Abridged from the

Dermatology Progress in Foundation Dermatology

Editor: Alan N. Moshell , MD.

Stratum Corneum Moisturization at the Molecular Level

Anthony V. Rawlings, Ian R. Scott, Clive R. Harding, * and Paul A. Bowsed Unilever Research, Edgewater Laboratory, Edgewater, N ew Jersey, U .S.A.; ' Unilevcr Research, Colworth Laboratory, Sharnbrook, Bedford; and tUnilever Research, Port Sunlight Laboratory, Bebington, Wirral, U.K.

n any living system, control of water translocation is essential corneocytes embedded in a lipid matrix (see Fig 1). The main func­ for survival. Being in close proximity to a non-aqueous envi­ tion of the epidermis is to produce the stratum corneum; a selec­ ronment this control is a fundamental property of our skin tively permeable outer layer that protects against water loss and and it uses mechanisms that are complex, elegant, and unique chemical insult. However, as will become evident from this review, in nature to achieve it. the barrier function of the stratum corneum is not its only function. IThe skin preserves water through intercellular occlusion (water The combination of the barrier properties of the stratum corneum permeability barrier) and cellular humectancy (natural moisturizing and its inherent cellular humectant capabilities moisturize the stra­ factor or NMF). The mechanisms for producing the water perme­ tum corneum, which is important for maintaining the flexibility of ability barrier and NMF are not only complex but also susceptible to the stratum corneum and its desquamation. The most characterized disturbance and perturbation. components of the stratum corneum are keratins, specialized cor­ This review begins with an overview of the vast amount of work neocyte envelope proteins, lipids, NMF, specialized adhesion struc­ that has led to a greater understanding of the mechanisms of mois­ tures (desmosomes), and enzymes. We will focus our attention on turization of the stratum corneum, including the nature of the water the last four components and discuss their relationship to stratum permeability barrier, and specifically examines the role of essential corneum moisturization. fatty acids, together with the role of NMF. The final parts of the review then examine the more recent knowledge on the biologic Barrier Lipids and the Role of Essential Fatty Acids General variation in barrier lipids (Variatiolls ill Stratum Comeum Lipids) and health is, in part, dependent on the nutritional balance of the diet other constituents that affect stratum corneum structure and func­ involving proteins, carbohydrates, lipids (saturated and unsatu­ tion, their variability, and particularly how they contribute to con­ rated), vitamins, and minerals. The inter-relationship between gen­ trolling moisturization and desquamation (New Irlsights illto Desqua­ eral health and skin health is reflected by the incidence of skin matioll and AbrlOrmal Desquamation: Comparisoll oj Ichthyoses and problems with nutritional deficiencies, e.g., scurvy, kwashiorkor, Xerosis) and finishes with a summary. and pellagra. For many decades it has been known that the absence of lipids in OVERVIEW the diet leads to unhealthy skin. In 1929 and 1930 Burr and Burr Stratum Corneum Structure Original histologic observations [1 ,2] showed that linoleic acid was an essential dietary requirement of the typical "basket weave" appearance of the stratum corneum in animals. Much later observations [3 - 6] indicated that the absence did not do justice to a full appreciation of its many functions that we of this essential fatty acid (EFA) from the diet results in a general know of today. Over the past three to four decades the stratum deterioration of the health of the animal and, specifically, in the corneum has been investigated by physical chemists, dermatolo­ development of a scaly skin condition with concomitant increase in gists, biologists, and more recently enzymologists. Initially consid­ trans-epidermal water loss. ered to be a homogeneous film, the stratum corneum has now been However, it was not until the late 1970s that Prottey and Houts­ shown to be a heterogenous structure composed of protein-enriched muller showed that the water permeability ofEFA-deficient skin of people [7] or animals [8] could be kept within normal ranges by Reprint requests to: Dr. Anthony V. Rawlings, Unilever Research, 45 direct topical application of linoleic acid or other EFA. Although River Road, Edgewater, NJ 07481. topically applied linoleic acid rapidly reversed the cutaneous mani­ Abbreviation: NMF, natural moisturizing factor; RXLI, recessive X­ festations of EFA deficiencies [5 - 8], the way it achieved this was linked ichthyosis. unknown. Potentially, the mechanism was postulated as the EFA

Copyright © 1994 by Dermatology Foundation, 1560 Sherman Avenue, Evanston, Illinois 60201 731 732 RAWLINGS ET AL THE JOURNAL OF IN VESTIGATIVE DERMATOLOGY

o I

Desqt"lo UlnUn STUATUY OISJUNCTUU o Cornco...:yle STJUTU~! Ce latnide 1 L'Pld BUB. I ;_ -- -..J C OR~EU~! 115%) -"'"- ., -~~

$TIUTU.\J COYP"CTUY n l~·· bol.l d l.l.pJd- ? _ - 1 ~ O < H/O HI t ILl,j ,,0 eornco(.'ylc~ lhllubr uu: CO Atine ~..... <: I' UU\l l all ~ ~ O H 0 1-1 I OH OH I-,x "\... ,1:):1 L I lnmcllar ST J('\TU~ ! Ceral11ide 2 Ceramide 3 0 ·1 1 t1U .. kentob, .. Un ~ pi 1. ,. Innllill. GI!A:--1ULO Sm ! (22%) (16%)

01~ 0 01~ 0 NH ST!UTU~! NH 0 11 S P!:--10SU~1 0 1-1 0 11 ~OH m---- ~Cerarnid e 4/5 (23%) STRATm! . ' JJJJJJ UASALE O l-:" O 0 1 ' ~ 0 Figure 1. Schematic representation of the epidermis. NH OH NH 0 1-1 0 11 /,~ 0 OH ~0 11 ~O Ceramide 6 II acting as a simple structural component, or via conversion to prosta­ ("10% ) (14%) glandins. Although linoleic acid does have classical structural roles in, for Figure 3. Chemical structures of stratum corneum ceramides. example, phospholipids, the slight differences between, say, oleic and linoleic acid does not explain the huge differences in EFA defi­ ciencies when oleic acid replaces linoleic acid in phospholipids. The the epidermal water permeability barrier it had to meet two criteria. role ofEFAs in barrier function is also not via prostaglandin synthe­ It had to contain linoleic acid or a derivative and it had to be detect­ sis, as the essential first step of conversion from linoleic acid to ab le in the region where the major barrier existed: the stratum arachidonic acid does not occur in skin [8]. T he curing effect of corneum or more specificall y the stratum compactum [13]. Elucida­ li noleic.acid is also no~ inhibited by indomethacin, a powerful pros­ tion of the way in which linoleic acid and other essential fatty acids tagland1l1-synthetase 1I1hlbltor, and a similar benefit in reducing control the loss of water from the skin has only emerged during the scaliness/water loss can be obtained using columbinic acid, which structurally carmot be converted to prostaglandins [9]. The specific last 10 years. .The detailed and pioneering work of Gray and colleagues in the role of EFAs was thus not via either of the mechanisms posulated at ITI1d to l at~ 197?s was pivotal and catalytic in opening up the whole the time. field of sk1l1 lipid research. Up to this point, most of the skin lipid Although early researchers recognized that the epidermis was resear~h had centered on phosph~lipids , which although rich in more impermeable than the dermis, it was not until the early 1940s that the precise location of the barrier was thought to reside in the l1l10ieJc aCid could not be 111volved 111 the water permeability barrier as they are almost totally absent from the stratum corneum. While stratum corneum [10]. Twenty years later, using organic solvent extraction of the stratum corneum, Matoltsy et at [11] demonstrated Gray al~d co-workers we.re ide~tifying a number of glycolipids that lipids played a critical role in barrier function. Middleton fur­ p.rese~1t I? skl1~ (14), and d ! sc~venng a unique epidermal glycolipid ther demonstrated [12] that lipids were also important in preventing nch 111 lin?letc aC id [15], US111g electron microscopic techniques, the loss of NMF from inside the corneocytes. Thus, for a lipid Elias and IllS co-workers [16 -18] were visualizing the bilayer nature of lipids in between the stratum corneum cell s. component to have a key role in the formation or maintenance of Building on Gray's biochemical studies, two main research centers ~hen e1uc.idate~ the structures of the linoleate rich lipids o present 111 the eptdermls. Bowser and co-workers determined un­ e~uivoca lly the structure [~9 ,20] of the unique epidermal glycoli­ pid, the acylglucosylceralTI1de, but proved that in itself it could not Nil ? II,.OH 0 ~- 0 1 ' 1 be a barrier lipid as it was absent from the stratum compactum [13]. I \ Stmultaneously With Downing and Wertz's group [21- 26] at Iowa, 0 1-1 011 Bo:vser e/ al [27] discovered a new class of structurally related cer­ ~0 11 Acylglucosylceratnide (AGC) amJdes, one of which was rich in linoleic acid and present in the o stratum compactum (Ceramide 1 or acylceramide; Fig 2). Compar­ ~oI\N'-J\A/VWWVVVv/ Ison of the structures of glucosylceramides suggested that they were "'YOH precursors of related cerarnides with acy lglucosylceramide being \f\/VVVV\I;-AOI-I the precursor of Ceramide 1. Much later, in metabolic studies, this Acylceratnide (AC or Ceratnide '1) was shown to be true [28]. o .Following on from this work, several other types of glucosylcera­ o ~Ides and ceram ld ~s were also discovered. In addition to the major

II li~oleoyl-nch speCles, five other chromatographically distinct cera­ Acylacid (AA) ITI1de and glucosylceramide classes could be separated and isolated

