Note: Descriptions are shown in the official language in which they were submitted.
CA 02132708 2002-08-14
77036-13
DESCRIPTION
Fluorescence Immunoassays Usina Fluorescent
Dyes Free of Aaareqation and Serum B:indina
Field of the Inver~tion
The present invention relates to methods for deter-
mining the presence or amount of antigenic substances in
samples. The invention is directed to fluorescence
immunoassays using particular fluorescence dyes which are
essentially free of aggregation and serum binding and,
thus, are particularly suited for the measurement of
antigenic substances in biological materials such as
serum, plasma and whole blood.
:10
Background o~ tie invention
The determination of the presence or amount of anti-
genic substances is commonly performed by immunoassay.
Immunoassay techniques are based on the binding of the
antigenic substance being assayed (the "target analyte")
and a receptor for the target analyte. Either the target
analyte or the receptor may be labeled to permit detec-
tion. Various labels have been employed for use in
immunoassays, including radioisatoDes, enzymes and
WO 93/19366 - PCT/US93/02470
~~.~~1~~ ' , ~.
~,.
2
fluorescent compounds. Many different types of immuno-
assays are known in the art, including competitive
inhibition assays, sequential additian assays, direct
"sandwich" assays, radioallergosorbent assays, radio-
s immunosorbent assays and enzyme-linked immunosorbent ,
assays.
The basic reaction underlying most immunoassays is
the binding of certain substance, termed the "ligand" or
"analyte", by a characteristic protein (receptor) to form
a macromolecular complex. These binding processes are
revers~.ble reactions; and the extent of complex formation
for particular analyze and receptor concentrations is
regulated by an equilibrium constant according to the law
of mass action. Thus, at equilibrium, some of the analyte
always exists unbound (free).
In a competitive inhibition immunoassay, the unknown
quantity of target anal~yte in the sample competes with a
known amount, of labeled target analyze for a limited
number of receptor binding sites. The reagents usually
consist of a labeled target analy~te, such as an antigen,
and a solid phase coupled receptor, such as an antibody.
The antigen to be assayed competes with the labeled anti-
gen for binding sites on the coupled antibodies. The
concentration of target axaalyte present in the sample can
be determined by measuring the amount of labeled target
a~aa~.yte - either "free" or .'b~und. " Th.is is an indirect
assay method where the amount of labeled antigen bound to
the antibodies is intrersely correlated with the amount of
antigen in the test'solution. Thus, low concentrations of~
target analyte in the sample will r~sul~ in low cdncentra-
t.ions of !' free" labeled taxget : analyte and high concentra-
tions of "bound!' labeled target analyte, and vzce versa.
The amount of "free"° or °°b~und'! labeled target analyze
is
measured using a suitable detector: Quantitative determi-
nations are made by comparing the measure of labeled tar-
g~t a~alyte with that obtained for calibrated samples
containing known quantities of the target analyte. This
S lJ ~ S°~I~tJTE S H E ~T'
y~ y)3/~~3~~ .. P~.'T/'US93/U2470
~v ;
3
method has been applied to the assay of a great number of
different polypeptide~ hormones, enzymes and immunoglobu-
l.ins. This method may also be used as a total liquid
system. .
S zt is apparent to those skilled in the art that it is
not absolutely necessary that the labeled analyte be iden-
tical to the un~.abeled target analyze. If there is a dif-
ference between the two, for example, if the labeled ana-
lyte is an analog of the target analyte, the reaction
between labeled and unlabeled analytes may be considered
to be competitive for the receptor binding sites; and the
reaction will still provide quantitative answers, provid- ,
ing the difference in affinity of the analyzes is not too
great . Whether or not true competition occurs in a sys'cem
consisting of labeled analyze, unlabeled analyze, and
receptor depends on the'nature of the labeled analyte and
the specificity of the receptor.
In sequential addition assays; the reagents used are
the same as in the comgetitive inhibition assay described
above: However, instead of incubating them at the same
time, the unlabeled antigen is first incubated with the
ar~ta.body, then the labeled antigen is added.
Direct immunoassay systems are also known in the art .
Such assays,, also termed "zmmunometric'! assays, employ a
labeled receptar~(antibody) rather than a labeled analyte
(ant~.gen) , In these assays the amount of labeled receptor
associated with the complex is proportioned to the, amount
of analyze in the sample. Immunometric assays are well-
suited to the detection of antigenic substances wrhich aver
~0 'able to; complex w~.th two or more antibodies at the same
f.2me: In such "two-site" or "sandwich'° assays, the anti-
geni:e substance has two antibodies bound to its surface at
d.if f~rent l.ocatians : In a t~rpical "forward" ' sandwich
assay, an antibody bound to a olid phase is f first con-
tacted with the sample being; tested to form a solid phase
antibody:ant~.gen complex. After incubation, the solid
support is washed to remove the residual sample, including
5 ~9 ~ ST1~°LIZE S H E ET
WO 93/19366 = 1'CC/L'S93/02~70
..
4
unreacted antigen, if any. The complex is then reacted
with a solution containing a known amount of labeled anti-
body. After a second incubation to permit the labeled
antibody to complex with the antigen bound to the solid
support through the unlabeled antibody, the solid support
is washed to remove unreacted labeled antibody. The assay
can be used'as a simple "yes/no" assay to determine whe-
ther the antigen is present. Quantitative determinations
can be made by comparing the measure of labeled antibody
with that of calibrated samples containing known quanti
ties of antigen. "Simultaneous" and "reverse" sandwich
"' assays are also known in the art. A simultaneous assay
involves a single incubatian step, both the labeled and
unlabeled antibodie s being added at the same time. A
reverse assay involves the addition of labeled antibody
followed by addition of unlabeled antibody bound to a
suitable solid support. The sandwich technique can also
be used to assay antibodies rather than antigens . Such an
assay uses as a first receptor an antigen coupled to a
0 solid phase. The antibodies being tested are first bound
to the solid phase-coupled antigen. The solid phase is
a
then washed, and then labeled anti-antibody (second
receptor? is added.
The radioallergosorbent technique (R.AST) is a method
for the determination of antigen-specific IgE. The method
uses a solid phase coup~.ed antigen and an immunoabsorbent
purified antibody labeled with a radioactive isotope . The
rctethod is used to detect reaginic antibodies against vari
,, :
lous antigens which elicit allergic reactions (allergens).
The reaginic antibodies react with allergen bound to a
s;;:
solid matrix. After washing of the solid phase, the ~
allergen-bound reaginic an ibodies are detected by their
ability to hind labeled antibodies against IgE . A variant
of RAST can be used for the determination of allergens.
The allergen to be tested is incubated with the reaginic
antibody. The mixture is then tested with BAST using the
same allergen coupled to the solid matrix. The allergen
SE~~S "TF SHED
WO y3/~~366 - ~CT/l.'S93/02470
~~~~i
in solution reacts with the reaginic antibodies and thus
inhibits the banding of these antibodies to the solid
phase-coupled allergen.
Another assay method for the determination of IgE is
5 the radioimmunosorbent technique ("RIST"). In this
method, the solid support is sensitized with anti-IgE and
increasing amounts of labeled IgE are added to determine
the maximum amount of IgE that can bind. A quantity of
labeled IgE equivalent to approximately 80% of the plateau
~.0 binding is chosen. In the test experiments, this amount
of labeled IgE is mixed with the serum containing the IgE
- to be tested. The test IgE competes with the labeled IgE.
The more IgE present in the test serum the less the amount
of labeled IgE that binds. Thus, by producing a standard
curve the amount of 2gE in a sample can be determined.
The above immunoassay methods can be applied to the
assay of many different biologically active substances.
g~ohg such substances are haptex~s, hormones, gamma globu-
lin, al~.ergens, viruses, virus subunits, bacteria, toxins
such ds those associated with tetanus arid animal venom,
and many drugs. Similar techniques can be used in non-
immunologa.cal systems with, for example, specific binding
proteins.
A~.th~~.gh some of the immunoassay methods described
above utilize radioactive labels, those skilled in the art
will appreciate that the assays can be adagted to use an
alternate label, for example, a fluorophore.
If the properties of the label are not altered by
'binding, for example, as in a radioimmunoassay,~'a separa
tion step is required to separate "free" from "bound"
labeled target analyte. Such assays, which require a
separation step, are called "heterogeneous" assays. If
the properties of the label'are altered in some way. when
it is bound, no separation step is required, and the
3 S imrtlunoassay is termed '~ homogeneous . "
The measurement of target analyzes in biological
fluids, such as serum, plasma and whole blood, reauires
su~s°rrr~~E s~~~r
WO 93/19366 - PC'f/US93/02470
.,
y : : ..1. .
ha n
6
immunoassay methods which are bath specific and sensitive.
Both the specificity and sensitivity of an immunoassay
depend on the characteristics of the binding interaction
between the target analyte and the receptor involved. For
example, the reaction must be specific for the analyte to
be measured and the receptor used should not bind to any
other structurally related compounds. In addition, by
choosing a receptor with a high affinity for the target
analyte, the sensitivity can be increased.
The label used to monitor the assay affects the
sensitivity of an immunoassay. Labels currently used for
immunoassay of target analytes in biological fluids
include radioisotopes (radioimmt~noassay, RIA), enzymes
(enzyme immunoassay, EIA); fluorescent labels (fluores-
cen~e immunoassay; FIA); and chemiluminescent labels
(chemiluminescent immunoassay, CzA).
RTAs are sufficiently sensitive for use in detection
in lour concentrations ~f analyt~s because of their low
background. They are disadvantageous in that they are
heterogeneous, thus requiring a separation step before
msas~,rernent of the bound and/or free portions of labelled
target analyte. RIAs involve the inconvenience and haz-
ards assoc~.ated with the handling and disposing of radio-
isotopes. In addition, they are labor intensive and have
a short shelf life due to the half-lives of radiolabels
and to chemical damage produced by the emitted radiation. ,
EIAs have the advantage of increased signal over
background, longer shelf life, lack of radiation hazards,
)
They are disadvantageous in that, r
and homogeneity.
because they invol~re enzyme kinetic reactions, they are
t:,~:
affected by the time of the kinetic measurements, as well
as by variat~.ons in temperature, gH and ionic strength.
The temperature of the enzyme incubation is particularly
critical, and variations of more than 0.5C can signifi-
cantly affect assay results: Thus, drifts in standard
curve may result from temperature fluctuation and incon-
sistencaes in sample handling. Enzyme activity may also ,
58~~5T~'6JT'E SHEET
d~0 93119366 - PC'I"/fS93/02470
be affected by constituents in biological samples, such as
plasma constituents. See gen:erall Strong, J.E. and
Altman, R.E., "Enzyme Immunoassay: Application to Thera-
peutic Drug Measurement," in P. Mover et al., Applied
Therapeutic Drug Monitorinct, American Association of
Clinical Chemistry (1984).
Chemiluminescent immunoassays (CTAs) offer a fairly
high degree of sensitivity (picomole per liter range) but
lack specificity in some instances. CIAs are disadvanta-
la geous because they are heterogeneous, require expensive
reagents, and are expensive to automate. See generally
Boeckx, R.L., "Luminescence: A New Analytical Tool for
Therapeutic Drug Monitoring," in P. Mover et al., Applied
Therapeutic Druq., Monitorincc, American Association of Clin
ical Chemistry (1984).
FIAs use fluorescent molecules as labels. Fluores-
cent molecules (fluoraphores) are molecules which absorb
light at one wavelength and emit light at another wave-
length. See Burd, J.F., "Fluoroimmunoassay ~-- Application
to Therapeutic Drug Measurement, " in P. Mover et al.. , Apps
lied Therapeutic Druct Monitorinct, American Association of
Clinical Chemistry (1984). Typically, an excitation pulse
of radiation is directed onto or into a sample, followed
by fluorescence of the sample, and the detection of the
fluorescence radiation.
FIAs may be either heterogeneous or homogeneous. As
noted above, homogeneous assays are usually sampler to
perform and are thus, more amenable to automation. How-
i , ~ , ~
ever, previously~known homogeneous FIAs are less sensitive
than heterogeneous FIAs because high background can limit
serasitivity. The heterogeneous FIA procedures can detect
smaller amounts of analyte than present homogenous FIAs,
but only because the separation and washing steps in the
assays serve to eliminate background interference from
biological substances. In solid phase fluorescent assays
the solid support can limit sensitivity at the wavelengths
of presently used fluors. In many cases the support
S ~! ~ S°~T'~ TE S ~ E ~~'
WC193f 19366 = PCT/L'S93/02470
8
itself will fluoresce at wavelengths of commonly used
floors such as fluoresceir~ (493 nm). FIAs also offer the
advantage of using stable reagents, a
Another assay method uses enzyme-enhanced fluores
cence technology which combines microparticle capture and
antigen-antibody reaction with an enzyme rate reaction
using' a fluorescent enzyme substrate. The rate reaction
is monitored by steady state fluorometric measurement. Tn
an enzyme-enhanced fluorescence assay, the analyte in
~0 question is "captured" by an antibody bound to a solid
phase and the solid phase is washed. An enzyme is then
_ bound to the captured analyte using an enzyme-anti analyte
conjugate. Excess reactants are washed away and the
amount of enzyme is measuxed by the addition of a i~on
fluorescent substrate. As the enzymatic reaction pro
ceeds, the non-fluorescent substrate is converted to the
fluorescent product. For example, an alkaline phaspha-
tase-labeled antibody can be used to catalyze the hydro-
lysis of 4-methylumbelliferyl phosphate substrate to the
fluorescent product methylumbelliferone. Thus, the rate
at which the fluorescent product is generated is directly
proportional to the concentration of analyte in the test
solution: Enzyme-enhanced fluorescence assays, like EIAs,
have the disadvantages associated with enzymes.
2S As discussed above; fluorescence is a phenomenon
exhibited by certain substances, which causes them to emit
light, usually in the visible range, when radiated by
another light source. This is not reflection, but crea-
tibn of news light ; Current commercially' available assay '
methods use fluorescein, which emits green light when
radiated by a light source containing blue light.
Zn addition to fluorescing; flunrescein (and other ~'
fluorophores) emit polarized light: That is, the light
emitted has the same direction of polarization as the
incident polarized light, if the fluorescein molecule is
held ' fixed with its transition moment parallel to the
electric field of the excitation. The amount of polariza
S tJ ~ S'~'t~'LJTE S H E ET'
~V~ 93/19366 - P~lf/US93102~170
9
E
tion in the emission can be defined in terms of the inten-
sity of the horizontally and~vertically polarized light,
as follows:
P = ( Iv - Th) r ( Iv + Th) ( 1 )
S where Iv = intensity of vertically polarized emission
Ih = intensity of horizontally polarized
emission
The maximum, or limiting value of polarization, for
fixed, randomly oriented molecules is 0.5 (Po).
A second eguation (the Perrin equation) defines
polarization in terms of physical parameters and Po:
lrp ~ ~r3 _ (~,rPo _ lr3) W + 3trry (2)
where t - fluoresce:ace lifetime, a constant
r = rotational relaxation time
1.5 Rotationrelaxation is further defined for spherical
molecules as
r = 3nV/R,T ( 3 )
where R = gas constant
T = temperature, °K
n = solution, viscosity
V' = volume of molecule
The rotational relaxation time is a measure of the
rite at vahi,ch a ' riiol~cule will rotate when free in a solo-
tion. Note that the rotational relaxation time will
typically be dependent primarily on the molecular volume
and' shape; since solution viscosity and temperature will
b~ essentially constant in a normal assay. Thus, rota-
tional relaxation time; and consequently, polarization,
are affected only by the hydrodynamic properties of the
molecule. The smaller a molecule is, the smaller its
rcatational relaxation time, and the faster it rotates
StJ~S ~'~ SHEET'
WO 93/1936 ' PC'f/iJS93102470
. ..
(e. g., r = 1 nsec for fluorescein, 100 nsec for large
antibody complexes). For a constant, small, .fluorescence
lifetime (~ nsec for fluorescein), a small molecule r
retains little of the original polarization when irradi-
5 aced by polarized light, because the molecule rotates
rapidly and then fluoresces. an the other hand, a large
molecule rotates slowly and for the same fluorescence
lifetime, still retains a large degree of the original
polarization when it fluoresces.
10 This dependency of polarization on molecular size can
be used to determine the presence or amount of drug.
- Using a fluorescent polarizing probe in a competitive
binding immunoassay provides a type of FIA called a fluor-
escence polarization immunoassay (FPIA). In this type of
assay, the smaller the molecule is, the smaller its rota-
tional relaxation time and the faster it rotates. Typi-
cally, antibody molecules are much larger than drug or
drug-probe molecules. For example, r 1 nsec for
fluorescein and 57 nsec for gamma globulin.
When there is a large amount of drug present, there
are very few binda.ng sites available for the drug-probe. ;,
As a result, most of the probe (fluorescein) is in the
form of small drug-probe molecules. As these molecules
rotate randomly and rapidly, a low polarization value
results: When there i.s a small amount of drug present,
much of the drug-probe is bound to the large antibody
molecules. These molecules rotate slowly, so the emitted
light will be highly polarized.
"' ' The relationship between polarization azad .drug con
centration can be determined by creating a standard, or
calibration; curve: 'This is done by running an assay
using a range of known drug concentrations, from the
lowest to highest expected concentrations, and plotting
the resulting values of polarization: Thereafter, for a
gi~ren ~ralue of polarization; Che drug concentration can be
determined from the standard curve.
SL1~5 'TE S~EE'T
WO 93119366 ' ~ PCf/U593102470
m
One advantage of the polarization technique is the
elimination of a step to separate unbound probe. Although
the unbound tracer is not physically eliminated from the
samples in FPIA, its contribution is readily assessed by
the polarization.
Another advantage in the FPIA technique is lack of
dependence on intensity. In equation (1) above for cal-
culating polarization using intensity, the ratio makes the
polarization value unitless, or independent of variations
1.0 in the intensity. Unlike most assays using a light mea-
surement, in which it is the intensity of the light that
is correlated to drug concentration (so any variations in
source light intensity will directly affect the sensitiv-
ity of the assay), the sensitivity of FPIAs is independent
of'intensity variations. Conventional FPIAs require sepa-
rate measurements of both blank and sample.
Theoretically, fluorometry is capable of being the
most sensitive of all analytic tools as it is possible to
detect single photon events. A problem which has plagued
fluorescence immunoassays has been discriminating the
,.
fluorescent signal of interest from background radiation.
The intensity of signal from background radiation may be
up to x.0,000 tames larger than the intensity of the
fluorescent signal of interest.
