Effects of the garlic components diallyl sulfide and diallyl disulfide on arylamine N-acetyltransferase activity in human colon tumour cells

Article abstract of DOI:10.1016/S0278-6915(98)00046-5

Diallyl sulfide (DAS) and diallyl disulfide (DADS), major components of garlic, were used to determine inhibition of arylamine N-acetyltransferase (NAT) activity in a human colon tumour (adenocarcinoma) cell line. Two assay systems were performed, one with cellular cytosols (9000 g suprnatant), the other with intact bacterial cell suspensions. The NAT activity in a human colon tumour cell line was inhibited by DAS and DADS in a dose-dependent manner in both system: that is, the greater the concentration of DAS and DADS in the reaction, the greater the inhibition of NAT activities in both systems. The data also indicated that DAS and DADS decrease the apparent values of K(m) and V(max) of NAT enzymes from human colon tumour cells in both systems examined. This is the first report to demonstrate that garlic components do affect human colon tumour cell NAT activity.

Full text of DOI:10.1016/S0278-6915(98)00046-5

Food and Chemical Toxicology 36 (1998) 761±770
E€ects of the Garlic Components Diallyl
Sul®de and Diallyl Disul®de on Arylamine
N-Acetyltransferase Activity in Human
Colon Tumour Cells
1
G. W. CHEN , J. G. CHUNG *, C. L. HSIEH and J. G. LIN
,3
2
3
3
1
2
3
Department of Surgery, Department of Medicine, Institute of Chinese Medical Science, China
Medical College, Taichung 400, Taiwan, ROC
(
Accepted 15 October 1997)
AbstractÐDiallyl sul®de (DAS) and diallyl disul®de (DADS), major components of garlic, were used to
determine inhibition of arylamine N-acetyltransferase (NAT) activity in a human colon tumour (adeno-
carcinoma) cell line. Two assay systems were performed, one with cellular cytosols (9000 g supernatant),
the other with intact bacterial cell suspensions. The NAT activity in a human colon tumour cell line
was inhibited by DAS and DADS in a dose-dependent manner in both system: that is, the greater the
concentration of DAS and DADS in the reaction, the greater the inhibition of NAT activities in both
systems. The data also indicated that DAS and DADS decrease the apparent values of K
m
and Vmax of
NAT enzymes from human colon tumour cells in both systems examined. This is the ®rst report to
demonstrate that garlic components do a€ect human colon tumour cell NAT activity. # 1998 Elsevier
Science Ltd. All rights reserved
Keywords: N-acetyltransferase; diallyl sul®de; diallyl disul®de; human colon tumour cells.
Abbreviations: 2-AAF=N-acetyl-2-amino¯uorene; 2-AF=2-amino¯uorene; N-Ac-PABA=N-acetyl-p-
aminobenzoic acid; DADS=diallyl disul®de; DTT=dithiothreitol; DAS=diallyl sul®de; EDTA=ethy-
lenediaminetetraacetic acid; FBS=foetal bovine serum; NAT=N-acetyltransferase; PABA=p-amino-
benzoic acid; PMSF=phenylmethylsulfonyl¯uoride.
INTRODUCTION
NAT has been shown to exist in di€erent forms in
various species, including rabbits (Hein et al.,
Exposure to environmental and occupational carci-
1
982a), hamsters (Hein et al., 1982b), rats (Hein
et al., 1991), mice (Chung et al., 1993), humans
Vatsis and Weber, 1993), pigeons (Andres et al.,
983), chickens (Deguchi et al., 1988), nematodes
Chung et al., 1996), insects (Whitaker and Goosey,
993), microorganisms [fungi (Fang et al., 1997)
nogens is thought to be responsible for a large pro-
portion of human cancers. The arylamines represent
one of the documented classes of chemicals known
to induce tumours in humans (Parker and Evans,
(
1
(
1
984). The arylamine carcinogens require metabolic
1
activation by host enzymes as a prerequisite to the
initiation of carcinogenesis in target organs and tis-
sues (Lower, 1982). N-Acetyltransferase (NAT) is a
drug-metabolizing enzyme that catalyses the N-
acetylation of arylamine and hydrazines. This
and bacteria (Chung et al., 1997)]. Several carcino-
genic arylamines and heterocyclic arylamines are
among the known substrates for NAT, and hence
there is much interest in the role of NAT in chemi-
enzyme also catalyses the formation of mutagenic cal carcinogenesis.
products from heterocyclic dietary carcinogens that
The acetylation of certain arylamines is poly-
are derived largely from cooked meat and ®sh morphic, with rapid, slow, and some intermediate
(
Nagao and Sugimura, 1993). NAT localizes in the acetylation phenotypes being shown in rabbits,
cytosol fraction of the liver and other organs or tis- hamsters and other mammals, and shown to be
sues and requires acetyl-coenzyme A for its activity. inherited as autosomal Mendelian recessive traits
(
Weber, 1987). This polymorphism is attributed to
allelic variants in the coding region of the NAT2
gene locus (Blum et al., 1990). Two NAT isozymes,
NAT1 and NAT2, have been reported that di€er
*
Author for correspondence at: Department of Medicine,
China Medical College, 91 Hsueh-Shih Road,
Taichung 400, Taiwan, ROC.
