Formation of a carcinogenic aromatic amine from an azo dye by human skin bacteria in vitro

Article abstract of DOI:10.1191/096032799678845061

Azo dyes represent the major class of dyestuffs. They are metabolised to the corresponding amines by liver enzymes and the intestinal microflora following incorporation by both experimental animals and humans. For safety evaluation of the dermal exposure of consumers to azo dyes from wearing coloured textiles, a possible cleavage of azo dyes by the skin microflora should be considered since, in contrast to many dyes, aromatic amines are easily absorbed by the skin. A method for measuring the ability of human skin flora to reduce azo dyes was established. In a standard experiment, 3 X 10¹¹ cells of a culture of Staphylococcus aureus were incubated in synthetic sweat (pH 6.8, final volume 20 mL) at 28°C for 24 h with Direct Blue 14 (C.I. 23850, DB 14). The reaction products were extracted and analysed using HPLC. The reduction product o-tolidine (3,3'-dimethylbenzidine, OT) could indeed be detected showing that the strain used was able to metabolise DB 14 to the corresponding aromatic amine. In addition to OT, two further metabolites of DB 14 were detected. Using mass spectrometry they were identified as 3,3'-dimethyl-4-amino-4'-hydroxybiphenyl and 3,3'-dimethyl-4-aminobiphenyl. The ability to cleave azo dyes seems to be widely distributed among human skin bacteria, as, under these in vitro conditions, bacteria isolated from healthy human skin and human skin bacteria from strain collections also exhibited azo reductase activity. Further studies are in progress in order to include additional azo dyes and coloured textiles. At the moment, the meaning of the results with regard to consumer health cannot be finally assessed.

Full text of DOI:10.1191/096032799678845061

Human & Experimental Toxicology
http://het.sagepub.com/
Formation of a carcinogenic aromatic amine from an azo dye by human skin bacteria in vitro
T Platzek, C Lang, G Grohmann, U-S Gi and W Baltes
Hum Exp Toxicol 1999 18: 552
DOI: 10.1191/096032799678845061
The online version of this article can be found at:
http://het.sagepub.com/content/18/9/552
Published by:
http://www.sagepublications.com
Additional services and information for Human & Experimental Toxicology can be found at:
Email Alerts: http://het.sagepub.com/cgi/alerts
Subscriptions: http://het.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://het.sagepub.com/content/18/9/552.refs.html
>> Version of Record - Sep 1, 1999
What is This?
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Human & Experimental Toxicology (1999) 18, 552±559
ã
1999 Stockton Press All rights reserved 0144 ± 5952/99 $15.00
http://www.stockton-press.co.uk/het
Formation of a carcinogenic aromatic amine
from an azo dye by human skin bacteria
in vitro
T Platzek*,1, C Lang2, G Grohmann2, U-S Gi3 and W Baltes3
1Federal Institute for Health Protection of Consumers and Veterinary Medicine, Postfach 330013, D-14191 Berlin,
Germany; 2Institute for Microbiology and Genetics, Technical University Berlin, Gustav-Meyer-Allee 25, D-13355 Berlin,
Germany; 3Institute for Food Chemistry, Technical University Berlin, Gustav-Meyer-Allee 25, D-13355 Berlin, Germany
Azo dyes represent the major class of dyestuffs. They are
metabolised to the corresponding amines by liver enzymes
and the intestinal micro¯ora following incorporation by
both experimental animals and humans. For safety
evaluation of the dermal exposure of consumers to azo
dyes from wearing coloured textiles, a possible cleavage of
azo dyes by the skin micro¯ora should be considered
since, in contrast to many dyes, aromatic amines are easily
absorbed by the skin. A method for measuring the ability of
human skin ¯ora to reduce azo dyes was established. In a
to metabolise DB 14 to the corresponding aromatic amine.
