ARTICLE
Acute irritant contact dermatitis (ICD) is a heterogeneous disorder,
with many differing cellular mechanisms contributing to its pathogenesis
[1]. One of the more recently identified of these is oxidative stress,
a situation in which an imbalance in pro-oxidant/anti-oxidant equilibrium
occurs within cells and tissues, resulting in damage to lipids, proteins,
carbohydrates and DNA [2, 3]. Such an imbalance may result from excessive
endogenous generation of free radicals during physiologic processes, but
may also occur through exposure to exogenous sources of free radicals,
such as tobacco smoke and ultraviolet radiation [4, 5]. Chemical irritants
may also generate reactive oxygen species, both directly and indirectly,
a well known example from a dermatological perspective being dithranol,
which rapidly auto-oxidizes in aqueous solution to produce singlet oxygen
and superoxide anion radicals [6, 7].
By way of defence against dangerously reactive species, a plethora of
antioxidant enzymes and systems have evolved across the spectrum of microorganisms,
plants and animals [8-10]. Recently, we applied quantitative image analysis
techniques in the immunocytochemical study of one of these enzymes, Cu,Zn-superoxide
dismutase (Cu, Zn-SOD), in man, and were able to demonstrate indirectly
the presence of oxidative stress at sites of experimentally induced acute
ICD [11]. Levels of Cu,Zn-SOD were reduced following topical exposure
to chemical irritants, including not only dithranol, but also sodium lauryl
sulphate (SLS), an anionic detergent not normally associated with the
direct generation of free radicals. As an extension to our studies of
oxidative stress in ICD, we have now focussed on another of the protective
enzyme systems present in many organisms, including man, namely the glutathione
S-transferases (GST). These are a widely distributed family of multifunctional
enzymes which catalyze the reaction between reduced glutathione and a
variety of exogenously and endogenously derived electrophilic compounds
[12]. Although there is some evidence of altered GST expression in animal
models of skin inflammation [13, 14], relatively little is known about
their activity in inflammatory responses in man, particularly that evoked
by chemical irritants. Since the pi class of GST has been reported to
predominate in normal human skin, with the alpha, but not the mu, isoenzyme
also being detectable [15], antibodies against GST pi and alpha were selected
and utilized in this quantitative immunocytochemical study of ICD.
Materials and methods
Subjects
Healthy, non-atopic, male subjects (n = 18; age range 18- 58 years,
mean age 32 years), with no past or present history of skin disease, took
part in the study. Approval was obtained from the Wycombe Local Research
Ethics Committee and all volunteers gave written, informed consent.
Patch testing and biopsy
Nine of the volunteers were patch tested with 0.2% (w/w) dithranol in
white soft paraffin (wsp), containing 0.25% (w/w) salicylic acid. The
remaining nine were patch tested with 5% aqueous (w/v) SLS (purity >
99%, Sigma Chemical Co., Poole, Dorset, UK). Four 8 mm Finn Chambers (Epitest
Ltd.) Oy, Rannankoukku, Tuusula, Finland) were applied in total, two on
the mid-volar region of each forearm. One chamber on each arm was filled
with the irritant (25 mg for dithranol/15 mul for SLS), the other with
a similar quantity of the appropriate vehicle control. The patches remained
in contact with the skin for either 5 hrs or 47 hrs, depending upon the
biopsy time. Immediately prior to biopsy, the intensity of the inflammatory
reactions was visually graded for erythema, according to the following
scale: 0, no visible reaction; 0.5, faint, patchy erythema; 1, weak erythema;
2, moderate erythema; 3, marked erythema; 4, intense erythema.
Each subject had five elliptical biopsies (4 mm diameter) removed, following
injection of lignocaine. Two biopsies were taken from the irritant patch
test sites, two from the vehicle control sites, and one from an area of
untreated, normal skin adjacent to the patch test sites on one arm. Two
time periods were selected for each subject from the following three sampling
times: 6 hrs (chamber application time, 5 hrs), 48 hrs or 96 hrs (chamber
application time for both, 47 hrs). Vehicle control sites were biopsied
at the same time as the irritant test sites, with the normal skin sample
being removed during the first biopsy session. Biopsies were immediately
embedded in OCT compound, and snap frozen and stored in liquid nitrogen.
A total of five to seven samples per irritant and control time point was
ultimately obtained.