0278-6915/98/$19.00+0.00 # 1998 Elsevier Science Ltd. All rights reserved. Printed in Great Britain
PII S0278-6915(98)00046-5

7
62
G. W. Chen et al.
signi®cantly in their intrinsic stabilities and acceptor type-dependent expression of arylamine N-acety-
substrate selectivities (Grant et al., 1991). cDNA lation capacity. Our initial choice of 2-
and genomic DNA clones coding for NAT have amino¯uorene (2-AF) and p-aminobenzoic acid
been isolated from the liver of humans (Blum et al., (PABA) as test substrates was based on previous
1
990; Ohsako and Deguchi, 1990). In humans, two studies in mice (Chung et al., 1993) and current
distinct NAT genes (NAT1 and NAT2) have been interest in comparing the metabolism of a carcino-
identi®ed and sequenced which encode for two gen (2-AF) to a non-carcinogen (PABA).
di€erent human liver arylamine NATs (Blum et al.,
1
990; Ohsako and Deguchi, 1990). Several lines of
evidence suggest that NAT1 and NAT2 represent
the monomorphic and polymorphic proteins, re-
spectively (Blum et al., 1990; Grant et al., 1991;
Ohsako and Deguchi, 1990). Interest in poly-
MATERIALS AND METHODS
Chemicals and reagents
DAS and DADS were obtained from Fluka
morphic acetylation of carcinogenic arylamines in Chemika Co. (Buchs, Switzerland). Ethylene-
humans is largely due to the statistical linking of diaminetetraacetic acid (EDTA), PABA, N-acetyl-p-
slow acetylator phenotypes having known occu- aminobenzoic acid (N-Ac-PABA), acetylcarnitine,
pational exposure to such compounds with a higher leupeptin, bovine serum albumin, phenylmethylsul-
risk of bladder cancer (Cartwright et al., 1982), and fonyl¯uoride (PMSF), Tris, dimethyl sulfoxide
studies linking rapid acetylator phenotypes to (DMSO), dithiothreitol (DTT) and carnitine acetyl-
increased risk of colorectal cancer (Ilett et al., transferase were obtained from Sigma Chemical Co.
1
987). Extrahepatic expression of acetylator-geno- (St Louis, MO, USA). Acetyl-CoA was obtained
type dependent NAT activity also has been reported from P-L Biochemicals Inc. (Milwaukee, WI, USA).
in tissues of human colon (Kirlin et al., 1989a) and 2-AF and N-acetyl-2-amino¯uorene (2-AAF) were
bladder (Kirlin et al., 1989b). Thus, the genetically obtained from K and K Laboratories (Plainview,
mediated variation in acetyltransferase activities NY, USA). All of the chemicals used were reagent
within target organs for arylamine-induced neo- grade.
plasm may indicate di€erent risks among the
human population.
Human colon tumour cell lines
Garlic (Allium sativum), as well as being used for
Human colon adenocarcinoma cell lines were
food, has been used for medicinal purposes for a obtained from the Veterinary Hospital (Taipei,
2
long time, and has been demonstrated to posses Taiwan). The cells were placed into 75 cm tissue
antibiotic (Cavallito et al., 1954), fungicidal (Moore culture ¯asks and grown at 378C under a humidi-
and Atkins, 1977), antihelmintic (Barone and ®ed 5% CO2 atmosphere in RPMI 1640 medium
Tonsey, 1977) and anti-thrombotic (Bordia, 1978) (Sigma), supplemented with 10% foetal bovine
properties. Garlic may be e€ective in the prevention serum (FBS) (Gibco BRL, Grand Island, NY,
or treatment of several chronic diseases, such as USA), 2% penicillin±streptomycin (10,1000 U/ml
cardiovascular disease, hypertension and cancer penicillin, 10 mg/ml streptomycin).
(
Boullin, 1981; Dausch and Nixon, 1990; Ernst
et al., 1985). Recently, it has been shown that garlic
Preparation of human colon tumour cell cytosols
7
components can possess anti-neoplastic cell proper-
ties (Sundaram and Miller, 1993). It was reported of the lysis bu€er (20 mM Tris±HCl, pH 7.5 (at
Approximately 5 Â 10 cells were placed in 2 ml
that the components of garlic, such as diallyl sul®de 48C), 1 mM DTT, 1 mM EDTA, 50 mM PMSF,
(
DAS) and diallyl disul®de (DADS), are e€ective 10 mM leupeptin) as previously described (Chung
inhibitors in the growth of neoplastic CMT-13 cells et al., 1993). The suspensions were centrifuged at
Sundaram and Milner, 1993). Particularly, it has 9000 g for 1 min in a Model 3200 Eppendorf/
(
been suggested that garlic intake can modify the Brinkman centrifuge, and the supernatant fraction
risk of colon cancer to women (Steinmetz et al., was subsequently centrifuged at 10,000 g for
1994). DAS and DADS have been shown to be 60 min. The supernatant was kept on ice for NAT
e€ective inhibitors for the promotion phase of activity and protein determinations.
