Allergen penetration through the skin

ARTICLE

Auteur(s) : Frédéric BERARD¹, Jean-Paul MARTY², Jean-François NICOLAS^1,3

¹ INSERM U 503, 69365 Lyon Cedex 07, France.
² Laboratoire de Dermopharmacologie et Cosmétologie, Faculté de Pharmacie, Université Paris Sud, 92296 Chatenay-Malabry, France.
³ Immunologie Clinique et Allergologie, CHU Lyon Sud, 69495 Pierre-Benite Cedex, France

Article accepted on 21/1/2003

In view of the large surface area of the skin and its direct accessibility to all the molecules present in the air as well as those in direct contact with it, the skin is quantitively the most important organ as regards the external environment. This “protective barrier” is not complete; the skin is permeable to practically all substances, only the degree of permeability varies. This principally depends on the physiological state of the skin, the physical/chemical properties of the substance whose entry it is intended to restrict and the nature of the environment’ (excipient, medium...) through which these substances are brought into contact with the skin. The human epidermis is particularly well adapted to its essential function as a barrier. It limits the entry of external agents and participates in the general homeostasis of the organism by regulating the trans-epidermal loss of water. It is the corneal layer which plays the biggest part in the barrier function of the epidermis [31, 38].
The penetration of allergens via the skin is well established for small, non-protein chemicals, and for haptens, which are responsible for delayed contact hypersensitivity in sensitized subjects (contact eczema or syn. allergic contact dermatitis). For a long time the skin was considered to be impermeable to large molecules (of a high molecular weight) and thus impermeable to allergenic proteins such as those of the pneumallergens or trophallergens. A threshold of 1,000 daltons was defined on the basis of physical/chemical tests on healthy human skin, above which the penetration of a molecule became impossible. This notion has always been in contradiction with the clinical observation of patients presenting contact urticaria to proteins and patients suffering from atopic dermatitis with positive skin tests to allergenic proteins [41, 9, 42].
The problem of the cutaneous penetration of allergens is thus that of crossing of the horny layer. As soon as this is crossed, the penetrating molecules are in direct contact with a network of epidermal dendritic cells, the Langerhans cells, (LC), which pick up these molecules and transport them to the lymph nodes, draining the penetration site. Taking into account the very dense network of dendritic LCs, it is in effect inconceivable that a molecule crossing the horny layer could penetrate without being in direct contact with the LCs. Recent studies have shown on the one hand that proteins do actually penetrate at the cutaneous level but also that it is possible to induce sensitization by exposure to allergenic proteins exclusively at the cutaneous level.

Cutaneous penetration - definitions

Percutaneous absorption corresponds to the transfer of a substance via the skin from the external environment as far as the blood. It can be defined as the sum of two phenomena: penetration of the molecules within the entire skin, followed by resorption in the blood or lymphatic systems from the papillary dermis and later, from the deep dermis.

The penetration stage, in physical terms, is a passive diffusion through each structure of the tegument: horny layer, Malpighian layer, dermis and cutaneous annexes. It depends on a transfer, which occurs at the interface of the environment/horny layer, without which no diffusion is possible. This indispensable stage, starting from the external environment or the medium, corresponds with the release of the molecule which will be diffused and therefore be put at the disposal of the organism. Once absorbed, the substance is distributed in the organism then, after having been metabolised or not, it is eliminated. The stages leading to percutaneous absorption are similar to those found in any other route of administration.

The skin barrier function

Within the skin, the epidermis forms the main barrier to cutaneous penetration and within the epidermis it is the horny layer which is charged with this role [30]. The horny layer (i.e. the stratum corneum) presents practically the same resistance to absorption as the entire skin. This means that any alteration of this layer will be accompanied by an increase in the skin penetration of molecules and will allow the absorption of molecules which would not have been able to penetrate a healthy epidermis. The total elimination of the stratum corneum by delamination with adhesive tape (stripping) leads to an increase in penetration by chemical agents [5, 21]. The resistance of live epidermis to diffusion is very weak compared to that of the horny layer, but it must, however, be admitted that when the latter is removed by stripping, the epidermis can have a barrier function, notably with regard to very hydrophobic molecules.
The dermis, in intact skin, only plays a very small part in the overall “barrier function”.
The resorption of absorbed molecules takes place through the blood (vascularisation of the dermal papillaries) and lymphatic systems. If the blood flow is insufficient to resorb the molecules as and when they arrive at the level of the deep dermis, local retention occurs. This phenomenon has been observed for numerous substances such as steroids (estradiol, progesterone, dexamethasone), toxic organophosphorous compounds (malathion, parathion, disopropyl fluorophosphate), non steroid anti-inflammatory drugs (indometacine, flufenamic acid, salicylic acid-based substances, ketoprofene, diclofenac) and enables us to better understand their mode of action at the level of structures situated below the application zone and where systemic diffusion and distribution by plasma does not seem to be necessary. The use of trace molecules like FITC (fluoresceine isothyocyanate) in murine models shows the rate of resorption of a hapten applied to the skin; in three minutes the FITC is found in the serum which is fluorescent and in less than an hour it is confined to the draining lymph nodes [50].