The problem of background detection is particularly
pronounced in assay of biological samples. Many of the
current fluorescence assays use the fluorescent molecule,
fluorescein. Fluorescein has an excitation maximum of 493
n~, and there are numerous substances in biological fluids
with overlapping excitation and emission similar to fluor-
escein. For example, in the analysis~of blood plasma, the
presence of a naturally occurring fluorescable material,
biliverd3:n, causes substantial background radiation. Such
i
compounds are highly fluorescent and contribute signifi-
cant background signals which interfere with the label's
signal, thus limiting the sensitivity of assays usin=
fluorescein labels.
S L! ~ S"1'~J'~E S ~# E E'T'
CVO 93/9366 ' PCT/US93/02470
12
Earlier attempts to overcome the problem of back-
ground radiation have met with limited success. One
technique far overcoming the problem involves discrimi- '
hating against background radiation on the basis of
wavelength. Filters have been used to reject detected
radiation at all but a narrowly defined wavelength band.
This technique has been less than successful principally
because the background radiation may also be at the same
wavelength as the desired fluorescence signal, accord-
ingly, still be passed through the filter and detected.
It has been recognized that for analysis of biologi-
cal fluids, it would be desirable to use a dye or label
which is excitable at radiations of wavelengths of greater
than background radiation. However, even though the back-
ground fluorescence of serum falls off at wavelengths
approaching 600 nm, significant decrease does not occur
until 650 nm or greater. Previous attempts to create dyes
of such wavelengths have been unsuccessful. See, e:g-,
Rotenberg, H. and Margarfit, R., Biochem. Journa1,229:197
0,985); and D.J.R: Lawrence, Biochem. Journal 51.:168
( 1952 ) .
A second technique attempting to discriminate the
desired fluorescent signal from the background is the so
called time gating approach. Here, the fluorescent signal
is observed in a short time window after the excitation.
The time window may be varied both in its length and ire
~-is starting time. Through the use of the variable time
window, the detected radiation may be observed at the,
,. ;
maximal time for detection sensitivity. Hi.storically,l
tk~is technique has used a fluorophore of very long decay
time (such as 1,000 nanosecands) to allow the background ,
fluorescence to substantially decay before detection of
the fluorescent signal of'interest. Generally however,
long, decay time fluorophores require longer times for
o~rerall analysis. i~ue to the long decay time, the light
source cannot be pulsed rapidly to coiiecc data, thus
requiring additional time for final analysis.
S tJ ~ S'~ITt~JTE 5 ~ E E~'
CA 02132708 2002-08-14
77036-13
13
Historir_ally, there have beers two excitation pulse
formats for transient state fluorescent analysis. One format
utilizes a single, relatively high power pulse which excites
the fluorophore. The transient state is typically monitored
by a high speed photomultiplier tube b~~ monitoring the analog
signal representative of r_urrent as a function. of time.
Single pulse systems require sufficiently high. power to excite
a large number of fluorescent molecules to make detection
reliable. The other principal format f:or transient state
fluorescent analysis is a digital format which utilizes
repetitive excitation pulses. Ordinarily, pulses of
relatively short, typically nanosecond duration, light with
power in the microwatt range are repetitively supplied to the
sample at rates varying from 1 to 10,600 Hz. Ordinarily, the
excitation source is a lamp, such as a Xenon-lamp.
Frequently, the decay curve is measured digitally by
determining the time to receipt of a photon. One commercially
available system uses repetitive pulses (such as at 5,000 Hz)
and pulses the photomultiplier tube at increasingly longer
times after the flash in order to obtain a time dependent
intensity signal. Detection in such systems has proved to be
very time consuming and insensitive. A single analysis can
take on the order of one hour, even at relatively high
fluorescable dye concentrations (e=g. , 10-~' M) .
Recently, significant advances have been made in the
area of fluorescable dyes. In one aspect, dyes being
excitable by longer wavelength radiation, such as in the red
and infrared wavelengths, are now available. These dyes are
described in two commonly assigned Arrro.enius, U.S. Patent No.
5,403,928, entitled, "Fluorescent Marker Components and
Fluorescent Probes".
Further significant advancements have been made in
increasing sensitivity through data collection and analysis
CA 02132708 2002-08-14
17036-13
14
techniques. As disclosed in Dandliker et al., U.S. Patent No.
4,877,965, entitled "Fluorometer", time gatingf techniques are
used in conjunction with data collection and analysis
techniques to obtain an improved signal relative to the
background. Generally, the '965 Patent considers the detected
intensity as a function of time to be composed. of signals from
various sources, including the desired signal source, and
various undesired background sources. Optimization of the
desired signal is achieved through data collection and
analysis techniques.
Further significant advancements have been made in
the ability to measure relevant materials in immunoassays.
For example, using the technique described in Dandliker, et
al., U.S. Patent No. 5,302,349, entitled "Transient State
Luminescence Assays", al.lows the bound and free form of
materials in a homogeneous assay to be determined. Generally,
the technique requires measurement of the time-dependent decay
of the intensity of parallel and perpendicular polarization
components. By measuring the time-dependent decay of various
polarization states, it is possible to determine the bound and
free forms of materials such as haptens, peptides, or small
proteins in a homogeneous analysis format. Significantly, no
separation of the bound and free materials is required.
Despite the significant and promising improvements
made in the field of fluorescable dyes, and in the data
CA 02132708 2002-08-14
77036-13
analysis aspects, the actual methods and apparatus for
achieving and detecting fluorescence have heretofore
remained relatively unchanged. Utilizing even the most
sensitive and best equipment, analysis can take an hour or
5 more, even at high concentrations of materials. When
fluorometry is used far immunoassay in a clinical context,
time for analysis and proper diagnosis can be absolutely
critical. Patient survival can depend on accurate, timely
analysis. Additionally, rapid testing would permit
l0 retests of patients without having them wait significant
periods of time, resulting in more rapid and accurate
diagnosis. As to sensitivity, it is extremely desirable
to be able to detect minute amounts of fluorescable mater-
ial. However, as the amount of fluorescable material in
15 a sample decreases, the ratio of the signal increases.
Further, since the time for analysis depends on the amount
of fluorescent radiation received from the detector, low
concentrations generally require substantially more time
to analyze.
Heretofore, the time required for analysis has been
prohibitively high. Known methods and apparatus for FIAs
have failed to provide rapid and accurate diagnosis and
analysis of samples. This has been so despite the clear
and known desirability of the use of FIAs. For example,
the drug digoxin, which is used to treat congestive heart
failure, has a narrow therapeutic range (,i.e., serum
levels of 0.5 to 2.5 ng/ml) and is generally toxic at
concentrations greater than 2.1 ng/ml. Present assays
using fluorescence-based methodologies require an extract-
ing process to remove interfering substances, such as
proteins, in order to detect digoxin at it:a therapeutic
levels. This additional extraction step increases the
time, cost and equipment needed to perform the assay.
From the above discussion it can be seen that,
although many different t~~pes of immunoassays currently
exist, none is satisfactory for measuring small quantities
of target analytes in biological ~':uids such as serum,
CA 02132708 2002-08-14
X7036-13
16
plasma and, especially, whole blood. Accordingly, an
object of the present invention is to provide improved
processes far assay of antigenic substances. More speci-
fically, the present invention provides fluorescence
assays which allow the detection of low levels of anti-
genic substances in biological samples such as serum,
plasma and whale blood. The present invention also pro-
vides homogeneous fluorescence assays which allow rapid
and accurate determination of low levels of antigenic
substances in biological samples.
Summary of the Invention
The present invention is directed to methods for
determining the pzesence or amount of a target analyze in
a sample by using, as a label for the target analyze or a
receptor which is capable of specifically recognizing the
target analyte, a fluorophore moiety comprising a lumi-
nescent substantially planar molecular structure coupled
to two stabilizing polyoxyhydrocarbyl moieties, one
located on either side of the planar molecular structure.
By "target anal.yte" is meant the antigenic substance being
assayed, for example an antigen. By "receptor" is meant
a molecule or molecular component capable of specifically
recognizing the target analyze, For example, an antibody
may be a receptor for an antigen.
Use of such detectable labels or marker components in
immunoassays is advantageous in that these: labels have
substantially the same intensities of parallel and perpen-
dicular components of transient state fluorescence emis-
sion in the presence and absence of biological fluids such
as serum. Thus, assay methods using these labels are
capable of detecting low concentrations of target analyte
in biological fluids.
The methods of the present invention are particularly
suitable for use with the improved luorescence detection
system described in commonly assigned U.S. Patent No. 5,323
CA 02132708 2002-08-14
77036-13
1?
008 entitled "Fluorometer Detection System ".
In one aspect, the present invention is directed
toward competitive inhibition assay procedures utilizing
particular labels. In this aspect, the present invention
is directed to a method of determining the; presence or
amount of a target analyte by contacting a sample sus-
pected of containing the target analyte with a known
quantity of added target anaiyte or analog thereof linked
to a fluorescent probe which includes a detestably labeled
marker component made up of a fluorophore moiety which
includes a luminescent substantially planar molecular
structure coupled to two solubilizing polyoxyhydrocarbyl
moieties, one located on either side of the planar mole-
cular-structure; contacting the sample with a receptor
capable of specifically recognizing the target ligand; and
determining either the amount of fluorescent probe bound
to receptor or free fluorescent probe. The amount of
bound or free fluorescent probe in the unknown samples may
2.0 be compared with blank samples and samples containing
known amounts of target analyte.
In a preferred embodiment, the resultant mixture of
sample, fluorescent probe and receptor is diluted before
measurement of the amount of bound and/or free fluorescent
~:5 probe. The dilution step allows for greater sensitivity
of the assay. Particularly preferred are dilutions of
2-fold to 100-fold, preferably about 7-fold to about
50-fold, and more preferably about 35-fold.
In one aspect, the present invention provides an
30 improvement in immunoassay procedures which utilize a
label for either the target analyte for analog thereof) or
the receptor. The improvement is the use of a fluorophore
moiety comprising a luminescent substantially planar mole
cular structure coupled to two solubilizing polyoxyhydro
35 carbyl moieties, one located on either side of the planar
molecular structure. Assays using this type of label are
advantageous in that they are free of serum binding and
CA 02132708 2002-08-14
77036-13
18
aggregation and are therefore, especially suitable for
testing biological samples such as serum, plasma, whole
blood and urine.
In another aspect, the present invention provides a
S method for performing a "sandwich" ar "two-site" immuno
assay comprising the steps of:
(a) contacting a sample suspected of con-
taining a target analyte with a first receptor
capable of specifically recognizing the target
analyte to form a complex of the target analyte
and the first receptor, the first receptor
being labeled with a fluorescent probe which
comprises a fluorophore moiety comprising a
luminescent substantially planar molecular
structure coupled to two solubilized polyoxy-
hydrocarbyl moieties, one located on either side
of the planar molecular structure;
(b) contacting the complex with a second
receptor capable of specifically recognizing
the target analyte to, the second receptor
being bound to a solid carrier, to farm a com-
plex of the first labeled receptor, the target
analyte and the second receptor bo~.xnd to the
solid carrier; and
(c) measuring either the amount of labeled
first receptor associated with the solid car-
rier or the amount of unreacted labeled first
receptor.
A sandwich-type assay may be either a heterogeneous
assay or a homogeneous assay. If it is heterogeneous, it
may incorporate the additional step of separating the
solid carrier from the unreacted labeled first receptor.
Homogeneous assays are generally preferred because they
are more rapid.
In another embodiment, the assay may incorporate the
additional step of relating the amount oL labeled first
receptor measured in the unknown sample to the amount of
CA 02132708 2002-08-14
77036-13
19
labeled first receptor measured in a control sample free
of the target analyse, or to the amount of labeled first
receptor measured in samples containing known quantities
of target analyze.
In another aspect, the present invention provides a
method for a simultaneous sandwich-type ass ay comprising
a method for determining the presence or amount of a
target analyte in a sample comprising the steps of:
(a) simultaneously contacting a sample
:LO suspected of containing a target analyze with
first and second receptors capable of specifi
cally recognizing the target analyte, the
first receptor being labeled with a fluorescent
probe which comprises a fluorophore moiety com
L5 ~ prising a luminescent substantially planar mole-
cular structure coupled to two solubilized poly-
oxyhydrocarbyl moieties, one located an either
side of the planar molecular structure, and the
second receptor being bound to a solid carrier,
20 to form a complex of the first receptor, the
target analyze, and the second receptor; and
(b) measuring either the amount of labeled
first receptor associated with the solid car-
rier or the amount of unreacted labeled first
25 receptor.
In another aspect, the present invention provides a
method for a simultaneous sandwich-type assay comprising
a method further comprising the step of relating the
amount of labeled f irsz receptor measured to the amount of
30 labeled first receptor measured for a control sample free
of the target analyze, or relating to the amount of
labeled first receptor measured with the amount of labeled
first receptor measured in samples containing known
amounts of target analyte.
35 In another aspect, the present invention provides a
sandwich-type fluorescence immunoassay method for measure-
ment of a target analyze which is capable of recognizing
WO 93/i9366 ' PCTIL'S93102470
two different receptors independently without mutual
interference. The method utilizes two receptors, each of
which is labeled with a different dye. For example, one
receptor is labeled with a first dye having absorption and
5 emissian maxima of 680 nm and 690 nm, respectively, and
the other receptor is labeled with a second dye having
absorption and emission maxima of 695 and 70S nm, respec-
tively: Detection and quantitation of the analyte can be
made using either steady state or transient state measure-
10 menu . In either case, for the example given, excitation
would be at 680 nm and detection would be at 705 nm. This
type of assay is based on energy transfer and is advan-
tageous in that, it is homogeneous.
In preferred embodiments the present invention is
Z5 directed to immunoassay of biological fluids, including
seem, plasma, whore blood and urine. Preferably, rid
blood cells in whole blood are lysed,prior to assay o~
whole blood samples: Preferred methods of lysing red
blood cells include addition of tearyl-lysolecithin,
20 palma.~oyl--lysolecithin and myristoyl lysolecithin.
Depending on the type of immunoassay used, the target
analyte may be an antigen. a hapten or an antibody; and
the receptor may b~ an antigen or antibody. The antibody
may be pr~lyclonal. or monoclonal. Preferably, the antibody
is a monoclonal antibody. Monoclonal antibodies useful in
the present invention may be obtained by the Kohle~ &
Milstein method reported in Nature 256:495-497 (1975). .
Alternatively, ~h.ey may be produced by recombinant
i , , ~ ~ j
(methods. Science 246:1275-1281 (1989) . ,
Tn one embodiment; the target analyte is a drug or a
metabolite o~ a drug. The drug may be a steroid, hormone
antiasthmatic, antineoplastic, ~n;tiarrhythmic, anticonvul-
sant> antiarthra:~ic, ant~.depressant, or cardiac glycoside.
'Examples of such drugs include digoxin; digitoxin, theo
phylline, Phenobarbital; thyroxine, N-;acetylproc~inamide,
primidone, amikacin, gentamicin, netilmicin, tobramycin,
earba.mazepine; e~hosuxima.dp, valproic acid, disopyram~.de,
St~~51"E SHEET
WO 93/19366 ' . PCTlUS93/02470 s
~s ~. c~ ~ f~
21
lidocaine, procainamide, quinidine, methotrexate, ami- y
triptyline, mortriptyline, imipramine, desipramine, van-
comycin, and cyclosporine. In a preferred embodiment, the
drug is digoxin.
In another embodiment, the target analyze is a pep-
tide, for example, a peptide hormone such as luteinizing
hormone, follicle stimulating hormone, human choriogonado-
tropin, thyroid stimulating hormone, angiotensin I, angio-
tensin II, prolactin or insulin. The peptide may also be
a tumor marker such as carcinoembryonic antigen. Or, the
peptide may be a virus or portion thereof, for example,
rubella virus or a portion thereof.
The methods Qf'the present invention provide ways of
measuring target analytes in concentrations of from about
I x 10'S M/L to about 1 x 10'13 M/L, and particularly in the
concentration range of from about 1 ~c 10-9 M/L to about 1 x
10'~a M/L. For measurement of drugs and their metabolites,
the present methods allow measurement in the range from
about 5 x 10'9 M/L to about 5 x 10-1z M/L, and particularly,
concentrations of from about 1 x 10'1° M/L to about 5 x
M/L. For measurement of peptides; the present methods
allow measurement in the range of from about 1 x 10'11 M/L
to about 1 x 1.0'lz M/L.
The measurement of amount of fluorescent probe
bound or free or both -- can be determined by measuring
steady~state fluorescence or by measuriztg transient state
fluorescence. In a preferred embodiment, the wavelength
,of~ light measured a.s greater khan about 500 nm, preferably
greater than about 650 nm, and more preferably greater
than about 680 n,m or 690 nm. Because the transient.state ,
detection system utilises a laser diode, it is necessary
for the dyes to have excitation maxima matched to the
d~.ode output wavelengths. Dyes have been made available
to match other commercially available laser diodes have
output wavelengths of 680; '690, 720, 750, or 780 nm.
Thus, the wavelength of the light measured may be greater
than about 580 nm, 690 nm, 720 nm, 750 nm ar 780 nm. The
S ~ ~ ST'1?~TE S H E E°~
'VVO 93/1936t~ ,' 1PCT/L'S93/02470
~~.~2'~U~
22
further into the red region of the spectrum one moves,
i.e., the greater the wavelength, the greater signal
ezirichment there is over background. .
In a preferred embodiment, detection and quantitation t
is performed using transient state measurement. Transient
state energy transfer offers improved measurements due to
optimization of the wavelengths of absorption and emir-
sion, as well as due to optimization of the decay times of.
the fist and second dyes. Such optimization allows
removal of Rayleigh and Raman scattering; and achieving
the best compromise between ef f iciency of transfer and the
undesired direct excitation of the second dye by the first
dye.
In one aspect, the present invention is directed to
immunoassays using de~~ctably labelled marker components
which comprise a fluorophore moiety which comprises a
sub~tanta.ally planar macrocycl~.c multidentate ligand
coordiz~ated to a central atom and two olubilizing poly-
o~yhydrocarbyl moieties, one linked on either Bide of the
plane of'the mul identa~e ligand to the central atom.
In one preferred aspect, the present invention is ,
directed to immunoassays using a marker component compris-
ing a fluorophore moiety which 'comprises a substantially
planar multidentate macrocyclic ligand coordinated to a
central atom capable of coordinating with two aacial lig-
ands which are coordinated to the central atom on either
s~,de Qf the macrocycli.c ligand.
Marker components used in the immunoassays of the
''
; ; ;,
;
~
present invention comprise a macrocyclic multidentate.
'ligand having two solubil~,zing polyoxyhydroearbyl moieties
one located'on either side of the plane-of the multiden-
ate ligand exhik~it no detectable non-specific binding to
serum components; and exhibit no detectable solvent sensi-
tivity. These marker components also exhibit enhanced
decay times which -approach their natural (fluorescent) ;
l i~et~;mes . .
51l 5'1'iTt~'~E S H E E'1'
.x'v Av,.
Wt T T' .:::l (-.
