add Automatic Sample Changing

Get All-day, all-night unattended operation. Store 35 samples for the diffractometer.

4.Add Direct Concentration Readout Systems

Get Analysis time reduced from minutes to seconds. For Diffractometry: Integrated peak intensities, background subtracted and ratioing to an integrated intensity, either stored or to be measured; direct concentration readout. Automatic slewing to next address position, automatic initiation of next data process. X-ray programs supplied , For Spectrometry: Programs for ratio measurements: determine and store m's & b's for each element: interelement corrections; direct concentration readout. Automatic data collection printout and slewing to next address. X-ray programs supplied.

pi,

P

I

Call Bill Kiley, Product Manager, X-Ray D iff rac to met ry and Spec t romet ry today, 914-664-4500, he'll give you full detailsor arrange for a demonstration here, in our Application Laboratory.

CIRCLE 187 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL. 45, NO. 11, SEPTEMBER 1973

883 A

I -

/-

There are nearly 400 quality Hach reagents available to you Everything from acetic acid to zinc test reagent. Most Hach formulations are b a s e d o n M ard Methods. They're packaged in convenient premeasured powder pillows or bottles for m a x i m u m s t a b i l i t y and m i n i m u m dosage waste Here are five of the most popular Hach reagents. a Formula 2533-" Nitrification Inhibitor powder for BOD testing Gives more accurate evaluation of the effectiveness of wastewater treatment processes and a way of calculating potential nitrogenous oxygen demand i n samples a New FerroZineB iron analysis reagent. It's fast becoming the new standard in trace analysis for ultrahigh purity water FerroZine" offers the highest test sensitivity available today a PA0 for dissolved oxygen testing A time proven successor to sodium thiosulfate solution PA0 gives same accuracy with maximum stability. Eliminates the nuisance of checking and restandardizing titrating solution or making adjustment calculations to compensate for thiosulfate deterioration. a New Formula 258gT" Fluoride Adjustment Buffer powder Ideal for fluoride electrode determinations One powder pillow dosage buffers samples, controls ionic strength and frees fluoride in solution from interfering substances without diluting the sample. a Dithizone powder pillows. A breakthrough in colorimetric lead analysis Hach's new procedure provides a highly sensitive test with no common metal interferences Write for more information today.

@CHEMICAL

COMPANY

P.O. Box 907*Ames, Iowa 50010 U S A . Phone: 515-232-2533 TWX: 910-520-1158

gen from unreacted iodide and other fractions has been achieved in many ways. For protein compounds these include gel filtration, adsorption on powdered cellulose or microfine silica particles, starch gel electrophoresis, and ion-exchange chromatography. The iodinated TME derivatives of nonprotein compounds have usually been purified by polyacrylamide disc gel electrophoresis. Incubation Procedure. As previously described, two factors that contribute to assay sensitivity are the character of the antiserum and the specific activity of the labeled antigen. Additional sensitivity can be achieved by dilution of the assay system-the more the antiserum and labeled antigen are diluted, the lower the concentration of test antigen that can be measured. However, dilution of the assay system also results in slower reactions (longer incubation times) and increased problems such as adsorption of the antigen to the test tubes, interference with the reaction by serum proteins, and damage to the labeled antigen. The assay system should be designed to achieve an optimal balance of these factors. Steroid hormones and certain other compounds which can be extracted from serum specimens can be assayed quickly because the interfering serum proteins will not be extracted. Often the extraction or purification steps (if required) can introduce a background or blank value in the assay. However, by running appropriate controls, this background can be subtracted from the assay results. Protein hormones are usually assayed on diluted serum, and the incubation times are much longer. Most RIA incubations are conducted a t 4°C to take advantage of the greater free energy of reaction and prevent microbial growth. Another variation that gives increased sensitivity is nonequilibrium incubation in which the test antigen and antiserum are first incubated in the absence of labeled antigen. The test antigen thus has a competitive advantage over the labeled antigen (which is added later) in binding to the limited number of antibody binding sites. Separation of Antibody-Bound and Free-Labeled Antigen Fractions, The original RIA methods for peptide hormones employed chromatoelectrophoresis (CME) to separate the bound and free-labeled antigen pools. This method depends on the ability of the peptides (including the Free fraction) to stick to the paper a t the origin, while the Bound fraction migrates. However, CME is limited in usefulness because it is slow and laborious. It has largely been replaced by a number of faster and simpler