In addition to OT, two further metabolites of DB 14 were
detected. Using mass spectrometry they were identi®ed as
3,3'-dimethyl-4-amino-4'-hydroxybiphenyl and 3,3'-di-
methyl-4-aminobiphenyl. The ability to cleave azo dyes
seems to be widely distributed among human skin
bacteria, as, under these in vitro conditions, bacteria
isolated from healthy human skin and human skin
bacteria from strain collections also exhibited azo
reductase activity. Further studies are in progress in order
to include additional azo dyes and coloured textiles. At the
moment, the meaning of the results with regard to
consumer health cannot be ®nally assessed.
standard experiment, 361011 cells of
a culture of
Staphylococcus aureus were incubated in synthetic sweat
(pH 6.8, ®nal volume 20 mL) at 288C for 24 h with Direct
Blue 14 (C.I. 23850, DB 14). The reaction products were
extracted and analysed using HPLC. The reduction
product o-tolidine (3,3'-dimethylbenzidine, OT) could
indeed be detected showing that the strain used was able
Keywords: azo dyes; skin bacteria; metabolism
Introduction
Azo colorants are the most important textile color-
ants which are divided, according to their solubi-
lity, into azo dyes (soluble) and azo pigments
(insoluble). They are synthesised from arylamines
by diazotation and coupling and are characterised
by the azo group (R-N=N-R). It is well-known that
soluble azo dyes when incorporated into the body
are split into the corresponding aromatic amines by
liver enzymes and the intestinal micro¯ora.1,2
Several aromatic amines are, or are suspected to
be, carcinogenic. It has therefore been stated that
`all azo colorants whose metabolism can liberate a
carcinogenic arylamine are suspected of having
carcinogenic potential'.3 In Germany, clothing
textiles containing azo dyes that may release certain
aromatic amines may not be marketed according to
the Regulations on Speci®c Commodities (under the
Foods Act).4 In the past, azo reductase activity has
been reported for the mammalian intestinal micro-
¯ora and mammalian liver enzymes.1,2 An analo-
gous activity of skin bacteria is not known. For
safety evaluation of the dermal exposure of con-
sumers to azo dyes from wearing coloured textiles,
it is relevant to consider a possible cleavage of azo
dyes by the skin micro¯ora since, in contrast to
many dyes, aromatic amines are easily absorbed via
the skin.5
In this study, the capability of human skin
bacteria to metabolise azo dyes was investigated
by measuring the reduction to the corresponding
amines. As model compound Direct Blue 14 (DB 14)
was used. DB 14 is a typical member of the group of
direct dyes which are de®ned as water-soluble dyes
capable of directly dyeing cellulose ®bres. In a
review of 1975 its use for dyeing textiles and leather
was addressed.6 More recent data are not available.
A standardised assay was established to determine
the reductive capacity of skin bacteria using
collections of bacterial strains and freshly isolated
ones from healthy human skin.
Materials and methods
Substances
*Correspondence: T Platzek
Received 1 February 1999; revised 26 April 1999; accepted 3 May
1999
Direct Blue 14 (Trypan Blue, C.I. 23850, DB 14) and
o-tolidine (3,3'-dimethylbenzidine, OT) were pur-
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
553
chased from Aldrich Inc. Since it was reported that
OT is a contaminant of DB 14 the content of OT in
DB 14 was investigated using HPLC and was found
to be less than 5 p.p.m.
prepared sweat simulant. Sweat simulant was
prepared by dissolving 0.5 g L-histidine-monohy-
drochloride-1-hydrate, 5 g sodium chloride and
2.2 g sodium dihydrogenphosphate-2-hydrate in
1 L distilled water and adjusting the pH to 6.8. The
dye was added to the bacterial suspension in sweat
simulant (7 mg DB 14 in a standard assay) and the
mixture was incubated at 288C for 24 h under
shaking (frequency 40 r.p.m.) in a water bath. The
reaction was stopped by cooling on ice, the pH was
adjusted to 11 with 10 N NaOH, and the amines were
eluted on Extrelut1 colunmns with 4620 mL t-
butylmethyl-ether. The t-butylmethylether extract
was then treated as described above.