A~r a r .i".
.!, x..,.w J
,. tw . .,...~. .f........v .. . .... n ..... .. ,. ..... , ..~.,..n. a
,......
. , v,.. .u .....>. e. ..~. .~. ~.. . .r ... ..7 n ... .... . .,..
CA 02132708 2002-08-14
77036-13
23
Preferred are fluorophores which produce fluorescent
light efficiently, i.e., which are characterized by high
absorbitivity at the appropriate wavelength anal high
fluorescence quantum yields. For certain applications,
preferred fluorophores have measured fluorescence decay times
cn the order of at least 2 nanoseconds and exhibit a high
degree of fluorescence polarization.
Preferred solubilizing polyoxyhydrocarbyl moieties
include water soluble carbohydrates such as glucose, sucrose,
maltotriose, and the like; water soluble carbohydrate
derivatives such as gluconic acid and mannitol and oligo
saccharides and water soluble polymers such as
polyvinylpyrrolidone, poly(vinylalcohol), poly(ethylenimine),
polyacrylic acid, polyacrylamide, ethylene oxide copolymers
such as Pluronic* (a propylene oxide copolymer, available from
BASF) and Tetronic* (BASF) polyol surfactants; and polyethers,
including water soluble polyoxyalkylene polymers, particularly
polyethylene glycol) ("PEG") and polyethylene glycol)
derivatives such as polyethylene glycol.) methyl ether,
polyethylene glycol) silicon derived ethers and the like.
In one aspect, the present irxvention is directed to
immunoassays using marker components comprising a fluorophore
moiety which comprises a substantially planar, multidentate
macrocyclic ligand coordinated to a central atom capable of
coordinating with two axial ligands and two polyoxyhydrocarbyl
moieties which are attached as axial ligands to the central
atom. Suitable central atoms are those to which may
coordinate two axial ligands and which are not of high enough
atomic number to cause extensive fluorescence quenching by
transition to the triplet state. Preferred elements for the
central atom include silicon, germanium, phosphorus, and tin,
*Trade-mark
CA 02132708 2002-08-14
77036-13
23a
especially preferred are silicon and germanium.
Depending on the type of immunoassay, these marker
components may be used as labels for labe7_ling an analyte,
antigen, antibody or other molecule. These marker compo-
CA 02132708 2002-08-14
77036-13
24
nents may be optionally functionalized so as to include a
linker arm which allows the marker Component to be linked
to the analyte, antigen, antibody or other molecule. A
variety of linker arms which may be suited to this pur-
e pose. The marker component is linked to the analyte,
antigen, antibody or other molecule using conventional
techniques.
The present invention is also directed to the use of
divalent peptide derivatives as analogs for large mole
lo cules in immunoassays. Preferably, a divalent hapten
consisting of two epitopes of the same specificity con-
nected by a linker about 10 nm long is used to bind to a
single antibody molecule, requiring approximately 26
residues,
1~ The present invention also includes assay methods of
involving cellular receptors located on the. plasma mem-
brane or isolated from cytosols and synthetic ligand
binders obtained by molecular imprinting.
Accordingly, it is a principal object of this inven
20 lion to provide improved FIAs with greatly enhanced sen
sitivity. It is yet another object of this invention to
provide FIA methods which allow rapid and accurate deter
minations, often within a matter of minutes..
CA 02132708 2002-08-14
77036-13
The present invention also provides particular fluor-
escent probes for use in immunoassays, for instance, see
Examples 3 and 11-18 below.
Definitions:
As used herein, the following terms have the follow-
ing meanings unless expressly stated to the contrary:
The term "target analyte" refers to the compound or
compound to be measured in an assay which may be any
l0 compound for which a receptor naturally exists or can be
prepared which is mono- or polyepitopic, antigenic or
haptenic, a single or plurality of compounds which share
at least one common epitopic site or a receptor. By
"analog" of a target analyte is meant a compound or com
pounds capable of competing with the target. analyte for
15 binding to a receptor.
The texzn "axial ligand" refers to a substituent
which, together with a macrocyclic ligand, forms a
coordination complex with a central atom. The axial
ligand lies normal to the plane described by the macro
20 cyclic ligand.
The term "fluorescent probe" refers to a marker
component comprising a fluorophore moiety which is bonded
to or coordinates either directly or via a linker arm to
an analyte, antigen, hapten, antibody or other molecule
25 which is used in an assay, such as a fluoroi.mmunoassay to
determine the presence of and/or quantitate a substance of
interest.
The term "solvent sensitivity" refers to changes in
tre fluorescence behavior of a molecule deyending on the
WO 93/19365 ~ PC.'f/L'S93/0247~
26
solvent system in use, most notably referring to differ-
ences in fluorescence behavior in aqueous solution in
comparison with organic solvents (such as DMF). Many
fluorophores which exhibit high fluorescence intensity in
organic solvents such as DMF show substantially decreased
fluorescence intensity in aqueous solution.
Fluorescence intensity is related to sample concen-
tration and the intensity of the exciting radiation. The
fluorescence intensity of a particular dye can be corre-
laced to its characteristic light absorptivity (extinction
coefficient) axed fluorescence quantum efficiency, as well
- as environmental factors.
The term "specific binding pair" refers to two dif
ferent molecules (or compositions) wherein one of~ the
molecules has an area on the surface or in a cavity which
specifically recognizes and binds to a particular spatial
and polar organization of the other molecule or molecular
complex involving other molecules.
The term "binding partner" refers to a molecule or
molecular complex which is capable of specifically recog
nizing or being recognized by a particular molecule or
molecular complex.
The term "bound" refers to the condition in which a
binding interaction has been formed between a molecule and
its specific binding partner.
The term "decay time" is the time which must elapse
in order for the concentration of excited molecules to
decrease from its initial concentration to 1/e of that
value .
The term °'receptor'° refers to a molecule or molecular
complex which is capable of specifically recognizing or
being recognized by a target analyte or analog thereof.
Brief Description of the Drawings
Fig. ~. depicts an HPLC analysis of crude caged dicar-
boxy silicon phthalocyanine dve preparation.
SUSTE SHEIE~"
CA 02132708 2002-08-14
77036-13
27
Fig. 2 shows the absorbance of caged dicarboxy sili-
con phthalocyanine dye in various so:Lverxts.
Fig. 3 describes the Jiatrori Analog Sy~~tem.
Fig. 4 depicts the decay time for caged dicarboxy
silicon phthalocyanine dye.
Fig. 5 shows serum interactions of purified caged
dicarboxy silicon phtha:Locyanine dye.
Fig. 6 depicts the absorbance spectrum of caged
dicarboxy silicon phthalocyanine dye-C12 linker.
Fig. 7 depicts the polarization of caged dicarboxy
silicon phthalocyanine dye-C12 linker at 680 nm.
Fig. 8 depicts the polarization of caged dicarboxy
silicon phthalocyanine dye-C12 linker at 69() nm.
Fig. 9 depicts an HPLC Chromatograph of: caged dicar-
boxy silicon phthalocyanine digoxin probe.
Fig. 10 depicts the structure of caged dicarboxy
silicon phthalocyanine digoxin probe.
Fig. 11 shows the absorbance spectrum of caged
dicarboxy silicon phthalocyanine digoxi.n probe in
2.0 methanol.
Fig. 12 shows the absorbance spectrum of caged
dicarboxy silicon phthalocyanine digoxin probe in FPIA
buffer.
Fig. 13 shows the decay time for caged dicarboxy
silicon phthalocyanine digoxin probe.
Fig. 14 shows the linearity of intensity for caged
dicarboxy silicon phthalocyanine digoxin probe.
Fig. 15 shows serum/urine interactions fox caged
dicarboxy silicon phthalocyanine digoxin probe.
:30 Fig. 16 depicts a comparison of TDx'~ and FAST-60
calibration curves.
Fig. 17 depicts the correlation of digoxin samples
assayed by TDx° and FAST-60.
Fig. 18 depicts the effect of dilution jump on non-
specific binding.
Fig. 19 depicts a digoxi:~ probe-serum calibration
curve .
*Trade-mark
CVO 93/y9366 PCT/L~S93/02470
28,
Fig. 20 depicts a calibration curve for a high sensi-
tivity digoxin assay.
Fig. 21 describes the FAST-60 digoxin assay
procedure.
Fig. 22 depicts digoxin correlation - TDx~ Serum vs.
FAST-60 Whole Blood.
Fig. 23 depicts digoxin correlation - Stratus° Serum
vs. FAST-60 Whole Blood.
Fig. 24 depicts digoxin correlation - FAST-60 Serum
vs. FAST-60 Whole Blood:
Fig. 25 depicts digoxin correlation - TDx° Serum vs.
FAST--60 Serum.
Fig. 26 depicts digoxin correlation - Stratus° Serum
vs. FAST-60 Serum.
Fig. 27 depicts the structure of caged dicarboxy
silicon phthalocyanine digitoxin probe.
Fig. 28 depicts the structure of caged dicarboxy
silicon phthalocyanine theoghylline probe:
Fig: 29 depicts the structure of caged dicarboxy
silicon phthalocyanine phenabarbital probe.
Fig. 30 depicts the structure of caged dicarboxy
silicon phthalocyanine thyroxine probe.
Fig. 3l dep~.ets the structure of caged dicarboxy
silicon phthalocyanine n-acetylpr~ocainamide probe.
Fig. 32 dep~.cts the structure of caged dicarboxy
silicon phthalocyanine primidone probe.
Fig. 33 depicts the structure of caged dicarboxy .
;,,, sihicon phthalocyanine phenytoin probe.
Fig. 34 dep~:cts a rubella antibody calibration curve . .
' for a sandwich assay.
Fig. 35;depicts a rubella peptide calibration curve ,
for an inhibition assay: ;
Fig. 36 depicts a rubella antibody calibration curve
for direct polarization.
S tJ ~ 5'tITF 5 ht E ~°3'
1~V0 93/19366 . w P('T/U~93/02470
29
Detailed Description of the Invention
The present invention provides fluorescence immuno-
assay methods which have dramatic increases in sensitivity
over previous methods, which can be easily performed "
because they require no separation step, and which can be
used to detect and quantitate low levels of target analyte
in biological samples such as serum, plasma, whole blood
and urine. The FTAs of the present invention may be per-
formed in small samples. For example, a digoxin assay may
be performed on a 20 ~C1 sample of serum, plasma or whole
blood, and the assay may be performed in about five min-
utes. The ability to perform FIAs on whole blood samples
is particularly significant because it allows assays t.o be .
performed at locations closer to the patient, such as
~.5 physicians' offices and emergenc~r rooms. The capability
of performing FIA.s rapidly is important because, in a
clinical context; patient survival can depend on accurate,
timely results.
The concept of sensita.vity in fluorescence measure-
menu can be usefully quantified by specifying the con-
centration of tie fluoro~hore in question at which the
fluorescence intensity fram the fluorophore is equal to
the intensity from the background. This manner o.f
expressing sensitivity emphasizes the fact that the
ser~sxtivity of f7.uorescence measurements is almost always
determined by the ability to discriminate between "signial"
and "background" and not by the absolute number of photons
av~ilable,from the "signal:" ;
The present invention provides methods for FIAs which ,
solve the problem with discriminating against background ::
radiation on the basis of wavelength. The probes used in
the method of the invention have excitation (and emission?
wavelengths greater than about 650'nm, preferably greater
than 680 nm. This wavelength shift into the infra-red
'35 range decreases background fluorescence, i.a., increases
si.:~nal-to-background ratio. This decrease in background
5l! ~ S'TiTI~'TE 5 H E E'~'
r ...~ :, :...; ..... ~ . ... ,.. .. . .:,.. . : . . :: ;.
v.". . ,
,,
. ,.. .
..,:, .:.::..;. .,..:.' '.:x;;,,:.. ,:~,::... . , ~...;;.~. :".i ;.'~.
.. ,:':' :..._., . ,..,:, ;i.,.,-.,.; .,:._~,~ , ;~;... :'.'. .. .,.,,.
Y'
> Fr..e .,.... .. ..,..,..:..., ....'.::1.. '...,-._._..,,:..
.,:::...,..,...,v...,.,s'.,:
~........:1J..:~..w.V,.,..;':'.. v...~.::. ::..., . ,.., ,.... ~... .
.."..,.s..,... ....,.. .
: .. ~.,... ....
Vd~ 931~936b PCT/LJS93/02470 _.
6~~.~G~~ ~~
fluorescence allows the use of fluorophore at far lower
concentrations than previously used in homogeneous FIAs.
The probes used in the present assay methods have low
dielectric constants, which applicants believe tend to '
S increase the Van der Waal interactions and hydrogen bond-
ing, thus accelerating the antigen/antibody reaction. In,
addition, applicants believe .that these probes compete for
the water of hydration, thus potentiating the antigen/
antibody reaction. In other words, the probes not only
10 substantially decrease non-specific bindings to serum
components, but applicants believe that they potentiate
- the immunochemical reaction.
For fluorescence polarization assays, the present
invention provides a further increase in sensitivity by
15 measuring transient state fluorescence rather than the
steady state signal. In the steady, state mode the signal
is constant over time enabling the determination of one
experimental~parameter, e.a., the polarization or the
anisotropy, both of which are related to molecular
20 rotational motion. In the transient state made the
signals vary in a systematic way with time. This varia-
tion represents a complex summation of the rates of decay
arid of wolecular rotation as it changes from moment to
moment in time.
25 The incxease in sensitivity from transient. state
measurements stems from two sources. First, that portion
of the background due to Rayleigh and Rarnan scattering
disappears, in ,about 10'15 sec and, so is clean~.y removed
before the transient state measurements start. This ,
30 portion of background is normally an important part of the
y-
total in steady state measurements. ,
Above and beyond the removal of scattering, the tran-
sient state measurements provide an additional powerful
means to discriminate between the desired signal and the
remaining fluorescence portion of the background. This
discrimination rests upon the time dependence cf the
polarized components in the fluorescence decay and makes
SU~S'T'~'UTE SKEET'
t-
CA 02132708 2002-08-14
'?7Q36-13
31
it possible to extract the desired signal only, simultane-
ously on the basis of the rate of decay of the excited
state and the rate of decay of the rotational distribution
imprinted by the excitation. Thus, transient state
methods allow signal to be distinguished from background
in ways not possible with steady state information alone.
In addition to these features, the probes exhibit a
high degree of polarization, necessary for mix and read
(homogeneous) fluorescence polarization assays. This
l0 increase in polarization translates into increased
sensitivity.
The methods of the present invention are particularly
useful when used with a time-correlated transient state
detection system, as described in commonly assigned
Studholme, et al., U.S. Patent No, 5,323,008 entitled
"Fluorometer Detection System °.
That system
features transient detection along wit.z detection of the
time-dependent polarization of the sample. The system
uses a laser diode which can be modulated at very high
frequencies, e-a., 14 MHz rats, and exhibits high output
power. Typically the laser "on" time is approximately 2-3
nanoseconds. Photons from the solution are detected using
a photomultiplier tube (PMT) operating in a single photon
counting mode. The photon event along with the relative
time of the photon event as compared with the laser pulse
time is determined. By storing the individual photon
event times a histogram of frequency of photons as a
function of time is venerated.
Data obtained in this manner can be analyzed as
described in Dandliker ~ ate", U.S. Patent No. 4,877,965,
entitled "Fluorometer" or as described by Studholme,
et al . , U. S . Patent 1'"10. 5,:323,008 entitled, "Fluorometer
Detection System ".
The methods of the present invention also include the
use of divalent peptide derivatives as analogs for large
Wl'~ 93/19366 ' PCT/L'S93/02470
32
molecules in immunoassays. Both polyclonal and monoclonal
antibody molecules are divalent. Due to the "chelate
effect," the binding of a low molecular weight mimic or
analog of a larger molecule will be stronger arid dissoci-
ation from the antibody will be slower if both antibody .
binding sites are utilised in the bonding. Tt is within
the scope of the present invention to arrange the struc-
ture of the peptide analog to have twa identical sequences
joined together by a linker of suitable length so as to
place the two peptide sequences, in their normal config-
uration in solution, in the most favorable position for
reaction with the two sites on the same antibody molecule .
Preferably, several such unspecific, divalent analogs
are used as a cocktail; rather than combining more~than
one epitope in the same analog molecule. The latter
arrangement would permit cross link~.ng of perhaps many
an;t~.body molecules which might be preferable in solid
phase assays,in which the formation of chains and cycles
could aid in adhering to a surface. Conversely, having
two identical epitop~s on the same analog molecule may
"inhibit" polymerization by strongly favoring, by prox-
imity factors, reaction with two sites on the same
molecule.
The use of divalent peptide derivatives as analogs
for large molecules in immunoassays is preferred in
solution, especially in conjunction with dilution jump,
due to the tighter binding afforded by the "chelate
a f fect, '! resin ing in an increase, in the sens~.tivity of
r ,
the immunoassay. ,
Preferably, a divalent hapten consisting of two
epitoges of the same specificity connected by a linker
about 20 nm long is used to bind to a single antibody
molecule. Taking into account the bond distances and
angles far simple peptides (L. Paining, The Nature of the
Chemical Bond, Cornell University Press (1960), p. 498)
and assuming a length of 0.380 nm per amino acid residue,
s c~ s°rrr~~rF s ~ E ~r
WO 93/ ~ 9366 ' PCT i'S93l02470
~,~x_~~'~~~~
33
this would require approximately 26 residues for a ~0 nm
i
length.
One approach for designing a divalent hapten with ,
such a linker is to synthesize the epitope with a 13 resi-
due linker terminating in a primary amino group. This
peptide is then reacted with the bis (3-isocyanatopropyl-
dimethylsilyl) derivative of dihydroxysilicon phthalocya-
nine. The resulting structure has the phthalocyaninine
moiety with two axial substituents, one on either side of
0 the molecular plane, each consisting of a thirteen residue
peptide linker leading to the peptide epitope. The mole-
cular plane of the dye moiety is perpendicular to the
direction of the linker. After combination of the two
peptide epitopes with the two binding sites of the 'same
~.5 antibody molecule th.e dye moiety may be located midway
between the two arms of the Y-shaped antibody molecule.
The polarization changes obtainable with this type of
structure may, not be as; great as yf the dye were linked
through a peripheral rather than an .axial bond, and the
20 absence of PEG may result in non-sped fic binding. How-
ever, if the dye moiety is held close to the antibody
i-
surface between the two Fab fragments after binding, it
may prove to be quite protected and rotate with the long-
est rotational decay 'time of the antibody (since the
2S molecular plane of the dye may lie parallel to the long .
axis of the antibody).
Alternatively, a divalent peptide hapten may be
designed to utilize the PEG protected dye linked through
i ,, . ; , '
a peripheral carboxyl to an amino group~on the~link~er'o
30 oz~; one of the peptide epitop~s, ~, the linkage could be
the e-amino group of a lysine residue located approxi-
mately midway on an interconnecting chain between the two
peptide epi~opes.
