Identification of Phytosphingosine-Based 1-O-Acylceramide in the Human Stratum Corneum and Investigation of Its Role in the Skin Barrier

1. Introduction

Ceramides (CER) form an essential permeability barrier in the human stratum corneum (SC) in combination with cholesterol and free fatty acids (FFA) [1,2]. A plethora of information on ceramides' biological and skin physiological functions has been published [3,4]. Among the three skin physiological lipids, the molecular diversity of ceramides is the most complicated. The permeability of the epidermal barrier varies depending on the amount of ceramides and compositional changes among different ceramide classes [5,6,7,8]. Classes of ceramides in the human stratum corneum are categorized depending on sphingoids and fatty acids binding to the amino group of sphingoids. The number of ceramide subclasses has increased with the advent of analytical methods [9,10,11]. Most recently, Suzuki M. reported that the total number of ceramide classes is 25 based on the newly identified sphingoid, 1,14-sphinganine, and a β-hydroxy fatty acid [12]. In addition to the ceramide classes, the chain length of ceramide is also an essential factor that determines the lipid lamellar organization's physical and chemical characteristics [13,14,15,16,17]. Since Mann et al. first showed that topically applied with ceramides could normalize abnormal skin barrier conditions, ceramide NP has been dominantly utilized over the last 30 years [18,19,20,21,22]. Recently, ceramide NS, NDS(NG), AP, and EOP have become available. Accordingly, the number of investigations on the effect of NS, NP, AP, and EOP, individually or in combination, on the barrier permeability, lipid organization, and interactions between specific ceramide classes increase [23,24,25,26,27].

1-O-acylceramide (1-OAC) was first identified in human and mouse epidermis in 2013 and named ceramide CER 1-O-ENS based on CER NS backbone and 1-O-EAS based on CER AS [28]. In 2017, the complexity of 1-OAC isolated from mouse epidermis was unveiled to find ceramide CER 1-O-ENS with C18-sphingosine as the major 1-OAC. As a minor class, ceramide CER 1-O-ENDS was also identified [29]. Another group also reported the presence of CER 1-O-ENS in fetal vernix caseosa and speculated a waterproof function of 1-OAC [30]. In addition, new 1-OAC having esterified ω-hydroxy acyl group such as CER 1-O-E(EO)S and CER 1-O-E(EO)H were identified in the reconstructed human epidermis obtained from the culture of human-derived epidermal keratinocytes [31]. However, the presence of 1-OAC having phytosphingosine as a sphingoid backbone (CER 1-O-ENP) has not been reported yet, though CER 1-O-EAP was found in cultured human keratinocytes [31].

The biological function of 1-OAC remains unknown. However, it is speculated that 1-OAC might play an essential role in the formation of lipid multilamellar organization although it consists of only 2~3% of the total ceramide. A report showed that 1-OAC affected skin permeability barrier function in mouse [32]. Identifying CER 1-O-ENS in fetal vernix caseosa would suggest the importance of 1-OAC in the epidermal permeability barrier function. The finding of a more than 2-fold increase of 1-OAC when the biosynthetic process of ω-esterified ceramides such as EOS and EOP was blocked in a mouse model further supports this notion [30]. This result suggested that 1-OAC could compensate for the deficiency of ω-esterified ceramides by forming an extended structure having C24 lignoceric acid esterified at the 1-O-position that can still support LPP phase in the lamellar organization [32]. However, the precise role of this new ceramide in lipid lamellar organization needs further investigation.

