In 1997, Wiechers1 introduced the concept of relative performance measurement to compare the moisturization of several neat emollients. The capacitance of skin treated with test products was measured by a corneometer and compared with glycerin-treated skin (defined as 100%) and untreated skin (defined as 0%) at given intervals, normally 6 hr after application. As one might expect, this test showed that all emollients were not the same in their capacity to moisturize skin. Emollients offered either low moisturization performance (0–30% relative performance moisturization (RPM)), medium RPM (30–70%), high RPM (70–100%), or excellent RPM (> 100%).1 Two structurally similar emollients interestingly imparted completely different moisturization performance: isopropyl isostearate (IPIS), with an RPM of 100%; and isostearyl isostearate (ISIS) with a score of only 15%, as illustrated in Figure 1.
Molecular modelling of all emollients in the study revealed that the mechanisms of IPIS and ISIS to impart moisture did not adhere to the two generally accepted means of moisturizing skin—i.e., acting either as a humectant or sponge, such as a hygroscopic molecule like glycerin does; or working as a plastic cling wrap or Saran Wrapa, as substantive molecules in petroleum jelly do.2 This hinted at a possible third mechanism of skin moisturization.
At the IFSCC Congress in 2002, a paper was presented describing additional work measuring the effect of neat emollients on skin moisturization using near-infrared reflectance spectroscopy (NIR).3 The NIR results for IPIS and ISIS were very different from those obtained with the corneometer. Figure 2 shows essentially no difference between IPIS and ISIS; both are good moisturizers.
How can the same molecule score differently for the same property, i.e., skin moisturization? To answer this, the first logical step is to investigate whether the same phenomenon indeed was measured.
The corneometer measures anything with a dipole moment and thus every type of water (bound, unbound, etc.) at a depth of 60–100 μm.4 Since the dipole moment of IPIS and ISIS is low, it could be concluded that water and not the moisturizer was in fact measured. Even in the case of glycerin, which has a high dipole moment, glycerin-treated skin had significantly higher readings than glycerin itself, suggesting that water indeed is measured, albeit indirectly.
NIR, on the other hand, measures water directly but much deeper—to a level of 1.3 mm. The wavelengths at which the overtone and combinations of fundamental vibrations can be seen depend on the mobility of the water molecules, hence a distinction can be made between bound and bulk water.3 The fact that the corneometer signal was much greater on IPIS-treated skin than on ISIS-treated skin suggests that the water is closer to the skin surface when this is treated with IPIS than with ISIS. This is because the strength of the signal depends on its proximity to the measuring probe. However, the equality in NIR signal on IPIS- or ISIS-treated skin suggests that the total amount of water in the skin as a whole is the same. Thus, the next step was to determine where in the skin the water was located.
Bouwstra et al. recently developed a method to determine the distribution of water in the skin via cryo-SEM.5 In short, following the topical administration of a moisturizer (or nothing, as a control), a section of full-thickness skin or isolated stratum corneum (SC) is quickly deep-frozen and cut perpendicular to the skin surface. Water that is present in the plane of the cut subsequently sublimates when the section is then brought under high vacuum. Since water is the only chemical that sublimates under these conditions, irregularities in the plane of the cut section will only be found where the water originally was located.
Spraying the skin sample with platinum and subsequently visualizing it by means of scanning electron microscopy allows these irregularities to be studied. Figure 3 shows three characteristic water distributions in human skin: a black (low contrast) layer on top of skin indicating low water content, or so-called upper non-swelling region (UNSR); a grayish (higher contrast) layer in the middle of skin showing high water content, or the so-called swelling region (SR); and another non-hydrated, black (low contrast) layer at the bottom of the SC, or so-called lower non-swelling region (LNSR). The occurrence of these high and low hydration zones beautifully corresponded with the occurrence of NMF as described by Rawlings et al.6
The cryo-SEM technique was also used after application of glycerin (the sponge molecule), petroleum jelly (the cling-wrap molecule), IPIS, ISIS or nothing (control).7 Human skin was dermatomed to a thickness of 250–300 µm
and equilibrated in Franz diffusion cells over a saturated Na2CO3 solution at 32°C, which typically results in a weight increase of 57% to 87%, relative to dry SC.5 Moisturizers were applied neat for 24 hr on the skin of at least three donors.
