Allergenic activity of an air-oxidized ethoxylated surfactant.

Abstract

Ethoxylated surfactants are used in household and industrial cleaners, topical pharmaceuticals, cosmetics and laundry products. Polyethers, e.g. ethoxylated surfactants and polyethylene glycols, are oxidized by atmospheric oxygen (autoxidized) when stored and handled. We have previously shown that a chemically well-defined non-ionic surfactant, the ethoxylated alcohol penta-ethylene glycol mono-n-dodecyl ether (C12E5), forms a complex mixture of autoxidation products when exposed to air. Predictive testing in guinea pigs showed that the surfactant itself is a non-sensitizer, but that oxidation products formed are skin sensitizers. The aim of this study was to investigate the sensitizing capacity of a total oxidation mixture of C12E5 obtained after autoxidation. The allergenic activity of different oxidation products is discussed as well as the clinical importance of the findings. This study shows that the non-ionic surfactant C12E5 containing 20% oxidation products is a sensitizing mixture. The result accords with what is observed for other compounds that are unstable when in contact with air, e.g. limonene and linalool, major fragrance terpenes. Studies regarding the clinical relevance of our findings should be performed. However, it is already clear from this study that precautions must be taken in handling and storage of ethoxylated surfactants to avoid formation of allergenic mixtures.

Full text

Page 1

Allergenic activity of an air-oxidized ethoxylated
surfactant
ANN-THERESE KARLBERG1,2, ANNA BODIN1,2AND MIHALY MATURA1,3
1Dermatochemistry and Skin Allergy, Department of Chemistry, Go ¨ teborg University, Go ¨ teborg,2National Institute
for Working Life, Stockholm, and3Department of Dermatology, Sahlgrenska University Hospital, Go ¨ teborg, Sweden
Ethoxylated surfactants are used in household and industrial cleaners, topical pharmaceuticals,
cosmetics and laundry products. Polyethers, e.g. ethoxylated surfactants and polyethylene glycols,
are oxidized by atmospheric oxygen (autoxidized) when stored and handled. We have previously
shown that a chemically well-defined non-ionic surfactant, the ethoxylated alcohol penta-ethylene
glycol mono-n-dodecyl ether (C12E5), forms a complex mixture of autoxidation products when
exposed to air. Predictive testing in guinea pigs showed that the surfactant itself is a non-
sensitizer, but that oxidation products formed are skin sensitizers. The aim of this study was to
investigate the sensitizing capacity of a total oxidation mixture of C12E5obtained after autoxidation.
The allergenic activity of different oxidation products is discussed as well as the clinical importance
of the findings. This study shows that the non-ionic surfactant C12E5containing 20% oxidation
products is a sensitizing mixture. The result accords with what is observed for other compounds that
are unstable when in contact with air, e.g. limonene and linalool, major fragrance terpenes. Studies
regarding the clinical relevance of our findings should be performed. However, it is already clear
from this study that precautions must be taken in handling and storage of ethoxylated surfactants to
avoid formation of allergenic mixtures.
Key words: alcohol ethoxylate; allergic contact dermatitis; autoxidation; ethoxylated surfactant;
guinea pigs; hydroperoxides; sensitization experiments; skin sensitization.#Blackwell Munksgaard,
2003.
Accepted for publication 27 November 2003
Ethoxylated surfactants are widely used in house-
hold and industrial cleaners, topical pharmaceut-
icals, cosmetics and laundry products. These
surface-active agents are emulsifiers as well as sus-
pending, wetting and solubilizing agents that give
final products physical stability (1). Surfactants
also affect the permeability of the skin, which
makes them suitable as penetration enhancers in
transdermal drug delivery systems (2). Non-ionic
surfactants are often preferred to ionic surfactants
in topical products, because they are considered to
cause less skin irritation (3).
It is well known that polyethers, e.g. ethoxyl-
ated surfactants and polyethylene glycols, are
easily oxidized by atmospheric oxygen (autoxi-
dized) (4, 5). We have previously shown that a
chemicallywell-defined
penta-ethyleneglycol
(referred to as C12E5) (Fig.1), forms a complex
mixture of autoxidation products when exposed
to air (6–10). Predictive testing in guinea pigs
ethoxylated
mono-n-dodecyl
alcohol,
ether
showed that the pure non-oxidized surfactant
itself is not a sensitizer, but that the majority of
investigated oxidation products are skin sensi-
tizers (Fig.1) (6–10).
The aim of the present study was to investigate
the sensitizing capacity of a mixture of the non-
ionic ethoxylated surfactant C12E5and its oxida-
tion products obtained after autoxidation. A
guinea-pig method was used for evaluation. The
allergenic activity of the different oxidation pro-
ducts is discussed as well as the clinical import-
ance of the findings.
Materials and Methods
Chemicals
3,6,9,12,15-Pentaoxaheptacosan-1-ol
ethylene glycol mono-n-dodecyl ether, C12E5,
98% pure) (Fig.1) was purchased from Nikkol
Chemicals Co., Ltd. (Tokyo, Japan) and, after
analysis, used without further purification. The
(penta-
Contact Dermatitis 2003: 49: 241–247
Printed in Denmark. All rights reserved
Copyright#Blackwell Munksgaard 2003
CONTACT DERMATITIS
ISSN 0105-1873