Routes for skin penetration

Two distinct pathways are available for penetration: 1) the transepidermal route where the substance will be diffused across the intercellular spaces of the horny layer or across the corneal cells themselves; 2) the cutaneous annex route which follows the pilo-sebaceous follicules and/or sweat glands. In the majority of cases penetration takes place via these routes together, both participating to the phenomenon and global penetration is the result of the conjunction of transepidermal and annex pathways.

Transepidermal penetration

The composite structure of the horny layer enables it to be presented schematically as a wall of ‘bricks’, made up of corneal cells (i.e. corneocytes) which are hydrophilic by nature of their protein content, surrounded by a lipophilic ‘cement’ made up of lipids which fill the extra-cellular spaces [14, 36]. In these conditions, diffusion through the horny layer can take place either directly by transcellular passage, with successive transfer through the cells and the extra-cellular spaces, or by intercellular passage through the winding spaces left free between the corneocytes.
Structural analysis of the horny layer shows that the lipids excreted in the intercellular spaces during the final stage of epidermic differentiation; free or esterified sterols, free fatty acids, triglycerides and sphingolipids, are organised in double layers which are arranged to separate the ‘hydrophilic’ and ‘lipophilic’ zones, thus creating an area of stratified diffusion with opposed physical/chemical properties [14, 31, 36]. The polar molecules can thus make their way towards the ‘hydrophilic’ regions of the intercellular lipidic layers while the ‘hydrophobic’ areas enable the nonpolar molecules to circulate.
The distance which a molecule travels is thus very different depending on whether it follows the transcellular pathway (the length of the diffusion pathway corresponding to the average thickness of the horny layer, i.e. from 10 to 40 μm) or the winding path which leads through the intercellular spaces filled with lipids, whose length has been estimated at several hundred μm.

Penetration via the cutaneous annexes

Over most of the body, the horny layer is crossed by the pilo-sebaceous follicules and the sweat glands, these structures offering zones of less resistance (“shunts”) to the penetration of molecules. In man, these two types of annexes represent less than 0.1 to 1 % of the total skin surface.
The anatomical structure of the sweat glands is complex and their resistance with respect to diffusion is difficult to evaluate. Because of this, their role as a penetration pathway is disputed, even if, despite the lack of recent studies, they seem indeed to participate in the penetration of iodine, histamine and adrenaline.
The pilo-sebaceous follicules are situated in epidermal invaginations and have always been considered as potential ‘shunts’, allowing a more rapid passage of molecules, in particular those which are diffused with difficulty across the horny layer [45]. Areas which are dense in follicules (the scalp, the arm pits) offer a higher permeability than those sites which are less hairy (forearms, palms).

Cutaneous and molecular factors involved in skin penetration

Diffusion through the skin depends on the heterogenous structural organisation of the different cutaneous layers, of their physical/chemical caracteristics (density, viscosity, solubility) and on the presence of proteins, as well as the potential for binding to these proteins [25]. The path of least resistance taken by the penetrating agent also depends on its relative affinity, or partition coefficient, for the different structures crossed and its volume.

