Development of a highly sensitive and specific enzyme-linked immunosorbent assay for detection of sudan I in food samples

Article abstract of DOI:10.1021/jf071005j

A highly selective and sensitive indirect competitive enzyme-linked immunosorbent assay (ELISA) for Sudan I was developed. Two hapten derivatives with different lengths of carboxylic spacer at the azo-bound para-position were synthesized and coupled to carrier proteins. The hapten-bovine serum albumin (BSA) conjugates were used as immunogens, while the hapten-ovalbumin (OA) conjugates were applied as coating antigens. The antisera which were obtained from four immunized rabbits were characterized in terms of sensitivity and specificity. At optimal experimental conditions it was found that IC ₅₀ and LOD values of seven pairs based on four antisera and two coating antigens were in the range of 0.3-2 ng/mL and 0.02-0.1 ng/mL, respectively. The most sensitive ELISA could be established with Sudan I-propionic acid-OA coating antigen and the antiserum which was obtained with the corresponding immunogen. The cross-reactivity values of the four antisera with Sudan II, III, and IV was estimated with 0.1-14.3%. No cross-reactivity was found with six edible colorants Sunset yellow, Amarant, Kermes, Indigotin, Bright blue and Lemon yellow, indicating high specificity for Sudan I. Six food samples were fortified with Sudan I and extracted by simple sample preparation. The methanolic extracts after dilution with methanol:water (5:95, v/v) were analyzed by the developed ELISA. Assay precision and accuracy was estimated by determination of three replicates. Acceptable recovery rates of 92.5-114% and intra-assay coefficients of variation of 5.9-24.8% were obtained. The data were validated by conventional HPLC method. As revealed, both methods were highly correlated (r= 0.9851, n = 7), demonstrating the applicability of the developed ELISA for Sudan I analysis in food samples.

Full text of DOI:10.1021/jf071005j

6424 J. Agric. Food Chem. 2007, 55, 6424−6430
Development of a Highly Sensitive and Specific Enzyme-Linked
Immunosorbent Assay for Detection of Sudan I in Food
Samples
DAN HAN,† MENG YU,† DIETMAR KNOPP,§ REINHARD NIESSNER,§ MEI WU,† AND
ANPING DENG*,†
College of Chemistry, Sichuan University, Chengdu, 610064, China, Chair for Analytical Chemistry
and Institute of Hydrochemistry, Technische Universita¨t Mu¨nchen, Marchioninistrasse 17,
D-81377 Mu¨nchen, Germany
A highly selective and sensitive indirect competitive enzyme-linked immunosorbent assay (ELISA)
for Sudan I was developed. Two hapten derivatives with different lengths of carboxylic spacer at the
azo-bound para-position were synthesized and coupled to carrier proteins. The hapten-bovine serum
albumin (BSA) conjugates were used as immunogens, while the hapten-ovalbumin (OA) conjugates
were applied as coating antigens. The antisera which were obtained from four immunized rabbits
were characterized in terms of sensitivity and specificity. At optimal experimental conditions it was
found that IC50 and LOD values of seven pairs based on four antisera and two coating antigens were
in the range of 0.3-2 ng/mL and 0.02-0.1 ng/mL, respectively. The most sensitive ELISA could be
established with Sudan I-propionic acid-OA coating antigen and the antiserum which was obtained
with the corresponding immunogen. The cross-reactivity values of the four antisera with Sudan II, III,
and IV was estimated with 0.1-14.3%. No cross-reactivity was found with six edible colorants Sunset
yellow, Amarant, Kermes, Indigotin, Bright blue and Lemon yellow, indicating high specificity for Sudan
I. Six food samples were fortified with Sudan I and extracted by simple sample preparation. The
methanolic extracts after dilution with methanol:water (5:95, v/v) were analyzed by the developed
ELISA. Assay precision and accuracy was estimated by determination of three replicates. Acceptable
recovery rates of 92.5-114% and intra-assay coefficients of variation of 5.9-24.8% were obtained.
The data were validated by conventional HPLC method. As revealed, both methods were highly
correlated (r ) 0.9851, n ) 7), demonstrating the applicability of the developed ELISA for Sudan I
analysis in food samples.
KEYWORDS: Sudan I; food samples; analysis; polyclonal antisera; ELISA; HPLC
INTRODUCTION
wings and chicken burgers on sale and in some commercial
products such as piccalilli and chilli sauce which led to a great
panic (4).