DMBA-induced skin tumours in the mouse
(Belman et al., 1989). However, there are no reports
to address DAS and DADS e€ects on tumour cell
NAT activity determinations
The determination of acetyl-CoA-dependent N-
NAT activity. Thus, our initial studies were focused acetylation of PABA and 2-AF were performed as
on the e€ects of DAS and DADS on the NAT ac- described by Chung et al. (1993). Incubation mix-
tivity of a human colon tumour cell line.
tures in the assay system consisted of a total
Human epidemiological studies suggested an as- volume of 90 ml: cell cytosol, diluted as required, in
sociation between the rapid acetylator phenotypes 50 ml lysis bu€er (20 mM Tris±HCl, pH 7.5, 1 mM
and risks of colorectal cancer. Thus, the hypothesis DTT, 1 mM EDTA), 20 ml of an acetyl-CoA recy-
for an association between acetylator phenotypes cling mixture of 50 mM Tris±HCl (pH 7.5), 0.2 mM
and colorectal cancer suggested an acetylator geno- EDTA, 2 mM DTT, 15 mM acetylcarnitine, 2 U/ml

Garlic components e€ects on arylamine N-acetyltransferase activity
763
Table 1. E€ects of DAS and DADS on human colon tumour cell NAT activity in cytosol
DAS
DADS
2
-AAF (nmol/min/mg N-Ac-PABA (nmol/min/mg 2-AAF (nmol/min/mg N-Ac-PABA (nmol/min/mg
protein)
Concentration
Control
protein)
protein)
protein)
1.28� 0.43
1.12� 0.40
0.82� 0.30
0.63� 0.21
0.48� 0.12
1.06� 0.32
0.88� 0.30
0.70� 0.24
0.53� 0.16
0.38� 0.09
1.28� 0.43
1.04� 0.38
1.06� 0.32
0.82� 0.36
d
1
1
5
1
mM
a
a
d
0 mM
0 mM
00 mM
0.74� 0.32
0.54� 0.19
0.66� 0.21
0.49� 0.18
f
b
b
e
e
c
c
f
0.39� 0.10
0.32� 0.08
Values are mean� SD; n = 3.
a
Di€ers between 10 mM DAS and control; P < 0.05.
b
Di€ers between 50 mM DAS and 10 mM DAS; P < 0.05.
Di€ers between 100 mM DAS and 50 mM DAS; P < 0.055.
c
d
e
Di€ers between 10 mM DADS and control; P < 0.05.
Di€ers between 50 mM DADS and 10 mM DADS; P < 0.05.
Di€ers between 100 mM DADS and 50 mM DADS; P < 0.01.
f
carnitine acetyltransferase, and AF or PABA at the expressed as nmol acetylated substrate/min/mg
speci®c concentrations. The reactions were started cytosolic protein.
by addition of 20 ml acetyl-CoA. The control reac-
Acetylation catalysed by NAT follows `ping-
tions had 20 ml distilled water in place of acetyl- pong' kinetics (Weber, 1987). Increasing the concen-
CoA. For the single point activity measurements, tration of Acetyl-CoA in an assay system contain-
the ®nal concentration of 2-AF or PABA was ing a saturated concentration of the acetylator
0
m
.1 mM and acetyl-CoA was 0.5 mM. The reaction arylamine will increase both apparent K and Vmax
mixtures with or without speci®c concentrations of values of the arylamine substrate. The kinetic con-
DAS or DADS were incubated at 378C for 10 min stant estimate with 0.1 mM acetyl-CoA should
and stopped with 50 ml 20% trichloroacetic acid for therefore approximate those achieved in vivo.
the PABA reactions, and 100 ml acetonitrile for the
Intact cell NAT activity determination
2-AF reactions. All of the reactions (experiments
and controls) were run in triplicate. The amounts of
Human colon tumour cells (in 1 ml RPMI 1640
acetylated product and remaining non-acetylated media with glutamine and 10% FBS) were incu-
6
substrate were determined by HPLC (Chung et al., bated with arylamine substrate at 1Â 10 cells/ml in
1
996). An aliquot of the reaction mixture was individual wells of 24-well cell culture plates with or
injected onto
(
(
a
C18 reversed-phase column without DAS or DADS co-treatment for the time
Spherisorb 4.6Â 250 mm) of a Beckman HPLC indicated at 378C in 95% air/5% CO . At the end
pump 168 and detector 126) and eluted at a ¯ow of incubation, the cells and media were removed
2
rate of 1.2 ml/min. For PABA and N-Ac-PABA, and centrifuged. For experiments with 2-AF, the
the solvent system was 50 mM acetic acid/CH
CN supernatant was immediately extracted with ethyl
86:14) with detection at 266 nm. The retention time acetate±methanol (95:5), the solvent evaporated,
3
(
for PABA was 8.0 min and that for N-Ac-PABA and the residue redissolved in methanol and assayed
was 11.0 min. For 2-AF and 2-AAF, the solvent for acetyl-2-AAF as described above. For exper-
system was 20 mM KH PO (pH 4.5)/CH
2 4 3
CN(53:47) iments with PABA, aliquots of the supernatant
with detection at 280 nm. The retention time for were assayed directly for N-Ac-PABA.