Glycerin was applied as a 35% w/v solution following equilibration over saturated Na2CO3 at 32°C since the pure glycerin was observed to be such a strong humectant that it extracted water from the SC (data not shown). Relative to an untreated control, the number of cells in the UNSR decreased with petroleum jelly (Figure 4a), and glycerin was shown to penetrate into the corneocytes of the SR that swelled dramatically (Figure 4b). With IPIS, the number of cells in the UNSR increased (Figure 4c), whereas with ISIS, the number of cells in the UNSR decreased (Figure 4d).
An increased UNSR implies that the water is hidden away deeper in the skin (IPIS), whereas a reduced UNSR (petroleum jelly) indicates that the water is closer to the skin surface. Finally, an increased SR means there is simply more water in the skin (glycerin). However, this was not at all in line with NIR and corneometer measurements. Table 1 summarizes these confusing results so far, and suggests that another mechanistic effect must be occurring.
The different distribution of water in the SC as a consequence of the application of the two emollients could have been caused by different types of interaction with the skin lipid organization. After all, a previous publication by Pilgram et al. showed that skin lipid organization has a prominent effect on skin barrier function.8 A schematic representation of this is shown in Figure 5.
In the liquid crystalline (Lα of fluid) phase (5a), lipids display lateral and rotational movements. In hexagonal (Lβ or gel) and orthorhombic (crystalline) phases, however, lateral movements of lipids are reduced. In the hexagonal packing, the hydrocarbon chains can freely rotate around their axes but in orthorhombic packing, the lipids are in a solid state and packed more closely in one direction, indicated by the
0.3 nm-spacing. In contrast to hexagonal and orthorhombic packing, the position of the lipids is not well-defined in the liquid phase (5b); the circles represent the acyl chains of the lipids.
The orthorhombic phase is characterized by very low permeability. The hexagonal phase, where the lipids are slightly further apart, exhibits an intermediate permeability, and the liquid phase, where the lipids are even further apart, has a high permeability. It is obvious that the quality of the barrier function influences the capability of the SC to retain water.
Two additional sets of experiments using small angle X-ray diffraction (SAXD) and Fourier-transformed infrared (FTIR) were undertaken to study the lamellar and lateral packing, respectively. With SAXD, the incorporation of skin moisturizers into the skin lipid bilayer structure can be investigated but the technique is not conclusive. An increase in the periodicity of the lamellar phase, which is the distance over which the structure is repeated, is evidence for the incorporation of skin moisturizers into the lipids although the lack of an increase can imply incorporation without any effect on skin lipid behavior.
Porcine skin lipids were extracted and treated as described previously, and ceramides, cholesterol, free fatty acids and moisturizer were mixed in a 2:1:1:1 molar ratio.9 In addition, human SC was isolated, rolled up and flattened, resulting in a tight “stack” of 45 SC layers. Dry stacks were immersed in neat IPIS or ISIS for a period of 24 hr at 32°C.9
The first set of experiments using isolated pig skin lipids showed that the addition of 20% IPIS reduced the formation of the short periodicity phase but promoted the long periodicity phase that is essential for skin barrier function.10 IPIS was therefore at least partially incorporated into the lipid bilayer structure. No changes were observed after the addition of ISIS but as noted, this did not mean the moisturizer was not incorporated into the skin lipid bilayer because it could be incorporated without having an effect on the periodicity of the lipid lamellae formed. Other methods, such as FTIR, may shed light on this possibility.9
In the second set of experiments using stacked human SC sheets, prior to the application of a skin moisturizer, the repeat distance measured 12.7 nm but when the moisturizers were added, the repeat distance increased (control = 12.7 nm; ISIS = 13.1 nm; and IPIS = 13.2 nm). This indicates that lipophilic moisturizers can be incorporated into the lamellar lipid phases of sprayed lipids (see Figure 6). However, it must be stressed that in repeated samples, this increase was not observed.9
A final set of experiments was therefore performed using perdeuterated moisturizers ISIS and IPIS in FTIR experiments. FTIR is a well-established technique in SC lipid research11 that can measure symmetric and asymmetric stretching as well as scissoring of C-H bonds in -CH2- groups. However, since both moisturizers and skin lipids contain –CH2– groups, moisturizer versus skin lipid C-H bond signals cannot be distinguished. Therefore, perdeuterated moisturizers were used. Within such moisturizers, every C-H bond is replaced with a C-D bond by replacing all hydrogen atoms with deuterium atoms. Perdeuterated moisturizers behave in the same way as standard moisturizers but the C-D bond vibrates at different frequencies, thus enabling the differentiation of signals.
