A strip liposome immunoassay for aflatoxin B1.

Article abstract of DOI:10.1021/ac010903q

A technique has been developed for the preparation of aflatoxin B1 (AFB1)-tagged liposomes encapsulating a visible dye. These liposomes have several useful potential analytical applications, one of which is demonstrated. A simple plastic-backed nitrocellulose strip is the basis for an assay for detecting AFB1. Samples containing aflatoxin B1 are allowed to migrate by capillary action along the strip into a zone containing immobilized antibodies; then aflatoxin B1-tagged, dye-containing liposomes are allowed to migrate into the same area, filling any remaining antibody sites. The liposomes that bound to the antibody zone exhibit an intense purplish pink color whose optical density is inversely proportional to the aflatoxin concentration in the sample. The device is capable of detecting aflatoxin B1 at levels down to 20 ng and could serve as a rapid procedure for visual screening of agricultural and food samples for AFB1 or, with densitometry, as an inexpensive quantitative assay.

Full text of DOI:10.1021/ac010903q

Anal. Chem. 2002, 74, 1493-1496
A Strip Liposome Immunoassay for Aflatoxin B1
J.-A. A. Ho*,†,‡ and R. D. Wauchope§
University of Georgia Coastal Plain Experiment Station, Tifton, Georgia 31793, Agricultural Research Service, U.S.
Department of Agriculture, Tifton, Georgia 31793, and Department of Applied Chemistry, National Chi-Nan University, Puli,
Nantou, 545 Taiwan
A liposome immunoassay (LIA), using fluorescent dye-loaded
liposomes whose surface has been tagged with the antigen of
interest, has been developed for a variety of assays, particularly a
flow injection automated procedure.9-13 As part of a project on
the development of a flow injection liposome immunoanalysis
(FILIA) system for AFB1, a procedure for tagging liposomes with
AFB1 using a bridging molecule was developed to overcome
hydrophobic disruption of the liposome structure. These lipo-
somes proved useful in a preliminary development of a test strip
semiquantitative assay for AFB1 that has potential as a rapid and
inexpensive field test.
A technique has been developed for the preparation of
aflatoxin B1 (AFB1 )-tagged liposomes encapsulating a
visible dye. These liposomes have several useful potential
analytical applications, one of which is demonstrated. A
simple plastic-backed nitrocellulose strip is the basis for
an assay for detecting AFB1 . Samples containing aflatoxin
B1 are allowed to migrate by capillary action along the strip
into a zone containing immobilized antibodies; then afla-
toxin B1 -tagged, dye-containing liposomes are allowed to
migrate into the same area, filling any remaining antibody
sites. The liposomes that bound to the antibody zone
exhibit an intense purplish pink color whose optical
density is inversely proportional to the aflatoxin concen-
tration in the sample. The device is capable of detecting
aflatoxin B1 at levels down to 2 0 ng and could serve as a
rapid procedure for visual screening of agricultural and
food samples for AFB1 or, with densitometry, as an
inexpensive quantitative assay.
EXPERIMENTAL SECTION
Reagents and Materials. All inorganic chemicals and organic
solvents used were reagent grade or better. AFBl, polyclonal
antibody against AFB1, cholesterol, poly(vinylpyrrolidone) (PVP,
MW 10 000), Tween-20, gelatin, Sephadex G-50, triethylamine
(TEA), Trizma base (tris[hydroxymethyl]aminomethane, Tris),
and n-octyl â-D-glucopyranoside (OG) were purchased from Sigma
(St. Louis, MO). Dipalmitoylphosphatidylethanolamine (DPPE)
and sulforhodamine B were purchased from Molecular Probes
(Eugene, OR). Polycarbonate syringe filters of 3-, 0.4-, and 0.2-
µm pore size were purchased from Poretics (Livermore, CA), and
plastic-backed nitrocellulose membranes with pore size of >3 µm
were from Schleicher and Schuell (Keene, NH). Dipalmitoylphos-
phatidylcholine (DPPC) and dipalmitoylphosphatidylglycerol
(DPPG) were obtained from Avanti Polar Lipids (Alabaster, AL).
N-Succinimidyl-S-acetylthioacetate (SATA), 4(4-N-maleimidophe-
nyl)butyric acid hydrazide (MPBH), and N-ethylmaleimide (NEM)
were purchased from Pierce (Rockford, IL). Carnation nonfat dry
milk (NDM) powder was obtained locally.
