Enhanced
ELISA:
how to measure
less than
10 picograms
of a specific protein
(immunoglobulin)
in less than 8 hours
ERIC
MACY,
Diuision
MICHAEL
of Clinical
Cahfornia,
KEMENY,1
Immunology
Los Angeles,
AND
and Allergy,
CA 90024,
USA,
ANDREW
Department
SAXON*
of Medicine,
and TDepartment
UCLA
of Medicine,
School of Medicine,
Guy’s Hospital,
London,
University
mit the measurement
of the antibody
dividual
B cells driven to high-rate
ABSTRACT
only
a relatively
short
culture
of
England
period,
produced
Ig secretion
as can
by inafter
be done
In this paper we outline a flexible and rapid method
to
measure
picogram
quantities
of isotype-specific
immunoglobulin
(Ig), including
IgE. Only
readily
or
with hemolytic
plaque
or enzyme-linked
immunosorbent assay (ELISA)
spot
assays
(4-5).
Radioimmunoassays can detect this level of Ig production,
but for
commercially
available reagents are required: isotypespecific, anti-human
Ig murine monoclonal
antibodies
(Mab) to coat microtiterplates,polyclonalalkaline
reasons
phosphatase-coupled
isotype-specific
F(ab)’2
or Fab’
fragments
as second antibodies,
and an enhanced
developing system that amplifies
the signal-to-noiseratioof
the quantitatively
bound
second
antibody.
The procedure is detailed in the appendix to enable easy application, even if one has no previous experience
with
ELISAs.
This system can be used to detect less than 10
picograms
of Ig in cultures
supernatants
of cells that
contain
mixtures
of various
Igs and it can be used to
detect
method
the product
of a single cell producing
1g. This
to measurement
of the
also will be applicable
quantities of lymphokines
minute
active molecules
and other
biologically
produced
in vitro and found in various
fluids
in vivo. MACY,
E.; KEMENY,
M.; SAXON, A.
Enhanced
ELISA: How to measure less than 10 picograms of a specific protein (immunoglobulin)
in less
than 8 hours. FASEB].
2: 3003-3009; 1988.
Key Words: IgE measurement
immunoglobulin
measurement
enhanced ELISAs
enzyme-linked
immunosorbent assay
immunoglobulin
synthetic rate
THE
ABILITY
TO REPRODUCIBLY
MEASURE
in
an
isotype-
specific
fashion
small quantities
of immunoglobulin
(Ig)1 produced
in vitro is important
for studies of human
humoral
immunity
(1). Naturally
occurring
plasma cells
can produce
2000 molecules
of Ig/cell
per second
(2).
Spontaneous
Ig-producing
peripheral
blood lymphoblastoid
cells or B lymphocytes
driven
to high-rate
antibody
production
produce
approximately
an order
of
magnitude
the ability
less, about
to measure
092-6638181002-30010.50.
1 pg/cell per day (3). Therefore,
less than 10 pg of Ig would per-
© FASEB
of speed, ease of assay, and
safety, we
have
adapted
ideas and methods
from our own and other
laboratories
with regard to ELISA
technology
to provide for a rapid and sensitive ELISA system to measure
all Ig classes
with
reagents
that
are available
to all inves-
This methodology
could readily
be adapted
for other proteins
found in picogram
concentrations.
tigators.
A sensitive
and
outlined.
recently
specific
Using
enhanced
ELISA
a combination
for IgE
of those
was
re-
other
isotype-specific
murine
monoclonal
(Mab)
anti-Ig’s as coating antibodies,
and Fab’ or F(ab)’2 alkaline phosphatase
(Alk-phos)-coupled
developing
antibodies in enhanced
sandwich
ELISA
assays, we have
confirmed
those findings
for IgE and extended
the
methodology, so that we are now able to assay for IgG,
IgA, 1gM, or IgE with sensitivity
and specificity
at less
than 10 pg and detect
the product
of a single Igproducing
cell.
agents,
MATERIALS
Regular
and
AND
enhanced
METHODS
ELISAs
Detailed
methods
and reagent preparation
are described
in the appendix.
