Journal of Immunological Methods 336 (2008) 251–254
Contents lists available at ScienceDirect
Journal of Immunological Methods
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j i m
Technical note
Rapid detection of ricin in cosmetics and elimination of artifacts associated
with wheat lectin
Jacqueline Dayan-Kenigsberg a, Agnès Bertocchi a, Eric A.E. Garber b,⁎
a
Unité Biotechnologie, Biochimie des Protéines et Macromolécules, Agence Française de Sécurité Sanitaire des Produits de Santé, 143 boulevard Anatole France, 93285
Saint Denis Cedex, France
b
Food and Drug Administration, Center for Food Safety and Applied Nutrition, HFS-716, 5100 Paint Branch Pkwy., College Park, MD, 20740, USA
a r t i c l e
i n f o
Article history:
Received 19 October 2007
Received in revised form 18 April 2008
Accepted 7 May 2008
Available online 5 June 2008
Keywords:
Ricin
Detection
Lectin
Cosmetics
Lateral flow devices (LFDs)
a b s t r a c t
Ricin can be detected in cosmetics at 0.005 µg/mL in the analytical sample using lateral flow
devices (LFDs). Wheat germ, an ingredient used in skin care products is also a potential source
of wheat lectin. False positives were observed when wheat lectin was added to LFDs from two
manufacturers, irrespective of whether the LFD was specific for ricin, Staphylococcus
enterotoxin B (SEB), or botulinum toxin. In contrast, pea and peanut lectins did not cause
false positives. Substitution of the buffer supplied with the LFDs with a buffer containing 2.5%
non-fat milk powder eliminated the occurrence of false positives. This substitution increased
the LOD to 0.01 µg/mL ricin, which is an acceptable level for screening cosmetics for
contamination by ricin.
Published by Elsevier B.V.
1. Introduction
Castor beans of Ricinus communis are poisonous to people
(Franz and Jaax,1997; Knight, 1979), animals (Okoye et al., 1987;
Purushotham et al., 1986) and insects (Czapla and Johnson,
1990). The most prominent toxin in castor beans is ricin. The
inhalation, intravenous, and intraperitoneal LD50 values of ricin
in mice are approximately 3–5, 5, and 22 μg/kg body weight,
respectively. Inefficient uptake from the intestinal tract reduces
the oral toxicity to approximately 20 mg/kg body weight (Franz
and Jaax, 1997). Topological application of 50 µg ricin displayed
no toxicity with mice (Franz and Jaax, 1997).
Ricin is a class II ribosome inhibiting protein (RIP-II)
consisting of two subunits linked by a disulfide bond
(Robertus, 1988). A-chain is a 267 amino acid, 32 kDa Nglycosidase that catalyzes the depurination of adenine A4324
of 28S rRNA and thereby blocks protein synthesis (Franz and
Jaax, 1997; Gluck et al., 1992; Lord et al., 1994; Robertus, 1988;
⁎ Corresponding author. U.S.-FDA, CFSAN, HFS-716, 5100 Paint Branch Pkwy.
College Park, MD 20740, USA. Tel.: +1 301 436 2224; fax: +1 301 436 2332.
E-mail address: [email protected] (E.A.E. Garber).
0022-1759/$ – see front matter. Published by Elsevier B.V.
doi:10.1016/j.jim.2008.05.007
Robertus, 1991). The second subunit, B-chain is a 262 amino
acid, 32 kDa lectin that targets the toxin to cells and
facilitates uptake (Lord et al., 1992; Robertus, 1991). The Xray structure of ricin has been elucidated (Katzin et al., 1991;
Montfort et al., 1987; Rutenber et al., 1991) and the enzymatic
properties of the toxin characterized (Chen et al., 1998; Day
et al., 1996; Endo et al., 1987). Extensive research has also
been conducted on the use of the toxin moiety in immunotargeted therapy (Kreitman, 1999 and 2003; Kreitman and
Pastan, 2006; Franz and Jaax, 1997; Schindler et al., 2001;
Thorpe, 2004) and on the development of vaccines to
counteract ricin toxicosis (Carra et al., 2007; Peek et al.,
2007; Vitetta et al., 2006).