There is a need for rapid and inexpensive field assays for
screening food and animal feed samples for mycotoxins. Immu-
noassays, which offer sensitivity, speed, and simplicity of opera-
tion, provide potential solutions for this need. Radioimmuno-
assays,1-2 antibody affinity columns,3 and enzyme-linked immuno-
sorbent assays (ELISA)4-7 have been developed for aflatoxin B1
(AFB1). Some of the immunoassays have been approved and
adopted by the AOAC.8 The Aflatest (Vicam L.P., Watertown, MA)
immunoaffinity column is widely used in the USDA and com-
modities at points of sale. However, the cost for a nonreusable,
single affinity column is about $7 U.S.
P reparation of Test Strips. A microprocessor-controllcd TLC
sample applicator, Linomat IV (CAMAG Scientific Inc., Wrights-
ville Beach, NC), was used to immobilize AFB1 antibody on the
plastic-backed nitrocellulose membranes. The membrane was cut
into 8 × 15.7 cm sheets, which were then mounted on a mobile
platform that moved at a constant rate in front of the airbrush
used to spray the antibody solution at concentration of 2 mg/ mL.
If a TLC sample applicator is not available, an alternative procedure
may be used in which antibody solution is manually dot-blotted
* Corresponding author. Department of Applied Chemistry, National Chi-Nan
University, No. 1 University Rd., Puli, Nantou, 545 Taiwan. Fax: +886-49-2917956.
E-mail: jah@nanu.edu.tw.
† University of Georgia Coastal Plain Experiment Station.
‡ National Chi-Nan University.
§ U.S. Department of Agriculture.
(1) Langone, J. J.; Van Vunakis, H. J. Nat. Cancer Inst. 1 9 7 6 , 56, 591-595.
(2) Sun, P. S.; Chu, F. S. J. Food Saf. 1 9 7 9 , 1, 67-75.
(3) Groopman, J. D.; Trudel, L. J.; Donahue, P. R.; Marshak-Tothstein, A.;
Wogan, G. N. Proc. Natl. Acad. Sci. U.S.A. 1 9 8 4 , 81, 7728-7731.
(4) Chu, F. S.; Ueno, I. Appl. Environ. Microbiol. 1 9 7 7 , 33, 1125-1128.
(5) El-Nakib, O.; Pestka, J. J.; Chu, F. S. J. Assoc. Offic. Anal. Chem. 1 9 8 1 , 64,
1077-1082.
(9) Rule, G. S.; Palmer, D. A.; Reeves, S. G.; Durst, R. A. Anal. Proc. Incl. Anal.
Commun. 1 9 9 4 , 31 (11), 339-340.
(6) Lawellin, D. W.; Grant, D. W.; Joyce, B. K. Appl. Environ. Microbiol. 1 9 7 7 ,
34, 94-96.
(10) Edwards, A. J.; Durst, R. A. Electroanalysis 1 9 9 5 , 7 (9), 838-845.
(11) Lee, M. Y.; Durst, R. A. J. Agric. Food Chem. 1 9 9 6 , 44, 4032-4036.
(12) Ho, J. A.; Durst, R. A. Anal. Chim. Acta 2 0 0 0 , 414, 51-60.
(13) Ho, J. A.; Durst, R. A. Anal. Chim. Acta 2 0 0 0 , 414, 61-69.
(7) Pestka, T.; Scott, P. M. J. Assoc. Off. Anal. Chem. 1 9 7 7 , 40, 1027-1031.
(8) Official Methods of Analysis, 15th ed.; Association of Official Analytical
Chemists.: Arlington, VA, 1990; Section 999.06.
10.1021/ac010903q CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/06/2002
Analytical Chemistry, Vol. 74, No. 7, April 1, 2002 1493

Figure 1. Conjugation of aflatoxin B1 to liposome (drawing not to scale).
onto the membrane. The antibody coat bands were allowed to
air-dry in the hood for 5 min and further dried under vacuum at
room temperature for 1.5 h. The coated nitrocellulose sheet was
then immersed in a blocking solution consisting of 0.5% poly-
(vinylpyrrolidone) and 0.07% nonfat dry milk in Tris-buffered saline
(TBS) at pH 7.0 for 45 min on a rotating shaker, followed by drying
under vacuum for at least 6 h. The prepared sheets were then
cut into 0.5 × 8 cm strips using a paper cutter, producing strips
with an antibody zone 1.5 cm above the bottom of the strip.
