ORIGINAL ARTICLE
Distribution of Bioactive Lipid Mediators in Human Skin Alexandra C. Kendall1, Suzanne M. Pilkington2, Karen A. Massey3, Gary Sassano4, Lesley E. Rhodes2 and Anna Nicolaou1 The skin produces bioactive lipids that participate in physiological and pathological states, including homeostasis, induction, propagation, and resolution of inflammation. However, comprehension of the cutaneous lipid complement, and contribution to differing roles of the epidermal and dermal compartments, remains incomplete. We assessed the profiles of eicosanoids, endocannabinoids, N-acyl ethanolamides, and sphingolipids, in human dermis, epidermis, and suction blister fluid. We identified 18 prostanoids, 12 hydroxyfatty acids, 9 endocannabinoids and N-acyl ethanolamides, and 21 non-hydroxylated ceramides and sphingoid bases, several demonstrating significantly different expression in the tissues assayed. The array of dermal and epidermal fatty acids was reflected in the lipid mediators produced, whereas similarities between lipid profiles in blister fluid and epidermis indicated a primarily epidermal origin of suction blister fluid. Supplementation with omega-3 fatty acids ex vivo showed that their action is mediated through perturbation of existing species and formation of other anti-inflammatory lipids. These findings demonstrate the diversity of lipid mediators involved in maintaining tissue homeostasis in resting skin and hint at their contribution to signaling, cross-support, and functions of different skin compartments. Profiling lipid mediators in biopsies and suction blister fluid can support studies investigating cutaneous inflammatory responses, dietary manipulation, and skin diseases lacking biomarkers and therapeutic targets. Journal of Investigative Dermatology (2015) 135, 1510–1520; doi:10.1038/jid.2015.41; published online 12 March 2015
1
Faculty of Medical and Human Sciences, Manchester Pharmacy School, The University of Manchester, Manchester, UK; 2Dermatology Centre, Institute of Inflammation and Repair, Faculty of Medical and Human Sciences, The University of Manchester, Manchester, UK; 3School of Pharmacy and Centre for Skin Sciences, School of Life Sciences, University of Bradford, Bradford, UK and 4Safety and Environmental Assurance Centre, Unilever, Sharnbrook, UK Correspondence: Anna Nicolaou, Faculty of Medical and Human Sciences, Manchester Pharmacy School, The University of Manchester, Stopford Building, Oxford Road, Manchester M13 9PT, UK. E-mail:
[email protected] The work was carried out in Bradford, West Yorkshire, UK.
Abbreviations: AA, arachidonic acid; AEA, N-arachidonoyl ethanolamide; 2-AG, 2-arachidonoyl glycerol; ALEA, N-alpha-linolenoyl ethanolamide; CB, cannabinoid receptor; COX, cyclooxygenase; CYP, cytochrome P450; DGLA, dihomo-gamma-linolenic acid; DHA, docosahexaenoic acid; DHEA, N-docosahexaenoyl ethanolamide; EPA, eicosapentaenoic acid; EPEA, N-eicosapentaenoyl ethanolamide; GC, gas chromatography; HDHA, hydroxydocosahexaenoic acid; HEPE, hydroxyeicosapentaenoic acid; HETE, hydroxyeicosatetraenoic acid; HETrE, hydroxyeicosatrienoic acid; HFA, hydroxy fatty acid; HODE, hydroxyoctadecadienoic acid; LA, linoleic acid; LC-MS/MS, liquid chromatography tandem mass spectrometry; LEA, N-linoleoyl ethanolamide; LOX, lipoxygenase; LT, leukotriene; NAE, N-acyl ethanolamide; OA, oleic acid; OEA, N-oleoyl ethanolamide; PA, palmitic acid; PEA, N-palmitoyl ethanolamide; PLA2, phospholipase A2; PLD, phospholipase D; PPAR, peroxisome proliferator-activated receptor; PUFA, polyunsaturated fatty acid; Rv, resolvin; SEA, N-stearoyl ethanolamide; S1P, sphingosine-1-phosphate; S1P1, sphingosine-1-phosphate receptor 1; SPE, solid phase extraction; TP, thromboxane receptor; TX, thromboxane Received 7 August 2014; revised 17 December 2014; accepted 23 January 2015; accepted article preview online 10 February 2015; published online 12 March 2015
1510 Journal of Investigative Dermatology (2015), Volume 135
INTRODUCTION The skin is rich in lipids that not only contribute to formation and maintenance of the epidermal barrier, but also perform critical roles in membrane structure and the functions of cutaneous cells. Bioactive lipid mediators are derived from complex membrane lipids and are produced upon request and in response to environmental and signaling stimuli. They contribute to both skin physiology and pathology, with evidence of involvement in psoriasis, dermatitis, acne, wound healing, and UV responses (reviewed in (Kendall and Nicolaou, 2013)). Dermal and epidermal compartments have diverse roles, while supporting each other with nutrients and transcellular signals (Edmondson et al., 2003; Yamaguchi et al., 2005; Yu et al., 2011), and exhibit considerable inter-family cross talk (Figure 1). Thus, the lipid complement of human skin and its compartments demand definition, to assist understanding of skin biology and to facilitate diverse applications including assessment of treatment targets in skin disorders. Of particular relevance because of their wide range of potential activities in the skin are the eicosanoid, endocannabinoid, N-acyl ethanolamide, and sphingolipid families. The eicosanoids are oxygenated metabolites of the 20-carbon polyunsaturated fatty acids (PUFA) arachidonic acid (AA, 20:4n − 6), eicosapentaenoic acid (EPA; 20:5n − 3), and dihomogamma linolenic acid (DGLA; 20:3n − 6). This family comprises the cyclooxygenase (COX)-derived prostaglandins © 2015 The Society for Investigative Dermatology
AC Kendall et al.
Cutaneous Bioactive Lipids
Plasma membrane
PLA2
Inhibition Activat
ion
N-acyl transferase, PLD
Sphingomyelinase
PUFA (e.g. AA, EPA, DHA ) LOX
COX
Hydroxy FA (e.g. HETE, HDHA, LT, LX, Rv, PD)
Ceramides
Acti
vati
on
Ceramidase, ceramide kinase, sphingosine kinase
FA-EA endocannabinoids (e.g. AEA, PEA)
C1P Sphingoid bases S1P
Prostanoids (PG, TX)
CB GPR55
S1P1
TRPV BLT
PPAR
G2A
NF-κB AP-1
Nucleus
Apoptosis proliferation differentiation protein synthesis
EP DP Migration TP degranulation IP cytokine release surface molecule expression
Cell
Figure 1. Schematic outline of bioactive lipid mediator production and related signaling events. AA, arachidonic acid; AEA, arachidonoyl ethanolamide; AP-1, activator protein 1; BLT, leukotriene B4 receptor; CB, cannabinoid receptor; COX, cyclooxygenase; C1P, ceramide-1-phosphate; DHA, docosahexaenoic acid; DP, prostaglandin D2 receptor; EP, prostaglandin E2 receptor; EPA, eicosapentaenoic acid; FA, fatty acid; FA-EA, fatty acid ethanolamide; G2A, G protein–coupled receptor 132; GPR55, G protein-coupled receptor 55; HDHA, hydroxydocosahexaenoic acid; HETE, hydroxyeicosatetraenoic acid; IP, prostacyclin receptor; LOX, lipoxygenase; LT, leukotriene; LX, lipoxin; NF-κB, nuclear factor of kappa-light-chainenhancer in B-cells; PD, protectin; PEA, palmitoyl ethanolamide; PG, prostaglandin; PLA2, phospholipase A2; PLD, phospholipase D; PPAR, peroxisome proliferator-activated receptor; PUFA, polyunsaturated fatty acid; Rv, resolvin; S1P, sphingosine-1-phosphate; S1P1, sphingosine-1-phosphate receptor 1; TP, thromboxane receptor; TRPV, transient receptor potential vanilloid; TX, thromboxane.