Cutaneous factors

Several parameters govern the penetration stage:
– The metabolism of cutaneous enzymes is qualitatively similar to that found in the liver, but only represents a few percent of hepatic values. Metabolic activity can be a determining factor in absorption, in particular of non-protein chemicals, as is shown by the results of using potassium cyanide which greatly reduces trans-cutaneous diffusion.
– Anatomical variabilities in cutaneous penetration explain the difference in absorption of the same composition administered under identical conditions (vehicle, concentration, dose) on human skin, according to the anatomical area treated [60]. The skin of the scrotum offers the greatest permeability, the palmo-plantar zone is the least permeable. These anatomical variations have been observed for different molecules, among which are hydrocortisone, the organophosphorous compounds, benzoic acid and water. They can be linked to differences in the structure of the horny layer: thickness of the membrane, size of the corneocytes, lipid content, and abundance of pilo-sebaceous follicules. They are particularly important for evaluating potential occupational intoxication in areas which are constantly exposed and for defining the optimal site for drug administration during the development stage of a transdermal product. As an example, the post-auricular zone is very permeable to scopolamine, the chest to trinitrine and the upper arm to clonidine. It should also be mentioned that there are important variations from one individual to another but the same differences from one site to another within a given individual are always found.
– Hydration of the skin is a preponderant parameter in percutaneous absorption. Water is the natural plasticizer of the horny layer, its average content is 5 to 15 % and this quantity can reach 50 % when the skin surface is covered by an occlusive bandage [62]. In the fifteen minutes following the removal of such a bandage which has been in place for 24 hours, the trans-epidermal water loss can reach 10 times the normal value. This elevation of the water content of the horny layer results in real changes in permeability, which is considerably increased, and this is also related to important physical/chemical changes.
– Age. All structural variations of the horny layer are capable of modifying cutaneous permeability. Skin ageing manifests itself by a drying of the integument, a reduction in thickness, a decrease in lipid content and by alterations in the functioning of the sebaceous glands [43]. All these changes relating to age do not lead to the disruption in cutaneous permeability found in the skin of new-born compared with premature babies. In the premature baby the cutaneous barrier is proportional to the degree of prematurity and is much less efficient than that of a baby born at term [20, 44].
– Skin pathologies. The disturbances induced by many dermatoses influence cutaneous penetration. Anomalies in the epidermal barrier can also induce cutaneous inflammatory lesions [13]. The case of atopy will be dealt with later. Epidermal lesions caused by stripping, by irradiation from high doses of ultra-violet, by scarification or by abrasion are able, to varying degrees, to increase skin penetration. This is more pronounced relative to the impenetrability when the horny layer was intact, but does not reach the maximum threshold of a lesion created by complete delamination with the aid of adhesive strips [5].
– Numerous other factors influence skin penetration, some of them appear surprisingly able to modify the skin barrier function. Thus psychological stress is responsible in mice and in men for reducing the skin barrier function to an extent proportional to the perceived stress [3, 16].

Physical/chemical factors of the penetrating agent

Several parameters govern penetration, among which the polarity of the molecules as well as their size (molecular weight) are the most important:
– The speed at which a substance enters the skin depends on the difference between its concentration on the surface and below the skin.
– The length of the pathway of the diffusion of the substance through the horny layer, depending on whether it takes the transcellular or intercellular route or both.
– The diffusion coefficient of the substance.
– The interaction of the substance with cutaneous molecules during transport can delay penetration.
The molecular size (molecular weight -MW) of the substance which is being diffused is a very important parameter when considering penetration through healthy skin. Thus the diffusion constant for a large molecule is lower than that of a small molecule. The relation between the diffusion constant and the molecular size (measured by the molecular weight -MW-, or by the molecular volume -MV-) has been modeled in various ways. Kasting et al. proposed a relationship of an exponential type which can be used to predict the permeability coefficients [22, 52].
However, it is possible for molecules of a high molecular weight, in particular for proteins, to achieve penetration. The most interesting studies result from research into new ways of administering drugs percutaneously [19].
Using liposome formulations or transplanting a fatty acid onto the protein increases the diffusion of molecules with a high MW by rendering them more lipophilic and thus more easily available at the level of the inter-corneocyte lipids [15]. The preparations thus obtained use the intercellular pathways for penetration. Numerous molecules, defined as percutaneous absorption enhancers, are able to increase cutaneous penetration: ethanol, polyethylene glycol, linolenic acid, limonene etc... The utilisation alone or in association with penetrating agents has been able to increase the diffusion of proteins like LHRH [6]. Ultrasound increases skin penetration and its use has enabled the percutaneous penetration of large molecules like insulin, interferon gamma and erythropoietine [33]. The administration of an electric current of weak or strong voltage results in the same phenomenon of an increase in permeability. The association of an electric current with a detergent (sodium lauryl sulphate) induces conditions of highly increased cutaneous passage with the creation of intracellular penetration pathways capable of crossing the lipidic cellular membrane, the corneocyte envelope and the interior of the corneocytes [61]. It is thus possible to introduce molecules larger than 150 kD, like the immunoglobulins, at a rate which reaches therapeutic levels in the tissues (10 to 100 μg/hour/cm²). Iontophoresis also enables calcitonine to penetrate [11]. The association of iontophoresis and chemical enhancers induces important morphological modifications at the level of the horny layer, with dissociation of the corneocytes and rupture of keratin filaments, resulting in a further increase in the flow of macro-molecules crossing the epidermis, induced by the iontophoresis alone [6].

