EPA-650/1-75-004

May 1975
Environmental Health Effects Research Series
         A FLUORESCENCE  IMMUNOASSAY
               TECHNIQUE  FOR  DETECTING
               ORGANIC ENVIRONMENTAL
                          CONTAMINANTS
                          U S Environmental Protection Agency

                          Office of Research and Development

                               Washington. DC 20460

-------
                                    EPA-650/1-75-004
A FLUORESCENCE IMMUNOASSAY


    TECHNIQUE  FOR DETECTING


     ORGANIC  ENVIRONMENTAL


            CONTAMINANTS


                      by

         Herbert R.  Lukens and Colin B. Williams

                 IRT Corporation
                 7650 Convoy Court
             San Diego, California 92138

                     and
  Stuart A. Levison, Walter B.  Dandliker, and Dennis Muryama

             Biophysical-Chemistry Unit
         Scripps  Clinic and  Research Foundation
      476 Prospect Street, La  Jolla, California  92037

              Contract No. 68-02-1266
                 ROAP No. 21AFM
             Program Element No. 1 El078

        EPA Project  Officer:  Dr. Ronald L. Baron

     Pesticides and Toxic Substances Effects Laboratory
         National Environmental Research Center
       Research Triangle Park, North Carolina 27711

                  Prepared for

         U. S. ENVIRONMENTAL PROTECTION AGENCY
          OFFICE OF RESEARCH AND DEVELOPMENT
              WASHINGTON, D.C. 20460

                    May  1975

-------
                         EPA REVIEW NOTICE

This report has been reviewed by the National Environmental
Research Center — Research Triangle Park, Office of Research
and Development, EPA, and approved for publication.  Approval
does not signify that the contents necessarily reflect the
views and policies of the Environmental Protection Agency, nor
does mention of trade names or commercial products constitute
endorsement or recommendation for use.
                     RESEARCH REPORTING SERIES

Research reports of  the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped  into
series.  These broad categories were established to facilitate
further development and application of environmental technology.
Elimination of traditional grouping was consciously planned to
foster technology transfer and maximum interface in related
fields.  These series are:

          1.  ENVIRONMENTAL HEALTH EFFECTS RESEARCH

          2.  ENVIRONMENTAL PROTECTION TECHNOLOGY
          3.  ECOLOGICAL RESEARCH
          4.  ENVIRONMENTAL MONITORING

          5.  SOCIOECONOMIC ENVIRONMENTAL STUDIES

          6.  SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
          7.  MISCELLANEOUS

This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS
RESEARCH series.  This series describes projects and studies
relating to the tolerances of man for unhealthful  substances or
conditions.  This work is generally assessed from a medical view-
point, including physiological or psychological studies.  In
addition to toxicology and other medical  specialities, study
areas include biomedical  instrumentation and health research
techniques utilizing anima Is — but always with intended applica-
tion to human health measures.
This document is available to the public for sale through the
National Technical Information Service, Springfield, Virginia 22161

                 Publication No.  EPA-650/1-75-004

-------
                                  ABSTRACT

This report describes the development and successful demonstration of an
immunological assay for the detection of low molecular weight organic
contaminants of environmental concern.

The specific technique described is a fluorescence polarization immuno-
assay, the theory of which is presented.

The preparation of the two required reagents, namely a fluorescent con-
jugate of the contaminant of interest, together with an antibody to the
contaminant, is described in detail.

The specific contaminant chosen for this study was 2-aminobenzimidazole
(MW = 133), a metabolite of certain fungicide agents used in agriculture.
The particular fluorescent moiety chosen to form the conjugate with
2-aminobenzimidazole was fluorescein.

A successful demonstration of the assay has been accomplished, and a detec-
tion sensitivity in the sub-nanogram/mJ, range obtained.

A high degree of specificity of the antibody for the hapten has been
demonstrated, and a successful quantitative recovery from an unknown
solution has been obtained.

This report is submitted in  fulfillment of Contract 68-02-1266 by IRT
Corporation, San Diego,  California,  under the sponsorship of the Environ-
mental Protection Agency.  The program was  initiated in November 1973 and
successfully completed in November 1974.
                                     111

-------
This page intentionally left blank.
               IV

-------
                                  CONTENTS
                                                                       Page
Abstract                                                                iii
List of Figures                                                         vii
List of Tables                                                           ix
Acknowledgments                                                          xi

Sections
I      Conclusions                                                       1
II     Recommendations                                                   3
III    Introduction                                                      5
IV     Methods                                                           13
V      Results                                                           19
VI     Discussion of Results                                             33
VII    References                                                        35
       Appendix l--The Physical Basis of Fluorescence Polarization
       Immunoassay                                                       37
       Appendix 2--Equilibrium Experiments                               45

-------
This page intentionally left blank.
               Vl

-------
                                   FIGURES
No.                                                                   Page
1      Rate of Change of Polarization as a Function of Inhibitor
       Concentration                                                   20
2      Rate of Change of Polarization as a Function of Inhibitor
       Concentration                                                   21
3      Standard Inhibition Curve                                       22
4      Comparative Inhibition:   Comparison Between
       2-Aminobenzimidazole and Benzimidazole                          26
5      Rate of Change of Polarization as a Function of Inhibitor
       Concentration                                                   28
6      Standard Inhibition Curve                                       29
7      Polarization as a Function of Time for Unknown Sample No.  1     30
8      Polarization as a Function of Time for Unknown Sample No.  2     31
                                    VII

-------
This page intentionally left blank.
               Vlll

-------
                                   TABLES
No.                                                                   Page
1      Relative Inhibition Produced by Benzimidazole and 2-ABZI        25
                                     IX

-------
This page intentionally left blank.

-------
                                ACKNOWLEDGMENTS

Much of the experimental work reported in this study was carried out in
the laboratories of the Biophysical Chemistry Unit, Scripps Clinic and
Research Foundation, La Jolla, California under the direction of Dr. W. B.
Dandliker, and is hereby gratefully acknowledged.

Many of the detailed laboratory analyses and processes were conducted by
Mr. A. Hitch of the Biophysical Chemistry Unit, and were carried out with
considerable care and skill.

The final manuscript was reviewed by Dr.  R.  Baron, Project Officer,  and
the authors are grateful for his many suggestions and constructive critique,
                                     XI

-------
This page intentionally left blank.
               XII

-------
                                  SECTION I


                                 CONCLUSIONS


The fluorescence polarization immunoassay (FPI) technique has been demon-

strated to be highly sensitive and specific for the detection of

2-aminobenzimidazole (2-ABZI) in aqueous solutions.  As a result of this

successful demonstration, the following major conclusions can be drawn.

        Immunological techniques can be applied to the detection of
        low molecular weight organic contaminants in aqueous solutions.

        A high degree of specificity of the contaminant antibody for
        the contaminant can be obtained, as evidenced by the difference
        in initial rate and equilibrium polarization values for 2-ABZI
        and benzimidazole.

        The detection sensitivity (see Section V) is in the sub-nanogram/
        mX. range, which compares favorably with other available analyti-
        cal techniques (Ref. 1).

        The assay can be conducted using initial rate data in one to
        two minutes by non-professional staff, which makes it particu-
        larly valuable for  the economic screening of large numbers of
        samples.

-------
This page intentionally left blank,

-------
                                 SECTION II

                               RECOMMENDATIONS

 In order to broaden the usefulness of this method and bring it to the
 stage of a practical analytical technique, a number of additional studies
 and developments are recommended.  The usefulness of the technique as
 presently applied to the detection of a single organic contaminant could
 be improved by:
        Development of a wide band fluorescence polarimeter capable
        of being operated in the field.
        Use of fluorescent conjugates with fluorescent wavelengths
        different from fluorescein (e.g., indocyanine green) to
        avoid natural competing fluorescences which occur in host
        samples of interest, e.g., blood serum and tissue.
        A study of the stability of the reagents, i.e., antibody
        and fluorescent conjugate, under normal operating environ-
        mental conditions, with recommendations for improving this
        stability where necessary.
 The demonstration of the ability of immunological techniques to detect low
 molecular weight organic contaminants lays the groundwork for the develop-
 ment of a rapid, sensitive, and inexpensive technique for environmental
 monitoring, e.g., reentry monitoring.  This involves the deposition of
 contaminant antibody on a solid substrate such as glass, plastic, or
paper, and monitoring of the physical characteristic of the surface to
 detect the presence of the contaminant.

