Development of a highly sensitive ELISA for the determination of PBAN and its application to the analysis of hemolymph in Spodoptera littoralis.

Archives of Insect Biochemistry and Physiology 30:369-381 (1995)
Development of a Highly Sensitive ELISA
for the Determination of PBAN and Its
Application to the Analysis of Hemolymph
in Spodoptera littoralis
M.-Pilar Marco, Gemma Fabriis, and Francisco Camps
Department of Biological Organic chemistry, C.1.D.-C.S.I.C., Barcelona, Spain
A highly sensitive enzyme linked immunosorbent assay (ELISA) for the determination of the pheromone biosynthesis activating neuropeptide (PBAN) has been
developed. Six antisera have been obtained that recognize the carboxyl terminal side of this peptide. Two immunogens have been rationally designed and
synthesized in order to direct antibody specificity, using as haptens PBAN or
PBAN(20-33) with a Cys residue attached to their amino-terminal side. The
Cys thiol group has been used to covalently bind the peptide to keyhole limpet
hemocyanin (KLH) by using N-succinimidyl-4-(rnaleidimidomethyl)cyclohexane carboxylate (SMCC) as a convenient heterobifunctional cross-linker. Several usable competitive immunoassays have been obtained by synthesizing
eight different coating antigens and screening the sera against all of them.
The best assay was obtained with antibody 4 using Cys-Hez-PBAN(20-33)
coupled to bovine serum albumin (BSA) through the Lys groups by using
the homobifunctional cross-linker dimethylpimelidate dihydrochloride (DMP)
as the coating antigen. The optimized assay allows to detect PBAN at concentrations as l o w as 1 fmol/well (Ijo = 2.5 fmol/well). An extraction procedure for the hemolymph has been developed that allows t o perform PBAN
measurements in this tissue even after a tenfold dilution. In these conditions matrix effect is negligible. Preliminary results on the presence of PBANlike immunoreactivity (PBAN-IR) i n the hemolymph of Spodoptera littoralis
females are reported. o 1995 WiIey-Liss, Inc.
Key words: pheromone biosynthesis activating neuropeptide, enzyme linked immunosorbent assay, sex pheromone, Spodoptera littoralis
Acknowledgments: We thank Isabel Milldn for rearing the insects used in this study and CICYT
for financial support (grant AGF-95-0185). M.P.M. also thanks CSlC for a Contract as Research
Associate in a Programme of the Spanish Ministry of Education and Science.
Received February 14, 1995; accepted May 10, 1995.
Address reprint requests to Dr. Gemma Fabrihs, Department of Biological Organic Chemistry,
C.1.D.-C.S.I.C., Jorge Cirona, 18-26, 08034-Barcelona, Spain.
0 1995 Wiley-Liss, Inc.
370
Marco eta[.
INTRODUCTION
It is well known that in some species of Lepidoptera the production of the
sex pheromone is controlled by PBAN*, a 33 aminoacid peptide that has been
isolated and characterized from brains of Helicoverpa zea (Raina et al., 19891,
Bornbyx mori (Kitamura et al., 1989), and Lyrnantria dispar (Masler et al., 1994).
Research on PBAN biosynthesis, mode of action, and catabolism will open a
new field of potential chemical agents for pest control. Unfortunately, the
conclusions from the studies regarding PBAN target site and mode of action
are at present still controversial (Raina, 1993). Some authors have suggested
that PBAN, biosynthesized in the brain-subesophageal ganglion complex (BrSOG), is released into the hemolymph and acts directly on the pheromone
gland. Alternatively, other researchers have reported that the presence of an
intact ventral nerve cord (VNC) is necessary for PBAN to reach the terminal
abdominal ganglia (TAG), which is its target organ. Octopamine is then presumably released from TAG neurons that innervate the pheromone gland,
where it stimulates pheromone production.
Questions such as identification of PBAN in the blood of adult female moths
and localization and characterization of the receptors for PBAN would undoubtedly clarify important aspects of the physiological mode of action of
PBAN. Immunoassays are sensitive, specific, rapid, and inexpensive techniques that have found a wide application in several areas of science. Intensive research devoted over the past several years to the development of
immunoassays has often shown that appropriate immunogen design can control specificity and sensitivity of an immunoassay. However, in recent immunochemical studies, using a radioimmunoassay, Rafaeli et al. (1991) found
PBAN-IR in the brain-subesophageal ganglion complex (Br-SOG), corpora
cardiaca (CC), thoracic ganglia, and TAG of Heliothis arrnigera at selected times
of the photoperiod, but not in blood extracts during the sampling periods.