Cutaneous penetration of non-protein chemicals

Non-protein allergens are chemicals of low molecular weight known as haptens, and they are responsible for contact dermatitis. Haptens are only immunogenic after covalent or non-covalent interaction with the amino acids of epidermal proteins. The great majority of haptens are electrophilic molecules which interact with the nucleophile residues of cutaneous proteins [29]. Metals do not bind in a covalent manner but establish weak interactions with the amino acids of cutaneous proteins.
Haptens are often derived from chemicals called pro-haptens, following skin metabolism stages which can, on the one hand, modify the diffusion and on the other, the allergenic capacity. Proof for the metabolization of pro-haptens into haptens comes from the model of dimethylbenzanthracene (DMBA), a polyaromatic hydrocarbon (PAH). Allergic contact dermatitis to DMBA only occurs in mice which can metabolize it and inhibitors of PAH metabolism reduce the intensity of the eczema reaction [4]. One of the implications of these observations is that the individual cutaneous metabolism is at least as important as penetration in achieving sensitization and then an allergic reaction.
Allergic contact dermatitis is a delayed hypersensitivity reaction due to the activation in the skin of hapten-specific T cells. Two stages are necessary to its development: a clinically silent stage, sensitization, and the effector stage [10, 27].
Sensitization: the hapten, having crossed the horny layer, interacts with the proteins in the cutaneous cells. In particular it meets the network of epidermal dendritic cells (DC), the Langerhans cells. The interaction of the hapten/Langerhans cell induces: i) activation and maturation of DC which migrate from the epidermis to the dermis and then rejoin the lymph nodes draining the hapten penetration site; ii) processing of haptenized proteins and exposure at the membrane of the haptenized peptides in the peptide binding groove of MHC class II and class I molecules of DC. In the lymph nodes the DC present haptenized peptides to naive T cells and activate those which express specific receptors. Thus the T cell clones CD4 + and CD8 + are generated and join the blood circulation and then the cutaneous tissue. It is interesting to note that CD8 + T cells are effectors in the allergic contact dermatitis reaction in murine models while CD4 + T cells are doted with regulatory properties [7, 23].
The effector stage: this occurs when the same hapten is applied on the skin. The hapten is diffused through the corneal layer and interacts with proteins of the epidermal cells. The hapten is metabolized by the cutaneous cells and is found in the cellular membrane in the form of haptenized peptides on MHC class I and class II molecules. The DC are thus capable of presenting the hapten to CD4 + and CD8 + T cells present in the dermis. The role of keratinocytes and other cutaneous cells only expressing MHC class I molecules appears to be essential for the activation of CD8 + effector T cells and has long been underestimated [2].
The penetration of haptens via the epidermis is mainly due to their polarity as their molecular weight is generally less than 1kD. Cutaneous penetration is therefore rapid for lipophilic molecules and more difficult for hydrophilic molecules. In contrast to protein allergens, the penetration of haptens is not a problem and is therefore not a limiting factor in their allergenic capacity.
The fact that a hapten can induce allergic contact dermatitis in an individual, or not, does not then depend on its capacity for diffusion through the horny layer but rests above all on its ability to interact with proteins and to induce important modifications in their physical/chemical structure [29]. The quantity of haptens in contact with the integument is certainly the most important factor in immunological tolerance of haptens. Studies in murine models show that, at low doses, haptens do not in fact have allergizing capacities. Initially, this immunological non-response was interpreted as an incapacity of the immune system to recognise haptens present in only very small quantities. In fact, the non-response is an active phenomenon since animals treated in this way with cutaneous exposure at very low doses become tolerant of doses of stronger haptens which typically induce contact dermatitis in non-treated animals [49].

Penetration of proteins through the skin

It was long considered very difficult, if not impossible, for protein molecules (molecules of high molecular weight) to penetrate normal skin. Penetration is increased in certain pathologies, in particular in patients with atopic dermatitis (AD) who present a severe constitutional xerosis. The proteins responsible for these allergic reactions are additionally endowed with enzymatic properties which favour penetration through the epithelium.

Penetration of proteins through the skin

Demonstration of the capacity of proteins to penetrate normal skin comes from clinical observations of urticaria and eczema at the site of skin contact with proteins [37, 54]. Urticaria on contact with latex in patients who have immediate hypersensitivity occurs in the minutes following putting on latex gloves. Skin contact alone can induce systemic manifestations (Quincke’s oedema, anaphylaxis) in patients who are very sensitized. Protein contact dermatitis, which is observed in butchers, is reproduced with skin tests using bovine or porcine meats. AD patients frequently test positive to pneumallergens and/or trophallergens with a clinical and histological appearance typical of contact dermatitis [9, 41]. These tests are positive even when they are carried out on normal skin on the back at times when there are no AD eruptions.
The use of murine experimental models confirmed these clinical observations. Application of the protein ovalbumine on non-sensitized animal tegument results in the generation of specific IgE, which indicates that penetration and activation of B cells and type 2 ovalbumine-specific T cells have taken place [57]. If application of the protein is repeated, a contact dermatitis develops with all the histological characteristics of AD: predominance of T cells, infiltration of eosinophils, production of type 2 cytokines (IL-4, IL-5) and IFNγ. These studies go even further as the authors show that, once mice are sensitized cutaneously by ovalbumin, simple inhalation of the protein can trigger a bronchial hyper-reactivity in these animals [48]. Thus, ovalbumin can penetrate healthy skin, can induce a humoral immune (IgE) and specific T cell response, as well as inducing clinical signs of cutaneous allergy (AD) and respiratory allergy (asthma).