AAF was 6.5 min, and for 2-AF was 9.0 min. All of
Protein determination
the compounds were quantitated by comparison of
the integrated area of the elution peak with that of
Protein concentrations in the human colon
known amounts of standards. NAT activity was tumour cell cytosols from individual donors were
Table 2. E€ects of DAS and DADS on human colon tumour cell NAT activity in intact cells
DAS
DADS
6
6
6
6
2-AAF (nmol/min/10
cells)
N-Ac-PABA (nmol/min/10
cells)
2-AAF (nmol/min/10
cells)
N-Ac-PABA (nmol/min/10
cells)
Concentration
Control
1.17� 0.41
1.10� 0.40
0.96� 0.28
0.90� 0.31
0.78� 0.26
0.62� 0.17
0.51� 0.11
1.17� 0.41
0.98� 0.42
0.77� 0.33
0.62� 0.18
0.41� 0.12
0.96� 0.28
0.89� 0.37
d
1
1
5
1
mM
a
a
d
0 mM
0 mM
00 mM
0.89� 0.31
0.76� 0.24
0.57� 0.19
b
b
e
e
0.71� 0.24
c
c
f
f
0.54� 0.14
0.42� 0.10
Values are mean� SD; n = 3.
a
Di€ers between 10 mM DAS and control; P < 0.05.
b
Di€ers between 50 mM DAS and 10 mM DAS; P < 0.05.
Di€ers between 100 mM DAS and 50 mM DAS; P < 0.055.
Di€ers between 10 mM DADS and control; P < 0.05.
c
d
e
Di€ers between 50 mM DADS and 10 mM DADS; P < 0.05.
Di€ers between 100 mM DADS and 50 mM DADS; P < 0.01.
f

7
64
G. W. Chen et al.
Fig. 1. E€ect of 2-AF and PABA concentrations on the production of 2-AAF and N-Ac-PABA by
human colon tumour cells. Tumour cells were incubated as described for 18 hr at speci®c concen-
trations of 2-AF (panel A) and PABA (panel B). 2-AAF and N-Ac-PABA were measured by HPLC
assay. Each point represents the mean of triplicate assays of three incubations of cells.
RESULTS
determined by the method of Bradford (1976) with
BSA as the standard. All of the samples were
assayed in triplicate.
E€ects of various concentrations of DAS and DADS
on human colon tumour cells NAT activity in cytosol
The possible e€ects of DAS and DADS on NAT
activity in human colon tumour cells, both in cyto-
sols and in intact cells, were examined by HPLC
Statistical treatment of data
Statistical analysis of the data was performed assessing the percentage of acetylation of 2-AF and
with an unpaired Student's t-test. The kinetic con- PABA. The means� SD (standard deviation) of
stants were calculated with the Cleland HYPER NAT activity co-treated with or without DAS or
Program (Cleland, 1967) that performs linear re- DADS with both substrates are given in Tables 1
gression using a least-squares method.
and 2. The data indicated that there was decreased

Garlic components e€ects on arylamine N-acetyltransferase activity
765
Fig. 2. E€ects of incubation time on 2-AAF and N-Ac-PABA production by human colon tumour
cells. Tumour cells were incubated with 2-AF (panel A) and PABA (panel B) at 30 mM and 60 mM for
the time shown. 2-AAF and N-Ac-PABA were measured by HPLC assay. Each point represents the
mean of triplicate assays of three incubations of cells.
NAT activity associated with increased DAS and spectively. In the presence of 1, 10, 50 and 100 mM
DADS both in cytosols and in intact cells. For the DADS, the NAT activities were inhibited 16.4,
cytosol examinations, in the presence of 1, 10, 50 34.2, 47.1 and 65% for 2-AF, and 7.3, 40.7, 53.8
and 100 mM DAS the NAT activities were inhibited and 56.3% for PABA, respectively.
1
2.5, 37, 50.8 and 72.5% for 2-AF, and 17, 34, 50
Metabolism of 2-AF and PABA by human colon
tumour cells
and 74.2% for PABA, respectively. In the presence
of 1, 10, 50 and 100 mM DADS, the NAT activities
were inhibited 18.9, 42.2, 58.9 and 70% for 2-AF,
Human colon tumour cells incubated with 2-AF
and 23.7, 37.8, 53.8 and 70% for PABA, respect- and PABA produced 2-AAF and N-Ac-PABA in
ively. For the intact cell examinations, the NAT ac- the culture media, whereas cells without substrate
6
tivity was 1.17� 0.41 2-AAF nmol/min/10 cells and substrate without cells did not lead to any
6
and 0.96� 0.28 N-Ac-PABA nmol/min/10 cells for detectable 2-AAF and N-Ac-PABA in the media.
acetylation of 2-AF and PABA, respectively. In the The e€ect of the concentration of 2-AF and
presence of 1, 10, 50 and 100 mM DAS, the NAT PABA on 2-AAF and N-Ac-PABA production
activities were inhibited 6.0, 24, 39.1 and 53.9% for are shown in Fig. 1(a and b). The decrease in 2-
2
-AF, and 6.3, 18.7, 35.5 and 48.9% for PABA, re- AAF production at 100 mM 2-AF was probably

7
66
G. W. Chen et al.
Fig. 3. E€ects of DAS or DADS on the production of 2-AAF and N-Ac-PABA by human colon
tumour cells. Tumour cells were incubated as described for 18 hr at the concentration of 2-AF (panel
A) and PABA (panel B) co-treatment with 50 mM DAS and DADS. 2-AAF and N-Ac-PABA were
measured by HPLC assay. Each point represents the mean of triplicate assays of three incubations of
cells.