Nil by thin-layer chromatography [23,29 - 31]. For the stratum cor­ ,(,0 11 neu~ ceramides these were named according to their polarity be­ \f\/VVVV\I;-AOI1 ~111nmg w~th the na~1e Ceramide 1 for the most non-polar and Sphingosine l1110lelc aCld - conta1l1111g species (Fig 3). These lipids consist of C W C 22 sph il~g osin~ (trans-4-sphingenine), dihydrosphinogosine Figure 2. Chemical structures of acylglucosylceramide, ceramide 1, and phytosphl11ogosl11e (4-D-hydroxy sphingenine) bases ami dated acylacid, and sphingosine. to long chain fatty acids, which mayor may not be 0'- or w-hydrox- VOL. 103, NO.5 NOVEMBER 1994 STRATUM CORNEUM MOISTURlZATION 733

droxyoctadecadienoic acid) are also essential for epidermal differen­ tiation [42]. In addition to the solvent extractable ceramides, a number of covalently bound lipids have been discovered [43,44]' These lipids were isolated from the stratum corneum by alkaJine hydrolysis fo l­ lowed by further solvent extraction. In human corneum, two (V-hy­ droxy fatty acid -containing ceramides were shown to be present (Cerami de A and Ceramide B) together with (V-hydroxy fatty acids and fatty acids. The fatty acid chain length distribution of the cova­ lently bound ceramides and the (V-hydroxy fatty acids were found to

OH be similar to the fatty acid profile of the epidermal acylglucosylcera­ mide and stratum corneum Ceramide 1, containing very-long­ ~ ~ chain fatty acids of 30 carbon atoms or more, some of which are OH unsaturated, indicating their source. The major normal chain fatty acid species covalently attached to the corneocytes were shown to be )--/0-{ )-0-{ palmitic, stearic, and oleic acid. Recently, cholesterol has also been ~Ot-! 01 1 0 11 U ti Ot l 0 11 OH 01-1 OH~ 01< reported to be covalently attached to the corneocyte envelope [45]. polyoxyacylceramide Regional cellular differences in other lipid species also occur. Decreases in cholesterol su lphate, phospholipid, and glucosylcera­ Figure 4. Hydroxylated variants of acylglucosylceramide (polyox­ mide levels occur in the stratum disjunctum [46] and, of the total yacylceramide). saponifiable fatty acids, 9% were of the very-long-chain type (C30 chain length) in the stratum compactum, whereas these were absent from the stratum disjunctum layers [47] . Presumably the fatty acids ylated and unsaturated. As mentioned, the (V-hydroxyl of Ceramide were derived from Ceramide 1 and these results imply that Cera­ 1 was further shown to be esterified to fatty acids, with linoleic acid mide 1 is critical for the barrier function of the stratum compactum. as the major acid present, which also occurred for porcine Ceramide Indeed Ceramide 1 is thought to playa major role in structuring 6a. The corresponding species in human stratum corneum, how­ stratum corneum lipids essential fo r barrier function [48]. Despite ever, consisted of a phytosphingosine ceramide in which the base these compositional changes, the intercellular lipid lamellae in the hydroxyl was further esterified to an a-hydroxy fatty acid [29]. stratum compactum and lower regions of the disjunctum, when More recently, Colarow [32] showed the presence of unsaturated viewed by electron microscopy, remain intact [48]. except in the fatty acids N-acy lated to human Ceramide 2, particularly linoleic very outermost layers of the stratum corneum [49,50]. These small acid, oleic acid, and a- together with )I-linolenic acids and Haman­ changes in lipid species, however, may impart different physical aka et af [33] showed that a major portion of the sphingoid base in properties to the stratum corneum lipids leading to regional differ­ human epidermis contains a trihydroxy-eicosphingosine. ences in barrier or other functional properties. Bowser and White A lipid species call ed acyl acid, which appears to be the (V-esteri­ [51] isolated the stratum compactum and demonstrated that this fied N-acyl fatty ac id portion of Ceramide 1 (Fig 2), has been region of the stratum corneum presents a major barrier to water observed in the epidermis and stratum corneum [1 9,20,34]. Inter­ permeation. It was suggested that the state of cohesion between the estingly, in sunburn peelings of human stratum corneum, acyl acid cells, tightly sandwiching the lipid lamellae, may be partly respon­ accounted for 2% of the total lipid, similar to the levels reported for sible for its barrier performance, even though this region is only five Ceramide 1 in normal skin. In addition, free sphingoid basses (ap­ cell layers thick. However, the lyotropic nature of the lipids found proximately 0.5% of the total extractable lipid) were shown to be in these different tissue regions may also contribute to the perme­ present in human [35] together with pig epidermis [36]. The major­ ability differences [5 2]. Although other lipid species are present in ity was sphingosine and dihydrosphingosine with only traces of the stratum corneum (small amounts of phospholipids, gl ucosylcer­ phytosphi.ngosine being ~resent . It is possibl~ that both of these amides, and cholesterol sulphate), the major lipids are ceramides, lipid speCies could be derIved from hydrolySIS of Ceramlde 1 or cholesterol, and fatty acids. All of these contribute to the water acylglucosylceramides. Recently, ceramide-hydrolyzing enzymes permeability barrier of the stratum corneum and their individual have been identified in the stratum corneum [37]. Their roles, how­ effects have been elucidated by the elegant studies of Eli as and ever, are uncertain. N evertheless, sphingosine may be involved in a stratum corneum - epidermis signaling function as it has been re­ ported to inhibit keratinocyte proliferation [38]. Other metabolites of Ceramide 1 also occur that are thought to be 600 important in the formation and maintenance of the stratum cor­ GPI 15- 29 GP 11 30-• 49 neum permeability barrier. Nugteren and his team [20] identified a 500 GP t!l 50-r..'i • • tri-hydroxy derivative (polyoxyacylceramide) of the linoleate-rich c . acylceramide in rat epidermis (Fig 4) that was thought to be impor­ ;; Gf' I VS GP lllp "005 • tant for barrier function. The acylglucosylceramide, thus, appears to K 400 • reside in the membrane coating granules, the contents of which are rg expelled into the intercellular space where they undergo conversion .s. z 300 to the acyl ceramide and then to polyoxyacylceramide. This ill siw 0 • metabolic sequence enables the right molecule to be located in the ...~ • ai 200 right place at the right time. How the polyoxyacyl ceramide u z • 0 achieves a barrier function is still uncertain. The possibilities are that u « it strongly hydrogen bonds to water, induces a polymerization reac­ u 100 c. tion [20], or offers suffi cient polarity to maintain the ordered inter­ cellular lamellar sheets [19] necessary for limiting water loss [18]. It 0 4 is also interesting to note that the dietary inclusion of a cyclooxy­ OEPTH IN STRATUM CORNEUM (M;crons) genase/lipoxygenase inhibitor (eicosatetraynoic acid) [39,40] pre­ vents the formation of the polyoxy acyl ceramide and gives rise to an Figure S. Profile of mean pyrollidone carboxylic acid (peA) con­ extremely dry, scaly skin [41] similar to that seen in EFA deficiency. centrations versus depth in stratum corneum from the hands of the However, other hydroxylated metabolites of linoleic acid (1 3-hy- young, middle-age, and old-age groups. 734 RAWLINGS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Mole perce nt Figure 5 shows typical depth-versus-concentration profiles, 0 to 20 30 w hich indicate, first, that the level of NMF declines markedly Gl XIPCA towards the surface of the skin and, second, that there is a significant ARG IORN ICIT age-related decline in the level ofNMF even, or in fact particularly, SEI1 deep within the stratum corneum. Aged skin, therefore, has low PR O NMF because of lower synthetic activities and not because of a GlY greater leaching of NMF during cleansing. This age-related defi­ HI SIUCA ciency in NMF production is, like most skin ageing symptoms, ASXIALA associated with actinic damage, as a similar trend is not found in arm lYS skin not regularly exposed to sunlight. THR ~ T hese observations suggest that the inabili ty to produce NMF (or lEU its loss from the superficial layers of the stratum corneum) might be VAL ~ a critical mechanism contributing to skin problems ranging from IlE ~ simple xerosis to severe psoriasis. It is important to understand the TYR ~ nature of the mechanism of NMF production so as to gain insight PH E ~ into the potential cause(s) of such conditions. CYS A critical first step in discovering this mechanism came in a rela­ _ F il ~gg rill MET tively little known publication [61] describing the time course of TRP IIIIIR NMF amlllO acids appearance of radioactive pyroglutamate (PCA; a major component of the NMF) after injection of a radioactive amino acid (glutamic Figure 6. Comparison of amino acids ofNMF with 6laggrin. acid) in mice. The finding of a 2 - 3-d delay before the appearance of PCA suggested that there might be a sequence