With the social development and the rising of living standard,
people pay more attention to food quality, especially to possible
adverse health effects of additives in food products. In Europe,
the Regulation (EC) No. 178/2002 of the European Parliament
and of the Council laid down the general principles and
requirements of food law (1). In May 2003, France gave a
warning against chilli powders from India that was found to be
contaminated with Sudan I (2). In February 2005, the British
Food Standard Agency (FSA) declared the withdrawal of foods
polluted by Sudan dyes (3). In China, from February until May
2005, Sudan I was also found in batches of roasted chicken
Sudan I [1-(phenylazo)-2-naphthol] belongs to Sudan dyes
which include Sudan I, II, III, and IV (molecular structures are
shown in Figure 1). These chemicals are inexpensive fat-soluble
azo-compounds which are mainly used as coloring additives in
manufacturing of some products, such as oils, wax products,
and ball point inks since their bright and vivid colors can
improve the luster of commercial products (5). Sudan I is
classified as a carcinogen by the International Agency for
Research on Cancer (IARC) (6). There is evidence that Sudan
I is potentially carcinogenic in rodents and causes damage to
genetic material since it can react with a given sequence of DNA
in vitro (7). Therefore, Sudan I and other Sudan dyes are not
permitted as additives in food production, though they can be
used fraudulently to intensify the colors of bell pepper and chilli
powders. However, they are still illegally utilized. Therefore,
* Author to whom correspondence should be addressed . Tel +86-28-
85471302; fax +86-28-85412907; e-mail: denganping6119@yahoo.com.cn.
Current address: College of Pharmacy, University of Kentucky, 725 Rose
St., Lexington, KY 40536.
† Sichuan University.
§ Technische Universita¨t Mu¨nchen.
10.1021/jf071005j CCC: $37.00
© 2007 American Chemical Society
Published on Web 07/10/2007

ELISA for Detection of Sudan I in Food Samples
J. Agric. Food Chem., Vol. 55, No. 16, 2007 6425
Figure 1. Molecular structures of analytes which were used for cross-reactivity testing.
control of Sudan dyes in food products is very crucial and the
development of a simple, economic, and rapid detection method
is urgently required.
resulting in obvious plasmon resonance scattering signals at 452
nm, which was used to determine 0.2-2.4 µM Sudan I;
however, sensitivity was low and selectivity rather poor (4).
Immunoassays are analytical methods which are based on
the specific interaction between an antibody and corresponding
antigen. The most significant advantages of ELISAs over the
traditional instrumental methods are their high sensitivity and
specificity, simple sample preparation, high throughput, and
therefore low cost per sample. Immunoassays, especially
enzyme-linked immunosorbent assays (ELISAs) have been
widely used for the determination of both large and small
analytes such as proteins, microorganisms (20, 21), antibiotics
(22), hormones (23), pesticides (24), halogenated and aromatic
pollutants (25, 26), and even heavy metals (27) in the biological,
medical, agricultural, and environmental area.
To our knowledge, there is no immunoassay available for
the detection of Sudan I, a notorious illegal additive in food
products. In this study, Sudan I derivatives with different lengths
of linking spacer bearing a carboxylic group at the end were
synthesized and then covalently coupled to different proteins
to prepare both immunogens and coating antigens. The poly-
clonal antisera (pAb) against Sudan I raised from immunized
rabbits were characterized and used to develop a competitive
ELISA. The proposed ELISA was used for the analysis of
The standard analysis method is based on liquid chromatog-
raphy approved by the European Commission (8). Other liquid
chromatographic methods associated with different detectors
such as UV (9), APCI-MS (10, 11), ESI-MS (12), DAD (13),
DAD-ESI-MS (14), ESI-MS/MS (15), capillary LC/Q-TOF-MS
(16), CL (17), etc., for the analysis of Sudan dyes in different
food samples have been reported. Generally, chromatographic
methods are expensive and time-consuming, mainly because of
extensive sample preparation. Therefore, other techniques for
detecting Sudan dyes were developed. One approach to reduce
the costs for sample preparation is molecular imprinting. Using
methacrylic acid and 4-vinylpyridine as functional monomers
and ethylene glycol dimethacrylate as cross-linker, an imprinted
polymer was prepared for Sudan I. It could successfully be used
for the extraction of the target analyte followed by HPLC
detection (18). The method showed good selectivity; however,
the course of washing steps for complex samples was time-
consuming. Sudan I, II, III, and IV were also separated by
capillary electrophoresis combined with UV-detection, with an
LOD of 96 ng/mL (19). Recently, Huang et al. found that Sudan
dyes react with silver nitrate to produce silver nanoparticles,

6426 J. Agric. Food Chem., Vol. 55, No. 16, 2007
Han et al.
Figure 3. Synthesis of Sudan I-4-butyric acid.
2-naphthol (0.09 g) to produce Sudan I-3-propanoic acid. The total
synthesized red powder of Sudan I-3-propanoic acid was 0.18 g.
Sudan I-4-butyric acid (Sudan I-C4) was synthesized using 4-(4-
aminophenyl)butyric acid as starting compound (Figure 3), e.g., 4-(4-
aminophenyl)butyric acid (0.10 g) was reacted with 2-naphthol (0.09
g) to produce Sudan I-4-butyric acid. The total synthesized red powder
of Sudan I-4-butyric acid was 0.17 g.
Figure 2. Synthesis of Sudan I-3-propionic acid.
fortified food samples and obtained data validated by a
conventional HPLC method.