CIRCLE 99 ON READER SERVICE CARD

384 A

ANALYTICAL CHEMISTRY

\i

45, NO. 11, SEPTEMBER 1973

methods. The double antibody (second antibody) method is now widely used. If the assay antiserum was made in a rabbit, for example, the antibodybound fraction (gamma globulins) can be specifically precipitated by reaction with an antiserum made against rabbit gamma globulins in another species such as a goat. The precipitate can then be centrifuged and counted, yielding the antibodybound labeled antigen pool. The method is applicable to antisera from any species as long as a different species is used to make the second antiserum. The antibody-bound pool can also be nonspecifically precipitated by salts such as ammonium sulfate. sodium sulfate, or trichloroacetic acid and by solvents such as dioxane. ethyl alcohol, acetone, or polyethylene glycol. A solutiqn of ammonium acetate and ethyl alcohol in water ha also been used with success. The free-labeled antigen pool can be nonspecifically adsorbed by substances such as talc, microfine silica, florisil, kaolin, cellulose powder, and anion-exchange resins. One of the most widely used adsorption methods is activated charcoal (Norit) coated with dextran. The latter is a permeable sieve which permits the smaller antigen to penetrate and bind to the charcoal but excludes the larger antibody fraction. A technique which has had some success and will probably be used more in the future is the solid-phase antibody method, in which the antibody is coupled to an inert material but is free to react in the RIA incubation. As the incubation proceeds, the separation of bound and free-labeled antigen fractions occurs simultaneously. One approach employs antiserum-coated test tubes which are merely emptied and counted after the incubation. Other approaches use antibodies coated or bound to polypropylene disks, poly (tetrafluoroethylene-g-isothiocyanatostyrene) powder, agarose and dextran gels, bromacetyl cellulose, or bentonite particles. Although solid-phase methods generally use more antiserum, their economy and simplicity are bringing them into more frequent use. Validation. After the ingredients have been obtained and the method has been established, an RIA must be validated. Primarily, assay validation consists of a demonstration that the assay is measuring what it is supposed to be measuring. This can be accomplished in several ways. Multiple dilutions of the physiological specimens should give lines parallel to the standard line. Biologically active compounds should show correla-

T h e Technicon@Sequential Multi-Sample (TSM@)Analyzer is a fully automated liquid chromatography system for amino acid analyses. It will process protein hydrolysates in just 75 minutes, and physiologic samples in 5 Y2 hours. The TSM Analyzer is sensitive enough to detect and analyze quantities as small as 2-5 nM of each amino acid with unprecedented resolution of each peak. Samples may be stored in pairs of sealed TSM cartridges, safe from diffusion loss. As many as 40 pairs of samples may be held ready for analysis at one time, for great operating convenience.

The system’s modular flexibility assures simple maintenance and service. Air-stable ninhydrin reagent is mixed with hydrazine sulfate within the system, eliminating stringent reagent preparation and storage requirements, while ensuring maximum reproducibility of reactions. If we’re right about what you need for amino acid analyses, we think you’ll want to know more about our TSM Analyzer. Please write Dept. 175,

-

Technicon Industrial Systems, Thrrytown,New York 1059 1

CIRCLE 214 ON READER SERVICE CARD

ALL OUR BOTTLES ARE NORMAL (they don’t breakdown under pressure)

In fact they respond to press u r e s . . . just squeeze them and they will dispense any quantity of liquid from a single drop to a jet stream . . . without leaking. Because cap and nozzle are molded as a single unit, leaks can’t develop even under the strongest pressure. The needle-nose nozzle won’t split, crack or sag. These normal plastic squeeze bottles are available in 30, 60, 125, 250, 500 and 1,000 ml. sizes. Caps available in natural, red, green, blue and yellow for positive color coding of wash liquid. For extra economy-buy them by the barrel from your favorite Mallinckrodt AR” Distributor o r write