Page 2

identity of the substance was checked using high
performance liquid chromatography-mass spec-
trometry (HPLC-MS) analysis and nuclear mag-
netic resonance (NMR) spectral characterization.
Dodecyltetraoxyethyleneoxy acetaldehyde (C12E4
aldehyde) (Fig.1), the major stable oxidation
product and sensitizer previously identified in
oxidized C12E5(7), was synthesized according to
prior experience (7).
Air oxidation of C12E5
A sample of neat C12E5(5g) used for sensitization
studies was kept for 10 months in an open
Erlenmeyerflask indaylight
sunlight, at room temperature without stirring.
The top of the flask was covered with aluminium
foiltoprevent contamination
evaporation. After such storage, the degree of oxi-
dation after 10 months, measured as per cent
remaining starting material, was determined to
be 80% using gas chromatography according to
previous experience (8). A peroxide value of 334mE
was determined according to the method in the
European Pharmacopoeia (11). The peroxide value
isthenumberthatexpresses,inmEofactiveoxygen,
the quantity of peroxide contained in 1000g of the
substance, as determined with iodine titration.
withoutdirect
and reduce
Study of the sensitizing capacity of autoxidized
C12E5
The sensitization study was performed using
female Dunkin–Hartley guinea pigs (weight,
350–470g) from AB Sahlins Fo ¨ rso ¨ ksdjursfarm,
Malmo ¨ , Sweden, that were randomly assigned to
exposed groups A (n¼15) and B (n¼15) and 2
control groups (n¼15þ15). Group A and 1 con-
trol group were induced according to a previously
published protocol (12) of the cumulative contact
enhancement test (CCET) (13). The guinea pigs
received an occlusive epidermal application on
their shaved upper back on days (D) 0, 2, 7 and
9 as induction. The exposed group was induced
with the autoxidized C12E5(20% w/w) in distilled
water, and the control group was induced with
just distilled water. At each of the 4 induction
occasions a filter paper (4?2cm), wetted with
about 200mg of the induction material, was
attached to the skin. All animals were injected
with Freund’s complete adjuvant (FCA) (Difco,
Detroit, MI, USA) (2?0.1ml), intradermally on
the upper back, on D7. Group B and the other
control group were induced according to the same
protocol but without injection of FCA. Challenge
testing was performed in all groups on D21 on the
shaved left flank using small Finn Chambers
C12H25
O
CH2
CH2
O
4
CH2
CH
OH
OOH
C12H25
O
CH2
CH2
O
4
CH2
CH2
OH
C12H25
O
CH2
CH2
O
n = 0 – 4
CH2
C
H
O
C12H25
O
CH2
CH2
O
C
H
O
H
O
CH2
CH2
O
n = 0 – 4
CH2
C
H
O
H
C
H
O
C12H25
O
CH2
CH2
O
CH2
C
H
O
H
O
CH2
CH2
O
CH2
C
H
O
O
CH2
CH2
OH
5
CH
C11H23
OOH
C12H25
O
CH2
CH2
O
C
H
O
n = 0 – 4
Alkyl polyethyleneglycol aldehydes
Formaldehydeb
Penta-ethylene glycol mono-n-dodecyl ether (C12E5)a
Alkyl polyethyleneglycol formates
Hydroxyaldehydes
4
Dodecyltetraoxyethyleneoxy acetaldehydeb
(C12E4 aldehyde)
3
Triethyleneoxy acetaldehydeb
16-Hydroperoxy-pentaoxaheptacosan-1-olb
1-Hydroperoxy-pentaoxaheptacosan-1-ol
2 - (Dodecyloxy)ethyl formatea
Fig.1. Oxidation products identified in C12E5after air exposure. Sensitization studies in guinea pigs according to a modified
cumulative contact enhancement test (CCET) method (12, 13) have been performed with major compounds from the different
groups of oxidation products.ano skin sensitizer;bskin sensitizer.
242 KARLBERG ET AL.