glutamic Acid ~ protein ~ NMF. coworkers (for review see [53]). The importance of the combination Our laboratory performed further studies [62,63] that culminated in of these lipids for efficient barrier function, however, has only re­ the discovery that a single protein in the epidermis was the sole cently been demonstrated [54]. It is also apparent that epidermal and source of all of the amino acid - derived components of the NMF. stratum corneum lipids may also have roles other than barrier func­ The most significant evidence supporting this theory is presented in tion. There is also some support that certain derivatives Inay be Fig 6, which shows the remarkable similarity of the amino acid "Chalones" - feedback biochemicals produced in the stratum cor­ composition of this protein with that of the NMF. This correlation neum that are able to regulate basal cell division or early differentia­ disproved the idea that NMF was some sort of residue from un­ tion events. needed proteins - the "dust bin hypothesis." Further support for Natural Moisturizing Factor It has been known for many years the specific and programmed nature of the production of NMF [55] that NMF is a mixture of amino acids, derivatives of amino came from the observation that several of the free amino acids pro­ acids, and specific salts. NMF is found exclusively in the stratum duced from the protein degradation are specifically enzymically corneum cells and can make up to as much as 10% of the dry weight modified to produce functional molecules. Glutamine is converted of the stratum corneum cells (corneocytes). The role of the NMF to pyrrolidone carboxylic acid, a potent humectant [64], whereas lies in the fact that its constituent chemicals are highly water soluble histidine is converted to urocanic acid, which was first considered an and hygroscopic. NMF will, therefore, absorb atmospheric water important UV absorber [65] and is now implicated in immune sup­ (even at relative humidities as low as 50%) to the extent that it pression [66]. dissolves in its own absorbed water- it is, thus, a very efficient This unique NMF precursor protein was further characterized humectant. Biologically, the significance of this is that it allows the and proved to be very unusual. Its molecular weight approached 500 stratum corneum to maintain levels of liquid water even when its kDa; it was highly phosphorylated and extremely susceptible to outermost layers are exposed to quite low humidities. breakdown during extraction procedures. The protein had, in fact, The conventional view of the importance of this liquid water was been studied in a variety of partially degraded forms by several that it plasticized the stratum corneum, keeping it resilient and groups [67 - 71]. The protein proved to be located in the keratohya- preventing cracking and flaking due to mechanical stresses. To a degree this view still holds, but, as will be discussed in more detail later, we now know that specific hydrolytic enzymes, critically important to allowing separation of corneocytes from each other, only function in an aqueous or semi-aqueous environment. Thus, there is a biochemical need as well as a biophysical need for NMF ./ peA binds water despite } / .'''""',oo.~, and the consequences of NMF deficiency are reflected in the disrup­ ) tion of both of these functions. The importance ofNMF is clear and the correlation of its absence