By means of FTIR, researchers can identify whether skin lipids are present in an orthorhombic phase, a hexagonal phase or in a liquid phase. The orthorhombic phase is the most rigid skin lipid configuration seen in the human SC; therefore it has the lowest skin penetration. The hexagonal phase is less rigid and thus represents a somewhat weaker barrier, whereas the fluid phase is a relatively dynamic configuration and constitutes a weaker barrier.8 At normal human physiological temperatures, skin lipids mainly are present in an orthorhombic configuration but some hexagonal phases are also present.8 So the question arises: Does the application of a skin moisturizer change the configuration of the skin lipid phases? FTIR can provide clarity.
Simply put, and with some scientific freedom, the occurrence of an orthorhombic phase is recognized in an FTIR spectrum by two peaks, whereas a hexagonal phase is characterized by a single peak.12 While all three peaks have different positions, since both phases are present in human skin at a normal physiological pH, researchers will see all three peaks. However, when skin temperature is increased, the orthorhombic phase transitions to a hexagonal phase and this is observed in the gradual transition of two peaks to one.
FTIR measurements of SC lipid samples treated with 20% ISIS and IPIS showed that at skin temperature, the vital orthorhombic lateral packing of SC lipids was more prominently present in the presence of moisturizers (see Figure 7).12, 13
This mechanism was named the “internal occlusion” mechanism12, 13 and although novel in knowledge, it must have existed for years since many cosmetic products do contain emollients like IPIS and ISIS, and formulations containing these emollients have demonstrated good moisturizing properties.14
A plethora of techniques have been used to measure skin moisturization of IPIS and ISIS. As a review, corneometer measurements showed vast differences in moisturization between the two, whereas NIR measurements indicated them to be much alike—but these could be different due to the different positions where these two methods measure. Additional research using cryo-SEM indeed indicated differences in water distribution profiles between the two, while SAXD investigations suggested that both moisturizers could be included in the skin lipid bilayer structure; however, the data was not conclusive. Finally, FTIR spectra of sprayed SC lipids with and without 20% moisturizer showed a stabilization of the orthorhombic phase that leads to the stabilization of skin barrier function, which in turn should lead to increased skin moisturization. But if the two moisturizers work via the same mechanism, how could their corneometer values be so different? Stabilization of the orthorhombic phase alone is not enough to explain this new mechanism of action of skin moisturization. There must be another requirement for molecules that stabilize the orthorhombic phase before they exert their activity.
At this point, it should be recalled that the FTIR studies were performed on mixed skin lipid studies where 20% of the molecules was moisturizer. The latter molecule must penetrate the SC in sufficient quantities to reach such a level. IPIS is a much smaller molecule and is therefore expected to penetrate the skin better than ISIS, which is significantly larger.15 Most likely, smaller quantities of ISIS than IPIS penetrate the skin, and IPIS penetrates the skin deeper. This means that at the same depth in the SC, the orthorhombic phase will be more stabilized with IPIS than with ISIS. This also implies that the stabilization will be deeper in the SC with IPIS than with ISIS. The cryo-SEMs in Figure 4 confirm this.
Since the barrier is more rigid at deeper levels in the skin when applying IPIS, the upper non-swelling region or UNSR is increased relative to normal skin. When ISIS is applied, the barrier is also strengthened but at a level closer to the skin’s surface. As a consequence, the water accumulates below this strengthened barrier, which leads to a reduced UNSR. However, because IPIS strengthens the barrier more than ISIS, more water should be “visible” to the corneometer when measuring at the level of the SC (Figure 1), whereas the differences become less pronounced or even undetectable when measuring with NIR spectroscopy (Figure 2). The most logical experiment remaining is to measure TEWL after a single and multiple applications.
This internal occlusion mechanism may be more important than initially thought. After all, when looking at the therapeutic opportunities to correct dry skin, Rawlings lists the following approaches: correcting ceramide deficiencies; normalization of the lipid lamellar structure; correction of the reduction of NMF levels; correction of the aberration of the corneocyte envelope maturation; or correction of impaired corneodesmolysis.16 However, the question to ask here is: “What is the cause and what is the effect?” Are all these symptoms not caused by an impaired barrier function? In other words, by stabilizing the orthorhombic phase, researchers may have discovered the fundamental answer to all skin moisturization problems.