Preparation of Immunogens and Coating Antigens. The carboxy-
lic acid derivatives Sudan I-C3 and Sudan I-C4, respectively, were
conjugated to BSA and OA by the DCC/NHS ester method as described
in the literature (29). Briefly, equimolar amounts (0.15 mmol) of Sudan
I-derivative, NHS, and DCC were dissolved in 300 µL of DMF and
the mixture was incubated overnight at 25 °C. The solution was
centrifuged at 12 000 rpm (e.g., 13 400g) for 10 min and the supernatant
added slowly to 100 mg of protein dissolved in 5 mL of 0.13 M
NaHCO3 under stirring. After incubation for 4 h at 25 °C, the solution
was centrifuged and the supernatant intensively dialyzed in 0.01 M
(NH4)2CO3 for 4 days with several changes of the dialyzing buffer
solution. Finally, the Sudan I-protein conjugates were lyophilized and
stored in the refrigerator until use. While the OA conjugates served as
coating antigens the BSA conjugates were used as immunogens for
antibody preparation.
Production of Polyclonal Antisera. Two adult New Zealand rabbits
were immunized with either Sudan I-C3-BSA or Sudan I-C4-BSA,
respectively. Each animal received a dose of 1 mg of immunogen which
was dissolved in physiological saline and emulsified with the same
volume of Freund’s complete adjuvant. Injections were made intrad-
ermally at 10 sites on the back of the animal. For booster immunizations,
the same volume of Freund’s incomplete adjuvant was used. Four
subsequent injections were given at 4-week intervals. Bleeding of
animals was performed 10 days after the last immunization and the
collected antisera, named C3-I, C3-II, C4-I, and C4-II were stored at
-60 °C until use. The antisera at appropriate dilutions were directly
used for the development of the ELISA.
Indirect Competitive ELISA. An indirect competitive ELISA
format was adopted for analyzing Sudan I. The ELISA procedure was
as follows. Coating antigen stock solution (1 mg/mL) was diluted with
carbonate buffer, pH 9.8, and 200 µL/well added to a 96-well microtiter
plate. The plate was incubated overnight at 4 °C and then washed with
PBST using an automated plate washer. Some binding sites not occupied
by the coating antigen were then blocked by the blocking buffer (280
µL/well) for 1 h at room temperature. After the plate was washed as
before, standard solutions or samples (100 µL/well) and diluted
antiserum (100 µL/well) were added and incubated for 1 h at room
temperature. After washing, GaRIgG-POD was added (200 µL/well)
and the plate incubated for 1 h at room temperature. Then, the plate
was washed and the substrate solution (200 µL/well) added. After
incubation with shaking for about 20 min, sulfuric acid (5%, 80 µL/
well) was added and the absorbance measured at 450 nm using a
microplate reader. Calibration curves were constructed in the form of
(B/B0) × 100% vs log C, where B and B0 was the absorbance of the
analyte at the standard point and at zero concentration of the analyte,
respectively.
MATERIALS AND METHODS
Chemicals. 4-(4-Aminophenyl)butyric acid, N,N′-dicyclohexylcar-
bodiimide (DCC), N-hydroxysuccinimide (NHS), dimethylformamide
(DMF), dimethyl sulfoxide (DMSO), 3,3′,5,5′-tetramethylbenzidine
(TMB), Tween-20, Sudan I, II, III, IV (HPLC grade), methanol (MeOH)
(HPLC grade), casein, bovine serum albumin (BSA), ovalbumin (OA),
goat anti-rabbit IgG-horseradish peroxidase conjugate (GaRIgG-POD),
and complete and incomplete Freund’s adjuvants were purchased from
Sigma (St. Louis, MO). 4-Nitrobenzaldehyde, malonic acid, pyridine,
sulfuric acid, acetic acid, 30% hydrogen peroxide, palladium-C catalyst
(5%), silica gel G and F, physiological saline, dichloromethane (DCM),
2-naphthol, and hydrochloric acid were purchased from Chang Zhen
Chemical Company (Chengdu, China). Sunset yellow, Amarant,
Kermes, Indigotin, Bright blue, and Lemon yellow used for testing
cross-reactivity were generous gifts from the Center of Disease Control
of Sichuan Province, China.
Apparatus. ELISA reader Sunrise Remote/Touch Screen, Columbus
plus (Tecan, Gro¨dig, Austria), microtiter plate washer M12/2R,
Columbus plus, (Tecan), spectrophotometer UV-2300 (Techcomp,
Shanghai, China), microtiter plate shaker KJ-201C (Oscillator, Jiangsu
Kangjian Medical Apparatus, Co., Ltd., China), pH meter PS-10
(Sartorius, Go¨ttingen, Germany), electronic balance BS 124S (Sartorius).
Microtiter plates were from Jiangsu (Haimen, China). Deionized-RO
water supply system DZG-303A was purchased from AK Company
(Chengdu, Joint Company between Chengdu and Taiwan).
Buffers and Solutions. (1) Coating buffer: 0.05 M carbonate buffer,
pH 9.8; (2) coating antigen stock solution: 1 mg/mL of coating antigen
prepared with coating buffer; (3) assay buffer: 0.01 M phosphate-
buffered saline (PBS) pH 7.4, containing 145 mM NaCl; (4) washing
buffer (PBST): assay buffer with 0.1% (v/v) of Tween-20; (5) blocking
solution: 1% of casein in assay buffer; (6) acetate buffer: 100 mM
sodium acetate acid buffer, pH 5.7; (7) substrate solution (TMB +
H2O2): 200 µL of 10 mg/mL TMB dissolved in DMF, 20 µL of 5%
H2O2, and 1 mL of acetate buffer were added to 20 mL of pure water;
(8) stop solution: sulfuric acid (5%).