Various technologies have been developed for the rapid
and sensitive detection of ricin (Huelseweh et al., 2006; Poli
et al., 1994; Shyu et al., 2002). One technology, lateral flow
devices (LFDs), requires minimal sample preparation and
instrumentation, making it ideal for field deployment as an
initial screen (Garber et al., 2005). Wheat germ and other
sources of plant lectins are routinely used in cosmetics and
may affect the reliability of assays employing glycosylated
proteins such as antibodies. The compatibility of three
common lectins with LFDs was examined along with the
252
J. Dayan-Kenigsberg et al. / Journal of Immunological Methods 336 (2008) 251–254
inclusion of 2.5% non-fat milk in the analytical buffer to
eliminate false positives. This study demonstrates the effectiveness of LFDs for the screening of cosmetics.
2. Materials and methods
Ricin used by Agence Française de Sécurité Sanitaire des
Produits de Santé (AFSSAPS) was the generous gift of B.
Beaumelle, Ph.D. of Montpellier University. The ricin displayed
an intravenous LD50 of 5.88 µg/kg body weight with six-week
old female BALB/c mice (personal communication, A. El-Zaouk
and D. Sauvaire, AFSSAPS). Lactose and 10 mM PBS were
purchased from Sigma Chemical Co. Purified ricin, ricin Achain, ricin B-chain, and agglutinin (RCA-120) used by the U.S.
Food and Drug Administration (FDA) were purchased from
Vector Laboratories (Burlingame, CA). Lectins derived from
pea (Pisum sativum), peanut (Arachis hypogea), and wheat
(Triticum vulgaris) were purchased from Sigma Chemical Co.
(catalog numbers L0881, L5380, and L9640, respectively).
LFDs for the detection of ricin, SEB, and botulinum toxin
were obtained from Tetracore, Inc. (Rockville, MD, USA) and a
U.S. Government supplier and used according to the manufacturer's instructions. The BioThreat Alert® LFDs manufactured by Tetracore, Inc. can be read either by eye or by using
the Guardian Reader™ (Alexeter Technologies, IL, USA). In
contrast, the results obtained with the LFDs from the U.S.
Government supplier could only be read by eye.
Cosmetics were purchased from local commercial suppliers.
Samples were spiked with the ricin provided by B. Beaumelle,
Ph.D. of Montpellier University. Eye make up samples were
diluted 1:1 with 10 mM PBS / 0.1% Tween-20 / 5% non-fat milk
(PBSTM) and analyzed using the LFDs according to manufac-
turer's instructions. The more viscous shampoo and body lotion
samples were first diluted 1:9 and 1:19, respectively, with PBS
and then diluted 1:1 with PBSTM to generate the 50% PBSTM
solution for analysis with the LFDs.
LOD values for ricin and lectin in buffer were determined
based on the concentration which generated a response that
exceeded the background response by four standard deviations. Average background responses were typically 0.002.
The concentration of ricin in a cosmetic necessary to generate
an average response greater than the manufacturer's recommended cut-off of 0.01 on the Guardian Reader™ defined the
concentration of toxin necessary to generate a positive.
3. Results
3.1. Cross-reactivity with wheat lectin
The cross-reactivities of Ricin BioThreat Alert® LFDs towards
lectins derived from pea, peanut, and wheat, were examined.
Neither the pea nor peanut lectin was detected by the LFDs at
concentrations as high as 1 mg/mL and 250 μg/mL, respectively.