in disease states with concomitant stratum corneum abnormality is 1 ___ Zone of stobie lIIoggrln. striking. In both psoriasis [56] and ichthyosis vulgaris [57] there is a virtual absence of NMF and the manifestation of this is severe --::==-__ ~...... '\ ~ Pro111aggrln converted cracking and flaking of the skin and failure of normal desquamation. I~ Into Iliaggrin. There is also evidence of insufficiency of NMF in severe dry flaky Profilnggrln synthesis. skin conditions [58 - 60]. Many studies have measured NMF levels extracted from the skin surface by washing/extraction techniques. We developed a sequential tape-stripping procedure that allowed ~ Keratin synthesis measurement of the concentration of NMF at different depths in 1 the stratum corneum [59]. In healthy skin that has not been exposed ) - -- Coli division. to surfactants the N MF concentration is independent of depth (until the very deepest layers of the stratum corneum are reached; see below), but in normal skin exposed to normal soap washing much of the NMF was washed out from the superficial stratum corneum Figure 7. Schematic of epidermis showing profilaggrin conversion [59]. to 61aggrin and peA. VOL. 103, NO.5 NOVEMBER 1994 STRATUM CORNEUM MOISTURIZATION 735 lin granules of the epidermis and to undergo a species-specific de­ zero. Immunohistochemical examination of the stratum corneum phosphorylation and proteolytic processing as the granular cell in this state shows that all cells at all levels in the corneum are filled made the transition into a corneocyte cell [72,73). The resulting with unconverted filaggrin. Removal of the occlusion results in a mixture of highly basic proteins proved to be identical to proteins rapid conversion of filaggrin to NMF in all but the innermost, most that had been independently discovered as being necessary in giving hydrated, layers of the stratum corneum. the correct formation and macrostructure of the keratin microfibrils The stratum corneum has thus developed an elegant process for in the stratum corneum-the so-called keratin matrix proteins [74]. adjusting itself to the environmental conditions to which it is ex­ The term filaggrin (filament aggregating protei,,) was coined for posed so that it is optimally moisturized. As yet, the molecular and these proteins [75] and the high-molecular-weight precursor was enzymatic nature of the system by which it achieves this water named profilaggrin. Although during the past few years, the gene activity-dependent control has not been elucidated and remains for profilaggrin has been sequenced and a great deal is now known one of the more challenging questions in understanding the once about the structure and the control of its expression, the precise apparently simple structure of the stratum corneum. nature and control of the phosphatases and proteases involved in its VARIATIONS IN STRATUM CORNEUM LIPIDS processing to the mature filaggrin protein remain to be fully eluci­ dated. Nevertheless, recent evidence that profilaggrin has a cal­ As already described, stratum corneum lipids are not only essential cium-binding domain [76] and that a certain stage of its processing is for the water barrier of the skin they also help to retain NMF within a calcium-dependent event [77] suggests that this protein may playa corneocytes to allow maximummoisturization of the outer layers of crucial role in coordinating the calcium-dependent events of termi­ the stratum corneum. As will be discussed later they also influence nal differentiation. the activity of certain enzymes within the tissue important for stra­ The role of filaggrin (summarized in Fig 7) does not, however, tum corneum maturation and desquamation. Variations in the levels conclude with its interaction with keratins to form the fiber/matrix and types of lipids present in the stratum corneum, therefore, can composite seen in the lower stratum corneum. Both radiolabel pulse influence stratum corneum barrier function, water content, and skin chase [62] and immunohistochemical [78,79] studies have shown condition. This section of the review will outline the recent knowl­ that filaggrin does not persist beyond the lower layers of the stratum edge of the biologic variation in stratum corneum lipids in human corneum. Approximately 2 - 3 d after its formation from profilag­ skin. grin, filaggrin, having been further enzymically modified by pepti­ Stratum corneum lipids are known to be influenced by genetic dyl deiminase, to reduce its interaction with keratin [73], is abruptly variation, ageing, dietary influences, seasonal effects, and environ­ and completely proteolyzed and certain of its constituent amino mental factors. These variables usually lead to reductions in the acids further metabolized [80-82] to form the NMF. It has only levels of stratum corneum lipids and pre-dispose to severe dry skin been in recent years that the control of this final process has been conditions. For instance, changes in the levels of stratum corneum investigated. It has proved to be a fine example of elegant biologic lipids are associated with several hereditary disorders (for review see control. [83]). The most well described condition is recessive X-linked ich­ The skin faces a genuine dilemma in producing NMF. To be thyosis (RXLI) in which there is a specific abnormality in sterol effective, a relatively large amount has to be present inside the metabolism [83). In contrast to the other ichthyotic states, subjects corneocyte to absorb sufficient water in the rather dry condition of with RXLI show excessively high levels of cholesterol sulphate in the outer stratum corneum. Such a high concentration of low­ the stratum corneum due to a deficiency of the enzyme steroid molecular-weight species would, however, have serious conse­ sulphatase. Although the stratum corneum lipid profiles in other quences for a cell deeper within the epidermis as it is exposed to an ichthyotic diseases have not been fully determined, reduced levels aqueous enviro~ment. The res~lting osmotic pressure wo~ld result of sphingosine have been found in a variety of subjects with vari­ in large quantities of water belI1g taken up and the cell disrupted. ous ichthyoses [84]. This could lead directly to cellular hyperpro­ The skin avoids the osmotic implications of producing NMF in its liferation in these conditions as sphingosine has been proposed viable, water-rich cells by using an osmotically inactive protein to feedback to the epidermis and down-regulate keratinocyte turn­ precursor, profilaggrin. over [85]. The stage at which filaggrin is converted to NMF appears to be Recently the biochemical analysis of stratum corneum of subjects just after the epidermal water permeability barrier has been formed with psoriasis has been performed [86]. Whereas cholesterol levels within the stratum compactum (l owermost part of the stratum cor­ were raised, fatty acids and ceramides, particularly ceramide sub­ neum). This avoids the cells being exposed to a water activity ap­ types 4, 5, and 6, were decreased. It has also been reported that the proaching 1.0 (the thermodynamic activity of pure water), which composition of their covalently-bound lipids differ compared with would produce an osmotic pressure of several hundred pounds per healthy stratum corneum [87). Decreases in the levels of Ceramide square inch. The skin must, therefore, not activate its protease sys­ B and increases in the w-hydroxy fatty acids together with free fatty tems until the corneocyte has moved far enough out into the dryer acids, particularly the levels of covalently-bound oleate and lino­ areas of the stratum corneum to be able to handle the osmotic effects leate, have been reported in the corneocytes of psoriatic subjects. of the NMF. Another way of looking at this is that the corneocyte Atopic dermatitis is also associated with abnormalities in stratum must not produce the NMF until it really needs it to prevent loss of corneum li pids and barrier function. Analysis of ceramides in these its liquid water from its outermost layers, thus maintaining a degree subjects was initially performed by the careful work ofImokawa and of suppleness for this region and enabling desquamatory enzymes to team [88]. Other investigators have now confirmed this [89] and function in a semi-aqueous environment. more recently it has been demonstrated that these subjects have However, a key piece of understanding was still required. The particularly low levels of Ceramide 1 linoleate [90]. Thus, a reduc­ environment of our skin is not constant-variable humidity, wind, tion in barrier lipids, particularly Ceramidc 1 linoleate, could ac­ sun exposure, etc. The skin, therefore, has to be able to control count for the abnormalities in stratum corneum function of these NMF levels, a problem it has solved by linking the activities of its subjects. Abnormalities in lamellar granule structure have been filaggrin-degrading systems to the water activity in the corneocyte. shown in atopic individuals [91], which could be directly due to a In vitro experiments [79] on isolated stratum corneum containing reduction in linoleoyl-acylglucosyl ceramide. Alternatively, how­ freshly synthesized filaggrin show that conversion of filaggrin to ever, modified epidermal differentiation may be the underlying NMF only occurs between water activities of 0.7 and 0.95. At problem as this molecule has been shown to influence corneocyte higher water activities filaggrin is stable; at lower water activities envelope formation ill vitro [92] . the whole tissue is too dry to allow hydrolytic enzymes to operate. The relationships between stratum corneum lipids and different In vivo experiments show the consequences of this. If the skin is racial groups are still poorly understood. Comparing stratum cor­ occluded for a long period [59] filaggrin conversion to NMF is neum lipids in several racial groups, Sugino et al [93] recently re­ blocked and the NMF level of the stratum corneum falls close to ported that stratum corneum ceramide levels were lowest in African 736 RAWLINGS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Americans compared with other racial types. This could imply this Total degradation particular ethnic group is more susceptible to stratum corneum of desmosomes and Surface layers - - of problems. lipid structure - -- - It is widely experienced that as we age we suffer from more skin stratum corneum problems. Although the skin problems arise from many factors, age-related reductions in the levels of stratum corneum lipids have Desmosome been shown in Japanese stratum corneum, particularly in the levels degradation and • of ceramides and fatty acids by Imokawa and group [88]. These v-- encapsulation ~~~=== ... ~ changes are probably related to poor differentiation rather than lipid by lipids extraction as a reduction in the number of epidermal lamellar gran­ ~ ules occurs with advancing age. Ceramide subtypes have also been t Intact desmosome reported to change with age in Japanese women [94]. Surprisingly, Lower layers increases in Ceramides 1 and 2 but decreases in Ceramides 3 and 6 normal lipid _ were found going from pre-puberty to adulthood. These changes configuration stratum o~orneum were not seen in men. In Caucasian populations, however, it was reported by Leveque and team [95] that, with increasing age, leg Figure 8. Schematic of stratum corneum showing events leading to stratum corneum was selectively depleted of sterol esters and tri­ desquamation. glycerides but not ceramides. More recently, measuring mass levels of lipids, Rawlings et at have shown that all stratum corneum lipid levels diminished with increasing age on the face, leg, and hand skin of Caucasians [96,97]. However, the relative levels of the ceramide together with the mechanical and desquamatoty properties of the subtypes did not change. Thus, at least for Caucasian and Japanese stratum corneum, changes in their levels or composition can lead to populations, stratum corneum lipids appear to be reduced with age­ disturbances in stratum corneum functioning. ing. These changes are expected for other racial groups. The likely NEW INSIGHTS INTO DESQUAMATION cause of age-related decline in lipid levels is probably due to the reduced epidermal lipid biosynthesis capabilities recently reported Terminal differentiation of the epidermis eventually leads to des­ by Ghadially et af [98] in murine epidermis. quamation, a process in which corneocytes are lost invisibly from It is also well known that skin condition deteriorates in cold, dry the surface of the stratum corneum [105]. Mechanistically this pro­ weather. Yet, despite this, there have been vety few studies examin­ cess is poorly understood. However, the enzymatic degradation of ing the seasonal influences on stratum corneum lipids. One study inter-corneocyte linking structures, or a reduction in inter-corneo­ has shown a general decrease in epidermal cerebrosides in winter cyte binding forces, must occur to allow for cell shedding. For many compared with summer [99]. More recently, Rawlings and co­ years, lipids [106] have been considered as cohesive structures workers have analyzed stratum corneum lipids from the face, hand, within the stratum corneum but there is now growing evidence that and leg skin of female Caucasians in the winter, summer, and spring other structures are also involved; namely, lectins and particularly months of the year [96]. Like the effects of aging, decreases in all desmosomes. major lipid classes were seen on all body sites during the winter Direct evidence for stratum corneum lipids being involved in ce ll months of the year. Although the levels of ceramide subtypes were cohesion has been provided by corneocyte reaggregation studies ill unchanged, the amounts of linoleate esterified to Ceramide 1 were vitro. Numerous workers have reaggregated previously dispersed reduced. corneocytes in the presence of stratum corneum lipids and found the It is apparent from these studies that many factors influence the physical properties of the reconstituted stratum corneum lipid-cell levels and types of stratum corneum lipids and it is possible that a films to be similar to the intact tissue [107,108]. However, Chap­ reduction in stratum corneum lipids leads directly to poor skin con­ man e/ al [109] suggested lipids actually may have an anti-cohesive dition. For instance, lower levels of stratum corneum lipids will be role preventing close opposition of adjacent corneocytes. In these more susceptible to extraction (e.g., washing) or perturbation of studies, when stratum corneum lipids were completely extracted, their structural organization that could lead to abnormalities in the inter-corneocyte forces were dramatically increased and stratum stratum corneum function and overall skin condition. Yet despite its corneum cells became tightly opposed. Close apposition of corneo­ prevalence, winter xerosis is still poorly understood. Nevertheless, a cytes occurs in both normal and abnormal stratum corneum [110]. picture is emerging that lipids influence the expression of this com­ Taken together these observations indicate that both the intercellu­ mon problem. Lifids are easily extracted from the stratum corneum lar and covalently-bound lipids may have a role in stratum corneum by so lvents [100 and surfactants [101] leading to their depletion integrity. A specific intercellular lipid thought to be involved in from the intercellular spaces of the stratum corneum. In acute treat­ stratum corneum cell cohesion is cholesterol sulphate [83]. These ment studies, this extraction leads to changes in the relative amounts findings indicate that lipid interactions must be reduced prior to of the different lipid species in the outer layers of the skin due to desquamation. In fact, using electron microscope techniques, Elias selective lipids being removed. During chronic treatments, how­ et af [47,111] have demonstrated changes in the structure of stratum ever, differences in stratum corneum lipid composition, but not corneum lipids during stratum corneum maturation and, more re­ total lipid levels, have been reported [102,103]. In these studies cently, Rawlings et af have shown that the classical lipid bilayer increases in free cholesterol, Ceramide subtypes 2 and 4, were ob­ structures seen in the lower layers of the stratum corneum disappear served, whereas cholesterol esters, long-chain fatty acids, and Cera­ in the outer layers [49,50]. These observations are supported by the mide 3 all decreased in concentration. More recently, similar findings of others demonstrating more fluid lipids in the peripheral changes in stratum corneum ceramide profiles have been reported in layers of the stratum corneum using spectroscopic techniques indi­ other experimentally-induced scaly skin, such as via tape stripping cating that disruption of stratum corneum lipids is occurring in the and surfactant treatment, indicating that the changes in lipid com­ outer layers of the stratum corneum. It is not yet clear how this position are related to changes in epidermal lipid biosynthesis rather occurs. Ceramides, particularly Ceramide 1, may be hydrolyzed by than lipid extraction. In contrast to these studies, the total levels of ceramidases [38,112] or surfactant-like sebaceous fatty acids may stratum corneum ceramides are decreased and the levels of fatty lead to bilayer disruption. However, lipid bilayers are reported to be acids are increased [104] in winter xerosis [49,50] and, unlike RXLI, present in desquamated epidermal cyst corneocytes [113]. The ef­ cholesterol sulphate levels are normal. fect of the covalently bound lipid fraction in desquamation is not yet Thus, stratum corneum lipids show marked biologic variation. As known. lipids influence the permeability barrier to water and NMF loss Cell-surface glycoproteins have also been implicated in stratum VOL. 103, NO.5 NOVEMBER 1994 STRATUM CORNEUM MOISTURlZATION 737