An ELISA has also been developed and used to quantify PBAN-IR in Heliothis peltigera (Gazit et al., 1992) and H. zea (Kingan et al., 1992). In the last
article, although PBAN-IR was found in Br-SOG and CC, amounts in the
*Abbreviations used: AKH = adipokinetic hormone; Anti@-AP = alkaline phosphatase-conjugated anti-rabbit IgG; AntilgG-HRP = horseradish peroxidase-conjugated anti-rabbit IgG; BrSOG = brain-subesophageal ganglion complex; BSA = bovine serum albumin; CC = corpora
cardiaca; CONA = conalbumin; Cys-Hez-PBAN = Hez-PBAN with a Cys aminoacid attached to
the N-terminal arninoacid; DEA = diethanolamine; DMF = dimethylformamide; D M P =
dimethylpirnelidate dihydrochloride; ECDI = 1-(3-dimethylaminopropyl)-3-ethyl carbodiimide;
ELISA = enzyme linked immunosorbent assay; HRP = horseradish peroxidase; KLH = keyhole
limpet hernocyanin; LPK = leukopyrokinin; OVA = ovalburnin; PBAN = pheromone biosynthesis
activating neuropeptide; PBAN-IR = PBAN-like imrnunoreactivity; PNP = p-nitrophenol; SMCC
= N-succinimidyl-4-(maleidimidomethyl)cyclohexane carboxylate; TAG = terminal abdominal
ganglia; TMB = tetramethylbenzidine; VNC = ventral nerve cord. Protein conjugates are abbreviated: BSA(DMP)(20-33) = Cys-Hez-PBAN(20-33) conjugated to BSA with DMP; BSA(ECD1)(20-33)
= Cys-Hez-PBAN(20-33) conjugated to BSA with ECDI; CONA(DMP)(20-33) = Cys-Hez-PBAN(2033) conjugated to CONA with DMP; CONA(ECD1)(20-33) = Cys-Hez-PBAN(20-33) conjugated to
C O N A w i t h ECDI; CONA(SMCC) = Cys-Hez-PBAN conjugated to C O N A w i t h SMCC;
CONA(SMCC)(20-33) = Cys-Hez-PBAN(20-33) conjugated to CONA with SMCC; KLH(SMCC) =
Cys-Hez-PBAN conjugated to K L H with SMCC; KLH(SMCQ(20-33) = Cys-Hez-PBAN(20-33) conjugated to KLH with SMCC; OVA(DMP)(20-33) = Cys-Hez-PBAN(20-33) conjugated to OVA
with DMP; OVA(ECD1)(20-33)= Cys-Hez-PBAN(20-33) conjugated to OVA with ECDI.
ELISA Development for PBAN
371
TAG and hemolymph were at or below the level of sensitivity of the assay.
The lack of PBAN-IR in the blood can be attributed to either an incorrect
timing of sampling, to a lack of sensitivity of the techniques employed, or
also because PBAN is not released into the blood in those cases. The aim of
the present paper has been (1) to develop a highly sensitive ELISA by rational preparation of suitable protein conjugates and (2) to demonstrate the applicability of the method to analyze insect hemolymph. Evidence for the
presence of PBAN-IR in the hemolymph of Spudoptera littoralis virgin females
is reported.
MATERIALS AND METHODS
Materials
Polystyrene microtiter plates (Maxisorb)were from NUNC (Roskilde, Denmark). Absorbances were measured using a Titertek Multiskan Plus ELISA
plate reader (Labsystems, Helsinki, Finland). Curve adjustments were performed with a commercial package (Genesis, Labsystems, Helsinki, Finland)
using a four parameter logistic equation.
Chemicals
Immunochemicals were purchased from Sigma Chemical Co. (St. Louis,
MO). Other chemical reagents were from Aldrich Chemical Co. (Milwaukee,
WI). Hez-PBAN, Hez-PBAN with a Cys aminoacid attached to the N-terminal aminoacid (Cys-Hez-PBAN), and Cys-Hez-PBAN(20-30) were prepared
by solid-phase synthesis at the Organic Chemistry Department (University
of Barcelona). Bom-PBAN was purchased from Peninsula laboratories
(Belmont, CAI, leucopyrokinin (LPK) and adipokinetic hormone I1 (AKH)
were obtained from Sigma Chemical Co. Oxidized Hez-PBAN was prepared
as reported (Kitamura et al., 1989).
Buffers
PBS is 0.2 M phosphate buffer, 0.8% NaC1, pH 7.5. Borate buffer is 0.2 M
boric acid-sodium borate, pH 8.7. Coating buffer is 0.5 M carbonate-bicarbonate buffer, pH 9.6. PBST is 0.05% Tween 20 in PBS. DEA buffer is 10%
diethanolamine (DEA) buffer, pH 9.8. Citrate buffer is a 0.1 M solution of
sodium citrate, pH 5.5.
Insects
S. littoralis larvae were reared on a wheat germ diet at 25 f 1"C, with a
1ight:dark cycle of 16h:8h. Sexes were separated as pupae, which were then
transferred to a reversed photoperiod chamber with the same photoregime.
Adults were segregated daily before the onset of the scotophase and fed with
a 5% sucrose solution.