Immunocytochemistry and
image analysis
A detailed account of the immunostaining and quantitative image analysis
methodologies employed is given in a previous publication [11]. Briefly,
4 mum sections were cut from three different areas of each biopsy and
dried overnight on Vectabond-subbed (Vector Laboratories, Peterborough,
UK) slides. Following fixation in acetone for 10 min, sections were incubated
for 30 min in the following rabbit polyclonal antibodies: anti-GST alpha,
1:100 and anti-GST pi, 1:100 (Novocastra Laboratories Ltd, Newcastle upon
Tyne, UK). Negative controls, using an irrelevant rabbit polyclonal antibody
were included. The Vectastain ABC Elite peroxidase kit (Vector
Laboratories), with 3',3 diaminobenzidine as chromogen, was used to visualize
the antibody/antigen reaction. All incubations were conducted at room
temperature. No counterstaining was performed. For consistency, test and
control samples were prepared and immunostained in parallel throughout.
Microscopy and analysis were performed blind, using a Zeiss Axioplan microscope
linked to a Power Macintosh 8100 computer, loaded with Optilab Pro 2.6
and GEMStain software (ME Electronics, Reading, UK), the latter permitting
measurements of the total quantity of stain present on an area basis (expressed
in terms of the total number of grey levels/mum2 epidermis)
to be made.
Routine haematoxylin and eosin staining was also performed on representative
sections from each biopsy.
Statistics
The mean staining density (+ SD) for each of the sample groups was calculated.
Irritant and vehicle control values were compared using the Wilcoxon matched
pairs signed ranks test. A significance level of p < 0.05 was applied.
Results
Intensity of patch test reactions
Visual assessments of the intensity of response for the irritant groups
are given in Table I.
Of the controls, water produced slight reactions (0.5) in five individuals
after 48 hrs only, whilst white soft paraffin caused a mild reaction of
0.5 in one individual alone, again after 48 hrs.
Histopathology
SLS induced small areas of mild spongiosis after 6 hrs, which became
more overt after 48 hrs, with marked parakeratosis also being present.
By 96 hrs, parakeratosis, spongiosis and, in some cases, acanthosis, were
apparent.
Dithranol gave little histopathological change after 6 hrs. At the two
later time periods, spongiosis and some swelling of upper keratinocytes
were seen, with one severe reaction showing marked cellular damage.
Little or no evidence of pathological change was present in the vehicle
controls.
Distribution of GST labelling
In normal skin samples, GST alpha was cytoplasmically located and distributed
throughout the epidermis, with a slightly greater intensity of staining
being present in the stratum granulosum. GST pi expression was similarly
evident within the cytoplasm at all levels of the epidermis, but with
this isoenzyme a greater density of staining was seen in the basal layers.
Positive immunolabelling for both isoenzymes was present in the sebaceous
glands and outer root sheath of hair follicles. Endothelial cells, fibroblasts
and scattered mononuclear cells within the dermis were also positively
stained.
There were no discernable changes in the density or distribution of
staining of either GST alpha or GST pi after 6 hrs exposure to the two
irritants. However, at the 48 hrs and 96 hrs time-points, the intensity
of GST alpha staining was considerably reduced within the epidermis of
both dithranol (Fig. 1)
and SLS treated sites (Fig. 2).
A reduction in labelling density, although to a lesser degree, was also
evident for GST pi in some of these samples (Figs.
3, 4). The inflammatory cell infiltrate present in most SLS and
dithranol biopsies at the two later time periods showed consistently strong
labelling for GST alpha and pi throughout.
Vehicle control samples closely resembled normal skin biopsies, both
in terms of intensity and distribution of staining for GST alpha and pi.
Quantification of GST labelling
Neither of the two irritants induced changes in the levels of the two
enzymes after 6 hrs. However, significant decreases in epidermal GST alpha
density were detected after 48 hrs and 96 hrs in both the dithranol (p
< 0.005 and p < 0.05, respectively) (Fig.
5) and SLS (p < 0.005 and p < 0.01, respectively) (Fig.
6) treated skin biopsies. Reductions in GST pi were less marked,
occurring in the 96 hrs samples only of dithranol patch tests (p <
0.05) (Fig. 7), and the
48 hrs samples only of SLS exposed sites (p < 0.05) (Fig.
8).
Vehicle controls did not differ significantly from normal skin samples
at any time point.
Discussion
This quantitative immunocytochemical study has demonstrated that the
levels of GST alpha in the epidermis decrease significantly in acute ICD,
with epidermal GST pi levels also reducing, although to a lesser extent.