Plasticware SCIENCE PRODUCTS DIVISION St. Louis, Mo. 63160 CIRCLE 156 ON READER SERVICE CARD

886 A

tion between the concentrations determined by bioassay and by RIA. The results of assays of drugs or other compounds that can be measured by instrumental or chemical means should also correlate with the RIA results. For substances such as hormones, enzymes, and antigens which are produced in excess in pathoiogic states, the assay results in normal subjects and patients should correlate clinically. Finally, if the concentration of hormone or antigen in the specimen can be influenced by physiologic changes or agents, the assay should be able to detect these changes. Accuracy. To a certain extent, some of the suggested methods for validating the RIA will assess its accuracy as well. The accuracy of an RIA can only be tested if there is an alternate accepted method to measure the same compound. For most peptide hormones, the only other method of determination has been bioassay which often is useful as a measure of relative activity but not for accurate determinations. In addition, some protein hormone RIA’s have used standard hormones from animal species, although the assays were used for measurement of the hormones in man. In these or similar circumstances, lack of the human standard and lack of knowledge of the degree of cross-reactivity prohibit any assessment of accuracy. For many other RIA’s (drugs, steroids), there are valid standards and reliable alternate methods of determination, so that accuracy can be easily assessed. Precision. Precision is the degree to which a group of determinations on a sample agrees with the mean of the group. It is generally expressed as the percent coefficient of variation (CV)

cv=-100 u .Y

or the index of precision ( A )

where g is the standard deviation of the determinations; f is the mean of the determinations; and b is the slope of the response line. For RIA’s using the logit method previously discussed, u is the standard deviation of the logit 7’0 B, and b is the slope of the regression line. Therefore, this precision estimate incorporates the % B value a t which the variation is determined and the slope of the standard line in addition to the CV. Whether X or CVis used, generally most RIA’s have had precision better than or equal to the corresponding bioassays. Quality Control. As with any analytical method, RIA’s require

ANALYTICAL CHEMISTRY. VOL. 45, NO. 11, SEPTEMBER 1973

standardized quality control procedures to insure consistently accurate and precise results on unknown specimens. In addition to such checks as antibody control tubes and standard antigen tubes, each run should contain aliquots from normal and abnormal serum pools or other types of specimens similar to the unknowns. If these pool results are outside a range defined by the mean f two standard deviations, then the assay results may be incorrect and require checking or repeat determinations. In this way, the assay performance can be carefully monitored and controlled.

RIA Applications The original RIA methods were applied to protein hormones. The list of protein hormones now assayed by RIA includes insulin, glucagon, growth hormone, luteinizing hormone. follicle-stimulating hormone, prolactin, thyroid-stimulating hormone, adrenocorticotropin, N and /3 melanocyte stimulating hormones, placental lactogen (chorionic sommatomammotropin), chorionic gonadotropin, vasopressin, oxytocin, bradykinin, gastrin, secretin, pancreozymincholecystokinin, calcitonin, parathyroid hormone, and thyroglobulin. RIA has helped toward identification of other forms or prohormones for such hormones as insulin, parathyroid hormone, growth hormone, and the gonadotropins. RIA has been extended to steroid hormones such as aldosterone, cortisol, deoxycorticosterone, androstenedione, testosterone, dihydrotestosterone, progesterone, 17-hydroxyprogesterone, estrone, estradiol, estriol, and 2-hydroxy-estrone and to other nonpeptide compounds such as prostaglandins, thyroid hormones, and cyclic nucleotides; drugs such as digoxin, digitoxin, medroxyprogesterone, methylprednisolone, morphine, lysergic acid diethylamide. and barbiturates; enzymes such as carbonic anhydrase, fructose-1,6-diphosphatase, carboxypeptidases, chymotrypsin, trypsin, and elastase: large proteins such as albumins, globulins, carcinoembryonic antigen, alpha fetoprotein, hepatitis-associated antigen; other tumor antigens; structural proteins; and many other substances too numerous t o list here. It is sufficient to say that ahy compound which can be made radioactive and which is immunogenic, or can be made immunogenic by coupling to a carrier. can be measured by RIA, provided the substance can be obtained in pure form.