However, the wheat lectin at ≥15 μg/mL purified protein, or as
N1000 ppm stone ground wheat, generated false positive
responses (see Figs.1 and 2). False positives were also generated
by wheat lectin with BioThreat Alert® LFDs for the detection of
SEB and botulinum toxin (data not shown) and with LFDs
manufactured by a U.S. Government source for the detection of
ricin and SEB. The addition of lactose at concentrations up to
100 mM in the buffer supplied with the LFDs failed to eliminate
the false positives observed with wheat lectin. However, no
false positives were observed upon preparing the samples in
50% PBSTM. Substitution of the LFD buffer with 50% PBSTM
Fig. 1. False positives due to wheat lectin with ricin LFDs and SEB LFDs. A) 0.5 μg/mL ricin in 50% PBSTM, B) 250 μg/mL lectin in the buffer supplied with the LFDs,
C) 250 μg/mL lectin in PBS, and D) 250 μg/mL lectin in 50% PBSTM. The presence of a second (T) red line at the position indicated by the arrows indicated a (false)
positive response. The presence of 50% PBSTM eliminated the occurrence of the second (T) red line due to the presence of wheat lectin. The ricin was obtained from
Vector Laboratories (Burlingame, CA). LFDs were obtained from Tetracore, Inc. and a U.S. Government source. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
J. Dayan-Kenigsberg et al. / Journal of Immunological Methods 336 (2008) 251–254
253
4. Discussion
Fig. 2. Detection of ricin and wheat lectin using LFD buffer and 50% PBSTM.
Ricin samples prepared in buffer supplied with LFDs;
ricin samples
prepared in 50% PBSTM;
wheat lectin in buffer supplied with LFDs; and
wheat lectin in 50% PBSTM. A decrease in response was observed with
10 μg/mL ricin (data not shown) characteristic of the “hook effect” which
entails excess, free ricin competing with captured ricin (complex) for the
detector antibody. The data represent the average of triplicate analyses with
error bars representing standard deviations. Background responses averaged
0.0021 + 0.0007 (n = 8); a cut-off of 0.01 was recommended by the
manufacturer for the classification of presumptive positives. The ricin was
obtained from Vector Laboratories (Burlingame, CA) and the LFDs were
produced by Tetracore, Inc. with the responses read using the Guardian
Reader™.
increased the LOD from 2.5 ng/mL to 10 ng/mL; the ricin from
Vector laboratories displayed a slightly greater increase in the
LOD with 50% PBSTM and slightly lower sensitivity than the
ricin used to spike the cosmetics.
3.2. Cosmetic samples
Table 1 lists the cosmetics tested and the responses
generated by samples containing 5 ng/mL ricin in the
analytical sample following dilution with PBS and PBSTM.
Samples derived from shampoo and body lotion required
dilution 1:9 and 1:19, respectively, with PBS prior to mixing
1:1 with PBSTM. As such, the concentration of ricin in the eye
make up, shampoo, and body lotion samples prior to dilution
were 10 ng/mL, 100 ng/mL, and 200 ng/mL respectively. The
additional dilution with PBS prevented the wetting and flow
problems typically observed when analyzing viscous samples
with LFDs. Further dilution of the shampoo samples, 1:19 with
PBS, improved the performance of the LFDs but was not
required for analysis.
Interestingly, three different commercial eye makeup
removers generated false positive responses with the LFDs
in the absence of ricin with an average response of 0.09 +
0.004. Inclusion of 50% PBSTM in the analytical sample
eliminated the problem and enabled the detection of 5 ng/mL
ricin. All of the cosmetics listed in Table 1 displayed
background responses of ≤0.003 in the presence of 50%
PBSTM except eye make up remover C (EMR C) and shampoo
D. EMR C and shampoo D displayed backgrounds of 0.01 and
0.004, respectively. It is therefore recommended that a
threshold greater than 0.01 be used with the Guardian
Reader™. The average responses for EMR C and shampoo D,
each containing 5 ng/mL ricin in the analytical sample, were
0.030 + 0.005 and 0.05 + 0.01, respectively; both considerably
above the background responses.