Preparation of the Protein Conjugates
SMCC-conjugates. To prepare the immunogens, Cys-Hez-PBAN (6.6 mg,
1.6 pmol) or Cys-Hez-PBAN(20-33) (3.4 mg, 1.6 pmol) were coupled to KLH
(10 mg, 3.3 nmol, 4.7 pmol Lys) using SMCC (53 pL, 1.6 pmol, 10 mg/mL in
372
Marco et al.
dimethylformamide [DMF])as described (Van Regenmortel et al., 1988).The
conjugates were purified on a Sephadex G-50 column, lyophilized and stored
dry at -80°C. The same procedure was applied to prepare the SMCC-conjugates for coating from conalbumin (CONA) (1 mg, 13 nmol), SMCC (10 pL,
0.29 pmol, 10 mg/mL in DMF), and Cys-Hez-PBAN (1.17 mg, 0.29 pmol) or
Cys-Hez-PBAN(20-33) (0.60 mg, 0.29 pmol). The solutions were dialyzed extensively against 0.01 M PBS (4 x 5 L) and milliQ water (5 L) at 4"C, lyophilized and stored at -80°C.
ECDI-conjugates. Cys-Hez-PBAN(20-33) (1.2 mg, 0.56 pmol, 1.16 pmol
Lys, 1 mg/mL in borate buffer), previously treated with citraconic anhydride
(7.8 mg, 69.6 pmol, 100 mg/mL in borate buffer) to protect the peptide Lys
amino groups, was reacted with l-(3-dimethylaminopropyl)-3-ethyl
cabodiimide
(ECDI) (33.3pmol, 6.3 mg, freshly prepared solution 20 mg/mL in borate buffer)
and the resulting adduct was then coupled to 2 mg of protein, BSA (29 nmol),
CONA (26 nmol), or ovalbumin (OVA) (25 nmol), following standard procedures (Van Regenmortel et al., 1988). The conjugates were finally dialyzed
for 3.5 h at room temperature against 1.5 L of 5% acetic acid and extensively
against 0.01 M PBS (4 x 5 L) and milliQ water (5 L) at 4"C, lyophilized and
stored at -80°C.
DMP-conjugates. A freshly prepared solution of DMP (497 pg, 1.92 pmol,
1 mg/mL of borate buffer) was added to a solution of Cys-Hez-PBAN(20-33)
(1 mg, 0.48 pmol) and the protein, BSA (1.4 mg, 21 nmol), CONA (1.7 mg, 22
nmol), or OVA (1 mg, 22 nmol), in 0.8 mL of borate buffer. The mixture was
stirred for 4 h at room temperature and overnight at 4°C. The solution was
extensively dialyzed against PBS 0.01 M (4 x 5 L) and milliQ water (2 x 5 L)
at 4"C, lyophilized and stored at -80°C.
Immunization Protocol and Collection of the Antisera
Six female white New Zealand rabbits (2-3 kg) were used to raise antibodies using Cys-Hez-PBAN(20-33)conjugated to KLH with SMCC [KLH(SMCC)(2033)l (rabbits 1, 2, and 3) and Cys-Hez-PBAN conjugated to KLH with SMCC
[KLH(SMCC)I (rabbits 4, 5 and 6) as immunogens. The immunogens were
dissolved in 0.2 M PBS, mixed 1:l with Freund's complete adjuvant, and 300
pg were injected intradermally at 10 sites on the back of the animals. Booster
injections were started 1 month after the initial immunization, and repeated
every 4-5 weeks, with 150 pg of the same conjugate solutions mixed 1:l with
Freund's incomplete adjuvant. Animals were bled 10 days after each boost (3
to 5 mL) from the marginal ear vein. Ten days after the last boost the rabbits
were exsanguinated under deep anesthesia, the blood was allowed to clot
and the clear serum was collected, mixed with NaN3 (final concentration
0.02%),and stored in aliquots at -80°C.
Antibody Titer Test
Plates were coated with Cys-Hez-PBAN(20-33)conjugated to CONA with
SMCC [CONA(SMCC)(20-33)](2.5 pg/mL in coating buffer, 100 pL/well),
sealed with adhesive plate sealers, and incubated overnight at 4°C. The following day the plates were washed 5 times with PBST and serum from rabbits 1 to 6, at 8 different dilutions (1/1,000 to 1/64,000 and 0 in PBST), was
ELISA Development for PBAN
373
added to the wells (100 pL/well) and incubated for 1 h at room temperature.
Plates were washed again 5 times with PBST and alkaline phosphatase-conjugated anti-rabbit IgG (antiIgG-AP) was added (115,000 in PBST, 100 pL/
well) and incubated for 1 h more at room temperature. The plates were
washed again, and the substrate (p-nitrophenol phosphate (PNP), 1 mg/mL
in DEA buffer, 100 pL/well) was added and incubated for 30 min at room
temperature. Absorbance values were read at 405 nm.