Page 3

(8mm i.d., Epitest, Helsinki, Finland) (12). The
animals were tested with the autoxidized C12E5at
the concentrations 1.0, 3.0 and 9.0% w/w in dis-
tilled water and with distilled water. 15mg of test
solution was applied in each Finn Chamber. After
24h the chambers were removed. Visual evalu-
ation of the reactions was performed at 48h and
72h after test application. The minimum criterion
for a positive reaction was a confluent erythema.
A second challenge was performed on the right
flank on D35. The animals were tested with aut-
oxidized C12E5(9.0% w/w) in distilled water and
with distilled water.
A third challenge was performed on the left flank
on D57. The animals were tested with autoxidized
C12E5(9.0%w/w),withdodecyltetraoxyethyleneoxy
acetaldehyde (C12E4aldehyde) (130mM, 5.0% w/w)
in distilled water and with distilled water.
The study was approved by the local ethics
committee in Stockholm.
Statistical analysis
The results of the sensitization study were statis-
tically evaluated using Fisher’s exact test. The
number of positive reactions in the exposed
group was compared with the number of reac-
tions in the control group. A P value of <0.05
was considered statistically significant.
Results
Sensitizing capacity of autoxidized C12E5
First challenge. Group A showed a significant
response to the highest concentration (9.0%) of
the oxidized C12E5 at the 72h scoring (8/15
exposed versus 0/15 controls)
lower concentrations gave no significant response
(Table1). In group B, a significant response was
obtained only to oxidized C12E5 3.0% at 48h
scoring (6/15 exposed versus 1/15 controls). No
significant response was obtained to the highest
concentration (9.0%) due to irritation, nor to the
lowest concentration (1.0%) due to few reactions
in the exposed group (Table1).
(Table1). The
Second challenge. Group A showed a significant
response to oxidized C12E59.0% at both scorings
(48h 7/15 exposed versus 1/15 controls; 72h 9/15
exposed versus 1/15 controls) (Table1). Group B
also showed a significant response to C12E59.0%
at both scorings (48h 9/15 exposed versus 2/15
controls; 72h 8/15 exposed versus 0/15 controls)
(Table1).
Third challenge. Group A showed a significant
response to oxidized C12E59.0% at the 48h scor-
ing (6/15 exposed versus 1/15 controls) and also
to
(C12E4aldehyde) (9/15 exposed versus 3/15 con-
trols at both scorings) (Table1). Group B gave a
significant response at the 72h scoring to both
C12E59.0% (9/15 exposed versus 2/15 controls)
and to C12E4aldehyde (11/15 exposed versus 5/15
controls) (Table1). More reactions were observed
in the control groups at the third challenge testing
compared to the previous testings (Table1).
dodecyltetraoxyethyleneoxyacetaldehyde
Discussion
This study shows that the ethoxylated non-ionic
surfactant pentaethylene glycol mono-n-dodecyl
ether C12E5, containing 20% oxidation products,
is a sensitizing mixture. In previous studies pure
(hy) (C12E5) was shown to be a non-sensitizer (6).
The present result is in accordance with what is
observed also for other compounds, e.g. the
major fragrance terpenes limonene and linalool
that are unstable when in contact with air. Due to
autoxidation these compounds form allergenic oxi-
dation mixtures on handling and storage at normal
room temperature (14–16). Patch testing in con-
secutive dermatitis patients has shown that the oxi-
dation mixtures of these compounds cause contact
allergyin thepopulation(17–19)thatisnot detected
when testing with the pure compounds (18).
In the present study, we performed 2 parallel
experiments to investigate the sensitizing cap-
acity of a chemically well-defined ethoxylated
alcohol, C12E5, after autoxidation at air exposure.
1 experiment (Group A and its control group) was
performed according to the ordinary predictive
guinea-pig method CCET (13), with epidermal
exposure combined with intradermal injection
of FCA in addition to the third induction.
In the parallel experiment (Group B and its
control group), no FCA was injected but only
epidermal exposure of oxidized C12E5was used
for induction. Overall, a similar result was
obtained in the 2 experiments (Table1). Interest-
ingly, more irritation was observed at first
challenge in the control group without FCA
treatment, compared to the control group treated
with FCA, resulting in no significant response to
the highest challenge concentration of oxidized
C12E5in the non-FCA-treated animals. At second
challenge a comparable response was obtained in
both experiments. This might be due to the fact
that C12E5was tested in only 1 concentration,
which reduces the risk of irritancy. At third chal-
lenge, a response was obtained to the oxidized
C12E5in both experiments and also to the C12E4
aldehyde, which is the aldehyde identified in the
highest amounts in the oxidation mixture of
oxidized C12E5 and previously shown to be a
ALLERGENICITY OF AN ETHOXYLATED ALCOHOL243