Briefly, microELISA
plates are coated
with monoclonal
antibodies
or antigen
for 2 h. The
plates are washed
three times and then blocked
with
bovine
serum albumin
(BSA) or ovalbumin
(OA) for
1/2 h. Serum
dilutions,
standards,
or culture
supernatants
to be assayed are added,
diluted
if needed
in
‘Abbreviations:
ELISA,
enzyme-linked
immunosorbent
assay;
Alk-phos,
alkaline
phosphatase;
INT, iodonitrotetrazolium blue;
BSA, bovine serum albumin; OA, ovalbumin;
OD, optical density;
p-NPP, para nitrophenyl phosphate;
Mab, murine
monoclonal
antibody; Ig, immunoglobulin;
BBS, borate-buffered
saline; PBS,
phosphate-buffered saline.
3003
BSA or OA, and incubated
for 2 h. The plates are then
washed
three more times. The second antibody,
Alkphos coupled,
is added for 2 h, and then the plates are
washed three times
each with two different
wash solutions.
The developing
reagents
are then sequentially
added. The complete
assay can be accomplished
in 8 h.
Data
1.0
O.D.IgE-PS100pI
o
o
Cells from a patient with atopic disease and high serum
IgE (>10,000 IU/ml) were obtained by buoyant density
centrifugation,
separated
into the B and T cell fractions, and cryopreservcd
in a standard
fashion (see ref
7 for description).
The cryopreserved
B cell fractions,
known to spontaneously
produce
IgE as well as IgG
and 1gM, were thawed, washed, and cultured at four
different
cell concentrations
(5000, 10,000 50,000, and
100,000/well)
in 60 replicate
microculture
wells in 200
l final volume in complete RPMI
(8). The supernatants
were harvested
on day 5 and Ig isotype
deter(IgE,
ELISA
samples.
on
IgG,
each
and
well’s
1gM)
were
supernatant
done
by
by enhanced
using
60-!il
RESULTS
The
most
Ig for which
difficult
it has been
to accurately
most important as well as
measure
in the
picogram
range is IgE, so we focused on the ability of this assay
system to measure
IgE. Figure
1 is an example
of an
enhanced
ELISA IgE standard
curve that shows a comparison
of the standard
IgE preparation
from
the
World Health Organization
(WHO)
with the IgE paraprotein
PS, a IgE myeloma
protein
used as a standard
in our own and in many other laboratories.
All data
shown in Fig. 1 were obtained from a single
96 wellenhanced
ELISA plate by using
a 30-mm
substrate
and
a 30-mm developer
incubation.
Although
the two standards
showed
somewhat
different
curves,
both were
linear
from
3.9 to more than 250 pg/well (r values
of
.99). Two different
but parallel PS assays are also shown
in Fig. 1, where the specified
number
of picograms
of
IgE were pipetted
into the wells as either 20 or 100 tl.
We were concerned that adding 20-gil samples might introduce
a greater
chance of error. With 2O-jl samples
the wells were filled with 80 /Ll of 0.5% BSA to a total
volume
of 100 ed/well,
and the curves were virtually
identical
to those obtained
with 100-tl samples.
When
standards
from the United
Kingdom
(UK),
National
Institutes
of Health (NIH),
and WHO
were compared
with PS, PS produced a twofold higher opticaldensity
3004
Vol. 2
Nov.
0
0.4
y -0.01 74 + 0.0015x
WHO:
y-0.0485+O.OO29
PS 100il:
PS 20tl: y=0.0302+0.003x
0.2
0
100
200
-
-
0.99
0.99
0.99
400
300
pg IgEweII
Figure 1.Enhanced
cell culture
minations
0.6
0
analysis
and
O.D. IgE . WHO
ID O.D.IgE.PS20sI
ELISA plates
were
read
on Bio-Tek instruments
model
EL31O hard wired
to a Macintosh
II computer.
The
data were processed
with the ELISA Master program,2
and graphed
using Cricket
Graph.
Limiting
dilution
curve fit, mean number
of precursors,
and 95% confidence intervals
were done with the maximum
likelihood method
(6).
Cells
Io
08
1988
IgE ELISA, WHO standard vs. PS myeloma.
Increasing amounts of WHO
and PS 1gB were assayed. After the
second antibody (Alk-Phos
Fab’ anti-IgE)
was washed
out, substrate was reacted
for 30 mm, followed by developer
incubation
for
30 mm. The PS IgE was added in either
100
or 20 jil (El) and
then topped up to 100 &l volume with BSS 1% BSA. The WHO
IgE
(#{149})
(0)
was added in 100 ol.
(OD)
than
than the NIH standard
and a twofold lower OD
the UK and WHO
standards
with equivalent
amounts
added in our assay (results
not shown).
