A rapid and inexpensive method to screen for common foods that reduce the action of acrylamide

A rapid and inexpensive method to screen for common foods that reduce the action of acrylamide, A rapid and inexpensive method to screen for common foods that reduce the action of acrylamide

Toxicology Letters 175 (2007) 82–88

A rapid and inexpensive method to screen for common foods

that reduce the action of acrylamide,

a harmful substance in food Koichi Hasegawaa,b, Satsuki Miwaa, Tomoko Tajimac, Kaname Tsutsumiuchia,c,

Hajime Taniguchid, Johji Miwaa,c,∗

aInstitute for Biological Function, Chubu University, 1200 Matsumoto,

Kasugai 487-8501, Japan

bGraduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan

cGraduate School of Bioscience and Biotechnology, Chubu University,

1200 Matsumoto, Kasugai 487-8501, Japan

dDepartment of Environmental Bioresource, Ishikawa Prefectural University,

308-1 Suematsu, Nonoichi, Ishikawa 921-8836, Japan

Received 20 August 2007; received in revised form 25 September 2007; accepted 25 September 2007

Available online 7 October 2007

Abstract

By DNA microarray and protein 2-DE screens for Caenorhabditis elegans genes up-regulated by acrylamide, we selected the gst-4 gene and constructed a gst::gfp fusion gene, which was used to transform C elegans into a biosensor for acrylamide This

biosensor detects acrylamide as a GFP-expression signal in a dose- and time-dependent manner When the biosensor was exposed

to acrylamide together with commercially available powdered green tea, GFP levels decreased to the control level, suggestive of acrylamide detoxification or prevention of GST induction The present methodology should be applicable for screening of not only harmful substances but also substances that reduce or counteract their harmfulness or action, with appropriately constructed visible biosensors

© 2007 Elsevier Ireland Ltd All rights reserved

Keywords: Biosensor; Acrylamide; GST; Food; Caenorhabditis elegans

1 Introduction

In April 2002, the Swedish National Food Agency

and Stockholm University reported that acrylamide was

formed in high concentrations of 30–2300␮g/kg

dur-ing the frydur-ing or bakdur-ing of carbohydrate-rich foods

∗Corresponding author at: Institute for Biological Function, Chubu

University, 1200 Matsumoto, Kasugai 487-8501, Japan.

Tel.: +81 568 51 6218; fax: +81 568 51 6218.

E-mail address:miwa@isc.chubu.ac.jp (J Miwa).

(Tareke et al., 2002); 64–5000␮g/kg were reported

in potato-based snacks (the Japanese National Food Research Institute (http://aa.iacfc.affrc.go.jp/en/), 2002; Tsutsumiuchi et al., 2004) The compound is produced

by the Maillard reaction during heat treatment of foods (Mottram et al., 2002; Stadler et al., 2002) The non-toxic acrylamide polymer is commonly used in chemical paper and fibers, soil stabilizers, plastic materials, gel electrophoresis, and so on In contrast, the acrylamide monomer, a known industrial hazard, has long been stud-ied and shown to exhibit neurotoxicity in vertebrates, 0378-4274/$ – see front matter © 2007 Elsevier Ireland Ltd All rights reserved.

doi: 10.1016/j.toxlet.2007.09.013

K Hasegawa et al / Toxicology Letters 175 (2007) 82–88 83 mutagenicity in somatic and germ cells, and

carcino-genicity in experimental animals Thus, the finding of

monomeric acrylamide in our daily diet focused the

world’s attention on this shocking public health problem

(Friedman, 2003)