There are basically two techniques by which this can be accomplished, and
 it is recommended that both be studied to determine the optimum and most
cost-effective system.   First, deposition of a metallic surface (indium
or nickel,  for example)  on a glass substrate,  bonding of the antibody to
the metallic surface,  and observation of the change in reflectivity which
occurs in the presence of the contaminant.   Second,  deposition of the
antibody on a plastic  or cellulose substrate,  exposure of the monitor to

-------
a fluorescent-labeled form of the contaminant, and then a study of the
inhibition of the bonding of this fluorescent conjugate by a sample
thought to contain the contaminant of interest.

Finally, the application of the technique to multi-residue analyses using
contaminants labeled with different fluorescent wavelength conjugates,
would reduce the time and effort involved in multi-residue, multi-sample
analyses.

-------
                                 SECTION III

                                 INTRODUCTION

BACKGROUND

Environmental pollution is currently creating extremely serious problems
on a global scale.  One facet of this pollution involves the ever-widening
agricultural use of chemicals such as pesticides, fertilizers, and similar
substances.  In addition, the industrial use of organic compounds is
increasing at an alarming rate.  An understanding of the gravity of the
situation can be obtained when it is noted that the annual production of
pesticides in the United States alone exceeds one billion pounds per year
(Ref. 2).  Repeated consumption of foodstuffs containing small quantities
of these substances can pose the threat of direct accumulative toxicity.
They can also seriously affect the ecological balance and produce insid-
ious disruptions of the food cycle.  This results from the tendency of
organisms in a food chain to sequentially concentrate many of the non-
biodegradable chemicals in tissue as they feed on organisms containing
only minute concentrations of them.  This phenomenon is called "biological
magnification" (Refs.  3,4).

Further, the spreading nature of this problem is indicated by the fact
that species which exist far from civilization also appear to be contam-
inated with such pesticide residues.  For example, eggs of the Bermuda
Petrel, which spends most of its life ranging the open sea at great dis-
tances from applications of DDT, have been found to contain up to 5.1 ppm
of DDE residues (Ref.  4).  Such accumulations of these nonbiodegradable
toxic chemicals could  also readily involve man as he consumes foods below
him in the eco-system  which are so tainted.

The impact of the problems associated with the use of synthetic organic
compounds is one which affects all members of society on a world-wide
basis irrespective of  geographic,  socio-economic, or ethnic background.

-------
Individuals come into direct contact with such compounds in industrial
processes involving, for example, plastics, waxes, optical brighteners,
and agricultural chemicals.  Indirect contact evolves through absorption
of these contaminants into the human food chain.  To determine the impact
of these contaminants on the environment, it is necessary to determine
their final distribution and concentration.  It is clear, then, that
sensitive and precise detection and measuring systems are required to
monitor the levels of these contaminants in the environment initially —
for example, in ocean waters, reservoirs, and streams, and finally in the
tissues of various living organisms.

The classical methods used in monitoring have been previously summarized
(Ref. 1), and it is beyond the scope of this paper to discuss them.  Of
all methods, gas chromatography-mass spectrometry has achieved the most
notable success in the detection of pesticides at the residue level.  In
an effort to improve upon the sensitivity and provide a considerably
simpler technique capable of field operation, the use of an immunochemical
technique has been investigated.  The use of such a technique has been
prompted by the extraordinary successes that have been recently achieved
in the measurement of biological substances by specific immunological
                                                               _g
reagents and techniques  (Ref. 5).  For example, as little as 10   mg/mJi
of several pituitary hormones can be directly detected by such approaches.
This level is equivalent to 0.001 ppb.  Available evidence indicates that
specific antibodies can be obtained against many organic compounds (Ref. 6).
In particular, antibodies have been obtained against DDT and malathion
(Refs. 7,8), and other pesticide residues (Ref. 1).   The availability of a
specific antibody against the structure or molecule to be detected immed-
iately opens up the possibility of using immunochemical methods as a highly
specific and sensitive (Ref.  5) detection technique.

Any means of applying an immunochemical reaction to a detection problem
ultimately relies upon a reaction occurring between a substance (antigen
or hapten) and its specific antibody.  Perhaps the most general means
by which this interaction can be employed in measurement and detection

-------
has come to be known as "competitive binding assay".  In principle, this
method requires two essential reagents.  These are a labeled form of the
substance to be detected or measured, and an antibody or receptor specifi-
cally directed against the substance.  The principle of the assay involves
a preliminary measurement of the binding of the labeled antigen (substance
being detected) with its antibody and then, a determination of the extent
of the inhibition of this binding by known quantities of the unlabeled
antigen, which corresponds to the unknown.  From these data, a standard
curve can be constructed which shows the degree of binding by the labeled
antigen under certain specified conditions as a function of concentration
of the unlabeled antigen or unknown added.

FUNDAMENTALS OF THE IMMUNOASSAY TECHNIQUE

Immunology as a subject is logically concerned with the immunity of living
organisms to harmful agents,  regardless of their origin, and includes
resistance to disease, hypersensitive reactions as exhibited by allergic
persons, and tolerance and rejection of foreign tissue such as those
encountered in organ transplants.  In this case, however,  a more restricted
view of the definition has been taken and has been applied to the mechanisms
and techniques involved in the detection and measurement of organic con-
taminants of environmental interest.

A substance, which when injected into an animal stimulates the animal to
produce antisera capable of reacting with it in a highly specific manner,
is referred to as an antigen,  and the specific protein produced is referred
to as an antibody.   These antibodies belong to a group of serum proteins
known as immunoglobulins.   The production of these antibodies as a result
of the injection of the antigen takes place over a period  of many weeks,
and depends upon the immunization schedule.   In general, "good" antigens
are usually of large molecular size (>40,000)  partially digestible by
enzymes and are recognized as  being foreign by the antibody-producing
animal.   It is immediately obvious,  of course,  that many compounds of
environmental  concern do not  have a large molecular weight and would,
therefore,  appear to be incapable of stimulating antibody  formation.

-------
Fortunately this is not the case, and so-called partial antigens or haptens
can be produced and are capable of reacting with specific antibody.  Haptens
or partial antigens are defined as antigens which alone cannot induce anti-
body  formation, but in conjugation with a suitable carrier can produce
antibody against themselves, as well as against the carrier-hapten complex.
Examples of such carriers  include ovalbumin, bovine serum albumin, fibrinogen,
and many others.  In summary, the hapten once conjugated with a suitable
carrier can stimulate antibody production.  The remarkable thing about this
antibody stimulation is that some antibody will be produced which is highly
specific in its reaction with the hapten alone.  It is this phenomenon
which allows the use of immunological techniques in the detection and
quantitation of organic contaminants of relatively low molecular weight,
and under a variety of practical circumstances.

REVIEW OF IMMUNOASSAY TECHNIQUES

The usual method of labeling the antigen to be identified in an immunoassay
requires the introduction  of a radioactive label.  When such a radiolabel
is used, an essential and  crucial step in the radio-immunoassay (RIA) is
to separate physically that portion of the labeled antigen which is bound
to the antibody from that  which is unbound or free.  Only in this way is
it possible, by radioactive counting, to determine what fraction of the
radiolabel remains bound,  or is being bound, in the presence of the unknown.

Alternatively, a direct way of implementing competitive binding principles
in an immunoassay is to employ a fluorescent label which allows the assay
to be carried out in principle, either by fluorescence polarization measure-
ments, or in some cases by fluorescence intensity measurements.  Unlike RIA,
no separation of the bound and free forms of the labeled antigen is necessary,
since a simple, rapid optical measurement gives the essential information
without physical separation.

Examples of other radioassays include radioreceptor assay, which employs a
partially purified tissue  receptor as the active reagent,  and immunoradio-
metric assay,  which uses radioactive labeled antibody.  In the latter case,

-------
 separation  of  bound  and unbound  antibodies  is  accomplished  by attachment
 to  the  antigen which is deposited  on  a  solid substrate.   Radioenzymatic
 assays  involve enzymes as the  reactive  agent.   Once  again,  a  separation
 procedure is required.  In this  case, separation  of  the  radioactive  pools
 requires separation  of two compounds, one formed  from  the other  by the
 enzyme  catalyzed  reaction.  These  techniques,  however, tend to be inferior
 to  RIA  and  FPI in terms of sensitivity  and  specificity (Ref.  1).