Enzyme activity of protein allergens

Little is currently known about the factors which determine the allergenicity of proteins found in the natural environment [8, 28]. One important step forward in this area was the discovery of the enzymatic properties (proteases, chitinases, lyases, amylases) of several respiratory and alimentary allergens [35, 47]. Thus, the current theory is that the enzymatic activity of allergens is necessary for their penetration through the epithelium, allowing direct access to the antigen-presenting cells.
The group of allergens which has been most studied is that of house dust, which is colonised by the acarian Dermatophagoides pteronyssinus (Der p), whose allergens are concentrated in their faeces. The allergic responses are the result of the contact of faeces with the respiratory epithelium by inhalation, or of contact with the skin directly. Der pl, the major allergen, is a cystein peptidase which shares sequence homologies with the catalytic site of a vegetal enzyme, papaine. At least three other allergens of Der p (Der p3, Der p6, and Der p9) are proteolytic enzymes whose amino acid sequences suggest that they belong to the group of serine peptidases. Der p allergens are proteolytic enzymes which are able to increase permeability through the bronchial epithelial cells by opening the tight junctions which form part of the systems of intercellular adhesion and regulation of paracellular permeability [55, 56]. The molecular targets of the Der p are occludines, members of the claudine family which are transmembrane proteins of the tight junctions [56]. Thus, Der p enzymes allow the penetration of allergens through the epithelial tegument enabling their uptake by CPA.
Easing the penetration of allergens is not the only factor capable of explaining their allergenic capacities. Der p1 polarizes the immune reponse towards the type 2 phenotype [12] while Der p9 induces the production of inflammatory cytokines by epithelial cells [24]. Taken altogether, these observations suggest that the presence in the air of proteins with an enzymatic action capable of altering the epithelial barrier not only promotes penetration of these proteins, but also the penetration of all the proteins present in the environment, including those which do not themselves have allergenic or enzymatic capacities.

Alteration of the skin barrier in AD

The physiopathology of AD is complex, associating cutaneous and immunological anomalies [18, 26, 53]. The principal cutaneous alterations concern the epidermis and in particular the horny layer and are responsible for the dry skin (xerosis) which is characteristic of the condition. Xerosis is responsible for the typical clinical appearance of AD patients: dry, scaly, rough, dull, slightly wrinkled skin. It is responsible for the pruritis which increases the penetration of pneumallergens.
The stratum corneum in AD patients is finer than in normal subjects and contains fewer intercellular lipids [59]. The lamellar organisation of the corneocytes is altered [40]. The hydrolipidic film on the surface is altered and washing aggravates this deficit [59]. The lipidic abnormalities mainly concern the ceramides, whose numbers in the corneal layer are very reduced, probably by a functional deficit in acylase sphingomyeline [34].
The alteration of the skin barrier in AD patients is shown by the reduction in the water-content of the stratum corneum and by the increase in transepidermal water loss. These anomalies are observed in the inflammatory zones but also in the skin areas with a clinically normal appearance [46, 58].
It is still not known whether the diminution of the barrier function of the stratum corneum is innate and pre-exists the development of the disease or if it is acquired and is the consequence of chronic cutaneous inflammation.
The skin barrier defects observed in AD are considered by some as a primitive, genetically determined abnormality [39, 51]. The murine model of spontaneous AD which develops in NC/Nga Tnd mice argues in favour of this hypothesis, as the animal presents water-retention anomalies as well as altered composition of the epidermal ceramides [1]. However, observation of a normalisation of the transepidermal water loss and the water content of the stratum corneum in AD patients who have been cured’ for several years goes against the idea of an intrinsic anomaly of the cutaneous barrier in AD [32].

Alteration of the barrier function could also be the consequence of cutaneous inflammation linked to the penetration of pneumallergens. One clinical study compared the alteration of the cutaneous barrier at the sites of atopy patch tests with the sites of patch tests to contact haptens, in AD patients who presented with nickel contact dermatitis [17]. This study compared the transepidermal water loss for each patient, on the sites of the two types of tests, with that of normal skin and showed that the barrier function of the epidermis is only decreased on the sites which are positive to pneumallergens. These studies show that the cutaneous inflammation induced by chemical contact allergens (type 1 T cells) is not responsible for a reduction in the barrier function, even in AD skin, while inflammation induced by contact allergen proteins (type 2 T cells) is able to severely alter the barrier function. Thus one can imagine that the alteration in the skin barrier induced by the pneumallergens increases the quantity of allergens which are able to penetrate, resulting in a vicious circle and perpetuating the eczema lesions.