In this report, we selected the gst-4 gene, which

encodes a human homologue of glutathione-requiring

prostaglandin d synthase and was most strongly

up-regulated after exposure to acrylamide (to be published

elsewhere), and constructed a gst-4::gfp translational

fusion gene to transform Caenorhabditis elegans into

a biosensor for acrylamide (MJCU017) As a whole

animal system, it can directly demonstrate the active

and perhaps harmful dose of a substance, such as

acry-lamide Here we report the use of this biosensor and

CL2166 (transcriptional reporter transgenic C elegans)

to establish a rapid and inexpensive method to screen for

common foods that reduce the action of acrylamide, a

harmful substance in food

2 Materials and methods

2.1 Nematode strains and culturing

C elegans, unc-119 (ed3) (Maduro and Pilgrim, 1996) and

CL2166 (dvIs19[pAF15(gst-4(727 bp)::gfp::NLS)]) (Link and

Johnson, 2002) were obtained from the Caenorhabditis

ele-gans Genetics Center (University of Minnesota, Minnesota,

Minneapolis, MN, USA) MJCU017 (Is [gst-4(1491 bp)::gfp

unc-119+]) was also used (to be published elsewhere) The

nematodes were cultured and handled essentially as described

byBrenner (1974)

2.2 GFP signal kinetics

Synchronized L1-stage animals were transferred onto

NGM plates seeded with E coli OP50 and grown at 20◦C for

48 h to reach late L4 stage Fifty L4-synchronized (CL2166)

or 100 (MJCU017) animals were placed into each well of

a B & W IsoPlate-96 (PerkinElmer, Massachusetts, MA,

USA) containing E coli OP50 (109cells/mL) and

acry-lamide (1 g/L, 500 mg/L, 400 mg/L, 300 mg/L, 200 mg/L,

100 mg/L, 50 mg/L) or control without acrylamide in S medium

(Stiernagle, 1999), at a total volume of 100␮L per well Plates

were sealed with optical adhesive covers (ABI Prism, Foster

City, CA, USA) and incubated at 20◦C or 25◦C The GFP

flu-orescence was measured once every hour with a Wallac 1420

ARVOsx multilabel counter (PerkinElmer) The GFP signal

value for each well was the mean value of three measurements;

whereas the value for each acrylamide concentration and the

control was the mean value from 8 to 12 wells

2.3 Co-feeding experiment

About 5000 transgenic animals (CL2166), synchronized at

late L4 stage as described above, were transferred into 10 mL

of S medium containing 1× 109cells/mL of E coli OP50,

500 mg/L of acrylamide with or without several common foods and grown at 25◦C The common foods used were powdered green tea (Ohi Ocha Koiaji, Itoen, Tokyo, Japan), instant cof-fee (Nescafe Gold Blend, Nestle Japan, Kobe, Japan), tomato juice (Delmonte Tomato Juice, Kikkoman, Chiba, Japan), and powdered sports drink (Pocari Sweat, Otsuka Seiyaku, Tokyo, Japan) After 2 h, 4 h, and 6 h of incubation, the transgenic animals in 3 mL of S medium were collected, washed with M9 buffer, and re-suspended in M9 buffer at 50 animals per 100␮L, which was put into each well (100␮L per well × 8–12) of a B &

W IsoPlate-96 Plates were monitored for the GFP-expression signal as was done for GFP signal kinetics

2.4 Statistical analysis Paired Student’s t-test (Microsoft Excel) was used to

deter-mine the significance of differences in the mean GFP signal values

3 Results

3.1 gst-4 expression pattern in vivo

We constructed a gst-4::gfp translational fusion gene

and made the transgenic line MJCU017 having a

chro-mosomally integrated gst-4::gfp fusion gene This line,

hereinafter called an acrylamide biosensor, emitted no detectable GFP signal in the absence of acrylamide, but emitted a very strong GFP signal from the whole body with 500 mg/L of acrylamide (Fig 1) In contrast, CL2166 line, which contains a chromosomally

inte-grated transcriptional reporter (gst-4 promoter drives gfp

transcription) (Link and Johnson, 2002), constitutively emitted GFP signals in the body-wall muscle without acrylamide, although its GFP signal pattern was the same as that for MJCU017 upon exposure to acrylamide (Fig 1)

3.2 gst-4 expression increases by acrylamide in a dose- and time-dependent manner

Fig 2 shows the kinetics of GST expression in response to acrylamide exposure at concentrations from