P reparation of AFB1 -Tagged Liposomes Containing Dye:
Overcoming the Hydrophobicity of AFB1. To provide antigenic
sites on the surface of the liposomes, it was necessary to conjugate
AFB1 to DPPE, a lipid component of liposomes. The reaction
sequence is illustrated in Figure 1. A multistep conjugation method
was employed in which SATA was used as the first coupling agent
to introduce a sulfhydryl group into DPPE via an amide linkage.
The conjugate, DPPE-acetylthioacetate (ATA-DPPE), was then
incorporated into the liposomes, providing protected sulfhydryl
groups on the surface of the liposomes. The second cross-linker,
MPBH, which contains a carbonyl-reactive hydrazide group on
one end and a sulfhydryl-reactive maleimide on the other end,
was used to derivatize an activated aldehyde group generated on
the AFB1. AFB1 was subsequently coupled to the liposome via
maleimide-sulfhydryl interaction.
evaporations to make sure no trace of triethylamine remained.
Finally, the dry residue was reconstituted with 1 mL of chloroform.
Dye-loaded liposomes were prepared by the reversed-phase
evaporation method.14,15 The lipids mixture consisted of a 5:5:0.5:4
molar ratio of DPPC, cholesterol, DPPG, and DPPE-acetylthio-
acetate. The lipids were dissolved in 8 mL of a solvent mixture
consisting of 6:6:1 volume ratios of chloroform, isopropyl ether,
and methanol, followed by a 1-min sonication at 45 °C under
nitrogen. Then 1.4 mL of a warmed aqueous solution of 150 mM
sulforhodamine B (SRB) in TBS, adjusted to pH 7.0-7.5 with 1
N NaOH, was added. After sonication of the solution for 6 min
more, the organic solvent was removed by evaporating under
nitrogen at 45 °C, leaving a dark purple, gel-like suspension of
liposomes. An additional 2.6 mL of SRB-dye was added, followed
by another 5 min of sonication at 45 °C. Liposomes were kept
warm before passing through three different pore sizes of
prewarmed polycarbonate filters (3.0, 0.4, and 0.2 µm) to produce
a homogeneous suspension. The liposomes were separated from
unencapsulated SRB and traces of organic solvent by gel filtration
on a 1.5 × 25 cm Sephadex G-50 column at room temperature,
followed by dialysis (MWCO, 12 000-14 000) in 0.01 M Tris-HCl,
0.15 M NaCl, pH 7.0 at 4 °C in the dark.
Since there is no ready-to-use functional group in the AFB1
molecule for direct conjugation with DPPE, the derivatization of
AFB1 with a maleimide group was necessary for its attachment
to the sulfhydryl liposomes. A total of 4.8 mg of AFB1 (15.4 µmol)
was dissolved in 1 mL of acetonitrile and 1 mL of 1.0 M HCl. The
mixture was heated to 60 °C for 3 h with occasional mixing and
A 5-mg sample of DPPE (7.2 µmol) was suspended in 1 mL of
0.7% triethylamine in chloroform and sonicated for 1 min at 45 °C
under nitrogen. Approximately 3.5 mg of SATA (14 µmol) was
added to the DPPE suspension and sonicated for 3 min under
nitrogen at 45 °C. The reaction flask was capped and placed on a
shaker at room temperatue for 20 min or until the solution became
clear. Organic solvent was removed by evaporating under nitrogen
at 45 °C, followed by three reconstitutions with chloroform and
(14) Szoka, F.; Olsen, F.; Heath, T.; Vail, W.; Mayhew, E.; Papahadjopoulos, D.
Biochim. Biophys. Acta 1 9 8 0 , 601, 559-571.
(15) O’Connell, J. P.; Campbell, R.-L.; Fleming, B. M; Mercolino, T. J.; Johnson,
M. D.; McLaurin, D.-A. Clin. Chem. 1 9 8 5 , 31, 1424-1426.
1494 Analytical Chemistry, Vol. 74, No. 7, April 1, 2002

then stood overnight at room temperature. The AF hemiacetal
(AFB2R) formed was extracted into chloroform. The phenolate ion
of AFB2R was prepared by evaporating off the chloroform under
nitrogen and redissolving the residue in 0.5 mL of 20% MeOH
and 0.5 mL of 0.1 M sodium pyrophosphate, pH 10. A total of 8.4
mg of MPBH (23.1 µmol) was dissolved in 55 µL of DMSO, and
the resultant mixture was added to the solution of phenolate ion
of AFB2R. The reaction mixture was then allowed to react on a
shaker for 6 h at room temperature in the dark. Concurrently,
the acetylthioacetyl liposomes were deprotected by adding hy-
droxylamine hydrochloride (100 µmol) to obtain free sulfhydryl
groups. The maleimide-derivatized AFB1 was then reacted with
sulfhydryl liposomes overnight at 4 °C. Unreacted sulfhydryl
groups on the liposome were subsequently capped with N-
ethylmaleimide. Excess unconjugated AFB1 molecules were
removed by dialysis against TBS (pH 7.0) or by ge1 filtration on
a Sephadex G-50 column.