(PGs), thromboxanes (TXs), and prostacyclin (PGI2), the lipoxygenase (LOX)–derived leukotrienes (LTs), lipoxins, E-series resolvins, and hydroxy fatty acids (HFAs), including hydroxyeicosatetraenoic acids (HETE) and hydroxyeicosapentaenoic acids (HEPE), derived from LOX and cytochrome P450 (CYP)–mediated reactions. Oxygenation of linoleic acid (LA, 18:2n − 6) gives rise to octadecanoids such as hydroxyoctadecadienoic acids (HODE), whereas docosahexaenoic acid (DHA, 22:6n-3) is the precursor of docosanoids such as D-series resolvins and protectins and hydroxydocosahexaenoic acids (HDHAs) (reviewed in (Massey and Nicolaou, 2011; Nicolaou, 2013)). A number of PG, HETE, HODE, and LT have been identified in human skin and contribute to keratinocyte proliferation, melanocyte dendricity, photocarcinogenesis, allergy, and inflammation (Ziboh et al., 2000; Scott et al., 2004; Honma et al., 2005; Satoh et al., 2006; Rhodes et al., 2009; Kendall and Nicolaou, 2013). The endocannabinoids N-arachidonoyl ethanolamide (anandamide; AEA) and 2-arachidonoyl glycerol (2-AG) are also derivatives of AA and function as endogenous lipid ligands of the two cannabinoid receptors (CB), CB1 and CB2 that are expressed throughout the skin in keratinocytes, melanocytes, fibroblasts, sebocytes, and hair follicles (Stander et al., 2005; Dobrosi et al., 2008; McPartland, 2008; Pucci et al., 2012). Cell and organ culture systems indicate that endocannabinoids are released by, and alter the func-
tions of, various skin cells (Dobrosi et al., 2008; Toth et al., 2011; Czifra et al., 2012; Pucci et al., 2012; Sugawara et al., 2012), and they have been implicated in several cutaneous pathologies (Kupczyk et al., 2009). A range of N-acyl ethanolamides (NAEs) can derive from membrane PUFA. Their relevance to cutaneous biology is illustrated by the activity of N-palmitoyl ethanolamide (PEA), which has been implicated in suppression of mast cell degranulation and inflammatory cytokine release, and explored for the treatment of atopic eczema (Eberlein et al., 2008; Petrosino et al., 2010; De Filippis et al., 2011). Sphingolipids are amides of sphingoid bases and an array of complex species can be derived through addition of fatty acids and head groups. Species present in the skin include free sphingoid bases and hundreds of ceramide species, as well as their phosphorylated versions (Masukawa et al., 2008; van Smeden et al., 2011; Rabionet et al., 2014; t'Kindt et al., 2012). The long-chain omega-esterified ceramides found in the stratum corneum are considered structural lipids pivotal for the integrity of the epidermal barrier, and recent studies have revealed their diversity of structure and physiological functions (Masukawa et al.., 2008; van Smeden et al., 2011; Iwai et al., 2012; Janssens et al., 2012; Breiden and Sandhoff, 2014). In contrast, non-hydroxylated ceramides and sphingoid bases are important mediators of immune cell regulation, with roles in homeostasis and inflammation (Chalfant and Spiegel, 2005; Uchida, 2014), and require further investigation to elucidate their contributions in the skin. The dermis and the epidermis exhibit distinct cell populations, enzyme profiles, and roles, and consequently lipid mediator expression could differ between the compartments. Although the epidermis comprises keratinocytes, Langerhans cells, melanocytes, and Merkel cells, the dermis contains fibroblasts, immune cells, hair follicles, sweat glands, blood vessels, and sensory nerves and provides the skin with elasticity and resistance to mechanical stress. Historically, the epidermis has been considered more important in terms of lipid biology because of the lipid-rich nature of the epidermal barrier. However, the dermis also displays considerable activity and provides biochemical support to the epidermis, with mediator cross talk between compartments––e.g. the epidermis exhibits impaired chain elongation of essential fatty acids and depends on the dermis for local production of longchain PUFA including AA, EPA, and DHA (Chapkin et al., 1986), whereas dermally derived 15-HETE impairs epidermal 12-LOX activity and dermally produced PGE2 stimulates keratinocyte proliferation (Kragballe et al., 1986; Conconi et al., 1996). Despite increased awareness of the role of lipid mediators in cutaneous physiology and disease, there is insufficient information on individual lipid mediator species and their variation through the skin. To address this, we undertook comprehensive analysis of (i) eicosanoids and related species, (ii) endocannabinoids, (iii) N-acyl ethanolamides, and (iv) non-hydroxylated ceramides and free sphingoid bases, in human skin tissue and blister fluid, using mass spectrometrybased targeted mediator lipidomics assays. Our aim was to assess the production, range, and variation of cutaneous lipid www.jidonline.org 1511
AC Kendall et al.
Cutaneous Bioactive Lipids
metabolites, including the contributions of dermal and epidermal compartments, in humans. This valuable information can provide insights into the mode of action of bioactive lipids, including the anti-inflammatory omega-3 fatty acids, and assist understanding of cutaneous biology, with potential for wide application in skin disease and its treatment. RESULTS Eicosanoids in human dermis, epidermis, and blister fluid
Eighteen species of prostanoids and 12 HFA were identified and quantified (Figure 2a; fold changes from dermal to epidermal expression are presented in Supplementary Table S1 online). The AA-derived prostaglandins PGE2, PGD2, PGF2α, and PGI2 (as its stable metabolite 6-keto PGF1α) were found in all tissues tested, with prevalence significantly higher in the epidermis (P = 0.0003, P = 0.003, P = 0.003, and P = 0.00004, respectively, compared with the dermis); TXA2 (measured as its stable metabolite TXB2) and PGJ2 were found predominantly in the epidermis. The EPA derived PGE3, PGD3, and PGF3α, and DGLA derived PGE1 and PGD1, although identified in both compartments were higher in the epidermis. Higher epidermal prevalence was also observed for all keto- and dihydro-keto PG (e.g. 15-keto-PGE2 and 13,14-dihydro-15-keto-PGE2); these derive from tissue– specific catabolic reactions reducing the bioactivity of primary PG (Tai et al., 2002). A large number of LOX-derived HFA products of DGLA, AA, EPA, LA, and DHA were found in both the dermis and the epidermis (HETrE, HETE, HEPE, HODE, and HDHA, respectively). Interestingly, the LA–derived 9-HODE and 13-HODE were the predominant species in both compartments, although found at higher levels in the epidermis (P = 0.015, P = 0.045, respectively) with 9-HODE at mean concentrations of 326.9 ± 88.2 pg per mg protein in the dermis and 2,728.5 ± 548.8 pg per mg protein in the epidermis and 13-HODE at 350.6 ± 83.5 pg per mg protein in the dermis and 1,416.7 ± 201.5 pg per mg protein in the epidermis. Although 13-HODE is considered a 15-LOX product, 9-HODE can be produced by a partially completed COX-2 reaction (Laneuville et al., 1995), its levels reflecting the high COX activity observed in both skin compartments. Concentrations of the 15-LOX products 15-HETrE, and 15-HETE were not significantly different between the dermis and the epidermis (Figure 2a; P = 0.75, P = 0.38 = 0.75, and P = 0.38, respectively). Although two more 15-LOX products were identified (EPA–derived 15-HEPE and DHA–derived 17-HDHA), these