Related Methods In the competitive protein-binding

(CPB) assay, the reactive agents are

Choosing a pH meter can be tough-there's quite a selection. And most of them do the essential measuring job pretty well. So if you want to make a smart buy, you have to look beyond the pretty faces. You have to look for things like:

More Distributors, Faster Service. Over 150 distributor locations. More than 1000 salesmen. So you can usually get the Corning meter you want within 24 hours.

The Largest SelectionAnalog and Digital. 10 pH instruments. More digitals too, including the least expensive digital on the market.

Portable pH-Under $100 The Corning Model 3. Dependable, portable pH. Guaranteed for 2 years.

Guaranteed Electrodes.

Expand Your Horizons.

Laboratory reference and pH electrodes are guaranteed for 6 months. In writing. And if any Corning Act/Ion'" ion selective electrode fails to do the job you bought it for, and we can't solve the problem, we'll give your money back.

More meters. More recorders, more electrodes. Longer, stronger guarantees. Fast, easy service from local distributors. Now isn't that what you want in a pH meter? Or a recorder? Contact your Corning distributor. Or call us, toll free. Corning Scientific Instruments, Medfield, Mass. (800) 225-3850.

24 Laboratory Recorders.

The Free Year. Our unique 2-year guarantee on every meter we make. An extra year of guaranteed performance -free.

for one year.

A recorder for any instrument application in your lab. Flexible, reliable modular design. Designed to meet present needs-expand to meet future needs. Guaranteed

CORNING SCIENT1FIC INSTRUMENTS CIRCLE 44

ON

READER SERVICE CARD

I pjpn I

MODULAR SYSTEMS

I

Announcing 5 New Precision Plug-In Units

MP-1004A Log Ratio Amplifier. 6 decades range: 10 to 10 A. 100 mvidecade outp u t . Built-in reference gives log of single input. Accepts positive or negative inputs.

MP-1030A High Voltage Power S u p p l y . - 500 to --1500 V. Adjustable current limit protects PM tubes. 3 operating modes: fixed voltage, programmable voltage and constant PM current.

MP-1039 Dual, FET Operational Amplifiers. Input current typically 3 pA at either input. 10l2 o h m s resistance at either input. Open loop gain 2 x lo4 minimum. 6V/psec slew rate.

MP-1040 Multiplier. Seven switch selected functions: A2/10;A x B h O ; 10A/B, B>O; 10A/B, B < O ; BO. Four quadrant multiplier.

MP-1041 Instrumentation Amplifier. Provides gains from 1 to 1000. Range switch and 10-turn potentiometer. Input impedance, differential and common mode, loT2ohms typ. Guarded inputs.

McKEE-PEDERSEN INSTRUM E NTS Phone 415-937-3630 Box 322, Danville, CA 94526 USA

naturally occurring proteins, usually nonimmune globulins, which have specific affinity for the hormone or a class of hormones. Examples include many steroid hormones, thyroxine, vitamin Ds and its metabolites, and vitamin B12. The radioenzymatic assay employs an enzyme as the reactive agent, and the separation of two radioactive pools involves separation of two compounds, one of which was formed from the other by the enzymatically catalyzed reaction. An example is the folic acid assay by use of folic acid reductase. The radioreceptor assay employs a partially purified tissue receptor as the reactive agent. Examples include adrenocorticotropin (ACTH), cyclic AMP, and cyclic GMP. The immunoradiometric assay differs from the RIA and other related methods. It uses a radioactively labeled purified antibody. Unreacted and reacted labeled antibodies are separated by adsorption to antigen which is bound to a solid support. Examples include insulin, growth hormone, calcitonin, and parathyroid hormone. Generally, the RIA has numerous advantages over related methods. Compared to CPB, RIA has greater sensitivity and specificity, so that the specimens do not require chromatography or other purification steps. The radioreceptor assay is restricted to compounds for which tissue receptors can be identified and isolated, although they do have one advantage of measuring “biologic activity” as opposed to “immunologic activity.” However, if the RIA is properly validated, there is no essential difference. The radioenzymatic assay may be able to measure compounds for which antibodies cannot be readily made but has the disadvantage of requiring separation of compounds of similar size and structure for quantitation. The immunoradiometric assay does not have some of the “blank” problems of the RIA but is costly in its use of antiserum and antigen and is more difficult to set up initially.