Previously, we reported the synthesis of 1-O-stearoyl ceramide NP [33]. The chemical structure of CER 1-O-ENP resembles an ‘anchor bolt’ having a hydrophilic head group in the center while bidirectionally splayed hydrophobic acyl groups. Based on this characteristic structure, we proposed a ‘Bidirectional Anchoring Model’ for the CER 1-O-ENP’s mode of action, which plays a role as a linker ceramide. To obtain a shred of evidence for this model, we performed physicochemical analyses and molecular dynamics (MD) simulations. Stratum corneum (SC) mimetic-nanovesicles (SCNVs) were prepared using CER NP/cholesterol/fatty acid (model SC lipids) in combination with CER 1-O-ENP. The result strongly suggested that CER 1-O-ENP tightened the multi-lamellar structure of SCNV, improving long-term stability [33]. All-atom MD simulation with the sandwich model framework of the LPP was conducted to gain more insight into the role of CER 1-O-ENP in lamellar organization and SC permeability. The results found that the presence of CER 1-O-ENP induces a compact lipid matrix in the lateral dimension of the SC model. In addition, the data demonstrated that CER 1-O-ENP retarded the penetration of ethanol through the lipid matrix [38]. When treated CER 1-O-ENP to a reconstructed epidermis (RHE), the integrity of the stratum corneum was well maintained while the untreated control was significantly disturbed [34]. The result again suggested that CER 1-O-ENP could foster the lipid lamellar structure.

Having these outcomes from in vitro characterizations and MD simulation, we wanted to identify the presence of CER 1-O-ENP in the human skin before investigating its influence on the human skin barrier functions in combination with CER NP. Identifying CER 1-O-ENP in the human epidermis should make it more physiologically meaningful when proven to positively affect the skin barrier function.

2. Materials and Methods

2.1. Identification of 1-O-Stearoyl-Ceramide NP (18:0-t18:1/18:0) in Human Stratum Corneum

Human stratum corneum samples were obtained by five tape strippings using D-squame standard tape (Cuderm, Dallas, TX, USA, diameter, 2.2 mm). Skin ceramides were extracted using methanol and the solvent extracts were dried under a nitrogen stream at 30^oC. To analyze ceramides, the extracts were dissolved in 100 μL of chloroform/methanol (1/9, v/v). Aliquots were subjected to liquid chromatography-high resolution mass spectrometry (LC-HRMS) to determine the 1-O-stearoyl-ceramide NP (18:0-t18:1/18:0) in the human stratum corneum samples. A Vanquish LC system coupled with a Q Exactive Focus Orbitrap mass spectrometer (Thermo Fisher Scientific Inc, Waltham, MA, USA) was used. A Kinetex C18 column (100 x 2.1 mm, 2.6 μm particle size, Phenomenex, Torrance, CA, USA) was used to separate samples. The mobile phase consisted of 10 mM ammonium acetate in 90% methanol (A) and 10 mM ammonium acetate in isopropyl alcohol/methanol (1/1, v/v) mixtures (B), and was set as 30% B (0 min), 95% B (15-20 min), and 30% B (20.1-25 min). The flow rate was set to 0.2 mL/min. A full scan in positive mode was performed at 70,000 FWMH resolution (full width at half maximum) and MS/MS spectra were acquired at m/z 200–900 at a resolution of 17,500 in the daughter ion scan mode. Parallel reaction monitoring (PRM) was also employed, and the PRM transition m/z 848.8066 (MH+) was used for the detection of 1-O-stearoyl-ceramide NP (18:0-t18:0/18:1).

2.2. Ceramides and Test Creams

A CER NP (HP-EcoCeramide LCS^TM) was produced using a process previously reported with some modifications to adjust the fatty acid compositions [33]. It was a mixture of ceramide NPs with different fatty acids that originated from 4 different natural oils, Shea butter, Moringa, Meadowfoam, and Macadamia. The percentile profiles of N-acyl moieties in the CER NP are C16(4.0), C16:1(0.9), C18:0(35.8), C18:1(44.1), C18:2(4.6), C20:0(1.9), C20:1(4.0), C22:0(0.9), C22:1(1.1), and C24(0.3). N-stearoyl phytosphingosine and N-oleoyl phytosphingosine are the two most abundant subclasses of this ceramide NP mixture. CER 1-O-ENP (Figure 1) was synthesized by conjugating stearic acid to the first hydroxyl group of the CER NP as described before. Test creams were prepared by mixing each ceramide with the base cream formulation to produce a targeted ceramide content. The base cream avoids glycerin and emollients as much as possible to minimize the effect on SC hydration measurement. 0.02%, 0.05%, 0.2%, and 0.5% of CER 1-O-ENP were formulated together with a vehicle cream containing 0.2% of CER NP as a control (Table 1).