50 mg/L to 1 g/L Columns indicate the mean val-ues± S.E.M of GFP signals at various acrylamide concentrations compared with that of the control (100%) from four independent experiments (or plates) GFP sig-nals increased in a dose- and time-dependent manner

in both transgenic nematodes, and the signals increased faster and their peaks were higher at 25◦C than at 20◦C. Because CL2166 recorded much higher sensitivity and signal peaks than did MJCU017, we chose the more

sen-Fig 1 GST-expression patterns in MJCU017 (A and B) and CL2166 (C and D) (A) MJCU017 without acrylamide (B) MJCU017 with 500 mg/L

of acrylamide (C) CL2166 without acrylamide (D) CL2166 with 500 mg/L of acrylamide Scale bars, 200 ␮m.

sitive and faster combination of CL2166 at 25◦C for

screening of common foods that reduce the action of

acrylamide

3.3 Powdered tea prevents gst-4 induction

To screen for common foods that reduce the action

of acrylamide, the acrylamide biosensors were grown

on S media containing 500 mg/L of acrylamide, with

or without common foods.Fig 3shows the results of

such “co-feeding” experiments with commercially

avail-able powdered green tea, instant coffee, tomato juice,

and powdered sports drink Columns indicate the mean

values± S.E.M of GFP signals from six (green tea) or

three (coffee, tomato juice, and sports drink)

indepen-dent experiments, as compared with that for the control

(100%) InFig 3A–C, Tea 1 was 8 g/L (powdered green

tea), Coffee 1 was 14 g/L (instant coffee), and PS 1 was

74 g/L (powdered sports drink), all at the makers’

recom-mended concentrations for drinking; and Tea 2, Coffee

2, and PS 2 or Tea 3, Coffee 3, and PS 3 were the 1/10

or 1/100 diluted concentrations, respectively InFig 3D, 1% tomato contained 100␮L of tomato juice in 10 mL

of S medium, and 0.1% tomato and 0.01% tomato were 1/10 and 1/100 diluted, respectively When the acry-lamide biosensors were cultured in S medium containing several common foods only, the GFP-expression signals differed little from that of the control without acry-lamide (Fig 3A–D); and with acrylamide only, the GFP signals increased time-dependently (Fig 3A–D) How-ever, when the acrylamide biosensors were cultured in

S medium containing acrylamide together with 8 g/L of powdered green tea (Tea 1 + AA) and 74 g/L of powdered sports drink (PS 1 + AA), the GFP-expression signals decreased to the control level (Fig 3A and C) After the signal measurements, the acrylamide biosensors were collected from each of their corresponding wells and checked for their viability and their GFP expression

K Hasegawa et al / Toxicology Letters 175 (2007) 82–88 85

Fig 2 Kinetics of GST expression by MJCU017 (A and C) and CL2166 (B and D) acrylamide biosensors exposed to acrylamide at concentrations from 50 mg/L to 1 g/L Columns indicate the mean values ± S.E.M of five independent experiments (plates) GFP signals were measured at 20 ◦C (A and B) or 25 ◦C (C and D) All nematodes were alive after measurement The significant differences in GFP signal relative to the control are

indicated for the first appearance at each acrylamide concentration as determined by the paired Student’s t-test; *P < 0.05.

under a fluorescence dissection microscope The

biosen-sors in the 74 g/L of powdered sports drink (PS 1 and PS

1 + AA) were all shriveled and dead (Fig 3C), and the

death of the animals should explain the lack of GFP

sig-nal All acrylamide biosensors, however, were alive in

the 8 g/L of powdered tea (Tea 1 and Tea 1 + AA) The

co-feeding result thus suggests that the powdered tea

used must have contained some substance(s) that

pre-vented acrylamide from inducing GST and renders this

tea a candidate for an acrylamide detoxifier or modifier

4 Discussion

Glutathione S-transferases (GSTs) comprise a large

family of enzymes whose members generally exist in

every organism from bacteria to humans (Vuilleumier

and Pagni, 2002) GSTs are considered one of the major

players in the phase II detoxification of both endogenous

products of oxidative stress and electrophilic

xenobi-otics GST binds with glutathione (GSH), a tripeptide thiol, and catalyzes its conjugation with target sub-strates to enable their excretion from cells About