Page 4

Table1. Results from sensitization studies on autoxidized C12E5in guinea pigs using a modified cumulative contact enhancement test (CCET) protocol (12, 13)
Number of animals with positive reactions Group Aa(n¼15)
Controls to Group A (n¼15)
Group Bh(n¼15)
Controls to Group B (n¼15)
Challenge
material
First
challengeb
Second
challengec
Third
challenged
First
challenge
Second
challenge
Third
challenge
First
challengeb
Second
challengec
Third
challenged
First
challenge
Second
challenge
Third
challenge
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
48h
72h
Oxidized C12E5(9.0%)
5
8e
7f
9g
6f
6
1
0
1
1
1
2
7
7
9g
8g
9
9g
3
4
2
0
4
2
Oxidized C12E5(3.0%)
2
3
1
0
6f
5
1
2
Oxidized C12E5(1.0%)
2
3
0
0
2
1
1
0
C12E4aldehyde (5.0%)
9f
9f
3
3
11
11f
8
5
Water
0
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
1
2
0
0
0
0
aInduction: 20% w/w autoxidized C12E5in distilled water;bD21;cD35;dD57;eSignificantly different from controls P<0.001;fSignificantly different from controls, P<0.05;gSignificantly
different from controls, P<0.01;hInduction: 20% w/w autoxidized C12E5in distilled water without FCA injections.
244KARLBERG ET AL.

