Hydrophobic Moiety of Capsaicinoids Haptens Enhancing Antibody Performance in Immunoassay: Evidence from Computational Chemistry and Molecular Recognition

Article abstract of DOI:10.1021/acs.jafc.1c03657

We previously found that the immune response to haptens is positively correlated with molecular hydrophobicity. The antibodies used in immunoassays for capsaicinoids (CPCs) in waste oil suffer from low affinity and loose recognition to structural analogues. To address this issue, four new haptens (hapten1-4), maximally exposing the hydrophobic alkane chain (noncommon moiety of CPCs), were designed and expected to produce antibodies with high affinity and accurate recognition to CPCs based upon our findings. The assumption was first evidenced by computational chemistry and animal immunization successively. Compared with four reported haptens (hapten5-8) that expose the hydrophilic vanillyl amide moiety (common structure of CPCs and other vanillin alkaloids), antisera from hapten1-4 showed an approximately 1000-fold increase in affinity and significantly improved recognition profiles for CPCs. The molecular recognition study showed that the high affinity of the antibody from new haptens mainly originated from hydrophobic forces. An indirect competitive enzyme-linked immunosorbent assay based on a monoclonal antibody from hapten1 was developed and exhibited limits of detection as low as 0.73-3.29 μg/kg for four CPCs in oils and with insignificant cross-reactivities for other eight vanillin alkaloids, which have been never achieved in previous reports.

Full text of DOI:10.1021/acs.jafc.1c03657

pubs.acs.org/JAFC
Article
Hydrophobic Moiety of Capsaicinoids Haptens Enhancing Antibody
Performance in Immunoassay: Evidence from Computational
Chemistry and Molecular Recognition
Yuchen Bai, Hui Jiang, Yingjie Zhang, Leina Dou, Minggang Liu, Wenbo Yu, Kai Wen, Jianzhong Shen,
Yuebin Ke, Xuezhi Yu, and Zhanhui Wang*
Cite This: J. Agric. Food Chem. 2021, 69, 9957−9967
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: We previously found that the immune response to haptens is positively correlated with molecular hydrophobicity.
The antibodies used in immunoassays for capsaicinoids (CPCs) in waste oil suffer from low affinity and loose recognition to
structural analogues. To address this issue, four new haptens (hapten1−4), maximally exposing the hydrophobic alkane chain
(noncommon moiety of CPCs), were designed and expected to produce antibodies with high affinity and accurate recognition to
CPCs based upon our findings. The assumption was first evidenced by computational chemistry and animal immunization
successively. Compared with four reported haptens (hapten5−8) that expose the hydrophilic vanillyl amide moiety (common
structure of CPCs and other vanillin alkaloids), antisera from hapten1−4 showed an approximately 1000-fold increase in affinity and
significantly improved recognition profiles for CPCs. The molecular recognition study showed that the high affinity of the antibody
from new haptens mainly originated from hydrophobic forces. An indirect competitive enzyme-linked immunosorbent assay based
on a monoclonal antibody from hapten1 was developed and exhibited limits of detection as low as 0.73−3.29 μg/kg for four CPCs in
oils and with insignificant cross-reactivities for other eight vanillin alkaloids, which have been never achieved in previous reports.
KEYWORDS: hydrophobicity, hapten design, capsaicinoids, computational chemistry, molecular recognition
findings provided a clear strategy to design haptens for
enhancing the resulting antibody; that is, the hydrophobic
moiety of the target molecule should be maximally exposed as
much as possible.
INTRODUCTION
■
Immunoassay, as a straightforward and cost-effective detection
modality, has gradually spread for use in the detection of
analytes with low molecular weight in clinical, environmental,
and food samples.1,2 Antibodies play a considerably important
role as the key reagent in immunoassay development, which
directly determinates the sensitivity, specificity, and stability of
subsequent assays. According to previous studies, the structure
and properties of haptens have been regarded as the most
important factors in the generation of a desired antibody in
terms of affinity and specificity, because the shape of an
antibody-binding pocket is originally derived from the
immunogenic hapten.3−5 However, little is known about how
the molecular structure and properties themselves influence
the strength of the immune response and then the performance
of the antibody.
Hydrophobicity is one of the most important and frequently
examined physical and chemical properties of molecules.
Research studies on the immunogenicity and haptenic
character of poly(ethylene glycol) (PEG) and other nano-
particles have provided evidence for a directly proportional
relationship between hydrophobicity and immunogenicity.6−8
Recently, we designed and synthesized two groups of hapten
molecules with various hydrophobicity values to investigate the
relationship between the properties of the small molecules and
the antibody response in terms of titer and affinity.5 The
magnitude of the antibody response was found to be positively
correlated with the degree of molecular hydrophobicity. These
Capsaicinoids (CPCs, shown in Table 1), mainly including
capsaicin (CPC), dihydrocapsaicin (DCPC), nordihydrocap-
saicin (NDCPC), and N-vanillylnonanamide (N-V), are the
main chemicals retaining the pungency and are usually
composed of hydrophobic alkane chains and hydrophilic
vanillyl amide moieties.9 Owing to their stable physicochemical
properties, CPCs have been confirmed as one of the most
representative specific markers for the authenticity evaluation
of edible oil. Adulteration of edible oil with refined waste oil
has been one of the most concerning problems of current food
quality control in developing countries.10 Thus, there is a high
demand for rapid, sensitive, and specific detection of CPCs to
ensure the quality of edible oils.11−13 At present, some
immunoassays for the determination of CPCs have been
described.14−16 In these studies, with the consideration of
producing antibodies that can recognize total CPCs, nearly all
the designed haptens were biased toward exposing the
Received: June 18, 2021
Revised: August 2, 2021
Accepted: August 5, 2021
Published: August 19, 2021
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967
© 2021 American Chemical Society
9957

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
a
Table 1. IC50 Values and CRs of Antisera to Four CPCs and Eight Analogues after the Fourth Immunization
a
The data of hapten5 were not given because it failed to induce a significant immune response (IC50 > 20,000 ng/mL).
common vanillyl amide moiety rather than the alkane chain
moiety of CPCs. However, the obtained antibodies showed
low affinities to CPCs with the half inhibition (IC50) varying
from 1.12 to 160,000 ng/mL, which are insufficient to develop
highly sensitive immunoassays for CPCs (Table S1).15,16
Perhaps even worse, the antibodies induced by these haptens
provided inaccurate recognition ability, that is, low specificity,
which inevitably recognize not only CPCs but also other
analogues that contain a vanillyl amide structure simulta-
neously, for example, guaiacol, 4-methylguaiacol, 4-ethyl-
guaiacol, and so on (Table 1). These analogues are naturally
present in several kinds of vegetable oils or used as food
additives that may coexist with CPCs during waste oil
treatment, which can result in false positive results.17,18
Clearly, to further improve the detection performance of
immunoassays to meet the practical detection requirements, a
more efficient hapten design strategy needs to be proposed to
produce antibodies with high affinity and specificity to CPCs.