Intact desmosome, lipid I.II!I·.'~ ABNORMAL DESQUAMATION: COMPARISON OF disruption, and imbalance ...... -, -- Surface layers ICHTHYOSES AND XEROSIS in lipid composition stratum °dorneum leading to abnormal ~ The ichthyoses and xerosis are clear examples of abnormal processes SC barrier of desquamatIon. Here the accumulation of visible scales or corneo­ t cyte clumps occurs on the surface of the skin. Although the precise causes of skin xerosis are likely to be multifactorial [132], the result­ Intact desmosome ant scaling is probably a consequence of perturbed degradation of and lipid disruption the corneocyte cohesive elements leading to aberrant desquamation. However, the relative contribution of changes in the composition and levels of lipids, lectins, and desmosomes to the disordered des­ Intact desmosome quamatory process is poorly understood. N evertheless, it is becom­ and normal lipid ing apparent that a lac k of des mosome degradation is common to all configuration these disorders. Several investigators [133 - 135] have now shown t~1at defects in desmosome catabolism occur in ichthyotic condi­ Figure 9. Schematic of stratum corneum showing abbe ration of tIOns, whIch directly leads to scaling and the accumulation of cor­ lipid and desmosome structure and degradation in xerosis. neocyte clumps on the surface of the skin. Desmosome retention a l ~o oc~urs in the surface layers of the stratum corneum in subjects WIth WlI1ter xerosIs [49,50] . In this respect, winter xerosis has been corneum cell cohesion [114]. One specific glycoprotein lectin, des­ shown to share similarities to these genetic, hyperkeratotic and quamin [115,116], a 40-kDa glycoprotein resistant to proteolytic scaling disorders. ' degradation [117] and localized to the corneocyte envelope, was In paral.leI w!th the effects on desmosome degradation, structural shown to be an endogenous stratum corneum lectin that has speci­ abnormalttles 111 stratum corneum lipid lamellae also occurs in ficity for amino sugars. Antibodies directed against desquamin pre­ wint~r xerosis and ichthyosis. Abnormalities in lipid structure are vented the reaggregation of dispersed corneocytes ill vitro, indicat­ seen 111 the outer layers of the stratum corneum in winter xerosis ing a potential cohesive role for this molecule. It is highly likely that (summarized in Fig 9) [49,50]. In contrast, abnormalities in lipid desquamin will not only playa role in intercorneocyte cohesion at ultrastructure ~ re seen throughout all the cellular layers of the stra­ the desmosomal linkages but also in lipid-depleted zones. tum corneu.m 111 autosomal recessive ichthyosis [133]. The former Due to the reports of extensive desmosomal degradation in the probably ames as a result of superficial damage to the stratum cor­ mid to outer regions of the stratum corneum [118,119], early inves­ neum and the latter from aberrations in differentiation. Although tigators largely ignored the involvement of desmosomes in cell des mosomes largely. lI1fluence . corneocyte cohesion [109,136], cohesion and desquamation. However, there is now mounting evi­ changes 111 . the phYSical properti es of the stratum corneum lipids dence that desmosomes persist into the outer layers of the stratum may als~ 1I1fluence mter-corneocyte cohesivity properties and corneum before their fin al degradation in the surface layers. Several th~reby ll1f1uence desquamatIOn. However, the precise relation­ investigators have now found that desmosomes, or their Inarker ships of these structural changes in lipid architecture and faulty proteins, are gradually degraded towards the surface of the stratum desquamation remain to be determined. N evertheless, two mecha­ corneum [49,50,111,119-121]. nisms are most likely operating. It is clear from these studies that stratum corneum lipids and In win.ter xerosis the li pid str u ~tura l abnormalities probably lead desmosomes are being degraded by enzymes in preparation for des­ to defectl~e desmosome degradatIOn and aberrant desquamation by quamation (summarized in Fig 8). Phospholipases and glucosylcer­ adversely 1I1fluencUlg the activities of hydrolases already present in ebrosidases hydrolyze their respective target lipids, allowing stra­ the stratum ~o rneum [47,123 - 130]. Desmosomal degradation is tum corneum maturation [47]. However, only the degradation of known to be 1I1fluenced by changes in water content of the stratum cholesterol sulphate has been specifically linked with the loss of corneum [131]. [,1 vivo, changes in stratum corneum water content surface corneocytes and desquamation [122]. Serine proteases have may occur as a result of the changes in stratum corneum lipid struc­ also been implicated in the process of the proteolytic degradation of ture ~nd loss of NMF, which may then affect enzyme activities as desmosomes, especially chymotrypsin-like [123 - 126] and possibly descnb ~ d abo~e: Enzyme activity may also be reduced by phase trypsin-like proteases [127]. One enzyme in particular, the stratum sepa~at1o~ of. hp~ds tnduced by the changes in lipid composition or corneum chymotryptic enzyme (SCCE), [126] has been proposed as by direct 11llubltion of the enzymes. The imbalance in the levels of a desquamatory protease due to its inhibition profile being similar to ceramides and fatty acids may be the true problem in winter xerosis that of desmosomal and Desmoglein 1 (a major desmosomal glyco­ [49,50]. Fatty acids pertur~ lipid membranes [137] and may explain protein) degradation together with cell loss ill vitro [123-126] . the structural abnormalities seen 111 these subjects. Although still There is also some evidence indicating that the desmosomal glyco­ co ~ ect ural, these changes may then lead to reduced activity of the proteins must be deglycosylated before efficient proteolysis can pro­ desquamatory enzymes. ceed [128 - 129]. As the stratum corneum desquamatory enzymes Although not excluding the changes in lipid-phase behavior in are believed to be extracellular [130), the enzymes mediate their the ichthyotic conditions that may affect enzyme activiti es as de­ action in a lipid matrix. Stratum corneum lipid-phase behav ior, scribed above, reduced des mosome hydrolys is may also occur from therefore, will influence enzymatic activity. Cholesterol su lphate the defective delivery of lamellar granule derived desquamatory may indirectly influence protease activity by influencing the type of enzymes as has been proposed by Menon et af [138]. In some cases a lipid phases present within the stratum corneum or by direct inhibi­ total deficiency of desquamatory enzymes may occur as in the case tion of the desquamatory enzymes as proposed by Williams [83]. of RXLI. The genetic variation in the putative desquamatory pro­ Similar to the enzymes involved in fil aggrin degradation, the stra­ tease types and levels is still not known but as enzymes playa key tum corneum desquamatory enzymes are probably also influenced role 11l the desquamatory process, several investigators are currently by environmental humidities. Recently, we have shown that des­ trying to es tablish their contribution to xerosis and ichthyosis. mosomal degradation, and therefore the desquamatory enzymes, are inhibited at low environmental humidities [131). This finding CONCLUDING REMARKS indicates that the maintenance of the water content of the stratum The complexity of the stratum corneum has unfolded over several corneum is vital for the normal orderly process of cell loss from the decades. Originally thought to be several layers of dead cells, it is surface of the skin. Stratum corneum lipids and NMFs will both now considered to be a highl y heterogeneous and elegant structure play a role in ac hieving this. whose main role is to protect terrestrial animals from desiccation 738 RAWLINGS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY and environmental challenge. Over a decade ago, Peter Elias [106] Undoubtedly, the molecular control mechanisms of desquama­ proposed a heterogeneous, two-compartment, "bricks and mortar" tion, desmosome, and lipid hydrolysis together with filaggrin deg­ model, in which the "bricks" are the corneocytes and the "mortar" radation will be elucidated over the coming years, but we know that is specialized lipids. This model now requires some modifications as these will be intimately associated with the water activity of the a result of considerable ongoing research from six main laboratories, stratum corneum. Little did Irvin Blank [145] suspect, when he including our own, to take into account the importance of stratum pointed out over four decades ago that water was important for corneum moisturization. Osmotically