Chemical azo cleavage
Chemical splitting of the dye was performed
according to the of®cial method by reduction with
sodium dithionate in a citrate-buffered aqueous
solution (pH 6) at 708C in a closed vessel.7 The
amines released in the process of reductive cleavage
were transferred to a t-butylmethyl-ether phase by
means of liquid-liquid extraction using Extrelut1 20
columns (Merck). The eluate was concentrated to
1 mL in vacuo. The remaining of the solvent was
removed by nitrogen. The residue was redissolved
in 500 mL methanol and directly analysed by HPLC.
HPLC separation
HPLS separation was achieved using the Beckman
System Gold. Stationary phase: LiChrosorb 60 RP-
select B (5 mm); column: 25064.6 mm, (M&W,
Berlin, Art.-Nr. 254671005); Eluent 1/2: Methanol
LiChrosolv/0.575 g ammonium hydrogenphos-
phate/L, 0.7 g disodium hydrogenphosphate/L, pH
6.9; Elution: Eluent 1/eluent 2 (60/40 V/V), ¯ow
1.0 mL/min; oven temperature: 408C, injection
volume 10 or 50 mL; Detector: Diode array detector
(DAD), wavelengths: 207 nm and 280 nm.
The detection limit of OT was 29 ng (136 pmol)
per chromatographic run (50 mL of 500 mL injected)
corresponding to 1.4 nmol OT per assay. The
recovery rate was determined by adding various
amounts of OT to the mixture (sweat simulant with
bacteria without DB 14) after the 24 h incubation
period. Depending on the amount added, the
recovery rate was 76 to 93%. Under these condi-
tions, the recovery rate from ®ve assay vials
containing the same amount of OT (4.3 mg) was
88+4%. Nevertheless, 3,3'-dichlorobenzidine was
used as an internal standard in most of the
experiments.
Bacteria
The following bacterial strains were used (strain
numbers are those of VLSF, Institut fuÈ r GaÈrungsge-
werbe und Biotechnologie, Berlin, FRG): Micro-
coccus luteus (Mi-0702), Micrococcus roseus (Mi-
0401), Micrococcus varians (Mi-0503), Staphylo-
coccus aureus (St-0411), Staphylococcus epidermi-
dis (St-0201). Additional strains used were St.
aureus DSM 1104, St. epidermis DSM 1798, and
St. epidermis DSM 22044. Strain Staphylococcus
aureus TU 5 ± 3 (previously designated as St.
epidermis) was used to establish the standardised
assay.8
Skin bacteria isolates were taken from the left
bend of the elbow of six healthy individuals. There
was a time interval of at least 24 h between the last
washing and the time of collection of bacteria.
Bacteria were grown and subcultured on Standard I
Agar (Merck, Germany). Morphologically different
types of colonies were identi®ed and one colony of
each type was randomly picked. Pure cultures were
obtained by repeated restreaking. Isolates A to E
were used in the subsequent assays. The isolates
were typed according to Bergey's Manual of
Systematic Bacteriology.9 Strains were stored on
nutrient Agar slopes at 48C or in nutrient broth plus
17% glycerol at 7708C.
Capillary gas chromatography/mass spectrometry
Apparatus: HP 5890A, coupled with Finnigan SSQ;
GC-column: J & W fused silica capillary column DB-
5; dimension 50 m60.32 mm, ®lm thickness
0.1 mm; injection volume 1 mL, on column tempera-
ture programme: 08C, 1 min, 108C/min, 2408C, 38C/
min, 2908C, 10 min, gas: helium; EI parameter:
Transline temperature 2808C, ion source 1508C,
ionisation energy: 70 eV, source pressure ca.
161076 Torr, cyclic scan 0.9 s, mass area 45 ± 350;
CI parameter: Reactant gas methane, source pres-
sure 7,000 Torr, mass area 80 ± 350.