Sudan I stock solution (1 mg/mL) was made by dissolving Sudan I
red powder in DMF. Sudan I standard solutions at the concentrations
of 0.1, 0.3, 1.0, 3.0, 10, 30, and 100 ng/mL were prepared by diluting
the stock solution with MeOH:water (5:95, v/v).
Synthesis of Sudan I Derivatives. Sudan I-3-propanoic acid (Sudan
I-C3) was synthesized in three steps as shown in Figure 2. Briefly,
3-(4-nitrophenyl)acrylic acid was first synthesized by reacting malonic
acid (5.72 g) and 4-nitrobenzaldehyde (7.50 g) in freshly distilled
pyridine with acetamide as a catalyst. Then the synthesized 3-(4-
nitrophenyl)acrylic acid (0.25 g) was reduced by hydrogen with
palladium-C catalyst (28), producing 3-(4-aminophenyl)propanoic (0.15
g). 3-(4-Aminophenyl)propanoic acid (0.10 g) was reacted with
Cross-Reactivity. The specificity of the generated antisera was
investigated by cross-reactivity (CR) experiments. Nine compounds
were selected (Figure 1) for testing CR. Sudan II, Sudan III, and Sudan
IV are fat-soluble, and their use for food production is prohibited. The
remaining six chemicals are water-soluble and as edible additives widely

ELISA for Detection of Sudan I in Food Samples
J. Agric. Food Chem., Vol. 55, No. 16, 2007 6427
Table 1. Optimized ELISA Parameters Including Coating Antigen, Antibody, and Peroxidase-Labeled Second Antibody (GaRIgG-HRP) and the
Corresponding IC50 Value
coating antigen
dilution
antibody
GaRIgG-HRP
dilution
IC50c (ng/mL)
name
ng/well
name
dilution
C3
−
OAa
1:10000
1:30000
1:50000
1:30000
1:5000
1:10000
1:10000
1:30000
20
6.7
4
6.7
40
20
20
6.7
C3-I
C3-II
C4-I
C4-II
C3-I
C3-II
C4-I
C4-II
1:100000
1:150000
1:1000000
1:70000
1:25000
1:50000
1:100000
1:50000
1:15000
1:20000
1:30000
1:20000
1:20000
1:20000
1:15000
1:20000
0.45
0.3
>100
−
1.7
−1.5
0.7−
0.7−
1.0−
0.7−
0.7−
2.0
2.0
1.5
1.5
1.5
C4
−
OAb
−OA: Sudan I-C3− −OA: Sudan I-C4− ) 6) which were run on
a C3 ovalbumin conjugate. b C4 ovalbumin conjugate. c Values were calculated from standard curves (n
6 consecutive days.
separate extractions were performed and each sample was determined
in triplicate. Unspiked samples were extracted in the same way and
used as blanks.
HPLC-UV Analysis. Analysis was performed according to the
procedure of Calbiani et al. (15). Sudan I standard solutions or sample
extracts were passed through a 0.45 µm filter prior to HPLC detection.
A HPLC system (Alltech) with a C18 column (250 mm × 4.6 mm,
5.0 µm particle size, Alltech) was equilibrated with mobile phase
consisting of 0.1% formic acid in MeOH and 0.1% aqueous formic
acid (85:15, v/v) at a flow rate of 1 mL/min. The volume of standard
or extract in each analysis was 20 µL. Sudan I was monitored at 480
nm by UV detector. The HPLC work station software (Alltech) was
used for the instrument control and data analysis. Peak areas were used
for quantification. The calibration curve for Sudan I was constructed
with standards of 0.2, 0.5, 1.0, 2.0, 5.0, and 10 µg/mL.
RESULTS AND DISCUSSION
Figure 4. ELISA standard curves for Sudan
I
with homogeneous
coating antigen C3−
Characterization of Sudan I Derivatives and Protein
Conjugates. For the production of high quality antibodies and
the development of highly sensitive and specific immunoassays,
it is inevitably necessary to modify the molecular structure of
the target analyte properly, e.g., to synthesize a hapten derivative
which bears a spacer to be coupled covalently to the carrier
protein. Both the position at the hapten molecule where the
spacer is attached and the length of the spacer may play an
important role for a successful antibody generation. Generally,
the molecular structure of the hapten should be left unchanged
as far as possible, and further a spacer length of 3-5 C-atoms
is considered most suitable (30, 31). Concluding, in this study
the benzene ring of Sudan I was modified at the para position
of the azo bond, i.e., the spacer is attached as far as possible
from the naphthol moiety. Two Sudan I-derivatives with
different spacer length were synthesized (Figures 2 and 3). The
structures of Sudan I-C3 and Sudan I-C4 were confirmed by
the NMR method. The data from the NMR method are as
follows: Sudan I-C3 1H NMR (200 MHz, CDCl3): δ 14.30 (b,
1 H), 8.55 (d, J ) 4.2 Hz, 1 H), 7.73-7.50 (m, 4 H), 7.41-
7.14 (m, 4 H), 6.89 (d, J ) 3.96, 1 H), 3.45 (b, 1 H), 3.05 (t,
combinations of coating antigen and antiserum.