 APPLICATION OF IMMUNOASSAY TECHNIQUES IN PESTICIDE RESIDUE  METHODOLOGY

 The application of immunological techniques to the analysis of pesticide
 residue levels appears to be of  relatively  recent origin, and Ercegovitch
 (Ref. 1) in 1971  gave an excellent review of work in this area.  This work
 appears to  be  limited to research  carried out  on  the herbicide aminotriazole
 and parathion  with sensitivities in the microgram range,  and  by  Centeno
 (Ref. 7) and Haas  and Guardia  (Ref. 8) using a tanned  cell  hemiglutination
 inhibition  test with detection limits of 0.1 \ig and 1.0  yg  for DDT and
 malathion,  respectively.

 APPLICATION OF FLUORESCENCE POLARIZATION TO IMMUNOASSAY TECHNIQUES (FPI)

 The essential  feature of applying this phenomenon to an  immunoassay  con-
 sists in first  labeling the antigen molecule with a fluorescent moiety
 and then observing the degree of polarization  of the fluorescent light,
 or  in certain  cases, the intensity of fluorescent light measured, when
 standard quantities  of the labeled antigen and antibody together with the
 unknown are allowed  to interact.   The dependence of polarization (and also
 occasionally,   the  fluorescent intensity) upon  the extent of reaction between
 the  antigen and antibody forms the basis for the quantitation  and immuno-
 assay.  Reaction between the antigen and antibody results in an increase
 in  size of  the  kinetic unit and in a retardation of the rotary brownian
motion,  which  in turn is manifested by an increase in the polarization of
 fluorescence.    In the presence of unlabeled antigen in the sample,  a smaller
percentage of the labeled antigen is bound to the antibody and in this

-------
circumstance the polarization observed will be lower.  Hence, the standard
immunoassay curve, which can be constructed from this type of data,  would
show the polarization of fluorescence for certain standard chosen experi-
mental conditions plotted as a function of the amount of unlabeled antigen.
                                     10

-------
OBJECTIVE

The objective of this program was to demonstrate the feasibility of using
the technique of fluorescence polarization immunoassay for the detection
of low molecular weight contaminants.  The structure of 2-aminobenzimida-
zole (2-ABZI), chosen as a model compound for this demonstration, is shown
below.
                                          Molecular weight  ^133
While this compound alone does not find extensive use in either agricul-
ture or industry,  it may be found as the result of the use of a number
of agricultural chemicals.   For example, the degradation of certain fungi-
cides results in 2-ABZI.   Consequently, the demonstration of a successful
fluorescence polarization immunoassay for this  compound would serve a
twofold purpose, namely the capability of immunological techniques to
assay for low molecular weight compounds,  and the sensitivity and speci-
ficity of the assay technique to a compound of  environmental interest.   The
scope of work for  the program was as follows.

     TASK 1:  From a list of available pesticides or metabolic
              derivatives selected by the  Contracting Officer,
              choose by analysis or experiment,  a single product
              that can be labeled with a fluorescent dye and can
              be used to  demonstrate the feasibility of a fluor-
              escence polarization immunoassay.
                                     11

-------
TASK 2:  Prepare a fluorescent sample of one selected pesticide
         or metabolic derivative using chemical and/or free
         radical induced labeling techniques.

TASK 3:  Prepare immunogens from the selected  pesticide or
         metabolic derivative and immunize animals with the
         immunogen to obtain antibody.

TASK 4:  The binding affinity of the fluorescent labeled com-
         pound developed under Task 2 for the  antibodies
         developed under Task 3 will be determined, and a
         standard immunoassay curve for the pesticide or deriva-
         tive will be constructed.

TASK 5:  Recovery experiments will be performed on known
         quantities of the selected compound,  added to natural
         water and one other medium specified  by the Contracting
         Officer.
                                12

-------
                                 SECTION IV

                                   METHODS

PREPARATION OF REAGENTS REQUIRED FOR THE ASSAY OF ENVIRONMENTAL
CONTAMINANTS

Of the techniques available for the synthesis of the fluorescent deriva-
tives, two have been considered.  First, classical organic syntheses, and
second, free radical labeling.  The organic synthesis adopted clearly depends
on the structure of the contaminant itself, and will vary widely.  Alter-
nately, a mixture of the substance to be labeled, together with the fluores-
cent dye, can be irradiated.  The multiplicity of free radicals formed
during the irradiation then affords a mixture of compounds, some of which
will generally be fluorescent-labeled derivatives (Ref. 9).

Preparation of an antibody against a contaminant begins by coupling it
to a highly immunogenic molecule such as ovalbumin, and introducing the
complex into an animal, for example rabbits, by means of intradermal injec-
tions.  The initial immunization yields "primary response" antibodies,
which are usually of fairly low specificity.  Booster immunizations can
be given at eight-week intervals and secondary response antibody collected
ten days after the booster.  This is generally more specific and of higher
titer.

PREPARATION OF FLUORESCENT DERIVATIVE OF 2-ABZI

Several organic syntheses were carried out in an effort to prepare a
fluorescent derivative of 2-ABZI.   Although several of the syntheses
proved to be unsuccessful,  they will be described for the purposes of
completeness.

CLASSICAL ORGANIC SYNTHESES

The initial synthesis effort was to prepare the conjugate Fluorescein-NH-H ,
where H  is a hapten derived from 2-ABZI via reaction at the 2-position of
2-ABZI.   Fluorescein amine  was used in the  preparation of this conjugate.
                                      13

-------
H  is prepared by reacting 2-ABZI with succinic anhydride.   The cyclized
product opens up on reacting with amines.   H  is then of the form:

             H    H    0            	

                                                   ,  and when conjugated
        C_C— C— C— N —   C
forms Fluorescein-NH-H..

It was determined that  this 2-position derivative of 2-ABZI was the suc-
cinate salt compound  (2-ABZIH lo'C^H.O., and consequently, alternate
syntheses were initiated.

Based upon the consideration that conjugation at the number two position
of 2-ABZI may result  in a weaker immunogen than might be the case for
conjugation at alternative positions, separate attempts were made to
conjugate at the one-position and five-position of 2-ABZI.  In each case
the effort was to prepare a butyric acid derivative, which could then be
coupled to either ovalbumin or fluorescein as desired.

Treatment of 2-ABZI with 4-chlorobutyric acid in a direct attempt to
prepare a one-position derivative resulted in a viscous, oily product
that was isolated by  chromatography, but would not crystallize.  An
indirect attempt to prepare the compound by first preparing the ethylester
of the butyric acid side chain was made.   This intermediate, ethyl-y-[l-2
aminobenzimidazol] butyrate, was crystalized (m.p. 95 to 97°).   The proton
magnetic resonance spectrum of the compound is consistent with the desired
1-position derivative of 2-ABZI.

Preparation of a 5-position derivative was started with y-phenyl-but.yric
acid,  which was nitrated at the para-position.   The nitro group was reduced
to the amino-group with zinc,  and then acetylated with  acetic anhydride.
                                     14

-------
The meta-position was then nitrated, and  the p-amino m-nitro phenyl  butyric
acid intermediate product was purified prior to  reducing the nitro group
and cyclization across the resulting adjacent amino groups  to  obtain the
desired product 2-ABZI with a butyric acid  size  chain at the 5-position.

In addition, a further attempt to product a conjugate at the 2-position
was made as follows.  2-ABZI and fluorescein isothiocyanate  (FNCS) were
reacted to produce a conjugate with fluorescein  coupled to  the 2-position
of 2-ABZI, by the following reaction.
         — N
           \
            /
C - NH2 + FNCS
                   THF
                                                     H
                                                  — N
                                     _ N
\    H   I
  C-N-C-N-F
The 2-FNCS-ABZI mixture was subjected to separation by a two-step, thin-
layer chromatography procedure.  Firstly, on silica gel with  15% methanol
in ethyl acetate until the faster moving unreacted dye was separated from
the labeled product.  The latter was removed from the silica  gel with
methanol and separated on cellulose TLC using borate buffer (pH 8.8).  The
fluorescent zone adjacent to the leading point contained the  desired product,
which was removed with methanol.  Since reactivity was successfully attained
with this conjugate at the 2-position, no further work was undertaken on the
1- and 5-position syntheses.