Screening of the Antisera and Coating Antigens
A two-dimensional titration protocol was used for the screening and
determination of the optimum concentrations of both coating antigens and
antisera to be used later in the competitive experiments. Following the
procedure described for titre analysis, CONA(SMCC)(20-33), Cys-HezPBAN conjugated to CONA with SMCC [CONA(SMCC)I, Cys-HezPBAN(20-33) conjugated to BSA [BSA(DMP)(20-33)1, CONA [CONA
(DMP)(20-33)], or OVA [OVA(DMP)(20-33)] with DMP and Cys-HezPBAN(20-33) conjugated to BSA [BSA(ECDI)(20-33)1, CONA [CONA
(ECDI)(20-33)],or OVA [OVA(ECDI)(20-33)1with ECDI were used to coat
plates at 6 different concentrations (from 6 to 0.07 pg/mL and 0 in coating buffer, 100 pL/well). The plates were washed the next day and the
antiserum from rabbits 1 to 6 was checked at different dilutions (1/1,000
to 1/64,000 and 0 in PBST) against the coating antigens. Plates were then
developed as previously described. Optimal conditions were chosen to
produce absorbances around 0.6 U in 30 min. Screening of the competitive experiments was performed as reported (Marco et al., 1993).
Optimized ELISA Protocol: Competitive Experiment
Microtiter plates were coated with BSA(DMPl(20-33)(0.3 pg/mL in coating buffer, 50 pL/well) overnight at 4°C. The following day, the plates were
washed 5 times with PBST and blocked with 3% nonfat milk for 1 h at room
temperature. The plates were washed again and the samples and standards
(eight different concentrations from 10,000 to 3.2 pM and 0), which had been
preincubated with the antibody 4 (1/25,000 diluted in PBST) overnight at
4"C, were added to the wells (50 pL/well) and incubated for 1 h more at
room temperature. After another washing step horseradish peroxidase-conjugated anti-rabbit IgG (antiIgG-HRP) (116,000 in PBST, 50 pL/well) was
added and incubated for an additional hour at room temperature. Plates were
washed and a solution of the substrate (tetramethylbenzidine [TMBI 0.01%,
H2020.004% in citrate buffer) was added (50 pL/well). The enzyme reaction
was stopped after 30 min at room temperature with 25 pL of 4 M H2S04and
the absorbances were read at 450 nm.
Cross-Reactivity Studies
Standard curves of Cys-Hez-PBAN, Cys-Hez-PBAN(20-33), oxidized HezPBAN, Bom-PBAN, LPK, and AKH (10 pM to 0.01 pM) were prepared in
PBST and used for ELISA determination to study the specificity of the assay.
Cross-reactivity values were calculated as follows: (IsoHez-PBAN /I5opeptide) x 100.
374
Marco et al.
Hemolymph Collection and Sample Preparation
Insects were decapitated and placed into a 500 pL Eppendorf tube with the
bottom tip severed, which was inserted into a 1.5 mL Eppendorf tube. The
Eppendorf tube was centrifuged at 2,000 rpm for 5 min and obtained 10 to 20
pL of clear hemolymph per female. Blood (20 pL) was collected with a 10 pL
microcapillary glass pipette, diluted with ethano1:water (5:4,180 pL), the mixture was vortexed for about 5 s and then centrifuged at 12,000 rpm for 5 min.
The clear supernatant (150-160 pL) was collected and dried down in a SpeedVac rotatory evaporator. Dry samples were stored at -80°C for ELISA analyses. Prior to ELISA determination, samples were redissolved in the same
volume of the solution of antibody (1/25,000 diluted in PBST). Hemolymph
was taken from 2-day-old adult virgin females either 2 h before or 2 h after
the onset of the scotophase. One group of animals were decapitated 2 h before lights off and hemolymph was collected 2 h into darkness. Hemolymph
extracts were prepared from one to two individuals.
PBAN Recovery Studies
Photophase hemolymph samples (20 pL) were diluted with ethano1:water
(5:4, 180 pL) containing Hez-PBAN to a final concentration of 100 pM, processed as described above, and used for ELISA determination. Hemolymph
extracts without Hez-PBAN and Hez-PBAN solutions in the absence of
hemolymph were also run simultaneously.
Matrix Effect Studies
Second-photophasehemolymph extracts, prepared as described above, were
serially diluted with PBST, spiked with Hez-PBAN to a final concentration of
100 pM, and measured by ELISA to determine the optimum blood dilution
factor. Then, Hez-PBAN standard curves were prepared in PBST and 1/10
PBST diluted hemolymph to determine the parallelism of the curves.
RESULTS
Immunoassay Development
As outlined in the discussion, two different immunogens were prepared to
obtain antibodies directed to the C-terminal part of PBAN. The first immunogen, KLH(SMCC),was synthesized by coupling Hez-PBAN, with an extra
Cys residue attached to the N-terminal aminoacid, to the carrier protein using SMCC as cross-linker. SMCC reacts with the Lys amino acids of the carrier protein through the activated acid moiety and, in a second step, it reacts
specifically with the Cys residue of the peptide through the maleidimido
group. A second immunogen was prepared following the same methodology using Cys-Hez-PBAN(20-33) as hapten. Because of the enhanced immunogenicity of KLH conjugated PBAN, maximum antibody titers were already
obtained after the second boost, and they remained almost constant after subsequent boostings.