The Ricin BioThreat Alert® LFDs in combination with the
Guardian Reader™ enabled the detection of ricin in
cosmetics at levels below those associated with toxic
responses. The primary problem associated with using
LFDs for the detection of ricin in cosmetics was flow
problems with viscous products. However, the high sensitivity of the LFDs made it possible to dilute the samples
sufficiently to overcome the flow problems, while still
being able to detect ricin at levels less than those associated
with toxic events.
The availability of a buffer that prevents false positives
due to the presence of lectins has implications affecting all
LFDs and immunoassays. The mechanism by which 50%
PBSTM prevented false positives or ‘nonspecific’ binding
between the capture and detector antibodies that comprise
LFDs is not understood. The explanation that the lectin
coupled the two antibodies via their glycosyl groups does
not explain why 100 mM lactose failed to suppress false
positives. The modest increases in the LOD values for ricin
with LFDs were acceptable considering the sensitivity of the
assays and elimination of a potential source of false
positives.
The mechanism by which eye makeup remover generated
false positives was also not understood. Fortunately, 50%
PBSTM eliminated these false positives. It is recommended
that samples of unadulterated eye makeup remover be
analyzed alongside suspicious samples and that a threshold
of greater than 0.01 (possibly 0.02) be employed when using
the Guardian Reader™. The ingredients listed on the eye
makeup remover did not indicate the presence of components that might contain lectins. Thus, it seems unlikely that
the mechanism causing the false positives with eye makeup
remover and with wheat lectin would be the same, though
the mechanism by which 50% PBSTM prevented these
nonspecific interactions may be related.
Table 1
Detection of ricin in cosmetics using lateral flow devices
Eye Makeup Remover A
Eye Makeup Remover B
Eye Makeup Remover C
Shampoo A
Shampoo B
Shampoo C
Shampoo D
Body Lotion A
Body Lotion B
Body Lotion C
Body Lotion D
Body Lotion E
Background
5 ng/mL ricin
0.0014 ± 0.0003 (n = 3)
0.0022 ± 0.0011(n = 3)
0.011 ± 0.004 (n = 9)
0.0020 ± 0.0004 (n = 3)
0.0022 ± 0.0005 (n = 3)
0.0011 ± 0.0003 (n = 3)
0.004 ± 0.001 (n = 3)
0.0027 ± 0.0004 (n = 3)
0.003 ± 0.001 (n = 4)
0.003 ± 0.001 (n = 3)
0.003 ± 0.001 (n = 3)
0.0019 ± 0.0008 (n = 3)
0.009 ± 0.003 (n = 4)
0.017 ± 0.005 (n = 3)
0.030 ± 0.005 (n = 4)
0.019 ± 0.006 (n = 3)
0.026 ± 0.005 (n = 3)
0.031 ± 0.006 (n = 5)
0.05 ± 0.01 (n = 3)
0.013 ± 0.002 (n = 4)
0.015 ± 0.003 (n = 4)
0.036 ± 0.004 (n = 4)
0.035 ± 0.005 (n = 4)
0.024 ± 0.001 (n = 3)
Cosmetic samples spiked with 0 μg/mL (Background) or containing 5 ng/mL
of ricin (courtesy B. Beaumelle, Ph.D., Montpellier University) in the
analytical sample after dilution with PBS and PBSTM. All cosmetic samples
were spiked with the ricin prior to dilution with PBS or PBSTM. Eye make up
remover samples were diluted 1:1 with PBSTM while the shampoo and body
lotion samples were diluted 1:9 and 1:19, respectively with PBS prior to
mixing 1:1 with PBSTM. The additional dilution with PBS was necessary to
reduce the viscosity. The responses were measured using BioThreat Alert®
LFDs (Tetracore, Inc.) with the Guardian Reader™.