were not found in all skin samples and their levels were close to the assay detection limit (20 pg and 10 pg on the column, respectively, Massey and Nicolaou, 2013). The 5-LOX– derived product of AA, 5-HETE, was found at a very low concentration (≤34.7 pg per mg protein) in both the epidermis and the dermis. The 12-LOX products 12-HETE, 12-HEPE, 13-HDHA, and 14-HDHA were significantly higher in the epidermis (P = 0.002, P = 0.027, P = 0.012, and P = 0.04, respectively) with 12-HETE being one of the predominant mediators found in both skin compartments (127.3 ± 10.5 pg per mg protein in the dermis and 570.4 ± 54.3 pg per mg protein in the epidermis). The epidermis had higher concentrations of 8- and 11-HETE, products of either oxidation or cutaneous CYP isoforms (Rhodes et al., 2009). The profile and level of eicosanoids found in suction blister fluid were similar to that in the epidermis, although with differences in some species (Figure 2a): 15-deoxy-Δ12,14 PGJ2 was found only in blister fluid, whereas 13,14 dihydro PGE1, PGJ2, PGD3, and PGF2α, all found in the dermis and epidermis, appeared to be minor species of blister fluid with concentrations close to the method detection limit (0.5–5 pg on the column) (Massey and Nicolaou, 2013). Endocannabinoids and N-acyl ethanolamides in human dermis, epidermis, and blister fluid
Nine endocannabinoids and NAE species derivatives of palmitic acid (PA; C16:0), alpha-linolenic acid (ALA; C18:3n − 3), linoleic acid (LA; C18:2n − 6), oleic acid (OA; C18:1n − 9), stearic acid (SA; C18:0), EPA, AA, and DHA were quantified. Although a larger number of NAE were detected, only those accurately identified and quantified based on the availability of commercial standards are reported. Both the dermis and epidermis expressed all species measured, mostly with comparable expression (Figure 2a). Derivatives of AA, EPA, and DHA––namely, AEA, eicosapentaenoyl ethanolamide (EPEA) and docosahexaenoyl ethanolamide (DHEA)––were significantly higher in the epidermis compared with the dermis (P = 0.016, P = 0.004, and P = 0.004, respectively) (Figure 2a and Supplementary Table S1 online). Finally, all these metabolites were also present in blister fluid, in ratios similar to the epidermis. Ceramides and sphingoid bases in the dermis, epidermis, and blister fluid
A huge diversity of cutaneous ceramides derives from the different sphingoid bases and acyl chains (Mizutani et al.,
Figure 2. Expression of eicosanoids, endocannabinoids, and N-acyl ethanolamides (a), and ceramides, phosphorylated ceramides, and sphingoid bases (b), in human dermis, epidermis, and blister fluid. All lipid mediators were analyzed by LC-MS/MS. The arachidonic acid–derived PGE2, 12-HETE, AEA, and 2-AG and C18 S and its derivatives C18 S1P, CER[N(22)S(18)], and d18:1/16:0 C1P are provided as example structures of the different classes of lipid mediators presented in this figure. Data for prostaglandins (PG), prostacyclin (PGI2) (measured as the stable derivative 6-keto PGF1α), thromboxanes (TX), hydroxy fatty acids, endocannabinoids, and N-acyl ethanolamides (NAEs) are expressed as pg per mg tissue protein or pg ml − 1 blister fluid. Data for ceramides, phosphorylated ceramides, and sphingoid bases are expressed as pmol per g protein or pmol μl − 1 blister fluid (the dermis and epidermis; n = 8 donors; labeled 1–8) and pg ml − 1 blister fluid (n = 3 donors, labeled 9–11). *Po0.05, **Po0.01, and ***Po0.001 when comparing the dermis with the epidermis. C18S: 18-carbon sphingosine (S); C18DS: 18-carbon dihydrosphingosine (DS); C18S1P: 18-carbon sphingosine-1-phosphate; C18DS1P: 18-carbon dihydrosphingosine-1-phosphate; Ceramide derivatives of sphingosine (S) with a non-hydroxy fatty acid (N) are named according to the number of carbons of the base (e.g. 16, 18, and so on) and fatty acid (e.g. 22, 24, and so on). Phosphorylated ceramides are denoted by the base (d18:1 representing sphingosine and d18:0 representing dihydrosphingosine) and the fatty acid (e.g. 16:0 represents palmitic acid).
1512 Journal of Investigative Dermatology (2015), Volume 135
AC Kendall et al.
Cutaneous Bioactive Lipids
Epidermis
Dermis
Lipid mediator
Blister fluid
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10 11 PGE1** PGD1** 13,14 dihydro-15-keto PGE1* PGF1α** 13,14 dihydro PGE1** PGE2*** PGD2** 13,14 dihydro-15-keto PGE2* TXB2 PGJ2* 15-deoxy-Δ12,14PGJ2 15-keto PGE2** PGF2α** 13,14 dihydro-15-keto PGF2α 6-keto PGF1α*** PGE3*** PGD3 PGF3α 9-HODE* 13-HODE* 15-HETrE 5-HETE 8-HETE* 11-HETE** 15-HETE 12-HETE** 11-HEPE** 12-HEPE* 14-HDHA* 13-HDHA* PEA ALEA LEA OEA SEA EPEA** AEA* DHEA** 2-AG
pg per mg protein 0-10
80-160
40-80 80-160
160-320
160-320
Prostanoids
320-640 640-1,280
COOH HO
OH
PGE2
COOH
OH
12-HETE
O OH
NH
AEA
O OH O OH 2-AG
Blister pmol per g fluid protein 0-10 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 1011 Epidermis
pmol per μl blister fluid 0-10 10-20 20-40
C18 S1P*
10-20 20-40 40-80
C18 DS1P*
80-160
40-80 80-160
160-320
160-320
C18 S C18 DS**
320-640 640-1,280
d18:1/16:0 d18:0/16:0 d18:1/18:0
1,280-2,560 2,560-5,120 5,120-10,240
N(24)S(16) N(22)S(18)
Ceramides
10,240+
O
d18:1/14:0
Phosphorylated ceramides
1,280-2,560 2,560-5,120 5,120-10,240
10,240+
Hydroxy fatty acids
Phosphorylated bases
320-640 640-1,280
1,280-2,560 2,560-5,120 5,120-10,240
Endocannabinoids and NAE
Sphingoid bases
0-10 10-20 20-40
10-20 20-40 40-80
Dermis
Lipid mediator
pg per ml blister fluid
10,240+
N(24)S(17) N(23)S(18) N(26)S(16)** N(24)S(18) N(22)S(20)
320-640 640-1,280 1,280-2,560 2,560-5,120 5,120-10,240 10,240+
N(24)S(19) N(24)S(20)** N(25)S(20) N(24)S(21) N(25)S(22) N(24)S(24)
O OH O P OH OH
OH OH
NH2
C18 S
O OH O P OH OH
C18 S1P
OH OH
NH
NH
O
d18:1/16:0 C1P
CER[N(22)S(18)]
O
NH2
www.jidonline.org 1513
AC Kendall et al.