In the present study, we proposed four novel haptens
maximally exposing the hydrophobic alkane chain moiety of
CPCs based on our previous findings, which is the non-
common structure of CPCs. The feasibility of these haptens
was first proved by computational chemistry in the aspects of
conformation and electrons to evaluate the similarity among
CPCs before total synthesis. For the purpose of comparison,
hapten5−8 exposing the hydrophilic vanillyl moiety were also
synthesized to produce antibodies. The effect of these hapten
structures on the characteristics of the resulting antibodies
obtained from animal immunizations was extensively inves-
tigated in terms of affinity and specificity. To further explore
the influence of hapten properties on antibodies, main
physicochemical descriptors (lipid−water partition coefficient
(LogP), molecular volume (Vm)/surface, etc.) of all haptens
were extracted and analyzed. Eleven monoclonal antibodies
(mAbs) were obtained, and the variable regions of two mAbs
from hapten1 and hapten7 were deconstructed to illustrate the
recognition mechanism to CPC. Finally, a highly sensitive
indirect competitive enzyme-linked immunosorbent assay
(icELISA) based on the mAb 2F7 from hapten1 was developed
for detecting four CPCs simultaneously in oil samples. The
performance of the developed icELISA was assessed using
fortified and recovery studies and further confirmed by high-
performance liquid chromatography-tandem mass spectrome-
try (HPLC−MS/MS).
MATERIALS AND METHODS
■
Reagents and Apparatus. CPC, DCPC, NDCPC, N-V, and 4-
(2-aminoethyl)-2-methoxyphenol were bought from J&K Scientific
Co., Ltd. (Beijing, China). Vanillylamine hydrochloride, 4-hydroxy-3-
methoxybenzylamine hydrochloride, guaiacol, 2-methoxy-4-methyl-
phenol, acetovanillone, 4-ethyl-2-methoxyphenol, and 4-hydroxy-3-
methoxybenzyl alcohol were bought from Shanghai Bide Pharma-
ceutical Technology Co., Ltd. (Shanghai, China). Ethyl 6-bromohex-
anoate, methyl 4-(bromomethyl)benzoate, bovine serum albumin
(BSA), bovine thyroglobulin (BTG), N-hydroxy succinimide (NHS),
N,N-dicyclohexylcarbodiimide (DCC), keyhole limpet hemocyanin
(KLH), horseradish peroxidase (HRP)-labeled goat antimouse IgG,
and casein were acquired from Sigma-Aldrich (St. Louis, MO, USA).
Incomplete Freund’s adjuvant (IFA), complete Freund’s adjuvant
(CFA), cell culture medium (DMEM), red blood cell lysis buffer,
PEG 1500, and fetal calf serum were obtained from Gibco BRL
(Carlsbad, CA, USA). Succinic anhydride, 1-(3-dimethylaminoprop-
yl)-3-ethylcarbodiimide hydrochloride, and adipic anhydride were
9958
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Figure 1. Synthetic routes of haptens.
supplied by Aladdin Chemistry Co., Ltd. (Shanghai, China). 3-
Methoxy-4-hydroxy benzaldehyde oxime was obtained from Saen
Chemical Technology Co., Ltd. (Shanghai, China). Ten edible oil
samples were bought from China Oil & Foodstuffs Co., Ltd. (Beijing,
China) and Yihai Kerry Group (Shanghai, China). The 10 real edible
oil samples were provided by the Shenzhen Center for Disease
Control and Prevention. The remaining chemical reagents were
supplied by Sinopharm Chemical Reagent, Inc. (Beijing, China).
Polystyrene microtiter plates and 96-well cell culture plates were
purchased from Corning incorporated (Costar 2592 and 3599,
Corning, NY). Buffer and solution used in this work can be found in
the Supporting Information.
Nuclear magnetic resonance spectrometry (NMR) DRX-300
(Bruker, Billerica, MA, USA) and high-resolution mass spectrometry
(HRMS) (Agilent Technologies, Santa Clara, CA, USA) were used to
confirm the hapten structures. The hapten−protein conjugates were
evaluated by matrix-assisted laser desorption ionization time-of-flight
mass spectrometry (MALDI−TOF−MS) (Bruker, Bremen, Ger-
many). The UPLC-MS/MS analysis was carried out on an ACQUITY
UPLC I-Class system with Xevo TQD, which was equipped with an
ACQUITY UPLC BEH C18 column (2.1 × 100 mm with 1.7 μm
particle size) (Waters, Milford, MA, USA).
functional level with the M06-2X basis set19,20 using the Gaussian 09
software (Gaussian, Wallingford, CT, USA). The descriptors of the
haptens including energy (E), dipole moment (μ), Mulliken charges,
energy of the highest occupied molecular orbital (EHOMO), and energy
of the lowest unoccupied molecular orbital (ELUMO) were extracted
from the Gaussian output file. The LogP was obtained using
ChemDraw (PerkinElmer, Waltham, MA, USA). The molecular
alignment was achieved by Molecule Overlay dialog in Discovery
Studio 2016 (BIOVIA Corp., San Diego, CA, USA). The surface area
(SA), Vm, polar surface area, and molecular polarity index (MPI) were
extracted using the Multiwfn 3.7(dev) code.21 The VMD together
with the Multiwfn 3.7(dev) code were used to analyze the ESP
surfaces on 0.001 a.u.
Synthesis of Haptens and Hapten−Protein Conjugates. A
synthetic scheme of the hapten1−6 is shown in Figure 1. Hapten7
and hapten8 are commercial compounds vanillylamine hydrochloride
and 4-(2-aminoethyl)-2-methoxyphenol, respectively.
Hapten1 was synthesized as follows: briefly, 3.05 g CPC and 2.76 g
potassium carbonate were added to acetonitrile and then stirred
evenly. An amount of 2.75 g ethyl 6-bromohexanoate was then added
and refluxed for 24 h. Acetonitrile was then removed after
concentration followed by ethyl acetate extraction, washing, drying,
and pacification steps. An equal volume of ethanol was added to the
above solution in tetrahydrofuran, and then 0.02 M NaOH solution
was added dropwise. The mixture was then reacted at 50 °C for 1 h.
After concentration, ethanol and tetrahydrofuran were removed. The
mixture was cooled to room temperature, and the pH was then
Electrostatic Potential (ESP) Analysis and Descriptor
Extraction by Computational Chemistry. Three-dimensional
(3-D) structures of molecules were built using GaussView 5.0
software (Gaussian, Wallingford, CT, USA). Then, all molecules were
optimized by density functional theory calculations at the TVZP
9959
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
adjusted to acidity by adding hydrochloric acid to obtain hapten1 with
and 0.001 mM KLH were dissolved in 0.1 M sodium bicarbonate
solution. After stirring for 10 min, 42 μL of 25% glutaraldehyde was
dropwise added and stirred at room temperature for 24 h. The
conjugates were dialyzed against PBS (pH 7.0) for 72 h. The synthesis
of the coating antigens of hapten7−8 was conducted similarly except
that the KLH was replaced with BSA.
The conjugation ratios of the hapten−BSA conjugates were
calculated using MALDI−TOF−MS by comparing the observed
molecular weights (MW) of the resulting conjugates with that of the
native protein.