Table 1. Characteristics of the Liposomes
mean diameter (nm)
volume of liposome (µL)
liposome concentration (number/ mL)
SRB concentration (mM)
SRB molecules per liposome
AFB1 molecules on the liposome surface
230 nm
6.37 × 10-12
3.14 × 1012
150
5.2 × 105
1.5 × 104
Safety Consideration. Aflatoxins are a highly toxic materials
that should be handled with care. Crystalline aflatoxin B1 and
organic solvents for use in the modification and production of the
conjugated liposomes and performance of the assay were handled
in a chemical hood with surgical gloves.
RESULTS AND DISCUSSION
P reparation of AFB1 -Tagged Liposomes. An initial attempt
to conjugate AFB1 to the liposome was made by preparing a
DPPE-AFB1 conjugate and incorporating it into the liposome.
However, the liposome produced did not show any affinity to anti-
AFB1 antibody. It is conjectured that the nonpolar AFB1 tended
to fold within the hydrophobic lipid bilayer, rather than pointing
outward on the outer surface of the liposomes to provide the
antigenic epitopes. To overcome this, a multistep conjugation
method as described above was used. The spacer between AFB1
and DPPE, and the hydrophilic nature of the phenolate ion of
AFB2R, overcame the problem of AFB1 folding within the lipid
bilayer. The liposomes produced are quite stable. The AFB1
liposomes are stable at temperatures up to the boiling point in
TBS, a result also reported by others.16,17
Stability Study and Characterization of Liposomes. Sulfo-
rhodamine B is a visible as well as fluorescent dye; thus, the
stability (intactness) of lipsomes means the maintenance of their
integrity, which can be determined by measuring fluorescence
intensity before and after lysis. Almost instantaneous lysis of
liposomes was observed at room temperature when a solution of
n-octyl â-D-glucopyranoside was added; total lysis of the liposomes
was achieved by addition of OG to a final concentration of 50 mM.
For these fluorescence tests, the SRB dye was excited at 541-nm
wavelength, and the fluorescent emmission intensity measure-
ments were made at wavelength 596 nm. Temperature effects on
liposome stability were conducted by adding 10 µL of liposome
solution to 3 mL of osmotically balanced TBS, pH 7.0, in a test
tube preheated to the required temperature in a water bath. The
liposomes were incubated for 5 min before measurement of
fluorescence intensity.
Characteristics of Liposomes. Having a narrow size range
for the tagged liposomes was extremely important in order to
obtain an even capillary migration on the test strips. Extrusion of
the liposome preparations through polycarbonate filters was found
to reduce the size heterogeneity. Liposomes passed twice through
polycarbonate filters of 3.0, 0.4, and 0.2 µm, arranged in tandem,
had a mean diameter of 230 nm with standard deviation of 50
nm. The 230-nm liposomes were used in all subsequent experi-
ments.
The diameter of the liposomes was measured with a Coulter
LS particle size analyzer (Coulter Corp., Miami, FL) using the
manufacturer’s method. Liposomes were visualized by transmis-
sion electron microscopy of negatively stained preparations to
check the integrity and size of the liposomes.
Test Strip P reparation and P rocedure. AFB1 is hydrophobic
and can only be dissolved in a relatively nonpolar organic solvents
such as chloroform or acetonitrile. However, those solvents were
found to cause instability and rupture of liposomes. Thus, a
procedure was designed to operate in a stepwise procedure to
avoid the contact of liposomes with disruptive solvents. A 75-
100-µL sample of AFB1 sample (in acetonitrile/ methanol (7:3, v/ v)
mixture) solution to be assayed was dispensed into a 10 × 75
mm glass test tube. The 5 × 80 mm antibody-coated test strips
were attached at the top to a 3 × 50 mm strip of filter paper and
inserted into the tube. The strip was left in the tube long enough
(3 min) to allow the AFB1 solution to migrate into the antibody
zone. The test strip was then transferred to another test tube
containing 75 µL of diluted liposome solution (the dilution varying
according to preparation); the strip was left in the tube until the
solution front reached the upper end of the extended filter paper
(8-9 min); the strip was then removed and air-dried. The color
intensity of the antibody zone was estimated either visually or by
scanning densitometry.