Cutaneous Bioactive Lipids
2009; van Smeden et al., 2011; Janssens et al., 2012; t'Kindt et al., 2012; Rabionet et al., 2014). Initial screening for ceramides for which information on specific transition ions suitable for mass spectrometric analysis was available, was based on the data reported by Masukawa et al (2008). Analysis of whole skin lipid extracts indicated the presence of ceramides with sphingosine, dihydrosphingosine, phytosphingosine, and 6-hydroxy sphingosine bases, and non-hydroxy, alpha-hydroxy, or ester-linked omega hydroxy fatty acids, belonging to 11 of the 15 ceramide families currently identified (Rabionet et al., 2014)––namely, CER[NS], CER[AS], CER [NH], CER[AH], CER[NP], CER[AP], CER[NDS], CER[ADS], CER[EOP], CER[EOS], and CER[EOH]. Lack of commercially available appropriate synthetic standards did not permit identification of individual species. We therefore focused on non-hydroxylated ceramides of the CER[NS] family and phosphorylated ceramides, sphingoid bases, and phosphorylated base species that could be analyzed using 17-carbon-containing species as internal standards (Figure 2b and Supplementary Table S1 online). Further to their role in epidermal barrier function, the CER[NS] family includes bioactive mediumchain (40-48-carbon) non-hydroxylated ceramides involved in cutaneous inflammation (Uchida, 2014). Overall, we identified two sphingoid bases––18-carbon sphingosine (C18 S) and 18- carbon dihydrosphingosine (C18 DS)––and their phosphorylated forms 18-carbon sphingosine1-phosphate (C18 S1P) and 18-carbon dihydrosphingosine-1phosphate (C18 DS1P). We also found 13 CER[NS] ceramides and four phosphorylated ceramides with long-chain nonhydroxy fatty acids (22–26 carbons) and sphingoid bases with 16–24 carbons. With the exception of phosphorylated ceramides that were expressed at similar levels, the epidermis showed greater expression of all these sphingolipids compared with the dermis. However, high inter-individual variability meant there was limited statistical significance, with only two ceramide species, CER[N(26)S(16)] and CER [N(24)S(20)] (P = 0.01, P = 0.005, respectively), two phosphorylated bases (C18S1P, C18DS1P, P = 0.024 and P = 0.018, respectively), and one sphingoid base (C18DS, P = 0.003) being statistically significantly higher in the epidermis (Figure 2b). Of the sphingolipid species found in blister fluid, the majority were in similar proportions to the dermis and epidermis (Figure 2b). However, some species were at noticeably low levels in the blister fluid, including C18S and the largest ceramides measured (CER[N(26)S(16)], CER[N(24)S(21)], CER[N(25)S(22)], and CER[N(24)S(24)]). Contribution of cutaneous fatty acids
The fatty acid precursors of the eicosanoids and NAE reported here comprise 77.1% and 61.9% of fatty acids in the dermis and epidermis, respectively (Figure 3d). Other, non-precursor, fatty acids make up the remainder (Supplementary Table S2 online). Despite similar proportions of LA in the dermis and epidermis (10.7 and 9.6% of total fatty acids, respectively), the epidermis demonstrated higher levels of the LA–derived 9and 13-HODE compared with the dermis (69.7 and 50.3% of total HFA detected in the dermis and epidermis, respectively), 1514 Journal of Investigative Dermatology (2015), Volume 135
and this was reflected in the composition of blister fluid (65.2% of total HFA detected in blister fluid) (Figure 3b). The eicosanoid precursors AA and EPA contributed in all classes of mediators detected. AA was found at a higher concentration in the epidermis compared with the dermis (2.7% vs. 0.7% of total, respectively), and this was directly reflected in the higher abundance of AEA in the epidermis and blister fluid (13.1% and 11.2%, respectively) compared with the dermis (8.3%). In all cases, AEA and EPEA levels were lower compared with DHEA (Figure 3c). Although cutaneous DHA was detected at very low levels in the epidermis and dermis (0.2% and 0.5%, respectively) and was only a minor contributor of epidermal, but not dermal, HFA (Figure 3b), it was detected as DHEA in the dermis, epidermis, and blister fluid (11.7%,16.5%, and 14.6%, respectively) (Figure 3c). The high levels of dermal OA (44.8% of total fatty acids) were reflected in the higher abundance of OEA (18.3% of total dermal NAE), whereas PA, which was the most abundant fatty acid in the skin, contributed as PEA to almost 40% of the NAE detected in the dermis, epidermis, and blister fluid (Figure 3c). Manipulation of lipid mediators by omega-3 PUFA supplementation ex vivo
Exogenous provision of EPA did not have a statistically significant effect on the levels of COX–derived PGE2 and LOX–derived 12-HETE in the dermis or epidermis, although it induced the formation of PGE3 and 12-HEPE, two less inflammatory eicosanoids that can attenuate PGE2– and 12HETE–mediated activities (Figures 4a–d). Conversely, DHA appeared to inhibit PGE2 with concomitant stimulation of 12HETE production, suggesting the diversion of cutaneous AA from COX to LOX–mediated metabolism. Both EPA and DHA stimulated the production of AEA, EPEA, and DHEA in the epidermis and dermis, indicating that their antiinflammatory activities may be mediated, at least in part, via the endocannabinoid system (Figures 4e and f). Finally, the formation of 18-HEPE, 17-HDHA, and 14-HDHA post supplementation shows that human skin can produce the biochemical precursors of the anti-inflammatory and protective lipids resolvins (RvE and RvD) and maresins (Figures 4c and d) (Serhan, 2014). DISCUSSION Our findings demonstrate the ability of both the dermis and epidermis to produce an array of bioactive lipids that may act locally or be transported and contribute to cross talk between these skin compartments. It is noteworthy that despite differences in their physiology, the dermis and epidermis produce the same mediators, although significant differences in the levels and ratios of certain species highlight their varying requirements. Higher production of AA, DGLA, and EPA-derived epidermal prostanoids can be attributed to increased expression and/or activity of COX isoforms and prostanoid synthases, and agrees with a higher epidermal concentration of AA (Figure 3), as well as reports of higher expression of the constitutive COX-1 in the epidermis (Kragballe et al., 1986). The presence
AC Kendall et al.
Cutaneous Bioactive Lipids
Dermis
N-acyl ethanolamides
Hydroxy fatty acids
Prostanoids
1.6%
1.5%
8.0%
DGLA
10.2%
AA
EPA
EPA 88.3%
LA
5.9%
2.4% 1.7% 1.1%
LA AA EPA DGLA DHA
EPA DGLA
4.1%
25.2%
50.3%
11.7% 8.3%
38.7%
11.6% 18.3% 6.2%
1.0%
1.0%
1.1%
0.9%
LA AA EPA DGLA DHA
32.0%
69.7%
PA ALA LA OA SA EPA AA DHA
DGLA AA EPA
94.2%
AA
40.9%
0.7% 5.1%
DGLA
AA
90.5%
3.0%
Blister fluid
Epidermis
65.2%
PA ALA
16.5% 34.7% 13.1% 6.3% 11.4% 11.3%
1.2%
14.6%
LA 36.3% OA 11.2% SA EPA 5.4% 11.5% AA 1.2% 14.0% DHA 5.8%
PA ALA LA OA SA EPA AA DHA
5.5%
Total fatty acids
0.7% 0.2% 0.7% 20.1% 19.9% 10.7%
0.1% 44.8% 2.9%
PA ALA LA OA SA EPA
2.7% 0.5%
0.5% 0.5% 16.0%
22.1%
AA DHA Other
23.9% 9.6%
24.3%
PA ALA LA OA SA EPA AA DHA Other
Figure 3. Contribution of cutaneous fatty acid precursors of lipid mediator populations found in the dermis, epidermis, and suction blister fluid. Prostanoids (a), hydroxy fatty acids (b), and N-acyl ethanolamides (c) (as quantified in Figure 2) are shown as a % of total mediators detected in each tissue, together with the % abundance of their precursor fatty acid (d) in the dermis and epidermis. Data are expressed as a % of total lipid mediators detected in the dermis and epidermis (n = 8 donors), as a % of total lipid mediators detected in the blister fluid (n = 3 donors), or as a % of total fatty acids detected in the dermis and epidermis (n = 5 donors). AA, arachidonic acid; ALA, α-linolenic acid; DGLA, dihomo-gamma-linolenic acid; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; LA, linoleic acid; OA, oleic acid; PA, palmitic acid; SA, stearic acid.