Thus
all four standard IgE preparations
assayed
within
a
fourfold
range and PS were in the middle of the range.
Figure
2 shows the plate-to-plate
variation
between
four IgE-enhanced ELISA platesdone on the same day
with two different substrate incubation times. There
is
no significant
plate-to-plate
variation
in signal from
standard curves with identical
incubation
times. With
the longer substrate
incubation,
there was a marked
increase in signal: 170-243%
over the range of dilutions
used.
Figure
3 shows typical saturation
between
1000 and
2000 pg of IgE/well in our IgE-enhanced ELISA when
developed
with 45-mm
incubations
for both substrate
and developer.
IgG, IgA, or 1gM added at up to 28,800
pg/well did not produce
a measurable
signal in the IgEenhanced
ELISA nor did all possible reciprocal
combinations with the other isotype-specific
enhanced
ELISAs
(data not shown).
The useful linear portion
of typical
enhanced
ELISA
curves has been between
0 and 500
pg Ig/well.
Equivalent
sensitivity
and specificity
were obtained
with enhanced
ELISAs
measuring
IgG, IgA, and 1gM
by using the specified coating and developing
antibodies
(see appendix).
Data for IgA, 1gM, and IgE assays are
shown in Fig. 4, which demonstrates
the increased
sensitivity
of the enhanced
development
compared
with
regular p-nitrophenyl phosphate (p-NPP) development.
Identical
primary
and secondary
antibodies
for each
isotype were used on each plate. One-half
of each plate
2Giftgenerouslyprovidedby Dr. Richard Deem, UCLA
ment of Microbiology and Immunology, Los Angeles,
USA.
The FASEB Journal
Depart-
CA 90024,
MACY
ET AL.
B
A
1.8
1.5
P1.1.83
#{149}
P111.11
#{149}
Plawes
#{149}
P1.1.12
45 mInutes withsubstrate
25 mInutes with substrate
1.2
0.9
0.6
0.3
0
7.75
15.5
31.25
62.5
125
250
500
1000
pglgE/w.II
0
7.75
15.5
31.25
pg
Figure 2.
Reproducibility
cubated
substrate
over
the linear
portions
of the curves.
250
is a 170-243%
There
500
1000
IgE/well
standards.
Plates 1 and
between
the pairs of plates
increase in signal with 45-mm
2 were inincubated
compared
incubation.
was developed
with the regular
reagents
(p-NPP)
and
the other half was developed
with the enhanced
reagents
to provide
a direct comparison
between
the two development
systems. The enhanced
ELISAs provided
much
better detection
of low levels of all Igs and could readily
distinguish less than 10 pg Ig from background.
The
enhanced
ELISA,
compared
with the regular
development, produced
ODs from 5.6 to 13.5 times higher for
10-pg Ig determinations,
with no significant
increase
in
the background
OD of the blank wells.
To demonstrate
the ability of the enhanced
ELISA to
identifythe Ig produced by individual cellsaftera relatively short culture,
B cells known
to spontaneously
secrete IgE as well as IgG and 1gM from an individual
with atopy, and a serum IgE greater than 10,000 IU/ml,
were cultured at limiting dilution (6). Sixty replicate
wells were
set up each with 5000, 10,000, 50,000, or
100,000 cryopreserved
B fraction
cells in a final culture
volume of 200 tl.After 5 days, supernatants from each
well were divided
into 60-1d aliquots
and assayed
for
IgE, IgG, and 1gM. Individual
wells that were positive
for IgE or IgG with a single precursor
produced
about
40 pg Ig/cellper 5-day culture,whereas 1gM-positive
wells produced
approximately
10 pg Ig/cell per 5-day
culture.
Figure 5 represents
limiting
dilution
curves for spontaneous
IgG-,
IgE-,
and 1gM-producing
cells. The
number
of B fraction
cells added
that produce
37%
negative wells corresponds to a mean of one precursor
per well. Based on B fraction
cells, one cell of 3500
(95%
confidence
interval,
2700-4400)
was an IgGproducing
cell, one cell of 5100 (95% confidence
interval, 4100-6500)
made IgE, and one cell of 30,800 (95%
confidence
interval,
23,700-40,100)
spontaneously
synthesized
1gM.