50 gst genes have been identified in the C elegans

genome (WormBase, http://www.wormbase.org/), and

gst-4 expression was reported to be dramatically induced

by oxidative stressors such as paraquat, juglon, plumba-gin, hyperbaric oxygen treatment, or the endocrine active substances diethylstilbestrol and progesterone (Link and Johnson, 2002; Tawe et al., 1998; Cutodia et al., 2001; Leiers et al., 2003; Reichert and Menzel, 2005) Our DNA microarray and protein 2-DE data obtained for

acrylamide also showed that gst-4 was the most highly

up-regulated gene (to be published elsewhere) For

these reasons, we selected the gst-4 gene as the most

suitable molecule for the acrylamide biosensor The dif-ferences in GFP signal sensitivity between MJCU017 and CL2166 might be caused by their transgenic struc-tures: CL2166 had a transcriptional reporter without the

Fig 3 Co-feeding experiments Columns indicate the mean values ± S.E.M of six (A) or three (B–D) independent experiments (plates) (A) Co-feeding with powdered green tea: AA, 500 mg/L of acrylamide; Tea 1, 8 g/L; Tea 2, 0.8 g/L; Tea 3, 0.08 g/L When biosensors were cultured in acrylamide with 8 g/L of powdered green tea (Tea 1 + AA), GFP-expression signals remained at the control level (B) Co-feeding with instant coffee:

AA, 500 mg/L of acrylamide; Coffee 1, 14 g/L; Coffee 2, 1.4 g/L; Coffee 3, 0.14 g/L (C) Co-feeding with powdered sports drink: AA, 500 mg/L

of acrylamide; PS 1, 74 g/L; PS 2, 7.4 g/L; PS 3, 0.74 g/L GFP-expression signals did not increase in 74 g/L of powdered drink (PS 1 + AA), but all biosensors died apparently from osmotic stress (D) Co-feeding with tomato juice: AA, 500 mg/L of acrylamide; Tomato 1%, 100 ␮L/10 mL; Tomato 0.1%, 10 ␮L/10 mL The significant difference in GFP signal of co-fed biosensors relative to the AA plate at each time point was determined

by the paired Student’s t-test; *P < 0.05.

gst-4 coding region, whereas MJCU017 comprised a

translational reporter with it

The acrylamide biosensor introduced here provides

an easily usable and available, rapid and inexpensive

method to detect acrylamide, a harmful substance in

food, not to mention that this C elegans-based biosensor

reproduces rapidly, whereby one biosensor reproduces

to become 8 million in a week Although man-made

machines offer superior sensitivity to detect lower

con-centrations of chemicals (Tsutsumiuchi et al., 2004),

they can only tell us the amounts of known

chemi-cal compounds In contrast to machines, our biosensor

not only tells us what these known chemicals do to

living forms, but it can also do the same for totally

unknown chemicals Since our method utilizes natu-ral biological responses of whole live animals against

a xenobiotic toxin, it also tells us the threshold con-centration of a toxin that is capable of triggering the animal’s natural response against it, as reported here

There are also other reports that transgenic C elegans

were developed as biosensors against xenobiotics For

example, a heat shock protein promoter-driven lacZ was

used as a reporter to detect a fungicide (captan) (Jones

et al., 1996), heavy metal pollution in water (Mutwakil

et al., 1997), dithiocarbamate fungicides (Guven et al.,

1999), and pharmaceutical compounds (EGFR kinase inhibitors) (Dengg and van Meel, 2004), and a

constitu-tive promoter let-858-driven lacZ was used for detecting

K Hasegawa et al / Toxicology Letters 175 (2007) 82–88 87 heavy metal and xenobiotics (3, 5-DCP) (Lagido et

al., 2001) By using the gfp reporter, gene expression

patterns can be rapidly and directly identified in

liv-ing animals without the need for fixliv-ing or stainliv-ing

A small heat shock promoter-driven gfp was used to

detect heat and oxidative stress, beta amyloid peptide

(Link et al., 1999), and microwave stress by a

non-thermal mechanism (de Pomerai et al., 2000), and a gst-4

promoter-driven gfp was used for detection of

oxida-tive stresses (Link and Johnson, 2002; Leiers et al.,

2003)