The development of a two-step competitive ELISA based on the coating
antigen format was accomplished by screening a library of coating antigens
ELISA Development for PBAN
375
and searching for a suitable competitor. For the preparation of these coating
antigens, we used the less expensive hapten, Cys-Hez-PBAN(20-33),BSA,
CONA, or OVA as carrier proteins and SMCC, ECDI, or DMP as cross-linkers. The titration experiments performed with all these coating antigens (Table
1) revealed that acceptably high antibody titers were obtained with both immunogens, allowing a considerable dilution of the antisera. Only the protein
conjugates prepared using ECDI gave slightly lower antibody titers.
Optimal antibody dilution and coating antigen concentration with all the
possible antibody/protein conjugate combinations were determined by a twodimensional titration experiment (Table 1).A tiered system was used starting
with a broad screening and finally resulting in one antibody/coating antigen
combination for full evaluation. Most of the assays showed moderate sensitivity and some of them presented high slope values, indicating significant
antibody affinity. Among the coating antigens prepared with ECDI, only
OVA(ECDI)(20-33)gave competitive assays with two of the antisera, although maximal absorbances were very low (data not shown). Although
CONA(SMCC)(20-33)gave competitive assays with three of the antisera tested,
the maximal absorbance/background signal ratios were never satisfactory
(data not shown). In contrast, DMP coating antigens gave the most competitive assays (Table 2). Among the different combinations, CONA(DMP)(20-33)
and BSA(DMP)(20-33)and antiserum 4 showed the best slopes, 15,, values,
and maximal absorbance/background signal ratios. Consequently, we focused
our attention on the optimization of the ELISA with these immunoreactants.
Blocking of the plates after the coating step along with preincubation of the
antibody with Hez-PBAN before the competition step, improved the maximal absorbance/background signal ratio and sensitivity in both assays. However, the best values were obtained with BSA(DMP)(20-33),which was selected
for routine use. Finally, the optimized assay, whose features are specified in
the four parameter equation in Figure 1, allows the measurement of PBAN
in a concentration range between 20 to 500 pM with an 150= 53 k 12 pM (2.6
k 0.6 fmol/well, n = 5).
Cross-reactivity studies were conducted to determine the specificity of the
assay by monitoring recognition by the antibody of a series of peptides structurally similar to PBAN. As shown in Table 3, similar binding affinities were
observed for both Bom-PBAN and Hez-PBAN, which makes the assay suitable for the analysis of PBAN from several sources. Obviously, haptens used
for immunization were detected by the antibody with a higher sensitivity.
Oxidized Hez-PBAN was also recognized by the same assay with a crossreactivity value of 64% and the octapeptide LPK exhibited a 27% cross-reactivity. It is important to note that the slope is slightly lower in the assay for
LPK, which indicates a decreased affinity of the antibodies to this peptide.
Evidently, AKH, a peptide with a completely different aminoacid sequence,
was not recognized by the antibody.
A series of experiments were performed to evaluate the effectiveness and reliability of the developed ELISA to determine PBAN in the insect hemolymph,
following our optimized extraction procedure. First, photophase hemolymph
extracts were serially diluted with PBST and spiked with a known amount of
Hez-PBAN. These experiments showed that the lowest dilution factor that
M
L
L
M
M
M
M
L
M
M
L
M
H
H
H
H
H
H
H
H
H
H
M
H
1
2
3
4
5
6
L
L
L
L
M
L
n.t.
H
M
M
H
M
M
n.t.
L
L
M
M
M
n.t.
H
H
H
H
*H, M, and L indicate the serum dilution range that gave absorbances of 0.6-0.7 after 30 min. H, more than 1/32,000; M, between 1/32,000 and 1 /
8,000; L, less than 1/8,000. n.t., not tested. The concentration of the coating antigens selected was that which ensured an optimum coverage of the
solid phase, ranging from 0.1 to 1 pg/mL.