Immunoassays, a class of ligand binding assays,
35 depend upon the strong and-selective binding of some
analyt~ of interest to antibody specific for that analyte .
Other molecular structures that have similar strong and
SL!~S T'E S~1EE°~'
WQ 93/9366 -- ~CT/US93/~D2470
6 ~ i~ 6~ t''~ ~.~ ,
34
selective binding for such an analyze can serve equally
well in designing an assay and such structures may have
some inherent advantages over antibody. Far example, ,
molecules which may have desirable properties in this
S context include cellular receptors located on the plasma ,
membrane or isolated from cytosols arid synthetic ligand
binders obtained by a process known as "molecular
imprinting."
The sensitivity of a ligand binding assay depends
upon the binding of f inity of association constant f K) of
the reaction between the analyte and the binding molecule .
For classical cytosolic steroid hormone receptors these Ks
are of the order of 109 M'~. In recent work with molecular
imprinting of synthetic polymers the binding constant for
1S -diazepam was found to be about 108 M'1 (Vlatakis et al . ,
Nature 361: 64S--647 (1993)). By contrast, the highest Ks
for antibody binding are of the order of 1012 M'1 for lig-
ands such as f~.uorescein and digo~in.
The magnitude of these Ks suggest that antibodies
bind far more tightly than do receptors or molecular
imprints. Binding processes are symmetrical and the ..
1,.
obsererad K depends upon both the "receptor" and the
"ligand" and the distinction between the two is made for
convenience. Applicants believe that because relatively
few Ks have been measured for receptars or molecular
imprinbs, there is no reason that the binding by these
molecu~.es should not be potentially as tight as by anti-
body,. Molecular, imprints also have the inherent advantage
lof being tailored for one specific molecule and the Ks can .
be improved by he'proper placement of hydrophobic, polar
hand ionic groups in the binding sites. Moreover, because
these molecules are synthetic, once the optimal structure
i.s.known large amounts should be readily obtainable.
Thus, the present invention includes methods of
involving cellular receptors located on the plasma mem
brave or isolated from cytosols and synthetic ligand
binders obtained by molecular imprinting.
S LI ~ ST't°TLJTE S #~ E ET
CA 02132708 2002-08-14
77036-13
I. Preferred Marker Components
The following is a brief description of the preferred
marker components to be used in the fluorescence immuno-
assays of the present invention. A more complete discus-
s sion is found in commonly assigned U.S. Patent No.
5,403,928.
A. P~r~~rred F,~~,uorophore Moi ties
Suitable fluorophore moieties comprise a luminescent
10 substantially planar molecular structure. Preferred are
fluorophore moieties in which the luminescent substan
tially planar molecular structure comprises a substan
tially planar macrocyclic multidentate ligand which
coordinates a central atom which may coordinate with two
15 axial ligands, one on either side of the: macrocyclic
ligand (~ having a trans orientation).
Preferred central atoms are elements which may form
octahedral coordination complexes containing two ligands
with a trans or axial orientation, on either side and
20 perpendicular to the planar macrocyclic ligand. For use
as fluorescent marker components in certain applications
the central atom should not have toa high atomic number
(about 30 or less) so that fluorescence is not diminished
through coupling interaction with orbitals of the central
25 atom.
Preferred multidentate ligands include nitrogen-
containing macrocycles which have conjugated ring systems
with pi-electrons. These macrocycles may be optionally
substituted, including substitution on bridging carbons or
30 on nitrogens. Suitable macrocycles include derivatives of
porphyrins, azaporphyrins, corrins, sapphyrins and por-
phycenes and other like macrocycles which contain elec-
trons which are extensively delocalized. In view of the
fact that they incorporate many of the above-noted char-
3~ acteristics, an especially preferred class of macrocycles
comprise porphyrin derivatives, and azaporphyrin deriva-
CA 02132708 2002-08-14
7'7036-13
36
tives (porphyrin derivatives wherein at least one of the
bridging carbons is replaced by a nitrogen atom). Azapor-
phyrin derivatives include derivatives of mono-, di- and
triazaporphyrin and porphyrazine. These rraacrocycles may
S optionally have fused aromatic rings. These azaporphyrin
derivatives include phthalocyanine, benzotriazaporphyrin
and naphthalocyanine and their derivatives. The prepara-
tion and fluorescent qualities of many of these compounds
are known and some are available commercially.
For certain applications, such as fluorescence polar-
ization assays, preferred are azaporphyrin derivatives
which exhibit a high degree of polarization, that is,
those which emit strongly polarized light. For these
applications, preferred are macrocycles having lower
degrees of symmetry, preferably having lower symmetry than
D",. One preferred group includes macrocyc:les having at
least one fused aromatic ring. Thus, preferred macro-
cycles include azapvrphyrin derivatives having fused
aromatic rings at positions which result in decreased
symmetry. Preferred classes of azaporphyrin derivatives
comprise derivatives of monoaxaporphyrin, d:iazaporphyrin,
and triazaporphyrin having lower than D~" symmetry,
B. P ed ub' o o a 1 of t'
Preferred solubilizing polyoxyhydrocarbyl moieties
include water soluble carbohydrates such as glucose,
sucrose, maltotriose and the like; water soluble carbo-
hydrate derivatives such as gluconic acid and mannitol,
and oligosaccharides; polypeptides such as polylysin and
naturally occurring proteins; and water soluble polymers
such as polyvinylpyrrolidone, polyvinyl alcohol), poly
(ethylenimine), polyacrylic acid, polyacrylamide, ethylene
oxide copolymers such as Pluronic~'T' (a polyether) and
Tetronicl" (BASF) polyol surfactants and, in particular,
polyethers such as other polyoxyalkylenes including poly
W~ 93/i9366 - PC.'C/US93/02470 .
,y:~YW ra
37
(ethylene glycol), or other water soluble mixed oxyalky-
lene polymers, and the like.
A particularly preferred class of solubilizing poly
oxyhydrocarbyl moietie s comprises polyethylene glycol)
S (PEG) and polyethylene glycol) derivatives, such as
polyethylene glycol) monomethyl ether. Other suitable
PEG derivatives include PEG-silicon derived ethers. Many
of these polymers are commercially available in a variety
of molecular weights. Others may be conveniently prepared
~.0 from commercially available materials, such as by coupling
of an amino-PEG to a haloalkyl silyl ar si,lane moiety.
When linked to a fluorophore moiety, these polyoxyhydro-
carbyl moieties imparf. particularly advantageous quali,ti~s
of solubility in aqueous solution with improved measured
15 fluorescence decay time, and improved fluorescence inten-
sity. Moxeover, the resulting marker components are water
so7.uble and show decreased non-specific binding. especi-
ally decreased binding tc ~exum albumin which has here-
tofore been a problem with certain fluorophores, parts--
20 cularly porphyxin or phtha~ocyanine dyes which have been
functioa~alized with groups such as sulfonate to imP~rt
increased water solubility to the molecule. Non-specifis
winding of fluorophore or marker component impairs the
accuracy of the resulting; immunoassay. These marker
25 components which comprise fluorophore linked to PEG may
also exhibit imprcwed fluorescence intensity in aqueous
solution with decreased quenching.
Suitable PEGs may vary in molecular; weight , from about
art~.culari mole~_'
200 to about 20,000 or more. Choice of a p
3.0 cular weight may depend on the particular fluorophore
chosen and its molecular weight'and degree of hydropho
lication for which
bicityas well as the particular app
the f~.uorophore-PEG complex is to be used.
SUSS 'TF SHEE'T
WO 93/19366
PCT/US93/02470 . .,,n.;.. :.
,...
38
C. Absorbance and Polarization Behavior of Preferred
Marker Components
These marker components which comprise a central atom ,
(for example, Silicon) coupled to two PEG moieties may be
characterized by measurements of transient state fluores- ,
cence. ~n such measurements the intensity of the two
components polarized either parallel or perpendicular to
the direction of polarization of the exciting pulse is
monitored over a time periAd equal to about 3 times the
ZO decay time of the marker component. Such curves reflect
extinction coefficient, quantum yield, decay time and
state of polarization and supply sensitive indications on
the chemical and physical condition of the marker
component.
~.5 For example, if the excited state is being deacti-
vated or converted to the triplet state the overall
intensities are lowered and the decay times shortened. If
the rotary brownian' ~rotian of the molecule is being
a~aered by an increase in viscosity or by being bound to
20 a large molecule; the ratio of the intensity of the parcel-
lel to the perpendicular component is increased.
Some marker components according to the present
invention show, within experimental error of about 5%, the
same intensities, decay time and polarization in DMF (an
25 organic solvent) as in ~.Ap (saline azide phosphate, an
aqueous neutral buffer). To'some extent these properties
are shared by other marker component preparations. A
distinctive and important property of the marker comps= r'
vents' of the'' present invention is ainsensitivity to (and
30 lack of binding to) the components in serum which is evi-
de~ced by a lack of shy measured effect of serum on the
intensities, decay time ar relative magnitudes of the
polarized components of the fluorescence. This property
is crucial for the marker components to be useful for
35 applications such as assays using biological materials.
S Ll ~ S~tT'LITE S H E E?' '
WO 93/19366 ' PCT/LrS93/02470
f
.. .,
39
ents
D, Pre aration of Preferred Marker Cornnon
According to one method of preparing the 'preferred
marker components of the present invention, the appropri-
ate fluorophore moiety having hydroxy or halide groups as
axial ligands is reacted with a reactive form of the solu-
bilizing polyoxyhydrocarbyl moiety in a ligand exchange
reaction according to the general reaction scheme:
Mcl-CA- (X) 2 + 2 (SM) ~ Mcl-CA- (SM) 2 + 2X
whexein Mcl denotes the macrocyclic ligand, CA the central
1C1 atom, X the displaced ligand and SM the solubilizing
moiety. This reaction may be carried out neat or, if
desired, in solvent. Suitable solvents include quinoline,
THF, DMF, imidazole and the like. Suitable reaction tem
peratures may vary, depending on the nature of the macro
~,5 cyclic starting material and the solubilizing group. The
reaction is generally complete in about 2 minutes to about
z~ hours: The r~actian mixture can be conveniently heated
under reflex or by mans 'such as a sand bath. For eon
venience, the reaction may be carried out at ambient
20 pressure:
It is believed that this reaction takes place in two
steps, with one polyoxyhydrocarbyl group coordinating as
an axial ligand at a tine.
y~hen used as fluorescent labels in fluorescence
2~ immunoassays, these marker components may be linked to one
member of a specific -Mnding pair ( "labelled binding part
ner") or an anal~g of such a member. The marker component
may be directly attached or conjugated thereto or attached
or conjugated via a linker arm.
30 II. Caged DicarbeY_y Silicon Phthalocyanine Dye
Example 1
Preparation of a Caned Dicarboxy Sil~.co~
_'-' phthalocyanine D~~e
unless otherwise stated, all chemicals used in the
3~ synthesis of ghthalocyanine derivatives were purchased
from Aldrich Chemical Co., Milwaukee, WI. Amino
SUSTE S~iE~T
CA 02132708 2002-08-14
;'7036-13
terminated polyethylene glycol and phthaloc:yanine deriva-
tives were synthesized according to published procedures.
See, e.a., Reference 18 of U.S. Patent No. 5,403,928.
5 A. Preparation of Diimino~so~.nd ine
In a three-neck, 100 ml round-bottom flask fitted
with a reflux condenser and a gas inlet tube was placed
phthalonitrile (12.8 g), and methanol (50 ml), and the
mixture was stirred while ammonia gas Was slowly intro-
l0 duced. In order to prevent the possible flow of the
reaction mixture into the ammonia source, and in-line trap
was employed. After the reaction mixture appeared to be
saturated with ammonia, 0.33 g of dry potassium tertbut-
oxide was added with stirring.
15 Stirring was continued and the reaction mixture was
heated to reflux for three hours with continued introduc-
tion of ammonia. Care was taken to avoid fouling of the
gas inlet with the crystalline product. At the end of the
reflux period a pale green solid had farmed. The solid
20 was collected by filtration and washed with a small volume
of cold (4°C) methanol. (This compound is appreciably
soluble in methanol.) This material was dried and used
for the next step without further purification. Yield was
'7 g (about 50%) .
25 B. re a ati i
1,2,4,5-Tetracyanobenzene (Pfaltz & Bauer, 0.5 g, 2,8
mMol ) was suspended in methanol ( 10 ml ) in a three-neck
round-bottom flask fitted with a reflux condenser and a
gas inlet tube. The mixture was stirred at 25°C without
30 external cooling while ammonia gas was rapidly introduced.
During the first two minutes of ammonia introduction the
temperature of the reaction mixture rose to greater than
50°C and the suspended solid dissolved r_ompl.etely. Within
5 minutes a precip.itate~began to form. Stirring at 40-
35 50°C with slow introduction of ammonia was continued for
1~0 93/19366 ' P~ I'/i;'S93/02470
r
. . ~., ~s~s~~~ o~
41
1 hour. The precipitated solid was collected by filtra-
Lion, washed with methanol, and dried. This product
exhibited a very low solubility in methanol.
C. Pre aration of Dic anosilicon hthaloc arsine
Dichloride (Compound I)
In a dzy 50 ml round-bottom f lack was placed dicyano-
diiminoisoindoline (350 mg, 1.8 mMol) along with diimino-
isoindoline (1.0 g, 5.9 mMol) and quinoline (Fluka, 20
ml) . The miaeture was stirred at 25°C while silicon tetra-
~.O chloride (Aldrich, 2.0 ml, 18 mMo1) was added dropwise
over a period of 1 minute. The f Task was then f fitted with
a reflux condenser (using teflon tape) and a calcium
chloride drying tube and stirred for one minute at 25°C.
At this time the reaction flask was lowered into a
large oil bath maintained at 180--185°C and efficient mag
netic stirring was continued for 30 minutes. The oil bath
was then remcwed and the contents of the flask were
allowed to cool to room temperature.
The dark reaction mixture was carefully treated with
water (5 m1) and then diluted with 45 ml of a 30% HC1
solution. The resulting dark precipitate was collected by
filtration on a ~.0 cm Buchner funnel. 'Washing with water
and then acetone left a blue solid (1 gram) which was air
dried and used without further purification for the next
2~ 'reaction step.
D ~ H drol sis of Dic anosilicon hthaloc arsine Dichloride
,. , , , ,
,~Compo~:nd I I )
The crude dicyanophthalocyanine from step (C)
( 1 grim) was placed in a f bask with a stir bar and 6 ml of
concentrated sulfuric acid and stirred at 50°C overnight.
The mixture was then carefully diluted with 4 ml water and
heated to 100°C for an additional 20 hours. Cooling and
dilution with water (20 ml) gave a blue precipitate which
was collected by ~ilt~ration and washed with water . The
soi.id was then transferred to a flask along with a stir
StI~S "~F S~IEF'T
CA 02132708 2002-08-14
77036-13
42
bar and 20 ml of a 1.0 M potassium carbonate solution and
stirred and heated at reflex for one hour. The suspension
was then slowly and carefully acidified with concentrated
HC1 and then filtered and the resulting solid was washed
with water and acetone and dried in a desiccator. This
material (0.7 g) was used without further purification in
the next step.
E. Preparation of 2~,3-Dicarboxvc~htha~5~ryani~ato-bis- (3-
~F~f-imidazol-1-Ylparbonyl? am~,n,opropvl-~dime,~hvlsil-
anQlatol silicon (Comgound IT,~~,
The crude dicarboxy silicon phthalocyani.ne dihydrox-
ide from step (D) (85 mg) was placed in a vial along with
a stir bar and imidazole (160 mg, ~2.3 mMol) and 1 ml of
dry DMF. The mixture was stirred for 5 minutes at 25°C
and then 3-isocyanatopropyldimethylchlorosilane (Petrarch,
110 ~cl, 0.68 mMol) was added to the stirred mixture over
a period of 0.5 minutes. The vial was capped in order to
exclude moisture and stirring at 25°C was continued for
20-40 hours. (A 40 hour reaction time appeared to result
in an improved yield.) The vial was then opened and the
dark blue mixture was diluted with methanol (4 ml) and
filtered through #545 Celite*to remove solids. The fil-
trate was concentrated on a rotovap using high vacuum and
a water bath maintained at 40°C. The dark residue was
2S then slurried with silica gel (1-3 g) and methanol (5 ml)
and the methanol was removed on a rotovap under aspirator
pressure. The blue residue was then suspended in toluene
and transferred to a silica gel column prepared from 15 ml
23-400 mesh silica gel (EM Science) and toluene. This
column had been washed with 50~r methanol in toluene.
Increasing the solvent polarity by increasing the
methanol content of the solvent to io% brought about the
migration of a distinct band which was collected. This
material was saved but not used for further
transformations.
*Trade-mark
PCTlUS93102470
...: i1r~ 93! 19366
~' ~~'~i~~
~ .. .
a3
t
Increasing the solvent polarity by slowly increasing
the methanol content of the eluant to 30% brought about .
the migration of a second blue band which. was collected
within a 20 ml volume of 30% methanol. This material was
transferred to a round bottom flask. Removal of solvent
on a rotovap under high vacuum at 25°C left a residue
which appeared to include an appreciable quantity of imi
dazole along with the blue dye. This material was used
without further purif ication for the next step . The yield
of compound III was approximately 3 mg.
g, Pre aration of Amine-Terminated Pol eth lane G1 col
Polyethylene glycol) monomethyl ether (Aldrich,
average M.W. 2000, 10 g, 5 mMol) was placed in a 100 ml
round-bottom flask along with a stir bar and 55 ml
toluene. The flask was fitted with a short-path distil
lation apparatus and immersed in a heating bath. Toluene
was slowly distilled at 760 mm Hg until the distillate was
no longer cloudy. This required the removal of about
15 ml of toluene.
The relatively water-free PEG solution was allowed to
coal to 40°C: When this temperature had been attained,
carbonyldiimidazole (Aldrich, 1.2 g, 7.5 mMol) was added
to the stirred solution in one portion. Stirring at 30
40°C was continued overnight with protection from atmos
ph~ric moistuz°e.
Water (100 ~sl, 3.75 mMol) was then added to the rear-
,. tion mixture and efficient magnetic stirring was, continued,
until the evolution of COZ gas could no longer be observed
(about 15 minutes). .
Most of the toluene was removed on the rotovap at a
30°C under high vacuum leaving a viscous, colorless oil.
This material was diluted with isopropanol (20 ml) and
added to a stirred solution of 1,2 ethylenediamine (Fluka,
6.7 ml, 100 mNi~l) in isopropanol (15 ml) over a period of
five minutes. After completion of the addition the clear
solution was maintained at 40pC for four hours.