PREPARATION OF 2-ABZI PROTEIN CONJUGATE

Although numerous protein carriers can be used in the preparation of a
successful immunogen, in this case the studies have been limited to oval-
bumin, since this carrier has been used successfully in much  of our earlier
work.   The preparation of immunogen was carried out by first  reacting tetra-
hydrofuran solutions of 2-ABZI and thiophosgene to produce the isothiocyanate
of 2-ABZI, which precipitated:
           H                  CSL
         "

               C —NH2 + C  =
                                      THF
                                                  'C - N = C = S
                                                          — N
                                      15

-------
The precipitate was filtered, washed, and air dried.  An 18-mg aliquot
was dissolved in 1 mH of dimethylformamide and slowly added to an 8 m£,
0.15 M saline solution of 500 mg of ovalbumin.  The solution was stirred
overnight, dial/zed, and freeze dried to give a yellow-tan product.
Spectral analysis  (U.V.) indicated the product was a conjugate of 2-ABZI
and ovalbumin in the mole ratio of ^2:1.  This indicated that the reac-
tion of 2-FNCS-ABZI with amino groups of the ovalbumin led to a successful
conjugation:
          H
        — N
        — N
           \
           /
C - N = C - S + H2N - Ov
'\     H    1    H
  C-N-C-N-Ov
                                     —N
This will be designated 2-Ov-ABZI.

The ovalbumin conjugate was mixed 1:1 with Freund's complete adjuvant and
2-mg portions were injected into two sites of each of three rabbits (female
New Zealand Whites).  After eight weeks, the rabbits were boosted and then
bled 10 days later (15 mH per rabbit); the blood was allowed to clot, and
the serum separated by centrifugation.  The globulins were separated from
the serum by precipitation with (NH.)-SO  (at 0.38 saturated ammonium
sulfate), followed by centrifugation.  The globulin fraction was solu-
bilized to its original serum concentration in 0.15 M saline solution for
testing by FPI.

RADIATION SYNTHESIS

Mixtures of 2-ABZI and fluorescein solutions, and mixtures of 2-ABZI and
ovalbumin solutions have been subjected to 17 megarads of bremsstrahlung
(E    = 10 MeV) from an electron linear accelerator, to examine the use
  max
of ionizing radiation chemistry in synthesizing both the immunogen with
2-ABZI hapten and the fluorescein conjugate with 2-ABZI.
                                     16

-------
A dose of 17 megarads is equivalent to the deposition of 1.1 x 10
of solution.  Assuming that about one molecule of desired product is formed
            3                                                            18
for every 10  eV deposited (G-product ~0.1), this would produce 1.1 x 10
molecules, or ^1.8 micromoles of product per m£ of solution.  This would
amount to about 1 mg of 2-ABZI-fluorescein, and about 91 mg of 2-ABZI-
ovalbumin per mi.

Since the radiation chemistry process gives mistures of all possible products,
the method seldom permits direct synthesis of a pure product.  The 2-ABZI
and fluorescein mixtures contained much more than 91 mg of product (visible
as highly colored precipitate, for example), and some separation of the
desired product was necessary.

These irradiated solutions proved to be too heavily irradiated (17 megarads
resulted in the denaturation of the ovalbumin).   Accordingly, new solutions
of 2-ABZI and ovalbumin, and 2-ABZI and fluorescein, were prepared and irradi-
ated.  These solutions were given 3.3 megarads of electron (10-MeV)  irradi-
ation.  However, since success was obtained with the chemical syntheses,
the workup of these irradiated conjugates was discontinued in order to
concentrate on the development of the immunoassay.   The practicality of
the radiation synthesis approach,  however, has been demonstrated on a
parallel program (Ref. 10).
                                     17

-------
This page intentionally left blank.
                18

-------
                                  SECTION V

                                   RESULTS

DETECTION SENSITIVITY

With the successful production of antibody to 2-ABZI, and also a fluores-
cent conjugate, a series of preliminary experiments were conducted to
estimate the detection sensitivity.  In these experiments, varying con-
centrations of 2-ABZI were incubated with the globulin aliquot (which
contains the antibody), labeled 2-ABZI was added, and the polarization
recorded as a function of time.  The results of these experiments are
shown in Figures 1 and 2.  The initial rate of polarization was diminished
in proportion to the logarithm of the 2-ABZI concentration, i.e., effec-
tively the kinetics are first order, which is demonstrated by the following
data points, which are shown in graphical form in Figure 3.
                   Log
      (picograms of 2-ABZI per m£)

                  3.64
                  4.12
                  4.64
                  5.12
Initial Polarization Rate
         (dp/dt)Q
(units of polarization/min)
           67. Ob
           32.4
           19.0
           12.6
            3.9
      a                 4
       Initial rate x 10 .
       Basic reaction of 1 yg 2-FNCS-ABZI with 200 yg of globulins
       in 3 ma of solution.
Least squares fitting of these data points gives the regression line:
          Y = 98.79 - 18.68X.
The observed data points show a standard deviation of 16%, relative, about
the regression line.   The value of (dp/dt)  in the absence of 2-ABZI
                                      19

-------
t

•z.
o
t—i
t—
•a:

»—t
o:
   o
   Q-
                                    As
  0.10
 0.090
  0.08
  0.07
RUN
0
S
R
INHIBITOR
CONCENTRATION
(ng/mfc)
0
4.43
13.3
 0.065
RT-08876



      Figure 1.
                                          I
                     10               20


                           TIME (MINUTES)
                                                          30
              Rate of Change of  Polarization as a Function of


              Inhibitor Concentration
                                  20

-------
   o

   p
   •=C
   O
   D.
  0.09
  0.08
  0.07
                    D
        A  A
                            A A
RUN
T
Q
P
INHIBITOR
CONCENTRATION
(ng/mi)
44.3
133
1330
RT-08875
10               20


       TIME  (MINUTES)
                                                         30
     Figure 2.  Rate of Change of Polarization as a Function of

                Inhibitor Concentration
                                  21

-------
      70
        I          I         I          I          I

O EXPERIMENTAL POINTS (a = 16% RELATIVE TO LEAST-
                        SQUARE LINE)
0 LEAST-SQUARE FIT            Y = 98.79 - 18.68 X


BASELINE
          ZERO  INHIBITOR
     60
 O


 X
     50
 DC
 O
 O-
     40
     30
     20
     10
                  I
                  I
I
                  12345


RT-08877            LOG CONCENTRATION OF 2-ABZI (pg/ml)


                Figure  5.  Standard  Inhibition  Curve
                                 22

-------
        -4
 (67 x 10   in the foregoing tabulation) is intercepted by the regression
 line at X = 1.7; i.e., when the concentration of 2-ABZI is 50 picograms
 per mi.  This level cannot be detected under the described conditions,
 since this concentration cannot be differentiated from zero inhibitor
 concentration, i.e., this concentration corresponds to the "noise" of
 the system.  However, changes in the initial rate of polarization of
      -4
 3 x 10   units are readily observable.  Thus, a rate of change of
       -4
 64 x 10   units can be differentiated from the noise level.  From Figure 3
 this can be seen to give a log concentration of 2-ABZI (pg/mX,) of approxi-
 mately 2, which corresponds to a minimum detection sensitivity for 2-ABZI
 of 100 pg/m£.

 SPECIFICITY

 In devising an immunoassay, the degree of specificity attained is a prime
 factor in determining the ultimate usefulness of the test.  No one parameter
 has ever been universally adopted as a measure of immunological specificity,
 but one of the best would seem to be the binding affinity between antibody
 and the antigen or hapten.  The binding affinity is a reasonable measure
 of specificity, since cross-reacting antigens or haptens usually show lower
 affinities than that of the structure contained in the immunizing antigen.
 The physical factors contributing to the binding affinity finally reduce
 to the number and strengths of all the atomic interactions between the
 antigenic determinant group and the antibody combining site.   The nature
 of these interactions includes hydrogen bonding, electrostatic attraction,
 hydrophobic bonding, and a variety of weaker dispersion forces.   The better
 the match between groups on antigen and antibody,  the greater will be the
 free energy of interaction when the two molecules  combine, and the greater
 the "specificity".

To estimate the specificity of this assay,  inhibition experiments involving
 2-ABZI  and benzimidazole were conducted.   The structure of these two'com-
pounds  and their respective molecular weights are  as  follows.
                                      23

-------
       2-Aminobenzimidazole  (MW =  133)
Benzimidazole (MW = 118)
                                                           -H
                        I
                        H
 In  these  experiments  antibody  (200 yL) was added to 3 mfc of buffer, and
 then  3  yL of benzimidazole  or  2-ABZI was added to achieve a 10~  M solu-
 tion  of the respective benzimidazole compound.  The fluorescence polari-
 zation  was read  at  timed  intervals after 50 yL of the 2-ABZI fluorescent
 conjugate was  introduced.