active chemicals present stratum corneum moisturization and plasticization, that the mecha­ within the corneocytes allow the "bricks" to act as "sponges" and nisms for water retention would be as complex and interlinked as swell , the network of bricks has its structure reinforced by desmo­ they are proving to be, and that water activity control would also somes, and a variety of enzyme systems are present that are activated turn out to be vital for specific enzyme reactions allowing optimal during different stages of stratum corneum maturation. stratum corneum maturation and function. It is now clear that the stratum corneum undergoes a series of alterations during its journey from the lower layers of the stratum compactum to the outer layers of the stratum disjunctum, with each REFEREN CES modification contributing to the overall properties of the stratum 1. Burr GO, Burr MM: A new deficiency disease produced by rigid exclusion offat corneum. It is now known that these modifications are important from the diet.) Bioi Ghem 82:345 - 367, 1929 for the three main functions of the stratum corneum: its barrier, 2. Burr GO, Burr MM: The nature and role of fatty acid in nutrition.] Bioi Gilem 86:587-621,1930 mechanical, and desquamatory properties [139] but as this review 3. BaSllayake V, Sinclair HM: The effect of deficiency of essential fatty acids upon has pointed out these functions are dependent on specific biochemi­ the skin. In: Popjak G, Le Breton E (cds.). Biochemistry oJLipids. Buterworth, cal transformations. The barrier properties of the stratum corneum London, 1956, pp 476-484 are highly dependent upon the levels and types of lipids within the 4. Dc Thomas ME, Branner RR, Peluffo RD: Position of eicosatrienoic acid in phosphatidyl choline and phosphatidyl ethanolamine &om rats in essential stratum corneum but superimposed upon this lipid barrier is its fatty acid deficiency. Biochim Biopllys Acto 70:472- 478, 1963 tortuous structure, which also contributes to permeability proper­ 5. Prottey C: Essential fatty ac ids and the ski n. Br) Dermatol 94:579-585, 1976 ties [140]. Changes in lipid composition also lead to changes in the 6. Prottey C: Investigations of fu nctions of essential fatty acids in the skin. Br) properties of lipids within the stratum compactum and stratum dis­ DermatoI97:29 - 38,1977 7. Prottey C, Hartop PJ, Press M: Correction of the cutaneous manifestations of junctum, which may influence the barrier properties of these two essential fatty ac id deficiency in man by application of sunflower seed oil to the tissue regions. However, the tightly coherent layers of the stratum skin.) bIVcst DcrmatoI64:228-234, 1975 compactum, due to the presence oflarge numbers of non-peripheral 8. Houtsmuller UMT: Effects of topical application of fatty acids. Prog Lipid Res desmosomes, certainly contributes to structuring this region prop­ 20:219 -224,1981 9. Houtsmuller UMT, Van der Beck A: Columbinic acid; a new type of essential erly. In addition, one should not ignore the influence of corneocyte fatty acid. Prog Lipid Res 20:889-896,1981 size, which enlarges with age [141] and during the winter months 10. Winsor T, Burch GE: Differential roles of layers of human epigastric skin on [142] and probably influences the barrier properties of the stratum diffusion of water. Arch Illtem Med 74:428 - 444,1944 corneum. 11. Matoltsy AG, Downes AM, Sweeney TM: Studies of the epidermal water barrier. Part II. Investigation of the chemica l nature of the water barrier.] The barrier properties of the stratum corneum also influence the b",est. DermatoI50:19-26, 1968 mechanical properties of the stratum corneum. However, a factor 12. Middleton JD: The mechanism of water binding in stratulU corneum. Br] largely ignored in this function is the NMF. Without the NMF, Derlllatol 80:437 -450, 1968 moisturization of the stratum corneum would not occur effectively 13. Bowser PA, White RJ: Isolation barrier properties and lipid analysis of stratum compactulTI, a discrete region of the stratum corneum. Br J Dermatol112: 1- in the outer layers of the stratum corneum, which could lead to its 14,1985 mechanical failure, e.g., cracking. Stratum corneum lipids, particu­ 14. Gray GM, Yardley HJ: Lipid compositions of cell isolated from pig, human, and larly the covalently-bound lipids, probably help to retain the NMF rat epidermis. ) Lipid Res 16:434 - 440, 1975 within the corneocytes and potentiate its moisturization capabili­ 15. Gray GM, White RJ, Majer JR: 1-(3-0-acyl)-beta-glucosyl-N-dihydroxypenta­ triacontadienoylsphingosine, a major component of the glucosylceramides of ties. Although not greatly appreciated, effective moisturization of pig and human epidermis. Biochilll Biophys Acta 528:127 -137, 1978 the stratum corneum also helps to maintain the barrier of the stra­ 16. Elias PM, Friend DS: The pemeability barrier in human epidermis.) Cell Bioi tum corneum. We have recently shown that stratum corneum lipids 65:180- 191,1975 become ruptured during mechanical extension of the stratum cor­ 17. Elias PM, McNutt NS, Friend DS: Membrane alterations during cornification of mammalian squamous epithelia: a freeze fracture tracer and thin section study. neum leading to barrier dysfunction [136]. This structural abnor­ Allot Rec 189:577 - 594,1977 mality occurs more readily when the stratum corneum is dehy­ 18. Elias PM, Brown BE: The mammalian cutaneous permeability barrier: defective drated. barrier function in essential fatty acid deficiency correlates with abnormal The replacement of damaged cells from the surface of the skin intercellular lipid deposition. Lob b",est 39:574-583,1978 19. Bowser PA, Nugteren DH, White Rj, Houtsmuller UMT, Proctey C: Identifi­ with fresh undamaged corneocytes is a mechanism that ensures cation, isolation and characterization of epidermal lipids containing linoleic continual protection of the epidermis. The desquamation process is acid. Biochilll Biophys Acto 834:419 - 428, 1985 now known to involve the degradation of surface lipid and protein 20. Nugteren DH, C hrist-HazelhofE, van der Beck A, Houtsmuller UMT: Metab­ species in the stratum corneum by lipases, glycosidases, and pro­ olism of linoleic ac id and other essential fatty acids in the epidermis of the rat. Biochilll Biopllys Acta 834:429-436,1985 teases. Like the enzymes involved in filaggrin degradation, the hy­ 21. Wertz PW, Downing DT: Glycolipids in mammalian epidermis: stmcture and drolases involved in desmosome degradation (and probably lipid function in the water barrier. Sciell ce 217: 1261 - 1262, 1982 degradation) are clearly dependent upon water for their activity. 22. Wertz PW, Downing DT: Acylglucosylceramides of pig epidermis; stmctu.re However, water activity within the stratum corneum is dependent determination.) Lipid Res 24:753 - 758,1983 23. Wertz PW, Downing DT: Ceramides of pig epidermis: structure determina­ upon the NMF and lipids, factors that, as this review has tried to tion.] Lipid Res 24:759-765,1983 point out, are influenced by many events. When adversely in­ 24. Wertz PW, C ho EI, Downing DT: Effect of essential fatty acid deficiency on flu enced, insufficient stratum corneum moisturization and water the epidermal sphingolipids of the rat. BiociJim Biophys Acta 753:350-355, content leads to defective desquamation. 1983 25. Wertz PW, Downing DT, Freinkel RK, Traczyk TN: Sphingolipids of the Many skin-care companies are now exploiting this knowledge stratum corneum and lamellar granules of fetal rat epidermis.] b",cst Dermatol usi ng specialized lipids and hygroscopic substances to affect stratum 83:193 - 195, 1984 corneum moisturization for the treatment of scaling disorders [143]. 26. Abraham W, Wertz PW, Downing DT: Linoleate-rich acylglucosylceramides It is interesting in this respect that ingredients in skin-care products of pig epidermis: structure determination by proton magnetic resonance.] Lipid Res 26:761 - 766, 1985 such as polyols (e.g., glycerol) not only act to improve stratum 27. Bowser et a/: Cosmetic composition for topical application to skin containing corneum flexibility and prevent lipid crystallinity [144], they also long chain fatty acid derivative (abstr). Patellt No. GB 2126:892A, 1982 improve the rate of desmosome hydrolysis probably through a com­ 28. Hedberg CL, Wertz PW, Downing DT: The time course of synthesis of epider­ bination of occlusivity, humectancy, and lipid-phase modulating mal lipid.] Illvest DermatoI91:169 - 174, 1988 29. Long SA, Wertz PW, Strauss JS, Downing DT: Human stratum corneum lipids properties. and desquamation. Arch Dermatol Res 277:284-287, 1985 VOL. 103. NO.5 NOVEMBER 1994 STRATUM CORN EUM MOISTURIZATIO N 739