KLH(SMCC)
KLH(SMCC)
(20-33)
Immunogen
Coating antigen
CONA
CONA
BSA
CONA
OVA
BSA
CONA
OVA
Antisera (SMCC) (SMCC)(20-33) (ECDI)(20-33) (ECDI)(20-33) (ECDI)(20-33) (DMP)(20-33) (DMP)(20-33) (DMP)(20-33)
TABLE 1. Titer of Antisera Measured in the Presence of Different Coating Antigens*
ELISA Development for PBAN
377
TABLE 2. Features of the Most Competitive Assays Obtained During the Screening of the
Different AntibodylCoating Antigen Combinations*
Coating antigen
BSA(DMP)(20-33)
CONA(DMP)(20-33)
OVA(DMP)(20-33)
Antiserum,
dilution factor
A/D
Is,, (pM)
1,1/20,000
3,l /80,000
4,1/25,000
5,1/25,000
6, 1 /80,000
4,1/25,000
1,1/10,000
4,1/5,000
5,1/10,000
6,1/20,000
2.3
3.3
4.8
5.5
2.1
3.2
2.1
2.5
2.1
2.5
296
6429
223
389
1138
217
2790
551
2635
1737
Slope
A max
0.6
0.4
1.o
0.5
0.9
0.8
0.4
1.4
0.4
0.8
0.801
0.197
0.266
0.331
0.263
0.751
0.551
0.398
0.731
0.538
r
0.965
0.936
0.968
0.987
0.925
0.947
0.954
0.962
0.963
0.851
*Maximal absorbance (A), slope (B), Is0 (C), and minimal absorbance (D) are the values of the
four parameter equation calculated for each assay: y = (A-D)/[I+(X/C)~]+D.Three well replicates were used for each concentration of PBAN in the standard curves.
gave no matrix effect was 1/10. In a second set of experiments, serial dilutions of Hez-PBAN were prepared in both a hemolymph extract diluted 1/
10 and PBST. The sigmoid curves obtained for this peptide in both cases
showed identical slope and sensitivity (Fig. 1). Finally, photophase hemolymph
extracts were prepared in the presence of known amounts of Hez-PBAN to
determine the recovery of this peptide, which was 82 16% (n = 7). It is
worth noting that no protease inhibitors had to be added to the extraction
solvent, as long as the samples were rapidly processed.
Determination of PBAN-IR in the Hemolymph
In order to demonstrate the applicability of this ELISA, we determined
PBAN-IR in S. littoralis hemolymph (Table 4). Hemolymph samples taken 1 h
0.2
Fig. 1. Standard curve for Hez-PBAN performed in PBST buffer (filled circles) or hemolymph diluted 1/10 in PBST buffer (open circles) showing the absence of matrix effect. Analysis was performed by the two-step competitive ELISA using 8 serial dilutions, in duplicate, of synthetic Hez-PBAN,
ranging from 0.1 KM to 0.01 pM. Four parameter logistic equation of an optimized routine assay is
as follows: y = 0.495/[1 + (~/48.7)'.*~]
+ 0.044. For general equation, see Table 2.
378
Marco eta!.
TABLE 3. Cross-Reactivity of PBAN Antiserum With Selected Peptides,
Peptide
Hez-PBAN
Cys-Hez-PBAN
Cys-Hez-PBAN(20-33)
Oxidized Hez-PBAN
Bom-PBAN
LPK
AKH
150(pM)
Slope
% Cross-reactivity
68
59
46
106
72
252
n.d.
1.1
1.2
0.9
1.1
1.2
0.7
100
115
148
64
95
27
n.c.
*Cross-reactivitydata were obtained by preparing standard curves in PBST. Two or three well
replicates were used for each concentration (10 pM to 0.01 pM) of the peptides. Is0 and slope
values were obtained from the four parameter equation of every standard curve. Cross reactivity values were calculated according to the equation: (I50H ~ Z - P B A N / peptide)
I~~
x 100.
n.d., not detected at concentrations below 10 pM; n.c., no cross-reactivity.
before lights off were always below the detection limit of the assay (200 pM
when measuring 1/10 diluted hemolymph). Conversely, all the samples taken
2 h into the dark period could be measured with concentration values around
600 pM. In contrast, concentrations of PBAN-IR in all hemolymph samples
taken 2 h into the scotophase from females that had been decapitated 2 h
before darkness were below the detection limit of the assay.
DISCUSSION
Since small amounts of PBAN can elicit a pheromonotropic response, physiological studies about PBAN mode of action rely on the availability of an
analytical tool able to detect trace amounts of this peptide. Immunoassays
are rapid, selective, and very sensitive analytical techniques that have proven
to be useful in several areas of pharmacology, toxicology, etc. In contrast with
other analytical procedures, small sample sizes are sufficient to perform quantification and clean-up procedures can often be avoided. However, the existence of PBAN-IR in the blood could not be confirmed with either of the three
immunoassays previously reported (Gazit et al., 1992; Kingan et al., 1992;
Rafaeli et al., 1991), probably because they lacked the required sensitivity.
The first aim of this paper was the development of a highly sensitive ELISA
TABLE 4. PBAN-IR of Hemolymph Extracts of S. littoralis Virgin Females*
[PBAN-IR]"
Scotophase
Photophase
<200 (7)
Intact
Decauitated
*
665 89 (20)
<200 (8)
*Extraction of hemolymph and preparation of samples were carried out as described in Materials and Methods. Analyses were performed by the two-step competitive ELISA using 8 serial dilutions, in duplicate, of synthetic Hez-PBAN, ranging from 10 nM to 3.2 pM. A dilution
of hemolymph extract of 1/10 was used, in duplicate.