Table 1. Ingredients for vehicle cream and test creams.

Table 1. Ingredients for vehicle cream and test creams.

2.3. Human Study

A double-blind intra-subject vehicle-controlled human study was performed. The human study aimed to investigate the effect of CER 1-O-ENP on the skin barrier functions in combination with the CER NP. Twenty women ages 20-29(3), 30-39(13), and 40-49(4) who volunteered for the study who had no history of previous abnormal skin conditions were recruited with written consent. The study scheme is outlined in Figure 2, and skin barrier parameters include SC hydration, retention of hydration, and SC cohesion. All procedures in studies involving human participants were performed by the ethical standards of the institutional research committee and the 1964 Declaration of Helsinki. This study was approved by the Institutional Review Board of Dongguk University Committee on Human Research (Approval no. DUIRB-2022051-01). Written consent of participation from each volunteer was documented according to the ethical standards.

2.4. Measurement of Skin Hydration, Retention of Hydration, and SC Cohesion

• The assessment of SC hydration

The assessment of SC hydration was also performed according to the previously described [32]. Samples were treated for 14 days in the case of the 1st dose-finding human study and 28 days for CER 1-O-ENP study. A Corneometer CM820 device (Courage & Khazaka, Cologne, Germany) was used for hydration measurement. The changed value of arbitrary units was calculated from each baseline and expressed as SEM.

• The SC cohesion

The SC cohesion was expressed as Δ TEWL between baseline and after four weeks of test cream application sites. The TEWL was measured to calculate SC cohesion immediately after tape-stripping 15 times with the D-Squame tape [36].

2.5. Microscopic Observations of Maltese Cross Appearance

Microscope images were observed for test creams formulated with CER 1-O-ENP and CER NP at two time points; immediately after the preparation and after 6 months of RT storage. The optical anisotropy was observed under the cross-polarized light microscope (Nikon Corporation, Tokyo, Japan).

2.6. Statistical Analysis

All data are represented as mean ± standard deviation (SD), with differences between means assessed with IBM SPSS Statistics for Windows version 23 (IBM Corp., Armonk, NY, USA). The data analysis was performed using ANOVA (one-way analysis of variance) followed by Tukey's honest significant difference. Significant was defined as a P-value less than 0.05

3. Results

3.1. Identification of 1-O-Stearoyl-Ceramide NP (18:0-t18:0/18:1) in Human Stratum Corneum

The MS/MS spectrum of 1-O-stearoyl-ceramide NP (18:0-t18:0/18:1) standard, which has a protonated molecular ion [M+H]⁺ at m/z 848.8066 yielded m/z 564.5334 (base peak) and m/z 282.2790, arising from the elimination of the 1-O-stearoyl group and further cleavage of the N-(9Z)-octadecenoyl substituent with loss water (Figure 3A), respectively. 1-O-Stearoyl-ceramide NP can also lose water, resulting in a fragment ion at m/z 830.7952. The part of the proposed fragmentation pathway of 1-O-stearoyl-ceramide NP (18:0-t18:0/18:1) standard is illustrated. 1-O-Stearoyl ceramide NP (18:0-t18:0/18:1) was also observed in human stratum corneum extracts. The MS/MS spectrum of m/z 848.8066 ion which was found in human stratum corneum extracts gave characteristic fragment ions at m/z 830.7919, 564.5336, and 282.2793 (mass error < 5 ppm) which are also observed in 1-O-stearoyl-ceramide NP (18:0-t18:1/18:0) standard (Figure 3B).