1
OA, 1:30 000 (e.g. 6.7 ng/well); antibody C3-I, 1:150 000; GaRIgG-HRP,
1:20 000; IC50 0.64 ng/mL. coating antigen C3 OA, 1:30 000 (e.g. 6.7
ng/well); antibody C3-II, 1:150 000; GaRIgG-HRP 1:20 000; IC50 0.35 ng/
mL; coating antigen C4 OA, 1:10 000 (e.g. 20 ng/well); antibody C4-I;
1:100 000; GARIgG-HRP, 1:20 000; IC50 0.70 ng/mL; coating antigen
C4 OA 1:30 000 (e.g. 6.7 ng/well); antibody C4-II, 1:150 000; GARIgG-
9
−
b
−
2
−
HRP 1:20 000; IC50 0.76 ng/mL. Each point represents the mean of
SD of three replicates.
±
used in food manufacturing. Standard solutions were prepared in the
concentration range of 0.001-1000 ng/mL with MeOH:water (5:95,
v/v) and applied to the ELISA. CR was expressed as percent IC50 values
based on 100% response of Sudan I. The IC50 can be considered a
measure (inverse) of the affinity of an antibody for a given analyte.
Food Samples. Six different food samples, chilli sauce A, chilli sauce
B, chilli powder, tomato sauce, preserved tomato juice, and fresh
tomatoes were collected from local supermarket in Chengdu (China).
While fresh tomatoes had to be homogenized before spiking with Sudan
I, the other five samples were directly used.
Details of the fortification experiment were listed in Table 3. Briefly,
1-5 g of sample was weighed into a glass tube with a glass stopper.
Appropriate amount of Sudan I dissolved in MeOH was added into
the tube to prepare a final concentration of 1-10 µg Sudan I/g of
sample. The fortified samples were vigorously shaken for 10 min and
kept at 4 °C overnight. Then, 5-50 mL of MeOH was added into the
tube. The extraction was performed by sonification for 20 min, followed
by centrifugation at 12 000 rpm (e.g., 13 400 g) for 5 min. The
supernatant was collected and measured by ELISA and HPLC. For
ELISA analysis, based on individual sample type, the extracts were
diluted at 1:100 to 1:1000 with MeOH:water (5:95, v/v) and directly
applied to ELISA using the combination C3-OA coating antigen and
C3-II antiserum (see Table 1). For HPLC detection, the extracts were
filtered by a 0.45 µm filter cellulose acetate membrane filter (Alltech,
Unterhaching, Germany) before injection. For each sample, three
1
J ) 7.50 Hz, 2 H), 2.76 (t, J ) 7.46 Hz, 2 H); Sudan I-C4 H
NMR (200 MHz, CDCl3): δ 14.32 (b, 1 H), 8.58 (d, J ) 8.57
Hz, 1 H), 7.75-7.51 (m, 4 H), 7.41-7.09 (m, 4 H), 6.90 (d, J
) 6.89, 1 H), 3.51 (b, 1 H), 2.76 (t, J ) 7.46 Hz, 2 H), 2.44 (t,
J ) 6.86 Hz, 2 H), 2.07 (m, 2 H).
All conjugates (Sudan I-C3-BSA, Sudan I-C4-BSA, Sudan
I-C3-OA, and Sudan I-C4-OA) were characterized by UV-
spectroscopy. The estimated molar ratio of Sudan I-C3-BSA
and Sudan I-C4-BSA was 37 and 17, respectively. A ratio of
20 was observed for both OA conjugates.
Optimization of ELISA Conditions. To improve the sen-
sitivity, the assay conditions including type and concentration

6428 J. Agric. Food Chem., Vol. 55, No. 16, 2007
Han et al.
II, C4-I, and C4-II), eight possible combinations were tested.
The experimental conditions including coating antigen concen-
tration and dilution of antiserum and secondary antibody were
optimized for each combination (Table 1). With exception of
the pair C3-OA coating antigen and C4-I antiserum which
could be used at very high dilutions, with all combinations quite
comparable optimal conditions were observed. In detail, coating
antigen concentration and antiserum and secondary antibody
dilutions varied in the range of 6.7-40 ng/mL, 1:25 000-1:
150 000 and 1:15 000-1:20 000, respectively. The calculated
IC50 values were lesser than 2 ng/mL. With an IC50 value of
0.3-1.5 ng/mL, the combination of C3-OA conjugate and C3-
II antiserum was superior. The aberrant data obtained with pair
C3-OA/C4-I are difficult to interpret. Obviously, there is a great
amount of high affinity C3-OA-recognizing antibodies present
in the generated antiserum which leads to significantly lower
affinity for Sudan I compared to other tested combinations. In
addition, no significant differences in assay sensitivity were
observed between homogeneous and heterogeneous combina-
tions of coating antigens and antisera. This can easily be
followed from Figures 4 and 5, in which typical standard curves
for Sudan I were constructed in the concentration range of 0.1-
100 ng/mL and the values of LOD at a signal-to-noise ratio of
3 (S/N ) 3) was within 0.02-0.1 ng/mL.