S9J~ST'd't'LJTE SiiE
WtD 93/19366 ' PCT/L'S93/02470 :,~'::.
44
At this time isopropanol (150 ml) was added to the
reaction mixture. The diluted solution was allowed to
stand at 4°C overnight, resulting in the formation of a
voluminous mass of white crystals. This solid was col-
S lected on a 10 cm Buchner funnel, and subsequently
recrystallized from isopropanol.
Drying under high vacuum over sulfuric acid afforded
7 grams of the crude amine, suitable for use as a reagent.
Stricture of the product was confirmed by IR.
ZO The amine content of polyethyleneglycol amine, pre
pared as outlined above, was determined to be > 70 moleo
- by the following method:
25 ml of ~.Oo solution of the amine in methanol was
allowed to react with an equal volume of a 6% solution of
1S malefic anh~rdride in THF: The reaction mixture was allowed
to stand for 0.5 hours at 25°C and was then diluted to 1..0
ml, with methanol. A S ~l aliquot of this final solution
was injected;on to an analytical RP18 reverse phase HPLC
column using 30% methanol in water as the initial mobile
~0 phase. Using n-propylamine as an internal standard
allowed for accurate-quantification of the UV-absorbing
ac~rl-PEG derivative; which was eluted in 80% methanol and
was detected at 254 nm:
Analysis of the infrared spectrum of amine-terminated
25 PEG can also provide a-convenient means of estimating the
product yield.
G. Reaction of Compound III with Amine-Terminated Poly-;,
eth~rl~ne Glycol ( Compound IZT)
The product of step (E) (Compound IIT) (3 mg, 5 x 10'3
30 mi~iol) , which had been obtained in partially purified form
by chromatography on silica gel, was dissolved in methanol
(1 ml): The mzxture was stirred while amine-terminated
PEG (product of step (F) . 100 mg, 5x10-'2 mMol) was added.
Th:e resulting deep blue solution was heated to reylux for
35 one hour.
S ~J ~ S'Ti°CCJT'E S H E E~'°
CA 02132708 2002-08-14
77036-13
Removal of methanol under aspirator pressure at 25°C
left a viscous blue ail which was taken up in water (0.5
ml) and applied to a small (10 ml wet volume) DEAE Sepha-
dex'~anion ion exchange column (Pharmacia, 3.5 meq/g, 40-
'~ 120 micron, basic form < 1M K~CO~) . The water-soluble
blue dye was retained quantitatively by the column. The
column was washed with water (15 ml) and the blue dye was
then eluted in greater than 70% yield with :10-20 ml of a
15% aqueous acetic acid solution.
10 Water and acetic acid were rernaved under high vacuum
and the blue residue was taken up in a small volume of
methanol and applied to a C18 reverse phase semi-prepara-
tive HPLC column. The major product., detected at 675 nm
as a single peak, eluted with 80% aqueous methanol (con-
15 taining 0.6% acetic acid) and comprised about 50% of the
sum of the material which was recovered from the column.
Fractions containing the major product were combined and
solvent was removed under high vacuum leavinc3 a blue resi-
due ( app rox . 0 . 5 mg , 10 ~' mMo 1 ) .
20 NMR (DCCl~: 8 -2.85 (5, 12H), b -2.29 (m, 4H), b -1.30
(m, 4H), b 1.80 (m, 4H), ~ 3.6 (br.s, 300-400H), b 8.39
(m, 6H), b 9.68 (m, 6H), b 10.56 (S, 2H). dote: Because
the sample had been previously dissolved in DzO, the acidic
protons, RCOOH, were not observed.
25 Fig. 1 is a typical HPLC chromatogram of this prepar-
ation (compound IV). The fraction containing compound IV,
with retention time of approximately 25-26 minutes, was
designated "B" fraction. The yields for a typical dye
preparation range from 25-65% of this fraction. Fractions
30 from several chromatographic runs were pooled, dried in
vacuo and analyzed. The absorbance of the "B" fraction
was measured in a Perkin-Elmer* spectrophotometer using
various solvents. As shown in Fig. 2, vex-y little "sol-
vent sensitivity" can be seen between methanol, dimethyl-
35 formamide and FPIA buffer (100 nM phosphate pH 7.5 with
0.01% gamma globulin).
*Trade-mark
WO 93!1'9366 ' PCT/US93/02470
J.
46
The fluorescence decay time for the "B" fraction was
determined to be 4.3 nanoseconds. The measurements were
made on the "Diatron Analog System". In the Diatron '
Analog System, transient-state fluorescence was detected
using a high speed, "gateable" photo-multiplier tube
(PMT). The combination of being able to rapidly change
the PMT gain and the use of high power laser pulses
enabled the viewing of the fluorescence decay of dyes with
a single excitation pulse. In practice, many pulses were
averaged to obtain improved data. These analog signals
coming from the PMT were captured by a digitizer which
took the analog signal and cut it into 512 time bins for
analysis. r
The Diatr~n Analog System is diagramed in Fig. 3.
The tunable dye laser used was a PTI model PL2300 nitrogen
laser with a dye laser module. By changing the ,laser dye
and adjusting the dye laser grating, 600 picosecond pulses
with peak power of near 40 KWatts could be generated at
wavelengths from 340 to 900 nm.
A beam splitter was used to send a portion of each.
pulse to a pulse detector which consists of a high speed
Hamamatsu photo diode. The resulting output of the photo-
diode was fed into a pulse shaper which converted the
resulting 800 picosecond (ps) pulse into a 100 nanosecond
2S (ns) purse. This 100 ns pulse was then used as a gate for
the Hamamatsu microchannel plate PMT whose gain was
changed by 10,000 within a 2 ns time period. The PMT
stayed at the high gain until the 100 ns was over. ;
The dye laser module, reaction cell and pulse ,
detector was positioned and connected such that the PMT
was dated to its high ~ensitivit~r state approximately 2 ns ,
after the laser pulse passed thrc~ixgh the reaction cell.
A filter was positioned in, front of the PMT to guard
against high scatter signals when required. A lens was
used to image the fluorescence onto the PMT microchannel
plate: Also, ~ rotatable polarizer was positioned in the
S l! ~ 5°~i'~°L!?'E ~' ~ ~
CA 02132708 2002-08-14
i'7036-13
47
output optical path to measure the time dependent polari-
zation of the fluorescence,
High voltage from 1000 to 3400 volts wa;a supplied to
the PMT, The output of the PMT was connected to a Tek
tronix 7912AD Programmable Digitizer.
A computer was used to trigger the laser. The laser
output was detected by the digitizer via a connection to
the pulse detector (not shown). A programmable sweep on
the digitizer set up the time spread to be measured after
the laser pulse from 10 ns to as high as several seconds.
Typically, the system was operated such that 512 data
points were generated over a 20 ns time period.
The natural log Eln) (intensity) of the dye prepara
tions was plotted versus time and subjected to least
square linear regression analysis. These data are shown
in Fig. 4.
The dye preparations were analyzed for their inter-
action with serum protein. The dye preparations were
adjusted to 5 x l0'' M/L in FPIA buffer. These dye prep-
arations were added to the following solutions to a final
dye concentration of S x 10'1' M/L: FPIA buffer, 0.5% bovine
gamma globulin, 5.0% bovine serum, 5,0% normal human
serum, 5.0% pooled human serum and 5.0% whole blood
lysate. These data are shown in Fig. 5. Typically, when
a dye binds to a protein non-specifically (as can be seen
with the "C" fraction), a significant increase in fluor-
escence polarization occurs. This makes it impossible to
distinguish the specific polarization due to antibody
binding from the non-specific due to protein-dye inter-
action. The "B" fraction showed only minimal interaction
aver buffer as determined by measurement of transient
state polarization.
Example 2
Linl~~rs
In certain polarization assays, it is advantageous to
use a spacer or linker. These .inker cr spacer arms are
*Trade-mark
!VO 93/d936b P~CT/(JS93/OZ~t70
48
useful when different ligands are terminated by either a !
carboxyl or amino group. Tn addition, such compounds are e.
important when the probe needs to be separated (stood off)
from the molecule with the antibody binding epitope. This
may be necessary to reduce the potential of non-radiated
transfer of energy when antibody binds the specific epi
tope and/or to eliminate stearic hinderance. These
linker/spacer arms are generally the same in both the
ligand-probe and ligand-protein immunogen used to raise
antibodies to the ligand; in order to create a specific
binding pair. In polarization immunoassays, it is desir-
able that the spacer create a relatively inflexible linker
moiety.
Because the caged diearboxy silicon phthalocyanine is
y5 ~a .c~rboxy-terminated dye, it is advantageous to have an
amino terminated dye coupled to a carboxy terminated lig
and. Various linkers (spacer arms) have been evaluated.
Such compounds include giperazine, ethylenediamine, hex
anediamine, 6-amixao hexanaic acid, 5-aminobutanoic acid,
2d I.2-amizaododecanoic said, alan~.ne and other amino acids .
The folloc~aing ' methodology for preparation of the
phthalocyanine-12 amino dodecanoic acid compounds is an
example of the genoral reaction for such linkers: To 1.o
mg of caged phthalocyanine dye (the "B" fraction of step
25 G of eacample 2) in 200 ~1 of 500 DMF in water was added
2 mg of 1-hydroxybenzotriazole (HOBT) and 1 mg 12-amino
dodecanoic acid. The suspension was gently warmed until
al.l ingredients were dissolved. This took from I-2 hours .
At this'time 3.0 mg of 3-dimethylaminopropyl carbodiimide~ ,
3d way added and mixed thoroughly. The reaction mixture was
allowed to react overnight at 4°C. The reaction became
slightly turbid and was clarified by centrifugation. The
new abducts were purified on reverse phase C-18 columns by
HPLC.
35 ~nthen 12-aminododecanaic acid was used as a linker,
the ~aolarization ira glyce~'ol of the dye increased from p =
0:280 to p 0.340. Concomitantly, a 10 nm shift :.0 890
SLJ~S'T~1J~'E SHEET'
VV~ 93/193bb ' PCT/US93/02470 s
Y
49
nm occurred, which matches commercially available 690 nm
laser diodes. The change increased the dynamic range of
the assay from 0.03 to greater than 0.30 millipolarization
in buff er when bound to an antibody molecule. In addi-
S tion, the 10 nm shift increased the signal-to-background
ratio by moving away from the excitation maximum of f luor
. escing background molecules found in biological fluids.
The absorbance spectrum in methanol for the purified
caged dicarboxy silicon phthalocyanine dye-linker is shown
10' in Fig. 6. There is a 10 nm shift from 680 for the frac
tion "a" dye (Fig. 2); to 690 nm for the dye-linker. In
addition, transient state fluorescence polarization was
measured on the Diatron Analog System described in Exam-
ple I, in FPIA buffer and glycerol at 680 nm and in
15 glycerol at 690 nm. These data are shown in Figs. 7
and 8.
IIx. Synthesis of Caged Dicarboxy Silicon Phthalocyanine
Dioxin Probe
Digoxin is a glycosylated steroid which, when used in
20 patients with, congestive heart failure, increases cardiac
r.
output, decreases heart size, venous pressure and blood
volume, and relieves edema. As noted above, digoxin has
a very marrow therapeutic range (serum levels of 0.5 to
2.S ng/ml) and is generally toxic at concentrations
25 greater than 2.l ng/ml. Accordingly, there is a need for
a digoxin assay which can accurately and precisely deter-
mine digoxin concentrations at these levels.
~~ , , ,
Example 3
Diexoxin Probe Preparation~ Caced Dicarboxy
30 Silicon Phthalocvanine-Dic~oxicxenin
A ~ Preparation of 3 -~Cetodictoxictenin
A mixture of 488 mg digoxigenin, 7.S ml toluene, 3.75
ml cyclohexanone; and SSO mg aluminum isopropoxide was
hewed under reflux for 2.3 hours and then concentrated
5~~5 TE S~dEET
W~ 93/19366 ' . P~I"/US93/02470 ,~~ ,
t~
,,
in_ vacuo to half of its original volume. Two hundred gel
water was added and the mixture was evaporated in vacuo to
dryness. The powdery solid was dried in vacuo over over-
night. The dry residue was stirred in 25 ml methanol and
5 the resulting mixture was filtered. The residue on the ,
funnel was washed with 25 ml methanol. The filtrate and
washing were combined and evaporated in vacuo affording
920 mg white solid.
B. Preparation of 3-Aminodiqoxigenin
10 A mixture of 92O mg 3-ketodigoxigenin, 693 mg ammo-
nium acetate, and 730 mg, NaBH2CN was stirred in 48 ml
methanol at room temperature overnight. Concentrated HC1
(35 ml) in 20 ml methanol and 5 ml water were added.
after gas evolution had_subsided, the sol~,rent was removed
15 in vacuo. The residue was stirred in 15 ml water and then
extracted with 2 x 20 ml methylene chloride. The water
phase was gummy materiel was evaporated in vacuo and the
residue was dried leaving a granular solid, This dry
solid was extracted'with 2 x 20 ml dimethylformamide (DMF)
20 and the solution centrifuged. The clear DMF solution was
,.
evaporated in vacuo affording 684 mg white solid. The
entire amount was dissolved in 25 ml methanol and the
solution was stared at -20°C overnight. A white crystal-
lino material; which had deposited, was removed by fil-
25 txation and washed with 300 ~1 cold methanol. The fil-
trate was applied to two washed EM 576 silica TLC plates .
~l,f~er being developed ~:n 90 ml chloroform + 25 ml methanol
the chrnmatdgram shcawed 7 bands visible under 254 nmi uv. ~, .
Band Number 1, Rf 0.10, was -rempved from the plates,
30 extracted with 4 x 40 ml methanol and the SiOz was cen-
trifuged out . The supernates were combined and evaporated
in ~racuo affording 70 mg white solid.
C. Synthesis of Probe
The digoxin probe was prepared as follows : 4 . 2 mg of
35 3-aminodigoxigenin was placed in a 3.0 ml reaction vial
S~~s~~ S~F
WO 93J~9366 ' PCI"/i;S93/02470
51
and dissolved with 100 ~.1 DMF . In a separate vial , 1 . 0 mg
of caged dicarboxy silicon phthalocyanine (Compound IV
from Example 1 (G) ) was dissolved in 400 ml DMF and then
transferred to the reaction vial along with 200 ul of wash
DMF for a total of 600 ~Cl. 4.2 mg of 1-hydroxybenzotria-
zale (HOBT) was added to the reaction vial, which was then
dissolved and mixed well. To make the final reaction mix-
tuts, 10.5 mg of 1-ethyl-3-(3-dimethylaminopropyl/carbodi-
imide)-HC1 (EDAC) was added and mixed thoroughly. The
reaction mixture containing digoxin-phthalocyanine probe
was stored at 4.0-8.0°C overnight.
D. Purification of Probe
The digoxin-phthalocyanine probe was purified as fol
lows: a slurry of S gm C-18 was made in acetone and
poured into a 1x15 cm glass column. The acetone was
removed by the application of light pressure, and the
column was equilibrated by the addition of 4 column vol-
umes of 70% methanol/30% water. The reaction mixture
cor~taining digaxin-phthalocyanine probe was applied to the
column and flushed with 70% methanol/30% water. The probe
was eluted with 80a methanol/20o water, concentrated by
vacuum and further purified by two subsequent passes on
H~~,C. After the second chromatograph an HPLC, the probe
was brought to dryness in vacuo. A portion was dissolved
in methanol and a portion was dissolved in 100 mM lVaPO~
buffer containing 0.1°~ sodium azide and l.Oa bovine gamma
globulin (pH 7.5). Fig. 9 depicts a chromatograph of the
HPDC method' semi-p~'eP C-18 column with a mobile phase and
gradient elution of meth~nol/wa~er. Fig. 10 depiets the
30- structure of the digoxin-phthalocyanine probe.
A~alysis of Probe
The probe was'analy~ed in a Perkin-Elmer spectro-
photometer (Lambda 4 c) in two solvents, methanol and
100 mM phosphate, pH 7.5: Figs. 11 and 12 are repre-
~~ntative spectra. Fluorescence decay time was determined
~ l! ~ 5°T1Z'L,'TE S I~ E ~'1'
W~ 93/19366 ' P~C'T/L'S93/02470 ~,
t:..:. .
,~
52
to be.4.7 ns using the Diatron Analog System described in
Example 1, (Fig. 13).
F. Linearity in Buffer and 1.0% Bovine Serum Albumin
To determine the sensitivity and linearity of the .
transient sate measurement system and development of a
digoxin assay, it wa's necessary to determine the concen-
tration of the phthalocyanine-digoxin probe. The probe
was purified by one additional pass through HPLC for a
total of three (3) passes through HPLC chromatography
using reverse phase C18 columns. The probe was dried
under vacuum and dissolved in a 100 mM phosphate buffex,
pH 7.5 containing 0.01% bovine gamma globulin. The
absorbance maximum was determined and the absorbance of
the probe solution was measured. Concentration was
l5 determined using the following formula: A = cxBC, where A
Absorbance, a = extinction coefficient; B = path length,
and C = concentration. Thus; C ~ A In this example:
aB
of = 1' . 6 X 105, B _ 1 cm, and A = 0 . 227 . A,ccordingly; the
concezztration of the stock phthalocyanine-probe solution
was ' = 1: 45 X 10'6 M/L. a
Based upon this value of 1:45 X 10'~ M/L, dilutions of
appropriate concentrations were prepared in FPIA buffer
'and in l.0% bov~:ne serum albumin. The results are shown
in Fig. 14:
The lxnear~,ty of intensity of probe from 6.5 X 10-9 ~
to 1 X 10°x3 M digoxin in~ both FPI:A. buffer and FPIA buffer
with. 1% BSS demonstrates the ab~.lity of the probej'to:'~~unG- ;
tion in a protein solutionwithout interaction with-bind-
ing-components.
As can be seen from'the data presented in Table 2;
the polarizations of free dye ("B°' fraction) and 'free
prolae in various sera are similar, with the polarizations
being slightly higher gor the digoxin prcabe: This is
'cons-intent with the increase in molecular size and asym-
;:metry o~ the probe. Tn addi.tion, the changes observed in
5 U ~ 5'°~'i~'L1°f F S ~l
a.
W4~ 93/~1~3E6 ' P~'/LJ593/02470
,.,:
53
buffer vs the serum solutions is consistent with a change s
in viscosity as defined by the following equation:
r = 3 nV
RT
where: R - gas constant; T - temperature (in °R). n -
solution viscosity, and ~l = volume of molecules.