Culture conditions
The strains were grown in a preculture (20 mL
Standard I broth, Merck No. 7782) at 288C for 62 h
on a reciprocal shaker (120 r.p.m.). One mL of the
bacterial solution was inoculated into 120 mL
Standard-I-Medium for the main culture which was
incubated at 288c for 48 h. If required, induction was
achieved by adding DB 14 (1 mg/mL) to the main
culture. In a typical experiment about 361011
bacteria (about 400 OD610) corresponding to 60 mL
grown culture, were used per assay. Cells of the
stationary phase (48 h growth) were harvested by
centrifugation and resuspended in 20 mL of freshly
Results
Chemical reduction of DB 14
According to the above-mentioned method DB 14
was reduced with sodium dithionate. The reaction
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
554
products were extracted and isolated as described
above. The expected cleavage product OT (see
Figure 1) was identi®ed on the basis of the retention
time (7.17 min) and the u.v. spectrum as compared
to an authentic standard.
respectively). No reaction products were seen when
DB 14 was incubated in sweat simulant under
standard conditions in the absence of bacteria.
Likewise reaction products were absent in the
bacteria suspension when no azo dye had been
included in the reaction mixture.
Reaction of DB 14 with St. aureus TU 5 ± 3
Similarly, three cleavage products were detected
in a total ion current chromatogram by GC/MS
analysis of the reaction products (see Figure 3).
Using EI- and CI-mass spectrometry, the compound
at the HPLC retention time 8.09 min was identi®ed
as 3,3' - dimethyl - 4 - amino - 4'-hydroxybiphenyl
(DAHBP) whereas the compound at retention time
19.66 min was assigned the structure of 3,3'-
dimethyl-4-aminobiphenyl (DABP) (see Figures 4
and 5).
Seven mg (7.3 mmol) DB 14 was incubated with
about 361011 cells of St. aureus TU 5 ± 3 for 24 h.
This strain which had been deposited in the
collection as St. epidermidis was chosen as a typical
representative of the major class of human skin
bacteria.10 The strain was reclassi®ed as St. aureus
during this study. The reaction products were
extracted and isolated as described above. The
HPLC separation of the reaction product is shown
in Figure 2. In addition to OT, two further u.v.-
absorbing compounds were detected during HPLC
separation (retention times 8.09 and 19.66 min,
The CI mass spectrum of DAHBP indicates a
molecular mass of 213 and the existence of one
nitrogen atom (quasi molecular peaks M+H+=m/z
Figure 1 Azo cleavage of DB 14 and formation of OT
Figure 2 HPLC separation of the reaction product following incubation of DB 14 with St. aureus TU5-3 (standard conditions)
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
555
214; M+28=m/z 242). Furthermore, the difference of
m/z 17 between the fragment ions m/z 169 and m/z
152 may be assigned to a hydroxyl group. This
group is additionally indicated by fragments m/z
107 and m/z 106 formed by phenyl cleavage. In the
case of DABP, the loss of the amino group is
indicated by the molecular peak m/z 197.
An additional reaction product was found when
the bacteria-dye mixture was incubated under
anaerobic and inductive conditions for 72 h (mole-
cular peak m/z 239, quasi molecular peaks:
M+H+=m/z 240; M+C2H4+=m/z 268). The spectra
indicate an N-acetylated derivative, 3,3'-dimethyl-
4-acetylaminobiphenyl.
assay resulted in a 8.1-fold increase in the amount of
OT formed (Table 1).
Due to the lack of standards for the metabolites
DABP and DAHBP which are by-products in the
reaction, the relative amounts of OT, DABP, and
DAHBP were quanti®ed based on the peak areas
measured in the HPLC analyses (Table 2). The data
have shown that the relative amount of OT is
dependent on the assay conditions and that the
amount of OT formed might not be fully representa-
tive of the total metabolic activity of the cells.
DB 14-reductive capacity of normal skin ¯ora
Additional strains representative of human skin
¯ora or strains freshly isolated from healthy
individuals were assayed for their reductive capa-
city using DB 14 as a substrate. The cells were
grown under standard conditions and OT as the
primary reaction product was quanti®ed (Table 3).
Interestingly, most of the strains analysed exhibited
reductive activity and produced the characteristic
cleavage product, OT.
Quantitative aspects
OT as the expected product of the reduction of DB
14 was quanti®ed in the standard assay and in
assays using varying experimental parameters
(Table 1). Depending on the assay conditions, 3.2
to 26.0 mg (15 to 123 nmol) OT were formed
following incubation of either 7 mg (7.3 mmol) DB
14 which resembles a reduction rate of 0.2 to 1.7%.