"Data are expressed in pM and are the mean k SEM of the number of replicates given within
parentheses.
ELISA Development for PBAN
379
by rational preparation of suitable protein conjugates to be used as both immunogens and coating antigens.
One of the most important stages in immunoassay development is the design of appropriate immunogens to obtain a suitable antiserum. In order to
enhance immunogenicity, small molecules, which usually do not stimulate
immunological response, are attached to a carrier protein. The position of
covalent coupling of the hapten to the carrier protein has a strong influence
on the selectivity and sensitivity of the immunoassay. The spacer arm used
as cross-linker is also important because, due to the shielding effect of the carrier
macromolecule, it directs antibody specificity so that the immunodominant part
of the molecule is that situated farthest from the attachment point. In the
development of this ELISA, the immunogens were designed taking into account that the pheromonotropic response may not be prompted by the complete 33 amino acid PBAN, as smaller peptides with the C-terminal sequence
of PBAN are also active (Fonagy et al., 1992; Gazit et al., 1990; Raina and
Kempe,1990). Therefore, we directed antibody specificity towards the C-terminal part of PBAN, so that any pheromonotropic peptide present in the
blood could be detected. With the above considerations in mind, this was
achieved by covalently attaching the carrier protein, KLH, to the N-terminal
side of the peptide so that its C-terminus would be highly exposed for antibody recognition. Coupling of KLH to the N-terminal portion of Hez-PBAN
was accomplished through an extra Cys residue, bound to the N-terminal
amino acid of PBAN, which specifically reacted with the sulfhydryl directed
moiety of the cross-linker, SMCC. Although in most of the immunoassays
the entire target structure is included in the immunizing hapten, a characteristic fragment can often be sufficient to generate antibodies with high affinity
to bind the whole compound. Such an approach is very useful if the target
substance is unstable, toxic, difficult, or expensive to obtain. Therefore, we
prepared a second immunogen using a 15 amino acid peptide with the Cterminal sequence of Hez-PBAN, functionalized with an extra Cys residue
bound to the N-terminal amino acid.
The development of competitive immunoassays require the preparation of
suitable coating antigens or enzyme tracers, depending on the format chosen. Since adsorption of peptides to the plates can lead to a lack of reproducibility of the assay due to their small size (Van Regenmortel et al., 1988), in
our coating antigen ELISA we used a protein conjugate as competitor, in contrast to the ELISAs previously described (Gazit et al., 1992; Kingan et al.,
1992). From the screening experiments we could corroborate that heterologous ELISAs give the most competitive experiments (Harrison et al., 1991).
In heterologous ELISA systems recognition of the coating antigen is weaker
than that of the target compound because a different protein, hapten, coupling position, or procedure from that employed in the preparation of the
immunogen is used in its preparation. As DMP reacts with Lys y-amino groups
under mild conditions with a high degree of specificity, in the DMP-conjugates covalent coupling probably occurred mainly through the. 27Lysamino
acid of Hez-PBAN(20-33), near to the C-terminal amino acid, thereby providing, besides protein and bridge, also site heterology affording the most effective competitive assays. In contrast, ECDI-conjugates did not always provide
380
Marco et al.
site heterology, whereas SMCC coating antigens did not fulfill either site or
bridge heterology, resulting in more deficient assays.
The optimized ELISA herein described can detect trace amounts of PBAN,
with a detection limit of 1 fmol/well. As we had planned, the assay is aimed
at the recognition of the C-terminal part of the peptide, as concluded from
the fact that LPK still cross-reacted with the antiserum. Although only 27%
of cross-reactivity was observed for this fragment, recognition (Iso = 252 pM,
12.6 fmol/well) is still higher than that for PBAN in the ELISA previously
reported (Gazit et al., 1992).The lack of specificity of this assay is convenient
in that PBAN and also shorter fragments derived from it, which could be
responsible for the pheromonotropic activity, are detected. However, it should
be mentioned that the assay might also detect fragments that are not active,
since Raina and Kempe (1990) reported that PBAN(25-33)and PBAN(29-33)
are much less active than PBAN at a dose of 10 pmol in H . zea. On the other
hand, the fact that the myotropic peptides of the FXPRL-NH2family are also
recognized raises the need for confirmatory methods to ensure the identity
of the peptide responsible for that immunoreactivity.
Using this ELISA, the presence of PBAN-IR in hemolymph of S. littoralis
virgin females was detected during the scotophase, at the time of pheromone
production. In contrast, no PBAN-IR was found in blood samples taken during the photophase, when only very low amounts of pheromone are present
in the gland. Likewise, no PBAN-IR was detected in blood samples taken
during the scotophase from females that had been decapitated during the
light period. This last result agrees with the observation that decapitation
during the photophase abolished normal sex pheromone production, probably because PBAN had not yet been released into the circulatory system
(Martinez and Camps, 1988). These results suggest that a pheromonotropic
peptide, either PBAN or a shorter fragment with the same C-terminal sequence, is present in the blood of S. littoralis virgin females at the time of
pheromone production. Since five peptides with almost identical terminal
sequences are coded by the PBAN gene (Ma et al., 19941, it is possible that
the five peptides are released into the blood and detected by this ELISA.