Bibliography The first section includes a number of reviews, symposia, and workshops on radioimmunoassay and related methods. The additional sections include references on specific subjects, although most of the references in the General section also cover these topics. General 1. S. A. Berson, R. S. Yalow, S . M. Glick, and J . Roth, Immunoassay of Protein and Peptide Hormones, Metabolism, 13, 1135 (1964). 2. S. A. Berson and R. S. Yalow, Immunoassay of Protein Hormones, in “The Hormones,” G. Pincus, K. V. Thimann, and E. B. Astwood. Eds., Vol4, pp 557-630, Academic Press, New York, N.Y., 1964. 3. J. T. Potts, Jr., I,. M. Sherwood, J . L

CIRCLE 155 ON READER SERVICE CARD

888 A

* ANALYTICAL C H E M I S T R Y , VOL. 45, NO. 1 1 , S E P T E M B E R 1973

H. O’Riordan, and G . D. Aurbach, Radioimmunoassay of Polypeptide Hormones, Aduan. Int. Med., 13,183 (1967). 4. S. A. Berson and R. S. Yalow, General Principles of Radioimmunoassay, Clin. Chim. Acta, 22,51 (1968). 5. E. Diczfalusy, Ed., “Immunoassay of Gonadotrophins,” Acta Endocrinol., 63, Supplementum 142 (1969). 6. M. Margoulies, Ed., “Protein and Polypeptide Hormones,” Part 1, Excerpta Med. Int. Congr. Series No. 161, Amsterdam, the Netherlands, 1969. 7. B. E. P. Murphy, Protein Binding and the Assay of Non-Antigenic Hormones, Her. Progr. Horm. Res., 25,563 (1969). 8. E. Diczfalusy, Ed., “Steroid Assay by Protein Binding,”Acta Endocrinol., 64, Supplementum 147 (1970). 9. G. H. Grant and W. R. Butt, Immunochemical Methods in Clinical Chemistry, AdLwn. Clin. Chem., 13,383 (1970). 10. J. W. McArthur and T. Colton, Eds., “Statistics in Endocrinology,” MIT Press, Cambridge, Mass., 1970. 11. F. G. Peron and B. V. Caldwell, Eds., “Immunologic Methods in Steroid Determination,” Appleton-Century-Crofts, New York, N.Y., 1970. 12. W.D. Odell and W. H. Daughaday, Eds., “Principles of Competitive Protein-BindingAssays,” Lippincott, Philadelphia, Pa., 1971. 13. K. E. Kirkham and W. M. Hunter, Eds., “Radioimmunoassay Methods,” Churchill Livingstone, Edinburgh, Scotland, 1971. 4. A. R. Midgley, Jr., G. D. Niswender, V. L. Gay, and L. E. Reichert, Jr., Use of Antibodies for Characterizationof Gonadotropins and Steroids, Rec. Progr. Horm. Res., 27.235 (1971). 5 . D. S. Skelley, L. f’. Brown, and P. K . Besch, Radioimmunoassay. Clin. C ‘ h ~ m .19, . 146 (1973). Principle and Treatment of Data 16. G. Scatchard, The Attraction of Proteins for Small Molecules and Ions, A n n . N . E.: Acad. Sci., 51,660 (1949). 17. D. Rodbard, P. L. Rayford, J . A. Cooper, and G. T. Ross, Statistical Quality Control of Radioimmunoassays, J. Clin. Endocrinol. Metab., 28, 1412 (1968). 18. R. Ekins and R . Newman, Theoretical Aspects of Saturation Analysis, Acta Endocrinol., 64, Supplementum 147, 11 (1970). 19. R. S. Yalow and S.A. Berson, Radioimmunoassays, in “Statistics in Endocrinology,” J. Cy. McArthur and T. Colton, Eds., pp 327-44, MIT Press, Cambridge, Mass., 1970. 20. W. G. Duddleson, A. R. Midgley, Jr., and G. D. Niswender, Computer Program Sequence for Analysis and Summary of Radioimmunoassay Data, Comp u t . Biomed. Res.. 5 , 205 (1972). Antiserum Preparation and Evaluation 21. J . Freund, The Effect of Paraffin Oil and Mycobacteria on Antibody Formation and Sensitization,Amer. J. Clin. Path., 21,645 (1951). 22. G. L).Niswender and A. R. Midgley, Jr,, Hapten Radioimmunoassay for Steroid Hormones, in “Immunologic Methods in Steroid Determination,” F. G. Peron and B. V. Caldwell, Eds., pp 149-73, Appleton-Century-Crofts,New York, N.Y., 1970. 23. B. A. L. Hurn and J . Landon, Antisera for Radioimmunoassay, in “Radioimmunoassay Methods,” K. E. Kirkham and W.M. Hunter, Eds., pp 121-42, Churchill Livingstone, Edinburgh, Scotland, 1971. 24. W. D. Odell, G . A. Abraham, W. R. Skowsky, M. A. Hescox, and D. A. Fish-