 The resulting  data  are shown in Table 1, and in graphical form in Figure 4.
 The results for  the case  in which no inhibitor has been added is also
 given.

 The significant  difference  between both the initial rate of change of
 polarization and the  equilibrium value of the polarization for the case
 of inhibition  by 2-ABZI and benzimidazole demonstrates the high degree of
 specificity of the  antibody for its hapten.

 RECOVERY  EXPERIMENT

 The objective  of this recovery experiment was to demonstrate the ability
 of the  fluorescence polarization technique to determine on an absolute
 basis the presence  of 2-ABZI in an unknown sample.

 For these measurements, new reagents were prepared.  Firstly, fresh tris
 buffer  at pH 8.0 and  0.01M  concentration was prepared and filtered through
 0.22y millipore  filter to remove any microorganisms.  Next, fluorescein-
 tagged, 2-aminobenzimidazole (FABZI) was prepared by reacting fluorescein
 isothiocyanate with 2-ABZI  in tetrahydrofuran, and purified by sequential
TLC procedures,  as  described in Section IV.
                                      24

-------
Table 1.  RELATIVE INHIBITION PRODUCED BY BENZIMIDAZOLE AND 2-ABZI
2-FNCS-ABZI
No Inhibitor
Time
(min)
(0)
0.28
0.38
0.75
1.00
1.58
2.5
4.0
5.0
p at
Pta)
(0.076)
0.0815
0.086
0.0895
0.095
0.100
0.105
0.118
0.125
time zero,
P-Po
0
0.0055
0.010
0.0135
0.019
0.024
0.029
0.042
0.049
p , obtained
vs
Time
(min)
(0)
0.30
0.55
1.0
2.0
3.0
5.0
7.0
2-FNCS-ABZI
2-ABZI Inhibitor
Pta)
(0.076)
0.076
0.076
0.074
0.074
0.074
0.0775
0.0775
10.0 0.076
by extrapolation of
P-P
F *o
0
0
0
-0.002
-0.002
-0.002
0.0015
0.0015
0
observed data.
2-FNCS-ABZI
vs Benzimidazole Inhibitor
Time
(min)
(0)
0.30
0.48
0.70
1.0
1.65
2.70
4.0
5.0
Pfa)
(0.084)
0.086
0.088
0.0895
0.0905
0.095
0.101
0.107
0.110
P-P
^ *o
0
0.002
0.004
0.0055
0.0065
0.011
0.017
0.023
0.026

-------
 o
ex
    0.06
    0.05  -
   0.04  -
   0.03  -
   0.02  -
   0.01
                                                    BENZIMIDAZOLE

                                                    INHIBITION
                                                   2-ABZI INHIBITION
  RT-10720
            Figure 4.  Comparative  Inhibition:  Comparison Between


                      2-Aminobenzimidazole and Benzimidazole
                                      26

-------
Previously unused immunoglobulin preparation, which had been  frozen  immed-
iately after preparation, was then thawed.  This preparation,  obtained  from
rabbits innoculated and boosted, was at normal serum concentration in the
thawed preparation.

A set of inhibition curves was then developed for varying concentrations of
inhibitor.  All polarization measurements were made after mixing the reac-
tants in 3.0 m£ of tris buffer.  The resulting curves are shown in Figure 5.
The initial rate of change of polarization for the four different levels of
inhibitor concentration shown, were as follows.
                                                                    o
                                           Initial Polarization Rate
                    Log
        (picograms of 2-ABZI per m&) _ (units of polarization/min)
                   3.60                               227
                   3.12                               433
                   2.60                               617
                   2.12                               842
       a                 4
        Initial rate x 10 .
These data points are shown in graph form in Figure 6.  Analyzing these
four points by least squares gives the following linear relationship.

     Log Inhibitor Concentration (pg/mJl) = -0.00247 dp/dt  + 4.171.

Two solutions were then prepared containing known but different quantities
of 2-ABZI.

The change in polarization as a function of time for these two unknowns,
containing the antibody, tris buffer, and 1 p£ of fluorescent-labeled
2-aminobenzimidazole was then recorded.   The results are shown in graphical
form in Figures 7 and 8.
                                                                     • -4
The estimated initial rate of change from these two curves is 730 x 10
            -4
and 670 x 10  , which from Figure 6 corresponds to 272 pg/m£ and 347 pg/m£,
respectively.   The quantities added to these unknowns corresponded to
270 pg/ml and 330 pg/mfc.
                                      27

-------
  t
 7Z
 O
 O
 Q-
0.30
0.28
0.26
0.24
0.22
0.20
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02h
                              I
RUN
A
B
C
D
E
INHIBITOR
CONCENTRATION
(ng/ma)
4.0
1.3
0.4
0.13
0
          I
RT-11525
  3       4
TIME (minutes)
     Figure 5.  Rate of Change of Polarization as a Function of
                Inhibitor Concentration
                                 28

-------
a:
o
Q-
1000
 900
 800
 700
 600
 500
 400
 300
 200
 100
          _L
                      I
I
I
          2.1           2.5              3.0              3.5
   RT-11526          LOG CONCENTRATION  OF  2-ABZI  (pg/m£)
                  Figure 6.  Standard Inhibition Curve
                                  29

-------
     t
     o
     I—H
     t—
     •=c
 o
 Q-
0.28
0.26
0.24
0.22
0.20
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
RT-11527
                             (dp/dt)  x 10  = 730
                             CONTAMINANT CONCENTRATION = 270  pg/mJL-
                              3       4
                           TIME (minutes)
            Figure 7.  Polarization as a Function of Time for
                       Unknown Sample No. 1
                                    30

-------
   t

  z
  o
  »-—t

  •a:
  1^1
  •—«
  o;
  <

  o
  Q-

 0.30


 0.28


 0.26


 0.24


 0.22


 0.20


 0.18


 0.16


 0.14


 0.12


 0.10


 0.08


 0.06


 0.04


 0.02
(dp/dt)Q x 104 - 670


CONTAMINANT CONCENTRATION =  330  pg/m£
                                        I
     0
RT-11528
       3        4


    TIME (minutes)
          Figure  8.  Polarization  as  a Function of Time for


                     Unknown  Sample No.  2
                                 31

-------
This demonstrates the ability of the technique to determine subnanogram
quantities of 2-ABZI in an unknown sample to approximately 5% of the true
value for the worst case analyzed in this experiment.
                                     32

-------
                                  SECTION VI

                            DISCUSSION OF RESULTS

 It  is  of  interest to compare  the  FPI  system  for  2-ABZI  to  that  of  a  typical
 case,  as  represented by dinitrophenol  (DNP)  (Ref.  11) and  its antibody,
 and to the  optimum  system  evaluated to date,  i.e.,  fluorescein  and its
 antibody  (Refs.  12,13).  The  relevant parameters  for  these systems are
 tabulated below.
Parameter
Concentration of unbound
binding sites3 - R
Heterogenity Factor - a
Average association
constant - K
o
Second order rate
constant - k

Detection limit
DNP
(Ref. 11)
10'5 Mb
0.6
4.5 x 107


5 x 107/M-sec
.7
10 M
2-ABZI
2.4 x 10"8 Mb
0.72
1.4 x 1010


2.5 x 106/M-sec
_9
10 M
Fluorescein
(Refs. 12,13)
7
1.0
1 x 10H


4 x 108/M-sec
-11
10 M
aAntibody from boosted animals.  In the case of fluorescein, very  late
 antibody was used.
 At normal serum concentration.
Q
 An inverse proportion is implied between a_ and the variety of binding
 sites.

Comparison of the three systems tabulated above shows that the FPI system
for 2-ABZI performs better than average, but not as well as for the best
test system evaluated to date, namely, fluorescein.

Despite the fact that fluorescein is a considerably larger and more com-
plex molecule than 2-ABZI, there is reason to believe that the FPI system
for the latter could be brought to perform nearly as well as the fluorescein
system.  In particular, it seems likely that affixing carrier protein to
                                      33

-------
one of the positions of the 6-membered ring of 2-ABZI would provide for
greater participation of the unique triple amine constellation in antibody
formation relative to the present case.