Inclusion of the azo dye in the culture medium
(`inductive conditions') resulted in a slight increase
in the reaction product (+15% as compared to the
standard conditions). A much more signi®cant
effect (+90%) was seen when the cells were
harvested in a different growth phase (late log to
early stat phase versus late stat phase under
standard conditions). The increase in OT formation
using a high cell concentration was non-linear: a
2.6-fold increase in the number of bacterial cells per
Discussion
Azo dyes are important colorants. More than 2000
of these substances have been synthesised, and
about 500 of them are based on carcinogenic
aromatic amines.11,12 Following incorporation they
are metabolised to the corresponding aromatic
amines. In the mammalian liver, azo compounds
are metabolised by cytosolic and microsomal
Figure 3 Total ion current chromatogram of GC/MS analysis of the reaction product following incubation of DB 14 with St. aureus
TU5-3 (standard conditions)
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
556
Figure 4 EI and CI mass spectra of DAHBP
enzymes, e.g. by reductive cleavage to the amines.
The intestinal micro¯ora plays an even more
important role.1,2 It was found that bacterial strains
in human faeces are capable of cleaving azo dyes.13
It has been known for decades that certain azo dyes,
which are split into carcinogenic aromatic amines
like benzidine, are carcinogenic in both experi-
mental animals and humans.14
Salmonella assay was developed and DB 14 was
found to be mutagenic under these conditions.15,16
The signi®cance of azo-reduction in the mutagen-
esis and carcinogenesis of azo dyes was reviewed
earlier.17 Additionally, for DB 14 reproductive
toxicity is worth mentioning.18
The reductive cleavage of azo dyes during
percutaneous absorption was investigated in vitro
using skin from mice, guinea pigs, and humans. All
species tested were capable of reductive cleavage of
The mutagenicity of benzidine and benzidine-
congener dyes is well documented. A modi®ed
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
557
Figure 5 EI and CI mass spectra of DABP
the dyes.19 Following epicutaneous treatment of rats
in vivo with a 14C-labelled azo dye, a signi®cant
amount of radioactivity was found in urine and
faeces. It was speculated that azo cleavage resulting
in the formation of aromatic amines is mediated via
the micro¯ora of the rat skin.20
The micro¯ora of the human skin is variable with
regard to bacterial strains and species. However,
staphylococci and micrococci represent major
Table 1 Formation of OT after incubation with S. aureus TU5-3
cells in sweat simulanta
Assay condition
OT (mg per assay)
Standard assay
3.2
3.7
6.1
26.0
Standard assay, inductive conditions
Cells from early-stationary growth phase
Concentrated cell suspension (2.66)
aData represent mean values of at least two independent
experiments
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
558
Table 2 Relative amounts of OT, DABP, and DAHBP after
incubation of DB 14 with St. aureus TU5-3 cells in sweat
simulanta
tional reaction product was found following an
incubation period of 72 h which was identi®ed as a
deaminated/N-acetylated derivative of OT. Similar
microbially-produced metabolites (deaminated and
N-acetylated derivatives) were found by Manning et
al. following incubation of a culture derived from
human faeces with the benzidine-based azo dye
Direct Black 38. The authors demonstrated that the
mutagenicity was increased by the human intestinal
micro¯ora.22,23
Area
of OT
Area of
DABP
Area of
DAHBP
Assay condition
(relative) (relative) (relative)
Standard assay
71 (1.0) 152 (2.1) 91 (1.3)
159 (1.0) 214 (1.4) 118 (0.7)
Cells from early-stationary
growth phase
Concentrated cell suspension 390 (1.0) 296 (0.8) 156 (0.4)
(2.66)
aData from one representative experiment
The wearing of textiles is assumed to be the
most widespread exposure to colorants, but the
exposure by this route was estimated to be less
than 1 mg per individual per day.24 The release
rate of dyes from coloured textiles was measured,
and from these data, the extent of exposure was
calculated to be 1 ± 10 mg per kg body weight.25,26
However, in the case of garments not having been
dyed according to state-of-the-art technology one
has to assume much higher release rates and
exposures.