due to toxicity and loss of viable cells. However, E€ects of 50 mM DAs and DADS on human colon
this observation was not evident in the N-Ac- tumour cell NAT activity in intact cells
PABA cultures. Increased time of incubation led
The NAT activity for the acetylation of 2-AF
to increased 2-AAF and N-Ac-PABA production
(
15, 30 , 60 and 100 mM) measured from intact cells
up to 24 hr (Fig. 2a,b). Figure 2 illustrates the
time course of 2-AAF and N-Ac-PABA pro-
duction from human colon tumour cells at both
was 1.68� 0.08, 3.10� 0.34, 6.48� 0.69 and
6
5
.89� 0.48 nmol/10 cells for acetylation of 2-AF
without co-treatment with DAS and DADS
Table 2). In the presence of 50 mM DAS and
PABA and 18 hr incubation times were used in DADS, the NAT activities were decreased about
most of the experiments. 43±52% and 49±55%, respectively (Fig. 3a). The
3
0 and 60 mM 2-AF and PABA. Based on the
(
data presented in Figs 1 and 2, 60 mM 2-AF and

Garlic components e€ects on arylamine N-acetyltransferase activity
767
Table 3. A€ects of DAS on human colon tumour cell NAT activity in intact cells
2
-AF
PABA
In cytosol
In intact cells
max (nmol/10
In cytosol
In intact cells
V
max (nmol/
min/mg
protein
Vmax (nmol/
6
6
V
min/mg
protein)
V
max (nmol/10
cells)
K
m
(mm)
K
m
(mm)
cells)
K
m
(mm)
m
K (mm)
Control
DAS
DADS
7.04� 1.26
3.00� 0.26
2.63� 0.12
28.57� 6.18
6.99� 0.72
4.20� 0.66
2.62� 0.49
40.00� 7.89
2.21� 0.18
1.91� 0.14
1.40� 0.10
13.16� 1.99
3.37� 0.48
3.08� 0.41
2.64� 0.29
14.70� 2.25
d
a
a
b
c
c
d
a
a
b
c
c
11.36� 1.57
23.20� 4.76
16.00� 2.88
11.23� 1.47
12.82� 1.66
10.87� 1.42
b
d
b
d
9.52� 1.44
8.06� 1.14
Values are mean� SD; n = 3.
The acetyl-CoA and DAS concentrations were 0.1 mM and 50 mM, and the kinetic constants were calculated from the modi®ed HYPER
Program of Cleland (1967).
a
Di€ers between 50 mM DAS or DADS and control; P < 0.05.
Di€ers between 50 mM DAS or DADS and control; P < 0.001.
b
c
Di€ers between 50 mM DAS or DADS and control; P < 0.005.
d
Di€ers between 50 mM DAS or DADS and control; P < 0.001.
NAT activity for the acetylation of PABA (15, 30, on NAT activity in human colon tumour cells. (1),
6
1
0
2
0 and 100 mM) measured from intact cells was NAT enzyme has been reported to exist in many
.18� 0.07, 2.46� 0.30, 5.12� 0.43 and 5.06� kinds of experimental animals, including humans,
.46 nmol/10 cells, respectively, for acetylation of and the NAT has been shown to be involved in
-AF without co-treatment with DAS and DADS. some chemical carcinogenesis (Grant et al., 1991;
6
In the presence of 50 mM DAS and DADS, the Minchin et al., 1992). (2), rapid and slow acety-
NAT activities were decreased about 48±51% and lations have been demonstrated to have a predis-
5
5±59%, respectively (Fig. 3b).
posing factor for the sensitivity of individuals to the
toxicity during exposure to many arylamines
(Weber and Hein, 1985). Therefore, the genetically
mediated variation in NAT activities within target
E€ect of 50 mM DAS and DADS on kinetic constants
of NAT in human colon tumour cells
In the presence or absence of DAS and DADS, tissues for arylamine-induced neoplasm may indi-
speci®c concentrations of 2-AF and PABA (0.373, cate di€erential risks within the human population.
0
.435, 0.543, 0.745, 1.102 and 2.205 mM, respect- (3), it was reported that colorectal adenoma and
ively) were added to the recycling mixtures for cancer risks were in relation to meat intake and
determining human colon tumour cell NAT kinetic acetylator status (Robert-Thomson et al., 1996). (4),
constants. For the cytosol examinations, the appar- inhibition e€ects on cultured cells tumour and bac-
ent Km and Vmax values were 7.04� 1.26 mM and teria have been associated with garlic (Cavallito
2
2
8.57� 6.18 nmol/min/mg protein for 2-AF, and et al., 1954; Sadhana et al., 1988; Sundaram et al.,
.21� 0.18 mM and 13.16� 1.99 nmol/min/mg pro- 1993). (5), consumption of garlic was inversely re-
tein for PABA, respectively (Table 3). When 50 mM lated with colon cancer risk in women (Steinmetz
DAS and DADS were added to the reaction mix- et al., 1994). (6), DAS and DADS a€ects
tures, the apparent K and Vmax values were Helicobacter pylori growth and NAT activity
m
decreased 59.5% and 60.3% , 63% and 72.7% for (Chung et al., 1997). (7), arylamine NAT activity
acetylation of 2-AF, 14.6% and 14.7% , 37.7% and towards 2-AF and PABA was detected in human
3
8.2% acetylation of PABA, respectively. For the colon tumour cell lines (Lai et al., 1997). Therefore,
intact cell examinations, the apparent K and Vmax it is of interest to study whether DAS and DADS,
m
values were 6.99� 0.72 mM and 40.00� 7.89 nmol/ the garlic components, would a€ect the NAT ac-
min/mg protein for 2-AF and 3.37� 0.48 mM and tivity of human colon tumour cells.