(+)
90% yield. HRMS (m/z) calc. For C24H37NO5 419.56, found
1
420.00; H NMR (300 MHz, DMSO-d6) δ (ppm) 11.99 (br, s, 1H,
COOH), 8.19 (t, 1H, NH), 6.88−6.71 (dd, 2H, J = 12 Hz, CHar),
6.84 (s, 1H, CHar), 5.36−5.33 (m, 2H, −HCCH−), 4.19−4.17 (d,
2H, CH2), 3.92−3.88 (t, 2H, CH2), 3.73 (s, 3H, CH3), 2.25−2.20 (m,
3H, CH2, CH), 2.14−2.09 (t, 2H, CH2), 1.95−1.93, (q, 2H, CH2),
1.72−1.67 (m, 2H, CH2), 1.58−1.49 (m, 4H, CH2, CH2), 1.43−1.23
(m, 4H, CH2, CH2), 0.94−0.84 (dd, 6H, CH3, CH3).
Hapten2 was synthesized similar to that described for hapten1.
DCPC was reacted with ethyl 6-bromohexanoate to obtain hapten2
Production of mAbs. Eight 8-week-old female BALB/c mice per
immunogen were immunized according to our previously described
procedure.22 Briefly, the first injection was 100 μg immunogen
emulsified with CFA followed by three subcutaneous injections of 100
μg immunogen with IFA (Table S2). Starting from the second
immunization, mouse antisera samples were collected in the tenth day
after each immunization, and the affinity and specificity were detected
by icELISA. After the fourth immunization, the mice exhibiting the
highest affinity were sacrificed for subsequent cell fusion. The
experimental procedures of fusion were described previously by our
group.22 The hybridoma cells were screened for antibody production
using noncompetitive ELISA and icELISA followed by subcloning
based on the limiting dilution method. The clones with high
inhibition were cloned three priors for ascites production. The
antibody was purified from ascites using the affinity chromatography
columns.
Extraction of Antibody Variable Region Sequences,
Homology Modeling, and Molecular Docking. Total RNA was
isolated from the hybridoma cells following the technical manual of
RNeasy Plus Micro Kit (Qiagen Japan Co. Ltd., Tokyo, Japan) and
then reverse-transcribed into cDNA using either isotype-specific
antisense primers or universal primers of SMARTScribe Reverse
Transcriptase (Clontech, Tokyo, Japan). Antibody fragments with
heavy chains and light chains were then amplified by polymerase
chain reaction (PCR). Different sets of primers listed in Table S3
were used for PCR, which was performed with 30 cycles of 95 °C, 60
s; 65 °C, 30 s; and 72 °C, 120 s. Amplified antibody fragments were
cloned into a standard cloning vector separately. Colony PCR was
performed to screen for clones with inserts of correct sizes. Bacteria
liquid were sent to Genscript (Nanjing, China) for sequencing after
the correct single clones were incubated overnight at 37 °C in LB
liquid medium with shaking at 250 rpm. The sanger sequencing
results of single-chain variable fragment antibody (scFv) 2F7 and
26A1 are presented in Figure S1.
3-D structure predictions for scFv 2F7 and 26A1 were generated by
the program Model Antibody of Discovery Studio based on the
primary structure of antibodies. According to the sequence similarity
and identity, the best matching overall template 1f4w and 2ddq had
been used to model antibody framework for 2F7 and 26A1,
respectively, which determined the relative spatial orientation of
light chain and heavy chain. The binding between two proteins (scFv
2F7 and 26A1) and CPC, DCPC, and ethyl-2-methoxyphenol was
performed using CDOCKER of Discovery studio. After docking
ligands into the top-ranked cavity of the antibody, the binding pose of
each complex with the highest score was selected.
Development and Optimization of icELISA. Each well of
microtiter plates was first coated with 100 μL of coating antigen in
carbonate buffer and then incubated at 4 °C overnight. After blocking,
50 μL of CPC standard with serial dilutions and 50 μL of diluted mAb
were added to the wells, and incubation for 30 min at 37 °C was
carried out. After washing two times, 100 μL of diluted HRP-labeled
goat antimouse IgG was then added followed by incubation for 30
min at 37 °C. The TMB substrate was added and incubated for 15
min at 37 °C after further washing. Then the chromogenic reaction
was inhibited using 2 M H2SO4 (50 μL/well), and the optical density
(OD) values of 450 nm were measured.
(+)
with 90% yield. HRMS (m/z) calc. For C24H39NO5 421.58, found
1
422.0; H NMR (300 MHz, DMSO-d6) δ (ppm) 11.99 (br, s, 1H,
COOH), 8.19 (t, 1H, NH), 6.88−6.85 (s, 1H, CHar), 6.84−6.74 (dd,
2H, J = 1.8 Hz, CHar), 4.19−4.17 (d, 2H, CH2), 3.92−3.88 (t, 2H,
CH2), 3.73 (s, 3H, CH3), 2.25−2.20 (t, 2H, CH2), 2.13−2.08 (t, 2H,
CH2), 1.71−1.67 (q, 2H, CH2), 1.72−1.67 (m, 2H, CH2), 1.56−1.40
(m, 7H, CH2, CH2, CH2, CH), 1.24−0.83 (m, 8H, CH2, CH2, CH2,
CH2), 0.86−0.83 (dd, 6H, CH3, CH3).
Hapten3 was synthesized similar to the previous procedure of
hapten1. CPC was reacted with methyl 4-(bromomethyl)benzoate to
(+)
obtain hapten3 with 70% yield. HRMS (m/z) calc. For C26H33NO5
439.55, found 439.9; 1HNMR (300 MHz, DMSO-d6) δ (ppm) 12.99
(br, s, 1H, COOH), 8.29 (t, 1H, NH), 7.97−7.94 (d, 2H, CHar),
7.65−7.62 (d, 2H, CHar), 6.94 (s, 1H, CHar), 6.92−6.71 (dd, 2H,
CHar), 5.35−5.32 (m, 2H, CHCH), 5.14 (s, 2H, CH2), 4.19−4.17
(d, 2H, CH2), 3.78 (s, 3H, CH3), 2.28−2.22 (m, 1H, CH), 2.14−2.09
(t, 2H, CH2), 1.94−1.93 (q, 2H, CH2), 1.54−1.49 (m, 2H, CH2),
1.31−1.29 (m, 2H, CH2), 0.93−0.83 (dd, 6H, CH3, CH3).
Hapten4 was synthesized similar to the previous procedure of
hapten1. DCPC was reacted with methyl 4-(bromomethyl)benzoate
to obtain hapten4 with 90% yield. HRMS (m/z) calc. for
(+)
1
C26H35NO5 441.57, found 441.90; HNMR (300 MHz, DMSO-
d6) δ (ppm) 12.99 (br, s, 1H, COOH), 8.29 (t, 1H, NH), 7.97−7.94
(d, 2H, CHar), 7.56−7.52 (d, 2H, CHar), 6.94 (s, 1H, CHar), 6.92−
6.71 (dd, 2H, CHar), 5.14 (s, 2H, CH2), 4.19−4.17 (d, 2H, CH2),
3.78 (s, 3H, CH3), 2.13−2.08 (t, 2H, CH2), 1.53−1.49 (m, 3H, CH2,
CH), 1.1.23−1.11 (m, 8H, CH2, CH2, CH2, CH2), 0.85−0.83 (dd,
6H, CH3, CH3).
Hapten5 was synthesized as follows: 0.20 g vanillin amine
hydrochloride was added to 6 mL of pyridine and stirred for 10
min. An amount of 0.12 g succinic anhydride was then added to the
mixture and stirred overnight at 80 °C. The reaction progress was
monitored by TLC (ethyl acetate/ methanol, 2:1). The final product
of hapten5 was obtained by filtering and drying. HRMS (m/z) calc.