Figure
6, which was generated
from the same limiting dilution
data depicted
in Fig. 5, shows that the
mean IgG, 1gM, and IgE produced
per well was linear
from 5000 to 100,000 B fraction cellscultured. This
demonstrates
the absence
of a cell density
effect and
ENHANCED ELISA
125
the enhancedELISA. Four plateswere used to assay identical PS IgE
for 25 mm (A) and plates 3 and 4 were incubated
for 45 mm (B). The variation
of
with substrate
at the same times is not significant
to 25-mm
62.5
validates
that we were measuring
the Ig from individual
The consistency
of the data shown by measuring the total amount
of IgG and IgE produced
in
wells containing
the largest number
of cells is within a
factor
of 2.5 of that calculated by multiplying the
predicted
number
of precursors
in the well by 40 pg
precursors.
Ig/precursor
produced
in 5 days.
If every positive
precursor
(i.e., spontaneous
Igproducing
lymphoblastoid
cell) was able to divide every
24 h, after 5 days each culture
with a single initial
precursor
would contain
a maximum
of 31 cells/day
of
Ig production
(16 cells + 8 cells + 4 cells + 2 cells + 1
precursor),
providing
an estimate
of approximately
1 pg Ig/cell per day. If the cells proliferated
less, the
estimate of the corresponding
amount
of Ig/cell would
increase.
S
S
2.0
1.6
1.2
0
0
.DE-PS1
0.8
0.4
0
400
800
1200
1600
2000
pg IgE/weII
Figure
3. IgE-Enhanced
ELISA
saturation
curve.
An IgE en-
hanced ELISA with PS IgE
was performed
with a 45-mm incubation period
for both substrate
and developer.
The linear
working
part of the curve was as much as approximately 500 pg/well, with
saturation
occurring around 2 ng/well.
3005
A
B
0.6
IgE (E) 60:
y
-
0.0014
+
S
r -0.98
0.0066x
IgA(E)
10:
y
0.0261
-
+
0.0168x
r -0.98
0.5
0.20
0
0.4
IgE(R)120
0
0.li
0
#{149}
IgE(E)60
In
S
h
IgA(R)50
A
IgA(E)10
0.3
0
0.1(
0
0.2
IgA(R)50:
y-0.0044+0.003x
r-0.99
0.1
r-0.90
0
5
10
15
20
25
30
35
pg IgA/well
pg IgE/well
C
IgM(E)11:
yO.0432+O.029lx
r-0.99
U
08
Figure 4. Comparison of the enhanced vs.regularELISA method
for detection of low levels of IgE (A), IgA (B) and 1gM (C). Plates
o
0.8
IgM(R)50
were incubated with Mab coating reagents (Ig standards
applied)
and then the lefthalfof each plate was developed by using the enhanced
ELISA method
(E); the right half was developed
with the
unenhanced
regular
(R) technique. Each isotype was compared
on
its own plate. The enhanced
ELISA was done with a 45-mm substrate
incubation
and then scanned
at 490 O.D. after a 60-mm
#{149}
IgM(E)11
0.4
developer
incubation.
The regular
ELISA was incubated
in the
dark for 120 mm with its substrate
and then read at 405 O.D..
Curve fitwas calculatedusing CricketGraph on a Macintosh II
r-0.99
02
computer.
0
5
10
15
20
25
30
35
pg 1gM/well
DISCUSSION
The enhanced
ELISA was originally described
by Stanley
and co-workers (9) in 1985, and the enhancing reagents
are now available from Gibco/BRL
(Bethesda
Research
Labs., Gaithersburg, Md.) for use with any Alk-Phosdeveloped
ELISA.
The reagents
may also be produced
from the constituent
chemicals
and enzymes
(see appendix).
Previously,
the rate of signal development
restricted the lower limits of detection
in ELISAs because
background
development
would rise to obscure low signals.
Steps such as developing in the dark for a period
of several hours to overnight
have been used to boost
the signal but this did not yield a major change in the
signal-to-noise
ratio. The enhanced
ELISA uses alkaline
phosphatase
coupled
to the developing
antibody
to
generate
NAD
from NADP.
The NAD
is cycled in
the development
step between
alcohol dehydrogenase
and diaphorase
to reduce
iodonitrotetrazolium
blue
(INT)
to a colored
form and thereby
rapidly
generate
a strong signal before the usual nonspecific
rise in background
signal noise occurs.