1
4.70� 2.25 nmol/min/mg protein for PABA, re-
The substrate speci®city for NAT1 is di€erent
spectively (Table 3). When 50 mM DAS and DADS from that for NAT2 in humans (Weber and Hein,
were added to the reaction mixtures, the apparent 1985). 2-AF was chosen in the present study
m
K and Vmax values were decreased 40% and because it is the common substrate for both NAT1
7
9
2.6%, 42% and 60% for acetylation of 2-AF, and NAT2, while PABA is a substrate for NAT1
.6% and 12.8%, 22.7% and 26.1% for acetylation only, and the authors' interest is comparing the
of PABA, respectively. Clearly, both K
m
and Vmax metabolism of a carcinogen (2-AF) to a non-carci-
values on the human colon tumour cells NAT were nogen (PABA). To simulate the conditions of the
decreased in the presence of DAS and DADS both human large intestine, a pH of 7.5 was selected for
in cytosol and intact cell examinations.
all of the experiments. This pH is close to that of
the normal large intestine. The data from the cyto-
sol and intact human colon tumour cell studies
showed that there were signi®cant di€erences in
DISCUSSION
There are many events that are likely to be prere- NAT activity between the control and DAS or
quisites for the observed e€ects of DAS or DADS DADS treatment groups. The present studies

7
68
G. W. Chen et al.
demonstrated that DAS or DADS can markedly colorectal cancer (Robert-Thomson et al., 1996).
inhibit the NAT activity in human colon tumour The relation between meat intake, acetylator status,
cells in both systems studied and the inhibition has and colorectal carcinogenesis may involve exposure
a
dose-dependent e€ect. Other reports already to heterocyclic amines that are generated by cook-
demonstrated that DAS and DADS does inhibit the ing meat for long periods at high temperatures. The
growth of neoplastic CMT-13 cells (Sundaram and present study may o€er some information about the
Milner, 1993). Currently, we are working on the in- e€ects of the garlic compounds DAS and DADS on
hibitory activity of DAS and DADS on human NAT activity, because NATs are expressed in
colon tumour cells. Essential oils of garlic are human ileum and colon (Ilett et al., 1987; Paci®c
known to contain approximately 60% DADS et al., 1986) and are able to metabolize arylamine
(
Dausch and Nixon, 1990); the present study also compounds. However, it is not known whether
indicated that DADS induced higher inhibition of decrease of NAT activity would decrease tumour
NAT activity in human colon tumour cells than did production or whether DAS or DADS could pre-
DAS. In other words, the potency of oil-soluble vent the development of colon cancer. Other investi-
organosulfur compounds from garlic increased as gators have demonstrated that elevated levels of
the number of sulfur atoms increased.
NAT activity are associated with increased sensi-
In the present study human colon tumour cell tivity to the mutagenic e€ects of many arylamines
lines produced acetylated metabolites from 2-AF (Einisto et al., 1991). Further investigation is
and PABA. The results also demonstrated that the needed.
acetylated metabolites produced increased gradually
as the incubation time increased up to 18 hr for 2-
AF and 24 hr for PABA. The decrease in 2-AAF
Conclusions
In summary, the present studies show that the
production at 100 mM 2-AF was possibly due to organosulfur compounds DAS and DADS present
toxicity and loss of viable cells. The results are in in garlic are not equally e€ective in the inhibition of
agreement with an earlier study on mice leucocytes human colon tumour cell NAT activity in vitro and
(
Chung et al., 1993), but the present study showed in intact cells, but both DAS and DADS did
no e€ects in treatment with PABA. It was also decrease NAT activity of the human colon tumour
determined that with an increase in the concen- cells. This report is also the ®rst to demonstrate
trations of 2-AF and PABA, the production of 2- that DAS or DADS from garlic can inhibit NAT
AAF and N-Ac-PABA was increased. Therefore, an activity of a human colon tumour cell line.
18-hr incubation time and 50 mM DAS or DADS
was selected for the intact cell NAT activity exam-
AcknowledgementsÐThis work was supported by grants
inations. The data indicated that DAS or DADS CMC-SCIWCM-97001 from the Supervisory Committee
for the Integration of Western and Chinese Medicine.
decreased the NAT kinetic constant in the cytosols
and in intact cells, but DADS induced higher e€ects
than DAS.
It is dicult at this point to extrapolate the quan-
REFERENCES
tity of garlic or its oil that would need to be con- Andres H. H., Kolb H. J. and Weiss L. (1983) Puri®cation
and
physical-chemical
properties
of
acetyl-
sumed by humans to potentially reduce the NAT
activity of human colon tumours. Nevertheless, an
intake of 20 g per day has been reported in some
areas of the world (Horwitz, 1981). This intake has
been correlated with a reduction in the incidence of
stomach cancers. DAS or DADS did not show
cytotoxicity at the concentrations used in the pre-
sent of study. The report showed that rapid acetyla-
tors are at higher risk for colorectal cancers
CoA:arylamine N-acetyltransferase from pigeon liver.
Biochimica et Biophysica Acta 746, 182±192.
Barone F. E. and Tonsey T. M. (1977) Isolation, puri®-
cation, identi®cation, synthesis and kinetics of activation
of the anticandidal component of Allium sativum and a
hypothesis for its mode of action. Mycologia 69, 793±
825.