of deactivated prostanoids, such as 15-keto-PGE2 and 13, 14-dihydro-15-keto PGE2, shows that the levels of potent proinflammatory PG such as PGE2 are under an active control, a step crucial for successful resolution of inflammation. Although the prostanoids found in blister fluid matched the epidermal profile (Figure 2a), some minor epidermal species were not detected, suggesting that skin biopsy samples are more appropriate when investigating low abundance lipid mediators. Interestingly, PGJ2 was found in the epidermis, but not blister fluid, whereas its metabolite 15-deoxy-Δ12,14 PGJ2 was found at high levels in blister fluid but not in the epidermis or dermis. As 15-deoxy-Δ12,14 PGJ2 is believed to act as a PPARγ agonist involved in anti-inflammatory proresolving signaling, its presence in blister fluid could potentially be linked to a cutaneous healing response to the trauma involved in raising suction blisters (Surh et al., 2011). The epidermal expression of HFA was also higher, with LA– and AA–derived species being the predominant mediators detected (Figure 2a). High production of 12-HETE
suggests increased expression/activity of 12-LOX and agrees with reports of three different isozymes highly expressed in the epidermis (Takahashi et al., 1993; Boeglin et al., 1998; Muller et al., 2002). 12-HETE has proinflammatory and neutrophil chemotactic properties indicating the potential involvement of 12-LOX in inflammatory skin disease (e.g. psoriasis) (Fogh et al., 1993; Baer et al., 1995). Although the LA-derived 15-LOX product 13-HODE was found at higher levels in the epidermis, the AA-derived 15-HETE and DGLAderived 15-HETrE were present at similar levels in both compartments, albeit at lower concentrations, in accordance with the much lower levels of their precursor fatty acids (Figure 3). 15-LOX products have anti-inflammatory activities and may have a specific role in the cross talk of the dermis and the epidermis, as demonstrated by the regulation exhibited by dermal 15-HETE on epidermal 12-LOX activity (Kragballe et al., 1986; Yoo et al., 2008). In addition, hair follicles, found in the dermis, are known to express high levels of a 15 S-LOX (Brash et al., 1997) that preferentially www.jidonline.org 1515
AC Kendall et al.
Cutaneous Bioactive Lipids
Epidermis
Dermis 4,000
Control
100
DHA
***
80 60 40
Control
3,000
EPA pg per mg protein
pg per mg protein
400 300 200
EPA
2,000
DHA
600 400 *** 200
20
2
PG
PG E
E
3
PG E
2
PG E
3
0
0
Epidermis
Dermis Control
200
***
***
***
500
DHA
150 100 50
EPA
400
DHA
300
**
200
*
100
DHA
200
**
100 10 8 6 4 2 0
pg per mg protein
**
*
A -H
D
H 14
D -H 17
H
A
E EP 18
12
-H
-H
ET -H 12
-H
***
Control
Epidermis
EPA 300
EP
E
A H D
H D 17
18
-H
-H
EP
A
E
E EP
E
-H 12
EPA
1,200 1,000 800 600 400 200
***
DHA **
** **
30 20 10
EA H
EA
A AE
D
D
H
EA
EA EP
AE
A
0 EP
ET -H 12
14
Control
Dermis 400
E
0
0
pg per mg protein
Control
***
EPA pg per mg protein
pg per mg protein
250
Figure 4. Effect of omega-3 PUFA supplementation on the production of cutaneous lipid mediators ex vivo. The skin was treated for 3 days with EPA or DHA (50 μM). PGE2 and PGE3 were extracted from the dermis (a) and epidermis (b), 12-HETE, 12-HEPE, 18-HEPE, 17-HDHA, and 14-HDHA extracted from the dermis (c) and epidermis (d), AEA, EPEA, and DHEA extracted from the dermis (e) and epidermis (f), and were analyzed by LC-MS/MS. Data are expressed as pg per mg tissue protein (n = 3 biopsies). *Po0.05, **Po0.01, and ***Po0.001 when comparing all data with control. PG, prostaglandin; HETE, hydroxyeicosatetraenoic acid; HEPE, hydroxyeicosapentaenoic acid; HDHA, hydroxydocosahexaenoic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; AEA, N-arachidonoyl ethanolamide; EPEA, N-eicosapentaenoyl ethanolamide; DHEA, N-docosahexaenoyl ethanolamide.
metabolizes AA over LA. Although the abdominal skin used in this study was not densely populated with hair, the presence of hair follicles in the dermis could possibly explain why 15HETE was present at comparable levels in the dermis and epidermis but 13-HODE was not. Finally, the 5-LOX product 5-HETE was found at very low levels in both skin compartments, suggesting a low cutaneous expression of 5-LOX. This supports current understanding that 5-LOX–derived leukotrienes, found in inflammatory skin disease such as psoriasis, are derived from infiltrating immune cells rather than resident cells (Ford-Hutchinson, 1993; Sadik et al., 2014). The significantly higher epidermal levels of the endocannabinoid AEA, and the NAE species EPEA and DHEA, are of interest, indicating their potential roles in homeostatic processes in resting skin. AEA has been reported to regulate 1516 Journal of Investigative Dermatology (2015), Volume 135
keratinocyte proliferation, sebum production, and melanogenesis (Conconi et al., 1996; Dobrosi et al., 2008; Pucci et al., 2012). It is noteworthy that, although all HFA derivatives of DHA contributed o2% of total class species, DHEA was one of the main NAE detected (11–16% of total; Figure 3c). DHEA is an endocannabinoid-like molecule exhibiting weak affinities for cannabinoid receptors (Kim and Spector, 2013). EPEA and DHEA have anti-inflammatory properties, reducing lipopolysaccharide (LPS)–induced IL-6 and monocyte chemotactic protein (MCP)-1 production in adipocytes, and LPS–induced ˙NO production in macrophages (Satoh et al., 2006; Uchida, 2014), and may operate similar anti-inflammatory effects in the skin. The saturated NAE species, PEA and OEA, derivatives of palmitic (PA) and oleic (OA) acid, respectively, were found at equally high
AC Kendall et al.