However, given the lack of specificity of this assay, we cannot disregard the
possibility that an immunoreactive peptide without pheromonotropic activity is coincidentally present in the hemolymph at the time of pheromone production. Future studies will be conducted to clarify these points.
In summary, the rational design of immunogens and coating antigens has
led us to the development of a very sensitive ELISA for PBAN, with a detection limit of 1 fmol/well. The assay has been characterized and we have demonstrated its utility to determine PBAN-IR in the hemolymph. An extraction
procedure has been developed that requires little or almost no sample cleanup to analyze the samples, thus diminishing the probabilities of the peptide
to be enzymatically or chemically degraded prior to the analysis. This extraction procedure affords high PBAN recoveries and reliable measurements with
samples diluted 1/10. The application of this ELISA to S. littoralis virgin females indicates the presence of PBAN-IR in the hemolymph of pheromoneproducing insects. Further determinations aimed at clarifying the physiological
mode of action of PBAN in this species will be reported elsewhere.
ELISA Development for PBAN
381
LITERATURE CITED
Fonagy A, Schoofs L, Matsumoto S, De Loof A, Mitsui T (1992): Functional cross-reactivities
of some locustamyotropins and Bornbyx pheromone biosynthesis activating neuropeptide. J
Insect Physiol38:651-657.
Gazit Y, Dunkelblum E, Benichis M, Altstein M (1990): Effect of synthetic PBAN and derived
peptides on sex pheromone biosynthesis in Heliothis peltigeru (Lepidoptera: Noctuidae). Insect Biochem 20:853-858.
Gazit Y, Dunkelblum E, Ben-Aziz 0,Altstein M (1992): Immunochemical and biological analysis
of pheromone biosynthesis activating neuropeptide in Heliothis peltigeru. Arch Insect Biochem Physiol19:247-260.
Harrison RO, Goodrow MH, Hammock BD (1991): Competitive inhibition ELISA for the striazine herbicide: Assay optimization and antibody characterization. J Agric Food Chem
39:122-128.
Kingan TG, Blackburn MB, Raina AK (1992): The distribution of PBAN immunoreactivity in
the central nervous system of the corn earworm, Helicovevpu zeu. Cell Tissue Res 270:229-240.
Kitamura A, Nagasawa H, Kataoka H, Inoue T, Matsumoto S, Ando T, Suzuki A (1989): Amino
acid sequence of pheromone-biosynthesis-activating neuropeptide (PBAN) of the silkworm,
Bornbyx mori. Biochem Biophys Res Commun 163:520-526.
Ma P, Knipple DC, Roelofs WL (1994):Structural organization of the Helicoverpu zeu gene encoding the precursor protein for pheromone biosynthesis-activating neuropeptide and other
neuropeptides. Proc Natl Acad Sci USA 91:6506-6510.
Marco M-P, Hammock BD, Kurth MJ (1993): Hapten design and development of an ELISA
(enzyme-linked immunosorbent assay) for the detection of the mercapturic acid conjugates
of naphthalene. J Org Chem 58:7548-7556.
Martinez T, Camps F (1988): Stimulation of sex pheromone production by head extract in
Spodopteru littoralis at different times of the photoperiod. Arch Insect Biochem Physiol
9:211-220.
Masler EP, Raina AK, Wagner RM, Kochansky JP (1994): Isolation and identification of a
pheromonotropic neuropeptide from the brain-subesophageal ganglion complex of Lyrnantriu
dispur: A new member of the PBAN family. Insect Biochem Mol Biol24:829-836.
Rafaeli A, Hirsch J, Soroker V, Kamensky B, Raina AK (1991): Spatial and temporal distribution of PBAN in Helicoverpu (Heliothis) urmigeru using RIA and in vitro bioassay. Arch Insect
Biochem Physiol18:119-129.
Raina AK (1993): Neuroendocrine control of sex pheromone biosynthesis in lepidoptera. Annu
Rev Entomol38:329-349.
Raina A, Kempe T (1990): A pentapeptide of the C-terminal sequence of PBAN with
pheromonotropic activity. Insect Biochem 202349-851.
Raina AK, Jaffe H, Kempe TG, Keim P, Blacher RW, Fales HM, Riley CT, Klun JA, Ridgway
RL, Hayes DK (1989): Identification of a neuropeptide hormone that regulates sex pheromone production in female moths. Science 244:796-798.
Van Regenmortel MHV, Briand JP, Muller S, Plau6 S (1988): Synthetic polypeptides as antigens. In Burdon RH, PH Van Knippenberg (ed): Laboratory Techniques in Biochemistry
and Molecular Biology, vol. 19. Amsterdam: Elsevier Science, pp 95-129.