Conclusions

The penetration of molecules which come into contact with the skin is a phenomenon which depends on numerous factors, but which is possible for small, non-protein chemical molecules as well as for proteins of a high molecular weight. Altogether, recent studies show that the barrier function of the epidermis only plays a small part in sensitization via cutaneous pathways in the main allergic conditions with cutaneous manifestations. This is explained by the very superficial subcorneal localisation of epidermal Langerhans dendritic cells, the principal antigen presenting cell responsible for the uptake of molecules which have penetrated. In allergic contact dermatitis it is accepted that sensitization occurs by epicutaneous pathways resulting in the induction in the lymph nodes of specific T cells capable of returning to the skin and inducing cutaneous lesions typical of eczema at the site of re-exposition to the hapten. In atopic dermatitis, the physiopathological processes are less clear. There is no doubt that eczema lesions can be triggered by cutaneous exposition to pneumallergens as patients develop eczema on contact with pneumallergens applied as patch tests. The possibility that patients can be directly sensitized by cutaneous exposition to pneumallergens is more controversial. Nevertheless, it is possible to immunize an animal with only cutaneous exposure to high molecular weight proteins with the induction of specific IgE and the development of the inflammatory lesions of atopic dermatitis [48]. Studies are in progress to determine if these mechanisms play a role in the development of atopic dermatitis in man and if the cutaneous penetration of allergens is not only responsible for the expression but also the induction of allergic immune responses [18, 53].

Acknowledgements. We are indebted to Jenny Messenger for translating this article. < 

References

1. Aioi A, Tonogaito H, Suto H, Hamada K, Ra C, Ogawa H, Maibach H, Matsuda H. Impairement of skin barrier function in NC/Nga Tnd mice as possible model for atopic dermatitis. Brit J Dermatol 2001; 144: 12-8.

2. Akiba H, Kehren J, Ducluzeau M, Krasteva M, Horand F, Kaiserlian D, Kaneko F, Nicolas J. Skin inflammation during contact hypersensitivity is mediated by early recruitment of CD8 + Tc1 cells inducing keratinocyte apoptosis. J Immunol 2002; 168(6): 3079-87.

3. Altemus M, Rao B, Dhabbar F, Ding W, Granstein R. Stress-induced changes in skin barrier function in healthy women. J Invest Dermatol 2001; 117: 309-17.

4. Anderson C, Hehr A, Robbins R, Hasan R, Athar M, Mukhtar H, Elmets CA. Metabolic requirements for induction of contact hypersensitivity to immunotoxic polyaromatic hydrocarbons. J Immunol 1995; 155: 3530-7.

5. Bashir S, Chew A, Anigbogu A, Dreher F, Maibach H. Physical and physiological effects of stratum corneum tape stripping. Skin Res Technol 2001; 7: 40-8.

6. Bhatia K, Singh J. Effects of linolenic acid/ethanol or limonene/ethanol and iontophoresis on the in vitro percutaneous absorption of LHRH and ultrastructure of human epidermis. Int J Pharm 1999; 180: 235-50.

7. Bour H, Peyron E, Gaucherand M, Garrigue JL, Desvignes C, Kaiserlian D, Revillard JP, Nicolas JF. Major histocompatibility complex class I-restricted CD8(+) T cells and class II-restricted CD4(+) T cells, respectively, mediate and regulate contact sensitivity to dinitrofluorobenzene. Eur J Immunol 1995; 25: 3006-10.

8. Bredehorst R, David K. What establish a protein as an allergen. J Chromatogr B Biomed Sci Appl 2001; 756: 33-40.

9. Castelain M. Atopic dermatitis and delayed hypersensitivity to dust mites. Clin Rev Allergy Immunol 1995; 13: 161-72.

10. Cavani A, Albanesi C, Traidl C, Sebastiani S, Girolomoni G. Effector and regulatory T cells in allergic contact dermatitis. Trends Immunol 2001; 22: 118-20.

11. Chang S, Hofmann G, Zhang L, Deftos L, Banga A. Transdermal iontophoretic delivery of salmon calcitonin. Int J Pharm 2000; 200: 107-13.

12. Comoy E, Pestel J, Duez C, Stewart GA, Vendeville C, Fournier C, Finkelman F, Capron A, Thyphronitis G. The house dust mite allergen, Dermatophagoides pteronyssinus, promotes type 2 responses by modulating the balance between IL-4 and IFN-gamma. J Immunol 1998; 160: 2456-62.

13. Elias P, Wood L, Feingold K. Epidermal pathogenesis of inflammatory dermatoses. Am J Contact Dermatitis 1999; 10: 119-26.

14. Elias P, Feingold K. Coordinate regulation of epidermal differentiation and barrier homeostasis. Skin Pharmacol Appl Skin Physiol 2001; 14: S28-S34.

15. Foldvari M, Baca-Estrada ME, He Z, Hu J, Attah-Poku S, King M. Dermal and transdermal delivery of protein pharmaceuticals: lipid-based delivery systems for interferon alpha. Biotechnol Appl Biochem 1999; 30: 129-37.