30. Wertz PW. Downing DT: Glucosylceramides of pig epidermis: structure deter­ granules. Source of the free amino acids. urocanic acid pyrrolidone carboxylic mination.] Lipid Res 24: 1135-1139.1983 acid in mamma.lia n stratulll corneum. Bioe"im Biol'''Ys Acta 719:110-117, 31. Wertz PW. Miethke MC. Long SA. Strauss JS. Downing DT: Composition of 1982 ccramiclcs from human stratuIll corneum and comedones. ] !til/cst Dcrm atol 64. Laden K. Spitzer R: The identification of a natural moisturising agent in skin.] 84:410-412. 1985 Soc Cosmel C!.em 18:351- 360. 1967 32. Colarow LJ: Quantification of ceramides with essential fatty acid moieties in 65. AngelinJH: Uroc,"icacid a natural sunscreen. Casmeticsa lld To iletries 91:47 -49. the human skin surface and blood plasma lipids. Platlar Chrom 3: 126 - 132. 1976 1992 66. DeFabo EC. Noonan FP: Mechanism of immune suppression by ultraviolet 33. Hamanaka S. Asayami C. Suzuki M. 1nagaki F. Suzuki A: Structure determina­ irradiation in vivo. J Exp Med 157:84 - 98. 1983 tion of glycosyl B-I -N (w-o-linoleoyl)acylsphingosines ofi1Uman epidermis. 67. Ugel AR, Idler W : Further characterisa tion of bovine keratobyalin.) Cell Bioi ] Biochem 105:684-690. 1989 52:453 - 464. 1972 34. Wertz PW. Downing DT: H ydroxyacid derivatives in human epidennis. Lipids 68. Ball RD. Walker GK. Bernstein LA: Histidine-rich proteins as molecular 23:415-418. 1988 markers of epidermal differentiati on.) Bioi C!. em 254:5861 - 5868. 1978 35. Wertz PW. Downing DT: Free sphingosine in porcine epidermis. Biochem 69. Balmain A. Loehren D. Fischer J . Al onso A: Protein synthesis during the fetal Biop"ys Acta 1002:213-217. 1989 devel opment of mouse epidermis. I. T he appearance of " histidine-rich pro­ 36. Wertz PW. Downing DT: Free sphingosine in human epidermis. ] ltlllest Der­ te in." D,,, Bioi 60:442-452. 1977 malo I94:1 59- 161.1990 70. Dale BA. Line SY: Evidence of a precursor form of stratum corneum basic 37. Wertz PW. Downing DT: Ceramidase activity in porcine epidermis. FEBS Lett protein in rat epidermis. Bioc!'emistry 18:3539-3545. 1979 268: 110- 112. 1990 71. Murozuka T. Fukuya ma K. Epstein WL: Immunochemical comparison ofhisti­ 38. Gupta AK. Fisher GJ. ElderJT. et 01: Sphingosine inhibits phorbol ester-induced dine-rich protein in kerato hya li n granules and cornified cells. Bioc"im Bio!,"),s inflammation. ornithine decarboxylase activity and the action of prorein ki­ Acta 579:334-345. 1979 nase C in mouse skin.) Itl vest DermatoI91:486-491. 1988 72. Lonsdale-Eccles JD. Haugen JA. Dale BA: A phosphorylated keratohyalin-der­ 39. Ahern DG. Downing DT: Inhibition of prostaglandin biosynthesis by eicosa- ived precursor of epidermal stratum corneum basic protein.] Bioi C!.em 5.8.11.14-tetraynoic acid. Bioc"im Biop"ys Acta 2 10:456-461. 1970 255:2235 - 2238. 1979 40. Hansson G. Malmsten C. Radmark 0: In: Pace-Asciak C . Granstrom E (cds.). 73. Harding C R. Scott IR: Histidine-rich proteins (filaggrins). Structural and func­ New COrllpre"etl sive Bioc"emistry, V615. Prostaglatldills alld Related Substa ll ces. tional heterogeneity during epidermal differentiation.] Mol Bioi 170:651 - Elsevier. Amsterdam. 1983. pp 155- 157 673.1983 41. Tobias LD. H amilton J G: The effect of 5.8.11.14-eicosatetraynoic acid on lipid 74. Dale BA. Holbrook K. Steinert PM: Assembl y of stratum Corneum basic protein metabolism. Lipids 14:181- 193. 1979 and keratin filaments in macrofibrils. Nalllrc (Lolldoll) 276:729-731. 1978 42. Miller CC. z iboh VA: Induction of epidermal hyperproliferation by topical n-3 75. Steinert PM. CantieriJS. Tell er DC. Lonsdale-EcclesJD. DaleBA: C haracteri­ polyunsa turated fatty acids on guinea pi g skin linked to decreased levels of sa tion of a class of catio nic protein s that specificall y interact with intermed iate 13-hydroxyoctadecadienoic acid.] Ill vest Derm atol 94:353 - 358. 1990 filaments. Proc Natl Acad Sci 78:4097-4101. 1981 43. Swartzendruber DC. Wertz PW. Kitko OJ. Madison KC. Downing DT: Evi­ 76. Presland RE. Haydock PV. Fleckman P. Nirunsuksiri W. Dale BA: C haracter­ dence that the corneocyte has a chemically-bound lipid envelope. ) Jtlvest ization of the human epidermal profil aggrin gene. Genomic organization and DermatoI 88:709-713.1987 identification of an S-100 like calcium binding domain at the amino terminus. 44. Wertz PW. Madison KC. Downing DT: Covalently bound lipids of human ] Bioi Chcm 267:23772-23781. 1992 stratum corneum.) Ill vest Dermatol 92: 109 - 111 . 1989 77. Resing KA . AI-Alawi N . Blomquist C. Fleckman p. Dale BA: Independent 45. Serizawa S. Ito M. Hamanaka S. Otsua F: Bound lipids liberated by alkaline regulation of rwo cytoplasmic processing stages of the intermediate filament­ hydrolysis after exhaustive extraction of pulverised clavus. Arch Derm atol Res associated protein fila ggrin and role of calcium in the second stage.) Bioi Chcm 248:472-475. 1993 268:25139 - 25 145. 1993 46. Gray GM. Yardley HJ: Different populations of epidermal cell s: isolation and 78. Scott I: Alterations in the metabolism of fila ggrin in the skin after chemical and lipid composition.) Lipid Res 16:441-448. 1975 ultraviolet induced erythema.) blllCSi DermatoI87:460-465. 1986 47. Elias PM. Menon GK. Grayson S.Brown BE: Membrane structural alterations in 79. Scort lR. Harding C R: Filaggrin breakdown to water binding compounds dur­ murine stratum corneum. Relationship to the localisa tion of polar lipids and in g development of the rat stratum corneum is controll ed by the water activity phospholipases. ] [livest DermatoI 91:3-10. 1988 of the environment. Dc" Bioi 11 5:84-92. 1986 48. Swartzendruber DC. W ertz PW. Ki tko OJ. Madison KC. D owning DT: Mo­ 80. Scott IR. BarrettJ G: Pyrrolidone carboxylic acid synthesis in guinea pig epider­ lecular models of the intercellular lamellae in mammalian stratum corneum.) mis. ] b lllest DcrmatoI81: 122-124. 1983 Ill vest Dermatol 92:251-257. 199 1 81. Delapp NW. Dieckman OK: Gamma glutamyl pepidase: a novel enzyme from 49. Rawlings A V. Hope J . Rogers J. Mayo AM. Scott IR: Mechanisms of desquama­ hai rl ess mouse epidermis.) b lllest Dermato I 90:490-494. 1988 tion: new insights into dty fl aky skin conditions. Proc 17t" IFSCC 2:865-880. 82. Scott IR: Factors controlling the expressed activity ofhistidillc ammonia lyase in 1992 the epidermis and rhe res ulting acculllulation of urocanic acid . Bioc!'em ) 50. Rawlings AV. Hope J. Rogers J. Mayo AM. Watkinson A. Scott IR: Skin 194:829-838. 1981 dryness-what is it? (abstr).] III vest DcrmatollOO:510. 1993 83. Williams ML: Lipids in normal and pathological desquamation. In: Elias PM 51. Bowser PA. White RJ. Nugteren DH: Location and nature of the epidermal (cd.). Advallces ill Lipid Research, Vol. 24. Academic Press. San Diego. 1991. pp permeability barrier. lilt] Cosmet Sci 8:125- 134. 1986 2 11 - 262 52. White R. W alker M: Thermotropic and lyotropic behav iour of epidermal li pid 84. Paige DG. Morse-Fisher N. H a.rper JI: T he quantification offree sphingosine in fractions. Bioe/1fI1I Soc Trotl s 18:881-882. 1992 the stratum corneulll of patients with hereditary ichthyosis. Br J Dermatol 53. Feingold KR: The regulation and role of epidermal lipid synthesis. In: Eli as PM 129:380 -383, 1993 (ed.). A d"al lCes ill Lil'id Rescore", Vol. 24. Academic Press. San Diego. 1991. pp 85. Wertz PW: Epidermal li pids. Semill Dermatol 11 :106 - 11 3. 1992 27-56 86. Motta S. MeUesi S. Sesana M: Study of interl amellar lipids in psoriatic scales 54. Man M. Feingold KR. Elias PM: Exogenous lipids influence permeability barrier (abstr). 18t!' World COllgress oj Dermatology 159. 1992 recovery in acetone-treated murine skin. Are!' Dermatol 129 :728-738. 87. W ertz PW. Madison KC. Downing DT: Covalently bound li pids of human 1993 stratum corneum.) b,,'cs t Dermatol 92: 109 - 111 . 1989 55. T abachnick J. Labadie JH: Studies on the biochemistty of epidermis. IV . The 88. Imokawa G. Abe A.Jin K. Higaki Y. Kawashima M. Hidano A: Decreased levels free amino acids. ammonia. urea and pyrrolidone carboxylic acid content of of ccramidcs in stratum Co rneum of atopic dermatitis: an etiologic factor in conventional and germ free albino guinea pig epidermis.) [liVest Dermatol atopic dry skin.) blllw Dermatol 96:523 - 526. J 99 1 54:24-31. 1970 89. Melnik B. Hollman J . Hoffman U, Yuh MS. Plewig G: Lipi d composition of 56. Marstein S. Jellum E. Eldjarn L: The conce ntration of pyroglutamic acid (2- outer stratum COrneum and nails in atopic and control subjects . A rcll Dermalo/ pyrrolidone-5-carboxylic acid) in normal and psoriatic epidermis. determined Res 282:549 -55 1. 1990 on a microgram scale by gas chromatography. Clillica Chimica A cta 49:389 - 90. Yamamoto A, Scri zawa S, Ito M. Sara Y: Stratum corneum lipid abnornla. litics in 395. 1973 atopic dermatitis. Arc!' Dermatol Res 283:219-223. 1991 57. Sybert VP. Dale BA. Holbrook KA: Ichthyosis vulgaris: identification of a defect 91. Werner Y. Lindberg M. Forslind B: Membrane-coating granules in "dry" non­ in filaggrin synthesis correlated with an absence of keratohyalin granules. ) eczematous skin of patients with atopic dcrmacitis. Acta Derm Venereal I"" estDermatoI 84:1 91- 194. 1985 67:385-390. 1987 58. J acobsen TM. Yukscl KU. GeesinJC. Gordon J S. Lane AT. Gracy RW: Effects 92. Uchida Y. Iwamori M. N agai Y: Activarion of keratinization of kcratinocytes of aging and xcrosis 011 the amino ac id composition of human skin. ] IU llest frol11 fetal rat sk in wirh N-(linoleoy l) w-hydroxy fatty acy l sphingosyl glu­ Dermatol 95:296 - 300. 1990 cose as a marker of epidermis. Bioe/lew Bioplrys Res Cow mllll 179 :1 62- 168 . 59. Scott IR. Harding CR: Physiological effects of occlusion-filaggrin retention 1990 (abstr). Derm atology 2000:773. 1993 93. Sugino K. Imokawa G, Maibac h HI: Ethnic difrerence of varied stratum cor­ 60. Horii I. N akayama Y. Obata M. T agami H: Stratum corneum hydration and neum function in relation to stratum corneum lipids (abstr)) Ill vest Derwatol amino acid content in xerotic skin. Br] Dermatol1 21 :587 - 592. 1989 101 :482. 1993 61. DeLapp NW. Dieckman OK: Biosynthesis of pyrrolidone carboxylic acid in 94. Denda M. Koyama J . Horii I. Takahashi M. Hara M. Tagami H: Age and hairless mouse epidermis.] [li Ves t DermatoI68:293-298. 1977 sex-dependent change in stratum corneum sphingolipids. Arc!' Derm atol Res 62. Scott IR. Harding C R: Studies on the synthesis and degradation of a high 285:41 5-417. 1993 molecul ar weight, histidin e-ric h phosphopro tein fro m mammalian epider­ 95. Saint-Leger D. Francois AM . LevcqueJL. Stoudcmayer TL. GroveGL. Kligman mis. Bioc"im Biophys Acta 669:65 - 78. 1981 AM: Age-associated changes in stratulll corn eum lipids and their rela tion to 63. Scott lR, Harding C R. Barrett J G: Histidine-rich protein of the keratohyalin dryness. Dermatologi(tl 177: 159-164. 1988 740 RAWLINGS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