3.2. Profiling of 1-O-Acylceramide NP (CER 1-O-ENP) in Human Stratum Corneum

CER 1-O-ENP in the human stratum corneum was profiled using LC-HRMS. MS/MS spectra of CER 1-O-ENP were obtained from the parallel reaction monitoring (PRM) of the protonated adduct ion ([M+H]+) (Figure 4). The extracted ion chromatograms (EIC) for the detected CER 1-O-ENP were shown in Figure S1. Based on established mass fragmentation patterns of CER 1-O-ENP (18:0-t18:0/18:1), nine species of CER 1-O-ENP were identified in human stratum corneum (Table 2). Interestingly, CER 1-O-ENP with an elemental composition of CnH2nO5N1 (n= 52-66), suggesting that N-acyl substituents consist of saturated form were mainly observed in human SC. Analyzing the fragments revealed 1-O-ENP ceramide molecules that differ in ester- and amide-linked fatty acid chain lengths. Multiple isobaric isomers were observed for most CER 1-O-ENP species, which existed as a mixture of two or seven species in the MS/MS spectra. For example, the MS/MS spectra of the ions of m/z 878.8535 (Figure 4A) showed that it included three ceramide NP moieties (fragment ions at m/z 594.5800, 622.6125, and 650.6443 corresponding to the dehydrated NP (t18:0/20:0), NP (t18:0/22:0), and NP (t18:0/24:0), respectively, from the cleavage of the 1-O-acyl substituent) (Figure 4B). In addition, the observation of the fragment ion at m/z 264.2683, corresponding to the d18:1-long chain base, reflected the presence of phytosphingosine resulting from cleavage of the N-acyl substituent and two water molecules from ceramide NP moieties. The confirmation of the suggested structures is further supported by the elemental composition extracted by high-resolution mass spectrometry (deviation within 5 ppm). Therefore, CER 1-O-ENP species at m/z 878.8535 was identified as the mixture of 1-O-ENP (18:0-t18:0/20:0), (16:0-t18:0-22:0), and (14:0-t18:0/24:0). Based on these kinds of fragmentation patterns, the structures of other 1-O-ENP were additionally identified (Table 2).

Table 2. Summary of l-O-acylceramide NP identified in the human stratum corneum.

Table 2. Summary of l-O-acylceramide NP identified in the human stratum corneum.

3.3. Effect of CER 1-O-ENP on Multilamellar Formation and Its Stability

Once test creams were prepared, we assessed the effect of CER 1-O-ENP on forming a multilayer lamellar structure in combination with CER NP, and evaluated the long-term stability of multilamellar structure. Ingredients of test creams are listed in Table 1. A typical concentric lamellar structure could be observed as a Maltese cross appearance under a cross-polarized microscope when ceramide is formulated with cholesterol and free fatty acid. This Maltese cross appearance, optical anisotropy, indicates appropriate lipid lamellar formation. It is generally accepted that a moisturizer with a more typical cross pattern is better for repairing the skin barrier. A month after the preparation of test creams, TC1 containing 0.2% CER NP alone showed almost no Maltese cross formation; even if a few remained, their size became smaller. Test cream TC3, prepared with 0.2% CER NP and 0.05% CER 1-O-ENP, presented numerous typical Maltese crosses. With increasing CER 1-O-ENP (TC5) concentration, the Maltese cross became more prominent in size and higher order in the multilamellar organization. It indicated that CER 1-O-ENP enhanced the formation of multilamellar structure. (Figure 5). After six months of storage at RT, the Maltese cross of TC3 was maintained well but with minor deformation, while TC5 retained the Maltese cross fairly well at almost the same time as the preparation time. TC1 showed a few Maltese crosses.