The limiting
factor in the enhanced
ELISA is the behavior of the antibodies
used. We use a variety of read+
3006
Vol. 2
Nov. 1988
ily available
high-affinity
mouse,
anti-human
heavy
chain-specific
monoclonal
antibodies
for the coating
step
and
Alk-Phos-coupled
Fab’ [anti-human
IgE
produced
by IQ(Bio)
Limited,
Cambridge,
UK, using
a rabbit anti-IgE
antiserum
from Dr. M. Kemeny]
or
F(ab)’2
[goat anti-human
IgA and 1gM from Tago,
Burlingame,
Calif. or anti-human
IgG (A-0287)
from
Sigma Biochemicals,
St. Louis,
Mo.] as second
antibodies. The use of Fab’ or F(ab)’2-labeled
reagents
as
second antibodies
reduces
nonspecific
Fc region interactions
and lowers background
color development
in
the assays.
Aliquots
of an alkaline
phosphate-coupled
Fab’ antiIgE reagent
and the Mab anti-human
IgE antibodies
(CIA-E-7.12
and CIA-E-4.15)
will be available
to research
investigators
from
the National
Institute
of
Allergy
and
Infectious
Disease/NIH
Serum
Bank,
Bethesda,
Md., in care of BioTek Research
Labs, 1600
East Gude Drive,
Rockville,
Md. 20850,
USA. The
other Alk-Phos-labeled
developing
antibodies
are commercially
available
as noted
previously.
Mab
antihuman
IgG, Hb 60; anti-IgM,
Hb 57, and anti-IgA,
CHE 86, are available
from the American
Type Culture Collection
(ATCC),
Rockville,
MD. Other
high-
The FASEB Journal
MACV El AL.
affinity,
heavy
chain-specific
monoclonal
antibodies
could be substituted.
Difficulties
in the measurement
of low levels of IgE
found in vivo and particularly
produced
in in vitro systems emphasize
the need for sensitivity
and specificity
of the enhanced
ELISA.
IgE was initially reported
to be
produced
by fetal cells as early as in the 11th week of
fetal gestation,
but it has been shown that it is likely to
be caused
by detection
of private
light-chain
determinants
by the early polyclonal
anti-IgE
antisera
(10).
Similarly,
the interpretation
of data regarding
the in
vitro production
of IgE from mononuclear
cells driven
by a variety
of stimuli
has been difficult
because
of
problems
in accurately
measuring
the small amounts
present,
even when
the issue of polyclonal
antisera
binding
to idiotypic
or similar
nonisotype
restricted
elements
in known
(1). Thus,
laboratories
report
as
much as a 100-fold difference
in the IgE obtained
with
the same culture
system, and measurement
of levels of
IgE below 5 ng/ml was highly variable
in an interlaboratory
evaluation,
although,
as has been noted, the
samples provided
in that assay may have contributed
to
some of the variation
(1). The ready availability
of the
enhanced
ELISA reagents
and antisera
for all of the Ig
isotypes
as provided
for later in this paper
should
obviate
these problems
for investigations
regarding
human
Ig.
The enhanced
ELISA is sensitive enough to detect Ig
produced
by individual,
spontaneous,
high-rate
Igsecreting
cells from a culture of only several days. Thus
the enhanced
ELISA system is readily adapted
to measuring Ig produced
by cells brought
together
in limiting
dilutions,
and has the intrinsic
ability to study interactions
between
individual
cells. Products
other than
Ig, if produced
in similar
quantities,
could also be
assayed.
3000
2000’
0
0
C
1000
0
20000
40000
B
60000
80000
100000
120000
fractioncells/well
Figure 6. The mean amounts of IgG, IgE, and 1gM produced in
each well in the limiting dilution data graphed in Fig. 5. The total
amount of IgG,
dilution
IgE,
was divided
number ofB
and
1gM produced
by the number
by
all the wells at each
of wells and then plotted
vs. the
fraction cellsadded.
In conclusion,
the enhanced
ELISA
is sensitive
below the range of 10 pg of IgG, IgA, 1gM, and IgE in
a 100-tl
sample. The assay can be completed
within 8 h
and isreadilyreproducible.
With the use of other coating proteins
or antibodies,
the assay is easily adaptable
for the assay
of antigen-specific
antibody
of any isotype
or, for that matter, any protein for which high affinityspecific
antibodies
are available.
The production
of
Fab’ or F(ab’)2
alk-phos-coupled
reagents
may be
viewed as a limiting
factor in achieving
the same sensitivity
for the assay
of any protein,
but such antibodies
can be prepared
by commercial
concerns
from an investigator’s
unlabeled
reagent
if they are not already
available.