The characteristics of liposomes are listed in Table 1. With
liposomes of 230-nm diameter, it is possible to calculate that the
average volume of a single liposome is 6.37 × 10-12 µL and the
volume entrapped (assuming a bilayer thickness of 4 nm) is 5.73
× 10-12 µL. Assuming the dye concentration inside of the
liposomes was equal to the original dye solution used, and
comparing the fluorescence of lysed liposomes to that of standard
SRB solutions, it is possible to calculate that there were ∼3.14 ×
1012 liposomes/ mL and that each liposome contained ∼5.2 × 105
molecules of dye. If the average surface area of the DPPC
molecules is 71 Å, and that of cholesterol 19 Å,18 it was estimated
that ∼1.5 × 104 molecules of AFB1 were on the outer surface of
each liposome given that 4 mol % of DPPE-sulfhydryl successfully
reacted with maleimide-derivatized AFB1.
(16) Kikuchi, H.; Carlsson, A.; Yachi, K.; Hirota, S. Chem. Pharm. Bull. 1 9 9 1 ,
39, 1018-1022.
(17) Rule, G. S.; Montagna, R.-A.; Durst, R. A. Anal. Biochem. 1 9 9 7 , 244, 260-
269.
(18) Israelchvili, J. N.; Mitchell, D. J. A. Biochim. Biophys. Acta 1 9 7 5 , 389, 13-
19.
Analytical Chemistry, Vol. 74, No. 7, April 1, 2002 1495

Figure 3. Aflatoxin B1 dose-response curve. Each point represents
the mean of three measurements; error bars represent (1 SD.
CONCLUSION
The U.S. Food and Drug Administration has set a maximum
allowable level of total aflatoxins at 20 parts per billion; commodi-
ties used for human and animal consumption must be tested to
ensure that aflatoxin levels are below this level. The Aflatest
immunoaffinity column is claimed to have a range of quantitation
from 1 to 50 ng with lower limit of detection of 1 ng; however,
this membrane test strip procedure is able to detect 18 ng of
aflatoxin B1, featuring a rapid and relatively inexpensive monitoring
method (∼$1 U.S./ assay) combined with instantaneous visualiza-
tion. It needs to be tested in real sample matrixes and under field
conditions to determine its limitations. In our laboratory, the total
assay time for a sample was less than 12 min, with the potential
of testing multiple samples simultaneously. These results also
suggest that AFB1-tagged, dye-loaded liposomes may be success-
fully used in a FILIA procedure. Such a technique would provide
an automated and still relatively inexpensive technique for Afla-
toxin B1.
Figure 2. Scanned image of strips showing the competition between
a series of AFB1 standards and constant amount of AFB1-tagged
liposomes. Each strip has 2.0 mg/mL anti-AFB1 antibody immobilized
on it: (1) AFB1-tagged liposomes only; (2) 10 ng of AFB1 + AFB1-
tagged liposome; (3) 100 ng of AFB1 + AFB1-tagged liposomes; (4)
1 µg of AFB1 + AFB1-tagged liposomes; (5) 10 µg of AFB1 + AFB1-
tagged liposomes.
Assay P erformance. The concept of the assay is to have an
immobilized antibody zone in the membrane strip that is first
exposed to free AFB1 in a sample solution, and then unbound
antibody sites in the zone are filled by a second exposure to a
solution of AFB1-tagged liposomes. Thus, the color exhibited by
the bound liposomes on the antibody zone is inversely proportional
to AFB1 present in the sample. Color intensity may be measured
semiquantitatively by visual examination (as might be done in the
field), but the use of a computer scanner and scan analysis
densitometry software (Biosoft, Ferguson, MO), which converts
the red coloration into gray-scale readings, can provide more
accurate quantitation. Figures 2 and 3 show the results obtained
with a series of AFB1 standards. The dose-response curve
exhibits the sigmoidal shape of competitive immunoassays. The
limit of detection estimated at 2 SD was determinated to be 18
ng for aflatoxin B1 with a 95% confidence limit.
ACKNOWLEDGMENT
The authors thank Dr. Richard A. Durst and Sui Ti Siebert of
Cornell University for valuable suggestions and assistance. This
work was supported by the American Peanut Foundation under
Grant 04-81898 and by the National Science Council in Taiwan,
ROC, under Grant 89-2113-M-260-010.
Received for review August 13, 2001. Accepted December
21, 2001.
AC010903Q
1496 Analytical Chemistry, Vol. 74, No. 7, April 1, 2002