1.0

0.4

f

-0.71

0.70 1

I

0

i

‘liine+hc;urs

2

4

3

‘

5

These traces were made with our Series 635 UV-visible Spectrophotometer - a scanning double-beam instrument with unmatched wavelength range of 190 to 900 nm. Photometric accuracy better than 0.002Abs. Stable. Reliable. Available with meter or digital readout. Supported by a complete range of accessories to suit your analytical needs. So much performance should cost you more.

D a K l t o/california/U.S.A./zug/SWITZERLAND/ georget own/ontario/CANADA/springvale/vic./ 4USTRALIA/wal ton-on-thames/surrey/ENGLAND/ xsay/FRANCE/darmstadt/W. GERMANY/torino/ ITALY/amsterdam/brussels/stockhnlm. ,~ CIRCLE 2 3 3 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL

45, NO

11, SEPTEMBER 1973

889 A

PI,

Produrrmnuf Antiwa for Radimm.

munoassap. m "Prmctplei uf (.'umpeti. t i v r I'rotrin-Hinding Assays." \V I). O d d and \ \ . 11. Daughaday. Eds.. pp T,^-R*l. Lippincutt. I'hiladelphia. Pa. 1471

25::.Vaitukaitis, J. B.Rohhins, E. Niesehlag, and G. T. Ross, A Method for Producing Smcific Antisera with

Protein Car&

VOLUMETRIC PERCENTAGE for aqueous and non-aqueous titrations. Special solutions to meet ASTM. APHA, USP and ACS requirements.