1
For C12H15NO5Na(+) 275.23, found 275.80; H NMR (300 MHz,
DMSO-d6) δ (ppm) 12.10−10.71 (br, s, 1H, COOH), 9.62−8.50 (br,
s, 1H, OH), 8.24 (t, 1H, NH), 6.82−6.81 (s, H, CHar), 6.71−6.64
(dd, 2H, CHar), 4.16−4.14 (d, 2H, CH2), 3.74 (s, 3H, CH3), 2.46−
2.44 (t, 2H, CH2), 2.38−2.36 (t, 2H, CH2).
Hapten6 was synthesized similar to that described for hapten5.
Vanillin amine hydrochloride was reacted with adipic anhydride to
obtain hapten6. HRMS (m/z) calc. for C14H19NO5Na(+) 303.29,
found 303.90.
The hapten1−6 were conjugated to a protein (KLH, BTG or BSA)
by the active-ester method through their active carboxylic acid groups.
First, the hapten (0.1 mM) was dissolved in 0.5 mL of DMF, and then
NHS (15.0 mg) and DCC (30.0 mg) were added, which was stirred at
room temperature overnight. Then, the solution (activated hapten)
was added dropwise into the dissolved KLH solution (0.001 mM
KLH in 10 mL of PBS), BTG solution (0.001 mM BTG in 10 mL of
PBS), or BSA solution (0.001 mM BSA in 10 mL of PBS),
respectively, and further stirred for 24 h. The conjugates were dialyzed
against PBS (pH 7.0) at room temperature for 72 h. Hapten-KLH
served as immunogens for hapten1−4, while hapten5−6 were
conjugated with BTG as immunogens because of the large amount
of precipitation generated when they reacted with KLH. The hapten−
BSA acted as coating antigens for hapten1−6.
Eight analogues of CPCs shown in Table 1 were used to evaluate
the cross-reactivity (CR) of assay. The CR was calculated using the
formula as follows:
The immunogens of hapten7−8 were prepared by conjugating with
KLH based on the glutaraldehyde method. Briefly, 0.1 mM of hapten
9960
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Figure 2. Molecular superposition, ESP energy, atomic charges, and physicochemical parameters of four CPCs including CPC, DCPC, NDCPC,
and N-V. (A) Superposition graph based on optimized structures of CPC (orange) and DCPC (purple), NDCPC (green), and N-V (yellow)
calculated at the TVZP/M06-2X level. (B) SA in each ESP range on the van der Waals surface of CPCs. The negative ESP regions are indicated in
blue, the positive regions in red, and the neutral regions in white. (C) Calculated partial Mulliken atomic charges of CPCs. Only those heavy atoms
to the four structures are depicted.(D) Physicochemical parameters of CPCs including MW, μ, SA, MPI, and LogP.
CR = (IC50 of CPC/IC50 of compounds) × 100%
obtaining antibody to a class of analytes, the planned hapten
obviously should preserve the common structure of target
analytes to the greatest extent and minimize the exposure of
the corresponding feature structure of the individual analyte.23
The CPCs, mainly including CPC, DCPC, NDCPC, and N-V,
all share the structures of various alkane chains and common
vanillyl amide as shown in Table 1. In previous reports, the
designed haptens that exposed the vanillyl amide moiety were
predominantly employed as summarized in Table S1.15,16 As
expected, the reported antibodies do recognize all CPCs;
however, they have low and different affinity to CPCs with
IC50 values of 1.12 to 160,000 ng/mL and thus obviously
cannot meet the high-sensitivity detection requirements. More
importantly, many compounds also possess the vanillyl amide
structure shown in Table 1 that may coexist with CPCs in the
procedure of oil production, consumption, and refinement,
which could be recognized by the antibodies derived from
haptens exposing the vanillyl amide moiety and then disturb
the specificity of the developed immunoassays. Therefore, a
novel hapten design strategy is highly demanded to produce
antibodies with uniform and high affinity as well as accurate
recognition to CPCs.
As shown in Table 1, although the vanillyl amide moiety of
four CPCs is identical and obvious differences are observed at
the alkane chain moiety, we thought that the hapten maximally
exposing the alkane chain might be a better strategy for
obtaining desired antibodies. As we reported, the hydrophobic
group of haptens could induce a higher immune response and
then exhibit stronger antibody performance than the hydro-
philic counterpart.5 Obviously, the alkane chains of CPCs
provide significantly high hydrophobicity. Therefore, as shown
in Figure 1, four novel haptens were designed and named
hapten1−4 by introducing carboxyl groups on the vanillyl
The optimal concentrations and pairing combinations of the
coating antigens and antibodies were optimized using a checkerboard
procedure. Under the optimized working concentration of antibody
and antigen, the ionic strength (concentrations of NaCl were 0, 0.01,
0.05, and 0.25 mol/L), pH value (5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, and
12.0), and organic solvent (proportion of methanol and acetonitrile
was 0, 2, 5, 10, 20, and 40%) were optimized.
Matrix Effect and Recovery. Ten kinds of soybean oil, maize oil,
peanut oil, canola oil, olive oil, sunflower oil, and blend oil were
collected from markets, and 10 real edible oil samples were from the
Shenzhen Center for Disease Control and Prevention. Standard
solutions of CPC, DCPC, NDCPC, and N-V at different
concentrations were added to 10.0 g oil samples and then extracted
twice with 5 mL of methanol. After centrifugation, the top organic
layer was separated from the oil layer. The organic layer (2 mL) was
then dried with a stream of nitrogen gas and reconstituted with 1 mL
of 20% methanol in 0.01 M PBS. Finally, 50 μL of reconstitution
solution was removed and mixed with 950 μL of 0.01 M PBS for
icELISA.
An amount of 1.0 g of edible oil sample was weighed into a 15 mL
centrifuge tube. Then, 1 mL of methylene chloride and 3 mL of 2%
sodium hydroxide solution were added. The mixture was vortex-
mixed for 10 min and then centrifuged at 4000 rpm for 10 min. The
pH of the upper aqueous phase was adjusted to 2−3 and transferred
into a preactivated C18 SPE column. The sample solution was slowly
passed through the SPE column followed by eluting twice with 3 mL
of acetonitrile. The eluent was then dried under a nitrogen stream at
50 °C, and the residues were redissolved in 1 mL of 50% (v/v)
aqueous methanol. The solution was subsequently transferred into an
injection vial with a 0.22 μm organic syringe filter for HPLC−MS/MS
analysis.
RESULTS AND DISCUSSION
■
Hapten Design and Computational Validation. The
objective of the present study was to produce mAbs with high
affinity and specificity to total CPCs. For the purpose of
9961
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
amide moiety of CPC or DCPC based on the above
explanation.
phenyl as the spacer arm ending with a carboxyl group for
hapten3 and hapten4. For comparison, hapten5−8 were
designed similar to previous studies to verify our hypothesis
by coupling to the carrier protein with the side of the CPCs
that was far from the vanillyl amide moiety (Figure 1). The
synthetic hapten1−6 were purified and identified by HRMS
and NMR (Figure S2), which demonstrated the successful
synthesis of hapten1−6. The BSA conjugates were charac-
terized successfully using MALDI−TOF−MS (Figure S3 and
Table S4).