Specificity of the Antisera. Specificity was evaluated by CR
testing. Obtained CR values were summarized in Table 2. It
became obvious that the CR pattern was very similar for all
antisera. Accordingly, no recognition of the tested edible
colorants was observed, i.e., the CR was below 0.01%. Even
the chemicals Sunset Yellow and Amaranth which share both
the naphthol ring and attached azo group with Sudan I were
not recognized by the antibodies because of the large sulfonic
acid group(s), obviously. Further, Sudan I showed by far the
highest binding followed by Sudan III with 5.0-14.3%. This
is not surprising because these two compounds exhibit highest
structural homology among all Sudan dyes. With a CR of only
0.1-5.3% for Sudan II and Sudan IV these compounds were
less recognized. This should be caused by the methyl group(s)
attached in ortho and para positions to the linking azo group.
To summarize, high specific antisera were obtained against
Sudan I after immunization with Sudan I derivatives which
contained either a C3- or C4-spacer arm attached to the benzene
ring.
Figure 5. ELISA standard curves for Sudan
combinations of coating antigen and antiserum.
I
2
with heterogeneous
coating antigen C3−
OA, 1:30000 (e.g. 6.7 ng/well); antibody C4-I, 1:70 000; GaRIgG-HRP,
1:20 000; IC50 0.70 ng/mL coating antigen C4 OA, 1:10 000 (e.g. 20
ng/well); antibody C3-I, 1:50 000; GaRIgG-HRP 1:20 000; IC50 0.75 ng/
mL; coating antigen C4 OA, 1:10 000 (e.g. 20 ng/well); antibody C3-
II, 1:50 000; GaRIgG-HRP 1:20 000; IC50 1.0 ng/mL. Each point represents
the mean of SD of three replicates.
b
−
9
−
±
Table 2. Cross-reactivity of the Polyclonal Antisera C3-I, C3-II, C4-I,
and C4II with Sudan I
−
IV and Other Six Edible Colorantsa
CR (%)
compound
C3-I
100
C3-II
100
C4-I
100
C4-II
100
Sudan I
Sudan II
5.3
0.5
1.4
3.4
Sudan III
5.8
14.3
5.0
9.5
Sudan IV
Sunset Yellow
Amaranth
Kermes
Indigotin
Bright Blue
Lemon Yellow
2.4
0.5
0.1
1.9
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
a Note: C3-I and C3-II are antisera obtained from two rabbits immunized with
immunogen of Sudan I-C3 BSA; while C4-I and C4II are antisera obtained from
the rabbits immunized with Sudan I-C4 BSA.
−
−
of the coating antigen, dilution of the antiserum and secondary
labeled antibody, etc. should be carefully optimized. In the
present investigation, it was performed according to two criteria,
i.e (1). to get an IC50 value as low as possible and (2) an
absorbance in the range of 0.8-1.5 absorption units for the zero
standard concentration (blank).
Fortification Experiment. In this study, six food samples
were fortified with Sudan I at a concentration of 1-10 µg/g of
sample and extracted with pure MeOH. Unspiked samples
served as blanks. For the determination with ELISA, the extracts
were 100-1000-fold diluted with MeOH:water (5:95, v/v)
depending on the individual sample (Table 3). Generally, a
simple dilution step is a good means to eliminate or reduce the
In this study, based on two different coating antigens (Sudan
I-C3-OA and Sudan I-C4-OA) and four antisera (C3-I, C3-
Table 3. Concentration of Sudan I in Spiked Food Samples as Determined by ELISA (n
) 3)
weight of
sample (g)
Sudan I
spiked concn
volume of
MeOH (mL)
dilution of
extract
concn measured
SD ( g/g)
intra-assay
RSD (%)
recovery
(%)
sample type
fresh
added (
µ
g)
(µ
g/g)
±
µ
1
1
1
5
100
1.14
±
0.09
7.9
114.0
tomato juice
preserved
1
2
2
5
200
2.07
±
0.22
10.6
103.5
tomato juice
tomato sauce
chilli sauce A
chilli sauce B
chilli powder
1
5
5
1
2
10
10
10
2
2
2
5
50
50
10
100
100
400
2.06
2.08
3.27
9.25
±
±
±
±
0.51
0.39)
0.76
0.55
24.8
18.8
23.2
5.9
103.0
104.0
103.0a
92.5
10
1000
a The concentration of Sudan I in unspiked chilli sauce B as measured by ELISA was 1.21
103.0%. No detectable Sudan I was found in the other blank samples.
µg/g of sample. Thus the recovery rate was calculated as: (3.27 − 1.21)/2
)

ELISA for Detection of Sudan I in Food Samples
J. Agric. Food Chem., Vol. 55, No. 16, 2007 6429
(5) Moller, P.; Wallin, H. Genotoxic hazards of azo pigments and
other colorants related to 1-phenylazo-2-hydroxynaphthalene.
Mutat. Res. 2000, 462, 13-30.
(6) Stiborova, M.; Martinek, V.; Rydlova, H.; Hodek, P.; Frei, E.