This paper introduces a third and, until now, novel mechanism of skin moisturization: stabilization of the orthorhombic skin lipid phase. This internal occlusion leads to a more sensorially pleasing answer to dry skin problems than externally occluding molecules such as mineral oil and petroleum jelly, which are heavy and must work externally, not solving the dry skin problem in the long run. The work described here clearly indicates that the answer one gets for a particular class of moisturizer depends not only on the mechanism of the skin hydration measuring equipment, but also on the mechanism of action of the moisturizer.
All types of moisturizers have their benefits but where some might simply be fighting the symptoms, others such as the orthorhombic phase stabilizers might be affecting the underlying cause of dry skin, which is an impaired barrier function.
Acknowledgement: The author wishes like to thank Joke Bouwstra, PhD, and Julia Caussin, PhD, of the University of Leiden, Netherlands, for all the years of very pleasant collaboration during his employment with Uniqema. The same thanks apply to former colleagues Wei Hansen, PhD, Marchel Schnieder and Tony Rawlings, PhD, of AVR Consulting.
1. JW Wiechers, Relative performance testing: Introducing a tool to facilitate cosmetic ingredient selection, Cosm & Toil, 112 (9) 79–84 (1997)
2. JW Wiechers and A Barlow, Skin moisturisation and elasticity originate from at least two different mechanisms, Int J Cosmet Sci, 21 425–435 (1999)
3. JW Wiechers, M Snieder, NAG Dekker and WG Hansen, Factors influencing skin moisturization signal using near-infrared spectroscopy, IFSCC 6(1) 19–26 (2003)
4. CW Blichman and J Serup, Assessment of skin moisture. Measurement of electrical conductance, capacitance and trans epidermal water loss, Acta Dermatol Venereol (Stockholm) 68 284–290 (1988)
5. JA Bouwstra et al, Water distribution and related morphology in human stratum corneum at different hydration levels, J Invest Dermatol 120 750–758 (2003)
6. AV Rawlings, IR Scott, CR Harding and P Bowser, Stratum corneum moisturization at the molecular level, J Invest Dermatol 103 731–740 (1994)
7. J Caussin et al, Lipophilic and hydrophilic moisturizers show different actions on human skin as revealed by cryo scanning electron microscopy, Exp Dermatol 16 891–898 (2007)
8. GSK Pilgram et al, Aberrant lipid organization in stratum corneum of patients with atopic dermatitis and lamellar ichthyosis, J Invest Dermatol 117 710–717 (2001)
9. J Caussin, GS Gooris, HWW Groenink, JW Wiechers and JA Bouwstra, Interaction of lipophilic moisturizers on stratum corneum lipid domains in vitro and in vivo, Skin Pharmacol Physiol 20 175–186 (2007)
10. JA Bouwstra, GS Gooris and M Ponec, The lipid organization of the skin barrier: Liquid and crystalline domains coexist in lamellar phases, J Biol Phys 28 211–223 (2002)
11. DJ Moore, ME Rerek and R Mendelsohn, Lipid domains and orthorhombic phases in model stratum corneum: Evidence from Fourier transform infrared spectroscopy studies, Biochem Biophys Res Comm 231 797–801 (1997)
12. J Caussin et al, Different mechanisms of action of skin moisturizer require different measurement techniques of skin moisturization, Conference Proceedings IFSCC Conference 2007, Amsterdam, The Netherlands (2007)
13. J Caussin, GS Gooris and JA Bouwstra, FTIR studies show lipophilic moisturizers to interact with stratum corneum lipids, rendering the more densely packed, Biochim Biophys Acta Biomem 1778 1517–1524 (2008)
14. JW Wiechers, CL Kelly, TG Blease and JC Dederen, The influence of emulsifiers on the skin penetration from emulsions, in: JW Wiechers (ed), Science and Applications of Skin Delivery Systems, Carol Stream, IL, USA: Allured Business Media (2008) ch 8, pp 125–138
15. RO Potts and RH Guy, Predicting skin permeability, Pharm Res 9 663–669 (1992)
16. AV Rawlings, Trends in stratum corneum research and the management of dry skin, Int J Cosm Sci 25 63–95 (2003)