The present 2-ABZI-Ov conjugate presents the hapten with the 6-membered
ring foremost in a manner that apparently minimizes the uniqueness of the
molecule as it appears to the antibody-forming system.  Both the small
production of antibody, which suggests that less than the usual number of
lymphocytes recognized the antigen as a foreign entity, and the compara-
tively small rate constant, k, which suggests a relatively large steric
factor in the antibody-hapten reaction, are consistent with this view.
It has been shown that the FPI method can be applied to the measurement
of 2-ABZI with great sensitivity and specificity.  The reaction between
the hapten and its antibody, as prepared in the present program, has been
defined in terms of order, equilibrium constant, antibody heterogeneity,
and rate constant.
                                     34

-------
                                 SECTION VII
                                  REFERENCES
 1.  Biros, F. J. (Ed.), Pesticides Identification at the Residue Level.
     Advances in Chemistry, Series 104, Am. Chem. Soc., Washington, D.C.
     (1971).

 2.  Statistical Abstract of the United States 1970, U.S. Department of
     Commerce.

 3.  Carson, R., Silent Spring, p. 22  (Riverside Press).

 4.  Environmental Pollution by Pesticides (Plenum Press, 1973).  See
     also Wurster, C.  F., and Wingate, D.  B.,  Science N.Y. 159, p. 979
     (1968).

 5.  Dandliker,  W. B., Kelly, R. J. Dandliker, J., Farquhar, Jr., and
     Levin, J.,  "Fluorescence Polarization Immunoassay.  Theory and
     Experimental Method," Immunochemistry 10, pp. 219-227 (1973).

 6.  Hawker, C.  D., "Radioimmunoassay and Related Methods," Analytical
     Chemistry 45, No. 11, pp. 878A-888A (1973).

 7.  Centeno, E. R.,  Johnson, W. J.,  and Sehon, A. H., "Antibodies to
     Two Common  Pesticides, DDT and Malathion," Int. Arch. Allergy 37,
     pp. 1-13 (1970).

 8.  Haas, G. J., and Guardia, E. J., "Production of Antibodies Against
     Insecticide-Protein Conjugates," Society for Exp. Biology and
     Medicine, Proc.  129, pp. 546-551  (1968).

 9.  Andersson,  L. 0., "Coupling of Dyes to Various Macromolecules by
     Means of Gamma Irradiation," Nature 222,  pp. 374-375 (1969).

10.  Lukens,  H.  R., Williams, C. B.,  Interim Report prepared under NSF
     GRANT GI-41892,  Intelcom Rad Tech, San Diego, California 92138
     (November 1974).

11.  Day, L.  A., Sturtevant, J. M., and Singer, S. J., Ann.  N.Y. Acad.
     Sci. 105, p. 611  (1963).

12.  Portman, A.  J.,  Levison, S. A.,  and Dandliker, W.  B., "Antifluorescein
     Antibody of High  Affinity and Restricted Heterogeneity as Character-
     ized by Fluorescence Polarization and Quenching Equilibrium Techniques,"
     Biochem. and Biophys.  Res. Comm.  43,  p.  207 (1971).
                                     35

-------
13.   Levison, S. A., Portman, A. J., Kiertzenbaum, F., and Dandliker, W. B.
     "Kinetic Behavior of Antihapten Antibody of Restricted Heterogeneity
     by Stopped Flow Fluorescence Polarization Kinetics," Biochem. and
     Biophys. Res. Comm. 43_, p. 258 (1971).
                                     36

-------
              APPENDIX  1
        THE PHYSICAL BASIS OF
FLUORESCENCE POLARIZATION IMMUNOASSAY
                 37

-------
This page intentionally left blank.
                38

-------
                            THE PHYSICAL BASIS OF
                    FLUORESCENCE POLARIZATION IMMUNOASSAY

To adequately understand the basic principles of fluorescence polarization
in an immunoassay, some basic discussion of the polarization phenomenon
itself must be presented.  In classical terms, the emission from a single
molecule may be regarded as radiation from a single oscillating dipole;
this radiation has an oscillating electric field parallel to the direction
of oscillation of the dipole and is said to be polarized in the same
direction.  Now, if a randomly oriented assembly of molecules is excited
by fully polarized light, their fluorescence is only "partially" polarized
(partially polarized light may be thought of as being a mixture of polarized
and unpolarized light), even if the molecules are prevented from rotary
brownian motion in solution, as may be seen by the following consideration.

For simplicity, assume that the direction of the absorption and emission
oscillators in a single molecule are the same and that they are rigidly
fixed with respect to the geometric axis of the molecule.  Furthermore,
assume the molecule is to be rigidly fixed in position during the interval
                                             -8
between absorption and emission (typically 10   second).   The probability
of absorption of light is proportional to the square of the magnitude of
the component of the electric vector of the exciting light in the direction
of the oscillator.  This probability is proportional to cos 9 (see Figure
Al-1), where 0 is the angle between the incident field E which is parallel
to the Z axis, and the direction of the absorption oscillator.  Because
the probability of absorption falls off as 8 increases, molecules oriented
so that 9 is small are preferentially excited, while those with large 9
have little chance of absorbing.  Since the absorption and emission oscil-
lators are parallel,  the emitted light will be partially polarized with a
degree of polarization P.  This quantity is defined (Ref. Al-1)  in terms
of the intensities I  polarized either parallel or perpendicular to the
incident electric field, and is given by
                                     39

-------
PROPAGATION
DIRECTION
                                         pSINO
RT-04089
        Figure Al-1.  Polarization Coordinate System
                              40

-------
 In Figure Al-1 the area  of one  face of  the  elemental  volume  is  equal  to

          (psin0A)  (pA9)

The elemental solid  angle is then given by
           2
               2
              P
          = sinGdedtf)
If the molecules in the medium are distributed randomly, the number of
oscillators excited in a given elemental solid angle must be proportional
to the solid angle and the probability of absorption of light by such an
oscillator.

However, since the probability of absorption of light is proportional to
the square of the magnitude of the component of the electric vector of
                                                                  2
the exciting light in the direction of the oscillator, i.e., « cos 6,
then, the number of oscillators excited in a given elemental solid angle is

          (cos 6 sine d6 d)

If the radiating oscillators have the same direction as the absorbing
oscillators, then the amplitudes of the electric vector, parallel to the
excitation vector and observed from the XY plane,  i.e., 6 = 90°, is:

          (I)(cos26 sine de d(J»") cose
                                    41

-------
 and the intensity which  is proportional  to  the square of the amplitude,
 is given by

          cos 6 sine  de  d<|> (cos6)  (I)2
               4
          = cos 6 sine de d

 integration over a sphere gives

                 •27T
          I.-  = /   /     cos e sine de  d<()
IT
The intensity of vibrations perpendicular  to the excitation vector is
computed similarly  as  follows.  The number of oscillators excited in a
given elemental  solid  angle is as before:
              2
          (cos 6 sine  d6 d)

The amplitude of the electric vector perpendicular to the excitation
vector and observed in the direction 6 = 90° is

          (l)(cos 6 sine de d)  sine cos

and the intensity is given by

             2               22
          cos 6  sine de d sin 6 cos 
               232
          = cos  6 sin  6 cos $ de d

Again, integration over a sphere gives

                .2ir  /.TT/2
                  /       223
          I  =         cos 4> cos 6 sin 6  de d<$>
 II
Jo  Jo
Evaluation  of  these two integrals  gives
                                   42

-------
          T    2ir      ,   T    2ir
          I  =       and   I  =
Now if the degree of polarization is defined by
          P = ^4t
then,
Thus, integration over the angles 6 and  (Figure Al-1)  shows  that  the
maximum value of P observed in the XY plane with linearly polarized  light
is one half.

Now, if the molecules are subject to rotary brownian motion  instead  of
being rigidly fixed, then the molecular rotation taking  place  between the
time of absorption and emission may be expected to result in values  of  P
lying between one half and zero.   The extent of this rotation  is  a  function
of molecular dimensions and structure, solvent  and temperature.   Low molecu-
lar weight compounds, such as inorganic ions, will give  rise to virtually
completely depolarized fluorescence.   Some  polarization  will be retained as
molecular size increases and considering two molecules of equal size the
fluorescence of the more asymmetric,  rigid  structure will be more highly
polarized.  If there is considerable internal flexibility within  a molecule,
very little polarization may be retained, because the fluorescent label
                                                      - 8
then may assume a wide range of positions within the 10   -sec  lifetime  of
the excited state,  even if the entire structure does not  rotate significantly
as a unit.
                                    43

-------
                                 REFERENCES
Al-1.  Feofilov, P. P., The Physical Basis of Polarized Emission.  Consul-
       tants Bureau, New York (1961).
                                      44

-------
      APPENDIX  2




EQUILIBRIUM EXPERIMENTS
           45

-------
This page intentionally left blank.
               46

-------
                          EQUILIBRIUM EXPERIMENTS

In order to provide a complete set of data concerning the kinetics of the
antibody-hapten reaction, a series of experiments to determine the equilib-
rium constant and the order of the reaction have been undertaken, and are
reported in this Appendix.