Cutaneous Bioactive Lipids
levels in both the dermis and epidermis. PEA is known to suppress cutaneous mast cell activity and may confer antiinflammatory activity (Mizutani et al., 2009; Wang and Ueda, 2009). OEA is known to affect food intake and the sleep-wake cycle, but its role in the skin is unknown (Honma et al., 2005). Supplementation with exogenous EPA and DHA showed that their cutaneous metabolism alters the profile of dermal and epidermal lipids, stimulating the production of anti-inflammatory species including COX– and LOX–derived products, and NAE (Figure 4). This provides information on how the skin can respond to nutritional interventions and suggests that n-3PUFA can also create a protective and pro-resolving environment, as supported by the identification of biochemical precursors for protectins and resolvins in the skin. Ceramides are components of the stratum corneum involved in epidermal barrier function, particularly the omega-esterified species with long acyl chains, with more than 300 species identified to date (Behne et al., 2000; van Smeden et al., 2011; Iwai et al., 2012; Rabionet et al., 2014). In this study, we focus our attention particularly on the family of non-hydroxylated medium-chain ceramides that can be found in cellular membranes, and, notably, we demonstrate the presence of 21 species in the epidermis, as well as in the dermis and blister fluid. This suggests their wider involvement in epidermal and dermal function, where they may mediate activities including regulation of apoptosis and inflammation (Stiban et al., 2008; Maceyka and Spiegel, 2014). Blister fluid showed lower expression of the longest ceramides measured, possibly indicating impaired release of these species during the blister formation. Further work is indicated to elucidate the prevalence of these non-hydroxylated species in epidermal and dermal cells and explain the intriguing finding of their comparable abundance in both skin compartments. Phosphorylated ceramides were also present in the dermis and epidermis at equally high concentrations, suggesting that they originate from cellular membranes and are not restricted to the differentiated epidermis. This is indicative of an active signaling role in both compartments that may include the ceramide-1-phosphate– and sphingosine-1-phosphate–mediated activation of cutaneous PLA2 (Chalfant and Spiegel, 2005). Ceramide-1-phosphate has also been recently reported to be involved in wound healing through acting in concert with eicosanoid production (Wijesinghe et al., 2014), and sphingosine-1-phosphate is a modulator of cutaneous immunity (Herzinger et al., 2007). In this study, we focused on the more common 18-carbon sphingosine and dihydrosphingosine bases and found that the free sphingoid base C18DS, known to promote keratinocyte differentiation (Paragh et al., 2008), demonstrated increased expression in the epidermis (Figure 2b). Future investigations could include analysis of the less–abundant sphingosines and dihydrosphingosines with other carbon chain-lengths, as well as the phytosphingosine and 6-hydroxy sphingosine bases. A limitation of the present study is the method used to separate the dermis from the epidermis––i.e. physical separation. The commonly used methods for their separation along the dermo–epidermal junction rely on long incubation with salts or enzymes, or rapid transition from high to low
temperatures, to degrade the basement membrane, and are inappropriate for the purposes of this study as they can induce oxidation or thermal degradation of the lipid mediators (Zhang et al., 2003; Oakford et al., 2011). Although physical separation of the dermis and the epidermis resulted in slight dermal contamination of the epidermis, care was taken that the less cellular dermis would remain completely free of any epidermal component. Following analysis, it was found that differences in lipid expression between the two compartments were reliably identified. Overall, this study shows that the dermis produces a wide range of lipid mediators and this production may be an important biochemical support for the epidermis, acting as an intermediary between the epidermal interface with the external environment and the subcutaneous circulatory and lymphatic systems. To our knowledge, this is the most complete analysis of bioactive lipid mediators in human skin to be reported. Profiles of lipid mediators can support the characterization of COX, LOX, and CYP isoforms, whose activity and pattern of expression depend on post-translational modifications and may contribute to cutaneous disorders (Aguiar et al., 2005; Kuhn and O'Donnell, 2006; Mbonye and Song, 2009). We have also shown an ability to manipulate the balance of bioactive lipids in favor of the protective omega-3–derived species, with implications for their mode of action in skin health and disease. Finally, direct comparison of lipid mediators found in suction blister fluid, a widely used technique for sampling cutaneous intercellular fluid, with skin lipid profiles, provides confirmatory information on study design to explore the potency and importance of lipid mediators in skin biology. MATERIALS AND METHODS Tissue samples Biobank skin. Skin samples were obtained from a biobank (Ethical Tissue, University of Bradford, Bradford, UK) with full ethical approval (Leeds East Research Ethics Committee reference 07/H1306/98+5). The skin was provided by 8 healthy donors (6 female, 2 male; 30–60 years; white Caucasian), undergoing elective abdominoplastic surgery, delivered to the biobank within 1 h of the operation. The tissue was washed and the adipose layer removed. Punch biopsies (3 mm) were cut, snap frozen and stored at − 80 ºC, or were cultured ex vivo. Ex vivo culture was performed in DMEM containing 100 U ml − 1 penicillin, 100 μg ml − 1 streptomycin, and 1.4 mM Ca2+(Promocell, Heidelberg, Germany), with or without the addition of EPA or DHA (50 μM) (Sigma Aldrich, Poole, UK), for 3 days (Tavakkol et al., 1999). For lipidomics assays, three punch biopsies per donor were used. Before analysis, the skin was divided into the dermis and epidermis (on ice, by scalpel, with the aid of visual inspection at × 40 magnification). Although epidermal samples demonstrated minor contamination with dermal tissue, care was taken that the dermal tissue was not contaminated with the epidermal tissue. Skin suction blister fluid and punch biopsies. Healthy human volunteers (n = 8; all females; 28–56 years; white Caucasian) were recruited by the Photobiology Unit, Dermatology Centre, Salford Royal Hospital, Manchester, UK. Ethical approval was www.jidonline.org 1517
AC Kendall et al.
Cutaneous Bioactive Lipids
obtained from the North Manchester Research Ethics Committee (reference 08/H1006/79). Written informed consent was obtained from participants and the study adhered to Declaration of Helsinki principles. Volunteers provided 5 mm skin punch biopsies and suction blisters from buttock skin for these analyses. Punch biopsy dermis and epidermis were separated as described above, and samples used for total fatty acid analysis. Suction blistering was performed using suction cups with a 1-cm central aperture as described previously (Rhodes et al., 2009). Skin blister fluid was aspirated with a 23-gauge needle, snap-frozen in liquid nitrogen, and stored at − 80ºC. Eicosanoid extraction and analysis Skin samples (30–60 mg) and blister fluid samples (40–90 μl) were extracted using ice-cold 15% (v/v) methanol solution. 12-HETE-d8 and PGB2-d4 (Cayman Chemicals, Ann Arbor, MI, USA) were used as internal standards, as published before (Masoodi and Nicolaou, 2006; Masoodi et al., 2008; Massey and Nicolaou, 2013).The extracts were semi-purified by SPE cartridges (C18-E; Phenomenex, Macclesfield, UK); the assay recovery was estimated at 96–98%. LC/ ESI-MS/MS was performed on an HPLC pump (Waters Alliance 2695) coupled to an electrospray ionization triple quadrupole mass spectrometer (Quattro Ultima, Waters, Elstree, Hertfordshire, UK). Results are expressed as pg per mg protein (skin) or pg ml − 1 (blister fluid), estimated using calibration lines covering the range 0.02–20 ng ml − 1. Detailed description of the experimental protocol is provided as Supplementary Methods SM1 online.
Endocannabinoid and NAE extraction and analysis Skin samples (30–60 mg) and blister fluid samples (40–90 μl) were extracted using ice-cold 2:1 (v/v) chloroform/methanol. Arachidonoyl EA-d8 and 2-arachidonoyl glycerol-d8 (Cayman Chemicals, Ann Arbor, MI) were used as internal standards. Lipid extracts were analyzed by LC/ESI-MS/MS. Recovery was estimated at 91%. Results are expressed as pg per mg protein (skin) or pg ml − 1 (blister fluid), estimated using calibration lines covering the range 0.02–50 ngml − 1. Detailed description of the experimental protocol is provided as Supplementary Methods SM2 online.