Table 1
Phthalocyanine Digoxin Probe: Serum Interactions
Comparison of milli-Polarization (mP)
O.So 5.Oo S.Oo
FPIA Gamma Bovine Normal
Buffer Globulin Serum ~-Iuman Serum
~~~n Fraction 8.0 20.0 22.0 30.0
Digoxin Probe 26.1 S1~~ 38.5 40.5
G. Serum Ura.ne Interactions
A comparison of a fluorescein-digoxin probe analyzed
on the Abbott TDx~ Fluorescence Polarization .Analyzer and
the phthalocyanine digoa~in probe analyzed with the Diatron
FAST-60 Analyzer is presented in Fig. 15. In this ex~m-
ple, both probes were tested at the normal (workizag)
concentration used when performing a digoxin assay with
the TDxT''' and, FA.~T--60 a~ialyzers ( i . a . , 2 . 5 ~ l0~lo and 5 . 5
X 15°11 M/~) . See' Example 4 for a description of the
p~,atron FAST-60' Analyzer. In this figv.re, the intensity
le~re~,s are plotted as background equivalents.
AS can' be se~era from Fig, i5, the fluorescein probe is'
only slightly detectable above background in 5o serum and
completely non-detectable in 10o urine. In contrast, the
phthalocyanine digoxin Pgobe is detectable at a very sig-
nificant level above background in both the same serum and
urine samples.
S~~S °~~ S~IEE't°
CA 02132708 2002-08-14
'17036-13
54
IV . Digoxin Assays
Example 4
Competitive Serum Assay for Diaoxin:
Seduential HindincLProcedure
Digoxin reacts with serum albumin and other serum
proteins at many reaction sites, Prabes made with a
fluorescent dye and digoxin will also react. "Nonspeci-
fic" binding or serum protein interactions were minimized
in this procedure by the action of the cyclodextrin, which
1G has an affinity for digoxin which exceeds digoxin's affi-
nity for constituents in serum. Thus the cyclodextrin
interferes with the binding of digoxin with serum consti-
tuents, but allows far binding of digoxin with digoxin
antibody. Thus, the assay was designed to allow both the
serum digoxin and the digoxin probe to react with the
digoxin antibody.
100 uL of serum sample was mixed with 25 ~L of rabbit
antidigoxin and 500 ~L Buffers (100 millimolar phosphate
buffer, pH - 7.6 with 0.01 bovine gamma-globulin, 0.5°c
gamma-cyclodextrin and 0.1~ sodium azide). The mixture
was incubated for 5 minutes. 25 ~.L of digoxin probe (as
prepared according to Example 31 and 200 ~L of Buffer2
(100 millimolar phosphate buffer, pH - 7.6 with O.Ol~c
bovine gamma-globulin and 0.1°-. sodium azide~ were added
and the mixture was incubated for 20 minutes.
In a study of 20 random human serum samples it was
found that the serum-digoxin probe interaction would vary
from sample to sample, and that the variation may be as
much as l0-15 millipolarization units. The buffers in the
present example were formulated to eliminate this varia-
tion to a relatively constant millipolarization of 70.
Transient state polarization was measured as
described in Studholme, et al . , U.S. Patent No. 5,323,008
entitled "Fluorometer Detection System ".
The transient state optical system
was installed in the Diatron "FAST-60 Analyzer," which
contains a laser diode operating at 685 nm was pulsed at
' PCT/US93/02470
i .":_:.
~~~~ ,
s
a 10 MHz rate. Typically, the laser "on" time was approx-
imately 3 nanoseconds. Photons from the solution were
detected using a photomultiplier tube (PMT) operating in
a single photon counting mode. The photon event along
5 with the relative time of the photon event as compared
with the laser pulse time was determined. By storing the
individual photon event times a histogram of frequency of
photons as a function of time was generated.
The Diatron FAST-60 Analyser includes a transient
10 state optical system installed in an automated fluores
cence reader designed ~.o measure fluorescence from immuno
- assay reactions: The reader contains the optical system,
motor control for position reaction cuvettes in front of
the optical system, thermal control to hold the system at
7,5 ~5°C and a computer link to control the reader, analyze
and display results and print those results. For immuno
assay use, the results were formatted into transient-state
polarization units,or, by using a calibration curve, the
results were transformed into concentration units of the
0 analyte being measured.
Commercial serum calibrators for digoxin determina-
tion (obtained from Abbott Laboratories) were analyzed to
obtain a standard curve. Sample blanks were prepared for
each sample or calibrator by performing the same steps
25 with the exception that buffer was added in place of the
digoxin probe. The sample blank was measured and sub-
tracted from the measurements for all reaction mixtures.
The procedures were performed at 25°C.
' ' Fig. 16 ' hows a comparison of digoxin calibr~atiori
30 curves by a standard fluorescence polarization procedure
(Abbott's TDx° Fluorescence Polarization Analyzer) and the
homogeneous sequential binding assay procedure described
in this Example. Fig. 17 displays a correlation plot of
37 serum samples assayed by a commercial digoxin test
35 system manufactured by: Abbott Laboratories (TDx° Digoxin
II In Vitro Test, Product #9511-60) and assayed by the
digoxin assay procedure described in this example. A
SU~STt~°LJTE SH~~'T'
p~./US93102470
d t:
_.
~~ 93/19366
~ ~ ~.
c
56
of 0.96 and a slope of 0.98 were determined.
correlation
a and y-intercept indicate no systematic bias.
The slop
F_xample 5
Competiti~re-.Serum_Assay f~ lgaxi~n'..
~ uf~ion Jur~~ Procedure_
tion jump Procedure described in this Example
The dilu
erformed in the presence of high
allows the assay to be p
s~~ri,; and was designed to reduce "non-
coxxcentrations of
,ions vahich compete with the antibody for
specif~.c" interac
' n and digoxin probe : While not wish° .
binding to the digoxi
b any particular theorya aPPlicants
~; ing to be bound y
when sample ~ antibody and d~.goxin probe are
believe that
reaction volume,, the "nonspecific"
incubated in a small
'initially compete with the antibody for
interacts~ns
in a.nd digoxin probe. then the solution
bixrding to digox
~~ bns ecif is" pxotein interactions tend' to
is diluted, the n P .
and only the specific antibody reaction
disappear rapidly
remains.
'ution jump procedure, 20Q ~L of '
To perf ~rm the dil
rotor (Abbott Laboratories) or serum sample was
2p serum calib
of antibody (rabbit anti-digoxin anti-
mixed with 250 ~.L
oxin probe and 1000 ~L, FPLA buffer.
body)', 250 ~,~, of d~.g
ated for 30 minutes at 35°C~ A
fihe mi~cture was incub
of the reaction mixture was removed and
variable volume
buf f er . For example 17 0 ~L of
~5 added to; 900 ~.L 'ot FPIA
in 900 ~L of buffer provided a fW al probe
reacted mixture
conc~ntraaion. of 5. x 10-11 M
r ~'' ure Bias diluted, the degr~e,of
As the reaction maxt
in decreased while the amount of specifl-
nonspecif is bind 9
shed nearly constant. As shown by
30 ca.lly bound Probe rema
de',icted graphically in fig. 18, for non-
the reault~s p
re (probe concentrat~.on of 3.5 x
diluted reaction mixtu
olarization was ~-89 mP : As the
IO-to M) , the resulting P
ted; the polarization decreased
reaction. mixture was dzlu
d alimit (near 152'mP) at a 7-fold dilu-
35 until it ruche
tion of the reaction mixture.
SU~ST~ 5'~~~'T
i
Wd 93/19356 P(.°T/L'~93/02470
57
An alternate dilution jump procedure was also done in
which 20 ~1 of commercial serum calibrator or serum sample
was mixed with 80 ~.1 of lysing/buffer (5 x 10'' M/L
stearyl-lysolecithin in .001 M/L Tris HC1 buffer at pH 8)
S diluent, and 10 ~l of rabbit anti-digoxin was added and
mixed. This mixture was allowed to incubate at 35°C for
minutes, At this time, 25 ~.1 of digoxin probe was added
a:nd incubated an additional 15 minutes. To this reaction
mixture 1.0 ml of FPIA buffer was added (dilution jump y
vortexed and the transient state polarization measurements
were made in the Diatron FAST-60 Analyzer described in
Example 4.
The calibration curve using the dilution jump pro-
cedure is shown in Fig. 19.
is Examt~le 6
Gomt~etitive Serum Assay for Diaoxin:
Signal-to-Background Comt~arisons
fox Transient-State and Steady-State Measurements
The signal~to-background ratios for steady state and
trahsient state measurement were determined as follows: '
.. The steady state fluorescence intensity measure-
ments were made on the Abbott TDx~ Fluorescence Polariza-
tion Analyzer using the '° Photo Check Mode , '° Both back-
ground and floor measurements wire taken by removing the
S reference solutions' from the calibration carousel and
substituting varying dilution of fluorescein from 1.7 X
10,'e M/L to 1 . 7 , X, 10'12 M/L in 1 . 0% bovine ,serum albumin . -
2. The steady state fluorescence of the caged sili
con ghthalocyanine-digoxin probe was measured in a modi-
3Q fled TDx~ Fluorescence Polarization Analyzer. . These
modifications were made by replacement of the input filter
with a' 680, 10 mm 1/2 bandwidth filter and the output with
a RG715 color Mass filter. The concentrations of the
probe solutions were determined in a similar manner as
35 those determined ~.n Example 3. Dilutions were prepared in
l.Oa bovine serum albumin in concentrations of 1.4 X 10'~
S UI ~ 5°f ~'i~'f F S H E ~~'
W(~ 93li936fi PCT/L'S93/02470 _ .
I
0 1.
M/L to 1.4 X 10-lz M/L. The measurements were made as in
1 above. '
3. The transient state measurements were made in the .
i
Diatron FAST-60 System using.the same solutions as in 2 ,
above, but at concentration of 1.4 X 10'g M/L to 1.4 10'1' _
M/L.
The data are summarized in Table 2. The signal-to-
background data is represented as a ratio (signal counts/
background count).
Table 2
Signal-ta-Background Ratio Comparisons
Serum Calibrator
680 nm Probe Intensity Blank Intensity Probe/
Fluorophore Seounts/15 sec) ~counts/15 sec) Blank
S teady-State 437,2 466,848 6.5
Transient-State 264,404 4,048 65.3
Thus; applicants have shown that steady state assays can
be configured with acceptable sa.gnal-to-background ratios
using the caged dicarboky silicon phthalacyanine digoxin
px'obe which is measured at 680 nm. There was an approxi-
mately l0-fold enhancement in this ratio When transient
state techniques were used to time-discriminate against
fast fluorescexs within the background and scattering
bands
zn immunoa~~ays wh~.ch measure analytes at very low
concentrations; for example, digoxin at 5 x 10-1 to 60 x
~~J,o' ~/L~ 'the concentrati;on of fluorescez~~ probe in' the ~ ,
fluorescein steady assay system is 2.5 x 10'1 M/L. These
assays require an extraction step to remove the digoxin
bound to serum proteins (approximately 25-40%) and fluor-
escers bound to proteins that interfere in the measurement
of fluorescence polarization. In serum, the background
fludroscenee is higk~ly polarized due to this protein bind-
ing, which can mimic specific polarization due to antibody
binding. These must ,be removed before the assay can be
nerfarmed. Many of these fluorescers are excited by 493
S tJ ~ S'~'~~t JT'E S ~ ~ ~'T'
. r ~:.. , , ;
r ~ r ..r,-.. :
.. , . ,. , ". .. . u~ ..
..u..._x:,.".. . r, , .".. ., ...... . , . .. .:r. ~ ,.. t . ~ . ... .. . _ ,_
,_....
..,. , ., .. _. ._ :~i:.. ..,... . , .. .. , . , .. ...-~.1. :.;-.
.
r
W~ 93/193fi~b ' PC'f/US93/02470
.. .
:': ,.~ a ~p
~~~~~~~ ~
59 ,
nm light, which corresponds to the excitation maximum of
these fluorescein based assay systems. In this example,
the signal to background is at best 2.5:1. However, when
unextracted serum is added as in a homogeneous assay using
the fluorescein steady state measurement technique, the
probe fluorescznce is not detectable at the concentration
'of serum needed to run the assay (Fig. 15).
This principle can be illustrated by the data in
Table 3. For a steady state fluorescein-based assay, the
fluorophore concentration at which fluoxophore signal
equals background is 1.6 x 10'9 M/L. This is far above the
concentration of fluorophore needed to perform an accept-
able digoxin assay, i . e-. , an assay which can detect and
quantitate digoxin at therapeutic levels. In other words,
the fluor cannot be measured over fluorescence of
background.
As can be seen from the data in Tables 2 and 3 , there
is approx~.mately an S-fold improvement in steady. state
caged dicarboXy si~.icon phthaZocyanine measurement over
steady state fluorescein measurement. This improvement
increases an additional 10-fold when transient state
measurements are made Casing the caged dicarboxy silicon
phtYa.alocyanine probe : Additionally, there is a dear 100--
fold imgx~ovement of the transient stake measurements over
the currently used fluorescein steady state measurements.
Table 3
Signal-to-aackaround Comparison: ,
.~~ , Fluorophore Caneentration Where
Fluor~hore Sicrnal Ernaals Background
' _Technolcaay Fluorophore Concentration
Steady'State Fluorescein 1:6 X 10'9 M/L
Steady State Phthalocyanine Probe 2.4 X 10'1° M/L y
Transient State Phthalocyanine Probe 1.4 X 10'11 M/L
5~U~5'~'tl°1.1°~°E SHEE'I°
P~.T/U593/02d70
WC~ 93/19366
' 60
Example 7
Competitive Serum Assay for DlCIOXIT1:
High Sensitivity Assay ,
In this assay, the sensitivity of the digoxin assay
described in Example 5 is increased by a factor of 10.
The concentration of caged dicarboxy silicon phthalocya-
nine digoxin probe was determined by the procedure out-
lined in Example 3 to be 4.2 X 10'lz M/L. In this assay the
total reaction valume was reduced by 50 percent and all
reactants were reduced 10-fold. To increase the polarize
tion values, the incubation times were increased to 5 and
_ l5 minutes for the sequential addition, competitive bind
ing format.
The procedure is as' follows: 50 ~C1 of lysir~g/buffer
(see Example 5 above) diluent was mixed with 2.0 ~.1 serum
calibrator or serum, plasma or whole blood sample and 5.0
~1 of rabbit anti-digoxin antibody. This mixture was y
incubated for 5 minutes and 2.5 ~1 of digoxin proY~e was
added and incubated an additional 15 minutes. After this
ineubation, 1.0 ml ,FPI~1. buffer was added as a dilution
jump. The transient stake polarization measurements were
G:
made on the Diatron FAST-60 Analyzer described in Exam-
Ple 4.
Sample blanks were prepared for each sample or
calibrator by performing the same steps, with the exeep
tion that buffer was added ~.n place of the probe. The
sample blank'was then measured and subtracted from the
measurements for the entire reaction mixtures.
;. ; F,ig ; 20 ~ displays a calibration curve for ccinmerdial
serum calibrators containing known cancentration of
digoxin, which were assayed using the high sensitivity
procedure.
Example 6
Pre~ai-ationo~ Whole Blood Calibrators
Whole bloocz was obtained from two donors by drawing '
blood into Vacutainer~' (Bector Dickinson) tubes containing
5~~5'~°1T~J~'E SHEF"f
WO 93/19366 ~'C.'T/U593102470
x~'a ~~. .s~! .;~ ;
;~ :~ .due 8
61
EDTA anticoagulant. The tubes were mixed thoroughly on a
standard laboratory sample rotator. Based on the average
specific gravity of blood being 1.056, a series o~ six
2 ml volumes of whole blood were weighed using standard
gravimetric technique. These samples were then spiked
using a USP grade digoxin (200 ng/ml) to final concen-
trations of 0, 0.5, 1.0; 2.0, 3.0, and 5.0 ng/ml whole
bland. 'fhe whole blood calibrators were stored at 4.0 -
8.0 °C and were used within two weeks.
ZO Example 9
W_ hole Blood Dictoxin Assay- Sictnal-to-Background
Ratio Serum Versus Whole Blood
Applicants determined the signal-to-background ratio ,
of the whole blood preparations which were pregared as
described in Example 8 and compared the whole blood (i.e.,
blank) intensity to the probe intensity. The resulting
values comparing the whole blood and sexwm signal-to-
background ratios are shown in Table 4. These measure-
ments were made at working digoxin probe concentration of
5 X 10'1'' M/L in the transient state system. It can be seen
that the net probe intensities remained constant even when
the background intensities fluctuated. In a typical
steady state fluorescein digoxin assay where the digoxin
is extracted by precipitation of proteins, the average
sa.gnal-to-noise ratio i s 2 to 1 at probe concentrations of
2.5 X 10'1 M/L, as c~ntrasted with those found by homoge-
neous transient state fluorescence for serum and whole
, .. ~ ~ i . . ~ ' ~
,'
. f
blood of 77.6:1 and 26.7:1, respectively.
Table
Signal-to-Background Ratio Comparisons a
for Serum and Whole Blood
z
Probe xn,tensity Blank Intensity
counts 15 sec? (counts/15 sec Probe/.Blank
Serum 113>2gc 1,466 77.6
Whole Blood 100,254 3,757 20.7
Sl3~STE SHEET
. . . ~.. .. : ..._.. ~ .... :.. .... . . ~..,.. ._... . . . . . . . .
:... . .. . .,.,. . ..
. ...... ., ....__ .. .. .. .__ . .. . . . ._.. . . , ... . . , . . .,
-
~....., , ... . .,... . .. z
... : . . .. ,. ~ . .. ,.. ...:, ..,,: ...., . .... , ., .... ..,
. ,. , ~. . ::.~. . ~.. . . . , . , . . .. .. . ... . . . .. ..
.. . . . ,. .. .. . . , .. ,... , . . .. ,
, ... . .. , . , .. ., . .
f
dV~ 93/19366 F°CT/US93/0247a ,
62
Examrle 10
Homo eneous Whole Blood Diqoxin Assav - Clinical Study
Previous whole blood immunoassays have been limited
by many factors. For example, separation steps are
S required in many assay systems, enzymes and other sub
stances released from red blood cells cause interference
in the assays, and the instrumentation is incapable of
measuring analyzes or reaction products through whole
blood hemolysates. Applicants have developed a homoge
° 10 neous whole blood assay system which offers the clinical
laboratory and other testing facilities significant advan-
tages over currently used methods, including decreased
J.abor cost, and decreased sample manipulation. In addi-
tion, with a homogeneous 5 to 1C minute assay, the pro-
15 cedure can be brought much closer to the patient, for
example, to the bedside, emergency care facilities clinics
and satellite testing facilities.