254
J. Dayan-Kenigsberg et al. / Journal of Immunological Methods 336 (2008) 251–254
5. Conclusion
Cosmetics suspected of containing ricin can be analyzed
using the BioThreat Alert® LFDs provided the samples are
sufficiently diluted to prevent flow problems. It is recommended that 50% PBSTM be used with immunoassays
whenever samples are suspected of containing wheat lectin.
In the event that a cosmetic product generates a positive
response with an LFD, the sample should be reanalyzed using
50% PBSTM as the analytical buffer and whenever possible
also confirmed using a second, independent method.
Acknowledgments
This work was supported by the Agence Française de
Sécurité Sanitaire des Produits de Santé (AFSSAPS, Saint Denis,
France) and the Food and Drug Administration (FDA, USA).
Thanks are expressed to Bruno Beaumelle, Ph.D. of the Biology
Department of Montpellier II University, Annie El-Zaouk and
Didier Sauvaire of AFSSAPS Montpellier for preparing, testing
and providing a ricin extract, and to Anne Decool of AFSSAPS
Saint-Denis for her involvement in documenting toxic waste
disposal issues. Appreciation is expressed to Peter Emanuel,
Ph.D. (Joint Program Executive Office, Department of Defense,
USA), Robert L. Bull, Ph.D. (Biological Defense Research
Directorate, Department of Defense, USA) and Lynn L. B.
Rust, Ph.D. (National Institutes of Health, USA) for their
support throughout the project. Appreciation is also expressed
to Thomas W. O'Brien, Ph.D. (Tetracore, Inc., USA), and Jennifer
Walker, M.S. (Tetracore, Inc. USA).
References
Carra, J.H., Wannemacher, R.W., Tammariello, R.F., Lindsey, C.Y., Dinterman, R.E.,
Schokman, R.D., Smith, L.A., 2007. Improved formulation of a recombinant
ricin A-chain vaccine increases it stability and effective antigenicity. Vaccine
25, 4149–4158.
Chen, X.-Y., Link, T.M., Schramm, V.L., 1998. Ricin A-chain: kinetics, mechanism,
and RNA stem-loop inhibitors. Biochemistry 37, 11605–11613.
Czapla, T.H., Johnston, I.A., 1990. Effect of plant lectins on the larval
development of European corn borer (Lepidoptera: pyralidae) and
southern corn rootworm (Coleoptera: chrysomelidae). J. Econ. Entomol.
83, 2480–2485.
Day, P.J., Ernst, S.R., Frankel, A.E., Monzingo, A.F., Pascal, J.M., Molina-Svinth,
M.C., Robertus, J.D., 1996. Structure and activity of an active site
substitution of ricin A-chain. Biochemistry 35, 11098–11103.
Endo, Y., Mitsui, K., Motizuki, M., Tsurugi, K., 1987. The mechanism of action of
ricin and related toxic lectins on eukaryotic ribosomes: the site and the
characteristics of the modification in 28 S ribosomal RNA caused by the
toxins. J. Biol. Chem. 262, 5908–5912.
Franz, D.R., Jaax, K.K., 1997. Ricin toxin. In: Sidell, F.R., Takafuji, E.T., Franz, D.R.
(Eds.), Medical Aspects of Chemical and Biological Warfare. Walter Reed
Army Medical Center, Washington USA, pp. 631–642.
Garber, E.A.E., Eppley, R.M., Stack, M.E., McLaughlin, M.A., Park, D.S., 2005.
Feasibility of immunodiagnostic devices for the detection of ricin,
amanitin and T-2 toxin in food. J. Food. Prot. 68, 1294–1301.
Gluck, A., Endo, Y., Wool, I.G., 1992. Ribosomal RNA identity elements for ricin
A-chain recognition and catalysis. Analysis with tetraloop mutants. J.
Mol. Biol. 226, 411–424.
Huelseweh, B., Ehricht, R., Marschall, H.J., 2006. A simple and rapid protein
array based method for simultaneous detection of biowarfare agents.