Belman S., Solomon T., Segal A., Block E. and Barany
G. (1989) Inhibition of soybean lipoxygenase and mouse
skin tumor promotion by onion and garlic components.
Journal of Biochemical Toxicology 4, 151±160.
Blum M., Grant D. M., McBride O. W., Heim H. and
Meyer U. A. (1990) Human arylamine N-acetyltransfer-
ase genes: isolation, characterization, chromosomal
localization, and functional expression. DNA and Cell
Biology 9, 193±203.
Bordia A. (1978) E€ect of garlic on human platelet aggre-
gation. Atherosclerosis 30, 355±360.
Boullin D. J. (1981) Garlic: a platelet inhibitor. Lancet i,
776±777.
(
Ladero et al., 1991). A metabolic characteristic of
potential relevance to colorectal carcinogenesis
involved polymorphic NAT2, an inherited pheno-
type in the rapid and slow acetylators as classi®ed
by drug metabolism (Lang et al., 1986) or genotype
(
Hickman and Sim, 1991). Many studies have
attempted to relate acetylation phenotype to cancer
susceptibility. Except for bladder cancer, in other
tumours including colon tumours, the relationship Bradford M. M. (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein uti-
lizing the principle of protein-dye binding. Analytical
not clear. Recently, it has been demonstrated that
between acetylator phenotypes and susceptibility is
Biochemistry 72, 248±254.
Cartwright R. A., Glasham R. W., Rogers H. J., Ahmed
R. A., Barham-Hall D., Higgens E. and Kahn
the combination of fast-acetylator phenotypes with
a high consumption of meat increased the risk of

Garlic components e€ects on arylamine N-acetyltransferase activity
769
M. A. (1982) Role of N-acetyltransferase phenotype in
bladder carcinogenesis: A pharmacogenetic epidemiolo-
gical approach to bladder cancer. Lancet ii, 842±846.
Cavallito C. J., Buck J. S. and Suter C. M. (1954) Allicin,
the antibacterial principle of Allium sativum. II:
Determination of the chemical structure. Journal of the
American Chemical Society 66, 1952±1954.
Chung J. G., Kuo H. H., Lin T. H., Ho C. C., Lee J. H.,
Lai J. M., Levy G. N. and Weber W. W. (1996)
Evidence for arylamine N-acetyltransferase in the nema-
tode Anisakis simplex. Cancer Letters 106, 1±8.
the polymorphic expression of N-acetyltransferase ac-
tivity in human colon and bladder cytosol. Proceedings
of the American Association for Cancer Research 30,
160±166.
Kirlin W. G., Trinidad A., Yerokun T., Ogolla F.,
Ferguson R. J., Andrews A. F., Brady P. K. and Hein
D. W. (1989b) Polymorphic expression of AcCoA-
dependent arylamine N-acetyltransferase and AcCoA-
dependent O-acetyltransferase activation of N-hydro-
xyarylamines by human bladder cytosol. Cancer
Research 49, 2448±2454.
Chung J. G., Levy G. N. and Weber W. W. (1993) Ladero J. M., Gonzalez J. F., Benitez J. and Vargas
Distribution of 2-amino¯uorene and p-aminobenzoic
acid N-acetyltransferase activity in tissues of C57BL/6J
E. (1991) Acetylator polymorphism in human colorectal
carcinoma. Cancer Research 51, 2098±2100.
Lai J. M., Chung J. G., Wang H. H., Chen G. W., Wu
L. T. and Tsai C. H. (1997) Human colon tumor aryla-
mine N-acetyltransferase characterization. Medical
Journal CMCH 2, 93±103.
Lang N. P., Chu D. Z. J., Hunter C. F., Kendall D. C.,
Flammang T. J. and Kadlubar F. F. (1986) Role of aro-
matic amine acetyltransferase in human colorectal can-
cer. Archives of Surgery 121, 1259±1261.
Lower G. M. (1982) Concepts in causality: chemically
induced human urinary bladder cancer. Cancer 49,
S
rapid and B6. A-Nat slow acetylator congenic mice.
Drug Metabolism and Disposition 21, 1057±1063.
Chung J. G., Wang H. H., Tsou M. F., Hsieh S. E., Lo
H. H., Yen Y. S., Chang S. S., Wu L. T., Lee J. H. and
Hung C. F. (1997) Evidence for arylamine N-acetyl-
transferase activity in the bacterium Helicobacter pylori.
Toxicology Letters 91, 63±71.
Cleland W. W. (1967) The statistical analysis of enzyme
kinetics data. Advances in Enzymology 29, 1±32.
Dausch J. G. and Nixon D. W. (1990) Garlic: a review of
its relationship to malignant disease. Preventive
Medicine 19, 346±361.
Deguchi T., Sakamoto Y., Sasaki S. and Uyemura
K. (1988) Arylamine N-acetyltransferase from chicken
liver. I. Monoclonal antibodies, immunoanity puri®-
cation, and amino acid sequences. Journal of Biological
Chemistry 263, 7528±7533.
Einisto P., Watanabe M., Ishidate M. and Nohmi
T. (1991) Mutagenicity of 30 chemicals in S. typhimur-
ium possessing di€erent nitroreductase or O-acetyltrans-
ferase activities. Mutation Research 259, 95±102.