96. Rawlings AV, Mayo AM, Rogers J, Scott IR: Aging and the seasons inlluence 122. Ranasinghe A W, Wertz PW, Downing DT, Mackenzie I: Lipid composition of stratum corneum lipid levels (abstr).J I"vest Dernrato/lOl:483, 1993 cohesive and desquamated corneoeytes from mouse car skin.] [li Vest Dermotol 97. Rawlings AV, Rogers J , Mayo AM: Changes in lipids in the skin aging process. 86:187-190, 1985 Bioeosmet Ski" Aging 1:31-45, 1993 123. Egelrud T, Lundstrom A: The dependence of detergent-induced cell dissociation 98. Ghadially R, Brown BE, Sequeira-Martin SM, Elias PM: Structural, functional in palmo-plantar stratum corneum on endogeneous proteolysis. J blllest Der- and lipid metabolic changes in the aged epidermal permeability barrier (abstr). 111010/95:456-459, 1990 J ["vest Dennato/lOl:402, 1993 124. Lundstrom A, Egelrud T: A chymottypsin-like proteinase that may be involved 99. Nieminen E, Leikola E, Koljonen M, Kiistala U, Mustakallio KK: Quantitative in desquamation in plantar stratum corneum. Arch Dermatol Res 283:108- analysis of epidermal lipids by thin layer chromatography with special refer­ 112,1991 ence to seasonal and age variation. Acta Derm Ve"ereo/47:327 - 338, 1967 125. Lundstrom A, Egelrud T: Stratum corneum chymotryptic enzyme. A proteinase 100. Imokawa G, Akasaki S, Hattori M, Yoshizuka N: Selective recovety of deranged which may be generally present in the stratum corneum and with a possible water-holding properties by stratum corneum lipids. J ["vest Dermotol involvement in desquamation. Acto Derlllatol Vellereol (StacH) 71:471-474, 87:758-761,1986 1991 101. Imokawa G, Akasaki S, Minematsu Y, Kawai M: Importance of intercellular 126. Egelrud T: Purification and preliminaty characterisation of stratum corneum lipids in water· retention properties of the stratUin cornCUU1: induction and chymotryptic enzyme. A protease that may be involved in "desquamation. J recovety study of surfactant dty skin. Arch Dermatol Res 281 :45 - 51, 1989 Ill vest Dermato/l0l:200-204, 1993 102. Fulmer AW, Kramer GJ: Stratum corneum lipid abnormalities in surfactant-in­ 127. Suzuki Y, Nonura J , Hori J , Koyama J , Takahash M, Horii I: Detection and duced dty scaly skin. J ["vest Dermotol 86:598 - 602, 1986 characterisation of endogenous protease associated with desquamation of stra­ 103. Denda M, Hori J, Koyama J , Yoshida S, N anba R, Takahashi M, Horij I , tum corneum. Arch Dcrlllatol Res 285:372-337,1993 Yamamoto A. Stratum corneum sphingolipids and free amino acids in experi­ 128. Walsh A, Chapman Sj: Sugars protect desmosomal glycoproteins from proteo­ mentally-induced scaly skin. Arch Dennatol Res 284:363-367,1992 lysis (abstr). Br J Derlllato/122:289, 1990 104. Saint-Leger D, Francois AM, Leveque JL, Stoudemayer TK, Kligman AM, 129. Rawlings A V , Scott IR: Cosmetic composition for application to skin contai ning Grove G: Stratum corneum lipids in winter xerosis. Dermatologiea 178:151 - proteases and glycosidases. Patellt No. W09319731, 1993 155, 1989 130. Egclrud T: Stratum corneum chymottyptic enzyme evidence of its location in 105. Oldland GF: Structure of the skin. In: Goldsmith LA (cd.). Biochemistry a"d the stratum cOflleum intercellular space. Ellr J DerlllatoI2:50- 55, 1992 Physiology oJthe Ski", Vol. 1. Oxford University Press, 1991, pp 3-63 131. Rawlings AV , Hope j, Watkinson AW, Harding CR, Egelrud T: The biologic 106. Eli as PM: Epidermal lipids, barrier function and desquamation.J Illvest Dermatol effect of glycerol (abstr).] /tlllest Derlllato/l00:526, 1993 80(suppl):44-49, 1983 132. Pierard G: What docs dty skin mean? IlItJ DemlOto/26:167-168, 1987 107. Kock WR, Berner B, Burns JL, Bissett DL: Preparation and characterization of a 133. Ghadially R, Williams ML, Hou SYE, Elias PM: Membrane structure abnor­ reconstituted stratum corneum film as a model membra ne for skin transport. malities in stratum corneum of autosomal recessive ichthyoses.] bwest Dermn­ Arch Dennatol Res 280:252-256, 1988 tal 99:755 - 763, 1992 108. Smith WP, Christensen MS, Nacht S, Gans EH: Effect oflipids on the aggrega­ 134. Mesquita-Guimaraes J. X-linked ichthyosis: ultrastructural study of 4 cases. tion and permeability of human stratum corneum.] ["vest Dermatol78:7 - 11, Derlllatologica 162:157 - 166, 1981 1982 135. Blanchet-Bardon CL, Anton-Lauprecht I, Puissant A, Schnyder UW: Ultra­ 109. Chapman SJ, Walsh A,jackson SM, Friedmann PS: Lipids, proteins and corneo­ structural features of icthyotic skin in Refsum's Syndrome. In: Marks R, and cyte adhesion. Arch Dermatol Res 283:1679-1732, 1991 Dykes Pj (cds.). Thelchtl,yoses. MTP Press Ltd., Lancaster, 1978, pp 65-69 110. Hou SYE, White SH,Menon GK, Grayson S, Ghadially R,Elias PM: Membrane 136. Rawlings AV, Hope j, Ackerman C, Banks J: Hydroxycaprylic acid improves structures in normal and essential fatty acid deficient stratum corneum: char­ stratum corneum extensibility at low relative humidities and protects the acteristics by ruthenium tetroxide staining and x-ray diffraction. J IrlVest Der­ desmosomes against mechanical damage. Prot 17th IFSCC lilt COllgress 1: 3- matoI96:215-223,1991 13,1992 111. Elias PM, Menon GK: Structural and lipid biochemical correlates of the epider­ 137. McKersie BD, Crowe jH, C rowe LM: Free fatty acid effects on leakage, phase mal permeability barrier. In: Elias PM (ed.). Advances i" Lipid Research, Vol. 24. properties and fusion offully hydrated model membrane. Bioe/relll Biophys Acta Academic Press, San Diego, 1991, pp 1 - 26 982:156-160, 1989 112. Wertz PW, Downing DT: Epidermal ceramide hydrolase (abstr).] [livest Derma­ 138. Menon GK, Ghadially R, Williams ML, Eli as PM: L.mellar bodies as delivety tal 94:590, 1990 systems of hydrolytic enzymes: implications for normal and abnormal desqua­ 113. Wertz PW, Swartzendruber DC, Madison KC, Downing DT: Composition mation. Br] Dermoto/126:337-374, 1992 and morphology of epidermal eystlipids.] [lIlIest Dermoto/89:419 - 425, 1987 139. R •• wlings AV: Skin Waxes. Their composition, properties, structures and bio­ 114. Brysk MM, Miller j: Concanavalin A- binding glycoprotein in human stratum logica l significance. In: Hamilton R, C hristie W (cds.). Waxes: Chelllistry, corneum.J bIVest Dermato/82:280-282, 1984 Mo/fCIIlar Biology olld Flillctiolls . The Oily Press, 1991, pp 223-259 115. Btysk MM, Rajaraman S, Penn P, Barlow E: Glycoproteins modulate adhesion 140. Potts RO, Francoeur ML: The influence of stratum corneum morphology on of terminally differentiated keratinoeytes. Cell Tisslle Res 253:657 - 663, 1988 water permeability.] blllest DermotoI96:495 - 499, 1991 116. Btysk MM, Rajaraman S, Penn P, Chen SJ: Endogenous lectin from terminally 141. Marks R, Lawson A, Nicholl S: Age-related changes in stratum corneum struc­ differentiated epidermal cells. Differelltiotioll 32:230-237, 1986 ture and function. In: Marks R, Plewig G (cds.). The StrawlII Com""• . 117. Btysk MM, Bell T, Rajaraman S: Sensitivity of desquamin to proteolytic degra­ Springer-Verlag, 1983, pp 175-180 dation. PatllObio/59:109-112, 1991 142. Herrmann S, Scheuber E, Plewig G: Exfoliative cytology: effects of the seasons. 118. Skerrow CJ: Desrnosomal proteins. In: Bereiter-Halin j, Matoltsky AG, Rich­ In: Marks R , Plewig G (cds.). The Stratllll' Comelllll. Springer-Verlag, 1983, pp ards SK (eds.). BiologyoJtheIntegllmelll2 Vertebrates. Springer Verlag, 1984, pp 181 - 185 762-787 143. Johnson AW: Dry skin. Recent advances in research and therapy: a continuing 11 9. Chapman SJ, Walsh A: Desmosomes, corneosomes and desquamation. An ultra­ education program for pharmacists. Dmg Store NelVs Jar tire Pharmacist 4:51 - structural study of ad ulr pig epidermis. Arch Der",otol Res 282:304 - 310, 1990 58, 1994 120. Fartasch M, Bassukas ID, Diepgen TL: Structural (elationship between epider­ 144. Froebe CL, Simion FA, Ohlmcyer H , Rhein LD, MattaiJ, Cagan RH, Friberg mal lipid lamellae, lamellar bodies and desmosomes in human epidermis: an SE: Prevention of stratum corneum lipid phase transition by glycerol-an ultrastructural study. Br J Der",oto/128:1-9, 1993 alternative mechanism for skin moisturization.] Soc Coslllet Chelll 41 :51 - 65, 121. Lundstrom A, Egelrud T : Evidence that cell shedding from plantar stratum 1990 corneum in vitro involves endogenous proteolysis of the desmosomal protein, 145. Blank IH: Factors which influence the water content of the stratum corneum.] desmogleinl.J bIVest DermatoI94:216-220, 1990 ll1 l1est Derlllato/18:433-440, 1952