Definition of the symbols used in the following discussion are:

     I = net fluorescence intensity
     H = net intensity of horizontally polarized component of
         fluorescence
     V = net intensity of vertically polarized component of
         fluorescence
     F = total molar concentration of 2-FNCS-ABZI in solution
     Q = molar fluorescence, I/F
     p = fluorescence polarization
     R = molar concentration of unbound antibody
     f as a subscript denotes that unbound 2-FNCS-ABZI is involved
     b as a subscript denotes that 2-FNCS-ABZI bound to antibody
     is involved.
     o
      The vertical is perpendicular to the plane defined by the
      direction of observation and the exciting light.
      Actually it is the concentration of unbound binding sites.

In addition,  identities that  prove useful in the analysis are given in
terms of the  foregoing symbols by:
     I = V+H  = QF; If = Vf+Hf = QfFf;  Ib  = V^ = Q^   ,
     V - Vf+Vb
     H = Hf+Hb
     F = VFb
     « - V-H
     P - VTH    '

                                     47

-------
The basic  equations  that  are used can be stated with  the  foregoing
symbology  and  identified  as follows.
           lf
                                                                        (2)

                (Fb/Ff)F
           Fb  =

The  final  polarization  values at equilibrium obtained between various
amounts  of 2-FNCS-ABZI  and antibody aliquots were measured to obtain
titration  curves.  This was done for three different amounts of antibody,
and  the  data  are given  in Table A2-1 and plotted in Figure A2-1.   It can
be seen  that  the titration curves intersect the polarization axis  at a
maximum  polarization value of 0.328, which represents the polarization,
P.,  of 2-FNCS-ABZI that is bound to antibody.  Separate measurements of
the  fluor-tagged hapten in the absence of antibody gave a polarization
value, pf,  of 0.045.

Comparison  of 2-FNCS-ABZI fluorescence intensity with that of known amounts
of fluorescein showed that 4 \iL of the tagged hapten added to the  3.2 mH of
buffer in  the instrument cell gave a concentration of 2.5 x 10~    M.  Thus,
the  conversion of aliquot size to a concentration value in Table A2-1 is
straightforward.  In addition, it was observed that there was no change
in fluorescent intensity as the fluorescent-labeled hapten was bound to the
antibody.   Consequently, Qf/Qh must be approximately unity.  Hence, it is
possible to obtain F,/F,. from Eq. (2) and the data in Table A2-1,  and there-
after F,  from Eq. (3).  These values are given in Table A2-2, and  plotted
as F, /Ff versus F in Figure A2-2 for the case of 200 yL of antibody.
                                                             -9
Extrapolation of this curve enables an estimation of 1.5 x 10   M  for the
                                                      .9
maximum value of F,  to  be made,  i.e., F,     «1.5 x 10   M.
                  b           '      '  bmax
                                      48

-------
    Table A2-1.  TITRATION OF 2-FNCS-ABZI VERSUS ANTIBODY IN
             3.2 m£ OF BUFFER, POLARIZATION VALUES
2-FNCS-ABZI
Concentration, in
i n
yL
4
6
8
12
16
20
22
32
Units of
2.
3.
5.
7.
10.
12.
13.
20.
10 M
5
75
0
5
0
5
75
0
Antibody Aliquot,
ML
60 120
0.263 0.
0.
0.195 0.
0.166 0.
0.137 0.
0.


311
300
283
262
234
209


0

0
0


0
0

200
.318

.308
.303


.267
.227
Table A2-2.   TOTAL, FREE,  AND BOUND 2-FNCS-ABZI IN THE PRESENCE OF
                   VARIOUS AMOUNTS OF ANTIBODY:
           CONCENTRATIONS,  F,  GIVEN IN UNITS OF 10
                                                  -10
M
Antibody Aliquot, yL
2-FNCS-ABZI,
F
2.5
3.75
5.0
7.5
10.0
12.5
13.75
20.0
60 120
VFf Fb Fb/Ff
3.52 1.95 19.0
10.2
1.15 2.67 5.67
0.761 3.24 3.44
0.489 3.28 2.06
1.40


Fb
2.38
3.42
4.25
5.81
6.73
7.29


200
Fb/Ff
39.0

15.5
11.7


3.83
1.86
Fb
2.44

4.70
6.91


10.9
' 13.0
                               49

-------
     0.38
     0.34
     0.30-
     0.26-
     0.22-
     0.18-
     0.14-
        I         I

O  200 yL ANTIBODY
D  120 yL ANTIBODY
O   60 uL ANTIBODY
                  4        8        12        16        20
          CONCENTRATION OF 2-FNCS-ABZI  IN UNITS OF  10"10  M
RT-10719
        Figure A2-1.  Titration Curves 2-FNCS-ABZI vs Antibody
                                  50

-------
   16
   12
                                               Fb  IN  UNITS  OF  10"10 M
RT-10718
                            10
15
  20
Fb/Ff
25
30
35
           Figure A2-2.   Fb/F£ vs  Fb,  for  the Case of 2-FNCS-ABZI vs 200 yL of Antibody
                         (see Table~A2-2)
40

-------
 It  should  be noted  that  the maximum molar concentration  of binding  sites,
 R     (i.e., the  binding  sites  in  free antibody),  is  the  same  as  F,
 max                                                              bmax
 Therefore, when  200 yL of the  antibody preparation is  added to 3.2  mi  of
 buffer,  the concentration of binding sites  for  2-FNCS-ABZI is 1.5 x 10  M.
 It  is  also true  that  R = FL     -  F, .
                           bmax    b
 EQUILIBRIUM CONSTANT

 Ordinarily, one  might expect the  equilibrium constant  for the reaction,
 F+R ^  FR to be given by,
                (FR)
                        F.CF,   -F.)
                         f v  bmax  bj

However, the curvature  of Figure A2-2 indicates a nonlinear function.  There-
fore, it is necessary to use the generalized isotherm of Sips  (Ref. A2-1
after the method of Dandliker et al . (Ref. A2-2) to properly express the
mass law.  Specifically,

          Log Ff - I log  F-IP-   - ^g KQ   ,                       (5)
                         \  bmax  b/

where K  is the average  association constant, and a is a heterogeneity
factor.

Equilibrium data taken  from Table A2-2 were used to calculate F_ and
F,/(F,    -F, ) , plotted  in Figure A2-3,  which is known as a Sips plot.

Least-squares analysis  indicates
Log Ff = 1.38 log
                             F    -F
                              bmax  b
                                       - 10.1592
The slope is I/a, which indicates a heterogeneity factor of 0.72.  The
value of the intercept is -log K_, which indicates that the equilibrium
constant, K   is 1.44 x 1010.
                                      52

-------
    -8.6

    -8.8

    -9.0

    -9.2

    -9.4

    -9.6

    -9.8

   -10.0
   -10.2

   -10.4

   -10.6

   -10.8
   -11.0

   -11.2
RT-10717
                 T
                  O  60 yL OF ANTIBODY
                  D 120 pL OF ANTIBODY
                  O 200 uL OF ANTIBODY
                  I
I
-1.0     -0.6     -0.2
           LOG  (Fu/
                                             0.6
                 1.0
       Figure A2-3.  Sips Plot Based on Data From Table A2-1
                                 53

-------
INITIAL RATE DATA

Initial rate data are presented in Figures A2-4 and A2-5, where it can be
seen that the rate of change of polarization, dp/dt, is linear over the
first few minutes; thus the initial values for dp/dt are easily defined
and obtained.

The equation governing the initial rate of change of polarization, (dp/dt) ,
has been formulated by Levison and Dandliker (Ref.A2-3),and may be expressed
as follows.