Ernst E., Weihmayr T. H. and Matrai A. (1985) Garlic
and blood lipids. British Medical Journal 291, 139±143.
Fang S. H., Chung J. G., Chang W. W. and Chang
S. H. (1997) Candida albicans arylamine-acetyltransfer-
ase: kinetics and characterization. Toxicology Letters 92,
1056±1066.
Minchin R. F., Reeves P. T., Teitel C. H., McManus
M. E., Mojarrabi T., Ilett K. F. and Kadlubar
F. F. (1992) N-and O-acetylation of aromatic and het-
erocyclic amine carcinogens by human monomorphic
and polymorphic acetyltransferase expressed in COS-1
cells.
Communications 185, 839±844.
Biochemical
and
Biophysical
Research
Moore G. S. and Atkins R. D. (1977) The fungicidal and
fungistatic e€ect of an aqueous garlic extract on medi-
cally important yeast like fungi. Mycologia 69, 341±348.
Nagao M. and Sugimura T. (1993) Carcinogenic factors in
food with relevance to colon cancer development.
Mutation Research 290, 43±51.
Ohsako S. and Deguchi T. (1990) Cloning and expression
of cDNAs for polymorphic and monomorphic N-acetyl-
transferase from human liver. Journal of Biological
Chemistry 265, 4630±4634.
1
09±116.
Grant D. M., Blum M., Beer M. and Meyer U. A. (1991)
Monomorphic and polymorphic human arylamine N-
acetyltransferase : a comparison of liver isozymes and
expressed products of two cloned genes. Molecular
Pharmacology 39, 184±191.
Hein D. W., Omichinski J. G., Brewer D. W. and Weber
W. W. (1982b) A unique pharmacogenetic expression of
the N-acetylation polymorphism in the inbred hamster.
Journal of Pharmacology and Experimental Therapy 220,
Ohsako S., Ohtomi M., Sakamoto Y., Uyemura K. and
Deguchi T. (1988) Arylamine N-acetyltransferase from
chicken liver. Journal of Biological Chemistry 263, 7534±
7538.
Paci®c G. M., Benicini C. and Rane A. (1986)
Acetyltransferase in humans: development and tissue
distribution. Pharmacology 32, 283±291.
Parker H. G. and Evans A. E. Z. (1982) Epidemiology of
aromatic amine cancers. In Chemical Carcinogens, 2nd
edn, pp. 277±302. American Chemical Society,
Washington, DC ACS Monograph 182.
Robert-Thomson I. C., Ryan P., Khoo K. K., Hart W. J.,
McMichael A. J. and Burtler R. N. (1996) Diet, acetyla-
tor phenotype, and risk of colorectal neoplasia. Lancet
8
±15.
Hein D. W., Rustan J. D., Bucher K. D., Furman E. J. and
Martin W. J. (1991) Extrahepatic expression of the N-
acetylation polymorphism toward arylamine carcinogens
in tumor target organs of an inbred rat model. Journal
of Pharmacology and Experimental Therapy 258, 232±
2
36.
347, 1372±1374.
Hein D. W., Smolen T. N., Fox R. R. and Weber
W. W. (1982a) Identi®cation of genetically homozygous
rapid and slow acetylators of drugs and environmental
carcinogens among established inbred rabbit strains.
Journal of Pharmacology and Experimental Therapy 223,
Sadhana J., Rao A. R., Kucheria K. and Bijani V. (1988)
Inhibitory action of garlic oil on the initiation of ben-
zo[a]pyrene-induced skin carcinogenesis in mice. Cancer
Letters 40, 193±197.
4
0±44.
Steinmetz K. A., Kushi L. H., Bostick R. M., Folsom
A. R. and Potter J. A. (1994) Vegetables, fruit, and
colon cancer in the Iowa Woman's health study.
American Journal of Epidemiology 139, 1±15.
Hickman D. and Sim E. (1991) N-acetyltransferase poly-
morphism: comparison of phenotype and genotype in
humans. Biochemical Pharmacology 42, 1007±1014.
Horwitz N. (1981) Garlic as a plat du jour: Chinese study Sundaram S. G. and Milner J. A. (1993) Impact of orga-
®
nds it could prevent G. I. cancer. Med. Tri. August 12.
nosulfur components in garlic on canine mammary
tumor cells in culture. Cancer Letters 74, 85±90.
Vatsis K. P. and Weber W. W. (1993) Structural hetero-
geneity of caucasian N-acetyltransferase at the NAT1
gene locus. Archives of Biochemistry Biophysics 301, 71±
76.
Ilett K. F., David B. M., Detchon P., Castledon W. M. and
Kwa R. (1987) Acetylator phenotype in colorectal carci-
noma. Cancer Research 47, 1466±1469.
Kirlin W. G., Andrews A., Ogolla F., Yerokun T.,
Ferguson R. J. and Hein D. W. (1989a) Comparison of

7
70
G. W. Chen et al.
Weber W. W. (1987) The Acetylator Genes and Drug
Response. pp. 21±55. Oxford University Press, New
York.
Weber W. W. and Hein D. W. (1985) N-acetylation phar-
macogenetics. Pharmacology Reviews 37, 25±79.
Whitaker D. P. and Goosey M. W. (1993) Puri®cation
and properties of the enzyme arylamine N-acetyltrans-
ferase from the house¯y Musca domestica. Biochemical
Journal 295, 149±154.