Page 5

sensitizer (7). At the third challenge testing, many
reactions were observed in both control groups,
which might be due to irritation because this
re-testing was performed on the same flank as the
one used at the first challenge testing. It might
also be due to sensitization of the controls due to
repeated exposure. The fact that the CCET
method without FCA application resulted in suc-
cessful sensitization further strengthens the latter
possibility. Also in the sensitization studies on
the C12E4aldehyde when 2 parallel CCET experi-
ments with and without FCA were performed, a
significant response was obtained in both experi-
ments (7). However, in that study more irritant
reactions were observed in the FCA-treated con-
trol group compared to the control group without
FCA treatment (7). The present results, together
with those previously obtained in the study on the
aldehyde (7), show that pretreatment with FCA is
not necessary for these compounds. The results
also illustrate the difficulty in testing surfactants
with irritant properties in a manner that shows
their allergenic activity only.
With the exception of molecules known to con-
tain labile groups, most non-ionic surfactants
composed of saturated hydrocarbon- and polyoxy-
ethylene-chains have previously been considered
to be relatively stable under conditions of pre-
paration, storage and use. However, as in the
case of simple ethers, polyether chains readily
undergo autoxidation (4), leading to profound
changes in physical properties and behaviour of
the surfactant (20, 21). The oxygen atom present
in ethers facilitates autoxidation at room tem-
perature, activation being some 20 times that
caused by a methyl group (22). Considering the
entire time period from production of the surfac-
tant to the final consumption of a formulated
product, the storage time used in this study (10
months in daylight at room temperature without
stirring) corresponds well to normal handling and
storage. During some periods antioxidants may be
present in the formulated product, but when these
are consumed the autoxidation will start again as
we have shown for other compounds (23).
Autoxidation of polyoxyethylene chains results
in a complex mixture of products. According to the
proposed mechanism the initial step is a radical-
mediated formation of hydroperoxides (10).
These primary oxidation products can subse-
quently undergo various reactions forming sec-
ondary oxidation products, such as esters,
aldehydes and alcohols, depending on the condi-
tions (10). Usually, such autoxidation products are
not identified on a molecular basis but just identi-
fied on a group basis. In our studies on C12E5,
we have identified specific primary oxidation
products, i.e. hydroperoxides (10), as well as
specific secondary oxidation products in the oxi-
dation mixture obtained after air exposure (6–9).
The alkylpolyethyleneglycol aldehydes (Fig.1)
constitute an important group among the oxida-
tion products identified, and the C12E4aldehyde
(Fig.1) was detected in an amount of about 1% in
samples of oxidized C12E5containing 70% of the
starting material (7, 8). Its sensitizing potential,
using the CCET method for predictive testing,
was shown to be equal to that of formaldehyde.
Related aldehydes with shorter ethyleneoxy chains
are present in smaller amounts in the oxidation
mixture. Significant cross-reactivity between the
aldehyde investigated and the next shorter homo-
logue (1 ethyleneoxy group less) was demonstrated
(7). The present study shows that the amount of
alkylpolyethyleneglycol aldehydes in the oxidized
C12E5was enough to cause elicitation.
The C12E4aldehyde originates from the very
quickly degraded 1-hydroperoxy-pentaoxahepta-
cosan-1-ol (Fig.1), with the hydroperoxy-group
positioned next to the primary alcohol at the
end of the ethyleneoxy chain (10). Due to its
instability we have not been able to isolate or
synthesize this hydroperoxide. However, of the
potential hydroperoxides that can be formed from
polyethyleneoxycompounds,
pentaheptacosan-1-ol with the hydroperoxy-group
sited at the a-carbon of the alkyl chain (Fig.1) is
likely to be the most stable. Different amounts
(0.7–5.7%) of this hydroperoxide were identified
in oxidation mixtures of C12E5oxidized under the
same conditions and containing 55% of the
original compound (10). This shows that the oxida-
tion mixtures can vary in composition even though
the amount of the original compound remaining in
the mixture is the same. The actual hydroperoxide
was synthesized and it was found to be a sensitizer
(10). It was not tested in the present study due to
shortage of material. However, in previous studies,
we have shown that hydroperoxides are the main
sensitizers in colophonium (24, 25) as well as in
oxidation mixtures of limonene (15, 17, 18) and
linalool (16, 19).
Alkyl polyethyleneglycol formates (Fig.1) were
also identified in the oxidation mixture of C12E5
(8). The sensitizing capacity of 2-(dodecyloxy)ethyl
formate was investigated and it was found to be a
non-sensitizer, which was not surprising because
esters rarely are sensitizers (26).
Of other compounds, the hydroxyaldehydes
(Fig.1) identified in the oxidation mixture are
chemically characterized as aldehydes of PEGs
(HO(CH2CH2)nOH). The sensitizing capacity of
triethyleneoxy acetaldehyde (Fig.1) was deter-
mined (9). It was found to be a weak allergen
16-hydroperoxy-
ALLERGENICITY OF AN ETHOXYLATED ALCOHOL 245