                                  cF,)"2"1   .                          C6)

A number of additional initial rate experiments were carried out with a
different level of concentrations of antibody and 2-FNCS-ABZI, than was
used in the detection sensitivity experiment, and are summarized in Table
A2-3.   Plotting log (R)  versus log (dp/dt) , as shown in Figure A2-6,
demonstrates that the value for N. is essentially unity.   By plotting
l°g (Ff) versus log [(dp/dt) /(R) ], as shown in Figure A2-7, N? is also
shown to be essentially unity.  Substituting N_ ^ 1 % NI  and Qh/Qf ~1 in
Eq. (6)  gives
          (dp/dt)Q = (Pb-Pf)k(Ro)(Ff)Q   .                              (7)

                    Table A2-3.  INITIAL RATE EXPERIMENTS
Reactants,
in Units
Antibody (R)
15.0
2.4
7.7
2.4
7.7
7.4
7.7
7.7
7.7
3.8
3.9
Concentration
of 10"10 M
2-FNCS-ABZI (Ff)
3.3
3.5
1.3
1.3
4.4
12.0
4.4
1.3
1.3
13.0
4.4
Initial Rate
(min)
•z
(dp/dt)Q x 10^
31.5
4.8
15.7
4.2
12.5
12.7
12.1
10.7
13.5
5.5
6.8
                                     54

-------
Figure A2-4.
           20
TIME (MINUTES)
Standard Inhibition Curve
                                          30
                  55

-------
    t
    o
    p
    h-J
    QC
    O
    a.
   0.09
   0.08
   0.07
RUN
T
Q
P
INHIBITOR
CONCENTRATION
(ng/mfc)
44.3
133
1330
         A  A
                                A
RT-08875
10               20

       TIME (MINUTES)
                                                          30
   Figure A2-5.  Rate of Change of Polarization as a Function of
                 Inhibitor Concentration
                                   56

-------
X

 o
Q.
•o
    1.4
    1.2
    1.0
    0.8
    0.6
    0.4
    0.2
 RT-10715
                0.2
0.4
                      Y = 1.06X + 0.21
0.6
0.8
1.0
1.2
                                                         10,
                      X, LOG (ANTIBODY CONCENTRATION X 10IU)
             Figure  A2-6.   Reaction Order with Respect to Antibody
                                      57

-------
     0.6
o
x
 -e
• It*
•°'    0.4
     0.2
                 0.2
  RT-10716
                                                 Y  =  -0.093X  +  0.30
                            0.4
                                    0.6
0.8
1.0
1.2
                             X, LOG (2-FNCS-ABZI x 1010)
     Figure A2-7.   Reaction Order with Respect to Fluor-Tagged Hapten
                                    58

-------
 INTEGRATED RATE DATA

One can effect a psuedo-first-order condition by using sufficient excess
of antibody to have (R) ~ (R)  over any time, t.  Then, the relationship

          -(dF./dt) = K'Ff   ,   can be applied where                   (8)

K' = k(R ) is the pseudo-first-order rate constant.

Integration of Eq.  (8) gives
                                                                       (9)
and since

          F
           fO    F   Ff + Fb
           Ff   Ff      Ff         Ff
substitution from Eq. (2) in Eq .  (9), with Qf/Q.  = 1, gives

                 P - P
                 p
                          = K't   .                                     (10)
Thus, K't can be calculated from polarization data.
                              -Q                             -10
A run involving R  = 2.32 x 10   M and 2-FNCS-ABZI = 2.5 x 10    M gave
the data listed in Table A2-4 and the calculated values of K't, also listed.
Figure A2-8 is drawn with K't versus t, and allows the estimation of K' as
5.8 x 10"  sec" , from which it follows that the second-order rate constant,
k, is approximately 2.5 x 10  M   sec
                                     59

-------
Table A2-4.  PSEUDO-FIRST-ORDER KINETIC DATA

                     -9                -10
WHERE  R  = 2.32 x 10   M, F = 2.5 x 10    M,



       P, = 0.040, AND P^    = 0.328
        f               bmax
t
(sec)
21
30
39
48
60
72
81
87
96
102
108
117
135
P
0.127
0.140
0.156
0.167
0.181
0.191
0.200
0.205
0.211
0.214
0.217
0.224
0.234
K't
0.357
0.424
0.513
0.579
0.673
0.739
0.811
0.851
0.901
0.927
0.949
1.014
1.120
t
(sec)
150
159
171
183
198
213
258
270
282
294
318
342

P
0.245
0.246
0.252
0.255
0.262
0.267
0.281
0.284
0.287
0.288
0.300
0.298

K't
1.238
1.256
1.332
1.373
1.473
1.544
1.813
1.879
1.949
1.974
2.151
2.245

                    60

-------
    2.2

    2.0

    1.8

    1.6

    1.4

    1.2

    1.0

    0.8

    0.6

    0.4

    0.2
                   50
100
              K't =  0.0058t  +  0.3143
                                           I
                         I
150         200
   TIME (sec)
250
300
350
RT-10714
            Figure A2-8.   Fluorescence Polarization, Pseudo-First-Order Representation

-------
                                 REFERENCES
A2-1.  Sips, R., J. Chem. Phys. 16, p. 490  (1948).

A2-2.  Dandliker, W. B., Shapiro, H. C., Meduski, J. W., Alonso, R.,
       Feigen, G. A., and Hamrick, J. R., "Application of Fluorescence
       Polarization to the Antigen-Antibody Reaction," Immunqchem.  l_,
       p. 165 (1964).

A2-3.  Levison, S. A., and Dandliker, W. B., "Effect of Phosphate  Ion on
       Ovalbumin-Antiovalbumin Kinetics, as Measured by Fluorescence
       Polarization and Quenching Techniques," Immunochem. 6_, p. 253
       (1969).
                                     62

-------
                                   TECHNICAL REPORT DATA
                            (I'leasc read luuructions on the re\ crsc before completing)
 1. REPORT NO.
 EPA-650/1-75-004
                              2.
                                                           3. RECIPIENT'S ACCESSIOI*NO.
 4. TITLE AND SUBTITLE
 A Fluorescence  Immunoassay  Technique for
 Detecting Organic Environmental  Contaminants
                                                           5. REPORT DATE
                   J975
             6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
 Herbert R. Lukens; Colin  B.  Williams;  Stuart A. Levison;
 Walter B. Dandliker; and  Dennis  Muryama
                                                           8. PERFORMING ORGANIZATION REPORT NO
               INTEL-RT 8098-004
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
 IRT Corporation, P.O.Box  80817,  San Diego, CA 92138

 Scripps Clinic and Research  Foundation
 476 Prospect Street, La Jolla,  CA 92037
              10. PROGRAM ELEMENT NO.
               1EA078
              11. CONTRACT/GRANT NO.

               68-02-1266
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Pesticides and Toxic Substances  Effects Laboratory
 National Environmental Research Center
 Research Triangle Park, North  Carolina 27711
                                                           13. TYPE OF REPORT AND PERIOD COVERED
               Final; 11/73-11/74
              14. SPONSORING AGENCY CODE
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
 This report describes the development and successful demonstration  of an immunological
 assay for the detection of  low  molecular weight organic contaminants  of environmental
 concern.  The specific technique  described is a fluorescence polarization immunoassay,
 the theory of which is presented.   The preparation of the two  required reagents, namely
 a fluorescent conjugate of  the  contaminant of interest, together  with an antibody to
 the contaminant, is described in  detail.   The specific contaminant  chosen for this
 study was 2-aminobenzimidazole  (MM  =  133), a metabolite of certain  fungicide agents
 used in agriculture.  The particular  fluorescent moiety chosen to form the conjugate
 with 2-aminobenzimidazole was fluorescein.  A successful demonstration of the as^say has
 been accomplished, and a detection  sensitivity in the sub-nanogram/m£ range obtained.
 A high degree of specificity of the antibody for the hapten has been  demonstrated, and
 a successful quantitative recovery  from an unknown solution has been  obtained.   This
 report is submitted in fulfillment  of Contract 68-02-1266 by IRT  Corporation,  San
 Diego, California, under the sponsorship  of the Environmental Protection Agency.  The
 program was initiated in November 1973 and successfully completed in  November 1974.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                           c. COS AT I Field/Group
 Fluorescence Polarization
 Immunoassay
 Organic  Contaminants
 Pollutant  Identification
  2-Aminobenzimidazole
  Benzimidazole
 8. DISTRIBUTION STATEMENT


             RELEASE UNLIMITED
19. SECURITY CLASS (This Report)
  UNCLASSIFIED
21. NO. OF PAGES

      75
20. SECURITY CLASS (Tins page)
  UNCLASSIFIED
                           22. PRICE
EPA Form 2220-1 (9-73)

-------