To verify the hypothesis of the novel hapten design strategy,
we conducted the computational chemistry to further compare
the similarity of four CPCs especially the alkane chain moiety
with the inspection of conformational, electronic, and
physicochemical aspects before total synthesis. As shown in
Figure 2A, the CPC was aligned with DCPC, NDCPC, and N-
V based on their lowest energy conformations without
hydrogen atoms for concise purposes. Consistent with two-
dimensional structures, the spatial conformations of four CPCs
showed a perfect overlap in the vanillyl amide moiety, but
visible position shifts were observed at the end of the CPC
alkane chain because of the discrepancies of the methyl
position and the length of the carbon chain. In the author’s
opinion, the observed structural difference of the alkane chain
among four CPCs may not significantly change the hydro-
phobicity of CPCs and then avoid deleteriously affecting
uniform recognition for four CPCs of the resulting antibody
because the hydrophobic force is thought to be the main
driving force between the small molecule and antibody.
Because ESP is closely related to the hydrophobicity of small
molecules and could be used as an indicator, the ESPs
displayed on van der Waals surfaces with atomic charge
distribution of four CPCs were obtained to confirm the
consistency of the alkane chain moiety. As expected from
Figure 2B, the ESPs on van der Waals surfaces of the four
CPCs were all found to be mainly distributed in the range of
0−10 kcal/mol with remarkably close values in each interval.
The vanillyl amide moiety of CPCs shows negative ESP
regions (blue) at nitrogen-containing groups and positive ESP
regions (red) at the oxygen-containing groups, respectively,
while the alkane chain moiety is predominantly present in
neutral regions (white), indicating the high hydrophobicity of
the moiety. These results showed that the conformational
differences of the terminal alkane chains on the C20 and C21
had a limited effect on the electronic distribution and then
hydrophobicity of the four CPCs (Figure 2B,C). Moreover, we
calculated some physicochemical parameters that may
influence the immunogenicity of haptens, including MW,
SAs, and several hydrophobic parameters. The summarized
data in Figure 2D showed no discernable difference between
four CPCs, and the variation of the alkane chains among CPCs
did not induce the observable change of physiochemistry. In
conclusion, the computational chemistry clearly provides
theoretical support for our hypothesis that the planned
haptens exposing the varied alkane chain moiety of CPCs
potentially could induce antibodies with uniform and high
affinity to four CPCs based on the hydrophobic similarity of
these CPCs together with the conformation and other
physicochemical parameters. The hypothesis of the hapten
design strategy was then further evaluated by total chemical
synthesis and animal experiments.
Screening and Characterization of Antisera and
mAbs. The antisera of eight groups of mice that immunized
with hapen1−8 were monitored from the second immuniza-
tion. CPC was used to determine the titer and affinity of
antisera with the homologous coating antigens. As shown in
Tables 1 and S5, the antisera derived from hapten1−4 showed
generally higher affinity to CPCs than those of hapten5−8,
with an obvious improvement of about 1000 times or even
higher. The results support our finding that the antibody
response to haptens is positively correlated with molecular
hydrophobicity. Furthermore, DCPC, NDCPC, N-V, and
other eight structurally related analogues were used to evaluate
the specificity (expressed by CRs) of antisera derived from
hapten1−8, except hapten5 that failed to induce a significant
immune response (IC50 > 20,000 ng/mL). Results summarized
in Table 1 represent that all antisera can recognize DCPC,
NDCPC, and N-V well, which was consistent with the
predicted results of computational chemistry (Figure 2). For
the mice immunized with hapten1−4-KLH, the antisera
showed ignorable CRs toward other CPC analogues (CRs <
0.1%) compared with those from hapten-6-BTG, hapten-7-
KLH, and hapten-8-KLH (CRs ranging from 2−72%). In
previous studies, haptens of CPCs were prepared by exposing
the common vanilla amide moiety, which was consistent with
hapten5−8 in this study. The antisera results provide strong
evidence that the previous reported hapten design strategy can
lead to a decrease in the specificity of the resulting antibody to
CPCs and thus result in false positive detection of the
developed immunoassay.16,17 In conclusion, the novel
hapten1−4 proposed in this study were superior in producing
antibodies with high affinity and specificity to CPCs than
hapten5−8, which confirmed our hypothesis. To further
confirm the key factors that affect the ability of haptens to
induce antibody responses, the physicochemical descriptors of
hapten1−8 such as LogP, Vm/surface, and so on were
extracted. As shown in Table S6, there is a significant
difference of the LogP between the two hapten groups: the
values of hapten1−4 were ranging from 4.925 to 6.221, while
the hapten5−8 were between 0.107 and 0.615, indicating that
the hapten1−4 are more hydrophobic. These data showed
again that the highly hydrophobic alkane chain moiety
accounts more for a strong antibody response than the
hydrophilic vanillyl amide moiety of CPCs. Therefore,
consistent with our earlier study, increasing hydrophobicity
of haptens is beneficial to antibody performance, which could
be a general and novel principle in the field of hapten design.5
Then, the mice with highest affinity to CPC were chosen,
that is, hapten1-KLH-3#, hapten2-KLH-4#, hapten3-KLH-8#,
and hapten7-KLH-7# for cell fusion. Eleven hybridomas were
finally obtained. In detail, mAb 2F7, 9B11, 14G9, and 11G1
were produced from hapten1, 15B3, 18E9, and 6H12 were
produced from hapten2, 8C4 and 1E11 were produced from
hapten3, and 7A1 and 26A1 were produced from hapten7,
Preparation and Identification of Haptens and
Conjugates. After the validation by computational chemistry,
we synthesized hapten1−4 by introducing carboxyl groups on
the vanillyl amide moiety of CPC or DCPC, thereby exposing
the alkane chain to the immune system, which was quite
possible to generate high affinity and specificity antibodies
against CPCs as mentioned above. Previous studies have
shown that the incorporation of phenyl into haptens as the
spacer arm could significantly improve immunological proper-
ties of the resultant immunogen.24,25 Here, we introduced a
9962
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Figure 3. Molecular docking of the Fv−CPC interaction. Molecular docking of scFv (A) 2F7 and CPC, (B) 26A1 and CPC, (C) 2F7 and DCPC,
and (D) 26A1 and ethyl-2-methoxyphenol generated using Discovery Studio software. The heavy chain and light chain are shown in orange and
blue, respectively; yellow, wheat, and cyan components for CDR L1, L2, and L3, respectively, and red, magenta, and gray components for CDR H1,
H2, and H3, respectively. Amino acid residues involved in the interactions with the ligand are labeled. Hydrogen bonds are shown as green dashed
lines and hydrophobic interactions are displayed as purple dashes. The hydrophobic regions of binding pockets are indicated in khaki, the
hydrophilic regions in blue, and the neutral regions in white.
respectively. The mAbs from hapten1−3 showed high affinity
and specificity to CPC with IC50 values ranging from 0.25−
0.77 ng/mL, which were nearly 3−10,000 times lower than
those of other antibodies reported before (Table S1).15,16 All
these results of mAbs demonstrated that the new designed
haptens based on our new hapten design strategy in this study
were more effective to produce mAb with high performances
than those reported previously.
Recognition Mechanism of mAbs to CPCs. To explain
the scientific basis behind the novel principle of hapten design
we proposed, structural studies on CPC in complex with the
scFv 2F7 and 26A1 were conducted after antibody sequencing,
which were derived from hapten1 and hapten7, respectively.