Proteomics 6, 2972–2981.
Katzin, B.J., Collins, E.J., Robertus, J.D., 1991. Structure of ricin A-chain at 2.5 .
Proteins Struct Funct. Genet. 10, 251–259.
Knight, B., 1979. Ricin: a potent homicidal poison. Br. Med. J. 1 (6159), 350–351.
Kreitman, R.J., 1999. Immunotoxins in cancer therapy. Curr. Opin. Immunol.
11, 570–578.
Kreitman, R.J., 2003. Taming ricin toxin. Nature Biotechnol. 21, 372–374.
Kreitman, R.J., Pastan, I., 2006. Immunotoxins in the treatment of hematologic malignancies. Curr Drug Targets 7, 1301–1311.
Lord, J.M., Wales, R., Pitcher, C., Roberts, L.M., 1992. Cell surface and intracellular
functions for galactose binding in ricin cytotoxicity. Biochem. Soc. Trans. 20,
734–738.
Lord, J.M., Roberts, L.M., Robertus, J.D., 1994. Ricin: structure, mode of action
and some current applications. FASEB J. 8, 201–208.
Montfort, W., Villafranca, J.E., Monzingo, A.F., Ernst, S.R., Katzin, B., Rutenber,
E., Xuong, N.H., Hamlin, R., Robertus, J.D., 1987. The three-dimensional
structure of ricin at 2.8 . J. Biol. Chem. 262, 5398–5403.
Okoye, J.O.A., Enunwaonye, C.A., Okorie, A.U., Anugwa, F.O.I., 1987. Pathological effects of feeding roasted castor bean meal Ricinus communis to
chicks. Avian Pathol. 16, 283–290.
Peek, L.J., Brey, R.N., Middaugh, C.R., 2007. A rapid, three-step process for the
preformulation of a recombinant ricin toxin A-chain vaccine. J. Pharm.
Sci. 96, 44–60.
Poli, M.A., Rivera, V.R., Hewetson, J.F., Merrill, G.A., 1994. Detection of ricin by
colorimetric and chemiluminescence Elisa. Toxicon 32, 1371–13137.
Purushotham, N.P., Rao, M.S., Raghavan, G.V., 1986. Utilization of castor-meal
in the concentrate mixture of sheep. Indian J. Anim. Sci. 56, 1090–1093.
Robertus, J.D., 1988. Toxin structure. Cancer Treat. Res. 37, 11–24.
Robertus, J.D., 1991. The structure and action of ricin, a cytotoxic Nglycosidase. Sem. in Cell Biol. 2, 23–30.
Rutenber, E, Katzin, B.J., Ernst, S., Collins, E.J., Mlsna, D., Ready, M.P., Robertus,
J.D., 1991. Crystallographic refinement of ricin to 2.5 Å. Proteins 10,
240–250.
Schindler, J., Sausville, E., Messmann, R., Uhr, J.W., Vitetta, E.S., 2001. The
toxicity of deglycosylated ricin A-chain containing immunotoxins in
patients with non-Hodgkin's lymphoma is exacerbated by prior radiotherapy: a retrospective analysis of patients in five clinical trials. Clin.
Cancer Res. 7, 255–258.
Shyu, H.F., Chiao, D.J., Liu, H.W., Tang, S.S., 2002. Monoclonal antibody-based
enzyme immunoassay for detection of ricin. Hybrid. Hybridomics 21,
69–73.
Thorpe, P.E., 2004. Vascular targeting agents as cancer therapeutics. Clin.
Cancer Res. 10, 415–427.
Vitetta, E.S., Smallshaw, J.E., Coleman, E., Jafri, H., Foster, C., Munford, R.,
Schindler, J., 2006. A pilot clinical trial of a recombinant ricin vaccine in
normal humans. Proc. Natl. Acad. Sci. U. S. A. 103, 2268–2273.