Sudan I is a potential carcinogen for humans: Evidence for its
metabolic activation and detoxication by human recombinant
cytochrome P450 1A1 and liver microsomes. Cancer Res. 2002,
62, 5678-5684.
(7) Stiborova, M.; Asfaw, B.; Frei, E.; Schmeiser, H. H.; Wiessler,
M.; Hradec, J. Mechanism of formation and P-32 postlabeling
of DNA adducts derived from peroxidative activation of
carcinogenic non-aminoazo dye 1-phenylazo-2-hydroxynaph-
thalene (Sudan-1). Carcinogenesis 1990, 11, 1843-1848.
(8) European Commission, Health and Consumer Protection Direc-
torate General, Committee IV-Food Safety in Production and
Currency, Part III-Physical and Chemical Substances Surveil-
lance. New Method Declaration, March 1999.
(9) Mazzetti, M.; Fascioli, R.; Mazzoncini, I.; Spinelli. G.; Morelli,
I.; Bertoli. A. Determination of 1-phenylazo-2-naphthol (Sudan
I) in chilli powder and in chilli-containing food products by GPC
clean-up and HPLC with LC/MS confirmation. Food Addit.
Contam. 2004, 21, 935-941.
Figure 6. Correlation of ELISA and HPLC analysis of fortified food
samples.
matrix effect on the assay. The found matrix effects were
different among the collected samples. Correspondingly, fresh
tomato juice, preserved tomato juice, and chilli sauce A could
be analyzed after only 100-fold dilution while tomato sauce,
chilli sauce B, and chilli powder had to be diluted at 1:200,
1:400, and 1:1000, respectively. Regarding blanks only in chilli
sauce B Sudan I was found at a concentration of 1.21 µg/g of
sample. Other samples were not contaminated. Assay precision
and accuracy was estimated by measuring three replicates.
Acceptable recovery rates of 92.5-114% and intra-assay
coefficients of variation of 5.9-24.8% were obtained. On the
other hand, for six fortified samples, the inter-assay coefficients
of variation were less than 30%.
(10) Tateo, F.; Bononi, M. Fast determination of Sudan I by HPLC/
APCI-MS in hot chilli, spices, and oven-baked foods. J. Agric.
Food Chem. 2004, 52, 655-658.
(11) Di Donna, L.; Maiuolo, L.; Mazzotti, F.; De Luca, D.; Sindona,
G. Assay of Sudan I contamination of foodstuff by atmospheric
pressure chemical ionization tandem mass spectrometry and
isotope dilution. Anal. Chem. 2004, 76, 5104-5108.
(12) Chen, H. W.; Zhang, X.; Luo, M. B. Desorption electrospray
ionization mass spectrometry for fast detection of Sudan dyes
in foods without sample pretreatment. Chin. J. Anal. Chem. 2006,
34, 464-468.
(13) Cornet, V.; Govaert, Y.; Moens, G.; Van Loco, J.; Degroodt, J.
M. Development of a fast analytical method for the determination
of Sudan dyes in chilli- and curry-containing foodstuffs by high
performance liquid chromatography photodiode array detection.
J. Agric. Food Chem. 2006, 54, 639-644.
Comparison of ELISA and HPLC Determination. The
HPLC calibration curve for Sudan I was constructed in the range
of 0.2-10 µg/mL with an LOD of 0.5 ng/mL (n ) 3). The
linear equation was y ) 59875x + 3348.5. To validate the data
obtained by ELISA all fortified food samples including the
unspiked chilli sauce B were also measured by HPLC. With
the selected parameters (see Materials and Methods) the
retention time of Sudan I was 24 min. As revealed, both methods
were highly correlated (r ) 0.9851, n ) 7) (Figure 6).
(14) Ma, M.; Luo, X. B.; Chen, B.; Sub. S. P.; Yao, S. Z.
Simultaneous determination of water-soluble and fat-soluble
synthetic colorants in foodstuff by high performance liquid
chromatography-diode array detection-electrospray mass spec-
trometry. J. Chromatogr. A 2006, 1103, 170-176.
(15) Calbiani, F.; Careri, M.; Elviri, L.; Mangia, A.; Pistara, l.;
Zagnoni, I. Development and in-house validation of liquid
chromatography-electrospray-tandem mass spectrometry method
for the simultaneous determination of Sudan I, Sudan II, Sudan
III and Sudan IV in hot chilli products. J. Chromatogr. A 2004,
1042, 123-130.
(16) Calbiani, F.; Careri, M.; Elviri, L.; Mangia, A.; Zagnoni, I.
Accurate mass measurements for the confirmation of Sudan azo-
dyes in hot chilli products by capillary liquid chromatography-
electrospray tandem quadrupole orthogonal-acceleration time of
flight mass spectrometry. J. Chromatogr. A 2004, 1058, 127-
135.
(17) Zhang, Y. T.; Zhang, Z. J.; Sun, Y. H. Development and
optimization of an analytical method for the determination of
Sudan dyes in hot chilli pepper by high performance liquid
chromatography with on-line electrogenerated Bro--luminol
chemiluminescence detection. J. Chromatogr. A 2006, 1129, 34-
40.