Such a finding is highly suggestive of the presence of oxidative stress
in the inflammatory lesions induced by chemical irritants, and extends
our earlier findings of reduced levels of another anti-oxidant enzyme,
Cu,Zn-SOD, in this same biopsy material [11]. As before, the changes were
not confined to the irritant, dithranol, which is a known generator of
reactive oxygen species, but also extended to the anionic detergent, SLS,
which is not generally associated with direct free radical production.
As stated above, little is known about the behaviour of GST in acute
ICD in man. Depressed levels have been demonstrated, however, in peripheral
blood lymphocytes collected from patients with chronic irritant hand dermatitis
[16] and there is evidence from rodent experimental models that cutaneous
inflammation, in general, may be associated with changes in the levels
of GST isoenzymes in the skin. Raza et al. [13] described a reduction
in GST activity of some 30-40% following the topical application of gasoline
in mice, whilst cutaneous inflammation subsequent to mechanical wounding
was, likewise, found to cause a significant reduction in GST, accompanied
by decreases in other anti-oxidant enzymes, including Cu,Zn-SOD and glutathione
peroxidase [14].
GSTs are a complex family of multifunctional enzymes which catalyze
the conjugation of reduced glutathione with a variety of electrophilic
compounds. The different classes of GST which exist in man and other organisms
exhibit different catalytic properties, suggesting separate or complimentary
activities [12]. Decreased levels of GST, as seen in the present study,
point to the formation, during oxidative metabolism, of such reactive
products as alkenes, epoxide derivatives and organic hydroperoxides [17],
the latter being a particularly favourable substrate for GST alpha, the
class which showed the greatest change in level with both dithranol and
SLS.
During the course of the cellular response to irritants, reactive oxygen
species (ROS) may be produced in excess through a variety of means. Dithranol,
as mentioned earlier, is a somewhat atypical irritant in that it provides
an exogenous source of oxidants by directly generating singlet oxygen
and superoxide anion radicals when in aqueous solution [6, 7]. For most
irritants, including SLS, the major sources of free radicals will be of
endogenous origin. Principal amongst these will almost certainly be infiltrating
neutrophils and macrophages, which generate and release an array of ROS
into the extracellular milieu [5]. They are a feature of virtually all
irritant patch test sites [18] and are particularly prominent at the later
time periods, coinciding with our observed changes to GST levels. Of possible
significance also, is the association between pro-inflammatory cytokines
and oxidative stress. Oxidative stress is known to induce the up-regulation
of cytokines such as IL-1, IL-8 and TNF-alpha [19, 20], but, paradoxically,
increased levels of TNF-alpha, a cytokine known to be produced in ICD
[21], can themselves induce oxidative stress [22].
In interpreting the data derived from this study, it is also important
to consider the fact that GSTs are not only involved in the detoxification
of products of oxidative metabolism, but, amongst other activities, also
play a role in the biotransformation of leukotriene A4 to leukotriene
C4 [23]. Leukotrienes may act as inflammatory mediators in
the response to at least some irritants [24] and the pathways involved
in their metabolism may therefore have an influence on the levels of GST
present in inflamed skin.
CONCLUSION
In conclusion, our findings add to the body of evidence that GSTs play
a role in cutaneous inflammation, and further support the hypothesis that
oxidative stress is one of the mechanisms which contributes to the pathogenesis
of ICD.
Acknowledgements
The authors would like to thank all the subjects who participated in
the study and the expert nursing assistance given by Sally Barth and Maria
Nicholson. Financial support was given by the British Occupational Health
Research Foundation and the Erasmus Wilson Dermatological Research Fund.
REFERENCES
1. Basketter DA, Gerberick GF, Kimber I, Willis CM. Contact irritation
mechanisms. In: Toxicology of Contact Dermatitis - Allergy, Irritancy
and Urticaria. John Wiley & Sons, Ltd., Chichester, 1999: 11-38.
2. Halliwell B. Free radicals, reactive oxygen species and human
disease. Br J Exp Path 1989; 70: 737-57.
3. Darr D, Fridovich I. Free radicals in cutaneous biology. J
Invest Dermatol 1994; 102: 671-5.
4. Fridovitch I. The biology of superoxide and superoxide dismutases.
In: Hayaishi D, Imamura S, Miyachi Y, eds. The Biological Role of Reactive
Oxygen Species in Skin. Elsevier, New York, 1987: 9.