16. Garg A, Chren M, Sands L, Matsui M, Marenus K, Feingold K, Elias P. Psychological stress perturbs epidermal permeability barrier homeostasis: implications for the pathogenesis of stress-associated skin disorders. Arch Dermatol 2001; 137: 78-82.

17. Gfesser M, Rakoski J, Ring J. The disturbance of epidermal barrier function in atopy patch test reactions in atopic eczema. Br J Dermatol 1996; 135: 560-5.

18. Goerdt S BR, Dippel E, Orfanos CE. Beyond inflammation: tolerance, immunotherapy and more. Eur J Dermatol 1999; 9: 507-13.

19. Kaiserlian D, Etchart N. Epicutaneous and transcutaneous immunization using DNA or proteins. Eur J Dermatol 1999; 9: 169-76.

20. Kalia Y, Nonato L, Lund C, Guy R. Development of skin barrier function in premature infants. J Invest Dermatol 1998; 111: 320-6.

21. Kalia Y, Alberti I, Naik A, Guy R. Assessment of topical bioavailability in vivo. The importance of stratum corneum thickness. Skin Pharmacol Appl Skin Physiol 2001; 14: S82-S86.

22. Kasting G. Kinetics of finite dose absorption through skin. 1. Vanillylnonanamide. J Pharm Sci 2001; 90: 202-12.

23. Kehren J, Desvignes C, Krasteva M, Ducluzeau MT, Assossou O, Horand F, Hahne M, Kagi D, Kaiserlian D, Nicolas JF. Cytotoxicity is mandatory for CD8(+) T cell-mediated contact hypersensitivity. J Exp Med 1999; 189: 779-86.

24. King C, Brennan S, Thompson P et al. Dust mite proteolytic allergens induce cytokine release from cultured airway epithelium. J Immunol 1998; 161: 3645-51.

25. Kompaore F, Marty J, Dupont DC. In vivo evaluation of the stratum corneum barrier function in blacks, caucasians and asians with two non invasive methods. Skin Pharmacol 1993; 6: 200-7.

26. Krasteva M, Jullien D, Choquet G, Nicolas JF. Physiopathologie de la dermatite atopique. Ann Dermatol Venereol 1998; 125: 785-9.

27. Krasteva M, Kehren J, Ducluzeau MT, Sayag M, Cacciapuoti M, Akiba H, Descotes J, Nicolas JF. Contact dermatitis I. Pathophysiology of contact sensitivity. Eur J Dermatol 1999; 9: 65-76.

28. Lehrer S, Horner W, Reese G. Why are some proteins allergenic ? Implications for biotechnology. Crit Rev Food Sci Nutr 1996; 36: 553-64.

29. Lepoittevin JP, Leblond I. Hapten-peptide-T cell receptor interactions: Molecular basis for the recognition of haptens by T lymphocytes. Eur J Dermatol 1997; 7: 151-4.

30. Leveque J, Ribaud C, Garson J. Caractérisation biophysique du stratum corneum. Relation structure-fonctions. Pathol Biol 1992; 40: 95-108.

31. Marty J, Dervault A. La couche cornée: barrière et réservoir. In: 7° cours de Biologie de la Peau. S. I. J Thivolet, ed, 1993; p. 31-42.

32. Matsumoto M, Sugiura H, Uehara M. Skin barrier function in patients with completely healed atopic dermatitis. J Dermatol Sci 2000; 23: 178-82.

33. Mitragotri S, Blankschtein D, Langer R. Ultrasound-mediated transdermal protein delivery. Science 1995; 269: 850-3.

34. Murata Y, Ogata J, Higaki Y, Kawashima M, Yada Y, Higuchi K, Tsuchiya T, Kawainami S, Imokawa G. Abnormal expression of sphingomyelin acylase in atopic dermatitis: an etiologic factor for ceramide deficiency. J Invest Dermatol 1996; 106: 1242-9.

35. Musu T, Grégoire C, David B, Dandeu JP. The relationship between the biochemical properties of allergens and their immunogenicity. Critical Review in Allergy and Immunology 1997; 15: 485-98.

36. Nemes Z, Steinert P. Bricks and mortar of the epidermal barrier. Exp Mol Med 1999; 31: 5-19.

37. Neubauer H, Rytter M, Kleine-Tebbe J, Haustein U. Protein contact dermatitis induced by isolated delayed-type hypersensitivity to meat proteins. Contact Dermatitis 1999; 41: 113.

38. Norlen L. Skin barrier structure and function. J Invest Dermatol 2001; 117: 823-9.

39. Ogawa H, Yoshiike T. A speculative view of atopic dermatitis: barrier dysfunction in pathogenesis. Journal of Dermatological Science 1992; 5: 197-204.