In addition to the acrylamide-detection method

pre-sented here, we have also shown that the same method

can be used to find common foods that may reduce or

inactivate the action of acrylamide, which was used as

a model compound of oxidative-stress-producing

xeno-biotics By this method, we found one candidate from

four commercially available common foods examined

(Fig 3) When biosensors were cultured in S medium

containing acrylamide at 500 mg/L together with 8 g/L

of powdered green tea, the GFP-expression signals

remained at the levels of the control while almost all

biosensors in 8 g/L of powdered green tea were healthy

This result apparently suggests that the green tea used

prevented GST induction despite the presence of the very

high concentration of acrylamide, which usually makes

biosensors (nematodes) very sick Although the animals

showed significantly increased GFP signals in 100 mg/L

of acrylamide after 6–9 h of exposure (Fig 2), for these

experiments, we chose the acrylamide concentration of

500 mg/L, which is 100 times the level of acrylamide

in some potato-based snacks and elicits its significant

response in only 3–5 h, as a suitably severe condition

under which to detect acrylamide-counteracting foods

After such screening, we can take the next steps of

identifying acting components in the tea and finding

the exact mechanism of how they work as well The

major components in the powdered green tea we used

are catechins, which comprise (−)-epicatechin (EC),

(−)-epigallocatechin (EGC), (−)-epicatechin-3-gallate

(ECG), and (−)-epigallocatechin-3-gallate (EGCG)

They have very high-antioxidant activity and reportedly

have a protective effect against a variety of cancers, such

as lung, prostate, and breast cancers as well as

com-mon diseases (Yang et al., 2002) Other components such

as vitamin C, caffeine, or the additives of dextrin and

cyclodextrin might also have taken part in inhibiting GST

induction

We have presented the acrylamide-detecting

biosen-sor that not only detects acrylamide but also and

importantly finds candidates for common foods that

may reduce or inactivate the action of acrylamide, as

a model compound of oxidative-stress-producing xeno-biotics One of the essential features of the present

methodology is that it requires no a priori knowledge

about either the substances to test for their effects on the animal or the genes to be used for constructing biosen-sors; we simply choose those genes that are up-regulated

by any given substances tested, whether they are known

or unknown Appropriately constructed visible biosen-sors can thus be tailor-made for specific substances we want to detect, or whose harmful effects we want to reduce, as demonstrated in this report The methodology should be applicable to detect a wide range of harm-ful substances in our diet or environment and to screen foods or other edible substances that could reduce their harmful effects and counteract their actions

Acknowledgement

This work was supported by a grant from the High-Tech Research Center Establishment Project from the Japanese Ministry of Education, Culture, Sports, Science and Technology

References

Brenner, S., 1974 The genetics of the nematode Caenorhabditis

ele-gans Genetics 77, 71–94.

Cutodia, N., Won, S.J., Novillo, A., Wieland, M., Li, C., Callard, I.P.,

2001 Caenorhabditis elegans as an environmental monitor using

DNA microarray analysis Ann N.Y Acad Sci 948, 32–42.

de Pomerai, D., Daniells, C., David, H., Allan, J., Duce, I., Mutwakil, M., Thomas, D., Sewell, P., Tattersall, J., Jones, D., Candido, P.,

2000 Non-thermal heat-shock response to microwaves Nature

405, 417–418.

Dengg, M., van Meel, J.C.A., 2004 Caenorhabditis elegans as model

system for rapid toxicity assessment of pharmaceutical com-pounds J Pharmacol Toxicol Methods 50, 209–214.