Sphingolipid extraction and analysis Skin samples (30–60 mg) and blister fluid samples (40–90 μl) were extracted using ice-cold isopropanol:water:ethyl acetate (30:10:60; v/v/v). The following internal standards were used: C17 S (used for C18 S), C17 DS (used for C18 DS), C17 S1P (used for C18 S1P), C17 DS1P (used for C18 DS1P), d18:1/12:0 C1P (used for all C1P species), and C25 Cer (used for all CER[NS] species) (Ceramide/ Sphingoid Internal Standard Mixture I, Avanti Polar Lipids, Alabaster, AL). The resulting lipid extracts were analyzed by LC/ESI-MS/MS (Bielawski et al., 2006; Kelly et al., 2011). The assay recovery was estimated at 75–79%. Results are expressed as pmol per g protein (skin) or pmol μl − 1 (blister fluid), relative to the appropriate internal standard (covering the range 0.005–8.0 nmol ml − 1). Detailed description of the experimental protocol is provided as Supplementary Methods SM3 online.
Fatty acid analysis Fatty acids were analyzed by gas chromatography (GC) in the dermis and epidermis as previously described (Pilkington et al., 2014). 1518 Journal of Investigative Dermatology (2015), Volume 135
Results are expressed as a percentage of total fatty acids. Detailed description of the experimental protocol is provided as Supplementary Methods SM4 online.
Protein content During lipid extractions, protein pellets were retained for analysis of protein content using a standard Bradford protein assay kit (Bio-Rad, Hemel Hempstead, UK) (Bradford, 1976). Proteins were extracted using 1 M NaOH and analyzed within the linear range of the assay to ensure accuracy.
Statistical analysis Statistical analyses of lipid mediator expression were performed using repeated measures ANOVAs with Greenhouse-Geisser corrections and Bonferroni post-hoc tests. Analyses were conducted using SPSS 20 software, and Po0.05 was considered significant.
CONFLICT OF INTEREST
The authors state no conflict of interest.
ACKNOWLEDGMENTS This work was funded by Unilever as part of its ongoing program developing novel non-animal approaches for assessing consumer safety. We thank Andrew Healey, Analytical Centre, University of Bradford, and Wayne Burrill, Ethical Tissue, University of Bradford, for the excellent technical support. SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper at http:// www.nature.com/jid
REFERENCES Aguiar M, Masse R, Gibbs BF (2005) Regulation of cytochrome P450 by posttranslational modification. Drug Metab Rev 37:379–404 Baer AN, Klaus MV, Green FA (1995) Epidermal fatty acid oxygenases are activated in non-psoriatic dermatoses. J Invest Dermatol 104:251–5 Behne M, Uchida Y, Seki T et al. (2000) Omega-hydroxyceramides are required for corneocyte lipid envelope (CLE) formation and normal epidermal permeability barrier function. J Invest Dermatol 114:185–92 Bielawski J, Szulc ZM, Hannun YA et al. (2006) Simultaneous quantitative analysis of bioactive sphingolipids by high-performance liquid chromatography-tandem mass spectrometry. Methods 39:82–91 Boeglin WE, Kim RB, Brash AR (1998) A 12 R-lipoxygenase in human skin: mechanistic evidence, molecular cloning, and expression. Proc Natl Acad Sci USA 95:6744–9 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–54 Brash AR, Boeglin WE, Chang MS (1997) Discovery of a second 15 S-lipoxygenase in humans. Proc Natl Acad Sci USA 94:6148–52 Breiden B, Sandhoff K (2014) The role of sphingolipid metabolism in cutaneous permeability barrier formation. Biochim Biophys Acta 1841:441–52 Chalfant CE, Spiegel S (2005) Sphingosine 1-phosphate and ceramide 1-phosphate: expanding roles in cell signaling. J Cell Sci 118:4605–12 Chapkin RS, Ziboh VA, Marcelo CL et al. (1986) Metabolism of essential fatty acids by human epidermal enzyme preparations: evidence of chain elongation. J Lipid Res 27:945–54 Conconi MT, Bruno P, Bonali A et al. (1996) Relationship between the proliferation of keratinocytes cultured in vitro and prostaglandin E2. Ann Anat 178:229–36
AC Kendall et al.
Cutaneous Bioactive Lipids
Czifra G, Szollosi AG, Toth BI et al. (2012) Endocannabinoids regulate growth and survival of human eccrine sweat gland-derived epithelial cells. J Invest Dermatol 132:1967–76 De Filippis D, Luongo L, Cipriano M et al. (2011) Palmitoylethanolamide reduces granuloma-induced hyperalgesia by modulation of mast cell activation in rats. Mol Pain 7:3 Dobrosi N, Toth BI, Nagy G et al. (2008) Endocannabinoids enhance lipid synthesis and apoptosis of human sebocytes via cannabinoid receptor-2mediated signaling. FASEB J 22:3685–95 Eberlein B, Eicke C, Reinhardt HW et al. (2008) Adjuvant treatment of atopic eczema: assessment of an emollient containing N-palmitoylethanolamine (ATOPA study). J Eur Acad Dermatol Venereol 22:73–82 Edmondson SR, Thumiger SP, Werther GA et al. (2003) Epidermal homeostasis: the role of the growth hormone and insulin-like growth factor systems. Endocr Rev 24:737–64 Fogh K, Iversen L, Herlin T et al. (1993) Modulation of eicosanoid formation by lesional skin of psoriasis: an ex vivo skin model. Acta Derm Venereol 73: 191–3 Ford-Hutchinson AW (1993) 5-Lipoxygenase activation in psoriasis: a dead issue? Skin Pharmacol 6:292–7 Herzinger T, Kleuser B, Schafer-Korting M et al. (2007) Sphingosine-1phosphate signaling and the skin. Am J Clin Dermatol 8:329–36 Honma Y, Arai I, Hashimoto Y et al. (2005) Prostaglandin D2 and prostaglandin E2 accelerate the recovery of cutaneous barrier disruption induced by mechanical scratching in mice. Eur J Pharmacol 518:56–62 Iwai I, Han H, den Hollander L et al. (2012) The human skin barrier is organized as stacked bilayers of fully extended ceramides with cholesterol molecules associated with the ceramide sphingoid moiety. J Invest Dermatol 132: 2215–25 Janssens M, van Smeden J, Gooris GS et al. (2012) Increase in short-chain ceramides correlates with an altered lipid organization and decreased barrier function in atopic eczema patients. J Lipid Res 53:2755–66 Kelly L, Grehan B, Chiesa AD et al. (2011) The polyunsaturated fatty acids, EPA and DPA exert a protective effect in the hippocampus of the aged rat. Neurobiol Aging 32:e1–15 Kendall AC, Nicolaou A (2013) Bioactive lipid mediators in skin inflammation and immunity. Prog Lipid Res 52:141–64 Kim HY, Spector AA (2013) Synaptamide, endocannabinoid-like derivative of docosahexaenoic acid with cannabinoid-independent function. Prostaglandins Leukot Essent Fatty Acids 88:121–5 Kragballe K, Pinnamaneni G, Desjarlais L et al. (1986) Dermis-derived 15-hydroxy-eicosatetraenoic acid inhibits epidermal 12-lipoxygenase activity. J Invest Dermatol 87:494–8 Kuhn H, O'Donnell VB (2006) Inflammation and immune regulation by 12/15lipoxygenases. Prog Lipid Res 45:334–56 Kupczyk P, Reich A, Szepietowski JC (2009) Cannabinoid system in the skin – a possible target for future therapies in dermatology. Exp