We have used this system to measure
antitetanus
toxoid and antidiphtheria
toxoid IgG and 1gM
produced
in vitro by a single cell as well as serum IgE
against
fetal calf sera with equivalent
results. This enhanced
ELISA
methodology
should be useful in providing sensitive
assays to measure
minute
quantities
of
biologically
active molecules
produced
in vitro, such as
interleukins,
as well as for measuring
levels of such proteins in body fluids.
E!I
This project was supported by U.S. Public Health Service grants
AI-07126, AI-15251, AI-15332, CA-12800, CA-23l75, and by American Cancer
Society
grant IM-482.
APPENDIX
20000
30000
60000
ELISA
procedure:
B fractIoncells/ well
1) Polystyrene
microtiter
plates, Immulon
2-MicroELISA
plates,
(Dynatech
Labs, Alexandria,
VA) are
coated with various Mab anti-human
Ig at 2 g/ml
in
borate-buffered
saline (BBS) (100 tl/well)
at room tem1gM production. 60 replicate wells were set up at each of the four
perature
in a humidified
box for from 2 h to overnight.
indicated
numbers
of cells per well (5,000, 10,000, 50,000, and
To measure
IgE, a combination
of Mab CIA-E-7.12
100,000). Each 220-tl well was split into 60-tlaliquots and assayed
(1 Lg/ml of each) is used (12). To assay
forlgG, IgE, and 1gM. Curve fitwas done by the maximum likeli- and CIA-E-4.15
hood method.
for IgG, Mab Hb 60 at 2 tg/ml
is used. To assay for
Figure
5. Limiting
dilution of peripheral
blood, spontaneous
Igproducing
lymphoblastoid
B cells. B fraction cells were set up in
5-day culturesin cRPMI and then assayed for total IgG, IgE, and
ENHANCED ELISA
3007
IgA, Mab CHE B6 at 2 sg/ml is used and for 1gM, Hb
57 at 2 ig/ml
is used. Mab Hb 60 and Hb 57 can be
obtained
from ATCC.
CHE
B6 was obtained
from
Dr.
F. Kearney
(University
of Alabama,
Birmingham) through
the ATCC. The Mab’s are grown as ascites in BALB C mice and enriched
by 50% ammonium
sulfate precipitation.
The coating antibodies
are kept in
BBS at a concentration
of 2 mg/ml and diluted
1:1000
before use. Protein
antigens,
such as tetanus
toxoid or
diphtheria toxoid, are used at 2 tg/ml in BBS. The
coating mixture
is used only once and then discarded.
2) After the coating
material
is discarded,
the wells
should
be washed
three
times
(300 tl/well)
with
phosphate-buffered
saline (PBS)/Tween
and then blocked
with 0.5% BSA in PBS/Tween
(150 1il/well) for 1/ h at
room temperature.
Use OA to block the plates and as
a sample
diluent
if the antibody
to be measured
is
directed
against
BSA or haptenated
BSA.
3) Discard
the blocking
solution
and then add either
J.
the
culture
supernatants
or
material
to
be
assayed
(diluted
tl/well.
with 0.5%
BSA) to a final volume
of 100
If less than 100 tl/well of culture supernatant
is
added, add 0.5% BSA to a total volume
of 100 tl/well.
Allow
the
plate
to incubate
in a humidified
box
least 2 h at room temperature.
4) Discard
the contents
of the wells. Wash
with PBS/Tween
(300 1il/well).Add
for
at
four times
100 tl/well of the
alkaline
phosphatase
(Alk-Phos)-coupled
developing
antibody
diluted
in 0.5% BSA. Rabbit
Fab Alk-Phosaffinity-purified
anti-IgE
from Dr. M. Kemeny
(Guy’s
Hospital,
London)
was used at a 1:3000 dilution.
The
goat Fab’2 Alk-Phos
anti-IgA
and anti-IgM
(Tago),
and anti-IgG
(Jackson
Immunoresearch,
Avondale,
Pa.) are used at one to two times
the working
dilution
provided
on the specification
sheet from the company.
This usually
works out to a 1:3000 dilution.
For the
data shown, all of the antisera
are diluted
1:3000. The
second antiserum
is incubated
at room temperature
in
a humidified
box for at least 2 h to overnight.
5) Discard
the second
antibody.
Wash three times
with PBS/Tween
using 300 tl/well.
Wash three times
with Tris/NaCl
using
300 tl/well.
At this point there are two options.
If the regular
(unenhanced)
development
system
is to be used, add
the regular
developing
(p-NPP)
solution
(100 id/well).