The modeled scFv 2F7 and 26A1 were docked with CPC as
shown in Figure 3A,B. The different orientations of CPC in the
binding pocket of antibody provide a tentative explanation for
the significant difference in binding affinity (over 1000-fold)
with 2F7 and 26A1 respectively. The CPC in the CPC-2F7
complex was observed in a “head-up, tail-down” configuration,
that is, the hydrophobic alkane chain moiety was deeply
embedded in the binding pockets with the opposite moiety
pointed outward (Figure 3A). Not surprisingly, the binding
pocket formed by 2F7 and CPC was largely hydrophobic, and
the alkane chain was deeply inserted into the hydrophobic
cavity surrounded by the hydrophobic residues VAL242,
ALA305, and PHE318, indicating that the high affinity of 2F7
to CPC mainly originated from hydrophobic forces, while the
location of the CPC was rotated by 180° in the binding pocket
formed by 26A1 and CPC, which exhibited a “tail-up, head-
down” configuration (Figure 3B). In this orientation, the
binding of 26A1 to CPC was mainly maintained by the
hydrogen bonds that are composed of HIS107, TRP55, and
GLY307 with the vanillyl amide moiety, and the surface of the
binding pocket was more hydrophilic. Therefore, the lack of
strong hydrophobic interactions for CPCs with a deep
hydrophobic cavity of antibody is considered to be the main
cause of low affinity of 26A1, indicating that increasing the
hydrophobicity of haptens is a useful strategy to improve the
performance of the resulting antibody. Then, to further
evaluate the contribution of hydrophobic force in antibody
recognition, the docking of DCPC (CR expressed as 96%)
with 2F7 was studied. As shown in Figure 3C, the DCPC binds
with 2F7 in a highly similar way to CPC, with a “head-up, tail-
down” pose in the hydrophobic binding pocket. The results
supported our hypothesis that the various alkane chains among
CPCs will not significantly change the hydrophobicity and
ensure the uniform and high-affinity recognition of antibody to
CPCs, which was already proved by computational chemistry.
On the contrary, the 26A1 showed exciting CRs to CPCs in
the range of 100−115% but also high CRs to analogues
9963
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Figure 4. Optimization of physicochemical parameters on the icELISA and the calibration curves assessed by the icELISA. The effect of (A) ionic
strength, (B) pH value, (B) methanol, and (D) acetonitrile on the icELISA (n = 3). (E) Calibration curve of icELISA based on mAb 2F7 for CPC
in buffer. (F) Calibration curves of icELISA for four CPCs in the diluted oil sample (n = 3).
ranging from 5 to 152%. For example, a CR of 119% between
26A1 and ethyl-2-methoxyphenol was observed. As expected
and observed in Figure 3D, the ethyl-2-methoxyphenol is in
the hydrophilic cavity of 26A1 that is mainly constituted by
TRP55, TYR38, GLY307, and SER308 despite the deletion of
the hydrophobic alkane chain moiety, demonstrating that the
26A1 predominantly recognizes the vanillyl amide moiety.
These results indicated that the 2F7 mainly recognized the
alkane chain moiety of CPCs while the 26A1 was more
inclined to the vanillyl amide moiety, which exactly manifested
the accurate recognition ability for CPCs of 2F7 because of
avoiding the interference of other vanillyl amide analogues
without alkane chains.
Optimization and Development of icELISA. Several
factors that influence the performance of icELISA were
evaluated. Coating heterologous antigens is an effective
strategy to increase the sensitivity of icELISA because of the
weaker recognition of the antibody compared to homologous
coating antigens.26 Thus, the optimal pairs of 11 mAbs and 8
coating antigens were then assessed, and the Amax, IC50, and
values <0.1). In addition, heterologous coating in our study
does not always cause a significant increase in the sensitivity of
icELISA as reported previously.27 This may be due to the high
degree of structural similarity between the coating antigens,
which only differ in the length and types of spacer arms. The
specificity of the antibodies was assessed using three CPCs and
eight structurally related compounds. The results are
summarized in Tables S8 and S9, presenting similar values
to that of antisera. Taken together, the pair of 2F7 and
hapten2-BSA was selected for the following work because of its
lowest IC50 and high specificity to CPCs.
The dilutions of immunoreagents were then optimized using
the checkerboard assay. The optimal dilutions of mAb, coating
antigens, and goat antimouse IgG are 1/30,000, 1/3000, and
1/3000, respectively. Several physicochemical factors including
ionic strength, pH value, and organic solvent were further
evaluated. It can be seen from Figure 4A that the ratio of Amax
/
IC50 was highest at an ionic strength of 0.01 mol/L, which
showed a significant improvement compared with that without
ionic strength. The ratio of Amax/IC50 was better at pH 8.0 in
the range of 5.0−12.0 (Figure 4B). In addition, the results of
the organic solvent tolerance test are illustrated in Figure
4C,D, indicating that the Amax/IC50 ratio decreased signifi-
cantly when the concentration of methanol achieving 10% or
the concentration of acetonitrile is over 5%. Therefore, the
Amax/IC50 values of icELISA for CPC from different pairs are
summarized in Table S7. The data obtained from 2F7, 9B11,
14G9, 11G11, 5B3, 18E9, 6H12, 8C4, and 1E1 with hapten5−
8-BSA as well as 7A1 and 26A1 with hapten1−5-BSA are not
given in Table S7 because of the negligible recognition (OD
9964
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Table 2. IC50, LOD, Linearity Range, Recovery, and CV of icELISA and HPLC−MS/MS for CPCs from Spiked Samples
icELISA
IC50
HPLC−MS/MS
amount spiked
recovery CV
LOD
linearity range
amount spiked
recovery CV
LOD
(μg/kg)
linearity range
CPCs
CPC
(μg/kg)
(%)
(%)
(μg/kg) (μg/kg)
(μg/kg)
(μg/kg)
(%)
(%)
(μg/kg)
1
2
5
1
2
5
2
5
10
3
10
30
81.7
92.3
108.1
76.5
93.5
114.3
92.3
96.9
98.4
109.2
88.4
84.8
4.8
4.5
2.0
6.3
5.7
3.3
10.5
5.9
6.3
6.3
10.0
5.3
0.73
0.18
0.04−0.78
0.04−0.83
0.06−2.13
0.17−3.81
2
93.0
94.3
98.0
91.0
93.0
94.5
92.0
95.0
93.2
84.5
90.3
93.0
7.6
5.3
6.1
7.6
6.2
6.0
5.2
4.5
4.7
8.3
6.8
5.9
30
60
2
30
60
2
30
60
2
30
60
2.0
1−100
1−100
1−100
1−100
DCPC
NDCPC
N−V
0.77
0.19
2.0
2.0
2.0
1.46
0.36
3.29
0.81
optimum conditions (0.01 M PBS and pH 8.0) of icELISA
were confirmed. Based on this, various concentrations of CPC
(0, 0.01, 0.03, 0.09, 0.27, 0.81, and 2.43 ng/mL) were
determined by the icELISA (Figure 4E). The IC50 and
analytical range of the developed icELISA were 0.11 ng/mL
and 0.03−0.43 ng/mL in buffer, respectively.
developed is comparable to the HPLC−MS/MS in sensitivity
and precision. Subsequently, 10 real edible oil samples from
the Shenzhen Center for Disease Control and Prevention were
detected by the established icELISA and HPLC−MS/MS at
the same time. Both the methods showed that no CPCs
residues was detected in all samples. These results demon-
strated that the developed icELISA can be used as a practical
screening tool to inspect the CPCs residues in oil samples. In
addition, compared to instrumental method, ELISA is more
suitable for field screening owing to its obvious advantages,
such as short detection time, simple sample preprocessing, and
low cost.