Dermatol 18: 669–79 Laneuville O, Breuer DK, Xu N et al. (1995) Fatty acid substrate specificities of human prostaglandin-endoperoxide H synthase-1 and -2. Formation of 12-hydroxy-(9Z, 13E/Z, 15Z)- octadecatrienoic acids from alphalinolenic acid. J Biol Chem 270:19330–6 Maceyka M, Spiegel S (2014) Sphingolipid metabolites in inflammatory disease. Nature 510:58–67 Masoodi M, Mir AA, Petasis NA et al. (2008) Simultaneous lipidomic analysis of three families of bioactive lipid mediators leukotrienes, resolvins, protectins and related hydroxy-fatty acids by liquid chromatography/ electrospray ionisation tandem mass spectrometry. Rapid Commun Mass Spectrom 22:75–83 Masoodi M, Nicolaou A (2006) Lipidomic analysis of twenty-seven prostanoids and isoprostanes by liquid chromatography/electrospray tandem mass spectrometry. Rapid Commun Mass Spectrom 20:3023–9 Massey KA, Nicolaou A (2011) Lipidomics of polyunsaturated-fatty-acid-derived oxygenated metabolites. Biochem Soc Trans 39:1240–6 Massey KA, Nicolaou A (2013) Lipidomics of oxidized polyunsaturated fatty acids. Free Radic Biol Med 59:45–55
Masukawa Y, Narita H, Shimizu E et al. (2008) Characterization of overall ceramide species in human stratum corneum. J Lipid Res 49:1466–76 Mbonye UR, Song I (2009) Posttranscriptional and posttranslational determinants of cyclooxygenase expression. BMB Rep 42:552–60 McPartland JM (2008) Expression of the endocannabinoid system in fibroblasts and myofascial tissues. J Bodyw Mov Ther 12:169–82 Mizutani Y, Mitsutake S, Tsuji K et al. (2009) Ceramide biosynthesis in keratinocyte and its role in skin function. Biochimie 91:784–90 Muller K, Siebert M, Heidt M et al. (2002) Modulation of epidermal tumor development caused by targeted overexpression of epidermis-type 12 S-lipoxygenase. Cancer Res 62:4610–6 Nicolaou A (2013) Eicosanoids in skin inflammation. Prostaglandins Leukot Essent Fatty Acids 88:131–8 Oakford ME, Dixon SV, August S et al. (2011) Migration of immunocytes across the basement membrane in skin: The role of basement membrane pores. J Invest Dermatol 131:1950–3 Paragh G, Schling P, Ugocsai P et al. (2008) Novel sphingolipid derivatives promote keratinocyte differentiation. Exp Dermatol 17:1004–16 Petrosino S, Cristino L, Karsak M et al. (2010) Protective role of palmitoylethanolamide in contact allergic dermatitis. Allergy 65:698–711 Pilkington SM, Rhodes LE, Al-Aasswad NM et al. (2014) Impact of EPA ingestion on COX- and LOX-mediated eicosanoid synthesis in skin with and without a pro-inflammatory UVR challenge − report of a randomised controlled study in humans.. Mol Nutr Food Res58: 580-90. Pucci M, Pasquariello N, Battista N et al. (2012) Endocannabinoids stimulate human melanogenesis via type-1 cannabinoid receptor. J Biol Chem 287: 15466–78 Rabionet M, Gorgas K, Sandhoff R (2014) Ceramide synthesis in the epidermis. Biochim Biophys Acta 1841:422–34 Rhodes LE, Gledhill K, Masoodi M et al. (2009) The sunburn response in human skin is characterized by sequential eicosanoid profiles that may mediate its early and late phases. FASEB J 23:3947–56 Sadik CD, Sezin T, Kim ND (2014) Leukotrienes orchestrating allergic skin inflammation. Exp Dermatol 22:705–9 Satoh T, Moroi R, Aritake K et al. (2006) Prostaglandin D2 plays an essential role in chronic allergic inflammation of the skin via CRTH2 receptor. J Immunol 177:2621–9 Scott G, Leopardi S, Printup S et al. (2004) Proteinase-activated receptor-2 stimulates prostaglandin production in keratinocytes: analysis of prostaglandin receptors on human melanocytes and effects of PGE2 and PGF2α on melanocyte dendricity. J Invest Dermatol 122:1214–24 Serhan CN (2014) Pro-resolving lipid mediators are leads for resolution physiology. Nature 510:92–101 Stander S, Schmelz M, Metze D et al. (2005) Distribution of cannabinoid receptor 1 (CB1) and 2 (CB2) on sensory nerve fibers and adnexal structures in human skin. J Dermatol Sci 38:177–88 Stiban J, Caputo L, Colombini M (2008) Ceramide synthesis in the endoplasmic reticulum can permeabilize mitochondria to proapoptotic proteins. J Lipid Res 49:625–34 Sugawara K, Biro T, Tsuruta D et al. (2012) Endocannabinoids limit excessive mast cell maturation and activation in human skin. J Allergy Clin Immunol 129:e8 Surh YJ, Na HK, Park JM et al. (2011) 15-Deoxy-Delta(1)(2),(1)(4)-prostaglandin J (2), an electrophilic lipid mediator of anti-inflammatory and pro-resolving signaling. Biochem Pharmacol 82:1335–51 t'Kindt R, Jorge L, Dumont E et al. (2012) Profiling and characterizing skin ceramides using reversed-phase liquid chromatography-quadrupole timeof-flight mass spectrometry. Anal Chem 84:403–11 Tai HH, Ensor CM, Tong M et al. (2002) Prostaglandin catabolizing enzymes. Prostaglandins Other Lipid Mediat 68-69:483–93 Takahashi Y, Reddy GR, Ueda N et al. (1993) Arachidonate 12-lipoxygenase of platelet-type in human epidermal cells. J Biol Chem 268:16443–8 Tavakkol A, Varani J, Elder JT et al. (1999) Maintenance of human skin in organ culture: role for insulin-like growth factor-1 receptor and epidermal growth factor receptor. Arch Dermatol Res 291:643–51
www.jidonline.org 1519
AC Kendall et al.
Cutaneous Bioactive Lipids
Toth BI, Dobrosi N, Dajnoki A et al. (2011) Endocannabinoids modulate human epidermal keratinocyte proliferation and survival via the sequential engagement of cannabinoid receptor-1 and transient receptor potential vanilloid-1. J Invest Dermatol 131:1095–104 Uchida Y (2014) Ceramide signaling in mammalian epidermis. Biochim Biophys Acta 1841:453–62 van Smeden J, Hoppel L, van der Heijden R et al. (2011) LC/MS analysis of stratum corneum lipids: ceramide profiling and discovery. J Lipid Res 52: 1211–21
Yamaguchi Y, Hearing VJ, Itami S et al. (2005) Mesenchymal-epithelial interactions in the skin: aiming for site-specific tissue regeneration. J Dermatol Sci 40:1–9 Yoo H, Jeon B, Jeon MS et al. (2008) Reciprocal regulation of 12- and 15lipoxygenases by UV-irradiation in human keratinocytes. FEBS Lett 582: 3249–53 Yu C, Fedoric B, Anderson PH et al. (2011) Vitamin D(3) signalling to mast cells: A new regulatory axis. Int J Biochem Cell Biol 43:41–6
Wang J, Ueda N (2009) Biology of endocannabinoid synthesis system. Prostaglandins Other Lipid Mediat 89:112–9
Zhang X-J, Chinkes DL, Wolfe RR (2003) Measurement of protein metabolism in epidermis and dermis. Am J Physiol Endocrinol Metab 284: E1191–201
Wijesinghe DS, Brentnall M, Mietla JA et al. (2014) Ceramide kinase is required for a normal eicosanoid response and the subsequent orderly migration of fibroblasts. J Lipid Res 55:1298–309
Ziboh VA, Miller CC, Cho Y (2000) Metabolism of polyunsaturated fatty acids by skin epidermal enzymes: generation of antiinflammatory and antiproliferative metabolites. Am J Clin Nutr 71:361 s–6
1520 Journal of Investigative Dermatology (2015), Volume 135