3.3. The Influence of CER 1-O-ENP on the Skin Hydration

Having these results, we investigated the effect of CER 1-O-ENP in combination with CER NP on the human skin barrier function. The content of CER NP in test creams was set to be 0.2%. The concentration of CER 1-O-ENP in the test creams was adjusted to be 10 or 25% of the CER NP used in the test creams, respectively. It is known that the concentration of 1-O-acylceramide was estimated to be 2~3% of the total ceramide in the human SC. Meanwhile, Ceramide NP’s concentration in the human skin barrier was around 22%, indicating the relative CER 1-O-ENP concentration to NP ranges from 9~13%. The relative content ratio of CER NP and CER 1-O-ENP in the test creams was adjusted within the range that reflects the content ratio of the human skin. As shown in Figure 6A, after four weeks of application, significant enhancement of SC hydration was observed from all the skin sites applied with the test creams TC1, TC2, TC3, and TC4 by 32%, 50%, 54%, and 43%, respectively. The hydration level of the skin applied with TC1, formulated by using only 0.2% CER NP, was considerably increased, as expected. The addition of 0.02% or 0.05% of CER 1-O-ENP to the TC1 brought about a dramatic enhancement of SC hydration. More than 20% of enhancement, which is statistically significant, was induced by adding CER 1-O-ENP compared to the SC hydration enhanced by the TC1 application. The long-lasting moisturization effect, or retention of hydration, was found to be even greater when the CER 1-O-ENP was combined with CerNP (Figure 6B). The moisture contents of all three test creams formulated with CER 1-O-ENP and CerNP together, TC2, TC3, and TC4 were maintained significantly higher than that of the baseline containing except TC1 three days after stop application. This result clearly showed that CER 1-O-ENP in the test cream confers a long-lasting moisturizing effect when applied to the human skin.

3.4. CER 1-O-ENP Fostered the SC Cohesion

TEWL was measured to calculate Δ TEWL before and immediately after tape-stripping 15 times with Desquame disc tapes (CuDerm, Dallas, TX, USA). All the Δ TEWL values obtained from the skin sites applied with test creams containing CER 1-O-ENP (TC2, TC3, TC4) were significantly less than that of the control cream (TC1) containing only ceramide NP (Figure 7). The results indicated that the stratum corneum has become more resistant against a physical force of the tape stripping. However, no additional improvement of SC cohesion was observed from the concentration of CER 1-O-ENP higher than 0.02% (TC3 and TC4). The CER 1-O-ENP to CER NP ratio seems appropriate at 10 to 25%, similar to the ratio found in human skin. In conclusion, CER 1-O-ENP significantly boosted the SC cohesion in combination with CER NP and other ceramides.

4. Discussion

This study aimed to investigate the physiological role of CER 1-O-ENP in combination with CerNP for several skin barrier parameters, for example, hydration, moisture retention, and SC cohesion. Although CER 1-O-ENS was identified as the primary class of 1-O-acylceramide, we produced CER 1-O-ENP since phytosphingosine is only affordable on a large scale. So far, no reports on the identification of CER 1-O-ENP have been published. We report for the first time the presence of CER 1-O-ENP in the human stratum corneum, specifically focusing on a subspecies known as 1-O-stearoyl ceramide NP (18:0-t18:0/18:1). This subspecies is the primary component of the CER 1-O-ENP examined in this study. According to a separate ongoing study by Liu with another group, CER 1-O-ENP was estimated to be about 1/4 of CER 1-O-ENS (unpublished data). It was also reported that about 30% of sphingosine ceramides converted into CER 1-O-ENS and other sphingosine-based 1-OAS [28]. Identifying CER 1-O-ENP in the human epidermis should make it more physiologically meaningful when proven to positively affect the skin barrier function. With this finding, CER 1-O-ENP could be classified as a genuine human skin ceramide.

Multilamellar lipid organization as a permeability barrier is a typical feature of the intercellular space of the skin barrier. It is manifested as a Maltese cross under a polarized light microscope [37]. As shown in Figure 5, CER 1-O-ENP stimulated the formation of a typical Maltese cross compared to that of base cream and CER NP cream. With the increasing concentrations of CER 1-O-ENP, the Maltese cross became thicker and more prominent. This may indicate that the number of lamellar layers in the Maltese cross has increased. More importantly, the striking enhancement of the stability of the Maltese cross after six months of storage at RT was observed from the test cream containing CER 1-O-ENP. This observation aligned with previous findings from different multilamellar vesicle experiments; CER 1-O-ENP demonstrated that increasing the CER 1-O-ENP application could make the multilamellar vesicle (SCNV model) more stable [33]. This finding was further supported by results from a different multilamellar nanovesicle model, where the addition of CER 1-O-ENP significantly promoted the formation of multilamellar nanovesicles and enhanced their stability during repeated freeze-thaw cycles. [39]. Our study, which utilized the RHE skin model, provides further confirmation that CER 1-O-ENP exerts a stabilizing effect on the lipid multilamellar matrix of the RHE stratum corneum [34]. This additional evidence underscores the significant role of CER 1-O-ENP in enhancing skin barrier properties.