V Indicators, buffers, clinical

reagents V Custom solutions available r / Accuracy to 0.05% r / Prompt delively

V Packed in 5-gal. cubitainen

and smaller quantities Write for

HAS TI

CHEMICAL CORPORATION Mmn&rtur.rs of Electronic Ckm;eds

hw.d and

Tyburn Road &Cedar Lane, Fairless Hills, Pa. 19030 'Phone: (215)295-5006

for Use in Radioimmu-

Radiolabeled Antigen 27. R. S. Yalow and S. A. Benon, Labeling of Proteins-Problems and Practices, Trans. N.Y. Acnd. Sei., 28, 1033 i1966). 28. A. E. Freedlender, Practical and Theoretical Advantages for the Use of 1-5 in Radioimmunoassay, in "Protein and Polypeptide Hormones," M. Margoulies, Ed., pp 351-3, Excerpta Med. Int. Cangr. Series No. 161, Amsterdam, the Netherlands, 1968. 29. G. C. Oliver, Jr., B. M. Parker, D. L. Brasfield, and C. W. Parker, The Measurement of Digitoxin in Human Serum by Radioimmunoassay,J. Clin.Inuest., 47. 1035 (1968). 30. A. L. Steiner, D. M. Kipnis, R. Utiger, and C. Parker, Radioimmunoassay for the Measurement of Adenosine-3'5'Cyclic Phosphate, R o e . Nat. Aead. Sei., U.S.,64,367 (1969). 31. W. T. Newton, J. E. McGuigan, and B. M. Jaffe, Radioimmunoassay of Peptides Lacking Tyrosine, J. Lob. Clin. Med., 75,886 (1970). 32. W. M. Hunter. The Preparation and Assessment of Iodinated Antigens. in "Radioimmunoassay Methods," K. E. Kirkham and W. M. Hunter, Eds., pp 3-23, Churchill Livingstone, Edinburgh, Scotland, 1971. Incubation, Separation of Bound and Free-Labeled Antigen 33. C. N. Hales and P. J. Randle, Immunoassay of Insulin with Insulin Antibody Precipitate, Biochem. J.,88,137 (1963). 34. V. Herbert, K. S. Lau, C. W. Gottlieb, and S. J. Bleicher, Coated Charcoal Immunoassay of Insulin, J. Clin. Endocrinol. Metab., 25, 1375(1965). 35. A. M. Eisentraut, N. Whissen, and R. H. Unger, Incubation Damage in the Radioimmunoassay for Human Plasma Glucagon and Its Prevention with Trasylol,Arner.J. Med. Sci., 255.137 (1968). 36. K. J. Catt and G. W. Tregear, Solid Phase Radioimmunoassay, in "Protein and Polypeptide Hormones,'' M. Margoulies, Ed., pp 45-8, Excerpta Med. lnt. Congr. Series No. 161, Amsterdam, the Netherlands, 1969. 37. D. Rodbard, H. J. Ruder, d. Vaitukaitis, and H. S. Jacobs, Mathematical Analysis of Kinetics of Radioligand Assays: Improved Sensitivity Obtained by Delayed Addition of Labeled Ligand,J . Clin. Endoerinol. Metab., 33,343 (1971). 38. W. H. Daughaday and L. S. Jacobs, Methods of Separating Antibody-Bound from Free Antigen, in "Principles of Competitive Protein-BindingAssays," W. D. Odell and W. H. Daughaday, Eds., pp 303-24, Lippincott, Philadelphia, Pa., 1971. Methods Related to Radioimmunoassay 39. B. P. Murphy and C. J. Pattee, Deter-

CIRCLE 42 O N READER SERVICE CARD 890 A

* ANALYTICAL CHEMISTRY, VOL. 45, NO. 11. SEPTEMBER 1973

mination of Thyroxine Utilizing the Property of Protein-Binding,J. Clin. Endoerinol. Metab., 24, 187 (1964). 40. B. E. P. Murphy, Some Studies of the Protein-Bindingof Steroids and Their Application to the Routine Micra and Ultramicro Measurement of Various, Steroids in Body Fluids by Competitive Protein-BindingRadioassay, J. Clin. Endocrinol. Metab., 21,913 (1967). 41. S. P. Rothenhurg. A Radiwnzymatic Assay for Folic Acid, Noture, 206, 1154 ( 1965). 42. R. J. Lentawitz. J. Roth, and I. Pastan, Radioreceptor Assay of Adrenocorticotropic Hormone: New Approach to Assay of Polypeptide Hormones in Plasma, Science, 170,633 (1970). 43. L. E. M. Miles and C. N. Hales, The Use of Labeled Antibodies in the Assay of Polypeptide Hormones, J. Nuel. Bid. Med., 13,

Charles D. Hawker is chief clinical chemist and director of research and development for Laboratory Procedures@,a division o f T h e Upjohn Co., Kalamazoo, Mich. Laboratory Procedures is a nationwide clinical lahoratory network. Dr. Hawker earned a BA degree in chemistry from Illinois Wesleyan University (Bloomington) in 1962, an MS degree in biochemistry from the University of Wisconsin (Madison) in 1965, and a PhD in hiochemistry from the University of Pennsylvania (Philadelphia) in 1967. He was the recipient of an NIH postdoctoral fellowship from 1967-69, and from 1967-71 engaged in research on the radioimmunoassay of parathyroid hormone at the University of Pennsylvania. He has been with Lahoratory Procedures since 1971 and is directing the development of radioimmunoassays for many compounds including protein hormones, steroid hormones, and drugs. Dr. Hawker is a member of PhiLamhda Upsilon, the American Chemical Society, the American Association for the Advancement of Science, the American Federation for Clinical Research, and the Endocrine Society.