5. Athar M, Agarwal R, Bickers DR, Mukhtar H. Role of reactive
oxygen species in skin. In: Mukhtar H, ed. Pharmacology of the Skin.
CRC Press, Boca Raton, Florida, 1992: 270.
6. Muller K, Wiegrebe W, Younes M. Formation of active oxygen
species by dithranol III. Dithranol, active oxygen species and lipid peroxidation
in vivo. Arch Pharm 1987; 320: 59-66.
7. Fuchs J, Packer L. Investigations of anthralin free radicals
in model systems and in skin of hairless mice. J Invest Dermatol
1989; 92: 677-82.
8. Shindo Y, Witt E, Han D, Epstein W, Packer L. Enzymic and
non-enzymic antioxidants in epidermis and dermis of human skin. J Invest
Dermatol 1994; 102: 122-4.
9. Applegate LA, Frenk E. Cellular defense mechanisms of the
skin against oxidant stress and in particular UVA radiation. Eur J
Dermatol 1995; 5: 97-103.
10. Fuchs J, Freisleben HJ, Packer L. Antioxidants in the skin.
In: Mukhtar H, ed. Pharmacology of the Skin. CRC Press, Boca Raton,
Florida, 1992: 250.
11. Willis CM, Reiche L, Wilkinson JD. Immunocytochemical demonstration
of reduced Cu, Zn-superoxide dismutase levels following topical application
of dithranol and sodium lauryl sulphate: an indication of the role of
oxidative stress in acute irritant contact dermatitis. Eur J Dermatol
1998; 1: 8-12.
12. Raza H, Agarwal R, Mukhtar H. Cutaneous glutathione S-transferases.
In: Mukhtar H, ed. Pharmacology of the Skin. CRC Press, Boca Raton,
Florida, 1992: 131-7.
13. Raza H, Qureshi MM, Montague W. Alteration of glutathione,
glutathione S-transferase and lipid peroxidation in mouse skin and extracutaneous
tissues after topical application of gasoline. Int J Biochem Cell Biol
1995; 27: 271-7.
14. Shukla A, Mohan Rasik A, Kumar Patnaik G. Depletion of reduced
glutathione, ascorbic acid, vitamin E and antioxidant defence enzymes
in a healing cutaneous wound. Free Rad Res 1997; 26: 93-101.
15. Raza H, Awasthi YC, Zaim MT, Eckert RL, Mukhtar H. Glutathione
S-transferases in human and rodent skin: multiple forms and species-specific
expression. J Invest Dermatol 1991; 96: 463-7.
16. Singh N, Hamann K, Clausen J. Decreased lymphocyte aryl hydrocarbon
hydrolase and glutathione S-transferase activites in patients with hand
dermatitis. Acta Derm (Stockh) 1981; 62: 61-3.
17. Mannervik B, Danielson UH. Glutathione transferases -
structure and catalytic activity. CRC Crit Rev Biochem 1988; 23:
283-337.
18. Willis CM, Stephens CJM, Wilkinson JD. Differential patterns
of epidermal leucocyte infiltration in patch test reactions to structurally-unrelated
irritants. J Invest Dermatol 1993; 101: 364-70.
19. Janssen YM, Van Houten B, Borm PJ, Mossman BT. Cell and tissue
responses to oxidative stress. Lab Invest 1993; 69: 261-74.
20. Verhasselt V, Goldman M, Willems F. Oxidative stress up-regulates
IL-8 and TNF-alpha synthesis in human dendritic cells. Eur J Immunol
1998; 28: 3886-90.
21. Enk AH, Katz SI. Early molecular events in the induction
phase of contact sensitivity. Proc Nat Acad Sci USA 1992; 89: 1398-402.
22. Adamson GM, Billings RE. Tumour necrosis factor induced oxidative
stress in isolated mouse hepatocytes. Arch Biochem Biophys 1992;
294: 223-9.
23. Mukhtar H, Raza H, Allyn DR, Bickers DR. Glutathione S-transferase
catalyzed metabolism of leukotriene A4 to leukotriene C4
in rodent and human skin. J Invest Dermatol 1990; 94: 557S.
24. Dickson FM, Lawrence JN, Benford DJ. Release of inflammatory
mediators in human keratinocyte cultures following exposure to a skin
irritant. Toxic in Vitro 1993; 7: 385-8.
|
Printable version