In the present study, we successfully designed and evaluated
novel haptens of CPCs by maximally exposing the hydro-
phobic moiety to produce antibodies with high affinity and
specificity. The results demonstrated that the antibodies
derived from novel haptens exhibited much higher affinity
and specificity to CPCs than those from haptens exposing the
hydrophilic moiety. We found that the degree of the antibody
response was positively correlated with hydrophobicity of
haptens and proposed that increasing hapten hydrophobicity
could become a new principle for hapten design, which were
proved by computational chemistry and molecular recognition
study. Finally, a rapid and reliable icELISA for screening CPCs
in oils with significantly improved sensitivity and specificity
was established. This work not only established an excellent
rapid screening method for total CPCs in oils but also
demonstrated a general strategy of rational hapten design.
Matrix Effect and Recovery. In the present study, the
matrix effect was evaluated using blank sample extract after the
pretreatment of the blended oil as described above to establish
the matrix addition standard curve of the four CPCs. As shown
in Figure 4F, the IC50 value (0.18 ng/mL) of CPC was almost
not affected by oil extract, indicating that the matrix effect can
be ignored using the sample preparation procedure (10-times
diluted oil samples) provided in our study. The other three
CPCs including DCPC, NDCPC, and N-V were also detected
in the matrix, and their IC50 values were 0.19, 0.36, and 0.81
ng/mL, respectively. By calculating the mean concentration of
20 blank oil samples plus three times the standard deviation,
the limits of detection (LODs) of the icELISA for four CPCs
(CPC, DCPC, NDCPC, and N-V) were determined as 0.73,
0.77, 1.46, and 3.29 μg/kg in oil, respectively, which are far
below the reported LODs for CPCs by other reports (Table
S1).16,17,26
The accuracy and precision of the developed icELISA were
evaluated by measuring the recovery and coefficient of
variation (CV) of spiked oil samples. CPC-negative oil samples
were spiked with four individual CPCs (CPC, DCPC,
NDCPC, and N-V) at three different concentrations (Table
2). Recoveries ranging from 76.5 to 114.3% were obtained,
with a CV lower than 10.5% in oil samples, which
demonstrates the reliability of the icELISA for the detection
of CPCs in oil.
ASSOCIATED CONTENT
■
sı
* Supporting Information
Comparison of icELISA and HPLC−MS/MS Analysis.
Although ELISA provides a relatively simple and rapid method
compared with traditional instrumental methods, confirmatory
research is still needed because of the possibility of false
positive results.28 Therefore, the HPLC−MS/MS was
developed to demonstrate the applicability of the developed
icELISA for the detection of CPCs residues in real oil samples.
The qualitative and quantitative ion pairs, regression equations,
and chromatograms of four CPCs are shown in Table S10 and
Figure S4. The established HPLC−MS/MS was validated in
terms of linearity, LODs, precision, and recoveries of four
CPCs. The results summarized in Table 2 show similar LODs
and recovery rates with icELISA. Thus, the icELISA we
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jafc.1c03657.
Sequence alignment, mass spectra and 1H NMR spectra
of hapten1−6, MALDI−TOF−MS, HPLC−MS/MS
chromatograms, reported haptens, antibodies, and
immunoassays for CPCs in the literature, information
of immunization procedures, sequences of primers,
hapten/protein molar ratios, detailed information of
antibody titers, molecular parameters of hapten1−8,
results of screening coating antigen, CR of mAbs, and
qualitative ion pairs, quantitative ion pairs, regression
equation, and R2 (PDF)
9965
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
Notes
AUTHOR INFORMATION
■
The authors declare no competing financial interest.
Corresponding Author
ACKNOWLEDGMENTS
■
Zhanhui Wang − Beijing Key Laboratory of Detection
Technology for Animal Derived Food Safety, Beijing
Laboratory for Food Quality and Safety, College of
Veterinary Medicine, China Agricultural University, Beijing
100193, People’s Republic of China; orcid.org/0000-
0002-0167-9559; Phone: +86-10-6273 4565;
This work was supported by the National Science Foundation
of China (Grant No. 32172905 and No. 31873025) and
Sanming Project of Medicine in Shenzhen (SZSM201611068).
REFERENCES
■
Email: wangzhanhui@cau.edu.cn; Fax: +86-10-6273 1032
(1) Li, Y.; Sun, Y.; Beier, R.; Lei, H.; Gee, S.; Hammock, B.; Wang,
H.; Wang, Z.; Sun, X.; Shen, Y.; Yang, J.; Xu, Z. Immunochemical
techniques for multianalyte analysis of chemical residues in food and
the environment: A review. TrAC, Trends Anal. Chem. 2017, 88, 25−
40.
(2) Lawler, K.; Kijanka, G. Clinical applications of immunoassays. In
Immunoassays: Development, Applications and Future Trends, Pan
Stanford: Singapore. 2017, pp. 161−190.
(3) Goodrow, M. H.; Hammock, B. Hapten design for compound-
selective antibodies: ELISAS for environmentally deleterious small
molecules. Anal. Chim. Acta 1998, 376, 83−91.
(4) Li, C.; Liang, X.; Wen, K.; Li, Y.; Zhang, X.; Ma, M.; Yu, X.; Yu,
W.; Shen, J.; Wang, Z. Class-Specific Monoclonal Antibodies and
Dihydropteroate Synthase in Bioassays Used for the Detection of
Sulfonamides: Structural Insights into Recognition Diversity. Anal.
Chem. 2019, 91, 2392−2400.
(5) Wen, K.; Bai, Y.; Wei, Y.; Li, C.; Shen, J.; Wang, Z. Influence of
small molecular property on antibody response. J. Agric. Food Chem.
2020, 68, 10944−10950.
(6) Seong, S. Y.; Matzinger, P. Hydrophobicity: an ancient damage−
associated molecular pattern that initiates innate immune responses.
Nat. Rev. Immunol. 2004, 4, 469−478.
(7) Li, B.; Yuan, Z.; Hung, H.; Ma, J.; Jain, P.; Tsao, C.; Xie, J.;
Zhang, P.; Lin, X.; Wu, K.; Jiang, S. Revealing the immunogenic risk
of polymers. Angew. Chem., Int. Ed. Engl. 2018, 57, 13873−13876.
(8) Li, B.; Xie, J.; Yuan, Z.; Jain, P.; Lin, X.; Wu, K.; Jiang, S.
Mitigation of inflammatory immune responses with hydrophilic
nanoparticles. Angew. Chem., Int. Ed. Engl. 2018, 57, 4527−4531.
(9) Lu, F.; Wu, X. China food safety hits the “gutter”. Food Control.
2014, 41, 134−138.
(10) Daood, H. G.; Halasz, G.; Palotas, G.; Palotás, G.; Bodai, Z.;
Helyes, L. HPLC determination of capsaicinoids with cross−linked
C18 column and buffer−free eluent. J. Chromatogr. Sci. 2015, 53,
135−143.