Develop
in the dark until the absorbance
at 405 nm of
the highest
concentrations
on the standard
curve
reaches
1 to 2. For the enhanced
development
system,
continue
with step 6.
6) Add
50 id/well
of substrate
solution
(see
later
If you make
equal
parts
watching
to pink
3008
Any nonspecifically
generated
NAD4
your own NADP
of substrate
and
bound
will cause
solution,
amplifier
Alk-Phos
or
signal
noise.
test it by adding
in a test
tube
and
for 15 mm for any sign of color development
or purple.
NADP
degrades
easily, and preVol. 2
Nov. 1988
in a plate
reader
until
the
absorbance
at 490
nm
in the
highest
concentration
wells is near 2. The wells are
typically saturated
in the IgE assay
with about 1 ng of
IgE. Complete
color development
generally
occurs
within 30 mm at 20#{176}C
or in an equivalent
period of
time to the incubation
casionally
it will take
allowed
with
the
substrate.
Oc-
slightly longer. If you make your
own amplifier,
it may change to a light pink on its own
within 1 day. This will cause a decrease
in the signal-tonoise ratio because of the increase in background
of the
assay;
therefore,
amplifier
solution
should
be made
fresh daily. The BRL amplifier
solution
is generally
more stable and may be stored reconstituted
according
to the
manufacturer’s
instructions.
Work
quickly
when
adding reagents
in steps 6 and 7. Keep the plates in the
same order, so that all the wells have exactly the same
amount
of time with all of the reagents
added.
Multichannel
pipettes
are required
to rapidly
add the substrate and developer.
8) Stop the assay, if desired,
with 20 1d/well of 0.75
M H2S04.
We have often found
it advantageous
to
serially
read the plate until the absorbance
of a predetermined
well on the standard
curve reaches a desired
point.
We used a BIO-TEK
EIA Autoreader
model
EL31O, which
has this capability, but other ELISA
readers also have this feature. Plate color, once stopped,
has been noted to be stable for at least 48 h if the plate
is kept from
drying
out.
The total time needed
is 2 h to coat the plate, 1/ h
to block, 2 h with the sample,
2 h with the second antibody, and about 1 h to develop. The total time needed
(exclusive
of washing)
is about
8 h. The assay is usually
done by precoating
the plates overnight,
which allows
the assay to be easily completed
in 1 day. It is theoretically possible
to store coated plates frozen,
but it has
been easier to store the coating
proteins
as concentrated
stock solutions
and coat plates as needed.
in-
structions)
and allow to incubate
at room temperature
for 35-45
mm. The longer the incubation,
the more
NAD
will be generated
and the greater
will be the signal produced
by the amplification
step. For this step the
wells should be uncovered.
This is the most critical step
in the assay.
spontaneously
weighed,
unopened
vials should be used fresh for every
assay.
If your test of the substrate
and developer
produces
spontaneous
color change,
it is possible to discard the substrate
in the wells and rewash four times
with Tris/NaC1
and repeat
step 6. We have had no
problems
with the BRL enhancing
kit reagents.
7) Add 50 id/well of the amplifier solution.
For the
remainder
of this step, the assay should
be carefully
watched.
If a large amount
of Ig is present
in some of
the wells, color development
will be noted quickly. Observe the plate for dark purple in the highest concentrations in the standard
curve, or scan every few minutes
Reagents
and
buffers:
The following reagents-buffers
are needed. Constituent
chemicals,
where
not specified,
were obtained
from
Sigma or an equivalent
supplier.
A) Coating
buffer to
dilute Mab anti-Igs
or other proteins,
BBS consisted
of
H3BO4
(mol wt 61.84),
10.3 g/liter;
NaCl
(mol wt
58.44), 7.3 g/liter:pH to 8.5. B) Substrate solutionuse as directed
in the BRL kit or make as follows. Tris-
The FASEB Journal
MACV El AL.
HC1 (mol wt 157.6), 11.82 g/liter; MgCl2 (4.9 M stock),
306 id/liter;
NaN3 (mol wt 65.0), 0.2 g/liter: pH to 8.8
at 20#{176}C.
Add 10 mg NADP
to each 40 ml just before
use; may use Sigma
preweighed
vials. This makes
enough
for eight 96-well plates by using 50 id/well.