Digoxin is widely distributed in body tissues. Serum
and plasma have been the accepted samples for the assay of
20 digoxin using the current commercially available test
kits . Studies have shown a relative constant relationship
between heart muscle and serum digoxin levels, thus vali-
dating the use of digoxin serum levels in monitoring
patients receiving the drug (Doherty, J.E., et al., 1978,
2S "Clinical Pharmacokinetics of Digitalis Glycosides." Pro
gress in Cardiovascular Diseases, Vol. XXT, No. 2 (Sept./
Oct.)). Because whole blood has not been routinely used
a,s ~a medium fox assay in digoxin therapy, the following
,, ; ,
study was undertaken to determine: (1) the distribution y
3c? of digoxin in seaum, plasma and red cell components of y
blood; t2).the percentage discrepancy, if any, in digoxin y
levels of serum, pl~.sma and whole blood assays; and
(3) correlation among two currently commercially available
assays - the Abbott TDx° serum assay and the Dade Stratus°
3S serum assay - and the assay of the present invention. The
clinical study was conducted using 43 patient samples,
collecting 1 EDTA tube for whole blood or plasma levels
5th~S?'F SHE
_ W~ 93/99366 P'CT/US93/02~d70 r
63
and 1 tube for serum levels. Each sample pair was ana-
lyzed for serum digoxin levels determined by the Abbott
TDx~, Dade Stratus~ and Diatron FAST-60 Systems. Plasma
levels were analyzed by Abbott TDx~ and Diatran FAST-60
Systems. Whole blood levels were analyzed by Abbott TDx
and Diatron FAST-60 Systems. The Abbott TDx~ System used
the TDx~ Digoxin II In Vitro Test, Product #9511-60
(Abbott Laboratories): The Dade Stratus System used the
Dade Stratus~ Digoxin Fluorometri.c Enzyc~e Immunoassay
(Dude Diagnostics Division of Baxter Healthcare Division,
Miami, Flora.da). The Diatron FAST-60 System used the
methods described ?z~ Example 5 and the apparatus described
in Examr~le 4 and Studholme, et al.; Lyon & Lyon Docket
No. 195/129.
In this study, the primary concern was whether the
whole blood digoxin values were similar to the serum
values. Thus, to reduce the number of variables, the
whale blood hysates were clarified by centrifugation
befare assay.
The study subjects were randomly selected patients
currently on active dig~xin therapy. The following sam- r
Ales wexe taken from each patient: (1) Red stopper Vacu-
tainerT''' tube (no EDTA) for serum collection (a minimum of
2 ml required); and (2) Purple stopper Vacutainerz~' tube
(EDTA) fear whole bload assay and for plasma preparation (a
ma.nimum of 4 ml required). Both tubes drawn at the same
time. All blood; serum and plasma was stored at 4C until
as~a~ed.: A1,1 assays xun within 24 ; hours aft=er , draw~.ng . , , .
The Diatron FAST-60 Digoxin Assay System consisted of
'(1) caged dicarboxy,silic~n phthalocyanine digoxin probe
ir_ FPIA buffer w~.th 1% bovine gamma globulin; (2) rabbit
anti-digoxin in FPIA buffer with 0.1% bovine gamma globu- -i.
lin; (3) iysing/buffer diluent; and (4) FPIA buffer
(100 mM phosphate buffer with 1% sodium azide end .O1%
3S bovine gamma globulin).
~ ~! ~ S'i't I'%'E '~ ~i E E'T
. . . ... . , ; .- , , , ; .; . . ...,.; , ,;; . : , ~; , . , ~ : ;~ ;
;:
r.. ~ .:.:.-.~ ..
. ~.:.: ~...... .....,.w. .... ,.:., ,.,. ... .. , ,.. ..... .,....
.....:....
!C.F.Y.?u......;.a......,.......~.... ~?.~.. :.:.~.. . . :;...
,... , . , .... . .. :.,. ... .. :... ...,.... , . .. . , . ,. ... ..
.... ,.... .:, .. . . .. .. ........ .... ...
~'~ y3/a936~6 F'~T/US93/02470 .
~: _
64
Assay procedures were performed as follows:
1. Serum
A. Abbott TDx~ Digoxin II In Vitro Test - .
Performed according to manufacturer's
instructions.
B. Dade Stratus~ - Performed according to
manufacturer's instructions. Serum values
performed by Pathology Medical Laborator
ies, 11160 Roselle Street, San Diego,
California 92121.
C. Diatron FAST-60 - See Fig. 21.
2. Plasma
A. Abbott TDxa Digoxin II In Vitro Test
Performed according to manufacturer's
instructions:
B. Diatron FAST-60 - See Fag. 21.
3. Whole Blood
A. vAbbott TDx~ Digoxin II In Vitro Test
Performed according to manufacturer's
instructions, except for the precipitation
step. Digoxin extraction from whole blood ~
was accomplished as follows: to 360 ~cl
whole blood, an equal volume of Abbott
-Precipitation Reagent (Digoxin II? was
added with .immediate vortexing for 30 sec
onds:' The conical tubes were centrifuged
at 10,000 RPMs for two minutes. The
,, ,; , slightly brownish supernate was; removed ; .
very careful~.y with a Pasteur pipette and ,
transferred to the sample cup, to avoid the
transfer of small particles. ,
B' Diatron FAST-60 - See Fig. 21.
Red blood cells were lysed prior to assay by addition
of lysing buffer (0:001 M/L Tris buffer, pY~ 8.0 containing
3~ 5 x Z0'; M/L stearyl-lysolecithin). Palmitoyl-lys~lecithin
and myristoyl-lysolecithin in Tris buffer are equally
~ l.6 ~ 5°t~t.lTE S ~d
.. :.. ~ ~ .r ..: ....., . , , ~ .; . .:-. . . . . ..,~ . ". : :.: , , ,,: . .
:.. , ,,., .
c: .,... . r :.:.: t
-c.,
..v;:,' .'".'.~. , . .av.::~ - , ~ ::~. :.~. ~'., . ;...,:.. , ..~.:':..n
r.~:~:. ~..., ,: ~ '.;...~.:r :=:, ._,.., ...-,:.. ......~.n~'.. ..
...:..~.;..~..:
a :.:., ..... .,. ,,....... ., .. ,,,, ..;,, ,,... .~.:..,: ~:. .".:........
..... ,. ...
a...,",. , .. ........,. ., ,..~,... ..:,.....,.. . .. : .. .. .,:.,...... .
,......... ,.; . ,..... ... .,... ..;..... ,. . "..". ...,.......,. .........
.. , .., .;.,, ...... ....., ...... ,.. ... ,.
"", . ...:... ...,.~. , :.,.. ,.... ,;,. . ,...; ....:.~,: ~ ~.~....~...;...,
. .. ,.... ,.,........,. . , ,..... .."...,. "..,;.. .. ..,.,.,.,, ...,..
~.:.. .... ..... .." . ...... , ,
y.>. .. , ,.. . .. r~. ~ ,... .. .. .. . . . .... . . , .. .. .. ..~ .. .....
. .. , .......
WO 93/19366 ' P~CTlLjS93/02470 '
~~~~~~~
effective. These reagents are preferred in that they do
not interfere with the ~_mmunoassay at this concentration,
and that red cell ghost particle size is reduced in 30 to
60 seconds, thus reducimg ar~y effect of light scatter dur-
5 ing the fluorescence measurements resulting in a homoge-
neous, non-separation whale blood assay.
The results of the testing are tabulated in Table 5.
The values from the TDx° and Stratus~ are the result of
single point testing. Because the FAST-60 procedure used
10 "manual gipetting" and automated instrumental analysis,
the samples were run in duplicate and the raw values
averaged.
The correlation data are found in Figs. 22 through
26. These include TDx~ serum versus FAST-60 whole blood
15 (Fig. 22), Stratus° serum ~rersus FAST-60 whole blood
(Fig. 23) FAST-60 serum versus FAST-60 whole blood
( F 'ig , 24 ) , TI7x~ serum versus FAST- 6 0 serum ( Fig . 2 5 ) and
Stratus~ serum versus FAST-6O serum (Fig. 26).
As can be (seen in Figs: 22 and 24, a high degree of
20 correlation exists between TDx~ serum versus FAST-60 whole
blood, with R -- 0.96 and 0.97, respectively. When the
correlation data in Fig. 24 is compared to Fig. 23, it can
be seen that the data is system consistent within the
FAST-60 System. Figs. 25 and 26 show a comparison of TDx~
25 serum versus FAST-60 s~:rum and Stratusa serum versus FAST-
60 serum assay data with co~r~elations of R - 0.96 and
0 . g7 , respectively . '.l-lgain., it could be interpreted that
alslight system bias exists:
The' raw' digoxin values f eir the assays aye found' in
30 Table 5. A comparison ,of the FAST-60 serum and FAST-60
whole blood systems with tie TDx° and Stratusp Systems
reveals only small differences between the mean values.
This is surprising and remarkable w~.en considering the
differences inherent in the various methods used,: A
35 composite CV of 12:7% was found when comparing data for
all assays.
5 ~ ~ 5~'i'tJ'TF ~ ~i E
W~ 93/19366 - PCT/U~93/02~70
'
66
TABLE S
Abbott Diatron PML
TDx FAST-60 StratusMean %C.V
ng/mL ng/mL ng/mL
S Serum Plasma WB Serum PlasmaWB Serum All All
1 0.87 0.80 1.00 0.72 0.82 0.92 0.80 0.85 10.0
2 2.51 2.08 2.75 2.78 2.02 2.08 2.50 2.39 12.6
3 2.16 2.21 2.46 2:14 1.92 2.OS 2.10 2.15 7.2
4 1.11 0.92 l.ll 1.28 1.07 1.24 0.70 1.06 17.3
0 S 0.85 1.03 1.41 1.25 1.24 1.46 0.80 1.15 21.0
6 1.85 1.96 1.83 1.85 1.87 1.95 1.80 1.87 3.0
7 1.0B 1.35 0.90 0.82 1.19 0.77 1.30 1.06 20.4
8 0.82 0.8? 1.04 0.80 0.76 0.88 0.70 0.84 12.0
9 2.79 3.13 3.23 3.93 4.04 4.34 3.40 3.55 14.6
1 5 10 0.96 0.82 0.98 0.97 0.73 0.75 1.00 0.89 12.2
- 11 0.78 0.88 1.21 0.92 0.96 1.00 1.00 0.96 12.8
12 1.80 1.74 2.43 2.26 1.85 1.90 2.00 2.00 11.9
13 0.60 0.61 0.6? 0.49 0.46 0.53 0.70 0.58 14.4
14 7..70 1.50 1.91 1.77 1.35 1.68 1.70 1.66 20.2
2 0 15 1.06 0.96 1.27 1.20 0.96 0.95 1.10 1.07 11.0
16 0.53 0.53 0.60 0.68 0.45 0.43 0.60 0.55 15.0
17 0.08 0.02 0.00 0.22 0.27 0.00 0.00 0.08
7,80.77 0.74 1.08 1.09 0.95 0.83 0.90 0.91 14.3
19 1:66 l.S2v 2.Z0 1:86 1.68 1.75 1.70 1.75 9.7
2 S 20 2.41 2.23 3.24 3.04 2.32 2.60 2.60 2.63 13.2
21 1.19 1.32 1.62 1.53 1.39 0.98 1.20 1.32 15.3
22 1. 1.13 1.51 0.53 1.12 0.94 ~ 0.90 1.03 26.7
OS
23 1.64 1.58 2.01 1.28 1. 1.55 1.40 1.55 14.2
A0
24 1.65 1,62 2.35 1.83 1.62 1.77 1.50 1.76 14.7
3 0 25 2.95 2.87 3.59 3.82 2:93 3.66 3.00 3.26 11.6
26 3.41 3.13 3.65 4.20 3.80 4.06 3.30 3.65 10.1
27 0.52 0.55 0:71 0.82 0:74 0.42 O.SO 0.61 22.5
28 2.54 2.48 2.91 2.92 2.56 3.00 2.60 2.72 7.4
29 1.93 1.84 2:31 2.07 2.02 2.20 1.70 2.01 9.6
~ ~ 30 2.64 3.02 3:44 3.45 4.64 3.55 2.60 3.33 19.3
31 3.64 3.49 4.34 3.51 3.83 4.21 3.60 3.80 8.4
32'!1:24 1.06 ~ 1:13 1.12 1.23 0.86 1.10 1:11 '10;~
33 0..68 0.55 0:53 0.61 0.44 0.81 0.70 0.62 18.5
3q 1.26 1.23 1.28 1,36 p,97 1.38 1.20 1.24 10.1
~ Q 3S 1.'74 1.67 2.05- 2.00 2.04 1.98 1.80 1.90 7.6 ,
36 1.16 0:95 1.17 1.17 0.87 0.99 1.00 1.04 10.8
37 0.85 0.98 1.12 0.93 0.96 0.86 0.90 0.94 9.0
38 2.65 2.77 3:06 3.26 3.30 2.93 2.90 2.98 7.5
39 0.95 0.96 1:22 1.19 1. 0.90 1.00 1.04 11.0
OS
4 S --".00.66 0.63 0:67 0.76 0.73 0.82 0.80 0.72 9.4
.~1'0.84 0.89 1.03 1.02 0.93 1.01 1.00 0.96 7.1
S U E3 STT~dT'E S H E El
PCT/US ;/0270
~
W~ 93/19366
,. i
~,~~~ ~~ ~ '
67
42 0.9:9 0,67 0.90 0.70 0.64 0.72
0.?0 0.69 16.3
43 1.20 1.16 1.54 1.43 1.16 1.06 1.30 1.26 12.4
Meanl.47 1.45 1.75 1.66 1.56 1.60 1.49 1.57 12.7
oCV57.3 57.4 57:1 62.5 66,6 67.6 58.5 60.0
P.:
In addition, the digoxin values for whole blood,
plasma and serum obtained on the FAST--.60 system were
extrapolated from a single composite calibration curve.
Thus, it would appear that satisfactory digoxin values
could be obtained using either specimen and extrapolation
from only one calibration curve: In a clinical setting
thin is important a.n that less sample needs to b~ drawn
from the patient, and also, any one of the blood specimen
can be used in the measurement of digoxin. This saves the
patient from a needless extra venipuncture and saves'the
1S laboratory time; and additional cost for time and
mater~:als .
The above examples involve the preparation end use of
a Caged dicarboxy silicon phthalocyarine digoxin probe.
Those skilled in 'the art will appreciate that other types
of fluorescent dic~oacin probes which Comprise a detectably
labeled marker component which COmpri.ses a fluorophore
moiety comprising a luminescent substantially planar
molecular structure coupled to two solubilizing polyoxy-
hydrocarbyl moieties; one located on either side of the
planar mol~CUlar structure;- can b~ prepared: Those
skilled in the art will also readily appreciate the' fact
~ha~ Caked diCarl~oxy silicon phthalocyanine probes can be
prepared for other analyzes, as well. For, example, small
anal~rtes such as amikacin, gentamicin, netilmicin, tobra-
;mycin, carbamaz~pine, ethosuxima.de, valproiC acid, diso-
PY~'amad~, lidocaine,' Procainamide, quinidine, meth~trex-
ate, amitriptyline, mortripyline, imipramin~, desipramine,
vanCOmycin and cyc3osporine arc part~.cularly suited for
the assays described here~.n due to their size.
3S For exempla, as described in Examples 11--1°T below,
sucn probes were prepared for digitoxin, theophylline;
phenobarbital, thyroxine, N-acetylnrocainamide, primidone
S~J~S TE SHEE?"
WO 93/19366 " PCT/US93/02~70
6 ,
and phenytain, Those skilled in the art will alsa recog
nize that caged dicarboxy silicon phthalocyanine probes
can be prepared for peptides. For example, as described
in Examples 18 and lg below, such probes were prepared for
rubella virus peptide.
Example 11
Synthesis of Caked
Dicarboxy Silicon Phthalocr~anine-Diqitoxin
A. Preparation of 3-Aminodic,~itoxicxenin
3-aminadigitoxigenin was prepared by procedures
- similar to those described in Example 3 above for the
digoxin probe.
E. Preparation of Probe
The digitoxin probe was pxepared as follows: 0.8 mg
2S of 3-aminodigitoxigenin was placed in a 3,0 ml reaction
vial and dissolved zn Z00 ~cl DMF. Caged dicarboxy silicon
phtha~.ocyanine ~ (1. 0 mg) was added to the reaction vial .
Also added to the reaction vial were 0.5 mg HOBT and 2.0
mg EDAC and the resulting mixture was thoroughly mixed.
The reaction mixture was stored overnight at about 4 to
8°C.
C. Purification of Probe
The digitoxin probe was purified by procedures
similar to those described in Example 3 for the digoxin
probe. The structure of the caged dicarboxy silicon
p'h~h~'~.ocyariine-digitoxi,n probe' is shown in Fig . 27 .
Example 12
Synthesis of Caged
Dica~boxv Silicon Phthalocvanine-Theophvlline
A: P~epara~ionof Theoph~rlline 8-Butyric Acid
,~ mixture of 3I.3 g glutaric anhydride, 25 g 5.5-
diamino-1,3-dimethyl uracil, and 300 ml N,N-dimethyl-
an~,line was heated under reflux for 4 hours. Upon
SIJ~STE SHEET'
l~Gl 93! 19366 ' PCT/ US93102470
W , j
i
i
69
cooling, the product crystallized from the dark, clear
reactian mixture. The crystals were collected by fil-
tration, washed with benzene, then crith methanol, and .
dried affording 187 g light yellow solid. ,
B. Preparation of Theo~hylline-8-(N-2-Aminoethvl~
Bu_tyramide
To a starred mixture of 453 mg theophylline-8-butryic
acid, 6 ml DMF and 4 ml (THF) was added 240 ~C1 triethyl-
amine. The resultant solution was cooled in ice and 220
~:0 ~.1 isrsbutylchloroformate was added. After 1 hour the
slurry was added-to 2 ml ethylenediamine cooled in ice:
The reaction mixture was maintained at 0°C for 6 hours and
then concentrated to dryness. The residue, upon frac-
~icinal crystallization from chloroform + ethanol provided
369 mg pure theoghylline-8-(~T-2-aminoethyl)butyramido.
~. Preparation of Probe
The theophylline probe was prepared as follows:
~:.2 mg 'theophylline-8-(N-2-aminoethyl)butyramide was
placed in a 3.0 ml reac ion vial and dissolved in 100 ~zl
DMF.' In a separate vial, caged dicarboxy silicon phthalo-
cyanine (1:0 mg) way dissolved in 400 ~cl'DMF and then
transferred to the reaction vial along with 200 u,l of wash
DMF (for a total of 600,1 DMF). To the r~aGtion vial wad
added 6.1 nng of IHt~BT; dissolved and mixed well. To make
the ffinal reaction mixture, 7.0 mg EDAC was added and the
resu~.ting' mixture,mix~d thoroughly., The reaction mixture
j ~ ; , ' ( i'
Y
was stored~overnight at about 4 to 8°C.
D. Purification of Probe
The theophylline probe was purified using procedures
simihar to hose descra.bed in Example 3 for tha digoxin
xarolae. The structure of the caged dicarboxy silicon
ohtk~alocyanzne-digoxigenin probe is shown in Fig: 28.