Previously, it was argued that, due to the low
degree of percutaneous absorption of certain dyes,
external exposure to dyes may not pose a signi®cant
risk to the consumer. The results of our experi-
ments, however, clearly demonstrate the ability of
bacteria of the human skin to metabolise azo dyes to
aromatic amines which are known to easily pene-
trate the skin. In this study, just a small portion of
the azo dye present in the assay was split.
Presumably, the formation rate is limited by the
number of bacteria of the assay and their metabolic
capacity. Additional experiments including co-
loured textiles and lipophilic dyes are in progress
to elucidate the quantitative aspects in more detail.
It is worth mentioning that, basically, conclusions
drawn from in vitro studies to the risk in vivo are of
limited value. So far, the risk for the consumer from
wearing textiles dyed with such colorants cannot be
®nally assessed.
Table 3 Quanti®cation of OT after incubation of DB 14 with
cells of different bacterial strainsa
Strain/
Isolate
o-Tolidine
(mg/assay)
Name
St-0201
St-0411
Staphylococcus epidermidis
Staphylococcus aureus
Micrococcus luteus
Micrococcus roseus
Micrococcus varians
Staphylococcus epidermidis
Staphylococcus epidermidis
Staphylococcus aureus
Micrococcus spec.
Coagulase-negative
Staphylococcus spec.
Coagulase-negative
Micrococcus spec.
Coagulase-negative
n.d.
6.1
10.4
29.8
14.8
2.9
0.0
6.6
0.0
Mi-0702
Mi-0401
Mi-0503
DSM 1798
DSM 22044
DSM 1104
Isolate A
(yellow I)
Isolate B
(white)
Isolate C
(beige)
Isolate D
(orange)
Isolate E
(yellow II)
6.4
13.1b
0.4b
0.5a
0.6
Coagulase-negative
n.d.
Coagulase-negative
0.0
aData represent mean values of at least two independent
experiments; bgrown under inductive conditions
groups of bacteria inhabiting the human skin. The
quantity varies extremely depending on age and site
(66102 to 26106 per cm2).9,21
Acknowledgements
In our experiments using skin bacteria, cleavage
of the azo dye DB 14 to the corresponding aromatic
amine OT was proved. Following incubation using
standard conditions (24 h incubation period),
further metabolites were detected and identi®ed as
deaminated (DABP) and deaminated/hydroxylated
(DAHBP) derivatives of OT, respectively. An addi-
Studies were supported by grants from the For-
schungszentrum Karlsruhe given to Prof Dr W Baltes.
The support of Prof Dr Stahl and Dr Specht, the help
of W Koch with the initial experiments, and the
excellent technical assistance of D Wittke are grate-
fully acknowledged.
References
1 Levine WG. Metabolism of azo dyes: Implication
for detoxi®cation and activation. Drug Metab Rev
1991; 23: 253 ± 309.
2 Chung K-T, Stevens SE, Cerniglia CE. The
reduction of azo dyes by the intestinal micro¯ora.
Critical Rev Microbiol 1992; 18: 175 ± 190.
3 DFG. Maximum concentrations at the workplace
and biological tolerance values for working
materials. Report XXIV, Commission for the
investigation of health hazards of chemical
compounds in the work area. 1988. VCH,
Weinheim.
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014

Azo cleavage by human skin bacteria
T Platzek et al
559
4 Ministry of Health. Regulations on Speci®c
Commodities (BedarfsgegenstaÈndeverordnung).
Bundesgesetzblatt I 1998. pp. 6 ± 36.
5 Specht K, Platzek T. Mittel zum FaÈrben und
AusruÈ sten von Textilien ± Anmerkungen zu
gesundheitlichen und analytischen Aspekten.
Deutsche Lebensmittel-Rundschau 1995; 91: 352 ±
359.
16 Prival MJ et al. Mutagenicity of benzidine and
benzidine-congener dyes and selected monoazo
dyes in a modi®ed Salmonella assay. Mutat Res
1984; 136: 33 ± 47.