Friedman, M., 2003 Chemistry, biochemistry, and safety of acry-lamide: a review J Agric Food Chem 51, 4504–5426 Guven, K., Power, R.S., Avramide, S., Allender, R., De Pomerai, D.I.,

1999 The toxicity of dithiocarbamate fungicides to soil nematodes,

assessed using a stress-inducible transgenic strain of

Caenorhab-ditis elegans J Biochem Mol Toxicol 13, 324–333.

Jones, D., Stringham, E.G., Babich, S.L., Candido, E.P.M., 1996.

Transgenic strains of the nematode C elegans in biomonitoring

and toxicology: effects of captan and related compounds on the stress response Toxicology 109, 119–127.

Lagido, C., Pettitt, J., Porter, A.J.R., Paton, G.I., Glover, L.A., 2001.

Development and application of bioluminescent Caenorhabditis

elegans as multicellular eukaryotic biosensors FEBS Lett 493,

36–39.

Leiers, B., Kampk¨otter, A., Grevelding, C.G., Link, C.D., Johnson, T.E., Henkle-D¨uhrsen, K., 2003 A stress-responsive glutathione

S-transferase confers resistance to oxidative stress in Caenorhabditis

elegans Free Rad Biol Med 34, 1405–1415.

Link, C.D., Johnson, C.J., 2002 Reporter transgenes for study of

oxidant stress in Caenorhabditis elegans Method Enzymol 353,

497–505.

Link, C.D., Cypser, J.R., Johnson, C.J., Johnson, T.E., 1999 Direct

observation of stress response in Caenorhabditis elegans using a

reporter transgene Cell Stress Chaperones 4, 235–242.

Maduro, M., Pilgrim, D., 1996 Conservation of function and

expression of unc-119 from two Caenorhabditis species despite

divergence of non-coding DNA Gene 183, 77–85.

Mottram, D.S., Wedzicha, B.L., Dodson, A.D., 2002 Acrylamide is

formed in the Maillard reaction Nature 419, 448.

Mutwakil, M.H.A.Z., Reader, J.P., Holdich, D.M., Smithurst, P.R.,

Candido, E.P.M., Jones, D., Stringham, E.G., de Pomerai, D.I.,

1997 Use of stress-inducible transgenic nematodes as biomarkers

of heavy metal pollution in water samples from an English river

system Arch Environ Contam Toxicol 32, 146–153.

Reichert, K., Menzel, R., 2005 Expression profiling of five

differ-ent xenobiotics using a Caenorhabditis elegans whole genome

microarray Chemosphere 61, 229–237.

Stadler, R.H., Blank, I., Varga, N., Robert, F., Hau, J., Guy, P.A., Robert,

M.A., Riedliker, S., 2002 Acrylamide from Maillard reaction

prod-ucts Nature 419, 449.

Stiernagle, T., 1999 Maintenance of C elegans In: Hope, I (Ed.), C.

elegans A Practical Approach Oxford University Press, Oxford,

pp 51–67.

Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., Tornqvist, M., 2002 Analysis of acrylamide, a carcinogen formed in heated foodstuffs.

J Agric Food Chem 50, 4998–5006.

Tawe, W.N., Eschbach, M.L., Walter, R.D., Henkle-D¨uhrsen, K., 1998.

Identification of stress-responsive genes in Caenorhabditis

ele-gans using RT-PCR differential display Nucleic Acid Res 26,

1621–1627.

Tsutsumiuchi, K., Hibino, M., Kambe, M., Oishi, K., Okada, M., Miwa, J., Taniguchi, H., 2004 Application of ion-trap LC/MS/MS for determination of acrylamide in processed foods J Food Hyg Soc Jpn 45, 95–99.

Vuilleumier, S., Pagni, M., 2002 The elusive roles of bacterial glutathione S-transferases: new lessons from genomes Appl Microbiol Biotechnol 58, 138–146.

Yang, C.S., Maliakal, P., Meng, X., 2002 Inhibition of carcinogenesis

by tea Ann Rev Pharmacol Toxicol 42, 25–54.