Examb l ~ 1: 3
SIJ ~S ~°F S H EE°T
WO 93/19366 ' PCT/L1593/02470
~:~. ,1
1Ya ':~ ' ~ ~~ ..
'f~ ~ r
Synthesis of Caned
Dicarboxy Silicon Phthalacyanine-Phenobarbital
A. Preparation of Nitra~henobarbital
Phenobarbital, 663 mg, was dissolved in 2.7 ml con- .
5 centrated sulfuric acid cooled in ice. With stirring, a
cold solution of 0.16 ml concentrated nitric acid in 0.65
ml concentrated sulfuric acid was added dropwise over a
period of ~ minutes. After 1/2 hour in the cold, the
reaction mixture was poured into ice water. The precipi-
10 tote was collected, washed with water, and dried in vacuo
affording 0.03 g white solid.
B. Preparation of Aminaphenobarbital
Nitrophenobarbital (225 mg) was stirred in a mixture
of 3 cnl concentrated HC1, 2 'ml acetic acid and 2 . 5 ml THF .
15 To the slurrr~r mixture was added a solution of 370 mg SnCl2
in 1 ml concentrated HCl and l ml acetic acid. After
stirring at room temperature fnr 2 hours, the reaction
mixture was concentrated to give an oily residue. To this
residue, 1~?aHCO~ solution was added until the pH was about
20 7. The precipitate was collected, washed with water and .
dried in vacuo leaving 449 mg white solid. This solid was
stirred in 10 ml THF ~.z~d centrifuged to remove the inor
ganic material. The supernatant lic~ua.d was evaporated and
the residue was dried in vacuo affordirxg 138 mg light yel
25 low solid.
C., Preparation of Probe
~' The ph~nobarb~.tal probe was prepared as follows:~l.2 ~ .
mg of 5-ethyl-5-(aminophenyl)barbituric acid (P-amino- .;
phenobarbi~al) was placed in a 3.0 ml reaction vial and ,
30 dissolved with -200 ~1 DMA':-'Tn a separate vial the caged
dicarboxy silicon phthalocyanine (1.0 mg) was dissolved in
200 ~1 DMF and then transferred to the reaction vial. To
the reaction vial was added 2.~ mg HOBT, dissolved and
mixed well. To make'~he final reaction mixture, 2.8 mg oz
~ L~ ~ STt~IJ'TF S ~i E ~'
WO 93/19366 ' F'C°f/U593102470
i
71
EDAC was added and mixed thoroughly. The reaction mixture ,
was stored overnight at about 4 to 8°C.
D. Purification of Probe
The phenobarbital probe was purified using procedures
similar to those described in Example 3 for the digoxin
probe. The structure of the caged dicarboxy silicon
phthalocyanine probe is shown in Fig. 29.
Examtale 14
Synthesis of Caged
Dicarboxy_ Silicon Phthalocvanine-Thvxoxine
A. Preparation of Th~rroacetic Acid Ethvlenediamine ;
To a stirred mixture of 100 mg thyroacetic acid in
ml pyridine was added 16 mg of N-hydroxysuccinimide and
27.6 mg N,N'-dicyclohexylcarbodiimide. The mixture was
1,5 stirred for 2 hours at, room temperature and transferred to
4°C for 18 hours. The crystals were removed by filtration
and 8.03 mg of ethylen~diamirie was added to the filtrate
J
w~.th stirring. This xeaction was allowed to proceed an
additional 24 hours at 4°C and was dried i~ va~uo result
ing in whitish-gray powder. The material was stored at
-20°C in a desiccator.
B. Preparation of Probe
The thyroxi:ne probe was prepared as fral~.ows: 1.0 mg
of tetraiodothyroacetic acid-ethylenediamine (Tetras-EDA)
was placed ixa a 3 . 0 ml "reaction vial and ,dissolved in 100 ,
r
~,1 D~~, In a separate vial, caged dicarboxy silicon
phthalocyanine (1.0 mg) was dissolved in 400 ~.1 DMF and
then transferred to the reaction vial along with X00 ~.1 of
wash DMF for a total'of 600 ~,1. To the reaction vial was
added ~:. 8 mg Ht,~B"~, dissolved and -mixed well . To the final
reaotioz~'mixture; 1.5 mg EDAC was added and the result~.ng
mixture mixed thoroughly. The reaction mixture was stored
cavernighc at about 4 to 8°C.
S U STI~tJ?°~ S ~i ~ ET
!~O 93/~936g ' PCT/~.JS93/02470 .
I2
C. _Purification of Probe
The thyroxine probe was purified using procedures
similar to those described in Example 3 for the digoxin
probe. The structure of the caged dicarboxy silicon
phthalocyanine-thyroxine probe is shown Fig. 30.
Example 15
Svnthesis of Caqed
Dicarbo Silicon ~Phthaloc anine-N-Acet 1 rocainamide
A. Preparation of Desethyl-N-Acet~lprocainamide
Desethyl-N-acetylprocainamide was prepared by dis-
solving 1.0 g of p-acetamidobenzoic acid and 0.7 g
N-hydroxysuccinimide in 20 ml pyridine. To this solution
was added l.4 g of N,N'-dicyelohexylcarbodiimide. The
reaction mixture was placed at 4°C for 18 hours, at which
~.5 time the~crystals were removed by filtration. The filtra-
tion was brought to room temperature and with stirring,
0.51 g N-ethyl,ethylenediamine was added. Stirring con-
tinued for 3 hours, the solution was cooled to 4°C and
allowed to react for an addition 24 hours at 4°C. The
second crop of crystals was removEd by ~~.ltration, dis-
solved in 25 ml of distilled water. The pH was adjusted
to 10'with sodium hydroxide to form a white precipitate of
desethyl-N-aoetylprocainamide. The resultant precipitate
was dried in vacuo and .stored at -20°C in a desiccator.
B. Preparation of Probe
The N-Acetylprocainamide probe was prepared as fol-
~i , ; ~ i
lows: I.O mg desethyl-N--Acetylprocainamide was placed in
a 3.0 ml reaction vial and dissolved with 100 ~.1 DMF. In
a separa a 'vial caged dicarboxy silicon phthalocyanine
(1.O mg) was dissolved irk 400 ~C1 DMF and then transferred
to the reaction vial along with 200 ~.l of wash DMF for a
- total of 600 ~.1. To the reaction vial was added 4.2 mg
HOBT, dissolved and mixed well. To make the final reac-
tion mixture, 10:5 mg EDAC was added and mixed thoroughly.
S!!~S 'fF SHED?'
WO 93/19366 ' PCTlUS93/02470
73
The reaction mixture was stirred overnight at about 4 to
8°C.
C. Purification of Probe
The N-Acetylprocainamide probe was purified using
procedures similar to those described in Example 4 for the
digoxin probe. The structure of the caged dicarboxy sili
con phthalocyanine-N-acetylprocainamide probe is shown in
Fig. 31.
Example l6
Synthesis of Caaed
Dicarboxy Silicon Phthalocyanine-Primidone
A. Preparation of Nitr~oprimidone
Primidone, 1.60 g; was dissolved with stirring in
8 ml concezztrated sulfuric acid and cooled in ice . A cold
solution of 4~5 gel concentrated nitric acid in 2 ml con
centrated sulfuric acid was added over a period of 10 min-
ut~s. After 2 hours at 0°C the reaction mixture was
poured into ice water neutralized with cold sodium hydrox-
ide: The precipitate was collected; washed with water and
2p dried in vacuo affording 1:79 g white solid.
~: partition of Aminoprimidone
Ni.troprimidone 1.79 g was dissolved with heating in
15 ml concentrated, HC1 and 3S ml THF. The salution was
allowed to cool to room temperature. A solution of 4.86
Sz~Cl2 in 3 ml concentrated HC1 and 3, ml THF was added over
a period of'10 minutes. After stixring at room~tempera-
tur~ overnight, the reaction mixture was made basic wi h
~~
ammonium hydroxide: The THF layer was separated and
evaporated to dryne~s> The residue was dried in vaeuo,
stirred a.n In ml THF and centrifuged to remove inorganic
material. The clean THF solution was evaporated in vacuo
to provide a residue which upon fractional crystallization
from THF and petroleum either yielded 519 mg pure amino
nrimiczane . .
S L~ ~ S°l'f F S H ~ E°
V!'O 93119366 ' P~CT/US93/0247U . ,
74
C. Preparation of Probe
The primidone prabe was prepared as follows: 0.8 mg
5-ethyl-5-(4-aminophenyl)hexahydropyrimidine-4,6-dione
(p-aminoprimidone) was placed in 100 ~.1 DMF in a 3.0 ul
reaction vial. To the reaction vial was added 1.0 mg .
caged dicarboxy silicon phthalocyanine (1.0 mg) and 3.1 mg
HOBT. To make the final reaction mixture, 3.9 mg EDAC was
added along with 150 ml DMF and the resulting mixture
mixed thoroughly. The reaction mixture was stored over
10. night at about 4-8°C.
D. Purification of Probe
The primidone: probe was purified using procedures
similar to those described in Example 3 for the digaxin
probe. The structure of the caged dicarboxy silicon
phthalocyanine-primid~ne probe is shown in Fig. 37.
Examt~la 17
S~nthes~.s of Cacxed
Da.carboxvSilicon Phthalocvanine-Phenytoin
1~, Preparation of Probe
The phenytoin probe was prepared as follows: 1.2 mg
of diphenylglycine was placed in a 3.O m1 reaction vial
and dissolved with 100 ~cl DMF . In a separate vial , dicar-
bo~yphthalocyana.ne (3.0 mg) was dissolved in 400 ml DMF
and then transferred to the reaction vial along with 200
~,l of wash DMF for a total of 600 ~1. To the reaction
vial was added 6.1 mg of ~iOBT, dissolved arid mixed well.
,, , i ,
To,make the final reaction, mixture,' 7.0 mg of EDAC was ,
added and mixed thoroughly. The reaction was stored at
4.0-$.0°C overnight.
B. Purification of Probe
The phenytoin-ph~halocyanine probe was purified using
procedure similar to those described in Example 3 for the
digaxin probe. The structure of the caged dicarboxy sili-
con phthalocyanine-phenytoin probe is shown in Fig. 33.
S~~S TE SHEET
CA 02132708 2002-08-14
77036-13
Exam~cl a
Rubella Anti-Ig~~robe
A. Labeling of Goat Anti-Human IaG
Caged dicarboxy silicon phthalocyanine dye (12
5 ,moles) prepared according to Example 1 and purified by
DEAE Sephadex* chromatography was mixed with 1 ml of
pyridine-pyridinium chloride buffer made by mixing 5 ml
1 M HCl with 0.5 ml pyridine. The solution was taken to
dryness in a sublimation apparatus and the excess pyridine
10 and pyridinium chloride was removed, thus assuring that
all acetate ion present would be removed. The dry resi-
dual dye Was dissolved in anhydrous dichloromethane to
make a 3.5 mM solution.
The carboxylic acid groups of the dye were converted
15 to the imidazolide by mixing 1 ml of 3.5 mM dye with '760
~1 of 0.46 M carbonyl diimidazole and allowing 1.5 hour at
room temperature for reaction, after which the solvent was
removed i~r v a
DMF was scavenged free of water and reactive amines
20 by adding carbonyl diimidazole to a final concentration of
0.1 M.
To 100 ~1 of scavenged DMF was added 10 ~ul of HZO and
100 ul of this mixture was added in the cold to the dry
activated dye. After 1 minute this mixture was added to
25 a mixture of 600 ~cl IgG solution containing 6 mg of goat
anti-human IgG and 100 ~l of 100 mM phosphate, pH 7.6.
The reaction was allowed to proceed for 4.5 hours at room
temperature and overnight at 4°C. A pardon of the reac-
tion mixture was equilibrated with 10 mM phosphate, pH 7.6
30 by two treatments in a Minicori concentrator iAmicon Cor-
poration, Danvers, MA, USA) and passed through a hydroxy-
lapatite column (Bio Rad Laboratories, Richmond, CA, USA)
equilibrated with 10 mM phosphate, pH 7.6. Free dye
eluted at this stage and the labeled antibody was recov-
3=~ ered by elution with 100 mM phosphate, pH 7.6.
B. Analysis of Probe
*Trade-mark
1W0 93/~93G6 PCT/US93/02470
~~~~ ~~
76
The labeled antibody was found. by absorbance measure-
menu at 280 nrn and 682 nm to contain an average of 1.3
moles of dye/mole of IgG. It was Shawn to react specifi-
cally in a solid phase sandwich assay in which adsorbed
rubella virus antigen was coated with human anti-rubella
which enabled reaction with the labeled antibody. Tran-
sient-state fluorescence intensity measurements in the 680
nm region were used to quantify the bound labeled anti-
body. Specificity was further tested by correlation with
a standard method. A series of 40 patient samples were ,
_ run; 4 were negative by bath methods, 35 were positive by
both while one was positive by the standard method and
negative transient state fluorescence.
In clinical pathology and medical screening, speci
ficity is defined as the proportion of individuals with
negative test results for the disease that the test is
intended to revel, i.~e., true negative results as a
proportion of the total. number of true negative and false
positive results. In this example, by this definition,
this assay demonstrated 1000 specificity. In addition,
sensa.tivity of a procedure can be defined as that pro-
portion of inda.viduals wa.th a paaitive test result for the
disease that the test intended to reveal, i.a., true pasi-
tive results as a proportion of the total true positive
and false negative results. These data indicate a 97.2%
sensitivity for this assay system. Although we report a
relatively small number of samples the performance of the
assay, demons;trat~s the use of an antibody labeled; with: the ,
marker components in a sandwich assay. See F'ig. 34.
Example l9
S~rnthesis of Caged
Dicarboxy Silicon Phthalocyanine Synthetic
Rubella Peptide
A. pre~arat~.on of Probe
A synthetic rubella peptide, for example, a portion
of the El protein of the rubella virus (Therien strain),
SII~ST'E S~i~ET
CA 02132708 2002-08-14
77036-13
77
can be synthesized by standard peptide synthesis proce-
aure. The coupling of caged dicarboxy silicon phthalo-
cyanine (prepared according to Example ly to the synthetic
rubella peptide was a four step process:
1. Dye activat~.gn_ - Sufficient caged dicarboxy
silicon phthalocyanine in dimethylformamide to give a
molar ratio of caged dicarboxy silicon phthalocyanine dye
to peptide of 1.3 was activated by adding 50 moles of
carbonyldiimidazole per mole of dye in dichloromethane to
l0 form an imidazole.
2. Decomggsition o_~ excess ca,,~bo~,vldiimidazole -
The dichloromethane was evaporated from the activation
mixture and water was added to decompose the excess
carbonyldiimidazole.
3. Cougling~ ~o Peptide - The solution was buffered
by adding for each ~cmole of carbonyldiimidazole used 1 ~1
of a mixture of 100 ~cl of pyridine, 2.38 ml water and 620
ul of 1 M HC1. The resulting solution was added to the
dry peptide. Alternatively, the carbonyldiimidazolide was
reacted with the peptide in DMF.
B. ~urifi~"tion cl~~ Probe
Purification of caged dicarboxy silieon phthalo-
cyanine labelled peptide from the reaction mixture was
carried out by high performance liquid chromatography on
a reversed phase C" column using a water-methanol
gradient.
Ex~mp 1 a 2 0
Immunoloaical Evaluation of Phthalocvanine-Rubella Probe
Two assay procedures were performed in order to
waluate the phthalocyanine-Rubella probe. The phthalo-
rvanine-Rubella Probe was diluted in a 0.01 Tris buffer
pH 8.0 containing 0.1 o bovine serum and 0.025°s Tweeri 20.
*Trade-mark
WO 93119366 FC,"T/CJS9310~470
:. , -.
a ~a .
78
The probe concentration was determined to be 1.1 x 10-1'
M/L. Rubella peptide calibrators were made by diluting in
the 'iris buffer to the following concentrations: 0.0, 1.0 .
x 10'12, 2.7 x 10y~2, and 5.4 x 10'2, 2.7 x 10'11, and 5.4 x
10'11. The antibody was made by hyper-immunizing a rabbit .
with the Rubella peptide. Dilutions were made in the Tris
buffer described above.
A. Com etitive Bindin Assa Se ential Format
To a series of small conical test tubes was added 25 '
~,l Tris buffer, 20 ~.1 antibody solution and 10 ~1 antibody
solution and l0 ~,l of each peptide calibrator. The tubes
were incubated at 3S°C for 10 minutes. At this time,
~.1 of probe was added to each tube and incubated an
addita:onal 20 minutes a~ 35°C. After incubation 1.0 ml of
15 iris buffer was'. added and transient Mate polarization
measurements were made.
The typical inhibition curve is shown in Fig 35. The
'data obtained clearly demons~ra~ed a sensitivity of l..o x
10'kx N1/L of peptide in a homogeneous fluorescence polariza
t~.on assay.
D, Antibody Titration Curve
To~a series o~ small conical test tubes, varying
di.lut~.ons ~f rubella 'antibody (20 ~cl) was incubated with
20 ~.1 Tris buffer and l5 ~1. probe for 20 minutes a~ 35°C.
After incubation 1.0 ml of Tris buffer was added ~:o each
tube; ,and the tra~.sient state, polarization was . measured.
~ ~
The data obtained are shown in Fig. 36. .
A typical antibody was obtained at a probe concentra-
tion of 2.0 x 10°11 MIL. As can be seen from the data, the
probe when depolarized in buffer has a polari.zata:on of S1
millipolarization units (mP) and when bound to antibody ~,,j'
exhibits a polarization of X15 mP, with a dynamic range of
164 mP. Thus, indicating the ability to use a homogeneous
polarization assay for detection of rubella virus anti-
bodies zn human serum samples.
5 ~ ~ 5'TTTLIT~ S H ~ ~'I°
yyp g3/ig3~,~ ' PCT/US93/02470
~~~f~ f ~~
79
Those skilled in the art will recognize that the
methods used in the above examples relating to rubella
peptide are applicable to other peptides. For example,
probes can be made for peptide hormones such as luteiniz-
S ing hormone, follicular stimulating hormone, human chorio-
gonadotropin, thyroid stimulating hormone. Angiotensin I,
Angiotensin II, prolactin and insulin. Probes can be made
for peptides such as tumor markers (for example, carcino-
embryonic antigen) as well.
20 To assist in understanding the invention, the results
of a series of experiments have been provided. The above
examples relating to the present invention should not, of
course, be construed as limiting the scope of the inven-
tion. Such variations of the invention, now known or
~.5 later developed, which would fall within the purview of
those skilled in the art are to be considered as falling
within the scope of the invention as hereinafter claimed.
SIJESTE SHEE'1°°