C) Amplifier solution - use as directed in the BRL kit
or make as follows. Dissolve
1.2 ml ethyl alcohol (200
proof)
and 100 mg iodonitrotetrazolium
violet (INT)
(mol wt 505.7), Sigma grade 1, using a stirring
bar at
20#{176}C
for 10-15 mm in a total volume of 40 ml 0.025 M
NaPO4
buffer,
pH
7.0,
with
0.02%
NaN3.
After
the
material
is completely
dissolved,
add 1.2 mg diaphorase
(EC 1.6.4.3) (from microorganisms),
Boehringer
(Indianapolis,
md.) and 7.5 mg alcohol
dehydrogenase
(EC 1.1.1.1), Boehringer.
Swirl gently to dissolve
and
use within several hours. This is enough
for eight 96well plates by using 50 1d/well. D) Blocking
solution
and standard
diluent:
BSA (fraction
V) or OA, 5 g/liter,
in PBS/Tween.
E) Stop solution:
0.75 M H2S04.
F) PBS/Tween:
Tween 20, 0.5 mi/liter, in PBS. G) Final
wash -Tris/NaC1:
Tris base (mol wt 121.14), 6.04 g/liter;
NaCl (mol wt 58.44), 8.76 g/liter; NaN3 (mol wt 65.0),
0.5 g/liter: pH to 7.5.
H) 10 X PBS stock with azide
(PBS) - NaC1 (mol wt 58.44), 30.4 g/liter; NaH2PO4.
H20 (mol wt 137.99),
15.4 g/liter;
Na2HPO4.7H20
(mol wt 268.07),
77.3 g/liter; NaN3 (mol wt 65.0), 2.0
g/liter:
pH to 7.0. Dilute
10-fold before use as PBS.
I) Carbonate
buffer-NaHCO3
(mol wt 84.01),
8.4
g/liter; MgCl2 (4.9 M stock),
674 1d/liter;NaN3
(mol
wt 65.0), 0.1 g/liter: pH to 10.3. J) Regular
developing
solution - Dissolve
0.05 g of p-Nitrophenyl
phosphate
(mol wt 371) in 30 ml of carbonate
buffer just before
use. This is enough
for three plates with a 100-sal/well.
ENHANCED
ELISA
REFERENCES
J. G.; ISHIZAKA, K. Workshop on measureof in vitro IgE synthesis and regulation
of IgE
synthesis.
j Allergy C/in. Immunol.
77: 544-554;
1986.
2. WILLIAMSON,
A. R. Control of antibody formation:
certain uncertainties.].
C/in. Pathol. Suppi. R. Coil. Pathol.
32: 76-84; 1979.
3. MACY,
E.; STEVENS, R. H. A restricted
component
of
1. MASSICOT,
ment
T cell help
in pokeweed
peripheral
759;
mitogen-stimulated
J.
blood cell cultures.
human
Immunol.
124: 752-
1980.
N. S. Detection of plaqueblood of actively immunized humans.].
Immunol. 118: 1480-1482;
1977.
5. SEDGWICK,
J. D.; HOLT, P. G. A solid-phase immunoenzymatic
technique
for the enumeration
of specific
antibody-secreting
cells. J. Immunol.
Methods 57:
301-309;
1983.
6. DE ST. GROTH, S. F. The evaluation of limiting dilution
assays.].
Immunol. Methods 49: R11-R23;
1982.
7. SAXON,
A.; BARNErF,
E. Human
auto-anti
idiotypes
regulating T cell mediated reactivity to tetanus toxoid.
4.
Tuosor.i,
forming
J.
8.
Clin.
P. H.;
cells
in
Invest.
peripheral
73: 342-348;
A.;
HASSNER,
HARRIS,
the
SAXON,
A.
1984.
Isotype-specific
suppressor
T cells for IgE synthesis
activated
by IgE-Anti-IgE
immune complexes.
J. Immunol. 132: 2844-2849;1984.
9.
STANLEY,
C. J.;
amplification
sitivity
of
89-95;
10.
J0HANNs50N,
can enhance
immunoassays.
1985.
S.;
KANOWITH-KLEIN,
sion of Fc epsilon
mic
sue.
C/in.
F.;
HOFMAN,
receptors
and
fetal
and adult
press.
IgE on human
j
A.; SELF, C. J. Enzyme
both the speed and the sen].
Immunol. Methods 83:
Immunol.
In
SAXON,
surface
A. Expresand
lymphopoietic
cytoplastis-
Receivedfor publication May 16, 1988.
Accepted for publication June 20, 1988.
3009