(11) Zhou, C.; Ma, D.; Cao, W.; Shi, H.; Jiang, Y. Fast simultaneous
determination of capsaicin, dihydrocapsaicin and nonivamide for
detecting adulteration in edible and crude vegetable oils by UPLC−
MS/MS. Food Addit. Contam., Part A 2018, 35, 1447−1452.
(12) Wang, L.; Jin, J.; Wang, S.; Wang, X.; Tian, Y.; Chen, J. A novel
method for the identification of illegal cooking oil (1): detection of
three capsaicinoids with liquid chromatography−mass spectrometry.
Chin. J. Chromatogr. 2012, 30, 1094−1099.
Authors
Yuchen Bai − Beijing Key Laboratory of Detection Technology
for Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China
Hui Jiang − Beijing Key Laboratory of Detection Technology
for Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China
Yingjie Zhang − Beijing Key Laboratory of Detection
Technology for Animal Derived Food Safety, Beijing
Laboratory for Food Quality and Safety, College of
Veterinary Medicine, China Agricultural University, Beijing
100193, People’s Republic of China
Leina Dou − Beijing Key Laboratory of Detection Technology
for Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China
Minggang Liu − Beijing Key Laboratory of Detection
Technology for Animal Derived Food Safety, Beijing
Laboratory for Food Quality and Safety, College of
Veterinary Medicine, China Agricultural University, Beijing
100193, People’s Republic of China
Wenbo Yu − Beijing Key Laboratory of Detection Technology
for Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China; orcid.org/0000-0003-0741-9640
Kai Wen − Beijing Key Laboratory of Detection Technology for
Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China
(13) Ma, F.; Yang, Q.; Matthäus, B.; Li, P.; Zhang, Q.; Zhang, L.
Simultaneous determination of capsaicin and dihydrocapsaicin for
vegetable oil adulteration by immunoaffinity chromatography cleanup
coupled with LC−MS/MS. J. Chromatogr., B: Anal. Technol. Biomed.
Life Sci. 2016, 1021, 137−144.
(14) Perkins, B.; Bushway, R.; Guthrie, K.; Fan, T.; Stewart, B.;
Prince, A.; Williams, M. Determination of capsaicinoids in salsa by
liquid chromatography and enzyme immunoassay. J. AOAC Int. 2002,
85, 82−85.
Jianzhong Shen − Beijing Key Laboratory of Detection
Technology for Animal Derived Food Safety, Beijing
Laboratory for Food Quality and Safety, College of
Veterinary Medicine, China Agricultural University, Beijing
100193, People’s Republic of China
Yuebin Ke − Key Laboratory of Molecular Epidemiology of
Shenzhen, Shenzhen Center for Disease Control and
Prevention, Shenzhen City 518055, China
Xuezhi Yu − Beijing Key Laboratory of Detection Technology
for Animal Derived Food Safety, Beijing Laboratory for Food
Quality and Safety, College of Veterinary Medicine, China
Agricultural University, Beijing 100193, People’s Republic of
China; orcid.org/0000-0002-4497-6182
(15) Yang, Q.; Zhu, J.; Ma, F.; Li, P.; Zhang, L.; Zhang, W.; Ding,
X.; Zhang, Q. Quantitative determination of major capsaicinoids in
serum by ELISA and time−resolved fluorescent immunoassay based
on monoclonal antibodies. Biosens. Bioelectron. 2016, 81, 229−235.
(16) Sun, J.; Liu, L.; Song, S.; Cui, G.; Kuang, H. Development of an
immunochromatographic strip assay for three major capsaicinoids
based on an ultrasensitive monoclonal antibody. Food Agric. Immunol.
2018, 29, 930−940.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jafc.1c03657
9966
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967

Journal of Agricultural and Food Chemistry
pubs.acs.org/JAFC
Article
(17) Kennison, K.; Gibberd, M.; Pollnitz, A.; Wilkinson, K. Smoke−
derived taint in wine: the release of smoke−derived volatile phenols
during fermentation of Merlot juice following grapevine exposure to
smoke. J. Agric. Food Chem. 2008, 56, 7379−7383.
(18) Kalaiyarasi, J.; Meenakshi, S.; Pandian, K.; Gopinath, S. C. B.
Simultaneous voltammetric determination of vanillin and guaiacol in
food products on defect free graphene nanoflakes modified glassy
carbon electrode. Microchim. Acta 2017, 184, 2131−2140.
(19) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(20) Zhao, Y.; Truhlar, D. G. The M06 suite of density functionals
for main group thermochemistry, thermochemical kinetics, non-
covalent interactions, excited states, and transition elements: two new
functionals and systematic testing of four M06−class functionals and
12 other functionals. Theor. Chem. Acc. 2008, 120, 215−241.
(21) Lu, T.; Chen, F. Multiwfn: a multifunctional wavefunction
analyzer. J. Comput. Chem. 2012, 33, 580−592.
(22) Wang, Z.; Mi, T.; Beier, R. C.; Zhang, H.; Sheng, Y.; Shi, W.;
Zhang, S.; Shen, J. Hapten synthesis, monoclonal antibody production
and development of a competitive indirect enzyme−linked
immunosorbent assay for erythromycin in milk. Food Chem. 2015,
171, 98−107.
(23) Wang, Z.; Zhu, Y.; Ding, S.; He, F.; Beier, R. C.; Li, J.; Jiang,
H.; Feng, C.; Wan, Y.; Zhang, S.; Kai, Z.; Yang, X.; Shen, J.
Development of a monoclonal antibody−based broad−specificity
ELISA for fluoroquinolone antibiotics in foods and molecular
modeling studies of cross−reactive compounds. Anal. Chem. 2007,
79, 4471−4483.
(24) Luo, S.; Chen, T.; Liu, C.; Zou, C.; Chen, C. Structural and
immunogenic effect of complete antigen linked by different spacer
arm. 2011 5th International Conference on Bioinformatics and
Biomedical Engineering, 2011, pp. 1−5.
(25) Mi, J.; Dong, X.; Zhang, X.; Li, C.; Wang, J.; Mujtaba, M. G.;
Zhang, S.; Wen, K.; Yu, X.; Wang, Z. Novel hapten design, antibody
recognition mechanism study, and a highly sensitive immunoassay for
diethylstilbestrol in shrimp. Anal. Bioanal. Chem. 2019, 411, 5255−
5265.
́
(26) Abad, A.; Montoya, A. Development of an enzyme−linked
immunosorbent assay to carbaryl. 2. assay optimization and
application to the analysis of water samples. J. Agric. Food Chem.
1997, 45, 1495−1501.
(27) Kim, Y. J.; Cho, Y. A.; Lee, H. S.; Lee, Y. T.; Gee, S. J.;
Hammock, B. D. Synthesis of haptens for immunoassay of
organophosphorus pesticides and effect of heterology in hapten
spacer arm length on immunoassay sensitivity. Anal. Chim. Acta 2003,
475, 85−96.
(28) Lu, P.; Yang, S.; Hu, D.; Ding, X.; Shi, M. Synthesis of hapten
and development of immunoassay based on monoclonal antibody for
the detection of dufulin in agricultural samples. J. Agric. Food Chem.
2013, 61, 10302−10309.
9967
https://doi.org/10.1021/acs.jafc.1c03657
J. Agric. Food Chem. 2021, 69, 9957−9967