(18) Puoci, F.; Garre, C.; Iemma, F.; Muzzalupo, R.; Spizzirri, U.
G.; Picci, N. Molecularly imprinted solid phase extraction for
detection of sudan I in food matrices. Food Chem. 2005, 93,
349- 353.
Conclusion. In this work, for the first time, antibodies against
Sudan I were prepared in rabbits and used to establish an indirect
competitive ELISA. The highly sensitive assay proved useful
for the analysis of Sudan I in different food extracts. As the
main advantage, the high selectivity of the ELISA became
evident. Using the optimized assay, other Sudan dyes such as
Sudan II, III, and IV exhibited a cross-reactivity below 15%,
while six tested edible colorants of similar molecular structure
did not show any binding to the raised antibodies. In addition,
interference caused by the sample matrix could be easily
overcome by a simple dilution step before immunochemical
analysis.
LITERATURE CITED
(1) EC Regulation No 178/2002. OJ L 31 (Jan 2, 2002).
(2) Waite, S.; Hansen, D.; McGinley, M. Easy isocratic HPLC
determination of Sudan dyes. LCGC North Am. 2006, Suppl. S
Feb, 46.
(3) Food Standard Agency. http://www.food.gov.gov.uk.
(4) Wu, L. P.; Li, Y. F.; Huang. C. Z.; Zheng. Q. Visual detection
of Sudan dyes based on the plasmon resonance light scattering
signals of silver nanopartices. Anal. Chem. 2006, 78, 5570-
5577.
(19) Mejia, E.; Ding, Y. S.; Mora, M. F.; Garcia, C. D. Determination
of banned sudan dyes in chili powder by capillary electrophoresis.
Food Chem. 2007, 102, 1027-1033.

6430 J. Agric. Food Chem., Vol. 55, No. 16, 2007
Han et al.
(20) Castro, M. C.; Garc_, R.; Rodr_, G. J.; Rodr_, G.; Marina, M.
L. Determination of soybean proteins in commercial heat-
processed meat products prepared with chicken, beef or complex
mixtures of meats from different species. Food Chem. 2007, 100,
468-476.
(21) Li, X. J.; Du, A. F.; Cai, W. M.; Hou, Y. H.; Pang, L. H.; Gao,
X. Evaluation of a recombinant excretory secretory Haemonchus
contortus protein for use in a diagnostic enzyme-linked immu-
nosorbent assay. Exp. Parasitol. 2007, 115, 242-246.
(22) Zhang, H. Y.; Duan, Z. J.; Wang, L.; Zhang, Y.; Wang, S. Hapten
synthesis and development of polyclonal antibody-based multi-
sulfonamide immunoassays. J. Agric. Food Chem. 2006, 54,
4499-4505.
(27) Knopp, D. Detection of trace metal ions in environmental
matrices by immunological methods - a review. Ecol. Chem.
Eng. 2006, 13, 383-397.
(28) Taylor, S. D.; Chen, M. J.; Dinaut, A. N.; Batey, R. A. A general
method for the synthesis of O-alkyl N,O′-arylphosphoramidates
and its application to the synthesis of a transition state analogue
for carbamate hydrolysis. Tetrahedron 1998, 54, 4223-4242.
(29) Franek, M.; Deng, A. P.; Kolar, V.; Socha, J. Direct competitive
immunoassays for the coplanar polychlorinated biphenyls. Anal.
Chim. Acta 2001, 444, 131-142.
(30) Sherry, J. P. Environmental chemistry - the immunoassay option.
Crit. ReV. Anal. Chem. 1992, 54, 217-300.
(23) Li, Z. L.; Wang, S.; Lee, N. A.; Allan, R. D.; Kennedy, I. R.
Development of a solid-phase extraction-enzyme-linked immu-
nosorbent assay method for the determination of estrone in water.
Anal. Chim. Acta 2004, 503, 171-177.
(24) Gao, H. B.; Ling, Y.; Xu, T.; Zhu, W. W.; Jing, H. Y.; Sheng,
W.; Li, Q. X.; Li, J. Development of an enzyme-linked
immunosorbent assay for the pyrethroid insecticide cyhalothrin.
J. Agric. Food Chem. 2006, 54, 5284-5291.
(25) Deng, A. P.; Kolar, V.; Franek, M. Direct competitive ELISA
for the determination of polychlorinated biphenyls in soil
samples. Anal. Bioanal. Chem. 2002, 373, 685-690.
(26) Matschulat, D.; Deng, A. P.; Niessner, R.; Knopp, D. Develop-
ment of a highly sensitive monoclonal antibody based ELISA
for benzo[a]pyrene. Analyst 2005, 130, 1078-7086.
(31) Marco, M-P.; Gee, S.; Hammock, B. D. Immunochemical
techniques for environmental analysis, 2. Antibody production
and immunoassay development. Trends Anal. Chem. 1995, 14,
415-425.
Received for review April 5, 2007. Revised manuscript received May
18, 2007. Accepted May 23, 2007. This research work was financially
supported by grants from the Scientific Research Foundation for the
Returned Overseas Chinese (No. 2006331-11-3) and from the Promotion
ProgramFoundationofSichuanUniversityofChina(No.0082204127090).
JF071005J