17 Chung K-T. The signi®cance of azo-reduction in
the mutagenesis and carcinogenesis of azo dyes.
Mutat Res 1983; 114: 269 ± 281.
18 Wilson JG. Experimental studies on the mechanism
of teratogenic action of trypan blue. J Chronic Dis
1959; 10: 111 ± 130.
6 WHO. Some aromatic azo compounds. IARC
Monograph 1975; 8: 267 ± 278.
7 BGVV. VorlaÈu®ge Methode zum Nachweis der
Verwendung verbotener Azofarbstoffe auf gefaÈrbten
textilen GegenstaÈnden. Bundesgesundheitsblatt
1966; 39: 78 ± 81.
8 Platzek T, Gi U-S, Lang C, Baltes W. Azo dyes are
split into aromatic amines by human skin bacteria
in vitro. Naunyn-Schmiedeberg Arch Pharmacol
1997; 356: (Suppl): R62.
9 Krieg NR, Holt JG. Bergey's Manual of Systematic
Bacteriology. 1984. Williams and Wilkins Co.,
Baltimore.
10 Marples MJ, Charles C. The Ecology of the
Human Skin. 1965. Thomas Publisher,
Spring®eld, Illinois.
11 Enquete-Kommission `Schutz des Menschen und
der Umwelt' des 12. Deutschen Bundestages. Die
Industriegesellschaft gestalten ± Perspektiven fuÈ r
einen nachhaltigen Umgang mit Stoff- und
MaterialstroÈ men. 1994. Economica, Bonn.
12 Platzek T. Wie groû ist die gesundheitliche
GefaÈhrdung durch Textilien wirklich? Melliand
Textilberichte 1996; 77: 774 ± 778.
19 Collier SW, Storm JE, Bronnaugh RL. Reduction of
azo dyes during in vitro percutaneous absorption.
Toxicol Appl Pharmacol 1993; 118: 73 ± 79.
20 Aldrich FD. Excretion of radioactivity from rats
and rabbits following cutaneous application of two
14C-labeled azo dyes. J Toxicol Environ Health
1986; 18: 347 ± 355.
21 Kloos WE, Schleifer KH. Isolation and
characterisation of staphylococci from human skin:
II. Description of four new species: Staphylococcus
warneri, Staphylococcus capitis, Staphylococcus
hominis, and Staphylococcus simulans. Int J Sys
Bacteriol 1975; 25: 62 ± 79.
22 Manning BW, Cerniglia CE, Federle TW.
Metabolism of the benzidine-based azo dye Direct
Black 38 by human intestinal mikrobiota. Appl
Environ Microbiol 1985; 1: 10 ± 15.
23 Cerniglia CE et al. Mutagenic activation of the
benzidine-based dye Direct Black 38 by human
intestinal micro¯ora. Mutat Res 1986; 175: 11 ± 16.
24 Anliker R, Steinle D. VerhuÈ tung von Risiken beim
Gebrauch und bei der Handhabung von
13 Ra®i F, Franklin W, Cerniglia CE. Azoreductase
activity of anaerobic bacteria isolated from human
intestinal micro¯ora. Appl Environ Microbiol 1990;
56: 2146 ± 2151.
14 Myslak ZW, Bolt HM. Occupational exposure to
azo dyes and risk of bladder cancer. Zbl
Arbeitsmed 1988; 38: 310 ± 321.
Farbmitteln. Textilveredlung 1990; 25: 42 ± 49.
25 ETAD (Ecological and Toxicological Association of
the Dyestuffs Manufacturing Industry). Final report
on extractability of dyestuffs from textiles. Project
A 4007 (1983).
26 ETAD (Ecological and Toxicological Association of
the Dyestuffs Manufacturing Industry).
15 Brown JP, Dietrich PS. Mutagenicity of selected
sulfonated azo dyes in the Salmonella/microsome
assay: use of aerobic and anaerobic activation
procedures. Mutat Res 1983; 116: 305 ± 315.
Extractability of dyestuffs from textiles over a
normal life time of use. Project G 1033 (1997).
Downloaded from het.sagepub.com at SYRACUSE UNIV LIBRARY on April 7, 2014