40. Pilgram G, Vissers D, van der Meulen H, Pavel S, Lavrijsen S, Bouwstra J, Koerten H. Aberrant lipid organisation in stratum corneum of pateints with atopic dermatitis and lamellar ichtyosis. J invest Dermatol 2001; 117: 710-7.

41. Reitamo S, Visa K, Kahonen K, Kayhko K, Stubb S, Salo OP. Eczematous reactions in atopic patients caused by epicutaneous testing with aeroallergens in atopic eczema. Br J Dermato 1986; 114: 303-9.

42. Ring J, Darsow U, Gfesser M, Vieluf D. The Atopy Patch Test’ in evaluating the role of aeroallergens in atopic eczema. Int Arch Allergy Immunol 1997; 113: 379-83.

43. Roskos K, Maibach H. Percutaneous absorption and age. Implications for therapy. Drugs Aging 1992; 2: 432-49.

44. Rutter N. Clinical consequences of an immature barrier. Semin Neonatol 2000; 5: 281-7.

45. Schaefer H, Lademann J. The role of follicular penetration. Skin Pharmacol Appl Skin Physiol 2001; 14: S23-S27.

46. Seidenari S, Giusti G. Objective assessment of the skin of children affected by atopic dermatitis: a study of pH, capacitance and TEWL in eczematous and clinically uninvolved skin. Acta Dermato Venereol 1995; 75: 429-33.

47. Shakib F, Furmonaviciene R. The significance of enzymic and other biological activities of proteins in relation to their capacity to serve as allergens. Clin Exp Allergy 2000; 30: 1056-7.

48. Spergel JM, Mizoguchi E, Brewer JP, Martin TR, Bhan AK, Geha RS. Epicutaneous sensitization with protein antigen induces localized allergic dermatitis and hyperresponsiveness to methacholine after single exposure to aerosolized antigen in mice. J Clin Invest 1998; 101: 1614-22.

49. Steinbrink K, Sorg C, Macher E. Low zone tolerance to contact allergens in mice: A functional role for CD8(+) T helper type 2 cells. J Exp Med 1996; 183: 759-68.

50. Steinbrink K, Kolde G, Sorg C, Macher E. Induction of low dose tolerance to contact allergens in mice does not require functional Langerhans cells. J Invest Dermatol 1996; 107: 243-7.

51. Taieb A. Hypothesis: from epidermal barrier dysfunction to atopic disorders. Contact Dermatitis 1999; 41: 177-80.

52. Taljera P, Kleene N, Pickens W, Wang T, Kasting G. Visualization of the lipid barrier and measurement of lipid pathlength in human stratum corneum. AAPS Pharma Sci 2001; 3: E13.

53. ThestrupPedersen K, Ellingsen AR, Olesen AB, Lund M, Kaltoft K. Atopic dermatitis may be a genetically determined dysmaturation of ectodermal tissue, resulting in disturbed T-lymphocyte maturation – A hypothesis. Acta Derm Venereol [Stockh] 1997; 77: 20-1.

54. Wakelin S. Contact urticaria. Clin Exp Dermatol 2001; 26: 132-6.

55. Wan H, Winton HL, Soeller C et al. Der p1facilitates transepithelial allergen delivery by disruption of tight junctions. J Clin Invest 1999; 104: 123-33.

56. Wan H, Winton HL, Soeller C, Taylor G, Gruenert D, Thompson P, Cannell M, Stewarts G, Garrod D, Robinson C. The transmembrane protein occludin of epithelial tight junctions is a functional target for serine peptidases from faecal pellets of Dermatophagoides pteronyssinus. Clin Exp Allergy 2001; 31: 279-94.

57. Wang LF, Lin JY, Hsieh KH, Lin RH. Epicutaneous exposure of protein antigen induces a predominant Th2-like response with high IgE production in Mice. J Immunol 1996; 156: 4079-82.

58. Werner Y, Lindberg M. Transepidermal water loss in dry and clinically normal skin in patients with atopic dermatitis. Acta Dermato Venereol 1985; 65: 102-5.

59. White M, Jenkinson D, Lloyd D. The effect of washing on the thickness of the stratum corneum in normal and atopic individuals. Brit J Dermatol 1987; 116: 525-30.

60. Ya-Xian Z, Suetake T, Tagami H. Number of cell layers of the stratum corneum in normal skin – relationship to the anatomical location on the body, age, sex and physical parameters. Arch Dermatol Res 1999; 291: 555-9.

61. Zewert T, Pliquett U, Vanbever R, Langer R, Weaver J. Creation of transdermal pathways for macromolecule transport by skin electroporation and a low toxicity pathway enlarging molecule. Bioelectrochem Bioenerg 1999; 49: 11-20.

62. Zhai H, Maibach H. Effect of skin occlusion on percutaneous absorption: an overview. Skin Pharmacol Appl Skin Physiol 2001; 14: S1-S10.