In this human study, the concentration of CER NP was fixed at 0.2% for every test cream with different CER 1-O-ENP contents because 0.2% of CER NP was the lowest concentration to present marginal yet marked enhancement of the skin barrier function. Therefore, we could expect that any CER 1-O-ENP was responsible for any changes from the topical application of test creams with varying concentrations. The results demonstrated that this is the case. Figure 6A shows that when 0.02% and 0.05% of CER 1-O-ENP were added to a control cream containing 0.2% CER NP, a significantly high skin moisturizing effect was observed compared to the control. However, no further improvement of skin barrier parameters was observed from the skin site treated with the test creams containing more than 0.05% CER 1-O-ENP. The results indicated that 0.02% or 0.05% of CER 1-O-ENP seemed appropriate in this formulation context, representing 1/10 ~1/4 of CER NP. This ratio is similar to that of the actual skin because 1-OAC only comprises less than 5% of the total ceramide in the skin barrier.

A significant decrease in the ΔTEWL (Figure 7) was observed only from test creams. CER 1-O-ENP suggested that CER 1-O-ENP fostered SC cohesion, most probably by increased corneodesmosomes. However, SC cohesion could also be strengthened by fostering the lipid multilamellar organization via the anchoring action of CER 1-O-ENP. Further study is required to elucidate whether CER 1-O-ENP can enhance the formation of corneodesmosomes.

Changes in CER composition concerning ceramide classes and chain lengths, commonly observed in abnormal skin, are also known to be the primary causes of impaired skin barrier function. However, more investigations are needed to elucidate molecular interactions among different ceramides in stratum corneum. Recently, Kono et al. reported a systematic review of the literature describing the effects of ceramide on human skin from clinical viewpoints and concluded that no report provided details of the composition of ceramide and concentrations of formulations [19,21,22]. In this study, we used defined dosages of ceramides with complete descriptions of their chemical nature. Some studies provided insights into the interactions between the individual ceramide classes, such as NS, AP, NP, and EOS, in forming lipid lamellar organization [25,26,27]. For example, CER NS was shown to have low miscibility with other lipids, and on the other hand, CER NP formed a strong in-plane H-bonding network [27]. Others reported that the ratio between CER NP and AP was critical concerning maintaining lamellar stability, and they found that a 2:1 ratio was optimal [25]. It is generally accepted that the presence of EO class ceramides such as EOS and EOP is required to establish a LPP organization [23,24]. In this study, we have shown a novel interaction between two individual ceramide classes, CER NP and CER 1-O-ENP, showing that a tiny amount of CER 1-O-ENP could significantly promote key skin barrier parameters by boosting the CER NP activity otherwise marginal. In this regard, our result, the boosting effect of CER 1-O-ENP on the skin barrier activity of CER NP, has provided new insight into the relative role of each specific ceramide class. It would be interesting to find if CER 1-O-ENP has similar boosting activity with other class of ceramide, such as CER NS or NG.

5. Conclusions

We identified 1-O-stearoyl ceramide NP (18:0/t18:0/18:1), a species of CER 1-O-ENP, along with other sub-species of diverse O-acyl and N-acyl chain length in the human stratum corneum. The results from the human study using a test cream co-formulated with the human skin-identical CER 1-O-ENP, and CER NP suggest there is a boosting effect of CER 1-O-ENP on the improvement of skin barrier functions by CER NP since a surprisingly small amount of CER 1-O-ENP is required. Our findings are the first report investigating the physiological role of CER 1-O-ENP in the human skin barrier.

6. Patents

KR 10-2621498